Performance of Artificial Intelligence-Based Algorithms to Predict Prolonged Length of Stay after Lumbar Decompression Surgery

Background: Decompression of the lumbar spine is one of the most common procedures performed in spine surgery. Hospital length of stay (LOS) is a clinically relevant metric used to assess surgical success, patient outcomes, and socioeconomic impact. This study aimed to investigate a variety of machine learning and deep learning algorithms to reliably predict whether a patient undergoing decompression of lumbar spinal stenosis will experience a prolonged LOS. Methods: Patients undergoing treatment for lumbar spinal stenosis with microsurgical and full-endoscopic decompression were selected within this retrospective monocentric cohort study. Prolonged LOS was defined as an LOS greater than or equal to the 75th percentile of the cohort (normal versus prolonged stay; binary classification task). Unsupervised learning with K-means clustering was used to find clusters in the data. Hospital stay classes were predicted with logistic regression, RandomForest classifier, stochastic gradient descent (SGD) classifier, K-nearest neighbors, Decision Tree classifier, Gaussian Naive Bayes (GaussianNB), support vector machines (SVM), a custom-made convolutional neural network (CNN), multilayer perceptron artificial neural network (MLP), and radial basis function neural network (RBNN) in Python. Prediction accuracy and area under the curve (AUC) were calculated. Feature importance analysis was utilized to find the most important predictors. Further, we developed a decision tree based on the Chi-square automatic interaction detection (CHAID) algorithm to investigate cut-offs of predictors for clinical decision-making. Results: 236 patients and 14 feature variables were included. K-means clustering separated data into two clusters distinguishing the data into two patient risk characteristic groups. The algorithms reached AUCs between 67.5% and 87.3% for the classification of LOS classes. Feature importance analysis of deep learning algorithms indicated that operation time was the most important feature in predicting LOS. A decision tree based on CHAID could predict 84.7% of the cases. Conclusions: Machine learning and deep learning algorithms can predict whether patients will experience an increased LOS following lumbar decompression surgery. Therefore, medical resources can be more appropriately allocated to patients who are at risk of prolonged LOS.

Non-English Primary Language Is Associated with Short-Term Outcomes After Supratentorial Tumor Resection

BACKGROUND

Despite research indicating that patients with non-English primary language (NEPL) have increased hospital length of stay (LOS) for craniotomies, there is a paucity of neurosurgical research examining the impact of language on short-term outcomes. This study sought to evaluate short-term outcomes for patients with English primary language (EPL) and NEPL admitted for resection of a supratentorial tumor.

METHODS

Using the Agency for Healthcare Research and Quality Healthcare Cost and Utilization Project New Jersey State Inpatient Database, this study included patients 18-90 years old who underwent resection of a supratentorial primary brain tumor, meningioma, or brain metastasis from 2009 to 2017. The primary outcomes were total, preoperative, and postoperative LOS. Secondary outcomes were complications, mortality, and discharge disposition. Univariable and multivariable analyses compared Spanish primary language (SPL), non-English non-Spanish (NENS) primary language, and EPL groups.

RESULTS

A total of 7324 patients were included: 2962 with primary brain tumor, 2091 with meningioma, and 2271 with brain metastasis. Patients with SPL (n = 297) were younger and more likely to have noncommercial insurance, lower income, and fewer comorbidities. Patients with NENS (n = 257) had similar age and comorbidities to the EPL group but had a greater proportion of noncommercially insured and low-income patients (P < 0.001). Multivariable analysis showed that patients with NENS had increased postoperative LOS (adjusted incidence rate ratio, 1.10; P = 0.008) and higher odds of a complication (adjusted odds ratio, 1.36; P = 0.015), and patients with SPL had higher odds of being discharged home (adjusted odds ratio, 1.55; P = 0.017).

CONCLUSIONS

Patients with NEPL have different short-term outcomes after supratentorial tumor resection that varies based on primary language. More research is needed to understand the mechanisms driving these findings and to clarify unique experiences for different populations with NEPL.

Machine Learning Prediction of Length of Stay in Adult Spinal Deformity Patients Undergoing Posterior Spine Fusion Surgery

(1) Background: Length of stay (LOS) is a commonly reported metric used to assess surgical success, patient outcomes, and economic impact. The focus of this study is to use a variety of machine learning algorithms to reliably predict whether a patient undergoing posterior spinal fusion surgery treatment for Adult Spine Deformity (ASD) will experience a prolonged LOS. (2) Methods: Patients undergoing treatment for ASD with posterior spinal fusion surgery were selected from the American College of Surgeon's NSQIP dataset. Prolonged LOS was defined as a LOS greater than or equal to 9 days. Data was analyzed with the Logistic Regression, Decision Tree, Random Forest, XGBoost, and Gradient Boosting functions in Python with the Sci-Kit learn package. Prediction accuracy and area under the curve (AUC) were calculated. (3) Results: 1281 posterior patients were analyzed. The five algorithms had prediction accuracies between 68% and 83% for posterior cases (AUC: 0.566-0.821). Multivariable regression indicated that increased Work Relative Value Units (RVU), elevated American Society of Anesthesiologists (ASA) class, and longer operating times were linked to longer LOS. (4) Conclusions: Machine learning algorithms can predict if patients will experience an increased LOS following ASD surgery. Therefore, medical resources can be more appropriately allocated towards patients who are at risk of prolonged LOS.

Length of Stay Beyond Medical Readiness in a Neurosurgical Patient Population and Associated Healthcare Costs

BACKGROUND

Length of stay beyond medical readiness (LOS-BMR) leads to increased expenses and higher morbidity related to hospital-acquired conditions.

OBJECTIVE

To determine the proportion of admitted neurosurgical patients who have LOS-BMR and associated risk factors and costs.

METHODS

We performed a prospective, cohort analysis of all neurosurgical patients admitted to our institution over 5 mo. LOS-BMR was assessed daily by the attending neurosurgeon and neuro-intensivist with a standardized criterion. Univariate and multivariate logistic regressions were performed.

RESULTS

Of the 884 patients admitted, 229 (25.9%) had a LOS-BMR. The average LOS-BMR was 2.7 ± 3.1 d at an average daily cost of $9 148.28 ± $12 983.10, which resulted in a total cost of $2 076 659.32 over the 5-mo period. Patients with LOS-BMR were significantly more likely to be older and to have hemiplegia, dementia, liver disease, renal disease, and diabetes mellitus. Patients with a LOS-BMR were significantly more likely to be discharged to a subacute rehabilitation/skilled nursing facility (40.2% vs 4.1%) or an acute/inpatient rehabilitation facility (22.7% vs 1.7%, P < .0001). Patients with Medicare insurance were more likely to have a LOS-BMR, whereas patients with private insurance were less likely (P = .048).

CONCLUSION

The most common reason for LOS-BMR was inefficient discharge of patients to rehabilitation and nursing facilities secondary to unavailability of beds at discharge locations, insurance clearance delays, and family-related issues.

Effects of the COVID-19 Pandemic on Operative Volume and Residency Training at Two Academic Neurosurgery Centers in New Orleans

BACKGROUND

Medical subspecialties including neurosurgery have seen a dramatic shift in operative volume in the wake of the coronavirus disease 2019 (COVID-19) pandemic. The goal of this study was to quantify the effects of the COVID-19 pandemic on operative volume at 2 academic neurosurgery centers in New Orleans, Louisiana, USA from equivalent periods before and during the COVID-19 pandemic.

METHODS

A retrospective review was conducted analyzing neurosurgical case records for 2 tertiary academic centers from March to June 2020 and March to June 2019. The records were reviewed for variables including institution and physician coverage, operative volume by month and year, cases per subspecialty, patient demographics, mortality, and morbidity.

RESULTS

Comparison of groups showed a 34% reduction in monthly neurosurgical volume per institution during the pandemic compared with earlier time points, including a 77% decrease during April 2020. There was no change in mortality and morbidity across institutions during the pandemic.

CONCLUSIONS

The COVID-19 pandemic has had a significant impact on neurosurgical practice and will likely continue to have long-term effects on patients at a time when global gross domestic products decrease and relative health expenditures increase. Clinicians must anticipate and actively prepare for these impacts in the future.

Readmission after neurosurgical intervention in epilepsy: A nationwide cohort analysis

OBJECTIVE

Hospital readmissions result in increased health care costs and are associated with worse outcomes after neurosurgical intervention. Understanding factors associated with readmissions will inform future studies aimed at improving quality of care in those with epilepsy.

METHODS

Patients of all ages with epilepsy who underwent a neurosurgical intervention were identified in the 2014 Nationwide Readmissions Database, a nationally representative dataset containing data from roughly 17 million US hospital discharges. Diagnosis of epilepsy was based on International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM)-based case definitions. Neurosurgical interventions for epilepsy: resective/disconnective surgery, responsive neurostimulation/deep brain stimulation, vagus nerve stimulation, radiosurgery, and intracranial electroencephalography were identified using ICD-9-CM procedure codes. Primary outcome was all-cause 30-day readmission following discharge from the index hospitalization.

RESULTS

There were a total of 2284 index surgical admissions. Overall, 10.83% (n = 251) of patients following an index epilepsy surgery admission were readmitted within 30 days. Factors independently associated with 30-day readmission for all epilepsy surgery admissions were: Medicare insurance (P < .01), discharge disposition that was not home (P < .01), higher Elixhauser comorbidity indexes (P < .01), longer length of stay (P < .01), and adverse events of surgical and medical care during index stay (P = .04). In the multivariate model, Medicare insurance (hazard ratio [HR] 1.81 [1.29-2.53], P < .01) and length of stay (HR 1.02 [1.01-1.04], P < .01) remained significant independent predictors for 30-day readmission. The most common primary reason for readmissions was epilepsy/convulsions accounting for 22.85%.

SIGNIFICANCE

Our results suggest that careful management of postoperative seizures and discharge planning after epilepsy surgery may be important to optimize outcomes and reduce the risk of readmission, particularly for patients on Medicare.