Deferred cord clamping (DCC) is beneficial for preterm infants, but there are concerns about the safety of DCC during Cesarean deliveries (CD) under general anesthesia (GA). We evaluated maternal and neonatal outcomes in preterm CD under GA vs. regional anesthesia (RA) after implementing 180 s of DCC. This retrospective single-center observational study included CD at < 33 weeks of gestation, delivered between January 2018 and December 2023. The cord was clamped before 180 s for concerns of maternal bleeding or infant apnea after 30-45 s stimulation. Multivariable regression analysis was used to assess the effect of anesthesia type and DCC on outcomes, adjusting for confounders. This study included 170 mothers and 194 infants, 84.9 % of the infants received DCC ≥ 60 s. The GA group had higher emergency CD and lower median duration of DCC (105 s vs 180 s, p = <0.001) compared to RA. GA was associated with lower odds (95 % CI) of UA pH < 7 [0.1, (0.0, 0.6)], base deficit ≥ 16 [0.0, (0.0, 0.5)], and higher odds of NEC [28.2, (1.4, 560.0)]. DCC ≥ 60 seconds was associated with lower maternal blood loss [Regression coefficient -698, (-1193, -202)], lower odds of transfusion [0.4, (0.1, 1.0)], DR resuscitation [0.4, (0.2, 0.8)], chronic lung disease [0.4, (0.2, 0.9)], and higher survival without major morbidities [2.8, (1.2, 6.8)]. DCC can be safely accomplished in majority of CD under GA with protocols to shorten DCC in cases where maternal or fetal safety is threatened. GA with DCC was not associated with increased neonatal resuscitation or major NICU morbidities and was associated with lower maternal hemorrhage and transfusion.

Materials and Methods

We conducted a single-center, retrospective study of all CD < 33 weeks gestation performed between January 2018 and December 2023. Institutional Review Board (IRB) approval was obtained (#24-007). Cases of lethal fetal anomalies were excluded.

Data on maternal demographics, pregnancy characteristics, delivery data, and maternal and infant outcomes were obtained from standardized reports in the electronic health record. Details on anesthesia timing, indication for CD, and indication for GA were confirmed with manual chart review. Anesthesia type was categorized as GA (if GA administration occurred before delivery of the infant) or RA, including spinal, epidural, or combined spinal epidural anesthesia. CD urgency was recorded by the delivering obstetrician at the time of surgery; “Emergency/STAT” was defined as need for immediate delivery due to impending fetal or maternal death or injury (e.g. fetal bradycardia.). Maternal blood loss was determined quantitatively: amniotic fluid quantity was marked on the suction cannister prior to placental delivery; all surgical drapes, chux, cannisters and lap pads were weighed at the conclusion of surgery and the weight of dry items and amniotic fluid subtracted. Antenatal steroid (ANS) for accelerating fetal maturation was administered prior to delivery of all infants < 34 weeks and Magnesium sulfate (4 gram bolus with or without ongoing infusion 1-2 gm/hr) for neuroprotection was administered prior to delivery of all infants ≤ 32 0/7 weeks whenever feasible. All CD were performed by in-house Obstetrics and Gynecology residents and their attending physicians. Anesthesia providers were available in house 24 hours a day.

Throughout the study period, our standardized institutional protocol for DCC recommended waiting at least 3 minutes before cord clamping. Contraindications to DCC included hydrops and recipient twin in twin-to-twin transfusion syndrome. Additional decisions to abort DCC early were based on discussion between the obstetric and neonatal providers in cases of significant maternal bleeding, cord avulsion, placental disruption or infant apnea after 30-60 seconds of tactile stimulation and/or airway suctioning.

Outcomes assessed for this study included maternal quantitative blood loss (QBL), PPH defined as blood loss > 1000 ml, and need for blood transfusion. Neonatal DR outcomes included umbilical artery (UA) and venous (UV) cord blood gas acidosis pH < 7 and base deficit (BD) ≥ 16, intubation, positive pressure ventilation (PPV), CPAP, and cardiopulmonary resuscitation (CPR). Neonatal NICU outcomes included admission temperature, severe IVH, late-onset sepsis, NEC, chronic lung disease (CLD), severe retinopathy of prematurity (ROP), severe neurologic injury including severe IVH and periventricular leukomalacia (PVL), length of stay, NICU death, and survival without major morbidities (late-onset sepsis, CLD, NEC, severe ROP, severe IVH or PVL).

Neonatal and maternal characteristics (Table 1 in associated manuscript) were compared between GA and RA groups using univariable generalized estimating equations (GEE), quantile regression, chi-squared tests, and Fisher’s exact tests, as appropriate. Average marginal effects were reported to quantify differences between groups. Neonatal and maternal outcomes (Table 2, 3) were analyzed using multivariable GEE models to evaluate the effect of anesthesia type and DCC ≥ 60 s. Models were adjusted for known or suspected confounders, including gestational age, antenatal steroid (ANS), and emergent CD. We report effect estimates as odds ratios and regression coefficients. ANS was dropped from several multivariable analyses due to lack of events. Umbilical venous pH < 7 and base deficit ≥ 16 were instead analyzed using a Fisher’s exact test due to sparsity of events. 

All infant-based regression models were clustered around pregnancy to account for multiple births. Robust standard errors were applied to all regression models. Linear, logistic, negative binomial, and Poisson models were selected based on the outcome and residual distributions. Descriptive statistics are presented as mean ± SD for normally distributed data, median (IQR) for non-normally distributed data, and n (%) for categorical variables. A p-value <0.05 was considered statistically significant.