Extended length of stay in open versus minimally invasive surgery with robotic-assisted sub-analysis for spinal nerve sheath tumor resection: a nationwide analysis

Spinal nerve sheath tumors are slow-growing neoplasms that arise from Schwann cell lineage and encompass schwannomas, neurofibromas, hybrid nerve sheath tumors, and malignant peripheral nerve sheath tumors. These lesions most commonly present as intradural extramedullary (IDEM) tumors, although extradural and dumbbell-shaped variants are also observed. Due to their typically benign behavior, gross total resection (GTR) remains the standard of care. However, there is a paucity of literature comparing the impact of open versus minimally invasive surgery (MIS) on postoperative extended length of stay (LOS). Prolonged hospitalization can increase healthcare costs, patient morbidity, and resource utilization. This study aims to compare the impact of MIS and open surgical approaches on extended LOS in patients undergoing resection of spinal nerve sheath tumors. Patients diagnosed with spinal nerve sheath tumors between 2004 and 2017 were identified from the National Cancer Database (NCDB) using ICD-O code 8680, 9560, 9490, 9540, and 9561. The cohort was stratified into four racial groups: White, Black, Hispanic, and Asian. Univariate analyses were performed to compare demographic, disease characteristics, and clinical outcomes. Additionally, a multivariate linear regression model was constructed to identify factors associated with extended length of stay, adjusting for sex, race, surgical modality (MIS, open, robotics), use of robotic surgery, facility type, insurance status, distance from facility to patient, comorbidities, age category, tumor behavior, and tumor size. Extended length of stay was defined as hospitalization exceeding the 75th percentile of the entire study population's length of stay. A total of 5,968 patients with spinal nerve sheath tumors were identified: 202 (3.4%) underwent MIS and 5,766 (96.6%) underwent open surgery. After 1:1 propensity score matching, 404 patients were equally distributed between the two groups. Prior to matching, MIS was more frequently used in the South Atlantic and East North Central regions compared to open surgery (29.3% vs. 21.4%; 20.1% vs. 16.1%; p = 0.008). Postoperative LOS was significantly shorter in the MIS group both before (4.4 ± 3.1 vs. 5.3 ± 3.5 days; p < 0.001) and after matching (4.4 ± 3.0 vs. 5.4 ± 3.5 days; p < 0.001). Patients treated with MIS were also less likely to experience an extended LOS both before (21.5% vs. 32.1%; p = 0.002) and after matching (21.5% vs. 35.4%; p = 0.002). On multivariable analysis, geriatric age (OR: 1.28; 95% CI: 1.12-1.46; p < 0.001), comorbidity burden (1 comorbidity: OR: 1.47; 95% CI: 1.25-1.72; ≥2: OR: 2.15; 95% CI: 1.72-2.68; p < 0.001), larger tumor size (OR: 1.02; 95% CI: 1.01-1.02; p < 0.001), and invasive behavior (OR: 1.41; 95% CI: 1.10-1.80; p = 0.007) were associated with increased odds of extended LOS. Male sex (OR: 0.83; 95% CI: 0.74-0.93; p = 0.001) and MIS approach (OR: 0.55; 95% CI: 0.36-0.80; p = 0.003) were associated with reduced odds. Robotic assistance did not significantly impact extended LOS (OR: 1.38; 95% CI: 0.61-3.01; p = 0.429). Gradient Boosting had the highest predictive performance among machine learning models (AUC: 0.594), followed by AdaBoost and logistic regression. SHAP analysis identified surgical approach, comorbidity score, tumor size, and behavior as the most influential features on extended LOS. MIS was associated with significantly lower odds of extended length of stay compared to open surgery for spinal nerve sheath tumor resection. Robotic assistance did not confer a significant additional benefit. These findings suggest that MIS may improve postoperative recovery and resource utilization in appropriately selected patients. Further prospective studies are needed to validate these results and clarify the role of MIS and robotic approaches in spinal tumor surgery.