Filter

Refine by Publication

Refine by Date

—

Refine by Access

By Author

Refine By Subjects

  • Print
  • Save

Browse

You are looking at 1 - 10 of 140 items for

  • Refine by Access: all x
  • By Author: Park, Paul x
Clear All
Restricted access

Risk factors for not reaching minimal clinically important difference at 90 days and 1 year after elective lumbar spine surgery: a cohort study

Travis Hamilton, Seokchun Lim, Edvin Telemi, Ho Jun Yun, Mohamed Macki, Lonni Schultz, Hsueh-Han Yeh, Kylie Springer, Kevin Taliaferro, Miguelangelo Perez-Cruet, Ilyas Aleem, Paul Park, Richard Easton, David R. Nerenz, Jason M. Schwalb, Muwaffak Abdulhak, and Victor Chang

OBJECTIVE

Patient-perceived functional improvement is a core metric in lumbar surgery for degenerative disease. It is important to identify both modifiable and nonmodifiable risk factors that can be evaluated and possibly optimized prior to elective surgery. This case-control study was designed to study risk factors for not achieving the minimal clinically important difference (MCID) in Patient-Reported Outcomes Measurement Information System Function 4-item Short Form (PROMIS PF) score.

METHODS

The authors queried the Michigan Spine Surgery Improvement Collaborative database to identify patients who underwent elective lumbar surgical procedures with PROMIS PF scores. Cases were divided into two cohorts based on whether patients achieved MCID at 90 days and 1 year after surgery. Patient characteristics and operative details were analyzed as potential risk factors.

RESULTS

The authors captured 10,922 patients for 90-day follow-up and 4453 patients (40.8%) did not reach MCID. At the 1-year follow-up period, 7780 patients were identified and 2941 patients (37.8%) did not achieve MCID. The significant demographic characteristic–adjusted relative risks (RRs) for both groups (RR 90 day, RR 1 year) included the following: symptom duration > 1 year (1.34, 1.41); previous spine surgery (1.25, 1.30); African American descent (1.25, 1.20); chronic opiate use (1.23, 1.25); and less than high school education (1.20, 1.34). Independent ambulatory status (0.83, 0.88) and private insurance (0.91, 0.85) were associated with higher likelihood of reaching MCID at 90 days and 1 year, respectively.

CONCLUSIONS

Several key unique demographic risk factors were identified in this cohort study that precluded optimal postoperative functional outcomes after elective lumbar spine surgery. With this information, appropriate preoperative counseling can be administered to assist in shaping patient expectations.

In Journal of Neurosurgery: Spine Publish Before Print

Restricted access

Does diabetes affect outcome or reoperation rate after lumbar decompression or arthrodesis? A matched analysis of the Quality Outcomes Database data set

Presented at the 2023 AANS/CNS Joint Section on Disorders of the Spine and Peripheral Nerves

James Mooney, Karim Rizwan Nathani, Daniel Zeitouni, Giorgos D. Michalopoulos, Michael Y. Wang, Domagoj Coric, Andrew K. Chan, Daniel C. Lu, Brandon A. Sherrod, Oren N. Gottfried, Christopher I. Shaffrey, Khoi D. Than, Jacob L. Goldberg, Ibrahim Hussain, Michael S. Virk, Nitin Agarwal, Steven D. Glassman, Mark E. Shaffrey, Paul Park, Kevin T. Foley, Dean Chou, Jonathan R. Slotkin, Luis M. Tumialán, Cheerag D. Upadhyaya, Eric A. Potts, Kai-Ming G. Fu, Regis W. Haid, John J. Knightly, Praveen V. Mummaneni, Erica F. Bisson, Anthony L. Asher, and Mohamad Bydon

OBJECTIVE

Diabetes mellitus (DM) is a known risk factor for postsurgical and systemic complications after lumbar spinal surgery. Smaller studies have also demonstrated diminished improvements in patient-reported outcomes (PROs), with increased reoperation and readmission rates after lumbar surgery in patients with DM. The authors aimed to examine longer-term PROs in patients with DM undergoing lumbar decompression and/or arthrodesis for degenerative pathology.

METHODS

The Quality Outcomes Database was queried for patients undergoing elective lumbar decompression and/or arthrodesis for degenerative pathology. Patients were grouped into DM and non-DM groups and optimally matched in a 1:1 ratio on 31 baseline variables, including the number of operated levels. Outcomes of interest were readmissions and reoperations at 30 and 90 days after surgery in addition to improvements in Oswestry Disability Index, back pain, and leg pain scores and quality-adjusted life-years at 90 days after surgery.

RESULTS

The matched decompression cohort comprised 7836 patients (3236 [41.3] females) with a mean age of 63.5 ± 12.6 years, and the matched arthrodesis cohort comprised 7336 patients (3907 [53.3%] females) with a mean age of 64.8 ± 10.3 years. In patients undergoing lumbar decompression, no significant differences in nonroutine discharge, length of stay (LOS), readmissions, reoperations, and PROs were observed. In patients undergoing lumbar arthrodesis, nonroutine discharge (15.7% vs 13.4%, p < 0.01), LOS (3.2 ± 2.0 vs 3.0 ± 3.5 days, p < 0.01), 30-day (6.5% vs 4.4%, p < 0.01) and 90-day (9.1% vs 7.0%, p < 0.01) readmission rates, and the 90-day reoperation rate (4.3% vs 3.2%, p = 0.01) were all significantly higher in the DM group. For DM patients undergoing lumbar arthrodesis, subgroup analyses demonstrated a significantly higher risk of poor surgical outcomes with the open approach.

CONCLUSIONS

Patients with and without DM undergoing lumbar spinal decompression alone have comparable readmission and reoperation rates, while those undergoing arthrodesis procedures have a higher risk of poor surgical outcomes up to 90 days after surgery. Surgeons should target optimal DM control preoperatively, particularly for patients undergoing elective lumbar arthrodesis.

In Journal of Neurosurgery: Spine Publish Before Print

Restricted access

What factors influence surgical decision-making in anterior versus posterior surgery for cervical myelopathy? A QOD analysis

Presented at the 2023 AANS/CNS Joint Section on Disorders of the Spine and Peripheral Nerves

Christine Park, Christopher I. Shaffrey, Khoi D. Than, Giorgos D. Michalopoulos, Sally El Sammak, Andrew K. Chan, Erica F. Bisson, Brandon A. Sherrod, Anthony L. Asher, Domagoj Coric, Eric A. Potts, Kevin T. Foley, Michael Y. Wang, Kai-Ming Fu, Michael S. Virk, John J. Knightly, Scott Meyer, Paul Park, Cheerag Upadhyaya, Mark E. Shaffrey, Avery L. Buchholz, Luis M. Tumialán, Jay Turner, Nitin Agarwal, Dean Chou, Nauman S. Chaudhry, Regis W. Haid Jr., Praveen V. Mummaneni, Mohamad Bydon, and Oren N. Gottfried

OBJECTIVE

The aim of this study was to explore the preoperative patient characteristics that affect surgical decision-making when selecting an anterior or posterior operative approach in patients diagnosed with cervical spondylotic myelopathy (CSM).

METHODS

This was a multi-institutional, retrospective study of the prospective Quality Outcomes Database (QOD) Cervical Spondylotic Myelopathy module. Patients aged 18 years or older diagnosed with primary CSM who underwent multilevel (≥ 2-level) elective surgery were included. Demographics and baseline clinical characteristics were collected.

RESULTS

Of the 841 patients with CSM in the database, 492 (58.5%) underwent multilevel anterior surgery and 349 (41.5%) underwent multilevel posterior surgery. Surgeons more often performed a posterior surgical approach in older patients (mean 64.8 ± 10.6 vs 58.5 ± 11.1 years, p < 0.001) and those with a higher American Society of Anesthesiologists class (class III or IV: 52.4% vs 46.3%, p = 0.003), a higher rate of motor deficit (67.0% vs 58.7%, p = 0.014), worse myelopathy (mean modified Japanese Orthopaedic Association score 11.4 ± 3.1 vs 12.4 ± 2.6, p < 0.001), and more levels treated (4.3 ± 1.3 vs 2.4 ± 0.6, p < 0.001). On the other hand, surgeons more frequently performed an anterior surgical approach when patients were employed (47.2% vs 23.2%, p < 0.001) and had intervertebral disc herniation as an underlying pathology (30.7% vs 9.2%, p < 0.001).

CONCLUSIONS

The selection of approach for patients with CSM depends on patient demographics and symptomology. Posterior surgery was performed in patients who were older and had worse systemic disease, increased myelopathy, and greater levels of stenosis. Anterior surgery was more often performed in patients who were employed and had intervertebral disc herniation.

In Journal of Neurosurgery: Spine Publish Before Print

Free access

Greater improvement in Neck Disability Index scores in women after surgery for cervical myelopathy: an analysis of the Quality Outcomes Database

Arati Patel, Sravani Kondapavulur, Gray Umbach, Andrew K. Chan, Vivian P. Le, Erica F. Bisson, Mohamad Bydon, Dean Chou, Steve D. Glassman, Kevin T. Foley, Christopher I. Shaffrey, Eric A. Potts, Mark E. Shaffrey, Domagoj Coric, John J. Knightly, Paul Park, Michael Y. Wang, Kai-Ming Fu, Jonathan Slotkin, Anthony L. Asher, Michael S. Virk, Regis W. Haid, Oren Gottfried, Scott Meyer, Cheerag D. Upadhyaya, Luis M. Tumialán, Jay D. Turner, and Praveen V. Mummaneni

OBJECTIVE

There is a high prevalence of cervical myelopathy that requires surgery; as such, it is important to identify how different groups benefit from surgery. The American Association of Neurological Surgeons launched the Quality Outcomes Database (QOD), a prospective longitudinal registry, that includes demographic, clinical, and patient-reported outcome data to measure the safety and quality of neurosurgical procedures. In this study, the authors assessed the impact of gender on patient-reported outcomes in patients who underwent surgery for cervical myelopathy.

METHODS

The authors analyzed 1152 patients who underwent surgery for cervical myelopathy and were included in the QOD cervical module. Univariate comparison of baseline patient characteristics between males and females who underwent surgery for cervical spondylotic myelopathy was performed. Baseline characteristics that significantly differed between males and females were included in a multivariate generalized linear model comparing baseline and 1-year postoperative Neck Disability Index (NDI) scores.

RESULTS

This study included 546 females and 604 males. Females demonstrated significantly greater improvement in NDI score 1 year after surgery (p = 0.036). In addition to gender, the presence of axial neck pain and insurance status were also significantly predictive of improvement in NDI score after surgery (p = 0.0013 and p = 0.0058, respectively).

CONCLUSIONS

Females were more likely to benefit from surgery for cervical myelopathy compared with males. It is important to identify gender differences in postoperative outcomes after surgery in order to deliver more personalized and patient-centric care.

In Neurosurgical Focus Volume 55 (2023): Issue 5 (Nov 2023): Diversity, Equity, and Inclusion and the Goal of Reducing Healthcare Disparities in Neurosurgery

Free access

Three-level ACDF versus 3-level laminectomy and fusion: are there differences in outcomes? An analysis of the Quality Outcomes Database cervical spondylotic myelopathy cohort

Vardhaan S. Ambati, Mohamed Macki, Andrew K. Chan, Giorgos D. Michalopoulos, Vivian P. Le, Alysha B. Jamieson, Dean Chou, Christopher I. Shaffrey, Oren N. Gottfried, Erica F. Bisson, Anthony L. Asher, Domagoj Coric, Eric A. Potts, Kevin T. Foley, Michael Y. Wang, Kai-Ming Fu, Michael S. Virk, John J. Knightly, Scott Meyer, Paul Park, Cheerag Upadhyaya, Mark E. Shaffrey, Avery L. Buchholz, Luis M. Tumialán, Jay D. Turner, Brandon A. Sherrod, Regis W. Haid Jr., Mohamad Bydon, and Praveen V. Mummaneni

OBJECTIVE

The authors sought to compare 3-level anterior with posterior fusion surgical procedures for the treatment of multilevel cervical spondylotic myelopathy (CSM).

METHODS

The authors analyzed prospective data from the 14 highest enrolling sites of the Quality Outcomes Database CSM module. They compared 3-level anterior cervical discectomy and fusion (ACDF) and posterior cervical laminectomy and fusion (PCF) surgical procedures, excluding surgical procedures crossing the cervicothoracic junction. Rates of reaching the minimal clinically important difference (MCID) in patient-reported outcomes (PROs) were compared at 24 months postoperatively. Multivariable analyses adjusted for potential confounders elucidated in univariable analysis.

RESULTS

Overall, 199 patients met the inclusion criteria: 123 ACDF (61.8%) and 76 PCF (38.2%) patients. The 24-month follow-up rates were similar (ACDF 90.2% vs PCF 92.1%, p = 0.67). Preoperatively, ACDF patients were younger (60.8 ± 10.2 vs 65.0 ± 10.3 years, p < 0.01), and greater proportions were privately insured (56.1% vs 36.8%, p = 0.02), actively employed (39.8% vs 22.8%, p = 0.04), and independently ambulatory (14.6% vs 31.6%, p < 0.01). Otherwise, the cohorts had equivalent baseline modified Japanese Orthopaedic Association (mJOA), Neck Disability Index (NDI), numeric rating scale (NRS)–arm pain, NRS–neck pain, and EQ-5D scores (p > 0.05). ACDF patients had reduced hospitalization length (1.6 vs 3.9 days, p < 0.01) and a greater proportion had nonroutine discharge (7.3% vs 22.8%, p < 0.01), but they had a higher rate of postoperative dysphagia (13.5% vs 3.5%, p = 0.049). Compared with baseline values, both groups demonstrated improvements in all outcomes at 24 months (p < 0.05). In multivariable analyses, after controlling for age, insurance payor, employment status, ambulation status, and other potential clinically relevant confounders, ACDF was associated with a greater proportion of patients with maximum satisfaction on the North American Spine Society Patient Satisfaction Index (NASS) (NASS score of 1) at 24 months (69.4% vs 53.7%, OR 2.44, 95% CI 1.17–5.09, adjusted p = 0.02). Otherwise, the cohorts shared similar 24-month outcomes in terms of reaching the MCID for mJOA, NDI, NRS–arm pain, NRS–neck pain, and EQ-5D score (adjusted p > 0.05). There were no differences in the 3-month readmission (ACDF 4.1% vs PCF 3.9%, p = 0.97) and 24-month reoperation (ACDF 13.5% vs PCF 18.6%, p = 0.36) rates.

CONCLUSIONS

In a cohort limited to 3-level fusion surgical procedures, ACDF was associated with reduced blood loss, shorter hospitalization length, and higher routine home discharge rates; however, PCF resulted in lower rates of postoperative dysphagia. The procedures yielded comparably significant improvements in functional status (mJOA score), neck and arm pain, neck pain–related disability, and quality of life at 3, 12, and 24 months. ACDF patients had significantly higher odds of maximum satisfaction (NASS score 1). Given comparable outcomes, patients should be counseled on each approach’s complication profile to aid in surgical decision-making.

In Neurosurgical Focus Volume 55 (2023): Issue 3 (Sep 2023): Cervical Spondylotic Myelopathy

Restricted access

Cervical laminoplasty versus laminectomy and posterior cervical fusion for cervical myelopathy: propensity-matched analysis of 24-month outcomes from the Quality Outcomes Database

Presented at the 2023 AANS/CNS Joint Section on Disorders of the Spine and Peripheral Nerves

Eunice Yang, Praveen V. Mummaneni, Dean Chou, Mohamad Bydon, Erica F. Bisson, Christopher I. Shaffrey, Oren N. Gottfried, Anthony L. Asher, Domagoj Coric, Eric A. Potts, Kevin T. Foley, Michael Y. Wang, Kai-Ming Fu, Michael S. Virk, John J. Knightly, Scott Meyer, Paul Park, Cheerag D. Upadhyaya, Mark E. Shaffrey, Avery L. Buchholz, Luis M. Tumialán, Jay D. Turner, Giorgos D. Michalopoulos, Brandon A. Sherrod, Nitin Agarwal, Regis W. Haid Jr., and Andrew K. Chan

OBJECTIVE

Compared with laminectomy with posterior cervical fusion (PCF), cervical laminoplasty (CL) may result in different outcomes for those operated on for cervical spondylotic myelopathy (CSM). The aim of this study was to compare 24-month patient-reported outcomes (PROs) for laminoplasty versus PCF by using the Quality Outcomes Database (QOD) CSM data set.

METHODS

This was a retrospective study using an augmented data set from the prospectively collected QOD Registry Cervical Module. Patients undergoing laminoplasty or PCF for CSM were included. Using the nearest-neighbor method, the authors performed 1:1 propensity matching based on age, operated levels, and baseline modified Japanese Orthopaedic Association (mJOA) and visual analog scale (VAS) neck pain scores. The 24-month PROs, i.e., mJOA, Neck Disability Index (NDI), VAS neck pain, VAS arm pain, EQ-5D, EQ-VAS, and North American Spine Society (NASS) satisfaction scores, were compared. Only cases in the subaxial cervical region were included; those that crossed the cervicothoracic junction were excluded.

RESULTS

From the 1141 patients included in the QOD CSM data set who underwent anterior or posterior surgery for cervical myelopathy, 946 (82.9%) had 24 months of follow-up. Of these, 43 patients who underwent laminoplasty and 191 who underwent PCF met the inclusion criteria. After matching, the groups were similar for baseline characteristics, including operative levels (CL group: 4.0 ± 0.9 vs PCF group: 4.2 ± 1.1, p = 0.337) and baseline PROs (p > 0.05), except for a higher percentage involved in activities outside the home in the CL group (95.3% vs 81.4%, p = 0.044). The 24-month follow-up for the matched cohorts was similar (CL group: 88.4% vs PCF group: 83.7%, p = 0.534). Patients undergoing laminoplasty had significantly lower estimated blood loss (99.3 ± 91.7 mL vs 186.7 ± 142.7 mL, p = 0.003), decreased length of stay (3.0 ± 1.6 days vs 4.5 ± 3.3 days, p = 0.012), and a higher rate of routine discharge (88.4% vs 62.8%, p = 0.006). The CL cohort also demonstrated a higher rate of return to activities (47.2% vs 21.2%, p = 0.023) after 3 months. Laminoplasty was associated with a larger improvement in 24-month NDI score (−19.6 ± 18.9 vs −9.1 ± 21.9, p = 0.031). Otherwise, there were no 3- or 24-month differences in mJOA, mean NDI, VAS neck pain, VAS arm pain, EQ-5D, EQ-VAS, and distribution of NASS satisfaction scores (p > 0.05) between the cohorts.

CONCLUSIONS

Compared with PCF, laminoplasty was associated with decreased blood loss, decreased length of hospitalization, and higher rates of home discharge. At 3 months, laminoplasty was associated with a higher rate of return to baseline activities. At 24 months, laminoplasty was associated with greater improvements in neck disability. Otherwise, laminoplasty and PCF shared similar outcomes for functional status, pain, quality of life, and satisfaction. Laminoplasty and PCF achieved similar neck pain scores, suggesting that moderate preoperative neck pain may not necessarily be a contraindication for laminoplasty.

In Journal of Neurosurgery: Spine Volume 39: Issue 5 (Nov 2023)

Free access

Introduction. Surgery for sacroiliac joint dysfunction: emerging techniques and assistive technologies

Paul Park, Kristen E. Jones, Yamaan S. Saadeh, Cristiano M. Menezes, and Juan S. Uribe

In Neurosurgical Focus Volume 55 (2023): Issue 1 (Jul 2023): Sacroiliac Joint Dysfunction: Diagnosis and Treatment Options

Free access

Robotic versus nonrobotic sacroiliac joint fusion

John H. Lee, Mark M. Zaki, Rushikesh S. Joshi, Joseph R. Linzey, Rakesh D. Patel, Paul Park, and Yamaan S. Saadeh

OBJECTIVE

Robot-assisted pedicle screw placement in spinal fusion has been well studied. However, few studies have evaluated robot-assisted sacroiliac joint (SIJ) fusion. The aim of this study was to compare surgical characteristics, accuracy, and complications between robot-assisted and fluoroscopically guided SIJ fusion.

METHODS

A retrospective review of 110 patients with 121 SIJ fusions done at a single academic institution was conducted from 2014 to 2023. Inclusion criteria included adult age and a robot- or fluoroscopically guided approach to SIJ fusion. Patients were excluded if the SIJ fusion was part of a longer fusion construct, not minimally invasive, and/or had missing data. Demographics, approach type (robotic vs fluoroscopic), operative time, estimated blood loss (EBL), number of screws, intraoperative complications, 30-day complications, number of intraoperative fluoroscopic images (as a surrogate for radiation exposure), implant placement accuracy, and pain status at the first follow-up were recorded. Primary endpoints were SIJ screw placement accuracy and complications. Secondary endpoints were operative time, radiation exposure, and pain status at the first follow-up.

RESULTS

Ninety patients were included who underwent a total of 101 SIJ fusions, of which 78 were robotic and 23 were fluoroscopic. The mean age of the cohort at the time of surgery was 55.9 ± 13.8 years; 46 patients were females (51.1%). No difference was found in screw placement accuracy between robotic and fluoroscopic fusion (1.3% vs 8.7%, p = 0.06). Chi-square analysis of robotic versus fluoroscopic fusion found no difference in the presence of 30-day complications (p = 0.62). Mann-Whitney U-test analysis found that robotic fusion had a significantly longer operative time than fluoroscopic fusion (72.0 vs 61.0 minutes, p = 0.01); however, robot-assisted fusions involved significantly lower radiation exposure (26.7 vs 187.4 fluoroscopic images, p < 0.001). No difference in EBL was noted (p = 0.17). No intraoperative complications were present in this cohort. Subgroup analysis comparing the 23 most recent robotic cases against the 23 fluoroscopic cases found that robotic fusion still was associated with significantly longer operative times than fluoroscopic fusion (74.0 ± 26.4 vs 61.0 ± 14.9 minutes, respectively; p = 0.047).

CONCLUSIONS

SIJ screw placement accuracy did not significantly differ between robot-assisted and fluoroscopic SIJ fusion. Complications overall were low and similar between the two groups. The operative time was longer with robotic assistance, but there was markedly less radiation exposure to the surgeon and staff.

In Neurosurgical Focus Volume 55 (2023): Issue 1 (Jul 2023): Sacroiliac Joint Dysfunction: Diagnosis and Treatment Options

Restricted access

Association of prolonged symptom duration with poor outcomes in lumbar spine surgery: a Michigan Spine Surgery Improvement Collaborative study

Travis Hamilton, Seamus Bartlett, Nachiket Deshpande, Moustafa Hadi, Jared C. Reese, Tarek R. Mansour, Edvin Telemi, Kylie Springer, Lonni Schultz, David R. Nerenz, Muwaffak Abdulhak, Teck Soo, Jason Schwalb, Jad G. Khalil, Ilyas Aleem, Richard Easton, Miguelangelo Perez-Cruet, Paul Park, and Victor Chang

OBJECTIVE

There is a scarcity of large multicenter data on how preoperative lumbar symptom duration relates to postoperative patient-reported outcomes (PROs). The objective of this study was to determine the effect of preoperative and baseline symptom duration on PROs at 90 days, 1 year, and 2 years after lumbar spine surgery.

METHODS

The Michigan Spine Surgery Improvement Collaborative registry was queried for all lumbar spine operations between January 1, 2017, to December 31, 2021, with a follow-up of 2 years. Patients were stratified into three subgroups based on symptom duration: < 3 months, 3 months to < 1 year, and ≥ 1 year. The primary outcomes were reaching the minimal clinically important difference (MCID) for the PROs (i.e., leg pain, Patient-Reported Outcomes Measurement Information System Physical Function (PROMIS PF), EQ-5D, North American Spine Society satisfaction, and return to work). The EQ-5D score was also analyzed as a continuous variable to calculate quality-adjusted life years. Multivariable Poisson generalized estimating equation models were used to report adjusted risk ratios, with the < 3-month cohort used as the reference.

RESULTS

There were 37,223 patients (4670 with < 3-month duration, 9356 with 3-month to < 1-year duration, and 23,197 with ≥ 1-year duration) available for analysis. Compared with patients with a symptom duration of < 1 year, patients with a symptom duration of ≥ 1 year were significantly less likely to achieve an MCID in PROMIS PF, EQ-5D, back pain relief, and leg pain relief at 90 days, 1 year, and 2 years postoperatively. Similar trends were observed for patient satisfaction and return to work. With the EQ-5D score as a continuous variable, a symptom duration of ≥ 1 year was associated with 0.04, 0.05, and 0.03 (p < 0.001) decreases in EQ-5D score at 90 days, 1 year, and 2 years after surgery, respectively.

CONCLUSIONS

A symptom duration of ≥ 1 year was associated with poorer outcomes on several outcome metrics. This suggests that timely referral and surgery for degenerative lumbar pathology may optimize patient outcome.

In Journal of Neurosurgery: Spine Volume 39: Issue 4 (Oct 2023)

Free access

Developing nonlinear k-nearest neighbors classification algorithms to identify patients at high risk of increased length of hospital stay following spine surgery

Shane Shahrestani, Andrew K. Chan, Erica F. Bisson, Mohamad Bydon, Steven D. Glassman, Kevin T. Foley, Christopher I. Shaffrey, Eric A. Potts, Mark E. Shaffrey, Domagoj Coric, John J. Knightly, Paul Park, Michael Y. Wang, Kai-Ming Fu, Jonathan R. Slotkin, Anthony L. Asher, Michael S. Virk, Giorgos D. Michalopoulos, Jian Guan, Regis W. Haid, Nitin Agarwal, Dean Chou, and Praveen V. Mummaneni

OBJECTIVE

Spondylolisthesis is a common operative disease in the United States, but robust predictive models for patient outcomes remain limited. The development of models that accurately predict postoperative outcomes would be useful to help identify patients at risk of complicated postoperative courses and determine appropriate healthcare and resource utilization for patients. As such, the purpose of this study was to develop k-nearest neighbors (KNN) classification algorithms to identify patients at increased risk for extended hospital length of stay (LOS) following neurosurgical intervention for spondylolisthesis.

METHODS

The Quality Outcomes Database (QOD) spondylolisthesis data set was queried for patients receiving either decompression alone or decompression plus fusion for degenerative spondylolisthesis. Preoperative and perioperative variables were queried, and Mann-Whitney U-tests were performed to identify which variables would be included in the machine learning models. Two KNN models were implemented (k = 25) with a standard training set of 60%, validation set of 20%, and testing set of 20%, one with arthrodesis status (model 1) and the other without (model 2). Feature scaling was implemented during the preprocessing stage to standardize the independent features.

RESULTS

Of 608 enrolled patients, 544 met prespecified inclusion criteria. The mean age of all patients was 61.9 ± 12.1 years (± SD), and 309 (56.8%) patients were female. The model 1 KNN had an overall accuracy of 98.1%, sensitivity of 100%, specificity of 84.6%, positive predictive value (PPV) of 97.9%, and negative predictive value (NPV) of 100%. Additionally, a receiver operating characteristic (ROC) curve was plotted for model 1, showing an overall area under the curve (AUC) of 0.998. Model 2 had an overall accuracy of 99.1%, sensitivity of 100%, specificity of 92.3%, PPV of 99.0%, and NPV of 100%, with the same ROC AUC of 0.998.

CONCLUSIONS

Overall, these findings demonstrate that nonlinear KNN machine learning models have incredibly high predictive value for LOS. Important predictor variables include diabetes, osteoporosis, socioeconomic quartile, duration of surgery, estimated blood loss during surgery, patient educational status, American Society of Anesthesiologists grade, BMI, insurance status, smoking status, sex, and age. These models may be considered for external validation by spine surgeons to aid in patient selection and management, resource utilization, and preoperative surgical planning.

In Neurosurgical Focus Volume 54 (2023): Issue 6 (Jun 2023): Machine Learning in Neurosurgery