Predictors of operative time and length of stay in robot-assisted transforaminal lumbar interbody fusions.

Abstract

Prolonged operative time and length of stay (LOS) are associated with increased complications and costs after lumbar fusions. However, with the application of robotic technology to transforaminal lumbar interbody fusions (TLIFs), operative times and LOSs have shortened. Accurate predictions of these metrics are crucial to improving operating room and hospital resource allocation. The objective of this study was to identify predictors of operative time and LOS in robot-assisted TLIF procedures. A single-institution retrospective review was performed of patients who underwent pedicle screw placement using the same robotic navigation platform. Patients who underwent TLIF for degenerative spine indications were included. Preoperative demographics, neurological deficits, and comorbidities were extracted. Operative time and LOS outcome variables were categorized as normal or extended (> 75th percentile). Univariable analysis was performed using the Kruskal-Wallis test, and significant variables (p < 0.10) were included in the multivariable analysis. All p values < 0.05 were considered statistically significant. One hundred seventeen patients were included, who were predominantly female (n = 67, 57%) and had a median age of 64 years. In univariable analysis, revision status (p = 0.063), BMI (p = 0.029), spinal level (p = 0.012), incision type (p = 0.013), number of levels (p = 0.001), and number of screws (p = 0.001) were associated with operative time, while Charlson Comorbidity Index (CCI; p = 0.003), American Society of Anesthesiologists class (p = 0.029), and Frankel grade (p = 0.042) were associated with LOS. Multivariable analysis showed that BMI ≥ 30 kg/m2 (odds ratio [OR] 1.163, 95% confidence interval [CI] 1.024-1.322) and number of levels > 2 (OR 1.655, 95% CI 1.401-1.956) were independently associated with operative time. Frankel grade < E (OR 1.146, 95% CI 1.023-1.283) and CCI > 2 (OR 1.285, 95% CI 1.112-1.486) were independently associated with LOS. This study shows significant predictors of extended surgical operative time and LOS in robot-assisted TLIF procedures. These findings can help with surgical planning and operating room allocation.