Mendelian randomization analysis of labor anesthesia and adverse neonatal outcomes

Despite the lack of data from randomized controlled trials, studies have indicated that labor anesthesia may be associated with neonatal asphyxia, neonatal respiratory distress and adverse neonatal neurological outcomes. Therefore, we performed a two-sample Mendelian randomization analysis to explore the potential causal relationships between labor anesthesia methods and adverse neonatal outcomes.

We collected genome-wide association study (GWAS) data, including on spinal (n = 3,780), epidural (n = 3,970), and other labor anesthesia methods (n = 4,094), as well as neonatal asphyxia (n = 499,936), neonatal respiratory distress (NRDS) (n = 499,974) and cerebral palsy (n = 496,311), attention-deficit hyperactivity disorder (ADHD) (n = 495,160), and intellectual disability (n = 363,663). Data on different delivery analgesia methods that were sourced from the Integrative Epidemiology Unit (IEU) OpenGWAS project were used as exposure data. Neonatal asphyxia, neonatal respiratory distress and neurological adverse outcomes sourced from the FinnGen consortium R12 were used as the outcome data. A two-sample MR was used to evaluate the effects of different delivery analgesia methods on neonatal asphyxia, neonatal respiratory distress and three adverse neurological outcomes in newborns to determine the existence of a causal relationship between them. The inverse-variance weighted (IVW) method was used for MR analysis and a series of sensitivity analyses were conducted. The MR-Egger intercept test was used to assess directional horizontal pleiotropy. Heterogeneity was evaluated using the Cochran's Q statistic. Instrument strength was assessed using F-statistics, with values greater than 10 indicating a low risk of weak instrument bias.

Spinal, epidural, and other methods of labor anesthesia were not found to be strongly associated with neonatal asphyxia (OR = 0.707, 95% CI = 0.176-2.832, p = 0.624; OR = 3.222, 95% CI = 0.973-10.664, p = 0.055; OR = 0.732, 95% CI = 0.166-3.230, p = 0.681, respectively), NRDS (OR = 0.941, 95% CI = 0.381-2.321, p = 0.894; OR = 1.116, 95% CI = 0.505-2.465, p = 0.786; OR = 0.801, 95% CI = 0.329-1.950, p = 0.624), cerebral palsy (OR = 0.930, 95% CI = 0.442-1.959, p = 0.849; OR = 0.636, 95% CI = 0.318-1.271, p = 0.200; OR = 1.112, 95% CI = 0.544-2.271, p = 0.771, respectively), intellectual disability (OR = 1.586, 95% CI = 0.917-2.743, p = 0.099; OR = 0.809, 95% CI = 0.454-1.440, p = 0.471; OR = 0.774, 95% CI = 0.380-1.575, p = 0.479, respectively), or attention deficit hyperactivity disorder (OR = 0.827, 95% CI = 0.621-1.102, p = 0.195; OR = 0.998, 95% CI = 0.739-1.346, p = 0.988; OR = 1.136, 95% CI = 0.771-1.673, p = 0.519, respectively). The sensitivity analyses, performed with Cochran's Q test and the MR-Egger intercept, showed little evidence of substantial heterogeneity or directional horizontal pleiotropy.

Our MR study based on genetic data does not support the existence of a causal relationship between different labor anesthesia methods and neonatal asphyxia, neonatal respiratory distress or adverse neonatal neurological outcomes. Thus, labor pain relief methods can be selected based on the mother's needs and condition without increasing associated risks.