Predicting extended operative time and length of inpatient stay in cervical deformity corrective surgery

It's increasingly common for surgeons to operate on more challenging cases and higher risk patients, resulting in longer op-time and inpatient LOS. Factors predicting extended op-time and LOS for cervical deformity (CD) patients are understudied. This study identified predictors of extended op-time and length of stay (LOS) after CD-corrective surgery. CD patients with baseline (BL) radiographic data were included. Patients were stratified by extended LOS (ELOS: >75th percentile) and normal LOS (N-LOS; <75th percentile). Op-time analysis excluded staged cases, cases >12 h. A Conditional Variable Importance Table used non-replacement sampling set of Conditional Inference trees to identify influential factors. Mean comparison tests compared LOS and op-time for top factors. 142 surgical CD patients (61 yrs, 62%F, 8.2 levels fused). Op-time and LOS were 358 min and 7.2 days; 30% of patients experienced E-LOS (14 +/- 13 days). Overlapping predictors of E-LOS and op-time included levels fused (>7 increased LOS 2.7 days; >5 increased op-time 96 min, P < 0.001), approach (anterior reduced LOS 3.0 days; combined increased op-time 69 min, P < 0.01), BMI (>38 kg/m(2) increased LOS 8.1 days; >39 kg/m(2) increased op-time 17 min), and osteotomy (LOS 2.0 days, op-time 62 min, P < 0.005). BL cervical parameters increased LOS and op-time: cSVA (>42 mm increased LOS; >50 mm increased op-time, P < 0.030), CO slope (>@-0.9 degrees increased LOS, >0.3 degrees increased op-time, P < 0.003.) Additional op-time predictors: prior cervical surgery (p = 0.004) and comorbidities (P = 0.015). Other predictors of E-LOS: EBL (P < 0.001), change in mental status (P = 0.001). Baseline cervical malalignment, levels fused, and osteotomy predicted both increased op-time and LOS. These results can be used to better optimize patient care, hospital efficiency, and resource allocation. (C) 2019 Elsevier Ltd. All rights reserved.