Search Results

You are looking at 1 - 10 of 29 items for

  • Author or Editor: Panagiotis Kerezoudis x
Clear All Modify Search
Restricted access

Anshit Goyal, Che Ngufor, Panagiotis Kerezoudis, Brandon McCutcheon, Curtis Storlie and Mohamad Bydon

OBJECTIVE

Nonhome discharge and unplanned readmissions represent important cost drivers following spinal fusion. The authors sought to utilize different machine learning algorithms to predict discharge to rehabilitation and unplanned readmissions in patients receiving spinal fusion.

METHODS

The authors queried the 2012–2013 American College of Surgeons National Surgical Quality Improvement Program (ACS-NSQIP) for patients undergoing cervical or lumbar spinal fusion. Outcomes assessed included discharge to nonhome facility and unplanned readmissions within 30 days after surgery. A total of 7 machine learning algorithms were evaluated. Predictive hierarchical clustering of procedure codes was used to increase model performance. Model performance was evaluated using overall accuracy and area under the receiver operating characteristic curve (AUC), as well as sensitivity, specificity, and positive and negative predictive values. These performance metrics were computed for both the imputed and unimputed (missing values dropped) datasets.

RESULTS

A total of 59,145 spinal fusion cases were analyzed. The incidence rates of discharge to nonhome facility and 30-day unplanned readmission were 12.6% and 4.5%, respectively. All classification algorithms showed excellent discrimination (AUC > 0.80, range 0.85–0.87) for predicting nonhome discharge. The generalized linear model showed comparable performance to other machine learning algorithms. By comparison, all models showed poorer predictive performance for unplanned readmission, with AUC ranging between 0.63 and 0.66. Better predictive performance was noted with models using imputed data.

CONCLUSIONS

In an analysis of patients undergoing spinal fusion, multiple machine learning algorithms were found to reliably predict nonhome discharge with modest performance noted for unplanned readmissions. These results provide early evidence regarding the feasibility of modern machine learning classifiers in predicting these outcomes and serve as possible clinical decision support tools to facilitate shared decision making.

Full access

Mohammed Ali Alvi, Redab Alkhataybeh, Waseem Wahood, Panagiotis Kerezoudis, Sandy Goncalves, M. Hassan Murad and Mohamad Bydon

OBJECTIVE

Transpsoas lateral interbody fusion is one of the lateral minimally invasive approaches for lumbar spine surgery. Most surgeons insert the interbody cage laterally and then insert pedicle or cortical screw and rod instrumentation posteriorly. However, standalone cages have also been used to avoid posterior instrumentation. To the best of the authors’ knowledge, the literature on comparison of the two approaches is sparse.

METHODS

The authors performed a systematic review and meta-analysis of the available literature on transpsoas lateral interbody fusion by an electronic search of the PubMed, EMBASE, and Scopus databases using PRISMA guidelines. They compared patients undergoing transpsoas standalone fusion (TP) with those undergoing transpsoas fusion with posterior instrumentation (TPP).

RESULTS

A total of 28 studies with 1462 patients were included. Three hundred and seventy-four patients underwent TPP, and 956 patients underwent TP. The mean patient age ranged from 45.7 to 68 years in the TP group, and 50 to 67.7 years in the TPP group. The incidence of reoperation was found to be higher for TP (0.08, 95% confidence interval [CI] 0.04–0.11) compared to TPP (0.03, 95% CI 0.01–0.06; p = 0.057). Similarly, the incidence of cage movement was found to be greater in TP (0.18, 95% CI 0.10–0.26) compared to TPP (0.03, 95% CI 0.00–0.05; p < 0.001). Oswestry Disability Index (ODI) and visual analog scale (VAS) scores and postoperative transient deficits were found to be comparable between the two groups.

CONCLUSIONS

These results appear to suggest that addition of posterior instrumentation to transpsoas fusion is associated with decreased reoperations and cage movements. The results of previous systematic reviews and meta-analyses should be reevaluated in light of these results, which seem to suggest that higher reoperation and subsidence rates may be due to the use of the standalone technique.

Restricted access

Mohammed Adeeb Sebai, Panagiotis Kerezoudis, Mohammed Ali Alvi, Jang Won Yoon, Robert J. Spinner and Mohamad Bydon

OBJECTIVE

Spinal peripheral nerve sheath tumors (PNSTs) are a group of rare tumors originating from the nerve and its supporting structures. Standard surgical management typically entails laminectomy with or without facetectomy to gain adequate tumor exposure. Arthrodesis is occasionally performed to maintain spinal stability and mitigate the risk of postoperative deformity, pain, or neurological deficit. However, the factors associated with the need for instrumentation in addition to PNST resection in the same setting remain unclear.

METHODS

An institutional tumor registry at a tertiary care center was queried for patients treated surgically for a primary diagnosis of spinal PNST between 2002 and 2016. An analysis focused on patients in whom a facetectomy was performed during the resection. The addition of arthrodesis at the index procedure comprised the primary outcome. The authors also recorded baseline demographics, tumor characteristics, and surgery-related variables. Logistic regression was used to identify factors associated with increased risk of fusion surgery.

RESULTS

A total of 163 patients were identified, of which 56 (32 had facetectomy with fusion, 24 had facetectomy alone) were analyzed. The median age was 48 years, and 50% of the cohort was female. Age, sex, and race, as well as tumor histology and size, were evenly distributed between patients who received facetectomy alone and those who had facetectomy and fusion. On univariate analysis, total versus subtotal facetectomy (OR 9.0, 95% CI 2.01–64.2; p = 0.009) and cervicothoracic versus other spinal region (OR 9.0, 95% CI 1.51–172.9; p = 0.048) were significantly associated with increased odds of performing immediate fusion. On multivariable analysis, only the effect of total facetectomy remained statistically significant (OR 6.75, 95% CI 1.47–48.8; p = 0.025).

CONCLUSIONS

The authors found that total facetectomy and cervicothoracic involvement may be highly associated with the need for concomitant arthrodesis at the time of index surgery. These findings may help surgeons to determine the best surgical planning for patients with PNST.

Restricted access

Lorenzo Rinaldo, Brandon A. McCutcheon, Meghan E. Murphy, Daniel L. Shepherd, Patrick R. Maloney, Panagiotis Kerezoudis, Mohamad Bydon and Giuseppe Lanzino

OBJECTIVE

The mechanism by which greater institutional case volume translates into improved outcomes after surgical clipping of unruptured intracranial aneurysms (UIAs) is not well established. The authors thus aimed to assess the effect of case volume on the rate of various types of complications after clipping of UIAs.

METHODS

Using information on the outcomes of inpatient admissions for surgical clipping of UIAs collected within a national database, the relationship of institutional case volume to the incidence of different types of complications after clipping was investigated. Complications were subdivided into different categories, which included all complications, ischemic stroke, intracerebral hemorrhage, medical complications, infectious complications, complications related to anesthesia, and wound complications. The relationship of case volume to different types of complications was assessed using linear regression analysis. The relationships between case volume and overall complication and stroke rates were fit with both linear and quadratic equations. The numerical cutoff for institutional case volume above and below which the authors found the greatest differences in mean overall complication and stroke rate was determined using classification and regression tree (CART) analysis.

RESULTS

Between October 2012 and September 2015, 125 health care institutions reported patient outcomes from a total of 6040 cases of clipping of UIAs. On linear regression analysis, increasing case volume was negatively correlated to both overall complications (r2 = 0.046, p = 0.0234) and stroke (r2 = 0.029, p = 0.0557) rate, although the relationship of case volume to the complication (r2 = 0.092) and stroke (r2 = 0.067) rate was better fit with a quadratic equation. On CART analysis, the cutoff for the case number that yielded the greatest difference in overall complications and stroke rate between higher- or lower-volume centers was 6 cases/year and 3 cases/year, respectively.

CONCLUSIONS

Although the authors confirm that increasing case volume is associated with reduced complications after clipping of UIAs, their results suggest that the relationship between case volume and complications is not necessarily linear. Moreover, these results indicate that the effect of case volume on outcome is most evident between very-low-volume centers relative to centers with a medium-to-high volume.

Restricted access

Mohammed Ali Alvi, Lorenzo Rinaldo, Panagiotis Kerezoudis, Leonardo Rangel-Castilla, Mohamad Bydon, Harry Cloft and Giuseppe Lanzino

OBJECTIVE

The impact of FDA approval of flow-diversion technology for the treatment of supraclinoid internal carotid artery aneurysms and the publication of the Carotid Occlusion Surgery Study, both of which occurred in 2011, on the utilization of extracranial-intracranial (EC-IC) bypasses is not known.

METHODS

The National Inpatient Sample (NIS) was queried for hospitalizations for EC-IC bypass performed from 2008 to 2016. Diagnoses of interest included an unruptured intracranial aneurysm (UIA), subarachnoid hemorrhage (SAH), carotid occlusive disease (COD), and moyamoya disease. The authors assessed trends in EC-IC bypass utilization for these diagnoses and the incidence of adverse discharges, defined as discharge to locations other than home, and the rate of mortality.

RESULTS

A total of 1640 EC-IC bypass procedures were performed at 558 hospitals during the study period, with 1148 procedures at 448 hospitals performed for a diagnosis of interest. The most frequent surgical indication was moyamoya disease (65.7%, n = 754), followed by COD (23.2%, n = 266), SAH (3.2%, n = 37), and a UIA (7.9%, n = 91). EC-IC bypass utilization for COD decreased from 0.21 per 100 admissions of COD in 2010 to 0.09 per 100 admissions in 2016 (p = 0.023). The frequency of adverse discharges increased during the study period from 22.3% of annual admissions in 2008 to 31.2% in 2016 (p = 0.030) when analysis was limited to procedures performed for a diagnosis of interest. Per volume, the top 5th percentile of hospitals, on average, performed 18.4 procedures (SD 13.2) per hospital during the study period, compared to 1.3 procedures (SD 1.3) that were performed in hospitals within the bottom 95th percentile. The rate of adverse discharges was higher at low-volume institutions when compared to that at high-volume institutions (33.8% vs 28.7%; p = 0.029). Over the study period, the authors noted a trend toward a reduced percentage of total surgical volume performed at high-volume hospitals (p < 0.001).

CONCLUSIONS

The authors observed a decrease in the utilization of EC-IC bypass for COD during the study period. An increase in the rate of adverse discharges was also noted, coinciding with more procedures being performed at lower-volume centers.

Free access

Anshit Goyal, Yagiz U. Yolcu, Aakshit Goyal, Panagiotis Kerezoudis, Desmond A. Brown, Christopher S. Graffeo, Sandy Goncalves, Terence C. Burns and Ian F. Parney

OBJECTIVE

With the revised WHO 2016 classification of brain tumors, there has been increasing interest in imaging biomarkers to predict molecular status and improve the yield of genetic testing for diffuse low-grade gliomas (LGGs). The T2-FLAIR–mismatch sign has been suggested to be a highly specific radiographic marker of isocitrate dehydrogenase (IDH) gene mutation and 1p/19q codeletion status in diffuse LGGs. The presence of T2-FLAIR mismatch indicates a T2-hyperintense lesion that is hypointense on FLAIR with the exception of a hyperintense rim.

METHODS

In accordance with PRISMA guidelines, we performed a systematic review of the Ovid Medline, Embase, Scopus, and Cochrane databases for reports of studies evaluating the diagnostic performance of T2-FLAIR mismatch in predicting the IDH and 1p/19q codeletion status in diffuse LGGs. Results were combined into a 2 × 2 format, and the following diagnostic performance parameters were calculated: sensitivity, specificity, positive predictive value, negative predictive value, and positive (LR+) and negative (LR−) likelihood ratios. In addition, we utilized Bayes theorem to calculate posttest probabilities as a function of known pretest probabilities from previous genome-wide association studies and the calculated LRs. Calculations were performed for 1) IDH mutation with 1p/19q codeletion (IDHmut-Codel), 2) IDH mutation without 1p/19q codeletion (IDHmut-Noncodel), 3) IDH mutation overall, and 4) 1p/19q codeletion overall. The QUADAS-2 (revised Quality Assessment of Diagnostic Accuracy Studies) tool was utilized for critical appraisal of included studies.

RESULTS

A total of 4 studies were included, with inclusion of 2 separate cohorts from a study reporting testing and validation (n = 746). From pooled analysis of all cohorts, the following values were obtained for each molecular profile—IDHmut-Codel: sensitivity 30%, specificity 73%, LR+ 1.1, LR− 1.0; IDHmut-Noncodel: sensitivity 33.7%, specificity 98.5%, LR+ 22.5, LR− 0.7; IDH: sensitivity 32%, specificity 100%, LR+ 32.1, LR− 0.7; 1p/19q codeletion: sensitivity 0%, specificity 54%, LR+ 0.01, LR− 1.9. Bayes theorem was used to calculate the following posttest probabilities after a positive and negative result, respectively—IDHmut-Codel: 32.2% and 29.4%; IDHmut-Noncodel: 95% and 40%; IDH: 99.2% and 73.5%; 1p/19q codeletion: 0.4% and 35.1%.

CONCLUSIONS

The T2-FLAIR–mismatch sign was an insensitive but highly specific marker of IDH mutation and IDHmut-Noncodel profile, although significant exceptions may exist to this finding. Tumors with a positive sign may still be IDHwt or 1p/19q codeleted. These findings support the utility of T2-FLAIR mismatch as an imaging-based biomarker for positive selection of patients with IDH-mutant gliomas.

Free access

Mohamed Macki, Mohammed Ali Alvi, Panagiotis Kerezoudis, Shujie Xiao, Lonni Schultz, Michael Bazydlo, Mohamad Bydon, Paul Park, Victor Chang and for the MSSIC Investigators

OBJECTIVE

As compensation transitions from a fee-for-service to pay-for-performance healthcare model, providers must prioritize patient-centered experiences. Here, the authors’ primary aim was to identify predictors of patient dissatisfaction at 1 and 2 years after lumbar surgery.

METHODS

The Michigan Spine Surgery Improvement Collaborative (MSSIC) was queried for all lumbar operations at the 1- and 2-year follow-ups. Predictors of patients’ postoperative contentment were identified per the North American Spine Surgery (NASS) Patient Satisfaction Index, wherein satisfied patients were assigned a score of 1 (“the treatment met my expectations”) or 2 (“I did not improve as much as I had hoped, but I would undergo the same treatment for the same outcome”) and unsatisfied patients were assigned a score of 3 (“I did not improve as much as I had hoped, and I would not undergo the same treatment for the same outcome”) or 4 (“I am the same or worse than before treatment”). Multivariable Poisson generalized estimating equation models were used to report adjusted risk ratios (RRadj).

RESULTS

Among 5390 patients with a 1-year follow-up, 22% reported dissatisfaction postoperatively. Dissatisfaction was predicted by higher body mass index (RRadj =1.07, p < 0.001), African American race compared to white (RRadj = 1.51, p < 0.001), education level less than high school graduation compared to a high school diploma or equivalent (RRadj = 1.25, p = 0.008), smoking (RRadj = 1.34, p < 0.001), daily preoperative opioid use > 6 months (RRadj = 1.22, p < 0.001), depression (RRadj = 1.31, p < 0.001), symptom duration > 1 year (RRadj = 1.32, p < 0.001), previous spine surgery (RRadj = 1.32, p < 0.001), and higher baseline numeric rating scale (NRS)–back pain score (RRadj = 1.04, p = 0.002). Conversely, an education level higher than high school graduation, independent ambulation (RRadj = 0.90, p = 0.039), higher baseline NRS–leg pain score (RRadj = 0.97, p = 0.013), and fusion surgery (RRadj = 0.88, p = 0.014) decreased dissatisfaction.

Among 2776 patients with a 2-year follow-up, 22% reported dissatisfaction postoperatively. Dissatisfaction was predicted by a non-white race, current smoking (RRadj = 1.26, p = 0.004), depression (RRadj = 1.34, p < 0.001), symptom duration > 1 year (RRadj = 1.47, p < 0.001), previous spine surgery (RRadj = 1.28, p < 0.001), and higher baseline NRS–back pain score (RRadj = 1.06, p = 0.003). Conversely, at least some college education (RRadj = 0.87, p = 0.035) decreased the risk of dissatisfaction.

CONCLUSIONS

Both comorbid conditions and socioeconomic circumstances must be considered in counseling patients on postoperative expectations. After race, symptom duration was the strongest predictor of dissatisfaction; thus, patient-centered measures must be prioritized. These findings should serve as a tool for surgeons to identify at-risk populations that may need more attention regarding effective communication and additional preoperative counseling to address potential barriers unique to their situation.

Full access

Benjamin F. Mundell, Marcus J. Gates, Panagiotis Kerezoudis, Mohammed Ali Alvi, Brett A. Freedman, Ahmad Nassr, Samuel F. Hohmann and Mohamad Bydon

OBJECTIVE

From 1994 to 2006 outpatient spinal surgery increased 5-fold. The perceived cost savings with outcomes comparable to or better than those achieved with inpatient admission for the same procedures are desirable in an era where health expenditures are scrutinized. The increase in outpatient spine surgery is also driven by the proliferation of ambulatory surgery centers. In this study, the authors hypothesized that the total savings in outpatient spine surgery is largely driven by patient selection and biases toward healthier patients.

METHODS

A meta-analysis assessed patient selection factors and outcomes associated with outpatient spine procedures. Pooled odds ratios and mean differences were calculated using a Bayesian random-effects model. The authors extended this analysis in a novel way by using the results of the meta-analysis to examine cost data from an administrative database of academically affiliated hospitals. A Bayesian approach with priors informed by the meta-analysis was used to compare costs for inpatient and outpatient performance of anterior cervical discectomy and fusion (ACDF) and lumbar laminectomy.

RESULTS

Sixteen studies with a total of 370,195 patients met the inclusion criteria. Outpatient procedures were associated with younger patient age (mean difference [MD] −2.34, 95% credible interval [CrI] −4.39 to −0.34) and no diabetes diagnosis (odds ratio [OR] 0.78, 95% CrI 0.54–0.97). Outpatient procedures were associated with a lower likelihood of reoperation (OR 0.42, 95% CrI 0.16–0.80), 30-day readmission (OR 0.39, 95% CrI 0.16–0.74), and complications (OR 0.29, 95% CrI 0.15–0.50) and with lower overall costs (MD −$121,392.72, 95% CrI −$216,824.81 to −$23,632.92). Additional analysis of the national administrative data revealed more modest cost savings than those found in the meta-analysis for outpatient spine surgeries relative to inpatient spine surgeries. Estimated cost savings for both younger patients ($555 for those age 30–35 years [95% CrI −$733 to −$374]) and older patients ($7290 for those age 65–70 years [95% CrI −$7380 to −$7190]) were less than the overall cost savings found in the meta-analysis.

CONCLUSIONS

Compared to inpatient spine surgery, outpatient spine surgery was associated with better short-term outcomes and an initial reduction in direct costs. A selection bias for outpatient procedures toward younger, healthier patients may confound these results. The additional analysis of the national database suggests that cost savings in the outpatient setting may be less than previously reported and a result of outpatient procedures being offered more frequently to younger and healthier individuals.

Restricted access

Anthony L. Asher, Clinton J. Devin, Panagiotis Kerezoudis, Hui Nian, Mohammed Ali Alvi, Inamullah Khan, Ahilan Sivaganesan, Frank E. Harrell Jr., Kristin R. Archer and Mohamad Bydon

OBJECTIVE

Patient satisfaction with treatment outcome is gaining an increasingly important role in assessing the value of surgical spine care delivery. Nationwide data evaluating the predictors of patient satisfaction in elective cervical spine surgery are lacking. The authors sought to decipher the impacts of the patient, surgical practice, and surgeon on satisfaction with outcome following anterior cervical discectomy and fusion (ACDF).

METHODS

The authors queried the Quality Outcomes Database for patients undergoing 1- to 2-level ACDF for degenerative spine disease since 2013. Patient satisfaction with the surgical outcome as measured by the North American Spine Society (NASS) scale comprised the primary outcome. A multivariable proportional odds logistic regression model was constructed with adjustments for baseline patient characteristics and surgical practice and surgeon characteristics as fixed effects.

RESULTS

A total of 4148 patients (median age 54 years, 48% males) with complete 12-month NASS satisfaction data were analyzed. Sixty-seven percent of patients answered that “surgery met their expectations” (n = 2803), while 20% reported that they “did not improve as much as they had hoped but they would undergo the same operation for the same results” (n = 836). After adjusting for a multitude of patient-specific as well as hospital- and surgeon-related factors, the authors found baseline Neck Disability Index (NDI) score, US geographic region of hospital, patient race, insurance status, symptom duration, and Workers’ compensation status to be the most important predictors of patient satisfaction. The discriminative ability of the model was satisfactory (c-index 0.66, overfitting-corrected estimate 0.64).

CONCLUSIONS

The authors’ results found baseline NDI score, patient race, insurance status, symptom duration, and Workers’ compensation status as well as the geographic region of the hospital to be the most important predictors of long-term patient satisfaction after a 1- to 2-level ACDF. The findings of the present analysis further reinforce the role of preoperative discussion with patients on setting treatment goals and realistic expectations.

Restricted access

Sung Huang Laurent Tsai, Anshit Goyal, Mohammed Ali Alvi, Panagiotis Kerezoudis, Yagiz Ugur Yolcu, Waseem Wahood, Elizabeth B. Habermann, Terry C. Burns and Mohamad Bydon

OBJECTIVE

The nature of the volume-outcome relationship in cases with severe traumatic brain injury (TBI) remains unclear, with considerable interhospital variation in patient outcomes. The objective of this study was to understand the state of the volume-outcome relationship at different levels of trauma centers in the United States.

METHODS

The authors queried the National Trauma Data Bank for the years 2007–2014 for patients with severe TBI. Case volumes for each level of trauma center organized into quintiles (Q1–Q5) served as the primary predictor. Analyzed outcomes included in-hospital mortality, total hospital length of stay (LOS), and intensive care unit (ICU) stay. Multivariable regression models were performed for in-hospital mortality, overall complications, and total hospital and ICU LOSs to adjust for possible confounders. The analysis was stratified by level designation of the trauma center. Statistical significance was established at p < 0.001 to avoid a type I error due to a large sample size.

RESULTS

A total of 122,445 patients were included. Adjusted analysis did not demonstrate a significant relationship between increasing hospital volume of severe TBI cases and in-hospital mortality, complications, and nonhome hospital discharge disposition among level I–IV trauma centers. However, among level II trauma centers, hospital LOS was longer for the highest volume quintile (adjusted mean difference [MD] for Q5: 2.83 days, 95% CI 1.40–4.26 days, p < 0.001, reference = Q1). For level III and IV trauma centers, both hospital LOS and ICU LOS were longer for the highest volume quintile (adjusted MD for Q5: LOS 4.6 days, 95% CI 2.3–7.0 days, p < 0.001; ICU LOS 3.2 days, 95% CI 1.6–4.8 days, p < 0.001).

CONCLUSIONS

Higher volumes of severe TBI cases at a lower level of trauma center may be associated with a longer LOS. These results may assist policymakers with target interventions for resource allocation and point to the need for careful prehospital decision-making in patients with severe TBI.