RESUMO
BACKGROUND: The incidence of failed spinal anesthesia varies widely in the obstetric literature. Although many risk factors have been suggested, their relative predictive value is unknown. The primary objective of this retrospective cohort study was to determine the incidence of failed spinal anesthesia for cesarean deliveries at a tertiary care obstetric hospital, and its secondary objectives were to identify predictors of failed spinal anesthesia in the obstetrics population and quantify their relative importance in a predictive model for failure. METHODS: With local institutional ethics committee approval, a retrospective review of our hospital database identified the incidence of failed spinal anesthesia for 5361 cesarean deliveries between 2010 and 2019. We performed a multivariable analysis to assess the association of predictors with failure and a dominance analysis to assess the importance of each predictor. RESULTS: The incidence of failed spinal anesthesia requiring an alternative anesthetic was 2.1%, with conversion to general anesthesia occurring in 0.7% of surgeries. Supplemental analgesia or sedation was provided to an additional 2.0% of women. The most important predictors of a failed spinal anesthetic were previous cesarean delivery (odds ratio [OR], 11.33; 95% confidence interval [CI], 7.09-18.20; P < .001), concomitant tubal ligation (OR, 8.23; 95% CI, 3.12-19.20; P < .001), lower body mass index (BMI) (kg·m -2 , OR, 0.94; 95% CI, 0.90-0.98; P = .005), and longer surgery duration (minutes, OR, 1.02; 95% CI, 1.01-1.03; P = .006). Previous cesarean delivery was the most significant risk factor, contributing to 9.6% of the total 17% variance predicted by all predictors examined. CONCLUSIONS: Spinal anesthesia failed to provide a pain-free surgery in 4.1% of our cesarean deliveries. Previous cesarean delivery was the most important predictor of spinal failure. Other important predictors included tubal ligation, lower BMI, and longer surgery duration.