Swedish randomized controlled trial of cardiotocography only versus cardiotocography plus ST analysis of fetal electrocardiogram revisited: analysis of data according to standard versus modified intention-to-treat principle

Objective. To undertake a renewed analysis of data from the previously published Swedish randomized controlled trial on intrapartum fetal monitoring with cardiotocography (CTG-only) vs. CTG plus ST analysis of fetal electrocardiogram (CTG+ST), using current standards of intention-to-treat (ITT) analysis and to compare the results with those of the modified ITT (mITT) and per protocol analyses. Methods. Renewed extraction of data from the original database including all cases randomized according to primary case allocation (n=5 049). Main outcome measure. Metabolic acidosis in umbilical artery at birth (pH<7.05, base deficit in extracellular fluid>12.0mmol/l) including samples of umbilical vein blood or neonatal blood if umbilical artery blood was missing. Results. The metabolic acidosis rates were 0.66% (17 of 2 565) and 1.33% (33 of 2 484) in the CTG+ST and CTG-only groups, respectively [ relative risk (RR) 0.50; 95% confidence interval (CI) 0.28-0.88; p=0.019]. The original mITT gave RR0.47, 95% CI 0.25-0.86 (p=0.015), mITT with correction for 10 previously misclassified cases RR0.48, 95% CI 0.24-0.96 (p=0.038) and per protocol analysis RR 0.40, 95% CI 0.20-0.80 (p=0.009). The level of significance of the difference in metabolic acidosis rates between the two groups remained unchanged in all analyses. Conclusion. Re-analysis of data according to the ITT principle showed that regardless of the method of analysis, the Swedish randomized controlled trial maintained its ability to demonstrate a significant reduction in metabolic acidosis rate when using CTG+ST analysis for fetal surveillance in labor.