13th Lumbar Spine Research Society Annual Scientific Meeting Oral and Poster Presentations

Free access

If the inline PDF is not rendering correctly, you can download the PDF file here.

Paper 1.1. Feasibility and Initial Clinical Experience of Lateral Interbody Fusion (LIF) using Prone Transpsoas (PTP) Access

Patel Ashish, MD1, Pimenta Luiz, MD2, Taylor William, MD3, Tohmeh Antoine, MD4

1DuPage Medical Group, Chicago, Illinois, United States, 2MultiCare Deaconess Hospital, Spokane, United States, 3University of San Diego School of Medicine, University of San Diego, San Diego, California, United States, 4Sao Paulo, Brazil, Sao Paulo, Brazil

Background/Introduction

Lateral interbody fusion (LIF) provides less invasive access to the anterior thoracolumbar spine in conditions requiring interbody correction and fusion. When performed as traditionally described with the patient in lateral decubitus, LIF requires either prone repositioning for supplemental posterior fixation / other surgical maneuvers or modifying traditional posterior techniques to be done while positioned lateral. However, the benefits of lateral anterior column access may be achievable with the patient prone, allowing for concomitant posterior techniques in a single-position setting more familiar to most spine surgeons.

Materials/Methods

Prone transpsoas (PTP) access was outlined and vetted in cadaveric simulations by a group of LIF-experienced spine surgeons. Clinical experience followed, prospectively capturing perioperative details to assess feasibility as well as measurable efficiencies or hurdles. The surgical technique is consistent with prior LIF descriptions, with the exception of prone positioning on a Jackson frame-type bed allowing the abdomen to hang freely and optimizing the positional effect on lumbar lordosis.

Results

Clinical experience includes 82 levels in 57 patients by 17 spine surgeons. Patient size varied (BMI 25-46, mean 31.3). Procedures were performed at between 1 (67%) and 4 (2%) lumbar levels; 65% inclusive of L4-5. Average time to dock the lateral retractor was 14 min, working through the retractor (transpsoas time) was 23 min. Most cases (65%) were supplemented using percutaneous pedicle fixation; 21% open pedicle fixation; 14% other. In some cases, posterior fixation was performed concurrent with the lateral approach (i.e., two surgeons working simultaneously). Concomitant procedures facilitated by prone position included direct decompression (30%), treatment at L5-S1 (11%), revision of posterior instrumentation (11%), and osteotomy/bony releases (9%). Estimated blood loss averaged 248 cc overall, 55 cc for the lateral procedure alone. Intraoperative complications included segmental bleeding in 2 cases, both controlled intraoperatively with no sequelae. Three planned lateral approaches were converted intraoperatively to TLIF without patient repositioning.

Discussion/Conclusion

Initial multicenter clinical experience suggests that PTP is not only feasible but creates efficiencies by allowing for single-position surgery maximizing both anterior and posterior column access and corrective techniques without increasing the complication profile.

Paper 01. The Effect of TNF Alpha Biological Medications in Spine Surgery on Reoperation Rates

Gaudiani Michael, BA1, Mroz Thomas, MD2, Winkelman Robert, MS, BS3, Coughlin Daniel, MD, Carbajal Valerie, BS1, Ravishankar Pavitra, BS1

1Case Western Reserve University, Cleveland, Ohio, United States, 2Cleveland Clinic Foundation, Cleveland, Ohio, United States, 3Center for Spine Health, Cleveland Clinic Foundation, Cleveland, Ohio, United States

Background/Introduction

While much has been published in the orthopedic joints literature regarding the increased risk of postoperative complications associated with Tumor necrosis factor-alpha inhibitor (TNF-AI) use, the association between TNF-AI use and post-operative complications has not yet been studied in the setting of spine surgery.

Materials/Methods

This study is a retrospective analysis of patients surgically treated with spinal fusion within a single large healthcare system from 1/1/2009-12/31/2018. There were three distinct cohorts of patients under study. A cohort of spine surgery patients prescribed TNF-AI within 90 days of surgery, a TNF-AI naïve cohort prescribed DMARD medications within 90 days of surgery, and a TNF-AI and DMARD naïve control cohort. The primary outcome of interest was reoperation within one year following surgery.

Results

Our study identified 97 TNF-AI, 97 DMARD, and 152 control patients. The median times TNF-AI medications were discontinued prior to surgery and restarted after surgery were 20 days [IQR:9-44] and 32 [IQR: 10-51], respectively. Reoperation up to one-year post-surgery occurred in 19% (n =18) of the TNF-AI group 15% (n=14) of the DMARD group and 6% (n=9) of the control group. The cox-proportional hazard model of reoperation within 1 year indicated that the risk of reoperation was 3.1 (95% CI:1.4 to 7.0) times higher in the TNF-AI group and 2.2 (95% CI 0.96 to 5.3) in the DMARD group compared to the control group. The reasons for reoperation for TNF-AI group were 40% (n=8) infection and 60% (n=12) other causes which included failure to fuse and adjacent segment disease. Reasons for reoperation at one year were 36% (n=5) infection and 64% (n=9) other causes for DMARD patients and 27% (n=3) infection with 73% (n=8) other causes for control patients.

Discussion/Conclusion

Patients taking TNF-AIs prior to surgery were found to have a significantly higher rate of reoperation in the 1 year following surgery compared to controls. The higher rate of reoperation associated with TNF-AI use prior to spinal fusion surgery represents the potential for higher morbidity and costs for patient.

Paper 02. Cost-effectiveness analysis of teriparatide therapy for the prevention of proximal junctional kyphosis/failure and subsequent revision after adult spinal deformity surgery

Okano Ichiro, MD1, Yang Jingyan, PhD1, Salzmann Stephan, MD1, Shue Jennifer, MS1, Sama Andrew, MD1, Cammisa Frank, MD1, Girardi Federico, MD1, Hughes Alexander, MD1

1Hospital for Special Surgery, New York, New York, United States

Background/Introduction

Recent studies have shown teriparatide therapy reduces the risk of proximal junctional kyphosis/failure (PJK/PJF) and subsequent revision after corrective surgery for adult spinal deformity (ASD). However, teriparatide usually costs more than other anti-osteoporotic treatments and its cost effectiveness for the prevention of these unfavorable sequelae is unclear. The aim of this study is to investigate the cost effectiveness of preoperative teriparatide therapy for ASD patients utilizing a Markov model of ASD corrective surgery.

Materials/Methods

A Markov model of ASD surgery was created incorporating the incidences of PJK/PJF and revision rates due to PJK/PJF in the surgically treated ASD patient with or without 12-month teriparatide treatment, revision rates for non-PJK causes, recurrence rates of PJK/PJF, post-operative mortality, and pre-/post-operative quality-adjusted life-years (QALYs), obtained from previously published literature. The effectiveness was measured in cost/QALY and the rates of patients who successfully achieved fusion without any correction loss due to PJK/PJF in each group.

Results

The Markov model simulation showed that the 10-year cumulative cost was higher in the teriparatide group ($131,000 vs $83,000). The percentage of patients without correction loss at 10 years after surgery was also higher in the teriparatide group (86.9% vs 76.7%). The 10-year average QALYs of each group were 0.59 in the no teriparatide group and 0.61 in the teriparatide group. The cost-effectiveness of 12-month teriparatide therapy measured in cost/QALY was $213,000/QALY, which was almost twice as high as the commonly used cost-effectiveness threshold ($100,000/QALY). Teriparatide therapy can be considered as cost-effective only if teriparatide therapy less than 6-months is as effective as 12-months of treatment for PJK/PJF prevention, or teriparatide is administrated only for high-risk patients whose 24-month PJK/PJF incidence rate exceeds approximately 40% without teriparatide ($94000/QALY).

Discussion/Conclusion

The results of our Markov model simulation suggest that teriparatide therapy is associated with lower correction loss, but is not a cost-effective option if it is routinely administered in all ASD patients regardless of bone status. Careful selection of high-risk patients for teriparatide therapy is warranted as a more cost-effective measure.

Paper 03. Differences in the Rate of Surgery for Lumbar Disc Herniation Between the Military and Civilian Healthcare Systems

Anderson Ashley, MD1, Pisano Alfred, MD2, Watson Nora, PhD1, Brooks Daniel, PhD1, Helgeson Melvin, MD3, Patel Tushar, MD4, Wagner Scott, MD2

1Walter Reed National Military Medical Center/Uniformed Services University of the Health Sciences, Bethesda, Maryland, United States, 2Walter Reed National Military Medical Center, Bethesda, Maryland, United States, 3Walter Reed Army Medical Center, Bethesda, Maryland, United States, 4Washington Orthopedics and Sports Medicine, Washington, District of Columbia, United States

Background/Introduction

In the military health system (MHS), surgery for lumbar disc herniation (LDH) can be performed at a military treatment facility (MTF) or deferred to the civilian economy (“purchased care”). It is unknown whether patients diagnosed with LDH and managed in the civilian sector are more likely to undergo surgery. Therefore, the purpose of this study was to determine whether surgery is more likely for patients deferred to purchased care than patients treated within the MHS for symptomatic LDH.

Materials/Methods

The Military Data Repository (MDR) was queried for all patients diagnosed with LDH from FY2011-2019, either within the MHS or in purchased care. Within this population, we identified all patients undergoing lumbar microdiscectomy or decompression. We then compared the rate of surgery when diagnosed within the MHS to the rate when diagnosed outside in the civilian sector. A Chi-squared test was utilized to compare categorical variables.

Results

We identified 195,612 patients who were diagnosed with LDH during the study period. Of these, 25,630 (13.1%) underwent a surgical procedure at either an MTF or at a civilian facility, while the vast majority (86.9%) did not require surgical intervention. We found that the rate of surgical intervention for patients diagnosed with LDH outside the military was statistically significantly higher than for patients diagnosed within the MHS (13.9% versus 11.0%, respectively, p<0.0001). Furthermore, roughly half (n=2,598, 44%) of the patients diagnosed within the MHS who required surgery were treated at a civilian facility.

Discussion/Conclusion

We found that patients diagnosed with lumbar disc herniation were statistically significantly more likely to receive surgery if managed outside of the military health system by a civilian facility. In addition, almost half of patients diagnosed with LDH within the MHS underwent surgery outside the system. The reasons for civilian deferral outside of the MHS are unclear, but further research is mandatory to assess the indications for, and cost impact, of this differential.

Paper 04. Long-Term Outcomes Following Intraosseous Basivertebral Nerve Ablation for the Treatment of Chronic Low Back Pain: 5 Year Treatment Results from a Prospective Randomized Double-Blind Sham-controlled Multi-center Study

Fischgrund Jeffrey, MD1, Rhyne Alfred, MD2, Macadaeg Kevin, MD3, Moore Greg, MD4, Kamrava Evish, MD5, Chris Yeung, MD6, Truumees Eeric, MD7, Schaufele Michael, MD8, Yuan Philip, MD9, DePalma Michael, MD10

1William Beaumont Hospital, Royal Oak, Michigan, United States, 2OrthoCarolina Spine Center, Charlotte, North Carolina, United States, 3Indiana University School of Medicine, Indianapolis, Indiana, United States, 4Pacific Sports and Spine, Eugene, Oregon, United States, 5The Spine Institute, Santa Monica, California, United States, 6Desert Institute for Spine Care, Phoenix, Arizona, United States, 7Ascension Texas Spine and Scoliosis Center, Austin, Texas, United States, 8Pain Solutions treatment Centers, Marietta, Georgia, United States, 9Memorial Orthopaedic Surgical Group, Long Beach, California, United States, 10Virginia iSpine Physicians, Richmond, Virginia, United States

Background/Introduction

Current literature suggests that degenerated or damaged vertebral endplates are a significant cause of vertebrogenic chronic low back pain (CLBP). Modic changes are an objective MRI biomarker for this specific subset of patients. Prior 2-year data from the treatment arm of a sham-controlled, randomized controlled trial (RCT) showed maintenance of clinical improvements at 2 years following ablation of the basivertebral nerve (BVN). This study reports the 5-year clinical outcomes in this cohort of patients.

Materials/Methods

A total of 117 US patients were treated successfully with BVN ablation in a previously reported pivotal RCT. Patient-reported outcomes of Oswestry Disability Index (ODI), visual analogue score (VAS), post-ablation treatments, and patient satisfaction were collected by an independent research nurse in a single study visit performed a minimum of five years post BVN ablation. Primary outcome was mean change in ODI from baseline. Comparisons between the post-ablation and baseline values were performed using a paired t-test at the 0.05 level of significance.

Results

100 of the 117 US treated patients (retention rate 85%) were available for review with a mean follow up of 6.4 years (range 5.4 to 7.8 years). Mean improvement in ODI was significant compared to baseline with a 25.95 point reduction from a baseline of 42.81 (p<0.001). Mean reduction in VAS pain score was 4.38 points on a 10-point scale (baseline of 6.74, p<0.001). 66% of patients reported a > 50% reduction in pain, 47% reported a >75% reduction in pain, and 34% of patients reported complete pain resolution. Responder rates were maintained at five years with a combined responder rate of 75% using thresholds of ≥ 15-point ODI and ≥ 2-point VAS for function and pain.

Discussion/Conclusion

CLBP patients treated with BVN ablation exhibit sustained clinical improvements in function and pain with high responder rates at a mean of 6.4 years following treatment. BVN ablation is a durable, minimally invasive treatment for the relief of vertebrogenic CLBP.

Paper 05. A Radiographic Analysis of Optimal Pelvic Fixation Trajectories

Scoville Jonathan, MD1, Mazur Marcus, MD1

1University of Utah, Salt Lake City, Utah, United States

Background/Introduction

Pelvic fixation is used to augment biomechanical support for fusion constructs. Three methods are well defined: the traditional iliac screw, the recessed iliac screw, and the S2-alar-iliac (S2AI) screw. Screw trajectories that use the anatomical landmarks of either the anterior inferior iliac spine (AIIS) or the supra-acetabular notch (SAN) trajectory have been described. These trajectories maximize the length and width of a screw placed in the ilium. This study compares the maximum screw sizes successfully placed using the three pelvic fixation techniques.

Materials/Methods

High-resolution CTs were obtained from 10 randomly selected patients. Using 3D reconstructed models of the lumbar spine and pelvis, 6 trajectories were plotted on each side, starting at the three recognized entry points (traditional iliac screw, recessed iliac screw or S2AI screw) and ending at the AAIS or SAN. The maximum possible screw length and width were measured. We compared the likelihood of successful screw placement for each trajectory. If a trajectory violated the bony architecture it was counted as a failure.

Results

The PSIS to the AIIS trajectory had the longest average screw length at 136.1 ± 6.6 mm, the PSIS to the SAN trajectory had the widest average screw at 12.3± 3.5mm. The traditional iliac screw trajectories had the highest success rates with 95% and 100% of screws reaching the AAIS and SAN, respectively, without breach. The failure rate was 55% for recessed iliac screws targeting either the AIIS or SAN, 75% for S2AI screws targeting the SAN, and 85% for S2AI screws targeting the AIIS. The differences in success rates for traditional iliac screws compared to the others were statistically significant using the Pearson Chi-square test (p<0.001).

Discussion/Conclusion

These findings suggest that a traditional iliac screw measuring up to 135 x 12mm can be reliably placed using a trajectory starting at the PSIS and ending at either the AIIS or SAN. This traditional iliac screw trajectory can be assisted by using a teardrop view on fluoroscopy or image-guidance. High resolution CTs depicting the entire pelvis should be considered for planning purposes if a large screw using one of the other methods is desired.

Paper 06. Right-sided Oblique Pre-psoas Access to the Lumbar Spine- Is it Safe?

Berry Chirag, MD1, Thawrani Dinesh, MD1

1Cincinnati VA Medical Center, Cincinnati, Ohio, United States

Background/Introduction

Oblique pre-psoas retroperitoneal approach to the lumbar spine for interbody fusion (OLIF) provides safe access to nearly all lumbar levels & in some instances to L5-S1 as well. A wide interval between psoas and aorta makes the left side a favored approach to the L2-5 discs. However, in some situations, a right-sided approach may be considered advantageous. Concavity of a scoliotic curve to the right, prior abdominal, retroperitoneal or peri-aortic surgeries from the left, and anomalous venous anatomy may be some such instances.

Materials/Methods

Seventy-five consecutive patients underwent lumbar interbody fusion through an oblique pre-psoas approach at a single institution. A retrospective chart review was performed. Of these, 63 were approached from the left (group L), and twelve from the right (group R). Operative (OR) time, estimated blood loss (EBL), hospital length of stay (HLOS), and early complications like ileus, groin pain, vascular, neurological and other approach-related complications were assessed and compared between the two groups. Indications included lumbar degenerative stenosis, deformity, and spondylolisthesis. All patients underwent posterior percutaneous pedicle screw instrumentation in the same operative setting or the next day.

Results

Demographics and operative indications were similar in the two groups. L5-S1 level was included in 12 of the 75 patients. OR time was significantly greater in group R (Mean Group L-257 min; Group R-335 min; p=0.0006). Differences did not reach statistical significance in EBL (Mean Group L-207.6ml; Group R-264.2ml; p=0.067), and HLOS (Mean Group L-2.62days; Group R-2.33days; p=0.263). Incidence of postoperative ileus, and groin pain on the side of approach were also not significantly different between the two groups. Ligation of ilio-lumbar or segmental veins for adequate exposure was required in 7 in group L (11%), and 5 in group R (42%). No vascular injuries occurred. Peri-operative blood transfusion was required in one patient in group L. Neurologic injury was noted in two in group L.

Discussion/Conclusion

Right-sided oblique pre-psoas approach is safe, but likely involves greater complexity and steeper learning curve. In our early experience with this technique, we did not find any increase in early complications when compared to left-sided approach.

Paper 07. A Novel Cost-Effective Training and Assessment Simulator for Spine Surgery

Sayari Arash, MD1, Louie Philip, MD2, Basques Bryce, MD2, Harada Garrett, MD2, Kunze Kyle, BS2, An Howard, MD2, Lopez Gregory, MD2

1Rush University Medical Center, IL, Illinois, United States, 2Midwest Orthopaedics at Rush, Chicago, Illinois, United States

Background/Introduction

Increasing complexity in spine surgery techniques has created a niche for improved surgical training. Currently, spine surgery simulation current includes synthetic bone activities, cadaver laboratories, and virtual reality simulators that are costly and often unaffordable for residency programs. The purpose of this study is to evaluate the construct validity of a novel training spine surgery simulator that is cost-effective, feasible, easily reproducible, and applicable in a clinical setting.

Materials/Methods

A spine surgery simulator was created using supplies from a local hardware store to evaluate trainees with varying levels of spine surgery experience. Tasks included: (1) the use of a hand-held high-speed burr in minimally invasive surgery (MIS) through a PVC pipe and in open fashion, (2) the use of a kerrison rongeur in MIS and open fashion, (3) gear shift awl triangulation, and (4) pituitary rongeur peg transfer through a tube. Subjects were divided into two groups based on level of training: novice (medical students and orthopaedic surgery residents) and experts (spine surgery fellows and spine surgery attendings). Time, accuracy, and correct technique were compared using chi-squared analysis and independent sample t-tests for categorical and continuous data, respectively.

Results

In total, 66 subjects were included, 56 were classified as novice and 10 subjects were classified as experts. The expert cohort was faster in completing MIS (p=0.009) and open (p=0.041) tasks using the kerrison rongeur, as well as the lateral pituitary rongeur peg transfers task (p=0.035) (Table 1). The expert cohort was more likely to use an appropriate technique with the open burr task (p=0.005), and were more accurate with gearshift awl triangulation (p=0.008, p=0.017). The novice cohort found the simulator more helpful in preparation for spine surgery (p=0.048) and hoped to have it included as part of their training (p=0.039).

Discussion/Conclusion

This study demonstrates the construct validity of a cost-effective novel surgical simulator for training and assessment of the fundamental techniques of spine surgery. The results of this study encourage the application of this simulator as a training tool for resident education and will serve as a foundation for future studies on surgical stimulation training.

Paper 08. The Effectiveness of a Free-Standing Lead-Shield in Reducing Spine Surgeon Radiation Exposure during Intraoperative 3-Dimensional Imaging

Guenthner Guy, BS1, Eddy Alexander, BS1, Sembrano Jonathan, MD1, Polly, Jr David, MD1, Martin Christopher, MD1

1University of Minnesota, Department of Orthopaedic Surgery, Minneapolis, Minnesota, United States

Background/Introduction

Intra-operative three-dimensional (3D) computed tomography (CT) imaging has become increasingly popular in spine surgery. The Minnesota Department of Health recently issued guidelines banning the use of lead shields as safety equipment during surgeries utilizing this imaging. However, there is no data to demonstrate the lead shield alone is sufficient, or if additional personal safety equipment is indeed required. Thus, the purpose of the present study is to measure radiation exposure behind OR lead shielding during 3D image-guided spinal surgery and determine if exposures fall below annual dose limits.

Materials/Methods

Twenty-seven spinal procedures using the O-Arm Imaging System were observed to determine radiation exposure to a spine surgeon standing behind a lead shield in the OR. Two dosimeters were used to measure scatter radiation in front of and behind lead shields. Dosimeter readings were compared to determine radiation attenuation by the lead shield. Regression analysis was performed to estimate the number of cases required to surpass annual exposure limits. Case numbers were calculated for the highest “worst case” and “average case” exposure values. The annual occupation exposure limit determined by the National Council on Radiation Protection (NCRP) is 50,000 microsieverts (μSv).

Results

Average surgeon radiation exposure per case was 0.694μSv behind the lead shield compared to 14.577μSv in front of the lead shield. The average radiation dose reduction by the lead shield was 13.962μSv per case, which is equivalent to an average of 95.65% radiation attenuation by lead shielding. If surgeons stand behind lead shields in the OR, the annual number of 3D image-guided spinal procedures required to surpass exposure limits is 15,479 and 67,060 based on “worst case” and “average case” analyses, respectively.

Discussion/Conclusion

Our study demonstrates intraoperative lead shields are very effective at decreasing radiation exposure to surgeons standing behind them. Additionally, surgeon radiation doses behind lead shielding fall far below annual exposure limits. Surgeons should not need additional protective equipment when a lead shield is used during cases with intra-operative 3D CT.

Paper 09. The Role of Neuromonitoring in Anterior Lumbar Interbody Fusions

shtewe Anan, MD1, Kesavabhotla Kartik, MD1, Dahdaleh Nader, MD1, Koski Tyler, MD1

1Northwestern University Feinberg School of Medicine, Northwestern, Chicago, Illinois, United States

Background/Introduction

The anterior lumbar interbody fusion is a common procedure performed for a number of conditions, particularly involving pathology at the L4-5 and L5-S1 levels of the spine. What is unclear is whether or not an anterior approach to the spine necessitates the use of intraoperative neuromonitoring.

Materials/Methods

Retrospective analysis of prospectively collected data at a single institution since 2010-2019, of all patients undergoing surgery for anterior lumbar interbody fusion (ALIF) of either one or two levels. The primary outcome was neurological injury or any action required by the senior surgeon as a response to neuromonitoring changes. Our secondary aim has been to assess the cost associated with ALIF neuromonitoring at a single institution.

Results

120 consecutive patients (62 males and 58 females) with average age of 58.8 years. Of these cases, 39 patients underwent a staged posterior lumbar fusion for deformity correction. Operative indications included symptomatic spondylolisthesis, radiculopathy, and neurogenic claudication. No neuromonitoring alerts or changes were noted intraoperatively within this cohort. No patients had SSEP/MEP changes that required action by the senior surgeon. Neuromonitoring is billed hourly and associated with costs ranging between $1000 and $2000 per case, with average ALIF duration between 2-4 hours.

Discussion/Conclusion

Neuromonitoring is a very important and vital tool when preforming certain spine surgeries like deformity correction, lateral approach to the lumbar spine and many more. Our data over the last 10 years for single surgeon includes 120 patients that had one or two-level ALIF, shows that monitoring doesn’t have any additive value to the procedure and doesn’t affect the normal course of the instrumentation. This is important since it may significantly reduce the time and the cost of such surgery.

Paper 11. The Association of Preoperative NarxCare Scores and Postoperative Opioid Use and Pain Scores in Lumbar Fusion Patients

Winkelman Robert, MS, BS1, Hoffmann Jacob, MD1, Kavanagh Michael, BS1, Steinmetz Michael, MD1, Mroz Thomas, MD2

1Center for Spine Health, Cleveland Clinic Foundation, Cleveland, Ohio, United States, 2Cleveland Clinic Foundation, Cleveland, Ohio, United States

Background/Introduction

Preoperative opioid consumption is associated with increased postoperative opioid requirements and adverse perioperative outcomes following elective surgery. Since preoperative opioid prescriptions often arise from non-surgical providers, quantifying patients’ preoperative opioid use presents a challenge to surgeons. The NarxCare score, a proprietary algorithm which analyzes prescription drug monitoring program data, can provide a single numerical summary of a patient’s recent history of opioid-use. The goal of this study is to assess the association of preoperative NarxCare scores with patient pain scores, postoperative opioid use, and length of stay following lumbar spine fusion surgery.

Materials/Methods

This study is a retrospective review of patients undergoing lumbar fusion surgery for degenerative disease at a single institution between 2/1/2019 and 8/31/2019. The institutional electronic medical record database was queried to obtain preoperative NarxCare scores, perioperative pain scores, and outpatient opioid medications prescribed in the 90 days following discharge from surgery. Patients with preoperative NarxCare scores greater than or equal to 200 were classified as having an elevated NarxCare scores. Multivariable linear regression models were developed to assess the association of preoperative Narx scores with perioperative visual analog scale pain scores, dosing of post-discharge opioid medication (in morphine equivalent doses [MEDs]), and length of stay (LOS).

Results

Overall, 260 patients met the inclusion criteria for the study and 114 (44%) were found to have elevated preoperative NarxCare scores. Patients with elevated preoperative NarxCare scores were found to have significantly higher pre- (+1.3, 95% CI: 0.7 to 1.8) and post-operative pain scores (+1.2, 95% CI: 0.6 to 1.9). Additionally, patients with elevated preoperative NarxCare scores required 57% more MEDs (95% CI: 30 to 91%) in the 90 days following discharge from surgery. Elevated preoperative NarxCare scores were not significantly associated with LOS.

Discussion/Conclusion

Elevated preoperative NarxCare scores were associated with higher perioperative pain scores and larger postoperative opioid requirements in the 90 days following discharge from surgery in patients undergoing lumbar fusion surgery. Elevated preoperative NarxCare scores could be used to identify patients at high-risk for poor perioperative pain control as well as prompt early referrals to pain management.

Paper 12. Does Immediate Postoperative Opioid Consumption Correlate with Intermediate and Long-term Outcomes in Patients Undergoing One- and Two-level Instrumented Posterior Lumbar Fusion?

Gannon Emmett, MD1, Larson Evan, MD1, Powers Sydney, BS1, Vincent Scott, MD1, Cornett Chris, MD1

1University of Nebraska Medical Center, Omaha, Nebraska, United States

Background/Introduction

Opioids are routinely used for management of postoperative pain following spine surgery. The purpose of this study is to investigate the potential relationship between increased opioid use immediately postoperatively and a patient’s outcomes. Most of the current literature has studied the effects of preoperative and chronic postoperative opioid consumption, however, there are limited studies investigating the relationships between immediate postoperative opioid use, VAS scores and LOS with patient outcomes.

Materials/Methods

A retrospective review of patients undergoing instrumented 1 or 2-level posterior lumbar fusion surgery by a single surgeon was performed. Descriptive statistics were used to summarize demographic and clinical characteristics for all patients. Pearson correlation coefficients were used to evaluate associations between patient outcomes as they relate to postoperative opioid use, visual analog scale (VAS) scores, and length of stay (LOS). Exclusion criteria included previous lumbar surgery, indications other than degenerative, and patients <19 years old. A total of 152 patients met inclusion criteria and were analyzed.

Results

Among 94 patients undergoing 1-level fusion, there was a significant inverse relationship between 6 Month SF-36 Bodily Pain and postoperative day (POD) 1, 2 and total opioid use (p=0.0328, p=0.0283, and p=0.0205). A significant inverse relationship was also found between 1 Year SF-36 Physical Function and POD 1 and 2 opioid use (p=0.0415, p=0.0200) and between 1 Year SF-36 Bodily Pain and POD 2 opioid use (p=0.0099). A significant inverse relationship was also found between POD 0 VAS scores and 6 Month SF-36 Physical Function, Bodily Pain, and Mental Health (p=0.0149, p=0.0006 and p=0.0171). In addition, there was a significant inverse correlation between LOS and 6 Month SF-36 Physical Function (p=0.0059). Among the 58 patients undergoing 2-level fusion, POD 0 and 1 VAS scores had a significant inverse correlation with 6 Month and 1 Year SF-36 Mental Health (p=0.0161, p=0.0021, p=0.0358, p=0.0103).

Discussion/Conclusion

Based upon these results, immediate postoperative opioid use, VAS scores, and LOS may be helpful in predicting patient outcomes at 6 months and 1 year following 1 and 2-level posterior lumbar fusion. These findings, however, must be interpreted in context with the limitations inherent to a retrospective study.

Paper 13. Postoperative Glucose Variability as a Predictor of Adverse Outcomes Following Lumbar Fusion

Patel Parthik, MD1, Canseco Jose, MD, PhD1, Chang Michael, BS1, Kurd Mark, MD1, Rihn Jeff, MD1, Anderson D., MD2, Hilibrand Alan, MD1, Kepler Christopher, MD, MBA1, Vaccaro Alexander, MD, PhD, MBA1, Schroeder Gregory, MD1

1Rothman Institute at Thomas Jefferson University, Philadelphia, Pennsylvania, United States, 2Department of Orthopaedic Surgery, Rothman Institute, Thomas Jefferson University, Philadelphia, Pennsylvania, United States

Background/Introduction

The incidence of lumbar fusion surgeries is increasing annually with documented improvements in fusion rates and health-related quality of life outcomes. While numerous risk factors have been associated with poor outcomes in lumbar fusions, recent studies have established a relationship between postoperative glycemic variability as a predictive marker for postoperative outcomes. Our study aimed to evaluate the predictive ability of postoperative glycemic variability in patients who have undergone one- to three-level lumbar fusions.

Materials/Methods

Upon Institutional Review Board approval, a retrospective cohort study was conducted with 850 total patients included in the final analysis. Patients were stratified into one of three tertiles based on degree of postoperative glycemic variability defined by coefficient of variation (CV). Surgical site infections were determined based on Centers for Disease Control and Prevention (CDC) definition and chart review. Demographic factors, surgical characteristics, inpatient complications, readmissions, and revisions were determined by chart review, telephone encounters, and available navigated data.

Results

Overall, our data showed a statistically significant difference in 90-day adverse outcomes when stratified by postoperative glycemic variability. In particular, patients with a high glycemic variability (third tertile) had higher odds of experiencing a readmission (OR: 4.01 [1.57, 10.23]; p: 0.004), surgical site infection (OR: 9.49 [2.73, 33.03]; p: <0.001), and undergoing revision surgery (OR: 3.91 [1.57, 4.85]; p: 0.005) compared to patients with a low glycemic variability (first tertile). No significant association was seen between the low and moderate glycemic variability groups. Patients with higher glycemic variability were more likely to experience longer hospital stay (β: 1.07; p: 0.001) and experience more complications (p < 0.001).

Discussion/Conclusion

Our study establishes a significant relationship between postoperative glycemic variability and inpatient complications, length of stay, and 90-day adverse outcomes. While hemoglobin A1c has classically been used as the principal marker to assess blood glucose control, our results show glycemic variability (CV) to be a strong predictor of postoperative adverse outcomes. Future high-quality, prospective studies are necessary to explore the true effect of glycemic variability, as well as its practicality in clinical practice. Nevertheless, fluctuations in blood glucose levels during the inpatient stay should be limited to improve patient results.

Paper 14. The Risks and Benefits of Bone Morphogenetic Protein Use in Single-level Lumbar Fusion Surgeries

Khalid Syed, MD1, Parola Rown, MS2, Edmondson Corbin, BS3, Behbahani Mandana, MD2, Mehta Ankit, MD2

1Rush University Medical Center, IL, Illinois, United States, 2University of Illinois at Chicago, Chicago, Illinois, United States, 3Rosalind Franklin University of Medicine and Science, chicago, Illinois, United States

Background/Introduction

The effectiveness and complication profile of bone morphogenetic protein (BMP) has not been studied in a large, propensity matched population despite common on and off label use in single-level lumbar fusion (LF). This study intends to assess the effectiveness and complication profile of BMP in single-level LF.

Materials/Methods

Study Design/Setting: Multicenter retrospective cohort study Patient Sample: 362,530 randomized and matched patients selected from 687,775 Medicare beneficiaries undergoing single-level lumbar fusion with and without the use of bone morphogenetic protein between 2004 and 2014. Outcome Measures: Fusion rates, demonstrated by pseudarthrosis and pseudarthrosis requiring surgical revision. Complications including hematoma, surgical site infections, durotomy, wound dehiscence, perioperative ileus, respiratory, urinary, cardiac, neurologic, thromboembolic, and implant complications. All cause readmission within 30 days and length of stay. Methods: Patient populations were matched based upon age, Elixhauser Comorbidity Index, discharge status, gender, region, the year of treatment, smoking status, osteoporotic status, and nutritional status.

Results

Bone morphogenetic protein use imparted significantly greater odds of pseudarthrosis (OR=1.59, 95% CI 1.53 to 1.66, p<0.001), pseudarthrosis requiring surgical revision (OR=2.95, 95% CI 2.71 to 3.321, p<0.001), implant complication (OR=1.61, 95% CI 1.57 to 1.64, p<0.001), and all cause readmission within 30 days (OR=1.18, 95% CI 1.15 to 1.22, p<0.001). Single-level LF with and without BMP impart similar rates of hematoma (8.4% vs. 8.0, p=0.18), surgical site infection (8.4% vs. 8.4%, p=0.84), durotomy (3.9% vs. 3.9%, p=0.72), wound dehiscence (3.5% vs. 3.9%, p=0.55), respiratory (22.0% vs. 22.2%, p=0.61), urinary (2.5% vs. 2.2%, p=0.62), cardiac (2.6% vs. 2.6%, p=0.99), neurologic (1.8% vs. 2.0%, p=0.19), and thromboembolic complications (13.4% vs. 13.2%, p=0.64). LF with BMP imparted significantly higher rates of perioperative ileus (8.3% vs. 7.5%, p=0.004).

Discussion/Conclusion

Results are consistent with a prior JAMA paper in 2008 on Medicare data on cervical spine fusion, but contrary to other Medicare analyses of lumbar fusion, published systematic reviews, meta-analyses, and industry sponsored RCTs for BMP FDA-approval studies and clinical case series. BMP may have a greater risk profile than previously reported and in fact result in more revision interventions among those patients that experience pseudarthroses.

Paper 15. A Comparative Analysis of 30-day Readmission, Reoperation, and Morbidity Between Lumbar Disc Arthroplasty Performed in the Inpatient and Outpatient Settings Utilizing the ACS-NSQIP Dataset

Katz Austen, MD1, Perfetti Dean, MD, MPH1, Job Alan, MD1, Willinger Max, MD1, Goldstein Jeffrey, MD1, Kiridly Daniel, MD1, Olivares Peter, BS1, Essig David, MD1

1North Shore LIJ Health System, Manhasset, New York, United States

Background/Introduction

Spine surgery has been increasingly performed in the outpatient setting, providing greater control over cost, efficiency, and resource utilization. However, short-term outcomes research evaluating the safety of this trend is limited. This is the first large-scale database study to compare short-term outcomes between inpatient and outpatient lumbar total disc arthroplasty (TDA). The objective of this study is to determine the differences in 30-day readmission, reoperation, and morbidity for patients undergoing lumbar TDA in the inpatient vs outpatient setting.

Materials/Methods

Patients who underwent lumbar TDA were identified using the NSQIP database. Regression was utilized to compare readmission, reoperation, morbidity, and specific complications between surgical settings, and to evaluate for predictors of primary outcomes.

Results

We identified 646 patients. Outpatient lumbar TDA independently predicted greater reoperation (p=0.031, OR=6.803), but not readmission or morbidity, on multivariate analysis. There were no significant differences in rates of specific complications. Inpatient operative time (141±79 min) was significantly (p<0.001) longer than outpatient operative time (106±44 min). Diabetes (p=0.001) and decreased platelet count (p=0.043) were associated with readmission in univariate analysis, but only diabetes predicted readmission in multivariate analysis (p=0.001, OR=6.481). Diabetes (p=0.013, OR=7.814) also predicted reoperation in multivariate analysis. Non-white race (p=0.032), baseline dyspnea (p=0.046), ASA-class ≥3 (p=0.001), decreased hematocrit (p=0.024), decreased platelets (p=0.049), increased operative time (p=0.003), and increased length of stay (p<0.001) were associated with morbidity in univariate analysis. ASA-class ≥3 (p=0.046, OR=2.988) and length of stay (p=0.007, OR=1.179) precited morbidity in multivariate analysis.

Discussion/Conclusion

Outpatient lumbar TDA was associated with an almost 7-times increased odds of reoperation after adjusting for comorbidities and factors associated with primary outcomes. Readmission and morbidity were similar between surgical settings. Higher ASA-class and specific comorbidities predicted poorer 30-day outcomes. These findings suggest that there may be a deviation when performing the procedure as an outpatient that negatively impacts reoperation rate. This may be related to intrinsic or extrinsic factors associated with outpatient surgery centers (i.e. quality control; equipment availability, access to laboratory draws, and device access; and support of medical personnel, device representatives, and specialty physicians). In sum, these findings can guide surgical setting given specific patient factors.

Paper 16. The Association of Comorbidity Burden on Lumbar Decompression Postoperative Outcomes

Jenkins Nathaniel, MS1, Parrish James, MPH1, Hrynewycz Nadia, BS1, Brundage Thomas, BS1, Alqam Musa1, Qadri Talha, BS1, Singh Kern, MD1

1Rush University Medical Center, IL, Illinois, United States

Background/Introduction

Minimal research has looked at the influence of comorbidity burden on postoperative patient-reported outcome (PRO) improvement and on the achievement of minimum clinically important difference (MCID) following lumbar decompression (LD). This study aims to detail the association between comorbidity burden and achievement of MCID following LD.

Materials/Methods

A prospectively maintained surgical registry was retrospectively reviewed for eligible spine surgeries between May 2015-July 2019. Inclusion criteria were primary, single or multilevel lumbar decompressions. Patients were excluded if they did not complete a preoperative Patient-Reported Outcomes Measurement Information System physical function (PROMIS-PF) survey. Stratification was based on Charlson Comorbidity Index (CCI) score: 0 points (no comorbidities), 1–2 points (low CCI), ≥3 points (high CCI). A Chi-squared or t-test was used to analyze baseline characteristics, perioperative outcomes, and scores for VAS-back, VAS-leg, ODI, SF-12 Physical Composite Score (PCS), and PROMIS-PF pre- and postoperatively (6-weeks, 12-weeks, 6-month, and 1-year). MCID achievement was compared between CCI groups using chi-squared analysis. MCIDs utilized were VAS-back=1.2, VAS-leg=1.6, ODI=12.8, SF-12-PCS=4.0, and PROMIS-PF=8.0.

Results

Of the 314 LD patients, 123 had no comorbidities, 100 had low CCI, and 91 had high CCI. Perioperative differences between CCI groups included increased operative time, hospital length of stay, and day of discharge in the CCI ≥3 subgroup. There were no significant differences between any pre- and postoperative PRO scores. CCI subgroups demonstrated no difference in the rate of achieving MCID for VAS-back, VAS-leg, and ODI. Patients with increased comorbidity had a lower rate of reaching MCID at 6 months for SF-12 PCS (p=0.008), at 6-weeks for PROMIS-PF (p=0.014), and the overall rate for both SF-12-PCS and PROMIS-PF (p=0.019 and p=0.000).

Discussion/Conclusion

Patients with increased comorbidity burdens undergoing LD had significantly longer procedure times and hospital stays, but their postoperative evaluations demonstrated no difference in achieving MCID for pain (VAS-back/VAS-leg) or disability (ODI) metrics up to 1-year. Patients in the high CCI group did, however, have a lower rate of achieving MCID for their physical function surveys. Our results suggest that comorbidity burden influence improvement in physical function following LD.

Paper 17. The Impact of Preoperative Depression on Clinician and Group Assessment of Healthcare Providers and Systems (CG-CAHPS) Survey Results in the Spine Surgery Setting

Khan Hammad, BS1, Rabah Nicholas, BS1, Levin Jay, MD1, Winkelman Robert, MS, BS1, Mroz Thomas, MD2, Steinmetz Michael, MD1

1Center for Spine Health, Cleveland Clinic Foundation, Cleveland, Ohio, United States, 2Cleveland Clinic Foundation, Cleveland, Ohio, United States

Background/Introduction

The Clinician and Group Assessment of Healthcare Providers and Systems (CG-CAHPS) survey is used to measure patient experience in the outpatient setting. CG-CAHPS scores are made publicly available and may be used by prospective patients when selecting spine surgeons or by hospital systems in physician incentive programs. While preoperative depression has been shown to correlate with worse patient-reported outcomes following spine surgery and lower patient satisfaction in the inpatient setting, no studies have examined its impact on patient experience with their spine surgeons in the outpatient setting. In the present study, we aim to examine the association between preoperative depression and CG-CAHPS scores in patients undergoing elective spine surgery.

Materials/Methods

Patients who underwent elective spine surgery between 2009-2017 and completed a CG-CAHPS survey at our institution were included in this study. Data was collected on patient demographics, Patient Health Questionnaire (PHQ-9) scores, and Patient-Reported Outcome Measurement Information System Physical Functioning (PROMIS-PF) scores. Patients with preoperative PHQ-9 scores of greater than or equal to 10 (moderate to severe depression) were included in the depressed cohort. Survey responses on several dimensions of the CG-CAHPS survey were compared between depressed and non-depressed patients using chi-square tests. High patient satisfaction was defined as a “top-box” survey response.

Results

Our cohort included 419 patients, of which 72 met criteria for preoperative depression. On average, depressed patients were younger (p<0.001) and had lower self-reported physical functioning (p<0.001) compared to non-depressed patients. 73.9% of depressed patients provided a “top-box” Overall Provider Rating compared to 90% of non-depressed patients (p<0.001). Further, depressed patients were more likely to feel that their spine surgeon did not listen to them carefully (p=0.003) and less likely to recommend their office to other patients (p=0.018).

Discussion/Conclusion

This study suggests that preoperative depression may be associated with lower patient satisfaction with spine surgeons in the outpatient setting. These results add to a growing body of literature highlighting depression as a modifiable risk factor for poor patient experience. Future studies should evaluate the impact of interventions for preoperative depression on patient satisfaction in the spine surgery setting.

Paper 18. Bariatric Surgery Imparts Increased Infection Risk Following Elective Instrumented Lumbar Fusions

Couch Brandon, MD1, Fourman Mitchell, MD, M. Phil1, Shaw Jeremy, MD, MS1, Wawrose Richard, MD1, Talentino Spencer, BS2, Boakye Lorraine, MD1, Donaldson William, MD1, Lee Joon, MD1

1University of Pittsburgh Medical Center, Pittsburgh, Pennsylvania, United States, 2University of Pittsburgh School of Medicine, Pittsburgh, Pennsylvania, United States

Background/Introduction

Recent literature has lauded the benefits of bariatric surgery prior to lumbar fusion, however, most of this data is derived from large database studies. Bariatric procedures while beneficial for certain medical ailments may paradoxically create conditions that are adverse for elective lumbar surgery due to malnutrition from gut resection, vitamin deficiencies and osteomalacia from mechanical offloading. The goal of this study was to evaluate the impact of prior bariatric surgery on elective instrumented lumbar fusion outcomes. It was hypothesized that patients who had undergone bariatric surgery would have an increased incidence of post-operative complication.

Materials/Methods

A retrospective cohort of prospectively collected patients was evaluated. Study patients underwent a bariatric procedure prior to elective instrumented lumbar fusion at a large academic medical center from 2012 to 2018 and were compared to comorbidity matched controls. The primary outcome was surgical site infection (SSI) requiring surgical debridement. Secondary outcomes were prolonged wound drainage that necessitated post-discharge antibiotics, hardware failure requiring revision, revision secondary to adjacent segment disease (ASD), and need for chronic narcotics or pain clinic. Statistical analysis was performed using Student’s two-way t-tests for continuous data. Data is stated as mean ± standard deviation, and significance was defined as p < .05.

Results

Twenty-five patients had bariatric surgery before their elective lumbar fusion. A randomly selected, comorbidity-matched group of 59 patients that underwent an elective lumbar fusion by the same surgeons during that span was used as a non-bariatric comparison group. Patients with a history of bariatric surgery had an increased incidence of SSI that required operative debridement, 90-day readamission, revision fusion due to ASD, and need for chronic narcotics or pain clinic (Table 1). Prolonged wound drainage and hardware failure were equivalent between groups.

Discussion/Conclusion

Bariatric surgery prior to elective instrumented lumbar fusion predisposed patients to a significantly higher risk of SSI, 90-day readmission, ASD, and chronic pain. Limitations include the selection bias inherent to retrospective analyses and the low sample size evaluated. Further prospective analysis of perioperative nutritional status on lumbar fusion outcomes is currently underway.

Paper 19. Predictive Algorithms Show Readmission is Associated with Body Mass Index in Patients Receiving Lumbar Spinal Fusion

Shahrestani Shane, MS1, Chen Xiao, BS1, Ton Andy, BS1, Ballatori Alexander, BS1, Wang Jeffrey, MD2, Buser Zorica, PhD1

1University of Southern California Keck School of Medicine, Los Angeles, California, United States, 2Keck Medical Center of USC, Los Angeles, California, United States

Background/Introduction

Body mass index (BMI) is often used as a proxy for physical fitness and is associated with increased surgical complication rates. However, literature documenting the effect of increasing BMI in the context of lumbar spinal fusion and its effect on potential readmission is limited. Here, we develop predictive algorithms to better understand the effect of increasing BMI on the probability of readmission in patients receiving lumbar spinal fusion.

Materials/Methods

Using the 2013-2016 National Readmission Database (NRD), we conducted a retrospective cohort analysis of 53,370 patients who were admitted into the hospital for lumbar spinal fusion and were found to have a BMI between 30 and 59.9. Of these patients, 5,719 readmissions were identified within the same year as the initial lumbar fusion. BMI groups were broken down as follows: BMI 30-34.9 (Group 1), BMI 35-39.9 (Group 2), BMI 40-44.9 (Group 3), BMI 45-49.9 (Group 4), BMI 50-59.9 (Group 5). Statistics were conducted using RStudio, and predictive algorithms were developed using generalized gaussian logistic regression models. Significance was determined by one-way ANOVA, Tukey multiple comparison of means, and Wald testing.

Results

Within the logistic regression model, BMI group was significantly correlated with the likelihood of post-discharge readmission (p<0.001), and an increase in BMI group level was found to increase the probability of readmission by 1.42% (Figure 1). Furthermore, comparison of the readmission rates between the BMI groups finds that Group 1 has a lower readmission rate than Groups 2, 3, 4, and 5 (p<0.001 for all groups), Group 2 has a lower readmission rate than Groups 3, 4, and 5 (p<0.001 for all groups), and Group 3 has a lower readmission rate than Group 5 (p<0.05). No significant difference in readmission rate was found between individuals in BMI Group 3 and Group 4, and Group 4 and Group 5. This finding suggests that post-fusion complications and readmission rates associated with increasing BMI may saturate for individuals with BMIs greater than 45.

Discussion/Conclusion

Patients hospitalized for lumbar spinal fusion procedures may have higher readmission rates as their BMI increases. Predictive algorithms allow for quantification of the dose-response relationship between increasing BMI level and readmission rates.

Paper 20. Lumbar Spine Fusion in Patients with Multiple Sclerosis: Is There a Higher Risk of Complications?

Sulovari Aron, BS1, Kurucan Etka, BS1, Bernstein David, MS, MBA1, Mesfin Addisu, MD1

1University of Rochester Orthopedics, Rochester, New York, United States

Background/Introduction

Multiple Sclerosis (MS) is a common cause of neurological disability in the US. Patient with MS can have deformity and degenerative changes of the lumbar spine, unrelated to the MS diagnosis, that may necessitate spinal fusion. The impact of MS diagnosis on outcomes of lumbar spine fusion has not been studied in a large series. The objective of this study was to evaluate comorbidities, complications and mortality in patients with and without MS undergoing primary lumbar spine fusion.

Materials/Methods

The US Nationwide Inpatient Sample (NIS) from 2003 to 2014 was used to create experimental MS and non-MS control cohorts undergoing primary lumbar spine fusion. Univariate and bivariate analyses were used to determine and compare outcome measures.

Results

842 MS patients undergoing lumbar spine fusion were compared to 165,726 non-MS patients undergoing lumbar spine fusion. MS patients undergoing primary lumbar spine fusion had higher rates of depression (p=.003) and lower rates of diabetes without chronic complications (p=.027). No significant differences were found in mortality or total perioperative complication rates.

Discussion/Conclusion

MS patients undergoing lumbar spine fusion had similar complication and mortality rates as compared to patients without MS. Overall, MS patients did have higher rates of depression and lower rates of diabetes as compared to the control group. The findings of this study are important for spine surgeons considering elective surgery for MS patients since patients with MS are living longer and developing degenerative conditions of the spine. Based on one of the largest studies of MS patients undergoing lumbar spine fusion, spine surgeons should be reassured on the low complication rate in this patient population.

Paper 21. Undergoing Elective Posterior Lumbar Fusions at “US News & World Report” Ranked Hospitals vs. Non-Ranked Hospitals

Malik Azeem, MBBS1, Phillips Frank, MD2, Yu Elizabeth, MD1, Kim Jeffery, MD1, Khan Safdar, MD1

1Ohio State University Wexner Medical Center, Columbus, Ohio, United States, 2Midwest Orthopaedics at Rush, Chicago, Illinois, United States

Background/Introduction

The US News and World Report (USNWR) hospital rankings remain the most publicly referenced system for identifying top hospitals for each specialty. The current study aims to evaluate whether there are any differences in 90-day outcomes and costs for patients undergoing elective 1-to-3 level posterior lumbar fusions at ranked and non-ranked hospitals.

Materials/Methods

The 2010-2014 USNWR hospital rankings were used to identify ranked hospitals in “Neurosurgery” and “Orthopaedics”. The 2010-2014 100% Medicare Standard Analytical Files (SAF100) were used to identify patients undergoing elective 1-to-3 level posterior lumbar fusions. The study cohort was divided into two groups, based on whether patients received surgery at a ranked hospital vs. a non-ranked hospital. Multi-variate logistic and linear regression analyses were used to assess for differences in 90-day outcomes and costs between ranked and non-ranked hospitals.

Results

A total of 149,863 patients undergoing elective 1-to-3 level posterior lumbar fusions were included in the study, out of which 15,722 (10.5%) underwent surgery in one of the 100 ranked hospitals across the study time period. Ranked hospitals were more likely to have a higher annual case volume, greater bed size, be teaching hospitals and be located in a low socio-economic area. On multi-variate analyses, there were no significant differences between ranked vs. non-ranked hospitals with regards to wound complications (4.6% vs. 4.3%; p=0.299), cardiac complications (8.3% vs. 7.8%; p=0.077), thromboembolic complications (3.0% vs. 2.3%; p=0.208), pneumonia (2.0% vs. 2.3%; p=0.482), urinary tract infections (10.6% vs. 9.7%; p=0.109), renal complications (4.2% vs. 4.0%; p=0.255), revision surgery (0.5% vs. 0.4%; p=0.886) and all-cause readmissions (10.0% vs. 10.0%; p=0.843). Patients undergoing surgery at ranked hospitals vs. non-ranked hospitals had a slightly lower odds of experiencing septic complications (1.0% vs. 1.3%; OR 0.78 [95% CI 0.64-0.94]; p=0.010), however these findings were largely not clinically significant. Ranked hospitals, however, had significantly higher risk-adjusted 90-day charges (+ 11,938; p<0.001) and 90-day costs (+ $1,923; p<0.001).

Discussion/Conclusion

Elective posterior lumbar fusions carried out on Medicare recipients at ranked hospitals have similar 90-day complications as compared to non-ranked hospitals, even though ranked hospitals have higher costs of care over the 90-day episode of care.

Paper 22. The necessity and risk factors of subsequent fusion after decompression alone for lumbar spinal stenosis with lumber spondylolisthesis: 5 years follow-up in 2 different large populations.

Urakawa Hikari, MD1, Samuel Andre, MD1, Vaishnav Avani, MBBS1, Katsuura Yoshihiro, MD2, Iyer Sravisht, MD1, McAnany Steven, MD2, Albert Todd, MD2, Gang Catherine Himo, MPH1, Qureshi Sheeraz, MD2

1Hospital for Special Surgery, New York, New York, United States, 2Hospital for Special Surgery/Weill Cornell Medical College, New York, New York, United States

Background/Introduction

Decompression without fusion is a common surgical treatment for patients with lumbar spinal stenosis (LSS) secondary to spondylolisthesis, but occasionally some of these patients will need a subsequent fusion surgery at a later time. The aim of this study is to examine the rate, timing and risk factors of subsequent fusion surgery for patients after decompression alone for LSS with spondylolisthesis.

Materials/Methods

Patients who had LSS with spondylolisthesis and underwent decompression alone at 1 or 2 levels as a primary lumbar surgery were extracted from both public and private insurance resources in a nationwide insurer database. Among them, patients who were followed less than 5 years were excluded. In each insurance population, the rates of patients who needed subsequent fusion surgery were examined at 1 year, 2 years and 5 years. The timing of the subsequent fusion surgery was also investigated. Risk factors for subsequent fusion surgery were evaluated by multivariate logistic regression controlling for age, gender, the presence or absence of claudication and comorbidities.

Results

5875 patients in the public insurance population and 1456 patients in the private insurance population were included in this study. The rates of patients who needed subsequent fusion surgery were 1.9% at 1 year, 3.5% at 2 years and 6.7% at 5 years in the public insurance population, whereas they were 4.3% at 1 year, 8.9% at 2 years, 14.6% at 5 years in the private insurance population. The period to subsequent fusion is 730 (365-1234) days in the public insurance population and 588 (300-998) days in the private insurance population. Age under 70 years old, presence of claudication, and rheumatoid arthritis (RA) /collagen vascular diseases were common risk factors for subsequent fusion at 5 years for both populations. For patients with degenerative spondylolisthesis, RA was also a risk factor for subsequent fusion at 2 years.

Discussion/Conclusion

The rate of subsequent fusion after decompression alone for lumbar spinal stenosis with lumbar spondylolisthesis was under 15% for both private and public insurance populations. Thus, it is important for patients who are younger, with claudication or RA /collagen vascular diseases be more careful with follow up post surgery.

Paper 23. The Timing and Incidence of Surgical Intervention Following Diagnosis of Lumbar Disc Herniation at Military Treatment Facilities

Anderson Ashley, MD1, Pisano Alfred, MD2, Braswell Matthew, MD3, Watson Nora, PhD1, Helgeson Melvin, MD4, Brooks Daniel, PhD1, Wagner Scott, MD2

1Walter Reed National Military Medical Center/Uniformed Services University of the Health Sciences, Bethesda, Maryland, United States, 2Walter Reed National Military Medical Center, Bethesda, Maryland, United States, 3Carolina Medical Center Atrium Health, charlotte, United States, 4Walter Reed Army Medical Center, Bethesda, Maryland, United States

Background/Introduction

Radiculopathy from lumbar disc herniation (LDH) is a major cause of morbidity and cost among US civilians and active duty service members. The purposes of this study were to determine survival of conservative management of lumbar disc herniation in the military health system (MHS), and to determine risk factors to fail conservative management.

Materials/Methods

The Military Data Repository (MDR) was queried for all patients diagnosed with lumbar disc herniation from FY2011-2018; the earliest such diagnosis in a military treatment facility was kept for each patient as the initial diagnosis. In this set of patients, we then identified patients who failed conservative management, finding the time to the first post-diagnosis encounter for lumbar microdiscectomy or lumbar decompression in either a military treatment facility or in purchased care in the civilian sector. Censoring was determined to be the first day of the last month each patient was eligible to receive MHS benefits. Patients were also queried for history of smoking status (yes/no) at any time during MHS care, age at the time of diagnosis, sex, MHS beneficiary category, and diagnosing facility characteristics (clinic or hospital; facility branch; and MEPRS service code). Multivariable Cox proportional hazards models were used to evaluate the associations of patient and diagnosing facility characteristics with time to surgical intervention over the study period.

Results

84,985 MHS beneficiaries including 62,771 active duty service members were diagnosed with LDH in a military treatment facility during the study period. 10,532 (12.4%) MHS beneficiaries, including 7,650 (10.9%) active duty service, failed conservative management onto surgical intervention with lumbar microdiscectomy or lumbar decompression. Multivariable Cox regression among all healthcare beneficiaries revealed that younger age, male sex, and history of smoking were each associated with higher risk of surgical intervention independent of diagnosing facility characteristics.

Discussion/Conclusion

LDH compromises the readiness of our troops and negatively impacts healthcare costs for the Department of Defense (DoD). MHS beneficiaries with LDH have a good prognosis with approximately 88% of patients successfully completing conservative management. However, younger patients with a history of smoking should be educated that they are more likely to fail conservative management than patients without those demographic features.

Paper 24. The Influence of Selective Serotonin Reuptake Inhibitors on Lumbar Arthrodesis

Pirkle Sean, BA1, Bhattacharjee Sarah, BS1, El DaFrawy Mostafa, MBBCh1, Leucht Philipp, MD, PhD2, Shi Lewis, MD1, Lee Michael, MD1

1University of Chicago, Chicago, Illinois, United States, 2New York University School of Medicine, New York, New York, United States

Background/Introduction

Contemporary literature has linked selective serotonin reuptake inhibitors (SSRIs) to decreased bone mineral density and increased rates of future bone fracture. With a recent murine model suggesting a potential role in the quality of fracture healing itself, we sought to determine whether patients prescribed these medications are at an increased risk for nonunion following spine fusion surgery.

Materials/Methods

We performed a retrospective analysis within the PearlDiver database. All single-level lumbar fusion patients were identified. The rate of nonunion diagnosis between 6-24 months following surgery was assessed. A stratified analysis of chronic SSRI use and a number of comorbidities was conducted, followed by a multiple logistic regression analysis of nonunion accounting for qualifying risk factors. Finally, subanalyses of individual procedure codes were carried out.

Results

8,494 single-level lumbar fusion patients were included. In the multivariate analysis, chronic SSRI (OR: 1.572, p = 0.003) and tobacco use (OR: 1.389, p = 0.034) were identified as independent risk factors for nonunion while patient age over 60 years (OR: 0.483, p < 0.001) was observed to be negatively associated with nonunion. In the individual procedure sub-analyses, SSRIs were significantly associated with nonunion in all three univariate analyses and observed to be an independent risk factor for nonunion in two of the three procedure populations.

Discussion/Conclusion

These data suggest that patients treated concomitantly for mental health disorders with SSRIs before arthrodesis may be at an increased risk of postoperative nonunion. Closer follow up may be indicated in this patient population.

Paper 25. Preoperative Disc Height Predicts TLIF Cage Subsidence

Pisano Alfred, MD1, Fredericks Donald, MD1, Steelman Theodore, MD2, Riccio Corey, MD1, Kang Daniel, MD1, Helgeson Melvin, MD3, Wagner Scott, MD1

1Walter Reed National Military Medical Center, Bethesda, Maryland, United States, 2US Army, Walter Reed National Military Medical Center, Bethesda, Maryland, United States, 3Walter Reed Army Medical Center, Bethesda, Maryland, United States

Background/Introduction

Post-operative subsidence of the transforaminal lumbar interbody fusion (TLIF) cages can result in loss of lordosis and foraminal height, and in some cases recurrence of nerve root impingement and radicular symptoms. Our objective was to determine factors associated with TLIF cage subsidence. Specifically, we sought to determine if preoperative disc height compared to cage height could be used to predict TLIF interbody cage subsidence.

Materials/Methods

We retrospectively reviewed all patients undergoing instrumented TLIF from two institutions between July 2004 and June 2014. Preoperative disc height was measured for the operative and adjacent level disc heights. The delta between cage height and disc heights were measured and compared between subsidence and non-subsidence groups. We identified 127 patients who met our inclusion criteria, with a mean clinical follow up of 27 months.

Results

Fifty four patients (44.3%) had evidence of interbody cage subsidence. Average cage subsidence was 5.5mm, with a range of 2.2 to 10.8 mm. Implant height was significantly higher in the subsidence group (12.6mm) when compared to the non-subsidence group (11.2mm). Additionally, the delta between cage height and preoperative disc height was significantly larger in the subsidence group delta (5.5mm vs. 3.9mm). Similarly, the delta between cage height and preoperative adjacent level disc height was also significantly larger (1.2mm vs. 3.8mm).

Discussion/Conclusion

This study demonstrates that preoperative disc height and preoperative adjacent level disc height can be used to template TLIF cage height to potentially decrease rates of subsidence. Larger cages subsided more frequently than smaller cages. Patients with evidence of subsidence had larger deltas between cage height and preoperative disc height. Similarly, a significant difference existed if preoperative adjacent level disc height was compared to cage height. This demonstrates that preoperative imaging should be used for TLIF cage height templating to avoid subsidence.

Paper 26. Does Sagittal Balance Matter in Short-Segment (1-3 levels) Posterior Lumbar Spinal Fusion? A Meta-Analysis

Chieng Lee Onn, MD1, Perez-Cruet Mick, MD1, Elfallal Wissam, MD1

1Oakland University William Beaumont School of Medicine, Royal Oak, Michigan, United States

Background/Introduction

Restoration of sagittal balance has been associated with better clinical outcome in spinal deformity.However its implications on short-segment lumbar spinal fusion for degenerative spine disease remains unknown.In this study we sought to determine the association between adjacent segment disease(ASD) and spinopelvic parameters in short-segement posterior lumbar spinal fusion.

Materials/Methods

With adherence to PRISMA guideline, comprehensive database searches were conducted to identify studies that evaluate ASD with spinopelvic parameters in short-segment lumbar fusion.ASD was defined as radiographic and symptomatic axial pain or radiculopathy at adjacent level following index surgery. Independent reviewers extracted data and assessed the outcome.A meta-analysis using random effect weight mean difference model (WMD) was constructed.

Results

A total of 13 studies comprising of 1716 patients underwent 1-3 levels posterior lumbar fusion for degenerative spondylisthesis and spinal stenosis. Of these, 439 patients developed ASD following index surgery while the rest of the patients serve as control. Both groups had comparable basic demographic features in term of age,gender,and BMI. Preoperatively, there were significantly higher pelvic incidence-lumbar lordosis(PI-LL) mistmatch in ASD group(WMD 5.1, 95% CI 1.4-8.8;p=0.007) and lower sacral slope(WMD -1.8, 95% CI -3.6 to 0.02;p=0.05). Interestingly, preop lumbar lordosis (LL) was not observed to be associated with ASD along with pelvic tilt(PT), pelvic incidence(PI), sagittal vertical axis(SVA).However postoperatively, significant loss of LL was then observed in ASD group(WMD -5.7, 95%CI -8.8 to -2.6;p=0.0003) along with much higher PI-LL mismatch(WMD 10.1, 95% CI 5.6 to 14.6).A subgroup analysis of patients with ASD who eventually treated with surgery was again revealing significant pre- &post-op PI-LL mismatch>10.

Discussion/Conclusion

This study demostrated that the relation of pelvis to spine is critical in treating patients with degenerative lumbar spine disease.Failure to correct or progression of PI-LL mismatch were significantly associated with ASD with posterior lumbar spinal fusion even in short-segment constructs.

Paper 27. Coronal Cobb Angle Correction with Pre-psoas (Oblique) and Lateral Lumbar Interbody Fusion in Adult Scoliosis Patients

Guinn Jeremy, BS1, Xi Zhou, MD1, Burch Shane, MD2, Berven Sigurd, MD3, Deviren Vedat, MD2, Chou Dean, MD1

1Department of Neurological Surgery, UCSF Medical Center, San Francisco, California, United States, 2Department of Orthopedic Surgery, University of California San Francisco, San Francisco, California, United States, 3Department of Orthopedic surgery, San Francisco, University of California, United States

Background/Introduction

Lateral lumbar interbody fusion (LLIF) and oblique (pre-psoas) lumbar interbody fusion (OLIF) can be used to treat adult scoliosis. However, few studies have compared coronal Cobb angle correction with LLIF versus OLIF. We evaluated Cobb angle correction in patients with adult scoliosis who underwent OLIF and LLIF.

Materials/Methods

A retrospective study of scoliosis patients undergoing multi-level OLIF or LLIF by 5 spine surgeons at our hospital from 2013 to 2018 was performed. Data collected included demographic variables, approach-related factors such as operative time, blood loss, and coronal Cobb angle changes. A two-tailed t-test and a two-tailed z-test were used to compare the data collected for the two techniques. Patients with posterior osteotomies were excluded.

Results

A total of 66 patients were included, and there were 33 OLIF patients (66 levels) and 30 LLIF patients (78 levels). The mean coronal Cobb angle changes for OLIF at L1-L2, L2-L3, L3-L4, and L4-L5 were 9.00° ± 1.41, 6.76° ± 7.91, 1.92° ± 6.70, and 3.38° ± 5.59, respectively. The mean degree changes for LLIF at levels L1-L2, L2-L3, L3-L4, and L4-L5 were 2.56° ± 3.17, 2.70° ± 3.56, 1.50° ± 2.81, and 2.48° ± 4.84, respectively. The average change for all levels was 3.89° and 2.19° for OLIF and LLIF, respectively, with a p-value of 0.062 when comparing the two.

Discussion/Conclusion

Coronal Cobb angle correction between OLIF and LLIF was not statistically different. Given the similar correction abilities between the two techniques, OLIF may be a feasible alternative to LLIF if Cobb angle correction is desired in adult scoliosis cases.

Paper 28. Concomitant Spinal Deformity and Sacroiliac Dysfunction

Heller Joshua, MD, MBA1

1Thomas Jefferson University Hospital, Philadelphia, Pennsylvania, United States

Background/Introduction

Spinal deformity including scoliosis and kyphosis is thought to contribute to sacroiliac (SI) dysfunction. The hypothesis is that spinal malalignment contributes to abnormal or asymmetric loading of the SI joints and subsequent SI dysfunction. The surgical treatment of spinal deformity including long segment fixation to the sacrum and ilium is also thought to contribute to SI dysfunction. Prior studies have shown an increase in SI dysfunction in those with lumbar fusion. The incidence of spinal deformity patients with concomitant SI dysfunction is unknown.

Materials/Methods

An IRB approved single center, single surgeon, retrospective review was performed to determine incidence of concomitant Adult Spinal Deformity (ASD) and SI dysfunction. Demographic, clinical, and radiographic information including long segment radiographs were then evaluated. Clinical data reviewed included relevant SI evaluation criteria, i.e. Fortin Finger Sign, Provocative Testing, and Response to Intraarticular SI Injection.

Results

A query of the EMR between January 2016 and June 2019 for ICD 10 any diagnosis of Scoliosis M41.* or Kyphosis and Lordosis M40.* returned a base population of 537 deformity patients. The addition of any diagnosis of Sacroiliitis M46.1 or Sacrococcygeal Disorders not elsewhere classified M53.3 narrowed the population to 109 patients. The incidence in the authors practice thus is 20.3%. Of the 109 patients 90 had imaging/data available for review. Median age was 62. 77.5% of the population are female. Patients with prior lumbar fusion were 3x more likely to undergo surgery for SI dysfunction. Radiographically sagittal, coronal, or combined malalignment was identified in 23.6%, 11.2%, and 44.9% of the patients respectively. 85% of patients had provocative testing: right positive 43%, left positive 28%, B/L positive 7%, negative 8%. Marked relief was observed in 2/3 of those that had SI intraarticular injection. The most frequent pattern observed in the population was: Positive sagittal imbalance with a Left coronal imbalance and Right SI Dysfunction (26.97%).

Discussion/Conclusion

Spinal deformity and its surgical treatment both likely contribute to SI dysfunction. The incidence of concomitant ASD and suspected SI disfunction is 20.3%. The most frequent pattern observed was: Positive sagittal imbalance with a Left coronal imbalance and Right SI Dysfunction (26.97%).

Paper 29. Revision surgery for adjacent segment degeneration after fusion for lumbar spondylolisthesis: is there a correlation with Roussouly type?

Duan Pingguo, MD1, Mayer Rory, MD1, Ruan Hui Bing, MD1, Chang Chih-Chang, MD1, Chou Dean, MD2

1UCSF Spine Center University of California, San Francisco, San Francisco, California, United States, 2Department of Neurological Surgery, UCSF Medical Center, San Francisco, California, United States

Background/Introduction

Revision surgery for ASD is known to occur after spinal fusion; however, it is unclear if rates of ASD are associated with certain Roussouly types. The aim of this study was to investigate whether there is an association between revision surgery rates for adjacent segment degeneration (ASD) and Roussouly type after L4-5 transforaminal lumbar interbody fusion (TLIF) for spondylolisthesis.

Materials/Methods

Patients who underwent L4-5 TLIF for spondylolisthesis at the University of California San Francisco from January 2006 to December 2016 with minimum 2-year follow up were retrospectively analyzed by Roussouly type. Revision surgery for ASD was noted and correlated by Roussouly type. Spinopelvic parameters were also measured for correlation. A value of p <0.05 was significant.

Results

There were 174 patients who met inclusion criteria, (59 males and 115 females). The average age was 62.3 (25-80) years. 132 patients had grade I spondylolisthesis, and 42 had grade II. Mean follow-up was 45.2 months (24 to 497). A total of 22 patients (12.6%) underwent revision surgery for ASD after L4-5 TLIF. When classified by Roussouly type, revision surgery rates for ASD were: 1, 14.3%; 2, 22.6%; 3, 4.9%; and 4, 15.6% (p = 0.013). Type 3 spines with normal PI-LL (8.85° ± 6.83°) had the lowest revision surgery rate (4.9%), and type 2 spines with PI-LL mismatch (11.06° ± 8.81°) had the highest revision surgery rate (22.6%), a 4-fold difference (p = 0.013). The PI-LL mismatch did not change significantly in each type post-operatively (p > 0.05).

Discussion/Conclusion

We found that there may be a correlation between Roussouly type and revision surgery for ASD after L4-5 TLIF for spondylolisthesis, with type 2 spines having the highest rate. Spinopelvic parameters may also correlate with revision surgery for ASD after L4-5 TLIF.

Paper 30. Preoperative Bone Health Optimization In Spine Surgery

Kadri Aamir, BS1, Anderson Paul, MD1, Binkley Neil, MD1, Hare Kristyn, PA-C1

1University of Wisconsin-Madison, Madison, Wisconsin, United States

Background/Introduction

Osteoporosis is associated with adverse outcomes in spine surgery, including pseudoarthrosis, hardware failure, fracture, revision surgery, and proximal junction kyphosis. Bone Health Optimization (BHO) is an intervention intended to improve the structure and biologic properties of bone to reduce the likelihood of these complications. This study characterizes a patient cohort referred for BHO to test the hypothesis that poor bone quality is common in spine surgery and that many such patients meet guidelines for osteoporosis treatment.

Materials/Methods

This retrospective study evaluated 104 patients referred for BHO by their surgeon based on suspected poor bone health who were age ≥ 50 and candidates for thoracolumbar surgery. Fracture Risk Assessment Tool (FRAX) risk factors and dual-energy X-ray absorptiometry (DXA) results were collected. When available, opportunistic computed tomography (CT) and trabecular bone score (TBS) were evaluated. World Health Organization (WHO) diagnostic and National Osteoporosis Foundation (NOF) treatment guidelines were applied.

Results

The mean age was 69.0 years, 81% were female, and 97% were Caucasian. A previous fracture occurred in 54%, and half had more than one fracture with two-thirds occurring after age 50. The mean historical height loss was 5.6 cm. The mean (SD) lowest T-score of the hip, spine, or wrist was -2.12 (0.65) in females and -1.97 (0.94) in males. The mean spine T-score was -0.43 (1.87), while the hip was -2.09(0.71), which was significantly different. Osteoporosis (T-score ≤ -2.5) was present in 49.5% of females and 32% of males. Opportunistic CT identified 58% likely having osteoporosis (HU of L1<100). TBS identified 32.9% with degraded and 26% with partially degraded bone microarchitecture. The FRAX 10-year major fracture and hip fracture thresholds of 20% for major osteoporotic fracture and 3% for hip fracture were exceeded in 79.6% of patients. NOF threshold for osteoporosis treatment was met in 82.5% of patients. Treatment was prescribed in 75% of patients (45% anabolic, 30% antiresorptive therapies).

Discussion/Conclusion

Osteoporosis, degraded bone microarchitecture, prior fracture, and elevated fracture risk were common. Given the high prevalence of impaired bone health in this cohort, we believe that bone health screening, including FRAX assessment, should be considered in selected spine surgical patients as part of pre-operative optimization for all adults age ≥ 50.

Paper 31. Vertebral Bone Quality Score and Body Mass Index Independently Predict Need for Reoperation After Lumbar Fusion

Ehresman Jeffrey, BS1, Ahmed A. Karim, BS1, Lubelski Daniel, MD1, Schilling Andrew, AB1, Pennington Zach, BS1, Cottrill Ethan, MS1, Witham Timothy, MD1, Sciubba Daniel, MD1, Perdomo-Pantoja Alexander, MD1

1Johns Hopkins University School of Medicine, Baltimore, Maryland, United States

Background/Introduction

Poor bone quality is a known risk factor for hardware failure and adjacent segment disease following lumbar fusion. Dual energy x-ray absorptiometry (DXA) is currently the most utilized tool to assess bone mineral density; however, previous studies have questioned whether bone “density” is equivalent to bone “quality”. One new method of analyzing bone quality is the vertebral bone quality (VBQ) score, which can be obtained from preoperative lumbar MRIs. To test whether this tool could provide additional insights in spine surgery patients, we decided to evaluate whether patients’ VBQ scores were associated with reoperation after lumbar fusion due to hardware failure or symptomatic adjacent segment disease.

Materials/Methods

We queried records of patients who underwent elective lumbar fusion for degenerative conditions between 2013 and 2017. Collected data included demographics, symptoms, surgical indications, operative information, and radiographic information (spinopelvic parameters, DEXA T-scores, VBQ scores). Patients who required reoperations after lumbar fusions due to hardware failure or symptomatic adjacent segment disease were combined into a case group. A control group of equal size was then created using patients who did not require reoperation and this group was individually matched by age, sex, and surgical indication to analyze risk factors for reoperation.

Results

Of the 351 patients who underwent first-time elective lumbar fusions within the reviewed time period, 46 patients required reoperation for symptomatic hardware failure or adjacent segment disease and 30 met the inclusion criteria. A control group of 30 individually age- and sex-matched patients who did not require reoperation was then created. The Reoperation group had significantly higher BMI (32.0 vs. 28.5, p = 0.019). There were no significant differences regarding spinopelvic parameters, Pfirrmann scores, or DEXA T-scores. There was, however, a significant difference in VBQ scores between the groups, with the Reoperation group having a higher VBQ score, indicating poorer bone quality (3.30 vs. 2.93, p = 0.030).

Discussion/Conclusion

This study found that bone quality, obtained using the VBQ score, is an important risk factor for reoperation after lumbar fusion surgery. Therefore, this MRI-based tool may be used to assist surgeons in preoperative planning for spine surgeries to reduce the risk of requiring reoperation.

Paper 32. A Deep Learning Approach to Efficiently Triaging Spine Surgery Patients Based Upon Computerized Intake Questionnaires

Schrum Mariah, BS1, Yoon Eric, BS2, Gombolay Matthew, PhD1, Yoon Sangwook, MD, PhD2

1Georgia Institute of Technology, Atlanta, Georgia, United States, 2Emory University Department of Orthopaedics, Atlanta, Georgia, Georgia

Background/Introduction

The volume of outpatient visits to a spine center or spine healthcare provider is often very high. The appropriate triage of patients to the correct spine provider (surgeon or non-surgeon) is an important decision point that can impact provider productivity and patient satisfaction. Triaging systems are often ineffective, resource intensive, or not-scalable. To address this problem, we developed a deep learning triage method that can be tuned and scaled.

Materials/Methods

We employed deep learning techniques with artificial neural networks to develop a system that can improve the triage of patients seeking appointment at a busy spine center. We obtained IRB approval for a retrospective review of 1200 computerized patient in-take questionnaires from our Spine Center. The questionnaires include demographic data, symptomatic information (location and duration of pain, aggravating and ameliorating factors etc), and health history data. This was used to train a complex neural network to predict disease. Then the pathology predictions are fed into the final output layer for prediction of whether the patient had surgery.

Results

Based on our chart review, under the conventional triage system, 70% of patients triaged to surgeons did not have surgery. Using the deep learning methods trained on the data set, we were able to tune the system to correctly identify 100% of surgical patients while triaging out 31% of those patients that are not surgical candidates. By tuning the model, we can still preserve the ability to identify 90% of surgical patients while triaging over half (54%) of patients who did not have surgery.

Discussion/Conclusion

This is the first study investigating the efficacy of using intake questionnaires for triaging patients for spine surgery via deep neural networks. Our results show that we are able to tune the algorithm to prioritize capturing surgical patients or reducing non-surgical patients. This system can be adapted to different priorities in an automated manner and to suit the needs of individual providers or spine center as a whole. Since this is an automated system, this system is also scalable without increasing per triage cost.

Paper 33. Utility of Neutral Sitting Lateral Radiograph to Evaluate Segmental Instability in Lumbar Spondylolisthesis

Sun Xu, PhD1, Zhou Qingshuang, MD1, Qian Bangping, PhD1, Wang Bin, PhD1, Zhu Zezhang, PhD1, Qiu Yong, PhD1

1Department of Spine Surgery, Drum Tower Hospital, Medical School of Nanjing University, Nanjing, Jiangsu, China

Background/Introduction

Degenerative lumbar spondylolisthesis (DLS) is a common disease and often results in back pain and neurogenic claudication, which frequently requires surgical intervention. However, segmental instability is a key factor in determining the treatment algorithm. This study aimed to investigate the olisthesis of the involved segmental degenerative lumbar spondylolisthesis (DLS) through slump sitting radiograph, and to determine the most useful diagnostic modality in the evaluation of instability.

Materials/Methods

This study recruited a consecutive series of patients with L4/5 DLS who underwent surgical intervention at our center between October 2018 and April 2019. All subjects were performed with slump sitting (Si), upright standing (U), flexion (F) and extension (E) radiographs, and the supine MR imaging (Su). Slip parameters were measured on all the sagittal images. Maximal translational and angular range of motion (ROM) was determined by comparing slip parameters between images.

Results

A total of 51 patients (14 men, 37 women), with a mean age of (58.5±7.5) years were included. In all, the Si radiograph was detected with the maximal SP and minimum SA (P<0.05). The Su sagittal T2-MR imaging was demonstrated with more anterior reduction than the E radiograph (P<0.05). The Si-Su modality revealed greater mean translational ROM than others (P<0.05). Not merely the Si-E modality but also the Si-U modality was detected with higher angular ROM than the F-E modility (P<0.05).

Discussion/Conclusion

The Si radiograph was demonstrated with maximal anterior listhesis and kyphotic SA than others, and the Su MR imaging could reveal greater listhesis reduction than the E radiograph. Utility of the Si lateral radiograph can improve our understanding of segmental mobility when evaluating DLS with the F-E modality.

Paper 34. Characterizing Disc Biomarker Profiles in Degenerative Lumbar Spondylolisthesis Patients

Choma Theodore, MD1, Kramer Jacob, BS1, Stoker Aaron, PhD1, Lee Naomi, BS, DVM1, Pundee Chaiyapruk, MD1, Moore Don, MD1, Cook James, PhD, DVM1

1University of Missouri-Columbia, Columbia, Missouri, United States

Background/Introduction

Intervertebral disc degeneration (IVDD)encompasses a spectrum of disorders that involve maladaptive biomechanical and cellular responses to chronic loading, insult, and or injury; and has been implicated in a substantial number of back and neck pain patients. Neither the mechanism nor impetus of IVD degeneration has been thoroughly characterized, nor is it known how the changes in mechanical stresses applied to the IVD due to instability effects the pathophysiology of IVDD. It was hypothesized that lumbar IVD tissues would show higher production of pro-inflammatory and degradative biomarkers in patients diagnosed with spondylolisthesis compared to patients without instability.

Materials/Methods

With IRB approval and informed patient consent, IVD tissues excised and then otherwise discarded at spinal surgery were obtained from patients (n=49, mean age = 54.0, 31 female, 15 diagnosed with spondylolisthesis). Four explants of combined NP and AF per disc segment were created with a 6 mm diameter biopsy punch. One set of AF and NP explants were cultured in basal media, the other set was exposed to 10 ng/mL IL-1β. The explants were cultured for 3 days in supplemented Dulbecco’s modified eagles medium. Media were collected on day 3 of culture and stored at -20°C until used for biomarker evaluation.

Results

Wit regard to pro-inflammatory biomarkers, IVD explants from both stable and unstable segments significantly increased (p≤0.001) the production of GRO-α, MCP-3, IL-2, IL-6, IL-8, MIP-1α, MIP-1β, and TNF-α with cytokine stimulation. IVD explants from stable segments significantly increased the production of MCP-1, while IVD explants from unstable segments increased the production of IL-4, and significantly less MCP-3 and significantly more RANTES (p=0.044) with cytokine stimulation. in terms of degradative biomarkers, IVD explants from unstable segments produced significantly more TIMP-3 (p=0.007), TIMP-4 (p=0.001), and MMP-8 (p=0.001) compared to stable segments.

Discussion/Conclusion

The data from this study indicate that spondylolisthesis is significantly associated with basal and cytokine-stimulated release of biomarkers related to inflammatory, degradative, and repair pathways from degenerative lumbar discs. Further characterization of the interplay of spinal instability and progression of IVD degeneration is required.

Paper 35. Development and Validation of an Large Ovine Animal Model for Biofilm Based Periprosthetic Spine Infection

Bailey Travis, MD1, Ong Jemi, BS1, Brodke Darrel, MD1, Ashton Nicholas, PhD1, Williams Dustin, PhD1, Shaw Jeremy, MD, MS2

1University of Utah, Salt Lake City, Salt Lake City, Utah, United States, 2University of Pittsburgh Medical Center, Pittsburgh, Pennsylvania, United States

Background/Introduction

Postoperative spine infections account for 28,000 cases yearly within the US. The ability to diagnose, treat and eradicate the infectious source is paramount to alleviating the burden placed on the patient and the healthcare system. The purpose of this study was to develop the first ovine large animal model for deep implant associated post-operative spine infection using biofilms as initial inocula.

Materials/Methods

Methicillin-sensitive Staphylococcus aureus (MSSA) biofilms were grown on titanium rods in a modified CDC biofilm reactor: brain heart infusion (BHI) broth was inoculated with ∼105 colony forming units (CFU)/mL, incubated for 24 hrs, then with an additional 24 hrs of flow. In five adult sheep, two surgical sites were utilized – one thoracic / one lumbar. Incisions were separated by at least 15 cm and randomized per animal as infected or sterile. At each site, pedicle screws were placed in a standard fashion. Beginning with the sterile site, an uninoculated rod was implanted. The sterile wound was then closed and sealed. The infected site followed in similar fashion with an MSSA inoculated implant. Each sheep followed the same protocol and were monitored for 28 days. Quantification of biofilm was performed prior to implantation and at explantation prior to necropsy. Tissue biopsies from the surgical site were quantified with a 7-fold dilution series.

Results

There was no cross contamination between infected vs uninfected surgical sites within the same animal. No animal required additional antibiotics other than the single preoperative prophylactic dose. Biofilm inocula produced robust and consistent biofilm on spinal hardware as confirmed by quantification and SEM imaging. Rods contained ∼109 CFU/implant prior to implantation. Roughly 107 CFU remained on contaminated implants. There was robust local tissue response creating a reliable deep spine infection.

Discussion/Conclusion

The present study developed an animal efficient ovine model for deep post-operative spine infection. This novel study produced a single animal capable of harboring both an infected and non-infected wound thereby reducing the number of animals needed for study and increasing statistical power secondary to internal controls. This study showed that MSSA biofilm inocula can produce a clinically relevant deep spine infection.

Paper 36. Characterization of Disc Biomarker Profiles for Patients undergoing Lumbar Discectomy vs Fusion

Choma Theodore, MD1, Moore Don, MD1, Kramer Jacob, BS1, Stoker Aaron, PhD1, Lee Naomi, BS, DVM1, Pundee Chaiyapruk, MD1, Cook James, PhD, DVM1

1University of Missouri-Columbia, Columbia, Missouri, United States

Background/Introduction

Intervertebral disc degeneration (IVDD) is prevalent and associated with painful degenerative conditions. Neither the mechanism nor impetus of IVDD has been wholly characterized. No therapies are known to restore the original state or function of the degenerative IVD. We routinely remove portions of disc in common procedures for microdiscectomy and interbody fusion. While discectomy is often performed on patients with earlier stage IVDD, and fusion on patients with later stage IVDD, it is not known if the metabolic profiles of the IVDs differ among these two cohorts. It was hypothesized that the release of pro-inflammatory and degradation biomarkers by lumbar IVD explants recovered from discectomy procedures would be significantly lower than those recovered from fusion procedures

Materials/Methods

With IRB approval and informed patient consent, diseased IVD tissue excised as part of a standard-of-care discectomy (n=14, mean age 40.85, 6 female) or fusion (n=35, mean age 59.85, 25 female) spinal surgery. Two explants of combined NP and AF per disc segment were created with a 6 mm biopsy punch and cultured for 3 days in supplemented Dulbecco’s modified eagles medium. Media were collected on day 3 of culture and stored at -20°C until used for biomarker evaluation. Media were analysed for multiple inflammatory biomarkers and growth factors.

Results

Pro-Inflammatory Biomarkers: Degenerative lumbar IVD explants harvested from patients undergoing discectomy released significantly (p≤0.021) more GRO-α, MIP-1α, and RANTES, and significantly (p=0.037) less MCP-3, compared to tissues harvested from patients undergoing fusions. Growth Factors: Degenerative lumbar IVD explants harvested from patients undergoing discectomy released significantly (p≤0.003) more PDGF-AA, PDGF-AB\BB, and VEGF compared to tissues harvested from patients undergoing fusions. Degradative Biomarkers: Degenerative lumbar IVD explants harvested from patients undergoing fusion procedures released significantly (p≤0.016) more MMP-2, MMP-7, and MMP-13, and significantly (p≤0.004) less MMP-8, TIMP-2, and TIMP-4, compared to tissues harvested from patients undergoing discectomies.

Discussion/Conclusion

The data from this study indicate that there are significant differences in the pro-inflammatory and degradative biomarker profiles of IVDs from patients undergoing lumbar discectomy versus fusion. Ongoing studies in our laboratory are aimed at further characterization.

Paper 37. PROMIS Scores Alone are Not a Reliable Measure of Patient Outcomes After Surgery for Lumbar Stenosis

Liu Andrew, MD1, Sulovari Aron, BS2, Maqsoodi Noorullah, BS1, Everett Clifford, MD1, Mesfin Addisu, MD2

1University of Rochester, Univ. of Rochester School of Medicine and Dentistry, Rochester, Michigan, United States, 2University of Rochester Orthopedics, Rochester, New York, United States

Background/Introduction

As the healthcare industry shifts to a value-based care model, patient reported outcome instruments such as the NIH-supported Patient-Reported Outcomes Measurement Information System (PROMIS), which evaluates patients’ global well-being, are increasingly being used to gauge the cost effectiveness of care. The Swiss Spinal Stenosis Questionnaire (SSSQ) is considered the gold standard validated disease-specific outcome measure for the treatment of lumbar spinal stenosis and has been found to be internally consistent, reproducible and responsive to clinical change. As PROMIS is further integrated into clinical care, its benefit must be evaluated relative to traditional PRO instruments such as the SSSQ. This study aims to establish the strength of relationship between three commonly used PROMIS measures and the SSSQ.

Materials/Methods

82 patients with spinal stenosis at a single academic medical center were enrolled. They filled out one SSSQ during their enrollment visit. Patients completed the PROMIS Adult Depression (AD), Physical Function (PF), and Pain Interference (PI) instruments at each visit. For patients who had undergone lumbar decompressive surgery, PROMIS scores from the first two postop visits were averaged. Regression analysis was performed to evaluate the pain and physical function domains of the SSSQ against their respective PROMIS counterparts (PI and PF), treatment satisfaction component of the SSSQ against postop PROMIS scores, and combined pre-treatment SSSQ against PROMIS AD.

Results

PROMIS PF correlates moderately well with the SSSQ physical function domain (R2 0.14, p=0.0008) while PROMIS PI correlates poorly with the SSSQ pain domain (R2 0.03, p= 0.13). PROMIS AD correlates relatively poorly to the pre-treatment SSSQ (R2 0.07, p= 0.02). In patients who have undergone surgery, postoperative PROMIS scores in all three domains correlate reasonably well with the treatment satisfaction component of the SSSQ [PROMIS AD (R2 0.13, p=0.01), PROMIS PI (R2 0.25, p= 0.003), PROMIS PF (R2 0.18, p= 0.003)].

Discussion/Conclusion

While postoperative PROMIS scores reflect patient outcomes after surgery for spinal stenosis reasonably well, preoperative PROMIS scores do not adequately capture the disease-specific impact of spinal stenosis on patients’ overall wellbeing. Therefore, PROMIS scores should not be used in isolation to assess outcomes in patients with spinal stenosis.

Paper 38. How Does Discharge Disposition Influence Patient-Reported Outcomes and Calls to Providers after Spine Surgery?

Bowles Daniel, MD1, Canseco Jose, MD, PhD1, Divi Srikanth, MD1, Anderson D., MD2, Rihn Jeff, MD1, Kurd Mark, MD1, Hilibrand Alan, MD1, Kepler Christopher, MD, MBA1, Vaccaro Alexander, MD, PhD, MBA1, Schroeder Gregory, MD1

1Rothman Institute at Thomas Jefferson University, Philadelphia, Pennsylvania, United States, 2Department of Orthopaedic Surgery, Rothman Institute, Thomas Jefferson University, Philadelphia, Pennsylvania, United States

Background/Introduction

Discharge disposition postoperatively has been shown to impact disability, injury, and healthcare utilization. There is scant literature regarding the effect discharge disposition settings have on patient-reported health-related quality of life (HRQOL) outcomes after surgery. The aim of this study was to identify associations between discharge disposition settings, postoperative HRQOL outcomes and healthcare utilization after lumbar surgery.

Materials/Methods

Patients undergoing 1-3-level lumbar decompression, 1-3-level lumbar decompression and fusion, or posterior cervical decompression and fusion (PCDF) were retrospectively identified. Patients were stratified into those discharged to home/self-care (home), those discharged to inpatient rehabilitation (acute), and those discharged to skilled nursing or long-term rehabilitation (subacute). HRQOL outcomes analyzed included the Visual Analogue Scale (VAS) Back/Neck and Leg/Arm pain scores, Short Form-12 Physical Component Score (PCS-12) and Mental Component Score (MCS-12), Oswestry Disability Index (ODI) and Neck Disability Index (NDI). Phone call data, including number of calls in the 1-year postoperative period and primary reason for calling, were obtained for secondary analysis.

Results

A total of 623 patients were included, with 480 patients in the Home group, 72 patients in the Acute group, and 80 patients in the Subacute group. Mean age was 70.3 [69.7, 71.0] years. Acute and Subacute patients had a higher average age compared to the home group (72.5 [70.9, 74.2] and 72.8 [71.2, 74.4] vs. 69.3 [68.6, 70.1] years, p <0.001). Females were more likely to be discharged to rehabilitation (29.7% vs. 20%, p < 0.001). HRQOL outcome analysis revealed that Home patients improved across all outcome measures, except VAS-Leg (p=0.276); whereas rehabilitation patients only improved in MCS-12 scores (p=0.018). Multiple linear regression found discharge to acute rehabilitation was a significant predictor for less improvement in PCS-12 (β=-5.328 [-10.524, -0.402], p=0.034). Overall readmission rates and number of postoperative provider phone calls were similar between groups.

Discussion/Conclusion

Our results suggest that home discharge after lumbar surgery is associated with improvement across various HRQOL outcome measures; whereas discharge to acute rehabilitation facilities is a predictor of worse physical health-related outcomes. Patients across all groups and surgery types utilized a similar amount of resources in terms of readmissions and postoperative telephone communication.

Paper 39. The Re-herniation After Decompression (RAD) Score Identifies Patients at Low-Risk for Re-herniation After Lumbar Decompression Surgery

Harada Garrett, MD1, Basques Bryce, MD1, An Howard, MD1, Samartzis Dino, PhD2

1Midwest Orthopaedics at Rush, Chicago, Illinois, United States, 2Midwest Orthopaedics at Rush University, Chicago, Illinois, United States

Background/Introduction

Re-herniation is a common, though unpredictable occurrence after surgical decompression for a herniated nucleus pulposus (HNP) that may lead to adverse patient outcomes. The purpose of this study was to create a prognostic tool to identify patients at greatest risk for re-herniation after surgery.

Materials/Methods

We performed a retrospective examination of patients receiving decompression of the lumbar spine for an HNP from 2006-2018. Baseline demographics (age, sex, BMI, American Society of Anesthesiologist Class, smoking history), clinical (symptom duration, herniation type, operative levels), and radiographic information (pelvic tilt, pelvic incidence, sacral slope, coronal angulation, lumbar lordosis) was collected in addition to initial and future herniation events. Re-herniations were identified on magnetic resonance imaging and included operative and non-operative cases. Multiple imputation and LASSO logistic regression was employed to build and validate a predictive model to identify patients at differing levels of risk for re-herniations after decompression surgery.

Results

A total of 517 patients (38.5% female) presented for surgical decompression of a lumbar HNP, in which 51 experienced re-herniations. Patients had a mean age of 45.2±13.9 years, BMI of 29.4±6.3 kg/m2, and 22.4±28.0 months of follow-up. Most herniations were paracentral (83.4%), followed by central (10.6%), and far-lateral (5.6%). Re-herniations were seen at a mean of 46.0±70.0 months after surgery. LASSO identified nine predictors for re-herniations. (Table 1) Validation of this model demonstrated excellent discrimination (AUC:0.834), sensitivity (71.4%), and specificity (87.6%). Within the testing cohort, 16.3% (17/104) of patients were identified as high-risk. Of these, 29.4% sustained re-herniations, while nearly all low-risk patients (98.7%, 85/87) experienced no recurrence.

Discussion/Conclusion

This is the first study to develop and validate a tool identifying patients with differences in risk for re-herniation after lumbar decompression. With further validation, this tool can be implemented in practice, and may accurately identify patients likely to experience lasting surgical benefit for lumbar disc herniations.

Paper 40. Outcomes Following Surgical Decompression for Gunshot Wounds to the Lumbar Spine: A Systematic Review and Meta-analysis

Platt Andrew, MD1, Herman Martin, MD, PhD1, Ramos Edwin, MD1

1University of Chicago, Chicago, Illinois, United States

Background/Introduction

Gunshot wounds to the spine account for 13% to 17% of all spinal cord injuries each year. Indications for surgical decompression of spinal gunshot wounds are highly debated in current medical literature. The lumbar spine has a unique anatomy which includes the termination of the spinal cord in the conus medullaris and presence of the cauda equina. Several studies have encouraged decompression of gunshot wounds to the lumbar spine. This present study was undertaken to examine whether surgical decompression of lumbar gunshot wounds results in a statistically significant increase in neurologic function as compared to conservative management.

Materials/Methods

A systematic review and meta-analysis was performed to assess superiority in terms of neurologic improvement and complications between patients undergoing operative decompression and non-operative management after sustaining gunshot wounds to the lumbar spine. Neurologic improvement was further analyzed in patients with complete and incomplete neurologic injuries on presentation.

Results

Four studies were included that met inclusion criteria. Eighty-one patients were included in the surgical arm and 36 patients were included in the non-operative arm. Among all patients, 59 of 81 patients undergoing surgical decompression showed neurologic improvement at follow-up compared to 25 of 36 patients undergoing non-operative management. Neurologic improvement slightly favored surgical decompression, however, the relationship was not significantly significant (OR 1.02; 95% CI 0.41, 2.52; p=0.96). Among patients with complete injuries, 7 of 10 patients showed neurologic improvement after surgical decompression compared to 2 of 7 patients undergoing non-operative management. Neurologic improvement was greater in the surgical arm however not statistically significant (OR 4.13; 95% CI 0.55, 30.80; p=0.17). Among patients with incomplete injuries, 24 of 34 showed neurologic improvement after surgical decompression compared to 16 of 20 undergoing non-operative management. Neurologic improvement was greater in the non-operative arm however not statistically significant (OR 0.38; 95% CI 0.10, 1.52; p=0.17).

Discussion/Conclusion

Our results suggest that patients undergoing surgical decompression after sustaining a gunshot wound to the lumbar spine may not have increased neurologic improvement compared to patients undergoing non-operative management. To the best of our knowledge, this is the first meta-analysis of patients with gunshot wounds to the lumbar spine.

Paper 41. Primary Single-level Lumbar Microdisectomy/Decompression at a Free-Standing Ambulatory Surgical Center vs. a Hospital-Owned Outpatient Department – An Analysis of 90-day Outcomes and Costs

Malik Azeem, MBBS1, Xie Jack, MS2, Retchin Sheldon, MD3, Phillips Frank, MD4, Xu Wendy, PhD3, Yu Elizabeth, MD1, Kim Jeffery, MD1, Khan Safdar, MD1

1Ohio State University Wexner Medical Center, Columbus, Ohio, United States, 2The Ohio State University, College of Medicine, Columbus, Ohio, United States, 3Ohio State University, Columbus, Ohio, United States, 4Midwest Orthopaedics at Rush, Chicago, Illinois, United States

Background/Introduction

While free-standing ambulatory surgical centers (ASCs) have been extolled as lower cost settings than hospital outpatient facilities/departments (HOPDs) for performing routine elective spine surgeries, differences in 90-day costs and complications have yet to be evaluated between the two types of treatment facilities

Materials/Methods

The Humana 2007-2017Q1 insurance dataset, of Medicare Advantage and Commercial beneficiaries, was queried using CPT codes 63030, 63056 and 0275T to identify patients undergoing primary single-level lumbar microdiscectomy/decompressions. Patients undergoing two-level surgery, laminectomies, fusions, revision discectomies and/or deformity surgeries were excluded. Service Location codes for HOPD (22) and free-standing ASC (24) were used to identify where patients received surgery. The two groups were propensity-score matched, using a 1:1 ratio, on the basis of age, gender, race, region and Elixhauser co-morbidity index (ECI). Multi-variate regression analyses were used to compare differences in 90-day complications, readmissions and costs between the two groups.

Results

A total of 1,077 and 10,475 primary single-level decompressions were performed in ASCs and HOPDs, respectively. Following a matching algorithm with propensity scoring, the two cohorts comprised of 990 patients each. There were no significant differences in the rates of all-cause 90-day complications (ASC=9.1% vs. HOPD=10.3%; p=0.362) and 90-day readmissions (ASC=4.5% vs. HOPD=5.3%; p=0.466). However, performing surgery in an ASC resulted in average cost savings of over $2000 per case compared with the same procedure performed in HOPDs. When stratified by payer, average costs were $5,814 in ASCs vs. $7,829 in HOPDs for Medicare Advantage enrollees and $10,116 in ASCs vs. $13,623 in HOPDs for commercially insured enrollees.

Discussion/Conclusion

Performing single-level decompression surgeries in an ASC, compared with HOPDs, was associated with approximately $2,000-$3,500 cost-savings per case with no statistically significant impact on complication or readmission rates.

Paper 42. Not Without My Attending: A Survey of Patient and Family Member Attitudes and Perceptions About Concurrent and Overlapping Surgery

Kim Andrew, MS1, Alluri Ram, MD2, Kang Hyunwoo, MD2, Wang Jeffrey, MD2, Hah Raymond, MD2

1Loyola University Chicago, Chicago, Illinois, United States, 2Keck Medical Center of USC, Los Angeles, California, United States

Background/Introduction

Despite the extensive practice of overlapping surgery in the medical field, most patients have very little knowledge about it and several studies investigating public perception of overlapping surgery have revealed strong disapproval of the practice. We sought to examine the attitudes and opinions of spine surgery patients and their family members toward concurrent and overlapping surgery, as well as toward surgical trainee participation in their spine surgeries.

Materials/Methods

A 36-item survey was given to 146 patients and their family members in the waiting area of an academic orthopaedic spine surgery clinic. The survey queried respondents’ baseline knowledge of and comfort level with concurrent and overlapping surgery, opinions on when they felt that the attending should be present in the OR during the procedure, comfort level with different levels of surgical trainees, and preferences regarding various aspects of the surgical procedure which should be disclosed during the informed consent process.

Results

Knowledge about concurrent and overlapping surgery was quite poor in our study population, with a little over half of respondents (58.22%) reporting that they felt comfortable with overlapping surgery. Respondents identified the critical period of the surgery as being the portion of the case when most patients feel the attending surgeon should be present, followed by the opening incision and then the prepping, draping and positioning. Almost all respondents reported that they would like their surgeon to disclose the participation of surgical trainees (residents and fellows) in their surgery. In addition, the 4th and 5th years of surgical training were associated with a significant increase in patient comfort with surgical trainee participation.

Discussion/Conclusion

Knowledge about concurrent and overlapping surgery remains poor in the general population, a troublesome finding given the widespread practice of the latter. The relatively low comfort level with overlapping surgery indicates the persistence of a strong negative perception toward the idea of surgeons “double-booking” cases. Surgeons who routinely schedule overlapping surgeries should seek to improve their informed consent process by counseling their patient on overlapping surgery, defining the different components of the surgery (including the critical periods), and disclosing the role of surgical trainees.

Paper 43. SINS and Predicting Failure of Radiation Therapy in Spinal Metastatic Disease

Baksh Nikolas, MD1, McKibben Natasha, BS1, Nguyen Thuy, DO1, Brown Simon, MD1, Jaboin Jerry, MD1, Lin Clifford, MD1

1Oregon Health and Science University, Portland, Maine, United States

Background/Introduction

The spinal instability in neoplasia score (SINS) was developed as a referral tool for medical providers treating patients with metastatic spine disease. There is limited research studying SINS correlation with rates of failure after radiation therapy. The goal of this study was to evaluate risk factors for failure of radiation therapy. We hypothesized that a higher SINS would correlate with failure.

Materials/Methods

We performed an IRB approved retrospective cohort study at a tertiary academic center. All patients with spinal metastasis being treated with radiation between September 2014 and October 2018 were identified. Pediatric patients and patients with myeloma, leukemia, lymphoma, sarcoma, cord compression, history of prior radiation or surgery, or inadequate follow up were excluded. Baseline demographics were recorded and the SINS calculated. variables analyzed were primary tumor, Karnofsky and ECOG scores, time to treatment, dosage, and type of radiation. The outcome was radiation therapy failure as defined by persistent pain, need for re-irradiation, or surgical intervention. Chi squared and fisher exact tests were used for analysis of the categorical variables. Continuous variables were analyzed with the student t-test. The multivariate regression model was built from the univariate analysis.

Results

583 patients were identified, of those 170 met the inclusion criteria. The median follow up was 218 days. 43 patients failed radiation therapy, 10 required repeat radiation, and 7 underwent surgery. 36 reported no pain relief, including some that required re-irradiation and surgery. Significant risk factors on univariate analysis were SINS grouping (<6, 7-12, >12) (p = 0.038), percent vertebral involvement (p = 0.002), biologically effective dose (BED) <43 (0.047), categorical Karnofsky (<50, 50-70, >80) (p = 0.003), continuous Karnofsky (p = 0.001), and ECOG (0-2, 3-4) (p = 0.025). The multivariate regression model had an AUC of 0.691 (95% CI: 0.605 – 0.777) a sensitivity of 0.8140 and a specificity of 0.5197. Adjusted odds ratios for the variables are shown in Table 3.

Discussion/Conclusion

Worse performance scores, lower BED, higher SINS and vertebral involvement were associated with radiation failure on univariate analysis. High SINS, increased Karnofksy score and male gender were predictive variables for radiation failure in the multivariate model.

Paper 44. Validating Patient Health Questionnaire-9 with Global Outcomes in Minimally Invasive Lumbar Discectomy

Parrish James, MPH1, Jenkins Nathaniel, MS1, Massel Dustin, MD2, Rush III Augustus, MD2, Brundage Thomas, BS1, Hrynewycz Nadia, BS1, Singh Kern, MD1

1Rush University Medical Center, IL, Illinois, United States, 2University of Miami Miller School of Medicine, Miami, Florida, United States

Background/Introduction

The Patient Health Questionnaire-9 (PHQ-9) is a widely used screening tool for evaluating depressive symptoms. The Short Form-12 (SF-12) and Veterans RAND-12 (VR-12) surveys have a longitudinal history of assessing both physical and mental outcomes after spine surgery. Research is scarce validating PHQ-9 scores with more rigorously tested mental health-focused patient-reported outcomes (PROs). This study aims to validate PHQ-9 as an appropriate metric for assessing mental health in patients undergoing minimally invasive Lumbar Discectomy (MIS LD).

Materials/Methods

A prospectively maintained surgical registry was retrospectively reviewed for patients who underwent single-level, elective MIS LD procedures for degenerative pathology from March 2016 until May 2019. Patients were excluded if they had not completed preoperative surveys. Demographic and baseline perioperative variables were collected. A Pearson’s correlation coefficient and time-independent partial correlation coefficient were used to evaluate the relationship of PHQ-9 with SF-12 MCS and VR-12 MCS scores (Table 1). Correlation strength was assessed by the following categories: 0.1 ≤ |r| < 0.3 = low; 0.3 ≤ |r| < 0.5 = moderate; |r| ≥ 0.5 = strong.

Results

A total of 239 patients underwent single-level MIS LD. The cohort was 32% female with an average age was 42.4 years, and 61% of the cohort was obese. Patient mean scores demonstrated statistically significant improvement by the VR-12 MCS and SF-12 MCS surveys compared to preoperative scores at all evaluated postoperative timepoints (p≤0.001 for 6-week, 12-week, 6-month and 1-year). When evaluating with Pearson’s correlation coefficients and with a partial correlation time-controlled coefficient, SF-12 MCS and VR-12 MCS were each observed to have strong correlations with PHQ-9 at preoperative, and each postoperative timepoint (Table 1).

Discussion/Conclusion

We observed that PHQ-9, SF-12 MCS and VR-12 MCS scores all have statistically significant improvement after patients undergo LD. PHQ-9 scores were strongly correlated with SF-12 MCS and VR-12 MCS scores when evaluated with both Pearson’s correlation coefficients and with partial correlation coefficients. Our results substantiate evidence from other surgical fields that PHQ-9 scores can be used as a valid tool to evaluate pre and postsurgical depressive symptoms.

Paper 45. Surgeon learning curve and clinical outcomes of minimally invasive anterior lumbar interbody fusion with posterior percutaneous instrumentation

Mirza Muhammad, MD1, Olson Sydney, BS2, Panthofer Annalise, BS2, Matsumura Jon, MD2, Williams Seth, MD3

1University of Missouri-Columbia, Columbia, Missouri, United States, 2University of Wisconsin-Madison, Madison, Wisconsin, United States, 3University Of Wisconsin-Madison, Department of Orthopedics and Rehabilitation, Madison, Wisconsin, United States

Background/Introduction

Minimally invasive anterior lumbar interbody fusion (mini-ALIF) enables indirect decompression, restoration of sagittal alignment, and reduction of spondylolisthesis while minimizing morbidity and blood loss. Mini-ALIF is associated with potentially life-threatening vascular injuries that occur while obtaining retroperitoneal access to the spine. The purpose of this study is to report the surgeon learning curve and complications associated with the initial learning phase of the mini-ALIF approach.

Materials/Methods

An IRB approved retrospective review of 138 patients who underwent mini-ALIF from 2010-2018 was conducted. All anterior approaches were performed by the same vascular access surgeon, with disc preparation by 3 different spine surgeons. Inclusion criteria was patients >18 years old undergoing a primary anterior retroperitoneal approach for the lowest two segments of the lumbar spine (i.e. L4-5, L5-S1). 127 patients met the criteria. Age, sex, BMI, ASA score, comorbidities (diabetes, PVD, nicotine use), diagnosis, length of stay, estimated blood loss, duration of procedure, vascular injuries and postoperative complication rates were obtained.

Results

There was no significant decrease in total operative time over the years. Total blood loss during the surgery significantly decreased until around case 30 before stabilizing. Mean EBL was 184 mL for L5/S1, 232 mL for L4/L5 and 458 mL for two-level ALIFs. The major vascular injury rate was 9.4% (12/127). Rate of vascular injuries declined over the course of the study with the highest rate of injury in the first 25 cases (32%). The post-operative complication rate also steadily declined with the highest rate in the first 25 cases (7 out of 21). There were 12 vascular injuries in patients with a BMI <35 (104 patients) and 8 with BMI >35.1 (23 patients). Odds ratio of vascular injury with BMI >35.1 was significant at 4.09 (1.4-11.7 CI, P = <0.008). Total operative time, rate of vascular injury and post-operative complications steadily increased with increasing BMI.

Discussion/Conclusion

The mini-ALIF approach for lumbar fusion is associated with a steep learning curve. About 25-30 cases are needed before the intra-operative and post-operative complication rates stabilize. Caution is warranted in operating on BMI >35, particularly in the initial learning phase.

Poster 01. Retrospective Analysis of Lumbar Radiculopathy Treatment: Do Care Pathways Emphasizing Conservative Measures Really Work?

Truumees David1, Duncan Ashley, RN, MBA2, Singh Devender, PhD2, Truumees Eeric, MD2

1Alamo Heights High School, San Antonio, Texas, United States, 2Ascension Texas Spine and Scoliosis Center, Austin, Texas, United States

Background/Introduction

Emphasis on exhausting all conservative treatments before resorting to surgical intervention is becoming more ubiquitous, especially when seeking authorization for invasive procedures from health insurance payors. We analyzed the treatments and outcomes, along with perception of improvement for lumbar radiculopathy patients who presented to an interdisciplinary spine clinic, which emphases conservative care first.

Materials/Methods

Institutional review board approved chart audits were conducted on patients treated in 2018 for lumbar radiculopathy. Demographics, number of visits, prescribed therapies, if epidural steroid injection was recommended, how many injections were performed, referral to surgeon and if surgery was performed was aggregated and analyzed. Patients were then contacted via phone and email and asked what percent improvement they felt was gained from the prescribed therapies.

Results

628 charts were audited, and attempted contacts made, with a 32% patient response rate (n=200). Injection therapy was recommended for 71.5% of patients, with the mean number of injections performed at 2.2±1.7, with an overall improvement rate of 53±38.5%. All age groups were prescribed other treatments, 48% receiving a combination of prescribed therapies. Global surgery rate was 15.1%, significantly below the national surgical intervention average of 33%. This below average surgical rate is associated with a cost avoidance upwards of $536,000 for the fiscal year. Overall, 65% of patients reported improvement from the treatments and therapies prescribed.

Discussion/Conclusion

A multifocal conservative approach including epidural steroid injections provides significant improvement for lumbar radiculopathy. A conservative approach to spine care prevents unnecessary surgeries and promotes care pathways that provide quality care with fiscal responsibility.

Poster 02. Effect of State Bill 273 Opioid Prescribing Limits in West Virginia

Sedney Cara, MD1, Haggerty Treah, MD1, Dekeseredy Patricia, RN2, Wood Nathan, MPH3

1West Virginia University School of Medicine, Morgantown, West Virginia, United States, 2West Virginia University, Morgantown, West Virginia, United States, 3West Virginia Board of Pharmacy, Charleston, West Virginia, United States

Background/Introduction

West Virginia leads the country in morbidity and mortality related to opiate use disorders (OUD). On March 27, 2018, new state opioid prescribing limits (SB 273) were enacted, significantly curtailing the ability to prescribe opioids for acute pain, for example in the post-operative period after spine surgery. This stands to significantly impact the availability of opioid pain medication to spine surgery patients; however, the population-level effects and success of this legislation in reducing community availability of opioids are not known.

Materials/Methods

An interrupted time series analysis was done utilizing opioid prescription data from the West Virginia Board of Pharmacy to compare prescribing patterns for the five month period before and after the enactment of SB 273. Generalized linear models were utilized, controlling for pre-intervention trends for both state and county level data. A one month “washout” was period was utilized. Variables including the total number of opiate prescriptions, first time opiate prescriptions, and quantity/day supply were selected due to their direct relation to the required components of SB 273.

Results

Though overall and first time opioid prescriptions decreased during the study period, the rate of change following the implementation of the bill was not significant when compared to the pre-intervention rate of change. However, there was a statistically significant decrease in days’ supply of medication at the state and county levels. County level data regarding the days’ supply per prescription largely followed the state trend towards decreasing amounts per prescription, with most counties having a significant decrease in amount per prescription after enactment of the law.

Discussion/Conclusion

Overall, the number of opioid prescriptions is decreasing in West Virginia, but the rate of change was not significantly affected by SB 273. The adoption of SB 273 appears to have effected a significant change in the days’ supply of prescribed opioids but not on overall prescription rates. This may impact patients after spine surgery. Regarding the overall decrease in number of prescriptions, there may be other societal factors at play, requiring more in-depth investigation to quantify effects.

Poster 03. The Impact of Surgical Chronology on Outcomes of Patients Receiving Lumbar Spine and Lower Extremity Joint Surgeries

Atli Karam, MS1, Smith Gabriel, MD2, Knusel Konrad, MS1, Benzel Edward, MD1, Krishnaney Ajit, MD1, Steinmetz Michael, MD3, Mroz Thomas, MD1

1Cleveland Clinic Foundation, Cleveland, Ohio, United States, 2University Hospitals Cleveland Medical Center, Cleveland, Ohio, United States, 3Center for Spine Health, Cleveland Clinic Foundation, Cleveland, Ohio, United States

Background/Introduction

The aging population and obesity epidemic indicate that increasingly more patients with lumbar spine pathology will likely also develop lower extremity joint (LEJ) disease (hip and/or knee). These patients may have co-existing lower back and extremity pain, the major source of which (spine vs. LEJ) is difficult to determine. There remains a lack of outcomes data of patients who receive surgeries for both an LEJ and the lumbar spine within a short period based on chronology of surgery type. This is the first report that evaluates the differences in surgical outcomes of patients receiving lumbar surgery following LEJ surgery compared to receiving lumbar surgery before LEJ surgery.

Materials/Methods

Retrospective chart review was performed of patients undergoing lumbar surgery between 2008-2015 and receiving hip/knee surgery within three years prior/ following lumbar surgery at a single institution. Pre- and post-operative outcome measures were assessed using the EuroQol five dimensions questionnaire (EQ5D) and Pain Disability Questionnaire (PDQ).

Results

670 patients were included. Patients receiving lumbar surgery first were 4.75 times more likely to reach the minimal important difference in PDQ. Patients receiving LEJ surgery first had a higher mean Charlson Comorbidity Index. Male patients were 37% less likely to be readmitted within 90 days of either procedure.

Discussion/Conclusion

Postoperatively, patients who received lumbar surgery before LEJ surgery had greater improvement than patients who received LEJ surgery before lumbar surgery. These findings suggest that in patients with a co-existing need for lumbar and LEJ surgery, greater consideration should be given to providing lumbar surgery before LEJ surgery. The findings of this study are applicable to multiple disciplines including primary care, orthopedics, and neurosurgery, and warrants further large-scale prospective investigation to determine the cause and generalizability of these surgical outcomes results.

Poster 04. Clinical Significance of Lateral Pedicle Screw Malposition in the Lumbar Spine

Smith Isabel, BS1, Golenbock Samuel, MS1, Hwang Raymond, MD, MEng, MBA1, Lee Gyu Ho, MA2, Fischgrund Jeffrey, MD3, Baker Kevin, PhD3, Park Daniel, MD3, Arnold Paul, MD4, Sasso Rick, MD5, Kim David, MD1

1New England Baptist Hospital, Boston, Massachusetts, United States, 2Columbia University, New York City, New York, United States, 3William Beaumont Hospital, Royal Oak, Michigan, United States, 4Carle Neuroscience Institute, Urbana, Illinois, United States, 5Indiana Spine Group, Carmel, Indiana, United States

Background/Introduction

The risks of pedicle screw malposition have largely focused on medial malposition and neurological injury. Although lateral malposition is considered relatively benign, no previously reported study has focused specifically on clinical outcomes associated with lateral screw malposition. We hypothesize that lateral pedicle screw breach has a negative effect on fusion rates and clinical outcomes following elective single-level posterior lumbar fusion (PLF).

Materials/Methods

Retrospective review of a prospective multicenter cohort of patients undergoing single-level instrumented PLF with local autograft bone for degenerative disease. 12-month postoperative computed tomography (CT) was reviewed for breach direction, severity of screw-associated pedicle breach and fusion status. Patients with lateral breach were compared to patients with no breach. Patients with non-lateral screw malposition were excluded. Outcomes measures included Oswestry Disability Index (ODI), Numerical Pain Rating Scale (NPRS) for back pain, and SF-36 physical function. Improvement in back pain was defined as a ≥35% change (minimal clinically important difference) at follow-up. Outcomes were analyzed using multivariable linear and logistic regression and were adjusted for age, procedure, level, and/or baseline pain score.

Results

154 patients underwent 12-month postoperative CT. 45 patients (30.6%) demonstrated one or more lateral breaches and were compared to 102 patients with no breach. After adjusting for baseline scores and fusion level, patients with one or more breaches had, on average, ODI score decreases that were 2.44 points less (p=0.454) and SF-36 score increases that were 2.04 points less (p=0.040) than patients with no lateral breach. Breach was also associated with a large decrease in the odds of achieving back pain relief (OR=0.37, 95%CI: 0.15, 0.91) after adjusting for baseline NPRS. There was no observed effect of breach on the odds of non-union (multiple breaches OR=1.97, 95% CI: 0.66, 5.97).

Discussion/Conclusion

The current study did not observe an association between laterally malpositioned pedicle screws and nonunion or disability scores. However, these results are consistent with a modest negative effect on physical functioning scores and self-reported back pain. This suggests that surgeons may anticipate poorer clinical outcomes among patients with laterally malpositioned screws.

Poster 05. Osteoporosis is Common and Undertreated Prior to Thoracolumbar Fusion

Bernatz James, MD1, Winzenried Alec, BS2, Mikula Anthony, MD3, Binkley Neil, MD4, Anderson Paul, MD4

1University of Wisconsin Department of Orthopedic Surgery, Madison, Wisconsin, United States, 2University Of Wisconsin-Madison, Department of Orthopedics and Rehabilitation, Madison, Wisconsin, United States, 3Mayo Clinic Rochester, Rochester, Minnesota, United States, 4University of Wisconsin-Madison, Madison, Wisconsin, United States

Background/Introduction

The prevalence of osteoporosis in thoracolumbar spine fusion patients is substantial and osteoporosis may portend poorer clinical outcomes. The implementation of a fracture liaison service (FLS) has been shown to improve bone health screening and treatment in the geriatric hip fracture population. In 2017 we established a FLS and in the last 18 months have expanded this service to preoperative bone health assessment. We hypothesize that preoperative osteoporosis is common, under-recognized and undertreated in thoracolumbar fusion patients. The purpose of this study is to determine if the FLS clinic increased preoperative osteoporosis screening rates and pharmacologic osteoporosis treatment as warranted.

Materials/Methods

This retrospective case series includes adults age 40+ who underwent elective thoracolumbar spine fusion (single or multiple levels) at a single tertiary care center prior to and after creation of a BHO referral clinic. Data collected included preoperative osteoporosis risk factors, prior dual-energy x-ray absorptiometry (DXA) testing, and prior osteoporosis pharmacotherapy. Fracture risk was estimated using the Fracture Risk Assessment Tool (FRAX) with and without bone mineral density (BMD) and the U.S. National Osteoporosis Foundation (NOF) criteria for screening and treatment were applied.

Results

Ninety-two patients were included in the pre-FLS group; fifty-three (58%) met criteria for osteoporosis screening and sixteen (17%) met criteria for osteoporosis pharmacotherapy. There were 117 patients in the post-FLS group; forty-three (37%) met criteria for osteoporosis screening and 10 (9%) met criteria for osteoporosis pharmacotherapy. After implementation of the FLS referral clinic, appropriate pre-operative screening increased from 19% to 42% and pre-operative osteoporosis treatment (when indicated) increased from 31% to 70% (table 1).

Discussion/Conclusion

Poor bone health is common in elective thoracolumbar fusion patients. The percentage of adults undergoing thoracolumbar spine fusion meeting criteria for osteoporosis pharmacotherapy is estimated to be 9-17%. A quality improvement program based on the concepts of the FLS that incorporates pre-operative bone health assessment can provide screening and osteoporotic care in selected patients.

Poster 06. Cost Comparison of Single Level Posterior and Anterior/Lateral Lumber Interbody Fusions

Gandhi Shashank, MD1, Virojanapa Justin, MD1, Latefi Ahmad, MD1

1Northwell Department of Neurosurgery, Manhasset, New York, United States

Background/Introduction

Some patients with single-level lumbar degenerative disc disease (LDDD) may be candidates for either posterior, anterior, or lateral lumbar interbody fusions. This study aims to compare costs and hospital metrics between posterior/transforaminal (PLIF/TLIF) and anterior/lateral lumbar (ALIF/LLIF) interbody fusions.

Materials/Methods

All patients from 2017-2019 at this institution who underwent PLIF/TLIF or ALIF/LLIF were assessed. Patient specific hospital metrics including post-anesthesia care unit (PACU) length of stay (LOS), intensive care LOS, hospital LOS, foley-catheter days, discharge disposition, use of assistive devices, 30-day-readmission rates, operating room cost, and hospital costs were assessed for each group.

Results

314 patients were assessed: 220 in the PLIF/TLIF group and 94 in ALIF/LLIF group. There was no difference in gender between the groups. The patients in the ALIF/LLIF group were younger (means 54.9 versus 60.2 years, p=0.012). The PLIF/TLIF group had significantly shorter mean foley-catheter days (2.3 versus 2.9 days, p=0.030). The PLIF/TLIF group compared to ALIF/LLIF had no significant difference in PACU LOS (0.35 versus 0.32 days, p=0.389), ICU LOS (3.7 versus 1.3 days, p=0.137), or hospital LOS (6.2 versus 6.8 days, p=0.406). There was similar postoperative use of assistive devices for ambulation (72.3% PLIF/TLIF versus 75.7% ALIF/LLIF, p=0.533). There was no difference in discharge to rehabilitation facility rates (24.5% PLIF/TLIF versus 20.2% ALIF/LLIF, p=0.405). 30-day-readmission rates were similar (10.9% PLIF/TLIF versus 9.6% ALIF/LLIF, p=0.724). However, PLIF/TLIF group had significantly lower operating room cost ($18,504 versus $23,775; p<0.001) and hospital cost ($74,350 versus $90,629; p=0.001).

Discussion/Conclusion

In the immediate 30-day postoperative period, PLIF/TLIF are more cost-effective and have better value than ALIF/LLIF with lower operating room and hospital costs and fewer foley-catheter days resulting in similar length of stay metrics, discharge to rehabilitation facility, and 30-day-readmission rates for single-level LDDD. Long-term data is needed to quantify outcomes and complications such as pseudoarthrosis and adjacent segment disease.

Poster 07. The Impact of Long-term Corticosteroid Use on Acute Postoperative Complications Following Lumbar Decompression Surgery

Tihista Mikel, MD1, Gu Alex, BS1, Wei Chapman, BS1, Weinreb Jeffrey, MD1, Rao Raj, MD1

1George Washington University Department of Orthopaedic Surgery, Washington, District of Columbia, United States

Background/Introduction

Studies have shown that patients on long-term corticosteroids have higher infection rates. However, the rates of infection and other complications following lumbar decompression surgery remains under-investigated. The aim of our study was to determine the impact of preoperative long-term corticosteroid usage on acute, 30-day postoperative complications in a subset of patients undergoing lumbar spine decompression surgery, without fusion or instrumentation. We hypothesize that patients on long-term corticosteroids will have higher rates of infection and other postoperative complications after undergoing lumbar decompression surgery of the spine.

Materials/Methods

A retrospective cohort study was conducted using data collected from the ACS National Surgical Quality Improvement Program (NSQIP) database data from 2005-2016. Lumbar decompression surgeries, including discectomies, laminectomies, and others were identified using CPT codes. Chi-square analysis was used to evaluate differences among the corticosteroid and non-corticosteroid groups for demographics, preoperative comorbidities, and postoperative complications. Logistic regression analysis was done to determine if long-term corticosteroid use predicts incidence of postoperative infections following adjustment.

Results

26,734 subjects met inclusion criteria. 3.9% were on long-term corticosteroids prior to surgical intervention, and 96.1% were not. Patients who were on long-term corticosteroids were more likely to be older (p<0.001), female (p<0.001), nonsmokers (p<0.001), and have a higher American Society of Anesthesiologist class (p< 0.001). Multivariate analysis demonstrated that long-term corticosteroid usage was associated with increased overall complications (odds ratio [OR]: 1.543; p < 0.001), and an independent risk factor for the development of minor complications (OR: 1.808; p < 0.001), urinary tract infection (UTI) (OR: 2.033; p = 0.002), extended length of stay (LOS) (OR: 1.244; p = 0.039), thromboembolic complications (OR: 1.919; p = 0.023), and sepsis complications (OR: 2.032; p = 0.024).

Discussion/Conclusion

Long-term corticosteroid usage is associated with a significant increased risk of acute postoperative complication development, including UTI, sepsis and septic shock, thromboembolic complications, and extended LOS, but not with superficial or deep in patients undergoing lumbar decompression procedures. Spine surgeons should remain vigilant regarding postoperative complications in patients on long-term corticosteroids.

Poster 08. Comparison of Ex Vivo Bone Morphogenetic Protein-2 (rhBMP-2)-Priming of Culture-Expanded Stem Cells Between Adipose-Derived and Bone Marrow-Derived Cells in a Posterolateral Lumbar Spinal Fusion Model

Perdomo-Pantoja Alexander, MD1, Holmes Christina, PhD2, Cottrill Ethan, MS1, Ishida Wataru, MD1, Lo Sheng-Fu, MD1, Witham Timothy, MD1

1Johns Hopkins University School of Medicine, Baltimore, Maryland, United States, 2Florida A&M University-Florida State University College of Engineering, Tallahassee, Florida, United States

Background/Introduction

Clinical delivery of rhBMP-2 often requires supraphysiological concentrations and is associated with complications. Therefore, the development of solutions for having successful outcomes of BMP-2 without undesired side effects is needed. It has previously reported that a short period of in vitro treatment with rhBMP-2 prior to transplantation enhanced bone marrow-derived stem cells (BMSC)-mediated bone regeneration in vivo in a calvarial defect model. In this study, we aimed to examine whether a brief period of in vitro priming with rhBMP-2 can enhance MSC-mediated fusion and to compare the response of adipose-derived (ADSC) and bone marrow-derived cells in a rat model.

Materials/Methods

We isolated ADSC and BMSC from syngeneic rats and cultured in vitro. When passage 1 (P1) ADSCs and BMSCs reached 80% confluency, they were primed for 24 hours with 1 ng of rhBMP-2 (Medtronic). After priming, 2x10^6 MSCs (P2) were seeded onto Vitoss (Stryker) bone graft substitute scaffolds for subsequent transplantation. Posterolateral intertransverse spinal fusion surgery at L4-L5 was performed on rats divided into 4 experimental groups: [1] Vitoss + rhBMP-2-Primed BMSCsP2; [2] Vitoss + non-primed BMSCsP2; [3] Vitoss + rhBMP-2-Primed ADSCsP2; and, [4] Vitoss + non-primed ADSCsP2. Fusion was assessed eight weeks post-surgery via micro-computed tomography (microCT) imaging, manual palpation, and histology.

Results

MicroCT imaging data indicate that culture-expanded rhBMP-2-primed ADSCs and BMSCs displayed significantly higher fusion mass volumes than non-primed ADSCsP2 and BMSCsP2, respectively (29.94 mm^3 vs. 25.56 mm^3, respectively, p=0.04; and 23.8 mm^3 vs. 17.79 mm^3, respectively, p=0.06). However, fusion mass volumes were not significantly different between primed ADSCsP2 and BMSCsP2. CT fusion scores were not significantly different between primed and non-primed groups. Primed ADSCsP2 also yielded higher manual palpation scores than primed ADSCsP2 (1.9 vs. 1.3, respectively, p=0.03).

Discussion/Conclusion

Preliminary results suggested that MSC BMP-2-priming enhances spinal fusion in our rat model. rhBMP-2-priming allowed an increased fusion mass volume and manual palpation score among ADSCsP2 compare to non-priming. rhBMP-2 primed BMSCsP2 showed a trend to a higher fusion mass volume compared to non-primed BMSCs. Ongoing histological studies will evaluate the characteristics and quality of the new bone formed within the fusion masses.

Poster 09. Efficacy of Isolated Local Bone Autograft for Posterolateral Lumbar Fusion

Goldstein Zachary, MD1, Boody Barrett, MD2, Vinayek Sheetal, BS2, Lee Evan, BS3, Sasso Rick, MD2, Smucker Joseph, MD2

1Indiana University School of Medicine, Indianapolis, Indiana, United States, 2Indiana Spine Group, Carmel, Indiana, United States, 3University of Pennsylvania, Philadelphia, Pennsylvania, United States

Background/Introduction

Local autogenous bone graft is frequently used during posterolateral lumbar fusion. Surgeons will also commonly add allograft or synthetic bone graft substitute as a bone graft extender due to the limited volume of local bone graft obtained. However, minimal data is available that rigorously reviews the fusion rates with CT scans using isolated local bone graft for posterolateral lumbar fusions.

Materials/Methods

Patients were obtained from a single site of a multi-center trial studying the efficacy of low intensity pulsed ultrasound on fusion rates. Thirty-six patients were initially enrolled, presenting with symptomatic stenosis failing conservative treatment requiring decompression and posterior instrumented fusion. Patients underwent blinded randomization into treatment (functional bone stimulator) and control (nonfunctioning bone stimulator) groups. Both groups received only autogenous local bone graft. CT scans were obtained at 6 and 12-months postoperatively to evaluate fusion. Three independent reviewers evaluated the CT scans using previously published metrics for determining fusion status and quality (Singh et al. 2006).

Results

The study was terminated by the sponsor prior to completion and multi-center data is unavailable. However, we were able to achieve completion of CT scans at 12 months in 22 patients, with 5/12 control and 2/10 treatment patients displaying evidence of definite fusion on CT scan at 12 months. No significant difference was found between treatment groups regarding fusion rates (p-value=0.27572). Overall 7/22 (31.8%) of patients demonstrated definite fusion at 12-month postoperative CT scan.

Discussion/Conclusion

Despite small sample size due to early study termination by the sponsor, the available data is unique due to the standardized use of advanced imaging to evaluate of fusion for isolated local bone grafting for posterolateral fusion. At both 6- and 12-month post-surgical intervals, CT-graded arthrodesis in a single-level instrumented lumbar posterolateral fusion with only local bone graft demonstrates a low rate of radiographic success. The application of externally low intensity pulsed ultrasound did not appear to favorably modify the arthrodesis rate. This available data adds to the current literature suggesting isolated autogenous local bone grafting for posterolateral fusions is insufficient and may require the use of bone graft extenders.

Poster 10. Patients with Bilateral Sacroiliac Joint Vacuum Sign have a Change in Pelvic Incidence between Supine and Standing

Mikula Anthony, MD1, Fogelson Jeremy, MD1, Oushy Soliman, MD1, Pinter Zachariah, MD1, Peters Pierce, MD1, Polly, Jr David, MD2, Elder Benjamin, MD, PhD1

1Mayo Clinic Rochester, Rochester, Minnesota, United States, 2University of Minnesota, Department of Orthopaedic Surgery, Minneapolis, Minnesota, United States

Background/Introduction

Pelvic incidence (PI) is an increasingly utilized spinopelvic parameter in the evaluation of spinal deformity patients and is thought to be a fixed parameter. The purpose of our study was to determine if patients with sacroiliac (SI) joint vacuum sign have a change in PI between the supine and standing position.

Materials/Methods

A retrospective chart review identified patients with a standing x-ray, supine x-ray, and CT encompassing the SI joints within six months. Patients were grouped according to their SI joints having either no vacuum sign, unilateral vacuum sign, or bilateral vacuum sign. PI was measured by two independent reviewers.

Results

Seventy three patients were identified with an average age of 66 years and a BMI of 30. Patients with bilateral SI joint vacuum sign (N=27) had an average absolute change in PI of 7.2 degrees (p-value = 0.045) between the standing and supine position compared to unilateral SI joint vacuum sign (N=20) patients who changed 5.2 degrees (p-value = 0.25) and patients without an SI joint vacuum sign (N=26) who changed 4.1 degrees (p-value = 0.90). Analysis of variance Tukey test showed a statistically significant difference in the change in PI between the bilateral SI joint vacuum sign patients and those without an SI joint vacuum sign (p-value = 0.023). Intraclass correlation coefficient between the two reviewers was 0.97 for standing PI and 0.96 for supine PI (p-values <0.0001).

Discussion/Conclusion

Patients with bilateral SI joint vacuum sign had a change in PI suggesting that there may be substantial motion across the SI joint between the standing and supine position.

Poster 11. Drivers of Cost in Primary Single-Level Lumbar Fusion Surgery

Hwang Raymond, MD, MEng, MBA1, Golenbock Samuel, MS1, Kim David, MD1

1New England Baptist Hospital, Boston, Massachusetts, United States

Background/Introduction

Allocating cost associated with resources utilized during healthcare delivery is challenging with traditional hospital accounting. Time-driven activity-based costing (TDABC) is an efficient method to accurately assign cost that has been well described in the management and health economics literature. TDABC allows the determination of difficult to quantify costs, leading to a more informative cost structure. The goal of this study was to characterize the variation in total hospital cost (THC) of lumbar fusion procedures and identify potential cost-saving opportunities.

Materials/Methods

Retrospective review of adult patients undergoing primary single-level lumbar fusion surgery for degenerative disease at a single institution from January 1, 2015 through December 31, 2017. Process maps were developed for preoperative, intraoperative and postoperative care. THC was composed of implant, medications, other supplies (e.g., gauze), and personnel costs and calculated using Avant-garde Health (Boston, MA). Linear regression and descriptive statistics were used to analyze THC variation.

Results

727 cases and 12 surgeons were analyzed. Approximately 50% of THC variation was associated with procedure choice alone. Patient characteristics explained 10% of THC variation. Implants (including biologics) accounted for 45% of cost, surgery personnel 27%, inpatient personnel 16%, medication and non-implant supply 8%. With reference to instrumented posterolateral fusion (PLF), THC ranged from 0.6x (in situ fusion, ISF) to 1.7x (lateral interbody fusion, LLIF). Implant cost variation ranged from 2.5x reference (LLIF) to 0.1x (ISF). There was a 1.7x difference between the highest THC surgeon and the lowest. The fusion type with the highest THC variation was transforaminal interbody fusion (TLIF). The surgeon with the highest TLIF THC was 1.5x more expensive than the surgeon with the lowest.

Discussion/Conclusion

Surgeon-based choices (fusion technique, implants) have the greatest effect on THC variation and represent the largest opportunities for cost savings. Primary single-level lumbar fusion THC is driven primarily by fusion type. Implants, including biologics, account for nearly half this cost. Future work will incorporate outcomes data to characterize the differential value conferred by fusion types with higher THC.

Poster 12. Femoral Head to Lower Lumbar Neural Foramen Distance and Postoperative Neuropraxia in Adult Degenerative Spine Patients

Ha Alex, MD1, Park Paul, MD1, Mathew Justin, MD2, Lombardi Joseph, MD1, Cerpa Meghan, MPH2, Lenke Lawrence, MD1

1New York Presbyterian - Columbia, New York, New York, United States, 2Columbia University Medical Center, New York, New York, United States

Background/Introduction

Lumbosacral fractional curves in adult spinal deformity (ASD) patients often have sharp coronal curves that result in significant pain and imbalance. Postoperative neuropraxia after fractional curve correction can lead to discomfort and unsatisfactory clinical outcomes. The goal of this study is to understand the relationship between postoperative neuropraxia and fractional curve correction using pre and postoperative radiographic measures

Materials/Methods

A radiographic review was performed on 66 ASD patients from July 2015-July 2018. The distance between the lower lumbar neural foramen (L4 and L5) to the ipsilateral center femoral head (FH-L4 and FH-L5) in the pre and postop setting were measured and compared (ΔFH-L4/L5). The largest distance in absolute value of ΔFH-L4/L5 on either the convexity or the concavity of the curve was used for the analysis. Logistic regression and Chi-square analyses were performed to study the relationship between the FH-L4/L5 and postop neuropraxia

Results

Of the 66 patients, 15 patients had postoperative neuropraxia. Patients without postop neuropraxia had an average ΔFH-L4 distance of 15.7mm compared to those with neuropraxia who had a ΔFH-L4 distance of 33.1mm (p<0.01). Patients without postop neuropraxia had an average ΔFH-L5 distance of 11.0mm compared to those with neuropraxia who had a ΔFH-L5 distance of 24.5mm (p<0.01). Chi-square analysis showed a 5.56-fold risk of developing neuropraxia if the ΔFH-L4 >20mm (RR- 5.56; p<0.01), and a 2.24-fold risk of developing neuropraxia if the ΔFH-L5 >20mm (RR- 2.24; p<0.01). Logistic regression analysis identified the odds of developing neuropraxia with a change in ΔFH-L4 >20mm of 19:1 (CI:3.8-95.7) and ΔFH-L5 >20mm of 37:1 (CI:6.4-211.7)

Discussion/Conclusion

The novel ΔFH-L4/5 distances can predict the chances of postoperative neuropraxia in ASD patients. The ΔFHL4/L5>20mm significantly increases the odds for patients to develop postoperative neuropraxia.

Poster 13. Preclinical Evaluation Of A Novel 3D-printed Composite Scaffold For Spinal Fusion

Nandurkar Tejas, MA1, Paul Jonathan, MPH1, Plantz Mark, BS1, Lyons Joseph, BS1, Marsh Parker, BS1, Minardi Silvia, PhD1, Yun Chawon, PhD1, Stock Stuart, PhD1, Hsu Wellington, MD1, Hsu Erin, PhD1

1Northwestern University Feinberg School of Medicine, Chicago, Illinois, United States

Background/Introduction

The majority of currently available treatments for spine fusion are inadequate alone, while recombinant human bone morphogenetic protein-2 (rhBMP-2), although capable of eliciting high rates of fusion as a stand-alone option, is associated with serious complications. To address these limitations, we have recently developed a novel 3D-printed hydroxyapatite (HA)/PLGA/demineralized bone matrix (DBM) scaffold as a growth factor-free bone graft substitute. We optimized the scaffold’s composition and architecture and herein evaluated the scaffold for efficacy, safety, and pro-osteogenic mechanism of action.

Materials/Methods

Experiments utilized the rat L4-5 posterolateral spine arthrodesis model, with 10 µg rhBMP-2 delivered on an absorbable collagen sponge (ACS) serving as the comparative control. Fusion efficacy was evaluated with plain radiography, manual palpation, and biomechanical testing (n=50). Additionally, rats underwent lumbar MRI and local soft tissue edema volume was measured as an index of inflammation (n=60). Mechanisms of action were investigated with temporal gene and protein expression studies (n=90).

Results

Fusion rates were not significantly different between the scaffold (93%) and rhBMP-2 (100%); however, fusion scores were greater with rhBMP-2 (p=0.008). Both treatments resulted in significantly reduced segmental range of motion (ROM; p<0.001) and greater stiffness (p=0.009) when compared with non-operated controls; this effect was greater with rhBMP-2 (L4-5 flexion-extension ROM: non-operated control: 28°; scaffold: 15°; rhBMP-2: 4°). Patterns of gene expression appeared similar between groups, with both rhBMP-2 and scaffold treatments resulting in significant elevations of several pro-osteogenic markers including Alp, Osx, and Runx2. Significantly greater inflammatory volume was observed on MRI in the rhBMP-2 group at 2 days postoperatively (p<0.001), with no differences noted thereafter.

Discussion/Conclusion

Our scaffold demonstrated success at eliciting spine fusion, yielding fusion rates comparable to those generated by rhBMP-2. While both treatments resulted in similar patterns of pro-osteogenic gene expression, our scaffold did not evoke the host inflammatory response seen with rhBMP-2 treatment. These results suggest that this novel biomaterial has promise as a bone graft substitute for use in spinal surgery, with potential to eliminate the requirement for and adverse effects of growth factor use.

Poster 14. Rate, Reasons, and Cost of Emergency Department Visits Surrounding Lumbar Discectomy

Burroughs Patrick, BS1, Moore Harold, BS2, Sabatino Matthew, MD3, Varthi Arya, MD4, Grauer Jonathan, MD3

1Yale School of Medicine, New Haven, Connecticut, United States, 2Weill Cornell Medical College, New York City, New York, United States, 3Yale School of Medicine Department of Orthopaedics and Rehabilitation, New Haven, Connecticut, New Haven, Connecticut, United States, 4Yale New Haven Hospital, Department of Orthopaedics and Rehabilitation, New Haven, Connecticut, United States

Background/Introduction

Lumbar discectomy is one of the most common spinal procedures in the United States. Although readmissions after such procedures have been studied, Emergency Department (ED) visits surrounding such procedures remain poorly studied.

Materials/Methods

The large insurance claims PearlDiver (Warsaw, IN, USA) 2007-2017 Humana Orthopaedics database was utilized. The database contains approximately 55 million patients, and includes demographic information, service locations and reimbursement payments. Patients who underwent lumbar discectomy where identified using Current Procedural Terminology (CPT) codes. This cohort was tracked for all ED visits in the 90 days before and after lumbar discectomy. These populations were assessed for number and distribution of ED visits.

Results

In total, 32,932 lumbar discectomy patients were identified. In the ninety days following lumbar discectomy, 5,691 (17%) of these patients were evaluated in the ED (Figure). Of those who had an ED visit, they had 1.8+/-1.5 ED visits (average+/-standard deviation) with a range of 1-21 visits, leading to a total of 10,085 ED visits. Payments related to these ED visits in the 90 days after lumbar discectomy accounted for a total cost of $66,714,715 USD (61% of the total payments for discectomy procedures of the entire study cohort ($108,856,427 USD). Putting the above into perspective, during the ninety days before lumbar discectomy, 8,178 (25%) of the patients had been to the ED for 1.7+/-1.4 visits with a range of 1-21 visits (Figure), accounting for 14,005 total visits costing $77,756,795 USD.

Discussion/Conclusion

The current study found that a large percentage of patients who underwent lumbar discectomy had been to the emergency department before and/or after the procedure. These encounters were associated with significant costs, with payments from 10,085 visits totaling $66,714,715 USD in the ninety days following surgery. With increased focus on patient satisfaction and minimizing cost of care, it is critical to factor the role of ED visits into the overall cost of care. Better understanding these visits may reveal mechanisms to reduce ED visits and associated costs.

Poster 15. Development and Validation of a Point-based Risk Stratification Model for Early Complications after Posterior Lumbar Fusion

Moore Harold, BS1, Schneble Christopher, MD2, Kahan Joseph, MD, MPH3, Lukasiewicz Adam, MD2, Burroughs Patrick, BS4, Pathak Neil, BS4, Varthi Arya, MD2, Grauer Jonathan, MD3

1Weill Cornell Medical College, New York City, New York, United States, 2Yale New Haven Hospital, Department of Orthopaedics and Rehabilitation, New Haven, Connecticut, United States, 3Yale School of Medicine Department of Orthopaedics and Rehabilitation, New Haven, Connecticut, New Haven, Connecticut, United States, 4Yale School of Medicine, New Haven, Connecticut, United States

Background/Introduction

Posterior lumbar fusion is a commonly performed procedure. To optimize patient outcomes and minimize related healthcare costs, there remains much value in identifying patients at risk for early post-operative adverse events. Accordingly, this study aims to develop and validate a risk stratification point system for the occurrence of perioperative adverse events in patients undergoing posterior lumbar fusion.

Materials/Methods

All patients who underwent primary posterior spinal fusions with and without instrumentation for up to three contiguous levels were identified from the 2012-2018 National Surgical Quality Improvement Program (NSQIP) database. Operations performed for fracture, infection, or neoplasm were excluded. Patient demographics and medical conditions along with 30-day adverse events were catalogued and 30,000 patients were randomly assigned to a development cohort. Multivariate logistic regression analysis was conducted to identify factors associated with adverse events. Next, a nomogram was developed with point values for identified predictive comorbidities, creating a point-based risk-stratification system. To validate the point-based model, the scoring system was applied to the validation cohort which consisted of the remaining 21,633 patients.

Results

Of the development cohort, 5,065 (16.9%) developed an adverse event. Pre-operative co-morbidities associated with adverse events can be found in Table 1. On the basis of the regression analysis and nomogram, patients were assigned points for independent factor associated with adverse events: age 65-80yo +1, age >80yo +2, hypertension +1, insulin dependent diabetes +1, dyspnea +1, corticosteroid use +2, anemia +2, dependent functional status +3, and end stage renal disease +5. For the validation cohort, 3,541 (16.4%) developed an adverse event and the risk-stratification model was able to be successfully applied to predict the risk of early adverse events (AUC = 0.61, Pearson’s goodness of fit p = 0.562).

Discussion/Conclusion

The occurrence of adverse events follow posterior lumbar fusion was associated with hypertension, insulin-dependent diabetes, dyspnea, chronic corticosteroid use, anemia, dependent functional status and ESRD. We developed and validated a point-based scoring system to risk stratify patients undergoing lumbar fusion in order to predict the likelihood they will sustain early adverse events.

Poster 16. The Clinical Utility of Immediate Postoperative Imaging Following Lumbar Fusion

Ronald Andrew, BS1, Gerges Christina, BS1, Shost Michael, BS1, Jella Tarun, BS, MPH1, Herring Eric, MD2, Smith Gabriel, MD2

1Case Western Reserve University School of Medicine, Cleveland, Ohio, United States, 2University Hospitals Cleveland Medical Center, Cleveland, Ohio, United States

Background/Introduction

The clinical utility of routine postoperative imaging following lumbar fusion surgery remains uncertain. Imaging of uncomplicated patients, ordered in the absence of specific indications, rarely produces alterations to a patient's clinical management and may not justify the cost of the study or the minimal risk of ionizing radiation exposure. Existing studies on the clinical utility of postoperative imaging in uncomplicated lumbar fusion patients have largely focused on imaging obtained post-discharge. We present a retrospective analysis that to our knowledge is the first study reporting on the clinical utility of routinely imaging lumbar fusion patients during the immediate (pre-discharge) postoperative period.

Materials/Methods

The medical records of patients who had undergone elective lumbar instrumented fusion for degenerative disease from 2013 to 2018 across one regional healthcare system were retrospectively analyzed. Review of inpatient care notes was used to assess if a patient received immediate postoperative imaging and how the practice affected the course of their clinical management. Routine immediate postoperative imaging was defined as the patient having received any lumbar spine imaging (i.e. CT, MRI, x-ray) prior to discharge as part of routine practice in the absence of specific indications (i.e. new neurologic deficit).

Results

Preliminary analysis has identified 56 patients who underwent elective lumbar instrumented fusion for degenerative disease and received routine immediate postoperative imaging. 68% (n=38) received only plain film x-rays, 18% (n=10) received only CT imaging, 13% (n=7) received both plain film x-rays and CT imaging, and 1% (n=1) received both plain film x-rays and MRI. 100% (n=56) of these patients who received immediate postoperative imaging also received intraoperative imaging with either fluoroscopy, CT scan, or both. Across all patients, there were no identified instances where routine immediate postoperative imaging results led to alterations in the course of the patient’s clinical management.

Discussion/Conclusion

The practice of routinely imaging patients following lumbar fusion surgery appears to provide little utility to their clinical care. Further research is needed to identify optimal imaging criteria for patients following lumbar fusion surgery.

Poster 17. Biomimetic Scaffold for Bone Augmentation and Spinal Fusion in Rabbit Models

Sears Candice, PhD1, Taraballi Francesca, PhD1, Stetco Eliana, MS1, Bratescu Rachel, MD2, Weiner Bradley, MD2, Tasciotti Ennio, PhD1

1Houston Methodist Hospital, Houston, Texas, United States, 2Houston Methodist Orthopedics & Sports Medicine, Houston, Texas, United States

Background/Introduction

Autologous bone graft is the current gold standard in fusion, however available quantity is limited and complications can include chronic donor site pain and morbidity. There is a demonstrated need for a graft option in spinal fusion that addresses these concerns. The purpose of our study was to create a manufacturable scaffold with high osteoinductivity designed to mimic the composition and spatial organization of trabecular bone and periosteum which was implanted in a bilateral spinal fusion model.

Materials/Methods

Three biomaterials were synthesized to fabricate the scaffold: (i) magnesium-doped hydroxyapatite/type I collagen composite (MHA/Coll); (ii) porous collagen/elastin composite (Coll/Elastin); (iii) fibrous collagen membrane (Coll). A monolithic scaffold was generated by assembling the 3 collagen slurries, which was characterized with FTIR, TGA, HPLC, and SEM. Scaffolds were implanted in rabbit bilateral spinal fusion models. Ultrasound elastography was used to evaluate mechanical changes at the soft tissue/bone interface as well as CT to quantify bone regeneration at 3, 6, 9, and 12 weeks.

Results

Figure 1 demonstrates the three-layer biomimetic scaffold. FTIR confirmed collagen interaction with positively charged MHA nanocrystals and TGA revealed 45% scaffold mineralization. Ultrasound elastography demonstrated an increase in stiffness at the soft tissue/bone interface during bone regeneration. CT quantification of bone at 3, 6, 9, and 12 weeks post implantation showed substantial trabecular and cortical de novo bone growth.

Discussion/Conclusion

A three-layer monolithic scaffold with spatial organization of trabecular bone and periosteum was fabricated to mimic the composition and structure of native bone. Initial results show that implantation into a rabbit spinal fusion model led to substantial trabecular and cortical de novo bone growth. Ultrasound elastography was successfully validated to monitor bone regeneration in spinal fusion. The ultimate goal of this investigation is to evaluate the suitability of these scaffolds for advanced regenerative medicine approaches for spinal fusion, and to provide a reasonable alternative to autologous bone graft.

Poster 18. Minimally Invasive Transforaminal Lumbar Interbody Fusion Results in Less Reoperation than Open Transforaminal Lumbar Interbody Fusion

Cheaney II Barry, BS1, Ko Diana, BS1, Stedelin Brittany, BS1, Than Khoi, MD2

1Oregon Health Sciences University, Portland, Oregon, United States, 2Duke University Health System, Durham, North Carolina, United States

Background/Introduction

Transforaminal lumbar interbody fusion (TLIF) is a well-established procedure traditionally performed in an open (O-TLIF) fashion; to minimize tissue trauma, the minimally invasive (MIS-TLIF) approach was developed. In this study, we compared reoperation rates in patients who underwent O-TLIF versus MIS-TLIF.

Materials/Methods

The authors retrospectively reviewed the medical records of patients who underwent a single-level TLIF procedure by neurological and orthopedic surgeons at a single institution, with at least 6-month follow-up. Patient demographics and surgical details were collected. Data was analyzed by SPSS.

Results

A total of 127 patients met inclusion criteria; 98 O-TLIF (77.2%) and 29 MIS-TLIF (22.8%). The groups had similar demographics. There was no difference in mean operative time (p=0.772) or mean hospital stay (p=0.630). MIS-TLIF had less blood loss (p=0.007) and O-TLIF had longer follow-up (p=0.018). Of the 98 O-TLIF procedures, there were 30 (30.6%) reoperations, including 6 (20.0%) due to infection, 10 (33.3%) ASD, 6 (20%) hardware malposition, 3 (10.0%) non-union, and 5 (16.7%) spondylosis. Of the 29 MIS-TLIF procedures, there were 3 (10.3%) reoperations, including 1 (3.4%) due to infection, 1 (3.4%) ASD, and 1 (3.4%) spondylosis. There was no difference in reoperations due to infection (p=1.000), ASD (p=0.454), hardware malposition (p=0.335), non-union (p=1.000), or spondylosis (p=1.000). Collectively, there was a difference between all-cause reoperations (p=0.031) and non-infectious reoperations (p=0.040).

Discussion/Conclusion

In this study, MIS-TLIF approach resulted in less all-cause reoperations and less reoperations due to adjacent segment disease, hardware malposition, non-union, and spondylosis combined when compared to O-TLIF approach.

Poster 19. Caudo-Cephlad Impacts: Human Lumbar Spine Tolerance To Injury

Baisden Jamie, MD1, Moore Jason, MS1, Pintar Frank, PhD1, Yoganandan Narayan, PhD1

1Medical College of Wisconsin, Milwaukee, Wisconsin, United States

Background/Introduction

Injury criteria for human neck developed from animal, cadaver, dummy tests, and modeling are used in automotive standards since 2001.It is still a research topic for the L-spine, for which standards do not exist. As axial loading of the L-spine occurs in automotive, falls, and military events, it is important to determine human tolerance in this mode. This study aims to determine injury risk curves using PMHS.

Materials/Methods

L-spines were subjected to impact in the caudo-cephalad direction. Forces at caudal and cephalad ends were recorded using load cells and transformed. CT images were used to determine vertebral level-specific BMD, disc/body areas, and column lengths. These variables & age were used in risk curve analysis. Repeated testing protocols were used to bracket the sub-injurious and injurious impacts. Further impact loading was ceased upon suspicion of injury, soft/hard tissue or joint laxity. Injuries were categorized: AIS, stable vs unstable, and single vs multilevel. Peak axial and shear forces at the two ends of the specimen were used as primary response variables with other stated parameters as covariates in the parametric survival analysis. Brier scores were used to identify the best metric that explained the underlying injury mechanism.

Results

The resultant forces at the caudal end for stable and resultant forces at the cepahald end for the unstable injuries were found to the best metrics according to the Brier score metric (1.15 and 3.74). Peak forces of 5.8 kN and 7.6 kN were associated with 50% probabilities for stable and unstable injuries, and the ±95% confidence intervals (CI) were such that the two curves were considered to have a good quality index. Individual and covariate-based risk curves and data parameters used in the present study are discussed.

Discussion/Conclusion

This study replicates injuries to the L-spine from caudo-cepahald impacts, develops injury risk curves using a host of biomechanical variables, and conducts experiments in the same laboratory setting without confounding effects of device-related issues. Results from this investigation will aid computational models including isolated component spine and total human body models for a better understanding of the biomechanics of load transfer within the spinal components and injury.

Poster 20. Recent Trends in Spinal Surgery Performed by American Board of Orthopaedic Surgery (ABOS) Part II Candidates Over the Last Decade

Yeung Caleb, MD1, Schoenfeld Andrew, MD2, Kang James, MD2, Makhni Melvin, MD2

1Massachusetts General Hospital, Harvard Medical School, Boston, MA, United States, 2Brigham and Women's Hospital, Harvard Medical School, Boston, Massachusetts, United States

Background/Introduction

Orthopaedic surgeons are eligible to take the ABOS Part II examination 20 months following fellowship completion. The ABOS keeps a database of all cases performed including indications and volume. Despite previous studies utilizing ABOS Part II data in other subspecialties, there is a paucity of information regarding complications of spine surgeons in their first two years of independent practice. Such data can help trainees anticipate case composition they might see in early practice and provide insight for fellowship programs in understanding emerging trends in practice within the field.

Materials/Methods

We performed a retrospective review of surgical cases submitted to the ABOS by candidates taking the Part II Spine examination between 2008-2017. Procedures and indications for surgery were determined using ICD and CPT codes. Chi-squared testing with p-value of <0.05 was considered statistically significant.

Results

37,539 patients were analyzed (average 3,754 cases each year; 49 cases performed per candidate per year). 760 candidates took the exam between 2009-2018 (average 76 per year). Average patient age was 54.7 years, with a relatively equal distribution of males/females. 87.2% of cases were for degenerative conditions, 4.7% for trauma, 1.3% for infection/tumor, 2.2% for deformity, and 4.6% for other categories. Single-level posterior spinal instrumented fusion (PSIF) represented 13.0% of all cases; multi-level was 5.9%. Lumbar microdiscectomy was 22.4% of cases, single level ACDF was 10.2%, and multi-level ACDF was 8.3%. Single-level CDA: 0.7%; multi-level CDA: 0.1%; single-level TLIF: 6.4%; multi-level TLIF: 0.9%. Single-level anterior lumbar interbody fusion/extreme lateral interbody fusion (ALIF/XLIF) was 4.9% of cases; multi-level ALIF/XLIF: 1.6%. Single-level PSIFs decreased from 2009 to 2018 (19.4% vs 8.3%, p<0.001), as did lumbar microdiscectomies (27.5% vs. 18.3%, p<0.001). Single and multilevel ALIF/XLIF cases, and single- and multi-level TLIF/PLIF cases, did not significantly change, nor did single- or multi-level ACDF (9.9% to 10.0%, and 8.0% to 7.6%, respectively). Single and multi-level CDA cases increased significantly (0.5% to 1.0%, p=0.008; and 0.0% to 0.3%, p=0.001).

Discussion/Conclusion

Recent spine fellowship graduates are performing more cervical disc arthroplasties and fewer single-level PSIFs and lumbar microdiscectomies. Techniques such as ALIF/XLIFs, ACDFs, and multilevel PSIFs have not changed significantly in recent years.

Poster 21. The Incidence and Clinical Impact of Proximal Junctional Kyphosis by Variations in Definition

Ton Andy, BS1, Alluri Ram, MD2, Kim Andrew, MS3, Hah Raymond, MD2

1University of Southern California Keck School of Medicine, Los Angeles, California, United States, 2Keck Medical Center of USC, Los Angeles, California, United States, 3Loyola University Chicago, Chicago, Illinois, United States

Background/Introduction

Proximal Junctional Kyphosis (PJK) is a common complication following corrective surgery for adult spinal deformities. Although well-recognized, there remain inconsistencies in criteria used to define PJK. Without a standardized definition, it remains difficult to thoroughly compare findings regarding PJK throughout the literature. To assess the need for unanimity in a definition, the current study analyzes differences in incidence of PJK, proximal junctional failure (PJF), and Oswestry Disability Index (ODI) scores between commonly cited definitions for PJK.

Materials/Methods

A retrospective review of 88 patients (50 female, 38 male) who had undergone corrective surgery for adult spinal deformities was conducted at a single institution from 2015 - 2019. Preoperative and postoperative sagittal Cobb angles were measured between the lower end plate of the upper instrumented vertebrae (UIV) and the superior end plate of both the supra-adjacent vertebrae (UIV+1) and two supra-adjacent vertebrae (UIV+2). The difference between the preoperative and final follow-up angle was used to determine the presence of PJK within the four commonly cited definitions: 1) UIV and UIV+2 > 10, 2) UIV and UIV+2 > 15, 3) UIV and UIV+2 > 20, 4) UIV and UIV+1 > 15. Incidence of PJK, PJF, and preoperative and postoperative ODI scores were compared across definitions using one-way ANOVA.

Results

A total of 43 patients fit at least one of four definitions used for PJK. All results are presented in Table 1. The incidence of PJK decreased in definitions with higher sagittal Cobb angle thresholds. Statistical analysis found no significant differences in preoperative and postoperative ODI scores between definitions. Additionally, no differences in absolute change or relative change of ODI scores were observed between groups. Finally, no significant differences were found for incidence of PJF across four different definitions of PJK.

Discussion/Conclusion

A uniform definition for PJK has yet to be established. In definitions with stricter criteria, the incidence of PJK expectedly decreased. Stricter definitions of PJK, however, found no difference in incidence of PJF or improve in ODI scores. Future studies should investigate varying definitions of PJK to assess if there is a radiographic parameter at which point PJK affects clinical outcomes.

Poster 22. Improved Efficacy of Tlif Procedures Due to a New Device: Preliminary Cadaver Study and a 209-Patient Retrospective Study in a Single Center

Peloza John, MD1, Millgram Michael, MD2, Rand Nahshon, MD2, Guyer Richard, MD3, Le Huec Jean-Charles, MD, PhD4, Ashkenazi Ely, MD2

1Center For Spine Care, Dallas, Texas, United States, 2Israel Spine Center, Assuta Hospital, Tel Aviv, Israel, Israel, 3Texas Back Institute, Plano, Texas, United States, 4Bordeaux University Hospital, Bordeaux, France

Background/Introduction

Transforaminal Lumbar Interbody Fusion (TLIF) procedures are commonly performed in the US, increasing in number as the population ages. A new powered FDA-approved device, designed to clean the end-plates for improved cartilage removal and shorten the time required for disc removal, can benefit this procedure’s outcome and reduce the time it requires. This study aims to assess the effect of device use in a preliminary cadaver study and a retrospective single center study.

Materials/Methods

The records of 209 single-level TLIF procedures conducted in a single hospital were reviewed. Overall the study included 143 procedures conducted using the device during 2014-2019 and 66 control procedures, conducted without the device during 2012-2019. Surgery duration, length-of stay and complication rates were extracted from the records and compared between both groups. In addition, a preliminary cadaver study was conducted on five lumbar levels by two surgeons. The number of instrument passes required for each surgeon with and without the device was measured and compared. After the conclusion of the procedures, the cadaver disc-spaces were cut open and the end-plates were observed for any perforations.

Results

The analysis revealed statistically significant reductions of 10 minutes in surgery duration and 0.5 days in the length-of-stay. In addition, the device group had less complications (2.8% vs. 6.1%), including fewer surgical-site infections (0.7% vs. 1.5%) and less readmissions (2.1% vs. 3%). Fewer patients in the device group (2.8% vs. 9.1%) complained on post-operative pain or weakness in the leg, possibly due to the longer surgery time. The number of instrument passes required to clean the cadaver disc-spaces was reduced by 18.5 on average when the device was used. As the figure shows, end-plate preparation was improved with the use of device compared with control. End-plate perforation occurred in 1 out of 6 end-plates in the device levels (16.7%) and in 1 out of 4 end-plates in the control group (25%).

Discussion/Conclusion

The study suggests that the device use can lead to a shorter procedure and hospitalization and potentially also reduced the complication rate, without deteriorating the clinical outcome.

Poster 23. Postoperative Bracing Practices after Elective Lumbar Spine Surgery: A Questionnaire Study

Pathak Neil, BS1, Scott Michelle, MD2, Galivanche Anoop, BS1, Justen Marissa, BS1, Hilibrand Ari, BS1, Moore Harold, BS3, Malpani Rohil, BS1, Varthi Arya, MD4, Grauer Jonathan, MD5

1Yale School of Medicine, New Haven, Connecticut, United States, 2Montefiore Medical Center University Hospital, Albert Einstein College of Medicine, Bronx, New York, United States, 3Weill Cornell Medical College, New York City, New York, United States, 4Yale New Haven Hospital, Department of Orthopaedics and Rehabilitation, New Haven, Connecticut, United States, 5Yale School of Medicine Department of Orthopaedics and Rehabilitation, New Haven, Connecticut, New Haven, Connecticut, United States

Background/Introduction

Surgeons may choose to use orthoses following spine surgery, though there is limited data available on optimal protocols and its utility across specific procedures. In 2009, Bible et al. conducted a questionnaire-based study; this previous investigation found postoperative bracing after elective lumbar spine surgery to have an overall frequency of 49%. The present study aims to provide a ten-year update regarding postoperative bracing practices after elective lumbar spine surgery.

Materials/Methods

A single-page questionnaire was distributed to all attendees of the Lumbar Spine Research Society Annual Meeting (April 2019). The questionnaire assessed demographic information, and asked surgeons to identify if they used orthoses postoperatively in a list of ten elective lumbar spine surgeries. Details regarding type of brace, reason for bracing, and duration of bracing were also collected. Chi-square tests were used to compare frequencies.

Results

Seventy-three of 88 U.S. attending surgeons completed the questionnaire (response rate: 83%). The majority of respondents were orthopaedic surgery-trained (78%), had completed fellowship (84%), practiced in an academic setting (73%), were in practice for 5-10 years (40%), and practiced in the Midwest (44%). The overall postoperative bracing frequency across all elective lumbar surgeries was found to be 27%. This frequency was significantly lower than the 49% in the referenced study from a decade earlier (Bible et al., 2009, p<0.0001). Respondents tended to brace least often for discectomies (10.8%) and most often for stand-alone lateral interbody fusions (43.5%) (p<0.0001). The most frequently cited brace was an off the shelf lumbar sacral orthosis (66%), and the most common reason for bracing was to improve pain. Surgeons cited bracing duration as 2-4 months (57%), followed by 3-8 weeks (36%). There was a statistically significant difference found in overall postoperative bracing frequency for private practice surgeons (50.4%) compared to academic surgeons (17.5%) (p=0.004).

Discussion/Conclusion

The present questionnaire study showed a clear decrease in postoperative bracing after elective lumbar surgery compared to a reference study from a year ago. In the absence of rigorous evidence of when to use such bracing, further prospective studies may be helpful in assessing the clinical utility of postoperative bracing for different types of elective lumbar spine surgery.

Poster 24. Impact of Allogeneic Blood Transfusion on Postoperative Morbidity and Mortality in Major Thoracic and Lumbar Posterior Spine Fusion and Instrumentation Surgeries

Kurra Swamy, MBBS1, DeMercurio Prisco, BS1, Gorji Reza, MD1, Li Fenghua, MD1, Tallarico Richard, MD1, Sun Mike, MD1, Lavelle William, MD1

1SUNY Upstate Medical University, Syracuse, New York, United States

Background/Introduction

Intraoperative blood loss increases morbidity and mortality in patients undergoing major spine surgery. A liberal strategy of blood transfusion (blood transfusion in anticipation of blood loss by surgeons) has been practiced in our institution for major spine surgery since 2013 and the clinical impression by surgeons of patient outcomes has been positive. The purpose of the study was to retrospectively investigate the impact of blood transfusions on outcomes in patients who underwent major spine surgeries.

Materials/Methods

Major spine surgery is defined as fusion and instrumentation surgery involved five or more levels of spine in thoracic, or lumbar, or combined spine levels. Patients who underwent major spine surgery between 2013 and 2017 were included in this study. Patient data included: demographics; amount of IBL; amount of blood transfused during surgery (units of PRBC); number of levels of fusion; and in-hospital and follow-up complications were collected and recorded. Patients were divided into two groups based PRBC transfusion used: Group 1(no blood transfusion, n=107) and Group 2 (PRBC transfused, n=222).

Results

Sample size, n=329 patients. Complications were: 24 patients (7.2%) in-hospital complications and 16 patients (4.8%) follow-up. In Group 1, 4 (3.7%) patients had in-hospital complications (1 pulmonary embolism (PE), 1 hospital acquired pneumonia, 1 atrial fibrillation (AFib), and 1 supraventricular tachycardia and postoperative hypotension (POH)). In Group 2, 20 patients (9%) experienced in-hospital complications (3 surgical site infections, 1 sepsis, 5 POH, 1 POH and PE, 4 DVTs (deep vein thrombosis), 3 pulmonary edema, 1 pneumothorax and 1 pleural effusion, 1 death). There was no difference in follow-up mortality between the groups. In Group 1, 3 patients had infection and 1 seroma; in Group 2, 9 patients had infections, 1 seroma and 1 paraspinal fluid collection. There was no correlation noticed between intraoperative blood transfusion and in-hospital complications (p=0.85) and follow-up complications (p=0.63), respectively (Table 1). However, intraoperative blood loss and older age trended toward postoperative complications (p=0.06, and p=0.05, respectively).

Discussion/Conclusion

Intraoperative blood transfusion in major spine surgery did not increase in-hospital and 9-month follow-up mortality. However, the results should be interpreted cautiously due to the small sample size.

Poster 25. Semitendinosus Ligament Reinforcement for Prevention of Proximal Junctional Failure

Alluri Ram, MD1, Kim Andrew, MS2, Ton Andy, BS3, Kang Hyunwoo, MD1, Hah Raymond, MD1

1Keck Medical Center of USC, Los Angeles, California, United States, 2Loyola University Chicago, Chicago, Illinois, United States, 3University of Southern California Keck School of Medicine, Los Angeles, California, United States

Background/Introduction

Investigation of reinforcement of the posterior ligamentous complex (PLC) has been limited to finite element analysis, cadaveric studies, or case series without a control group. In this study we sought to describe the clinical results of PLC augmentation using a semitendinosus allograft and compare these results to a control group.

Materials/Methods

Patients with adult spinal deformity who underwent fusion from the thoracic spine to pelvis with semitendinosus allograft to reinforce the PLC (allograft group) or without PLC augmentation (control group) were identified. Minimum follow-up was 6 months. The following radiographic measurements were calculated for each group: Upper instrumented vertebrae (UIV) tilt, UIV + 2 Sagittal cobb angle, thoracic kyphosis (TK), sagittal vertical axis (SVA), lumbar lordosis (LL), pelvic incidence (PI), pelvic tilt (PT), T1 sagittal tilt, and T1 pelvic angle. PJK (UIV + 2 ≥ 20), PJF (PJK requiring revision surgery), and Oswestry Disability Index (ODI) were compared between both groups. P≤ 0.05 was considered significant.

Results

Forty-nine patients in the allograft group and 34 in the control group were identified. Average follow-up was 20 months. Demographic variables, preoperative ODI, % revision, % osteotomy, % anterior column procedures, and UIV level were similar in both groups. Preoperative and postoperative spinopelvic parameters were also similar in both groups. Final PJK incidence was similar in both groups (P=1.00) (Figure 1). PJF incidence was significantly higher in the control group (P=0.01) (Figure 1). Postoperative ODI scores were significantly lower in the allograft group (42 v 55, P=0.01) and demonstrated a significantly greater relative % decrease from preoperative ODI when compared to the control group (31% v 10%, P=0.02).

Discussion/Conclusion

In a well-matched case-control cohort of adult deformity patients undergoing thoracic to pelvis fusion, the use of semitendinosus allograft to reinforce the PLC significantly lowered the incidence of PJF over the study period. Zero patients required revision for PJK in the allograft group while 6 required revision in the control group. Functional outcomes were also significantly better in the allograft group. The use of semitendinosus allograft to reinforce the PLC at the UIV may be a feasible technique to decrease PJF in adult spinal deformity surgery.

Poster 26. Biomimetic 3D-printed Titanium-Alloy Interbody Spacers Demonstrate Uniform Bone Growth over 12 Weeks

Van Horn Margaret, PhD1, Beard Roland, MS1, Cunningham Bryan, PhD2, Mullinix Kenneth, MS2, Bucklen Brandon, PhD3

1Musculoskeletal Education and Research Center, A Division of Globus Medical, Inc., Audubon, Pennsylvania, United States, 2MedStar Union Memorial Hospital, Baltimore, Maryland, United States, 3Globus Medical, Inc., Audubon, Pennsylvania, United States

Background/Introduction

A primary clinical outcome of interbody fusion surgery is achievement of bony fusion. Biomimetic implants may promote formation of new bone. A uniform rate of bone formation during healing may lead to optimized fusion. The purpose of this study was to investigate bone growth rate (BGR) following lateral lumbar interbody fusion (LLIF) across different interbody materials.

Materials/Methods

Eighteen sheep underwent a two-level LLIF using polyetheretherketone (PEEK), titanium-alloy (TAV), or 3D-printed biomimetic TAV (3DP) spacers (n=6 levels per group per time point). Animals were sacrificed at 6 and 12 weeks postoperatively. Cadaver sheep (n=4 levels per group) were implanted with bone graft packed spacers to quantify the initial 0 week time point. Microcomputed tomography quantified bone volume (BV) through the spacer and BGR between time points was analyzed.

Results

The 3DP-treated group demonstrated a more consistent BGR over 12 weeks compared to PEEK and TAV groups. The BGR for the 3DP group was 10.6mm3/week and 9.6mm3/week from 0 to 6 weeks and 6 to 12 weeks, respectively. Bone growth in PEEK and TAV groups was delayed, evidenced by not surpassing the remodeling phase until after 6 weeks and an inconsistent BGR between early and late time intervals (PEEK: 1.5 and 17.0mm3/week, respectively; TAV: -2.1 and 17.7mm3/week, respectively) (Figure 1).

Discussion/Conclusion

This study demonstrated biomimetic 3DP TAV interbody spacers facilitated uniform bone growth through the spacers over 12 weeks of healing, with lower bone volume observed with PEEK and TAV implants at the 6 week time point. Biomimetic 3DP implants lead to robust intervertebral fusion.

Poster 27. Effect of Lower Lumbar Fusion on the Cone of Economy

Patwardhan Avinash, PhD1, Muriuki Muturi, PhD2, Havey Robert, MS2, Blank Kenneth, PhD2, Yang Seunghyuk, MS1, Ghanayem Alexander, MD1, Matteini Lauren, MD1, Wojewnik Bartosz, MD3

1Loyola University of Chicago, Maywood, Illinois, United States, 2Edward Hines, Jr. VA Hospital, Hines, Illinois, United States, 3Loyola University Medical Center, Maywood, Illinois, United States

Background/Introduction

The upright posture should fall within the “cone of economy,” requiring the least amount of muscle energy to maintain the upright posture. In this study we used a novel laboratory model and cadaver specimens to quantify the effect of lower lumbar fusion on the “cone of economy”.

Materials/Methods

T10-sacrum cadaveric specimens underwent biomechanical tests in the native state, after L4-S1 immobilization, and finally after creating a “hypolordotic” fusion across L4-S1. Specimens were subjected to sagittal vertical axis (SVA) testing, in which the horizontal offset of T10 relative to sacrum (T10 SVA) was varied (Fig. 1). Shear force at T10 and moment at sacrum were measured.

Results

Fig. 1 depicts the T10 shear force response of a specimen to changing T10 SVA. Native spine response (blue curve), demonstrates a wide “neutral zone” (blue hash marks). It indicates that to change the T10 SVA between about 1.5cm posterior to 4cm anterior to the starting upright posture requires only about 6 Newtons of shear force. A hypolordotic L4-sacrum fusion (orange curve) shifts the position of the T10 vertebra nearly 40mm anterior of the starting neutral upright posture. In this posture T10 SVA is forward, requiring constant posteriorly directed shear force to maintain the posture. From this hypolordotic posture, it was not possible to return T10 to its position in the starting upright posture. An estimated posterior shear force of 35-40 Newtons would be required (dotted orange curve), exposing the lower lumbar spine to elevated forces and moments.

Discussion/Conclusion

The intact spine response demonstrates the existence of a “cone of economy” in T10 SVA, such that if T10 remains within this zone, the shear force required to maintain the upright posture would vary little from its minimum. After hypolordotic L4-S1 fusion, the size and shape of the “cone of economy” changes and substantially larger magnitudes of forces are needed to bring T10 back within a specified range around the upright standing posture. Improving L1-S1 lordosis so that the T10 vertebral position approaches the normal upright posture is the goal of flatback corrective surgery to allow patients to stand upright with reduced muscle burden.

Poster 28. The Effect of Modifiable Risk Factors on 30, 60, and 90-day Readmission Rates in Patients Following Lumbar Spine Fusion

Ton Andy, BS1, Shahrestani Shane, MS1, Chen Xiao, BS1, Ballatori Alexander, BS1, Wang Jeffrey, MD2, Buser Zorica, PhD1

1University of Southern California Keck School of Medicine, Los Angeles, California, United States, 2Keck Medical Center of USC, Los Angeles, California, United States

Background/Introduction

The increasing prevalence of lumbar spinal fusion has been accompanied by an increase in costly and untimely readmissions - a concern shared among hospitals, physicians, and patients alike. Many studies emphasize perioperative and intraoperative care, few highlight the autonomy and control of patients over their own outcomes. Therefore, this study explores the impact of modifiable risk factors on 30, 60, and 90-day readmission rates following lumbar spine fusion.

Materials/Methods

In a retrospective cohort analysis, patients with lumbar spine fusions who were non-electively readmitted within 30 (n=5,179), 60 (n=7,122), and 90-day (n=9,678) periods were identified within the 2016 Nationwide Readmissions Database (NRD) using ICD 10 codes. Patients were then stratified by the following modifiable risk factors: Alcohol use, tobacco or nicotine use, malnourishment, and hyperlipidemia. Routine follow-ups were excluded. Statistical analysis was conducted using Welch’s paired two sample t-test to compare differences in 30-, 60-, and 90-day readmissions in patients diagnosed with the identified modifiable risk factors versus those without.

Results

30-day readmission rates were significantly higher in patients with malnourishment (p=0.0488) and hyperlipidemia (p<0.0001) compared to those without. No significant differences in 30-day readmission, however, were found between patients with tobacco/nicotine (p=0.0596) and alcohol (p=0.0907) usage compared to those without. 60-day readmission rates were significantly higher in patients with malnourishment (p<0.0001), hyperlipidemia (p<0.0001), tobacco or nicotine use (p=0.011), and alcohol use (p=0.0007) compared to those without each respective comorbidity. At 90-days, readmission rates were significantly higher in patients with malnourishment (p<0.0001), hyperlipidemia (p<0.0001), tobacco or nicotine use (p<0.001), and alcohol use (p<0.0001) compared to those without each respective comorbidity.

Discussion/Conclusion

Malnourished or hyperlipidemic patients were readmitted significantly more at 30 days than patients without those comorbidities, though tobacco/nicotine and alcohol were non-significant. However, all modifiable risk factors demonstrated a significant difference in readmission rates at 60 and 90 days compared to patients without those risk factors. Future studies should investigate the impact of these modifiable risk factors on long-term outcomes for lumbar spine fusion patients.

Poster 29. Single Position Robotic Assisted Surgery versus Conventional Minimally Invasive Surgery during LLIF: An In Vitro Comparison

Brady Robert, MD1, Protopsaltis Themistocles, MD2, Larson Jeffrey, MD3, Frisch Richard, MD4, Huntsman Kade, MD5, Lansford Todd, MD6, Maulucci Chris, MD7, Hayward II Gerald, BS8, Harris Jonathan, MS8, Bucklen Brandon, PhD9

1Coastal Orthopaedics, Norwalk, Connecticut, United States, 2Rothman Institute, New York University, Milburn, New Jersey, United States, 3Coeur d'Alene Spine and Brain, Coeur d'Alene, Idaho, United States, 4Southeastern Spine Institute, Mt Pleasant, South Carolina, United States, 5Salt Lake Orthopaedic Clinic, Salt Lake, Utah, United States, 6South Carolina Sports Medical and Orthopaedic, North Charleston, South Carolina, United States, 7Tulane Medical Center, New Orleans, Lousiana, United States, 8Musculoskeletal Education and Research Center, A Division of Globus Medical, Inc., Audubon, Pennsylvania, United States, 9Globus Medical, Inc., Audubon, Pennsylvania, United States

Background/Introduction

Lateral lumbar interbody fusion (LLIF) provides restoration via indirect decompression of the neural elements, while minimizing potential vascular complications associated with ALIF. Posterior fixation may be applied through various techniques including conventional minimally invasive surgery (CMIS) requiring the patient to be repositioned prone to provide access to both pedicles. Conversely, robot-assisted navigation (RAN) of pedicle screws can be utilized from a single-position, without flipping the patient. RAN is theorized to reduce patient surgical time, radiation, and blood loss due to positioning and workflow effects.

Materials/Methods

Twelve unembalmed human torsos were implanted with two level static LLIF cages, followed by posterior bilateral pedicle screw fixation using either CMIS (n=6) or RAN (n=6). Preoperative computed tomography (CT) RAN workflow utilized CT scans of the specimen taken offsite and transferred to the robotic system during setup. Screw planning was performed using these CT scans, and then merged with intraoperative fluoroscopy. Surgical times and surgeon radiation exposure were measured. Patient flip time from a consecutive patient series was included.

Results

Significant differences in surgical time and radiation dosages were found between groups. Surgical time for preoperative RAN and CMIS was 64.7 min±4.1 min and 123.0 min±13.7 min, respectively, as shown in Table 1. Time per screw for RAN and CMIS workflows was 2.7±0.6 min and 4.3±1.3 min, respectively. RAN was significantly different for total operative time and time per screw in comparison to CMIS (p<0.05). Radiation dosage and time were separated into interbody and posterior fixation procedures separately, and sorted by imaging workflow. RAN and CMIS radiation dosages during posterior fixation were 0.4±0.2 rad and 2.7±1.6, respectively as shown in Table 1.

Discussion/Conclusion

Significant differences were found in both surgical time and radiation exposure between conventional MIS and robot assisted navigation with RAN resulting in shorter surgical times and less radiation exposure to the surgeon than CMIS. Consideration should be given to single-position LLIF procedures that utilize RAN to instrument the spine with bilateral pedicle screws.

Poster 30. Biomechanical Investigation of an L5 Pedicle Subtraction Osteotomy for Sagittal Malalignment: An In-Vitro Study

Janjua Burhan, MD1, Hayward II Gerald, BS2, Harris Jonathan, MS2, Thai Evan3, Bucklen Brandon, PhD4

1Mercy Health, Rockford, Illinois, United States, 2Musculoskeletal Education and Research Center, A Division of Globus Medical, Inc., Audubon, Pennsylvania, United States, 3Drexel University, Philadelphia, Pennsylvania, United States, 4Globus Medical, Inc., Audubon, Pennsylvania, United States

Background/Introduction

Adult spinal deformity can result from progression of prior (childhood/adolescence) spinal kyphoscoliosis or an iatrogenic degenerative spinal disease. Sagittal malalignment has been established with poor health-related quality of life. Planned osteotomies may help achieve optimal sagittal correction. The objective of this study is to investigate the effect of an L5 pedicle subtraction osteotomy (PSO) on sagittal correction, as well as range of motion, following L2–S2 fixation.

Materials/Methods

Six cadaveric specimens (L1–pelvis) were tested in flexion-extension (FE), lateral bending (LB), and axial rotation (AR) to ±12.5 Nm. Constructs included intact and PSO with fixation from L2–S2 with S2-alar iliac fixation (L2–S2+PSO) with lordosis measurements recorded for each construct. Lordotic angles, measured from the inferior endplate of L4 and superior endplate of S1, were recorded for intact and post-L5-PSO specimens to determine the impact of sagittal correction.

Results

Mean lordotic angles [±SEM] in the intact and L2–S2+PSO specimens were 30.12°[±7.56°] and 46.81°[±7.75°], respectively, with L2–S2+PSO having significantly higher lordotic angles (p<0.05). L2-S2+PSO significantly reduced motion compared to intact across all bending planes (15.4%, 15.3%, and 50.9% in FE, LB, and AR, respectively).

Discussion/Conclusion

In this cadaveric study, significant correction was achieved with an L5 PSO, similar to other biomechanical studies investigating mid-lumbar PSOs. However, the most caudal lumbar level (L5) osteotomy provided the greatest contribution to overall spinopelvic sagittal alignment. Supplemental pelvic fixation plays an additive role in construct stability.

Poster 31. Biomechanical Effects of Vertebroplasty on Osteolytic Lumbar Spine Metastasis

Choi Hoon, MD, MS1, Khandelwal Prashant, MS1, Porwal Vaibhav, MS1, Baisden Jamie, MD1, Yoganandan Narayan, PhD1

1Medical College of Wisconsin, Milwaukee, Wisconsin, United States

Background/Introduction

The incidence of spinal metastasis has increased due to longer survival of cancer patients and improved imaging. Percutaneous ablation and vertebral augmentation have gained renewed interest to manage debilitating pain and improve quality of life. Biomechanical effects of vertebral augmentation in lumbar spine metastasis are unclear.

Materials/Methods

A validated finite element osteoligamentous lumbosacral spine model was used. Bone density and elastic modulus were reduced to simulate osteoporosis. Osteolytic L3 vertebral metastasis was simulated by reducing elastic modulus and Poisson’s ratio of cancellous bone. Two models, M50 and M80, simulated 50% and 80% vertebral body involvements respectively. Vertebroplasty was simulated by injecting 4 cc of PMMA through each pedicle. 1750 N of follower forces were applied to simulate sitting. Vertebral body forces in response to axial loading were obtained.

Results

Greater vertebral body involvement by metastasis led to reduced vertebral body forces. M80 model after vertebroplasty showed minimal improvement in vertebral body forces. M50 with vertebroplasty, however, showed restoration of vertebral body forces at both index and adjacent levels beyond those of the metastasis-free model.

Discussion/Conclusion

There is minimal biomechanical benefit of vertebroplasty in those with extensive metastatic involvement of the vertebral body. Those with 50% or less vertebral body involvement can have significant improvement of the spinal biomechanics after vertebral augmentation. However, increases in forces beyond baseline in adjacent segments could predispose the patient to compression fractures.

Poster 32. Machine Learning for Predictive Modeling of 90-day Readmission, Major Medical Complication, and Discharge to a Facility in Patients Undergoing Long Segment Posterior Lumbar Spine Fusion

Jain Deeptee, MD1, Durand Wesley, BS2, Burch Shane, MD3, Daniels Alan, MD4, Berven Sigurd, MD5

1Washington University in St Louis, St. Louis, Missouri, United States, 2Alpert Medical School of Brown University, Providence, Rhode Island, United States, 3Department of Orthopedic Surgery, University of California San Francisco, San Francisco, California, United States, 4Alpert Medical School of Brown University, Department of Orthopaedics, Providence, Rhode Island, United States, 5Department of Orthopedic surgery, San Francisco, University of California, United States

Background/Introduction

Surgery for adult spinal deformity is effective for treating spine-related disability, however has high complication and readmission rates. The objective of this study was to develop predictive models for post-operative outcomes after long segment lumbar posterior spine fusion (LSLPSF).

Materials/Methods

In this retrospective case control study, patients who underwent LSLPSF (3 or more levels) were identified in State Inpatient Database. Data was queried for discharge-to-facility (DTF), 90-day readmission, and 90-day major medical complications, and demographic, comorbid, and surgical data. Data was partitioned into training and testing sets. Multivariate logistic regression, random forest, and elastic net regression were performed on the training set. Models were applied to the testing set to generate AUCs. AUCs between models were compared using the method by DeLong et al.

Results

37,852 patients were analyzed. The DTF, 90-day readmission, and 90-day major medical complication rates were 35.4%, 19.0%, and 13.0% respectively. For DTF, the logistic regression AUC was 0.77, vs. 0.75 for random forest and 0.76 for elastic net (p<0.05 for all comparisons). For 90-day readmission, the logistic regression AUC was 0.65, vs. 0.63 for both random forest and elastic net (p<0.05 for all comparisons). For 90-day major medical complications, the logistic regression AUC was 0.70, vs. 0.69 for random forest and 0.68 for elastic net (p<0.05 for all comparisons).

Discussion/Conclusion

This study created comprehensive models to predict discharge to facility, 90-day readmissions, and 90-day major medical complications after LSLPSF. This information can be used to guide decision making between the surgeon and patient, as well as inform value based payment models.

Poster 33. Risk Factors for Blood Transfusion in Patients Undergoing Single-Level Anterior Lumbar Interbody Fusion

Lee Ryan, BS1, Lee Danny, BS2, Fassihi Safa, MD3, Heyer Jessica, MD3, Iweala Uchechi, MD4, Weinreb Jeffrey, MD3, Ramamurti Pradip, BS3, Yu Warren, MD5

1Hospital for Special Surgery, New York, New York, United States, 2George Washington University School of Medicine, Washington, District of Columbia, United States, 3George Washington University Department of Orthopaedic Surgery, Washington, District of Columbia, United States, 4New York University, Langone Medical Center, New York City, New York, United States, 5Division of Spine Surgery; Precision Orthopaedics and Sports Medicine, Gaithersburg, MD - Maryland, United States

Background/Introduction

Given that blood transfusions are associated with a multitude of complications, there have been increasing efforts made to minimize blood loss and identify patients at risk of requiring transfusion. To the authors’ knowledge, there have been no previous reports on independent risk factors that increase the need for blood transfusions in anterior lumbar interbody fusion (ALIF). As ALIF continues to be performed for a variety of indications, this study aimed to identify risk factors for blood transfusions in ALIF patients so that perioperative management can be optimized to reduce patient morbidity.

Materials/Methods

10,232 patients who underwent single-level ALIF were isolated from a multi-institutional surgical registry. 977 (9.55%) were transfused intraoperatively or postoperatively within 72 hours from the start of surgery. Analysis of Variance (ANOVA), Fischer’s exact tests, and Pearson’s chi-squared analyses were utilized to assess for differences in patient demographics and preoperative comorbidities. Significant demographic factors and comorbidities were input into a logistic regression model in order to determine significant risk factors for intraoperative/postoperative blood transfusions. The predicted probabilities given by the logistic regression model were used to generate a Receiving Operating Characteristic (ROC) curve to assess the discriminatory ability of the regression model in assigning patients into the transfused or non-transfused cohorts based on the controlled variables.

Results

Age >60 years (OR:1.834, p<0.001), female gender (OR:1.746, p<0.001), white ethnicity (OR:1.302, p=0.047), insulin-dependent diabetes mellitus (OR:1.670, p=0.001), dyspnea (OR:2.445, p<0.05), open wound/wound infection (OR:2.360, p=0.022), corticosteroid use (OR:1.557, p=0.009), hematologic disorders (OR:1.829, p=0.026), transfusions within 72 hours preoperatively (OR:9.733, p<0.001), significant weight loss (OR:2.819, p=0.026), and longer operative time (>256 min) (OR:4.801, p<0.001) all independently increased the risk of blood transfusions for single-level ALIF. The Area Under ROC curve (AUROC) was 0.755 (p<0.001), indicating that the logistic regression model was able to adequately predict the need for transfusions with respect to the entered variables.

Discussion/Conclusion

Several demographic factors and comorbidities are associated with increased rates of blood transfusion for single-level ALIF. By identifying these risk factors, surgeons can risk-stratify ALIF surgical candidates and optimize them appropriately prior to surgery, thereby reducing the risk of blood transfusion and the associated complications.

Poster 34. Preliminary Findings of the Clinical Efficacy of Tranexamic Acid in Posterolateral Lumbar Fusion: A Prospective Randomized Double-Blind Control Trial

Basques Bryce, MD1, Harada Garrett, MD1, Khan Jannat, BS2, Goldberg Edward, MD2, An Howard, MD1, Colman Matthew, MD3

1Midwest Orthopaedics at Rush, Chicago, Illinois, United States, 2Rush University Medical Center, Chicago, Illinois, United States, 3Midwest Orthopaedics at Rush University, Chicago, Illinois, United States

Background/Introduction

Tranexamic acid (TXA) is widely regarded as an efficacious pharmacotherapy to minimize bleeding during a large variety of surgical procedures. Despite this, TXA is not routinely used in degenerative lumbar spine cases, and the optimal dosing regimen is not known. The purpose of this study was to explore the clinical utility of TXA in posterolateral lumbar fusions (PLF) while considering differences in intravenous and oral dosing schedules.

Materials/Methods

We performed a prospective randomized double-blind study investigating effects of TXA on blood loss in short-length open PLFs. Exclusion criteria included a current/recent history of a thromboembolic events, allergy to TXA, and/or pregnancy/breastfeeding. Patients were randomized to one of four study arms: 1) Oral placebo, 2) Intravenous TXA at time of incision, 3) 1950mg Oral TXA 1 hour before incision, and 4) 1950mg oral TXA 1 hour before incision, and 1950mg oral TXA once per day for three days after surgery. Outcome measures included postoperative hemoglobin, hematocrit, and hospital length of stay. Differences in baseline characteristics between study arms was assessed using a combination of chi-square & ANOVA, while assessment of outcomes was performed using multivariate linear regression.

Results

74 patients (41/74; 55.4% female) were enrolled, with a mean age of 65.9±12.6 years, body mass index of 32.4±6.2 kg/m2, and 2.0±1.4 levels fused. For each treatment group, 21, 17, 20, and 16 patients were assigned to study arms 1-4, respectively. There were no differences in baseline demographic or operative characteristics, though preoperative hematocrit values significantly differed across groups (p=0.035). (Table 1) There were no differences between groups in assessed outcomes after controlling for these baseline parameters.

Discussion/Conclusion

TXA can be used safely in PLF, though there was no clear benefit in hemostasis when compared to placebo. Though possibly underpowered, these findings suggest that TXA may not have a clinically meaningful application in short-length open PLF.

Poster 35. Variation of Bone Mineral Density within Each Vertebral Body in the Lumbar Spine

Vaishnav Avani, MBBS1, Louie Philip, MD2, Chaudhary Chirag, MBBS1, Lee Ryan, BS1, McCarthy Michael, MD1, Iyer Sravisht, MD1, McAnany Steven, MD3, Albert Todd, MD3, Gang Catherine Himo, MPH1, Qureshi Sheeraz, MD3

1Hospital for Special Surgery, New York, New York, United States, 2Midwest Orthopaedics at Rush, Chicago, Illinois, United States, 3Hospital for Special Surgery/Weill Cornell Medical College, New York, New York, United States

Background/Introduction

With increasing use of instrumentation in spine surgery, assessment of bone mineral density(BMD) is becoming important in order to optimize outcomes. Hounsfield units(HU), a surrogate for BMD in the lumbar spine, can be measured on standard lumbar CT. Little has been published on BMD variation within each vertebral body(VB). Thus, the purpose of this study was to examine variations in BMD within lumbar VBs.

Materials/Methods

Lumbar CT scans were evaluated in consecutive patients. Patients with fracture, infection, tumor, or prior surgery were excluded. HU were measured on axial cuts: below superior endplate(SE), at the waist(W) and above inferior endplate(IE), and in the anterior(A), middle(M) and posterior(P) 1/3rd at the level of the waist for each VB. T-test was used for within-level comparisons, and repeated measures ANOVA for comparing trends amongst levels.

Results

50 patients (mean age:55 yrs, mean BMI:26.4, M/F: 27:23) were included. - L1: SE was lower than W (p=0.008) and IE (p=0.005), with no difference between A, M and P. - L2: IE was higher than SE (p=0.021) and W (p=0.024); and P was lower than A (p=0.027) and M (p=0.002) - L3: IE was higher than W (p=0.005); and P was lower than M (p-0.002) - L4: IE was higher than SE (p=0.003) and W (p=0.002); and P was lower than A (p=0.001) and M (p<0.0001) - L5: IE was higher than W (p=0.031); and P was lower than A (p<0.0001) and M (p<0.0001) - S1: SE, W and IE were all different (p<0.0001), with a downward trend seen from SE to IE; and P was lower than A (p<0.0001) and M (p<0.0001) Comparison of change from superior-to-inferior and anterior-to-posterior was statistically significant, with S1 differing from all other levels.

Discussion/Conclusion

At all levels, a trend towards higher BMD at IE compared to SE and/or W (except S1), and lower BMD at P compared to A and M was seen. This may be attributable to greater weight-bearing going from superiorly to inferiorly, and transmission of a majority of weight through the anterior column. These findings can help improve selection of implants and guide optimal placement.

RF 01 Combined Annulus Fibrosus Repair And Nucleus Pulposus Replacement In An In-vivo Ovine Study

Kirnaz Sertac, MD1, Sloan Stephen, MS2, Wipplinger Christoph, MD1, Härtl Roger, MD1, Bonassar Lawrence, PhD2

1Weill Cornell Medical College, New York City, New York, United States, 2Cornell University, Ithaca, New York, United States

Background/Introduction

Objective of the current study is to assess the efficacy of combined annulus fibrosus (AF) using high-density collagen (HDC) gel and nucleus pulposus (NP) repair using a hyaluronic acid (HA) gel in an in-vivo sheep model.

Materials/Methods

We performed an anterolateral, retroperitoneal pre-psoas approach to access the IVDs L1-6 in a total of 8 skeletally mature Finn sheep. IVDs were randomized into 5 groups 1) intact, 2) injured via 3x10mm box annulotomy and removal of 200mg of NP; 3) injury and HDC gel patch for AF repair, 4) injury and injection of a HA gel into the NP and 5) injury and HDC AF repair and NP HA replacement. At 6 weeks postoperatively, sheep were sacrificed and underwent post-mortem 3T-MRI scans as well as gross anatomical and histological evaluation. Disc height index (DHI) analysis and Pfirrmann grading (PG) were performed on each segment using MR images.

Results

Intact control discs were not degenerated and had an average PG of 1 while injured, untreated discs had significant degeneration with an average PG of 3. Discs receiving the combined injection and collagen AF patch individually showed fewer signs of degeneration than injured alone and the combined treatment resulted in the least amount of degeneration with PG not significantly different than the intact controls. DHI confirmed the trends seen in the PG, where injured discs lost 20% of the intact disc height, the individual NP and AF repairs restored 5-10% of intact disc height, and the combined repairs preserved 90% of the intact disc height.

Discussion/Conclusion

PG and DHI results demonstrate that individual NP and AF repairs are able to prevent disc degeneration better than no treatment at all, however the greatest preservation of disc health was seen with combined AF and NP repairs.

RF 02. Abaloparatide Enhances Fusion and Bone Formation in a Rabbit Spinal Arthrodesis Model

Moore Harold, BS1, Kumagai Hiroshi, MD2, Hahn William, BS3, Koo Alexander, BS1, Purdue Edward, PhD3, Iyer Sravisht, MD3, Cunningham Matthew, MD, PhD3

1Weill Cornell Medical College, New York City, New York, United States, 2Department of Orthopaedic Surgery, University of Tsukuba, Tsukuba, Ibaraki, Japan, 3Hospital for Special Surgery, New York, New York, United States

Background/Introduction

Lumbar arthrodesis (spinal fusion) is a common orthopedic surgical technique with a broad range of indications. Pseudarthrosis, or failed bony fusion, is a costly complication in a significant number of cases often requiring revision surgery. Recently, the use of osteoporosis drugs has been applied post-operatively to try to reduce the rate of pseudarthrosis after spinal fusion. Abaloparatide is a PTH-related protein analog with anabolic effects on bone formation that is currently FDA-approved for treatment of postmenopausal osteoporosis in women at high risk for fracture. The purpose of this study was to explore the hypothesis that abaloparatide has a significant effect on intervertebral lumbar fusion rate, quality, and volume in a rabbit model.

Materials/Methods

With approval from IACUC for all procedures, 24 skeletally mature male New Zealand White rabbits underwent posterolateral spinal fusion surgery. Starting 4 days postoperatively, half of the rabbits received daily subcutaneous injections of abaloparatide 25µg/kg/d) and the other half received subcutaneous injections of equivalent volume of saline solution. 6 weeks after surgery, all rabbits were euthanized, in vivo radiographs were obtained, and lumbar spine segments were explanted. These spine segments were subjected to blinded manual assessment of fusion (MAF) testing, microCT imaging, biomechanical force testing, and bone histomorphology. Region of Interest (ROI) quantification was performed on microCT images to measure bone fusion mass at the site of fusion. Post-sac radiographs were analyzed using a scoring system (“0” [no bone] through “5” [definite fusion]) and scores were averaged for abaloparatide and control groups.

Results

A total of 21 rabbits completed the study. Successful fusion occurred in 45% (5/11) of control animals vs 100% (10/10) of abaloparatide animals. Radiographic scores yielded a mean of 4.17 in the abaloparatide-treated animals versus 3.39 in control animals (P < .001, unpaired t-test). ROI quantification using microCT found significantly higher bone volume in abaloparatide treated animals (Figure 1A) compared with controls (Figure 1B) (1173mm3 vs. 520mm3, P < 0.001, unpaired t-test) (Figure 1C-D).

Discussion/Conclusion

Abaloparatide increased the rate and quality of lumbar fusion in a rabbit spinal fusion model and may be useful in the clinical setting to reduce the rate of pseudarthrosis after spine surgery.

RF 03. The Effect of Local Steroid Application on Bone Regeneration in a Rodent Posterolateral Spinal Fusion Model

Plantz Mark, BS1, Minardi Silvia, PhD1, Kannan Abhishek, MD1, Nandurkar Tejas, MA1, Paul Jonathan, MPH1, Lyons Joseph, BS1, Marsh Parker, BS1, Yun Chawon, PhD1, Hsu Erin, PhD1, Hsu Wellington, MD1

1Northwestern University Feinberg School of Medicine, Chicago, Illinois, United States

Background/Introduction

Intraoperative local steroid administration has been shown to reduce soft tissue swelling and dysphagia after cervical spine surgery. However, the potential negative effects on bone regeneration have prevented this practice from becoming commonplace. This study aims to assess this risk by evaluating the impact of local steroid on bone healing in both in vitro and in vivo models.

Materials/Methods

Forty-five female Sprague-Dawley rats underwent bilateral L4-5 PLF with absorbable collagen sponge (ACS) containing low-dose rhBMP-2 (1.0 μg per animal). Three groups (n = 15 each) received either no steroid [ACS/rhBMP-2], low-dose triamcinolone (0.5 mg/kg) [ACS-LD/rhBMP-2], or high-dose triamcinolone (2.5 mg/kg) [ACS-HD/rhBMP-2], administered locally. Harvested spines underwent manual palpation to obtain a fusion score, wherein 0 = no fusion, 1 = unilateral bridging bone, and 2 = bilateral bridging bone. A fusion score of 1 or greater indicates successful fusion. Micro-computed tomography was used to quantify new bone volume. Histological analysis was used to assess for new bone and collagen formation. In vitro, hBM-MSCs were grown in 3D cultures on ACS with comparable doses of triamcinolone. Gene expression of pro-osteogenic markers (RUNX2, SP7, ALPL, COL1A1, BGLAP, SPP1) and alkaline phosphatase (ALP) activity assays were used to assess for differences in osteogenic differentiation.

Results

There was no difference in fusion score (FS), fusion rate (FR), or new bone volume (BV) between the control group (FS: 0.58 ± 0.58, FR: 38%, BV: 32.37 ± 22.43 mm³), low-dose steroid (FS: 0.62 ± 0.59, FR: 57%, BV: 40.08 ± 23.20 mm³), and high-dose steroid (FS: 1.09 ± 0.81, FR: 33%, BV: 25.50 ± 21.75 mm³) groups, assessed by one-way ANOVA and post-hoc Tukey’s t-test. Representative microCT and histological images from each group (ACS/rhBMP-2, ACS-LD/rhBMP-2, and ACS-HD/rhBMP-2) are summarized in figure 1. For in vitro assessment, there were no differences in the expression of pro-osteogenic markers or ALP activity between the groups.

Discussion/Conclusion

Local triamcinolone – at both a clinical dose (0.5 mg/kg) and a supratherapeutic dose (2.5 mg/kg) – did not significantly influence multiple indicators of bone regeneration in this model. These results suggest that the clinical benefits of local steroid administration may outweigh the theoretical risks.

RF 04. Recent Complications in Spinal Surgery Performed by American Board of Orthopaedic Surgery (ABOS) Part II Candidates Over the Last Decade

Yeung Caleb, MD1, Schoenfeld Andrew, MD2, Kang James, MD2, Makhni Melvin, MD2

1Massachusetts General Hospital, Harvard Medical School, Boston, MA, United States, 2Brigham and Women's Hospital, Harvard Medical School, Boston, Massachusetts, United States

Background/Introduction

Orthopaedic surgeons are eligible to take the ABOS Part II examination 20 months following fellowship completion. The ABOS keeps a database of all cases performed including their indications and complications. Despite previous studies utilizing ABOS Part II data in other subspecialties, there is a paucity of information regarding complications of spine surgeons in their first two years of independent practice. Such data can help trainees anticipate potential rates and types of complications they will face early in practice and hold better-informed patient discussions.

Materials/Methods

We performed a retrospective review of surgical cases submitted to the ABOS by candidates taking the Part II Spine examination between 2009-2018. Chi-squared tests were used for categorical data and the T-test for analysis of means, with a p-value of <0.05 considered significant.

Results

37,539 patients were analyzed (average 3,754 cases each year; 49 cases performed per candidate per year). 760 candidates took the exam between 2009-2018 (average 76 per year). Average patient age was 54.7 years, with a relatively equal distribution of males/females. Overall mortality rate was 0.5%; re-operation rate was 5.2%; re-admission rate was 5.7%. Overall medical complication rate was 8.5% and overall surgical complication rate was 15.6%, representing a 24.7% complication rate overall. Incidence of surgical and overall complications increased significantly with a 19.7% and 20.2% higher incidence rate ratio, respectively, between 2008 and 2017 (p<0.001 and p=0.002). By contrast, mortality decreased significantly, with a 52.2% lower incidence rate ratio between 2008 and 2017 (p=0.013).

Discussion/Conclusion

We noted a higher overall complication rate (24.7%) than the literature-reported rate of 16.4%. Surgical and overall complications appear to be increasing while mortality is decreasing. This may be multifactorial in the setting of higher acuity patients, more frequent reporting or assessment of complications, or changes in training, and further studies may help to elucidate the contribution of each of these factors. Encouragingly, however, mortality is decreasing, suggesting that though complication rates appear to be rising, overall patient safety particularly with regard to life-threatening complications is improving.

RF 05. The Effect of Lumbar Corticosteroid Injections on Postoperative Infection in Lumbar Arthrodesis Surgery

Fredericks Donald, MD1, Pisano Alfred, MD1, Seavey Jonathan, MD2, Steelman Theodore, MD3, Wagner Scott, MD1, Helgeson Melvin, MD2

1Walter Reed National Military Medical Center, Bethesda, Maryland, United States, 2Walter Reed Army Medical Center, Bethesda, Maryland, United States, 3US Army, Walter Reed National Military Medical Center, Bethesda, Maryland, United States

Background/Introduction

Although widely utilized, the effect of lumbar corticosteroid injections on the outcomes of subsequent surgery is unclear. Some literature suggests an elevated risk of infection while several other recent series found no significant association between preoperative corticosteroid injections and postoperative infection. Objective: Characterize the association between lumbar corticosteroid injections and postoperative infection rate for patients in the Military Health System undergoing lumbar arthrodesis.

Materials/Methods

The Military Health System Data Repository was searched for all patients undergoing lumbar arthrodesis from 2009 to 2014. CPT codes were used to identify the subset of patients who also received preoperative lumbar corticosteroid injections. These patients were stratified by timing, type, and number of injections. Infection rates were compared to the control group of patients who did not receive preoperative lumbar corticosteroid injections.

Results

The search identified 3,403 patients who had undergone lumbar arthrodesis from 2009-2014 within the Military Health System. 612 patients had received lumbar corticosteroid injections prior to surgery (348 epidural, 264 facet). The control group consisted of the remaining 2,791 patients. Overall post-operative infection rate was 1.47% with an infection rate in the injection group of 1.14% versus 1.54% in the control group. When stratified by time, infection rates ranged from 0% to 1.85% in the injection groups. No differences between injection and control groups reached statistical significance in any subgroup analysis.

Discussion/Conclusion

Post-operative infection rate is not significantly increased in patients receiving LCSIs prior to lumbar arthrodesis. No differences were observed in infection rates based on timing, type, or number of injections prior to surgery.

RF 06. Clinical and Radiographic Outcomes of Surgical Management for Lumbar Degenerative Spondylolisthesis

Narain Ankur, MD1, DiSegna Steven, MD1, Carroll Thomas, BA1, Holmes Sara, BS1, Connolly Patrick, MD1, DiPaola Christian, MD1, Stauff Michael, MD1

1University of Massachusetts Memorial Medical Center, Worcester, Massachusetts, United States

Background/Introduction

Degenerative lumbar spondylolisthesis is a common cause of pain and disability. A variety of surgical treatments exist, including: decompression, decompression with non-instrumented fusion, and decompression with instrumented fusion. Instrumented fusion can also be used in conjunction with interbody techniques. Identification of a preferred surgical technique is controversial. The aim of this study is to compare the clinical and radiographic outcomes of surgical management for lumbar degenerative spondylolisthesis among various techniques using a prospectively collected registry.

Materials/Methods

A prospectively maintained registry of patients undergoing operative treatment for degenerative lumbar spondylolisthesis was reviewed. Patients with at least 6 months follow-up were included. Variables collected included demographics, operative characteristics, patient-reported outcomes (PROs), and radiographic measurements. PROs included the Short-Form 8 Health Survey (SF-8), Oswestry Disability Index (ODI), and Sciatica Bothersome Index (SBI). Analyses were performed on the entire cohort and stratified by procedures with adequate representation. These procedures included decompression plus non-instrumented fusion (NIF, n=6), posterior spinal instrumented fusion (PSIF, n=17), or transforaminal lumbar interbody fusion (TLIF, n=31). One-way analysis of variance or chi-square analysis were used to compare demographic and operative characteristics. Paired t-tests and linear regression were used to evaluate changes in PROs.

Results

63 patients were included. Average age was 65.1 years, and average BMI was 32.3 kg/m2. For the entire cohort, there were significant improvements in ODI (preoperative 59.4 vs. postoperative 37.8), SBI (14.1 vs 7.6), and SF-8 physical composite score (PCS, 31.4 vs 41.2) at final follow-up. When stratified by procedure type, there were significant improvements in ODI and SF-8 PCS for NIF, PSIF, and TLIF groups. Only the TLIF group reached a minimal clinically important difference (MCID) in ODI (ΔODI = -27.7). Radiographic measures including slippage, dynamic slippage change, and disk height were not associated with PRO improvement differences.

Discussion/Conclusion

This preliminary analysis demonstrated improvements in PROs after all types of operative treatment for lumbar degenerative spondylolisthesis. However, only patients undergoing TLIF reached MCID for ODI, suggesting that interbody fusions may provide greater clinical benefit. Continued recruitment of patients into this registry is required and will provide opportunities to compare the efficacy of surgical treatment options for degenerative spondylolisthesis.

RF 07. Risk Factors for 30 and 90 day Readmission after Lumbar Decompression Surgery

Reyes Ariana, BS1, Canseco Jose, MD, PhD1, Divi Srikanth, MD1, Kurd Mark, MD1, Anderson D., MD2, Rihn Jeff, MD1, Hilibrand Alan, MD1, Kepler Christopher, MD, MBA1, Vaccaro Alexander, MD, PhD, MBA1, Schroeder Gregory, MD1

1Rothman Institute at Thomas Jefferson University, Philadelphia, Pennsylvania, United States, 2Department of Orthopaedic Surgery, Rothman Institute, Thomas Jefferson University, Philadelphia, Pennsylvania, United States

Background/Introduction

Past studies have reported risk factors for hospital readmission after lumbar spine surgeries; however, there is a paucity of literature on readmission predictors after lumbar decompression alone. The purpose of this study was to determine risk factors for inpatient readmission after isolated lumbar decompression surgery.

Materials/Methods

After Institutional Review Board (IRB) approval, patients over the age of 18 who underwent 1- to 4-level lumbar decompression surgery between 2017 and 2018 at a single, academic institution were retrospectively identified. Patients were divided into three groups based on postoperative readmission status: 1) no inpatient readmission or related events, 2) inpatient readmission, or 3) ED/Observation Unit/Urgent care visit evaluation without inpatient readmission. Demographics and medical history were extracted from the electronic medical record. Univariate analysis was used to compare demographics and comorbidities between groups. Backwards stepwise regression analysis was performed to determine independent predictors for 30- and 90-day readmission.

Results

A total of 2636 patients were included in the final analysis. There was a significant difference between the groups for age (p<0.001), American Society of Anesthesiologists (ASA) classification (p=0.013), and length of stay (p<0.001). The no readmission group had a lower median age 57.0 [44.0;68.0], higher proportion of patients with an ASA of 1 (27.2%), and higher proportion of patients who were non-smokers (64.7%) compared to the other two groups. The inpatient readmission cohort had the highest proportion of current smokers (20.3%). Regarding patient comorbidities, there was a significant difference between the three groups for renal failure (p=0.047), rheumatoid arthritis (p=0.026), bleeding disorders (p=0.047), and diabetes (p=0.027). In the final logistic regression model, age (OR:1.018 [1.004, 1.033], p=0.013), current smoking status (OR:2.027 [1.298, 3.164], p=0.002), length of stay (OR:1.083 [1.011, 1.160], p=0.022), and a history of renal failure (OR: 2.551 [1.027, 6.339], p=0.044) were found to be significant predictors for inpatient readmission after lumbar decompression surgery. The AUC for the final model was determined to be 0.68.

Discussion/Conclusion

Overall, age, current smoking status, length of stay, and a history of renal failure were found to be significant independent predictors of 30- and 90-day readmission after lumbar decompression surgery.

RF 08. The Re-herniation After Decompression (RAD) Score Identifies Patients at Low-Risk for Re-herniation After Lumbar Decompression Surgery

Harada Garrett, MD1, Basques Bryce, MD1, An Howard, MD1, Samartzis Dino, PhD2

1Midwest Orthopaedics at Rush, Chicago, Illinois, United States, 2Midwest Orthopaedics at Rush University, Chicago, Illinois, United States

Background/Introduction

Re-herniation is a common, though unpredictable occurrence after surgical decompression for a herniated nucleus pulposus (HNP) that may lead to adverse patient outcomes. The purpose of this study was to create a prognostic tool to identify patients at greatest risk for re-herniation after surgery.

Materials/Methods

We performed a retrospective examination of patients receiving decompression of the lumbar spine for an HNP from 2006-2018. Baseline demographics (age, sex, BMI, American Society of Anesthesiologist Class, smoking history), clinical (symptom duration, herniation type, operative levels), and radiographic information (pelvic tilt, pelvic incidence, sacral slope, coronal angulation, lumbar lordosis) was collected in addition to initial and future herniation events. Re-herniations were identified on magnetic resonance imaging and included operative and non-operative cases. Multiple imputation and LASSO logistic regression was employed to build and validate a predictive model to identify patients at differing levels of risk for re-herniations after decompression surgery.

Results

A total of 517 patients (38.5% female) presented for surgical decompression of a lumbar HNP, in which 51 experienced re-herniations. Patients had a mean age of 45.2±13.9 years, BMI of 29.4±6.3 kg/m2, and 22.4±28.0 months of follow-up. Most herniations were paracentral (83.4%), followed by central (10.6%), and far-lateral (5.6%). Re-herniations were seen at a mean of 46.0±70.0 months after surgery. LASSO identified nine predictors for re-herniations. (Table 1) Validation of this model demonstrated excellent discrimination (AUC:0.834), sensitivity (71.4%), and specificity (87.6%). Within the testing cohort, 16.3% (17/104) of patients were identified as high-risk. Of these, 29.4% sustained re-herniations, while nearly all low-risk patients (98.7%, 85/87) experienced no recurrence.

Discussion/Conclusion

This is the first study to develop and validate a tool identifying patients with differences in risk for re-herniation after lumbar decompression. With further validation, this tool can be implemented in practice, and may accurately identify patients likely to experience lasting surgical benefit for lumbar disc herniations.

RF 09. Fusion and Decompression versus Decompression only for Lumbar Facet Cysts

Section Jarren, MD1, Maqsoodi Noorullah, BS2, Mesfin Addisu, MD1

1University of Rochester Orthopedics, Rochester, New York, United States, 2University of Rochester, Univ. of Rochester School of Medicine and Dentistry, Rochester, Michigan, United States

Background/Introduction

Lumbar facet cysts are associated with radiculopathy, neurogenic claudication and a higher risk of dural tears. Consensus on the best method to address these lesions is lacking (fusion versus decompression only). Outcomes after surgery are of increasing interest to spine surgeons, particularly when seeking to offer the most conservative, yet effective, option. Our objective is to describe complications, outcomes and recurrence rates following surgical intervention of facet cysts.

Materials/Methods

A retrospective analysis of patients diagnosed with lumbar facet cysts who underwent treatment for their condition at a single academic tertiary care facility between 2008 and 2018. Patients, underwent surgical treatment (fusion + decompression [DF] or decompression only [DO]). All patients were evaluated for the surgical approach, facet cyst level, post-op complication, cyst recurrence, EBL and length of hospital stay. Adjacent segment disease rates were also assessed.

Results

Fifty-three patients were assessed. The mean age was 65 (43-93) including 27F and 26M. Thirty-four patients underwent DO and 19 patients underwent DF. The majority of the cysts 58% (n=32) were at L4-L5, followed by 22% (n=12) at L5-S1, and 16% (n=9) at L3-L4. The recurrence rate in the DO group was 14.7% (5/34). 3 patients underwent additional surgery for these recurrences. There were no cyst recurrences in the DF group. The DF patients had significantly higher blood loss (avg: 394.2ml) compared to the DO patients (avg: 106.3ml) (P <0.01). Dural tear rates were similar in the DO patients (8.8%) vs. DF (5.3%) (P= 0.58). The DF group had increased length of hospitalization (5 days) compared to the DO group (2.4 days) (P <0.01). There were higher rates of adjacent segment disease 14.3% (3/21) in the DF group requiring further surgeries.

Discussion/Conclusion

Patients that undergo DF have an increased risk adjacent segment degeneration and need for revision procedures. However, three patients treated initially with DO underwent revision procedures with fusions. The rate of cyst recurrence is higher among those treated by decompression alone.

RF 10. The Effect of Obesity on Perioperative Morbidity During Oblique Lumbar Interbody Fusion

Xi Zhuo, MD, PhD1, Guinn Jeremy, BS2, Burch Shane, MD3, Mummaneni Praveen, MD2, Mayer Rory, MD4, Chou Dean, MD2

1University of California, San Francisco, San Diego, California, United States, 2Department of Neurological Surgery, UCSF Medical Center, San Francisco, California, United States, 3Department of Orthopedic Surgery, University of California San Francisco, San Francisco, California, United States, 4UCSF Spine Center University of California, San Francisco, San Francisco, California, United States

Background/Introduction

Obese patients have been shown to have longer operative times and more complications from surgery. However, for obese patients undergoing minimally invasive surgery (MIS), these differences may not be as significant. In the lateral position, it is thought that the obesity is less of an issue because gravity pulls the visceral fat away from the spine; however, this observation is primarily anecdotal and expert opinion. We aim to report the effect of obesity on the oblique lateral lumbar interbody fusion (OLIF) and perioperative morbidity in obese and non-obese patients.

Materials/Methods

A retrospective review of patients undergoing OLIF by three spine surgeons and one vascular surgeon at the University of California San Francisco from 2013 to 2018 was performed. Data collected included demographic variables, approach-related factors such as operative time, blood loss, expected temporary approach related sequalae, and overall complications. Patients were categorized according to their body mass index (BMI). Obese was defined as BMI ≥30, and severely obese was defined as BMI ≥35.

Results

There were 238 patients, 95 males and 143 females. There was no significant difference between the obese and non-obese groups in terms of gender, levels fused, or smoking status. For the entire cohort, there was no difference in operative time, blood loss, or complications between obese and non-obese patients. However, a subset analysis of the 77 multi-level OLIFs which included L5-S1 demonstrated that the operative times for the non-obese group was 223.55 ±57.93 minutes and 273.75 ±90.07 minutes for obese group (p=0.004). In this subset, the expected approach related sequelae rate was 13.2% for the non-obese group and 33.3% for the obese group (p=0.039). The two subset groups had similar blood loss (p=0.476) and complication rates (p=0.876).

Discussion/Conclusion

Obesity and morbid obesity generally do not increase the operative time, blood loss, approach related sequelae, or complications following OLIF. Obese patients who undergo multi-level OLIF which includes the L5-S1 level have longer operative times and a higher rate of expected approach related sequelae. Obesity should not be considered a contraindication to multi-level OLIF, but patients should be informed of potentially increased morbidity if including the L5-S1 level.

RF 11. Teriparatide Treatment Increases Hounsfield Units in the Vertebral Body out of Proportion to the Pedicles and Lamina of the Lumbosacral Spine

Mikula Anthony, MD1, Bernatz James, MD2, Fogelson Jeremy, MD1, Nassr Ahmad, MD1, Freedman Brett, MD1, Currier Bradford, MD3, Bydon Mohamad, MD1, Yaszemski Michael, MD, PhD1, Anderson Paul, MD4, Elder Benjamin, MD, PhD1

1Mayo Clinic Rochester, Rochester, Minnesota, United States, 2University of Wisconsin Department of Orthopedic Surgery, Madison, Wisconsin, United States, 3Mayo Clinic College of Medicine, Rochester, Minnesota, United States, 4University of Wisconsin-Madison, Madison, Wisconsin, United States

Background/Introduction

Teriparatide treatment has been used to improve bone density when considering spinal fusion surgery in patients with osteoporosis. This treatment has been shown to improve Hounsfield units (HU), a standardized measurement for radio density that can be obtained on routine CT scans, in lumbar vertebral bodies but its effect on other anatomic structures within the lumbosacral spine is unknown. The purpose of our study was to determine the effects of teriparatide on the HU of the pedicles, lamina, and vertebral bodies of the lumbosacral spine.

Materials/Methods

A retrospective chart review was performed from 1997 to 2018 across all campuses at our institution. We identified patients who had been treated with at least 6 months of Teriparatide with a lumbar or abdominal CT scan before and after treatment. HU were measured in the pedicles, lamina, and vertebral bodies from L1-S2. Levels with spinal hardware were excluded, with 58% of patients having at least one level. Measurements were made by two independent reviewers.

Results

Fifty-two patients (46 women, six men) were identified for analysis with an average age of 67 years who underwent an average of 20.9±6.5 months of Teriparatide therapy. Mean (± standard deviation) HU improvement in the vertebral bodies throughout the lumbosacral spine (L1-S2) was from 110±49 to 134±58 (p-value=0.029), a 22% improvement. HU in the pedicles increased from 174±75 to 199±85 (p-value=0.11), a 14% improvement. The lamina improved from 325±129 to 377±121 (p-value=0.039), a 16% improvement. There was no correlation between pedicle size and improvement in HU (Pearson correlation coefficient -0.0005, p-value=0.99).

Discussion/Conclusion

Teriparatide treatment improved Hounsfield units in the vertebral body out of proportion to the improvement observed in the pedicles and lamina from L1-S2. This may indicate that improved screw fixation following Teriparatide therapy is largely due to fixation within the vertebral body, but further biomechanical studies are warranted.

RF 12. Retrospective Assessment of the Use of Liposomal Bupivacaine in Lumbar Fusions in Immediate Post-operative Hospital Care

Katsevman Gennadiy, MD1, Allison Andrew, PharmD, Fang Wei, PhD, Confer Jennifer, PharmD, CPS, BCCCP, Elhamdani Shahed, MD, Hoyt Alastair, MD, Garavaglia Jeffrey, PharmD, BCPS, Marsh Robert, MD, PhD

1West Virginia University, Morgantown, West Virginia, United States

Background/Introduction

Liposomal bupivacaine (LB) is FDA-approved for administration into surgical sites for post-operative analgesia. The liposomal formulation allows for sustained effects up to 72 hours, making it an appealing adjunct in multi-modal analgesia for surgery.

Materials/Methods

A retrospective study assessed the immediate post-operative course in patients undergoing lumbar interbody fusion with and without the use of LB. Visual analog scale (VAS) pain scores and amount of opioids consumed were recorded at 12 hour intervals for 72 hours post-operatively.

Results

A total of 122 patients (97 LB versus 25 control group) were reviewed. Median hospital length of stay was shorter in the LB cohort compared to controls (1.94 versus 3.08 days, respectively; p=0.0043). When assessing the percentage of discharges between the two groups at 12-hour intervals, there were significantly more discharges in the LB cohort at 36-48 hours (p=0.0226). There was a significant decrease in intravenous opioids consumed at 48-60 hours in the LB cohort compared to controls (p=0.0494), a difference not detected at other time points or with oral or total opioids. Mean VAS scores were significantly higher in LB cohort compared to controls at 0-12 hours (5.2 versus 3.9, respectively; p=0.0079), but insignificantly different subsequently up to 72 hours. The LB cohort and controls were not significantly different in total amount of opioids consumed, overall pain scores, or with regards to how the opioid amount consumed or pain scores changed over time. However, the amount of opioids consumed and pain scores were statistically significantly different between individual time points within groups, with a general trend for decreased opioids consumption and VAS scores over time.

Discussion/Conclusion

The use of LB in spine fusions may increase pain acutely due to the trauma of injection although its sustained effect may improve muscular pain at ∼48 hours post-operatively, resulting in fewer intravenous opioids required and earlier discharges.

RF 13. Rate of Reoperation Following Limited Microscopic Decompression of Lumbar Facet Cysts

Thompson Kirk, MD1, Johnson Nicholas, MD2, Neiberger Angelina3, Stokman James, MD1, Darden Bruce, MD1, Spector Leo, MD1, Milam Alden, MD1, Laxer Eric, MD1, Segebarth P. Brad, MD1, Chapman, Jr. Todd, MD1

1OrthoCarolina Spine Center, Charlotte, North Carolina, United States, 2Carolina Medical Center Atrium Health, charlotte, United States, 3Davidson College, Davidson, North Carolina, United States

Background/Introduction

The surgical treatment of lumbar facet cysts has long been debated. While some surgeons elect for limited microscopic decompression of such cysts, other surgeons advocate for primary decompression and fusion over concerns for recurrence and instability. Previous studies have suggested the rate of reoperation following facet cyst excision to be as low as 2%, while others have reported this rate to be markedly higher.

Materials/Methods

A retrospective review of the electronic medical records of patients treated at our institution using the CPT code 63267 for laminectomy for excision or evacuation of intraspinal lesion other than neoplasm, extradural, before January 2018 was conducted. Inclusion criteria included 1) operative treatment with facet cyst resection and limited decompression, 2) >2 years from index procedure and 3) age >18 years of age at the time of surgery. The primary outcome measure was revision surgery at the index level for recurrence or instability. Secondary outcome measures include radiographic assessments of preoperative and postoperative spondylolisthesis and VR12 and ODI patient-reported outcome measures.

Results

162 patients met inclusion criteria. 25 patients have minimum 2-year follow-up in clinic without undergoing revision procedures and 120 are being contacted to obtain 2-year study data. So far, 17 patients have undergone revision surgery at the index level. The revision rate for this group was 40.5%. There were 12 revision decompression/fusion procedures (70.6%) and 5 revision decompressions (29.4%). 4 patients (9.5%) underwent a second reoperation and 1 patient (2.4%) underwent a third re-operation. There were no differences between the groups with regard to the presence of pre-operative spondylolisthesis, or patient demographics.

Discussion/Conclusion

If the revision rate following limited decompression for lumbar facet cysts is higher than is currently proposed in the literature, alternative techniques may be considered during the index procedure. Our revision rate of 40.5% may underestimate revisions as it assumes patients without two-year follow-up did not undergo revision surgery outside of our facility. This data should help guide surgeons during the informed consent discussion prior to surgical decompression of lumbar facet cysts. Final analysis will include 2-year phone follow-up.

RF 14. TLIF Operations Employing Limited Midline Exposure and Cortical Trajectory Fixation Lead to Superior Early Term Patient Reported Outcomes

Siyaji Zakariah, BS1, Piracha Ali, BS1, Khalid Alina, BS1, Kittaneh Haneen, BS1, Hornung Al, BS1, Maschhoff Clayton, BS1, Alam Omar, BS2, Qadri Talha, BS2, Harada Garrett, MD3, Colman Matthew, MD1

1Midwest Orthopaedics at Rush University, Chicago, Illinois, United States, 2Rush University Medical Center, IL, Illinois, United States, 3Midwest Orthopaedics at Rush, Chicago, Illinois, United States

Background/Introduction

Cortical bone trajectory transforaminal lumbar interbody fusion (TLIF) screw fixation is a relatively new technique that currently lacks evidence describing its outcomes. The aim of this study was to compare radiographic and patient-reported outcomes between open TLIF, posterolateral lumbar fusion (PLF), and cortical TLIF approaches.

Materials/Methods

We retrospectively examined all patients receiving a TLIF, PLF, or cortical TLIF for degenerative lumbar spine pathology from 2017-2019. Demographics (age, sex, BMI), comorbidities (diabetes, American Society of Anesthesiologists [ASA] Class), and radiographic parameters (segmental lordosis [SL], lumbar lordosis [LL], pelvic incidence [PI], pelvic tilt [PT], PI-LL mismatch) were obtained and further analyzed. Patient-reported outcomes included Oswestry Disability Index (ODI), Visual Analogue Scale (VAS)-back, VAS-leg, 12-item Short Form (SF-12), and Veterans RAND 12 (VR-12) scores. Comparison of baseline characteristics was performed using chi-square and ANOVA. Multivariate regression was used to assess outcomes associated with procedure type.

Results

A total of 23 cortical TLIF, 63 TLIF, and 89 PLF patients were identified for study (Table 1). Mean ages were 58.7±12.8, 65.3±10.5, and 69.1±11.3 years for cortical TLIF, TLIF, and PLF patients respectively (p<0.001). Given the recency of the cohort, mean follow-up was 6.97±5.61, 8.00±6.12, and 9.15±6.43 months for cortical TLIF, TLIF, and PLF patients, respectively. There were no differences in complication rate or radiographic parameters. Patients receiving a cortical TLIF had significantly better 5-weeks postoperative patient-reported outcomes across all surveys (ODI:p<0.001, VAS-back:p=0.008, VAS-leg:p=0.025, SF-12:p=0.012, and VR-12:p=0.004). Further, cortical TLIF patients had superior final follow-up VAS-leg (p=0.026) and VR-12 scores (p=0.018).

Discussion/Conclusion

Compared to open TLIF, cortical TLIF had superior postoperative and final follow-up patient-reported outcomes. This suggests that cortical TLIF may be an acceptable alternative to PLF or open TLIF which leverages the advantages of interbody reconstruction in a less-invasive setting. Future long-term studies are warranted to compare the longevity of these outcomes.

RF 15. Consumer Price Discrepancies for Lumbar Discectomy Across US Hospitals

Butler Alexander, MD1, Rivera Sebastian, MD1, Donnally Chester, MD1, Ghasem Alexander, MD1, Granger Caroline, BS1, Li Deborah, BS1, Wang Michael, MD1

1University of Miami, Miami, Florida, United States

Background/Introduction

The concept of consumer driven health care is of growing interest and application in reducing the burden of healthcare costs on the American economy. A small number of previous studies have described poor price availability for surgical procedures, though none in the field of spine surgery. The authors sought to determine the availability of consumer price information for a common spine surgery in the United States and to identify discrepancies in prices provided.

Materials/Methods

153 hospitals in the United States were contacted via telephone calls under the guise of a patient requesting a self-pay price for elective lumbar discectomy. Complete and partial prices were recorded, when available. Hospitals were grouped based on profit status (nonprofit, government, private), teaching status (nonteaching, teaching), and geographical region (Northeast, South, Mid-west, West). Statistical analysis was performed to compare rates of price availability and mean prices between hospital groups.

Results

34 (23.0%) of a total of 148 hospitals included in final analysis were able to provide complete price information. An additional 70 (47.3%) were able to provide a partial price. Teaching hospitals had a significantly higher rate of complete price availability compared to non-teaching hospitals (27.8% teaching vs. 10.0% non-teaching), while non-teaching hospitals had a higher rate of partial price availability compared to teaching (65.0% non-teaching vs. 40.7% teaching) (p=.02). The mean complete price for microdiscectomy when provided was $27,342.36 (n=40), and the mean partial price when provided as a facility fee only was $13,857.50 (n=64). No significant differences in mean complete prices were determined based on hospital profit status, teaching status or geographic region. When compared to government and non-profit hospitals combined, private hospitals had significantly lower partial-prices ($14,666 + 8,469 (n=63), $7,386 + 5,663 (n=7), p=0.03).

Discussion/Conclusion

A cost-conscious patient seeking to undergo a common spine surgery in the United States will likely be met with difficulty and few transparent options. A high degree of variability exists with regards to accessibility and comprehensiveness of pricing information. Efforts to define and improve value of care will likely be challenged by a lack of understanding of cost.

RF 16. Robotic-assisted LSF Surgery Leads to Increased Risk of Revision Surgery, Readmission, Complications, and Prolonged Opioid Utilization Compared to Non-robotic LSF Surgery

Yang Daniel, BS1, Li Neill, MD2, Patel Shyam, MD2, Kleinhenz Dominic, MD2, Daniels Alan, MD2

1Alpert Medical School of Brown University, Providence, Rhode Island, United States, 2Alpert Medical School of Brown University, Department of Orthopaedics, Providence, Rhode Island, United States

Background/Introduction

Surgeons have increasingly adopted robotic-assisted spinal fusion in efforts to increase accuracy of pedicle screw placement. However, it remains to be demonstrated whether the technique offers advantages with regards to need for revision surgery, readmission, postoperative complication rates, and opioid use.

Materials/Methods

The PearlDiver Research Program (www.pearldiverinc.com) was used to identify patients undergoing primary LSF between the years of 2007 and 2017. These patients were divided into two cohorts: 1) LSF with robotic assistance (CPT-0054T, CPT-0055T, CPT-20985, CPT-61783 coded the same day as surgery) and 2) LSF without robotic assistance. Statistical matching was performed using demographic and relevant comorbidity variables. Postoperative outcomes were assessed for each cohort with multivariable logistic regression. All regression modeling controlled for age, sex, race, and Charlson Comorbidity Index. A p-value of .05 was set as the level of significance.

Results

Matching between cohorts of robot-assisted LSF and unassisted LSF yielded 2528 patients in each matched cohort. Risk of 30 day readmission was increased for robot-assisted LSF patients (13.7% vs. 10.3%, aOR=1.39, 95%CI 1.17-1.65, p=0.0002) (Table 1). Comparing the rate of 30 day complications, this was 17.1% in robot-assisted LSF and 12.2% in unassisted LSF patients, showing increased risk (aOR=1.50, 95%CI 1.27-1.76, p<0.0001). Robot-assisted LSF patients had higher risk of respiratory (aOR=1.56, 95%CI 1.21-1.90, p=0.0006), surgical site (aOR=1.56, 95%CI 1.14-2.16, p=0.0061), and implant-related complications (aOR=1.74, 95%CI 1.20-2.54, p=0.0038). Robot-assisted LSF patients were more likely to fill opioid prescriptions at all studied postoperative time points (p<0.0001). Robot-assisted LSF patients experienced an estimated 1.8 times the risk of revision compared to unassisted LSF patients (12.9% vs. 7.6%, aOR=1.80, 95%CI 1.48-2.18, p<0.0001). Whereas in unassisted LSF 46.5% of revisions were due to instrument failure and 14.1% were due to infection, in robot-assisted LSF 36.6% of revisions were due to instrument failure and 21.5% were due to infection (p<0.0001).

Discussion/Conclusion

This study found that robot-assisted LSF led to increased rate of revision surgery, readmission, postoperative complications and opioid use compared to conventional LSF. These findings suggest a need for further studies evaluating the efficacy of robot-assisted surgery before more widespread adoption of this technology.

RF 17. Radiographic Analysis of Lumbar Disc Arthroplasty Reoperations

Shah Darshan, MD1, Holauchock Griffin, PA-C1, Bates Taylor, MD1, Childs Benjamin, MD2, Jorgensen Anton, MD1

1San Antonio Military Medical Center, San Antonio, Texas, United States, 2William Beaumont Army Medical Center, El Paso, Texas, United States

Background/Introduction

Lumbar disc arthroplasty (LDA) is a motion preserving treatment for degenerative disc disease. Reoperations may be related to either poor patient selection or errors in surgical technique. To date, there has been little investigation into the surgical parameters for proper implant placement that can decrease the risk for revision. The goal of our study was to critically evaluate radiographic parameters of LDA that are associated with higher risk of reoperation to help guide implant placement.

Materials/Methods

This was a retrospective analysis of lumbar disc arthroplasties performed at a single academic institution from 2011-2017 with minimum 2 year follow up. Radiographic analysis included measurement of pre/post-operative disc level lordosis, functional spinal unit lordosis, and disc height. Implant containment within the disc space was measured as well as coronal plane centrality. Follow up was a minimum of 2 years. Patients who underwent a reoperation of the lumbar disc arthroplasty level were compared to patients who did not have reoperation. Logistic regression was used to determine radiographic parameters associated with odds of reoperation

Results

Between 2011 and 2017 83 LDAs were performed. 25 cases were excluded due to incomplete records or insufficient follow up. In the final analysis we were able to include 14 implants that underwent reoperation and 44 implants as controls. Average follow up for controls was 3.8±1.5 years and 4.6±2.1 years for cases. There was a significant difference in the mean preoperative disc height (14.0±3.4 vs. 11.1±3.7, p=0.031) and post-operative disc height (22.9±2.9 vs 21.0±2.8, p=0.04) for implants that underwent reoperation versus controls. Implants that were not completely contained within the boundaries of the upper and lower vertebral bodies had 5.4 times higher odds (95% CI [1.5-19.6]) to undergo reoperation. Devices not contained within the boundaries of the lower vertebral body had 4.5 times higher odds (95% CI [1.3-16.2]) to undergo reoperation compared to those that were contained.

Discussion/Conclusion

The findings of this study suggest that increased pre-operative and post-operative disc height and poor implant containment within disc space boundaries of the upper and lower vertebral body are associated with a higher risk of LDA reoperation.

RF 18. Primary Versus Revision: A Comparison of Outcomes Following Minimally Invasive Lumbar Decompression

Parrish James, MPH1, Jenkins Nathaniel, MS1, Hrynewycz Nadia, BS1, Brundage Thomas, BS1, Kazmi Ali1, Katiki Aishwarya1, Singh Kern, MD1

1Rush University Medical Center, IL, Illinois, United States

Background/Introduction

Lumbar spinal stenosis may lead to significant long term symptoms, such as radicular back and leg pain, and can be effectively treated with decompression surgery. Research is limited on the postoperative outcome differences between patients undergoing primary and revision lumbar decompressions. The purpose of this study is to investigate the differences in postoperative outcomes among primary and revision minimally invasive lumbar decompression procedures.

Materials/Methods

A prospective surgical registry was reviewed for lumbar decompression spine surgery patients between 2013 and 2019. Inclusion criteria were primary or revision, single or multilevel lumbar decompressions. Patients who did not complete preoperative surveys were excluded. Patient were subgrouped by primary or revision procedure. A chi-squared analysis, Students’ t-test, and linear regression were utilized as appropriate to compare baseline characteristics, perioperative variables and PROs. Pre- and postoperative (6-weeks, 12-weeks, 6-months, and 1-year) scores collected included: Visual Analogue Score (VAS) back, VAS-leg, Oswestry Disability Index (ODI), 12 Item Short Form Physical Composite Score (SF-12-PCS), and Patient-Reported Outcomes Measurement Information System physical function (PROMIS-PF) (Table 1).

Results

516 patients were included. The cohort mean age was 46.7 years. 31.1% were female and 42.4% were obese. Operative time was the only perioperative characteristic that was significantly different between groups (45.6 vs 51.4, p=0.017). Preoperative mean differences were observed among primary and revision groups with SF-12-PCS and PROMIS-PF (p=0.042, p=0.033). Both subgroups had significant improvement for all PROs for both primary and revision, with the exception of VAS back at 6-months (p=0.474). At the 1-year evaluation, there were no differences between primary and revision PRO scores. Postoperative differences between primary and revision groups were observed for VAS back (12-weeks p=0.011, 6-months p=0.003), VAS leg (6-months p=0.042) and ODI (12-weeks p=0.014).

Discussion/Conclusion

In this study, primary and revision lumbar decompression patients experienced minor differences in VAS back, VAS leg, and ODI during midterm recovery (3-6 months), and experienced similar improvement at the 1-year time point (VAS back, VAS leg, ODI, SF-12-PCS, PROMIS-PF). Compared to primary lumbar decompressions, patients considering revision may ultimately experience similar postoperative improvement.

RF 19. Unacceptably High Revision Rate after Microdiscectomy for Far Lateral Disc Herniation at Early Follow-up

Stokman James, MD1, Li Katherine, MD2, Thompson Kirk, MD1, Segebarth P. Brad, MD1, Laxer Eric, MD1, Milam Alden, MD1, Darden Bruce, MD1, Anderson David, MD1, Spector Leo, MD1, Chapman, Jr. Todd, MD1

1OrthoCarolina Spine Center, Charlotte, North Carolina, United States, 2Carolina Medical Center Atrium Health, charlotte, United States

Background/Introduction

Surgical treatment for far lateral disc herniations (FLDH) tends to have poorer and more variable outcomes compared to paramedian disc hernations. There is, however, a paucity of literature regarding surgical outcomes for FLDH and the incidence of revision surgery.

Materials/Methods

A retrospective review of patients treated at our institution using CPT code 63056 (posterolateral extradural decompression) between Jan 2012-September 2017 was performed. Records were reviewed to confirm that a microdiscectomy for FLDH was performed. Inclusion criteria included 1) primary diagnosis of FLDH, 2) minimum of two years from index microdiscectomy and 3) age between 18 and 70. Our primary end point was revision surgery. We also looked at pre-operative and post-operative leg pain, back pain and motor weakness.

Results

149 patients met inclusion criteria. In our preliminary review we found a 56% revision rate (29/52) in those patients that had a minimum of 2 year follow-up. Of all patients meeting inclusion criteria there was a revision rate of at least 19% (29/149). With this significant finding, we are currently in the process of collecting 2-year follow-up on the remaining 97 patients. So far, the 29 revision surgeries have consisted of 19 (65%) fusions, 9 (31%) decompressions and 1 (3%) incision and debridement. Two patients underwent a second revision surgery. There was no change in employment or unemployment status from pre to post surgery with a median return to work of 2.3 months. For patients with 2-year follow-up and no revision surgery, 65% noted resolution of one or more of the pre-operative symptoms (leg pain, back pain, or weakness). There were no statistical differences in sex, age or BMI between those who underwent revision surgery vs those who did not.

Discussion/Conclusion

In preliminary review we found an unexpectedly high failure rate of at least 19%. As a result, we are currently contacting the remaining patients to determine our true revision rate and final analysis will be completed over the next two months. If the revision rate for FLDH microdiscectomy is sufficiently high it may influence patient counselling and surgical decision making.

RF 20. Caudal Lumbar Disc Herniations are More Likely to Require Discectomy for Symptom Resolution

Gupta Anmol, BS, MBA1, Upadhyaya Shivam, MD1, Yeung Caleb, MD1, Ostergaard Peter, MD1, Fogel Harold, MD1, Cha Thomas, MD1, Schwab Joseph, MD1, Bono Christopher, MD1, Hershman Stuart, MD1

1Massachusetts General Hospital, Harvard Medical School, Boston, MA, United States

Background/Introduction

Lumbar spine anatomy varies from level to level. This may have an impact on the rate of resolution of symptoms following a lumbar disc herniation, and in turn, impact the likelihood of successful non-operative treatment following a lumbar disc herniation. In this study, we assessed the impact that the level of a lumbar disc herniation has on the likelihood that a patient will require surgery after at least six weeks of non-operative management.

Materials/Methods

Using the ICD-10 code M51.26, we identified patients at a single academic institution from 2015-2016 who received a diagnosis of primary lumbar radicular pain, had an MRI imaging showing an acute or sub-acute lumbar disc herniation, and who underwent at least six weeks of non-operative management. Non-operative management was defined as any two of the following: medication use, steroid injection, or physical therapy. Patients experiencing symptoms suggesting cauda equina syndrome and those with progressive motor deficits were excluded from analysis.

Results

Five hundred patients met inclusion/exclusion criteria. Twenty-nine (5.8%) had L3-L4 herniations, 245 (49.0%) had L4-L5 herniations, and 226 (45.2%) had L5-S1 herniations. Overall, 451 (90.2%) patients did not undergo surgery within one year of the LDH diagnosis. Patients who did not receive surgery had an average herniation size which occupied 31.2% of the canal, whereas patients who received surgery had disc herniations which occupied 31.5% of the canal on average. While the size of a herniation, age, sex, and race failed to show a statistical association with the likelihood that a patient would require surgery, the level of the disc herniation demonstrated a strong association. L3-L4 and L4-L5 herniations had odds ratios of 0.19 and 0.45, respectively, relative to L5-S1 herniations (p-value = 0.0047). Patients were more than twice as likely to require a surgery on an L5-S1 herniation in comparison to an L4-L5 herniation (p <0.05).

Discussion/Conclusion

Patients with caudal lumbar disc herniations were more to require surgery after six weeks of conservative management than those with disc herniations in the mid lumbar spine. Most patients, regardless of the level of lumbar disc herniation, benefited from non-operative treatment and exhibited resolution of symptoms without surgery.

VA 01. Biomimetic Platelet-Derived Growth Factor Delivery System with Osteoinductive Bone Particles and Adipose Derived Stem Cells Enhance Fusion in Biodegradable Grafts in Posterolateral Lumbar Spinal Fusion

Perdomo-Pantoja Alexander, MD1, Rindone Alexandra, BS1, Holmes Christina, PhD2, Noller Kathleen, BS1, Grayson Warren, PhD1, Witham Timothy, MD1

1Johns Hopkins University School of Medicine, Baltimore, Maryland, United States, 2Florida A&M University-Florida State University College of Engineering, Tallahassee, Florida, United States

Background/Introduction

Biodegradable biomaterials are a promising graft material for spinal fusion due to their ability to degrade as new bone tissue grows into the graft while also providing mechanical stability during the early stages of bone healing. However, a stem cell source is needed for the bone to regenerate into the material, such as adipose-derived stem cells (ASCs) that have demonstrated to enable spinal fusion. Recently, platelet-derived growth factor-BB (PDGF) has surged as an attractive alternative to BMP-2, since it has shown to directly promote the osteogenesis of ASCs when applied at physiological concentrations. Using a newly designed, biodegradable polycaprolactone-decellularized bone (PCL-DCB) hybrid, Rindone et al. developed a biomimetic PDGF delivery system using heparin-conjugated, decellularized bone particles (HC-DCB) that enhances ASC-mediated bone regeneration. In vivo, we tested this novel ASCs and PDGF-BB delivery system in biodegradable PCL-DCB spinal fusion grafts.

Materials/Methods

ASCs were isolated from syngeneic rats and cultured in vitro. Cultured-expanded ASCs (P2) were seeded onto PCL-DCB scaffolds along with HC-DCB particles (40 mg/mL in fibrin) and 20 ug/mL PDGF for subsequent transplantation. Dorsolateral spinal fusion surgery at L4-L5 was performed on Lewis rats divided into 5 experimental groups (2 control and 3 treated groups): [1] PCL only (Control 1); [2] PCL-DCB + HC-DCB particles (Control 2); [3] PCL-DCB + 2.5x106 ASCP2/side + HC-DCB particles; [4] PCL-DCB + 2.5x106 ASCP2/side + HC-DCB particles + PDGF; and, [5] PCL-DCB + HC-DCB particles + PDGF. Fusion was evaluated eight weeks post-surgery.

Results

MicroCT analysis revealed that fusion was reached in all treated groups [3,4,5], achieving maximum CT fusion and manual palpation scores. No statistical differences in new bone volume were found between treated groups (p>0.05), which was higher than in control groups [1,2] (p<0.05). Furthermore, histological staining demonstrated that the groups with PCL-DCB grafts were well-integrated with the host transverse processes, showing expression of CD31, Osteocalcin and RUNX2 in these groups.

Discussion/Conclusion

Biodegradable PCL-DCB grafts enabled spinal fusion when combined with osteoinductive HC-DCB particles. Although presence of both cells and growth factors was not required to achieve spinal fusion in combined PCL-DCB + HC-DCB grafts, ASCs and PDGF-BB enhanced bone formation.

VA 02. Possible Sex-based Difference in Response to recombinant Human Bone Morphogenetic Protein-2 (rhBMP-2) in a Rat Posterolateral Fusion Model

Paul Jonathan, MPH1, Plantz Mark, BS1, Lyons Joseph, BS1, Nandurkar Tejas, MA1, Marsh Parker, BS1, Minardi Silvia, PhD1, Stock Stuart, PhD1, Yun Chawon, PhD1, Hsu Wellington, MD1, Hsu Erin, PhD1

1Northwestern University Feinberg School of Medicine, Chicago, Illinois, United States

Background/Introduction

Previous research pertaining to differences between sexes has demonstrated hormonal involvement in osteogenesis. Estrogen modulates bone activity, by inhibiting osteoclast bone resorption, and bone morphogenetic protein-2 (BMP-2) signaling pathways in human osteoblast cells. In this study, we aimed to evaluate and authenticate the presence of sex-dependent differences in osteogenesis and overall response to rhBMP-2 in a rat posterolateral spinal fusion model.

Materials/Methods

Twenty-eight Sprague-Dawley rats – 14 male and 14 female – at age 12-16 weeks underwent L4-L5 posterolateral fusion with bilateral placement of an absorbable collagen sponge loaded with 10 μg of rhBMP-2/animal (5 μg/implant). Fusion was evaluated at 8 weeks postoperatively with plain radiographs and blinded manual palpation. Fusion scoring was assessed by 3 blinded investigators using an established scoring system wherein 0 = no fusion, 1 = unilateral fusion, and 2 = bilateral fusion. A fusion score of 1 or greater was considered successfully fused. MicroCT imaging (N=3/group) and subsequent analysis was used to evaluate and quantify new bone formation.

Results

Radiography and manual palpation showed 100% fusion rates for both sexes. However, manual palpation scores differed significantly with females having lower fusion score (female: 1.26±0.31 vs males: 1.55±0.33, *p=0.001). Both radiography and MicroCT demonstrated noticeably smaller fusion mass volumes in females compared to males. MicroCT axial cross-sections of representative L4-L5 fusion beds are illustrated below (Figure 1A). MicroCT evaluation showed significant differences in females and males, with females having significantly lower total volume (female: 262.361 ± 86.949mm3 vs. male: 731.896 ± 237.476 mm3, *p=0.000), significantly higher bone volume fraction (female: 0.565 ± 0.181 vs. male: 0.237±0.100, *p=0.000), and significantly higher trabecular number (female: 6.212 ± 1.078/mm vs. male: 2.963 ± 0.776/mm, p=0.000*).

Discussion/Conclusion

Females produced more dense, but lower volume of new bone relative to males in a rat posterolateral spinal fusion model. This study shows that there may be sex-dependent differences in bone regeneration induced by rhBMP-2. The details of interactions between sex hormone signaling and the BMP-2 pathway still need to be elucidated to provide a mechanism to explain such findings.

VA 03. Preoperative Lumbar Epidural Steroid Injections Administered within 6 weeks of Microdiscectomy are Associated with Increased Rates of Revision

Bhattacharjee Sarah, BS1, Pirkle Sean, BA1, Shi Lewis, MD1, Lee Michael, MD1

1University of Chicago, Chicago, Illinois, United States

Background/Introduction

Lumbar epidural steroid injections (LESI’s) are widely utilized in the management of lower back pain. However, as recent studies continue to shed light on potential negative effects from these injections, defining a safe interval for their use preoperatively is of particular importance. In this study, we investigated the effects of preoperative LESI timing on the rates of recurrent microdiscectomy.

Materials/Methods

This was a retrospective study using the HORTHO subset of the PearlDiver national insurance claims database. Microdiscectomy patients were stratified by the timing of their most recent LESI prior to surgery into bimonthly cohorts (0-2 months, 2-4 months, 4-6 months). Dates closer to surgery were further stratified into biweekly cohorts (0-2 weeks, 2-4 weeks, 4-6 weeks, 6-8 weeks). The 6-month revision rate was assessed for all patients and compared using chi-squared tests between each injection cohort and a control group of patients with no injection history within 6 months prior to surgery. Risk factors were determined using univariate analyses with variables trending towards significance included in follow up multivariate analyses. Multivariate analyses were then performed to evaluate the effect of injection timing on microdiscectomy revision.

Results

12,786 microdiscectomy patients were identified of which 1,090 (8.52%) received injections within 6 months prior to surgery. In our multivariate analysis, we observed a significant increase in the 6-month revision rates in patients who received injections within 6 weeks prior to surgery (Odds Ratio [OR]: 1.900, 1.218 - 2.963; p = 0.005) compared with those who received no injections within 6 months before surgery (Table 1). No other significant differences were observed.

Discussion/Conclusion

In this national database analysis, we observed that microdiscectomy performed within 6 weeks following LESI’s are at a higher risk for recurrent surgery while microdiscectomy performed after a period greater than 6 weeks from most recent LESI showed no association with increased risk. These results suggest 1) that LESI’s impact the rate of recurrent surgery after microdiscectomy and 2) that the safe interval between LESI’s and future microdiscectomy is roughly a 6-week period. Further research into this relationship is necessary.

VA 04. Analysis of Postoperative Outcomes of Sacroiliac Joint Fusion to Treat Sacrum Pain, Sacroiliitis, Sacral Instability, or Spondylosis; a National Readmission Database Study

Ballatori Alexander, BS1, Shahrestani Shane, MS1, Ton Andy, BS1, Chen Xiao, BS1, Wang Jeffrey, MD2, Buser Zorica, PhD1

1University of Southern California Keck School of Medicine, Los Angeles, California, United States, 2Keck Medical Center of USC, Los Angeles, California, United States

Background/Introduction

The sacroiliac joint has been implicated as a source of chronic low back pain. Many have turned to sacroiliac joint fusion after failure of conservative care to relieve symptoms. However, identifying patients who will respond positively from the sacroiliac joint fusion is still unclear. This study examines the postoperative outcomes of sacroiliac joint fusion based on the primary diagnoses of sacrum pain, sacroiliitis, sacral instability, or spondylosis.

Materials/Methods

Using the 2016 National Readmission Database, we conducted a retrospective cohort analysis of 1,272 patients who received a sacroiliac joint fusion for diagnosed sacrum pain (n=288), sacroiliitis (n=536), sacral instability (n=135), or spondylosis (n=155) using ICD-10 coding. We then collected all of the individuals that had a non-elective readmission, separated them by their primary diagnosis, and analyzed for postoperative complications including infection, urinary tract infection, pain, osteomyelitis, thromboembolisms, sepsis, pneumonia, pseudoarthrosis, novel lumbar pathology, nervous system complications, and need for revision surgery. Statistical analysis was conducted in R. Tukey multiple comparisons of means was used to compare complications by diagnosis.

Results

Sacroiliitis had the highest rate of non-elective readmissions (17.1%), followed by those diagnosed with sacrum pain (14.2%), spondylosis (11.6%), and sacral instability (9.6%), but these rates were not significant between any groups on pairwise comparison. However, those with a primary diagnosis of sacroiliitis had a significantly higher average number of readmissions (1.80±1.16 readmissions) than those with a primary diagnosis of sacrum pain (1.24±0.58 readmissions, p=0.013), but no differences existed between other groups. After comparing the different categories of postoperative complications, significant differences in postoperative infection rates were noted between patients with a primary diagnosis of sacroiliitis and those with a primary diagnosis of sacrum pain (p=0.008).

Discussion/Conclusion

Patients diagnosed with sacroiliitis had the highest rate of non-elective postoperative readmissions following sacroiliac joint fusion, and patients diagnosed with sacral instability had the lowest rates of non-elective postoperative readmission. The reasons for readmission were not significantly different across each group, except for infection rates of those with a primary diagnosis of sacroiliitis being significantly higher than those diagnosed with sacrum pain.

VA 05. The Effects of Marijuana Use on Lumbar Spinal Fusion

Jakoi Andre, MD1, Kirchner Gregory, MPH2, Kerbel Yehuda, MD3, Khalsa Amrit, MD3

1Orthopedic Health of Kansas City, Kansas City, Missouri, United States, 2Drexel University College of Medicine, Philadelphia, Pennsylvania, United States, 3University of Pennsylvania, Philadelphia, Pennsylvania, United States

Background/Introduction

As marijuana becomes legalized throughout the United States, its medicinal and recreational usage is becoming more mainstream. Clinicians currently have little guidance regarding both short-term and long-term effects of marijuana usage on surgical interventions. While the rate of lumbar spinal fusion in the United States continues to grow, the effect of marijuana usage on fusion remains uncertain. The purpose of this study was to evaluate marijuana usage and its effect on outcomes following transforaminal lumbar interbody fusion (TLIF).

Materials/Methods

102 patients who underwent TLIF performed by the same surgeon were followed for 12 months. Patients were self-reported for marijuana usage (n=36). Patient reported outcome measures included preoperative Oswestry Disability Index (ODI), 6-month ODI and 12-month ODI, as well as length of stay (LOS), complications, return to operating room (OR), revision surgery, and confirmed fusion. Continuous variables were compared using the independent two-sample t test or analysis of variance (ANOVA), whereas categorical variables were analyzed using the chi-square or Fischer exact tests. Adjusted analysis was performed using a multivariate logistic regression model.

Results

Marijuana usage was associated with a younger population (p<0.001), but showed no difference regarding sex or BMI compared to the non-usage group. There was no statistically significant difference in complications, return to OR, or revision surgery between groups. When controlling for factors such as age and pre-operative ODI, multivariate analysis demonstrated that marijuana usage did not limit post-operative ODI reduction. The marijuana usage group demonstrated shorter LOS (2.42 vs. 3.00 days, p=0.020). Fusion rates at 12 months were similar between groups (96% vs. 92.3%, p=0.678). ODI was similar between groups at all time points.

Discussion/Conclusion

Perioperative outcomes were similar in patients who underwent TLIF regardless of marijuana usage.

If the inline PDF is not rendering correctly, you can download the PDF file here.

Contributor Notes

Disclaimer: The Journal of Neurosurgery Publishing Group (JNSPG) acknowledges that these abstracts are published as submitted and did not go through JNSPG’s peer-review or editing process.

Metrics

All Time Past Year Past 30 Days
Abstract Views 0 0 0
Full Text Views 223 223 188
PDF Downloads 91 91 88
EPUB Downloads 0 0 0