In Reply:

In Reply: We thank Drs. Reichman and Grisaru-Granovsky for their thoughtful comments regarding our population-based analysis of contemporary outcomes in term breech birth.1 Their remarks highlight essential considerations in interpreting real-world data on this longstanding obstetric question. Our study was designed to complement, rather than replicate, the Term Breech Trial2 by using a large U.S. vital records cohort and rigorous propensity score matching with doubly robust estimation to approximate comparable delivery groups under current practice conditions. Our primary objective was to inform bedside counseling in the setting clinicians most commonly encounter in the United States: a patient with a singleton, nonanomalous, term breech fetus presenting for care in a system in which most breech births are managed by cesarean delivery, with some delivering vaginally, often in labor or advanced labor. The authors question that grouping women who underwent emergent cesarean delivery after trial of labor with those who had planned elective cesarean delivery may mask important associations. We agree and acknowledge that our analysis was based on the actual mode of delivery rather than the intended mode, because an intention-to-treat approach cannot be implemented in vital records data. This pragmatic approach remains clinically relevant, reflecting real-world scenarios in which management decisions often are made after labor onset. The inclusion of a small proportion of cesarean deliveries performed after trial or induction of labor would bias our results toward the null rather than inflating between-group differences. Despite this conservative bias, we observed higher risks of adverse neonatal outcomes associated with vaginal breech births (composite: 7.2% vs 6.3%; neonatal mortality: 0.4% vs 0.1%). This finding is consistent with that reported in the Term Breech Trial; however, it is smaller in magnitude. The authors correctly note that, within the matched cohort, 21.3% of patients who ultimately had cesarean delivery underwent induction of labor and 19.4% had a documented trial of labor (Table 1).1 We agree that these figures are the relevant ones for the matched analysis, and we appreciate the correction in our interpretation. We agree that parity is a salient determinant of both maternal and neonatal outcomes. In this analysis, parity was incorporated as a key covariate in both the propensity score and outcome models, achieving excellent balance after matching (eg, in the propensity score model, nulliparous: 30.7% vs 30.2%; two or more prior deliveries: 42.3% vs 42.9%; Table 1), meeting our prespecified balance criterion (standardized mean differences less than 0.1).1 However, stratifying all outcomes by parity would have produced unstable estimates due to the infrequent occurrence of these events. An exploration of parity-specific effects may be worthy of future consideration. We concur that every neonatal death is clinically meaningful. Our use of the term low absolute risk reflects the rarity of events in both groups and, in comparison with the Term Breech Trial, not a minimization of their significance. In our analysis, short-term maternal morbidity excluding perineal lacerations was lower with vaginal breech birth (0.3% versus 0.7%), a difference that coexists with the neonatal findings. These population-level estimates should inform balanced counseling that considers both neonatal and maternal risks, including downstream maternal risks associated with current and future cesarean deliveries. We share the correspondents' commitment to refining evidence on breech delivery. Further research incorporating planned intent, parity-stratified analyses, and clinician-experience metrics will enhance shared decision making and ensure that patients with term breech presentations have safe and informed options.