OBJECTIVEHealthcare costs continue to escalate. Approaches to care that have comparable outcomes and complications are increasingly assessed for quality improvement and, when possible, cost containment. Efforts to identify components of care to reduce length of stay (LOS) have been ongoing. Spinal anesthesia (SA), for select lumbar spine procedures, has garnered interest as an alternative to general anesthesia (GA) that might reduce cost and in-hospital LOS and accelerate recovery. While clinical outcomes with SA or GA have been studied extensively, few authors have looked at the cost-analysis in relation to clinical outcomes. The authors’ objectives were to compare the clinical perioperative outcomes of patients who received SA and GA, as well as the direct costs associated with each modality of care, and to determine which, in a retrospective analysis, can serve as a dominant procedural approach.METHODSThe authors retrospectively analyzed a homogeneous surgical population of 550 patients who underwent hemilaminotomy for disc herniation and who received either SA (n = 91) or GA (n = 459). All clinical and billing data were obtained via each patient’s chart and the hospital’s billing database, respectively. Additionally, the authors prospectively assessed patient-reported outcome measures for a subgroup of consecutively treated patients (n = 75) and compared quality-adjusted life year (QALY) gains between the two cohorts. Furthermore, the authors performed a propensity score–matching analysis to compare the two cohorts (n = 180).RESULTSDirect hospital costs for patients receiving SA were 40% higher, in the hundreds of dollars, than for patients who received GA (p < 0.0001). Furthermore, there was a significant difference with regard to LOS (p < 0.0001), where patients receiving SA had a considerably longer hospital LOS (27.6% increase in hours). Patients undergoing SA had more comorbidities (p = 0.0053), specifically diabetes and hypertension. However, metrics of complications, including readmission (p = 0.3038) and emergency department (ED) visits at 30 days (p = 1.0), were no different. Furthermore, in a small pilot group, QALY gains were statistically no different (n = 75, p = 0.6708). Propensity score–matching analysis demonstrated similar results as the univariate analysis: there was no difference between the cohorts regarding 30-day readmission (p = 1.0000); ED within 30 days could not be analyzed as there were no patients in the SA group; and total direct costs and LOS were significantly different between the two cohorts (p < 0.0001 and p = 0.0126, respectively).CONCLUSIONSBoth SA and GA exhibit the qualities of a good anesthetic, and the utilization of these modalities for lumbar spine surgery is safe and effective. However, this work suggests that SA is associated with increased LOS and higher direct costs, although these differences may not be clinically or fiscally meaningful.
CORRIGENDUM: Correction of IRB number: Comparison of clinical outcomes according to the history of previous gastrectomy in patients undergoing pancreatoduodenectomy: a propensity score matching analysis
