Predictors of General Anesthesia Receipt During Cesarean Deliveries in the U.S. Military Health System: A...

Clinical practice guidelines indicate general anesthesia should be avoided for cesarean deliveries, when possible. In the U.S. Military Health System, understanding multifaceted predictors of general anesthesia for cesarean deliveries can be used to enhance value-based obstetric care efforts.

This retrospective observational cohort study included adult patients who experienced a cesarean delivery in a military treatment facility between January 2018 and October 2024. Potential covariates included patient characteristics, diagnoses, and pregnancy care information. Pre-delivery general anesthesia receipt was evaluated using a generalized additive mixed model with a random effect for facility. Sensitivity analyses repeated the model with (a) a sample restricted to patients with gestation ages >35 weeks and (b) a multinomial outcome comparing neuraxial only to general only and neuraxial-to-general conversion.

Of patients with available anesthesia information (N = 15,181), the general anesthesia rate was 4.8%. After applying additional exclusion criteria for the primary analysis, the rate was 4.4% in the final sample (N = 11,921); of whom, 51% had also received neuraxial anesthesia. There were 34 unique facilities represented in the data. In the primary model, the odds of general anesthesia receipt was higher for Black compared to White patients (OR 1.31, 95% CI, 1.01-1.69, P = .04), those who were not military-connected compared to military-connected (OR 1.29, 95% CI, 1.05-1.60, P = .02), patients with fetal abnormalities and stress diagnoses compared to those without (OR 2.61, 95% CI, 2.15-3.16, P < .001), those with umbilical cord complications diagnoses compared to those without (OR 1.30, 95% CI, 1.04-1.64, P = .02), and patients with other labor and delivery complications diagnoses compared to those without (OR 1.74, 95% CI, 1.38-2.19, P < .001). General anesthesia odds were lower for patients whose health plan sponsors were senior enlisted service members (OR 0.73, 95% CI, 0.59-0.91, P = .01) and officers (OR 0.73, 95% CI, 0.53-0.999, P = .049) compared to junior enlisted service members, as well as patients who received care justifying non-elective delivery compared to those who did not (OR 0.62, 95% CI, 0.50-0.78, P < .001). Lastly, there were significant, nonlinear relationships between gestational age (P < .001) and direct care percent (P = .01) with general anesthesia receipt. The estimated probability (95% CI) of general anesthesia was highest at 30 weeks' gestation (17%, 11%-23%), which decreased through 41 weeks' gestation (4%, 3%-6%), before increasing slightly at 42 weeks (5%, 3%-7%). As direct care percent increased, the probability of general anesthesia decreased, for example patients receiving 25% (8%, 5%-10%) and 50% (6%, 5%-7%) of pre-delivery care in the direct care system had higher probabilities than those receiving 75% (5%, 4%-5%) and 100% (4%, 4%-5%). Lastly, there was a significant random effect for facilities, indicating probabilities were different across delivery locations (P = .003). Sensitivity models had similar findings, with nuanced differences in the multinomial model across general anesthesia groups.

The present findings could be used to value-based obstetric care efforts around regional anesthesia and cesarean delivery outcomes in the U.S. Military Health System.