Justin Mathew

Justin Mathew

Auto-generated profile, based on publicly available information

Justin Mathew

Department of Orthopaedic Surgery, Columbia University Medical Center, New York, NY, USA

Research Focus & Metrics

Academic Career Length

7Years (2020-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
28 February 2022 in Global Spine Journal
Verified
Journal Article
Open access
CLCesar D. Lopez
Cesar D. Lopez
VBVenkat Boddapati
Venkat Boddapati
JLJoseph M. Lombardi
Joseph M. Lombardi
NLNathan J. Lee
Nathan J. Lee
JMJustin Mathew
Justin Mathew
RLRonald A. Lehman
Ronald A. Lehman
Objectives: This current systematic review sought to identify and evaluate all current research-based spine surgery applications of AI/ML in optimizing preoperative patient selection, as well as predicting and managing postoperative outcomes and complications. Methods: A comprehensive search of publications was conducted through the EMBASE, Medline, and PubMed databases using relevant keywords to maximize the sensitivity of the search. No limits were placed on level of evidence or timing of the study. Findings were reported according to the PRISMA guidelines Results: After application of inclusion and exclusion criteria, 41 studies were included in this review. Bayesian networks had the highest average AUC (.80), and neural networks had the best accuracy (83.0%), sensitivity (81.5%), and specificity (71.8%). Preoperative planning/cost prediction models (.89,82.2%) and discharge/length of stay models (.80,78.0%) each reported significantly higher average AUC and accuracy compared to readmissions/reoperation prediction models (.67,70.2%) ( P < .001, P = .005, respectively). Model performance also significantly varied across postoperative management applications for average AUC and accuracy values ( P < .001, P < .027, respectively). Conclusions: Generally, authors of the reviewed studies concluded that AI/ML offers a potentially beneficial tool for providers to optimize patient care and improve cost-efficiency. More specifically, AI/ML models performed best, on average, when optimizing preoperative patient selection and planning and predicting costs, hospital discharge, and length of stay. However, models were not as accurate in predicting postoperative complications, adverse events, and readmissions and reoperations. An understanding of AI/ML-based applications is becoming increasingly important, particularly in spine surgery, as the volume of reported literature, technology accessibility, and clinical applications continue to rapidly expand.

23 September 2020 in Global Spine Journal
Verified
Journal Article
Open access
PDF file
NLNathan J. Lee
Nathan J. Lee
MFMichael Fields
Michael Fields
VBVenkat Boddapati
Venkat Boddapati
JMJustin Mathew
Justin Mathew
DHDaniel Hong
Daniel Hong
LLLawrence G. Lenke
Lawrence G. Lenke
Study Design: Retrospective cohort. Objective: To provide a national-level assessment of the short-term outcomes after spinal deformity surgery in pediatric patients with cerebral palsy. Methods: A national, prospectively collected database was queried to identify pediatric (≤18 years) patients with cerebral palsy, who underwent spinal fusion surgery from 2012 to 2017. Separate multivariate analyses were performed for the primary outcomes of interest including extended length of stay (>75th percentile, >8 days), and readmissions within 90 days after the index admission. Results: A total of 2856 patients were reviewed. The mean age ± standard deviation was 12.8 ± 2.9 years, and 49.4% of patients were female. The majority of patients underwent a posterior spinal fusion (97.0%) involving ≥8 levels (79.9%) at a teaching hospital (96.6%). Top medical complications (24.5%) included acute respiratory failure requiring mechanical ventilation (11.4%), paralytic ileus (8.2%), and urinary tract infections (4.6%). Top surgical complications (40.7%) included blood transfusion (35.6%), wound complication (4.9%), and mechanical complication (2.7%). The hospital cost for patients with a length of hospital stay >8 days ($113 669) was nearly double than that of those with a shorter length of stay ($68 411). The 90-day readmission rate was 17.6% (mean days to readmission: 30.2). The most common reason for readmission included wound dehiscence (21.1%), surgical site infection (19.1%), other infection (18.9%), dehydration (16.9%), feeding issues (14.5%), and acute respiratory failure (13.1%). Notable independent predictors for 90-day readmissions included preexisting pulmonary disease (odds ratio [OR] 1.5), obesity (OR 3.4), cachexia (OR 27), nonteaching hospital (OR 3.5), inpatient return to operating room (OR 1.9), and length of stay >8 days (OR 1.5). Conclusions: Efforts focused on optimizing the perioperative pulmonary, hematological, and nutritional status as well as reducing wound complications appear to be the most important for improving clinical outcomes.

8 July 2021 in Global Spine Journal
Verified
Journal Article
Open access
PDF file
NLNathan J. Lee
Nathan J. Lee
SZScott L. Zuckerman
Scott L. Zuckerman
IBIan A. Buchanan
Ian A. Buchanan
VBVenkat Boddapati
Venkat Boddapati
JMJustin Mathew
Justin Mathew
RLRonald A. Lehman
Ronald A. Lehman
Prospective single-cohort analysis. To compare the outcomes/complications of 2 robotic systems for spine surgery. Adult patients (≥18-years-old) who underwent robot-assisted spine surgery from 2016-2019 were assessed. A propensity score matching (PSM) algorithm was used to match Mazor X to Renaissance cases. Preoperative CT scan for planning and an intraoperative O-arm for screw evaluation were preformed. Outcomes included screw accuracy, robot time/screw, robot abandonment, and radiation. Screw accuracy was measured using Vitrea Core software by 2 orthopedic surgeons. Screw breach was measured according to the Gertzbein/Robbins classification. After PSA, a total of 65 patients (Renaissance: 22 vs. X: 43) were included. Patient/operative factors were similar between robot systems (P > .05). The pedicle screw accuracy was similar between robots (Renaissance: 1.1%% vs. X: 1.3%, P = .786); however, the S2AI screw breach rate was significantly lower for the X (Renaissance: 9.5% vs. X: 1.2%, P = .025). Robot time per screw was not statistically different (Renaissance: 4.6 minutes vs. X: 3.9 minutes, P = .246). The X was more reliable with an abandonment rate of 2.3% vs. Renaissance:22.7%, P = .007. Radiation exposure were not different between robot systems. Non-robot related complications including dural tear, loss of motor/sensory function, and blood transfusion were similar between robot systems. This is the first comparative analyses of screw accuracy, robot time/screw, robot abandonment, and radiation exposure between the Mazor X and Renaissance systems. There are substantial improvements in the X robot, particularly in the perioperative planning processes, which likely contribute to the X’s superiority in S2AI screw accuracy by nearly 8-fold and robot reliability by nearly 10-fold.

7 January 2021 in Global Spine Journal
Verified
Journal Article
Open access
PDF file
VBVenkat Boddapati
Venkat Boddapati
NLNathan J. Lee
Nathan J. Lee
JMJustin Mathew
Justin Mathew
MHMichael B. Held
Michael B. Held
JPJoel R. Peterson
Joel R. Peterson
KRK. Daniel Riew
K. Daniel Riew
Study design: Retrospective cohort study. Objective: Respiratory compromise (RC) is a rare but catastrophic complication of anterior cervical spine surgery (ACSS) commonly due to compressive fluid collections or generalized soft tissue swelling in the cervical spine. Established risk factors include operative duration, size of surgical exposure, myelopathy, among others. The purpose of this current study is to identify the incidence and clinical course of patients who develop RC, and identify independent predictors of RC in patients undergoing ACSS for cervical spondylosis. Methods: A large, prospectively-collected registry was used to identify patients undergoing ACSS for spondylosis. Patients with posterior cervical procedures were excluded. Baseline patient characteristics were compared using bivariate analysis, and multivariate analysis was employed to compare postoperative complications and identify independent predictors of RC. Results: 298 of 52,270 patients developed RC (incidence 0.57%). Patients who developed RC had high rates of 30-day mortality (11.7%) and morbidity (75.8%), with unplanned reoperation and pneumonia the most common. The most common reason for reoperations were hematoma evacuation and tracheostomy. Independent patient-specific factors predictive of RC included increasing patient age, male gender, comorbidities such as chronic cardiac and respiratory disease, preoperative myelopathy, prolonged operative duration, and 2-level ACCFs. Conclusion: This is among the largest cohorts of patients to develop RC after ACSS identified to-date and validates a range of independent predictors, many previously only described in case reports. These results are useful for taking preventive measures, identifying high risk patients for preoperative risk stratification, and for surgical co-management discussions with the anesthesiology team.

19 February 2021 in Global Spine Journal
Verified
Journal Article
Open access
PDF file
NLNathan J. Lee
Nathan J. Lee
VBVenkat Boddapati
Venkat Boddapati
JMJustin Mathew
Justin Mathew
MFMichael Fields
Michael Fields
MVMeghana Vulapalli
Meghana Vulapalli
KRK. Daniel Riew
K. Daniel Riew
Retrospective cohort. Provide a comparison of surgical approach in the treatment of degenerative cervical myelopathy in patients with OPLL. A national database was queried to identify adult (≥18 years) patients with OPLL, who underwent at least a 2-level cervical decompression and fusion for cervical myelopathy from 2012-2014. A propensity-score-matching algorithm was employed to compare outcomes by surgical approach. After propensity-score matching, 627 patients remained. An anterior approach was found to be an independent predictor for higher inpatient surgical complications(OR 5.9), which included dysphagia:14%[anterior]vs.1.1%[posterior] P-value < 0.001, wound hematoma:1.7%[anterior]vs.0%[posterior] P-value = 0.02, and dural tear:9.4%[anterior]vs.3.2%[posterior] P-value = 0.001. A posterior approach was an predictor for longer hospital length of stay by nearly 3 days(OR 3.4; 6.8 days[posterior]vs.4.0 days[anterior] P-value < 0.001). The reasons for readmission/reoperation did not vary by approach for 2-3-level fusions; however, for >3-level fusions, patients with an anterior approach more often had respiratory complications requiring mechanical ventilation(P-value = 0.038) and required revision fusion surgery(P-value = 0.015). The national estimates for inpatient complications(25%), readmissions(9.9%), and reoperations(3.5%) are substantial after the surgical treatment of multi-level OPLL. An anterior approach resulted in significantly higher inpatient surgical complications, but this did not result in a longer hospital length of stay and the overall 90-day complication rates requiring readmission or reoperation was similar to those seen after a posterior approach. For patients requiring >3-level fusion, an anterior approach is associated with significantly higher risk for respiratory complications requiring mechanical ventilation and revision fusion surgery. Precise neurological complications and functional outcomes were not included in this database, and should be further assessed in future studies.

8 October 2020 in Global Spine Journal
Verified
Journal Article
Open access
PDF file
NLNathan J. Lee
Nathan J. Lee
ZSZeeshan M. Sardar
Zeeshan M. Sardar
VBVenkat Boddapati
Venkat Boddapati
JMJustin Mathew
Justin Mathew
MCMeghan Cerpa
Meghan Cerpa
RLRonald A. Lehman
Ronald A. Lehman
Study Design: Consecutively collected cases. Objective: To determine if a machine-learning (ML) program can accurately predict the postoperative thoracic kyphosis through the uninstrumented thoracic spine and pelvic compensation in patients who undergo fusion from the lower thoracic spine (T10 or T11) to the sacrum. Methods: From 2015 to 2019, a consecutive series of adult (≥18 years old) patients with adult spinal deformity underwent corrective spinal fusion from the lower thoracic spine (T10 or T11) to the sacrum. Deidentified data was processed by a ML system-based platform to predict the postoperative thoracic kyphosis (TK) and pelvic tilt (PT) for each patient. To validate the ML model, the postoperative TK (T4-T12, instrumented thoracic, and uninstrumented thoracic) and the pelvic tilt were compared against the predicted values. Results: A total of 20 adult patients with a minimum 6-month follow-up (mean: 22.4 ± 11.3 months) were included in this study. No significant differences were observed for TK (predicted 37.6° vs postoperative 38.3°, P = .847), uninstrumented TK (predicted 33.9° vs postoperative 29.8°, P = .188), and PT (predicted 23.4° vs postoperative 22.7°, P = .754). The predicted PT and the TK of the uninstrumented thoracic spine correlated well with postoperative values (uninstrumented TK: R2 = 0.764, P < .001; PT: R2 = 0.868, P < .001). The mean error with which kyphosis through the uninstrumented thoracic spine can be measured was 4.8° ± 4.0°. The mean error for predicting PT was 2.5° ± 1.7°. Conclusion: ML algorithms can accurately predict the spinopelvic compensation after spinal fusion from the lower thoracic spine to the sacrum. These findings suggest that surgeons may be able to leverage this technology to reduce the risk of proximal junctional kyphosis in this population.