Effectiveness of fetal scalp blood sampling for the prevention of cesarean section for suspected fetal distress during labor
Fetal blood sampling (FBS) is applied as additional test to cardiotocography (CTG), mainly to avoid unnecessary operative delivery in case of suspected fetal distress. It is potentially effective in reducing the intervention rate. However, the method is invasive and data on its cost-effectiveness are limited. This hampers the implementation of consistent use of FBS. The primary aim of this study is to determine if FBS is cost effective in preventing CS in case of suspected fetal distress during the first stage of labor. The secondary aim is to determine to what extent FBS affects perinatal outcome.
A multicenter RCT embedded in a cohort with informed consent during pregnancy and randomization in case of an abnormal CTG during labor (FIGO/STAN criteria).
Women with a singleton pregnancy of 36 weeks or over with an indication for continuous CTG monitoring are included in the cohort. When fetal distress is suspected during the first stage of labor women are randomized for either FBS or no-FBS.
In case of suspected fetal distress, randomization for decision making using FBS (intervention) or decision making using CTG only (control).
The primary outcome of the study is the CS rate for any indication. Secondary outcomes are a composite of poor perinatal outcome, maternal complications and womens birth experiences. A cost-effectiveness analysis of the two strategies, FBS or no FBS, will be performed. For the budget impact analysis, we will report on the estimated change in annual costs if consistent FBS, or delivery management on the basis of CTG only becomes the strategy of first choice in the Netherlands.
In the Netherlands, two types of fetal monitoring are used; CTG with and without ST-analysis of the fetal electrocardiogram (STAN). In order to show superiority of a strategy with FBS:
In CTG WITHOUT STAN, 2 groups of 140 women are needed. This number is sufficient to show a reduction in CS rate of at least 15% (from 80% to 65%)(two-sided test; alpha 0.05, beta 0.20).
In CTG WITH STAN, 2 groups of 60 women are needed to show a similar reduction in CS rate (from 95% to 80%) (one-sided test; alpha 0.05, beta 0.20).
The 400 women in the RCT will also be used to compare poor perinatal outcome, because if a strategy with FBS reduces the CS rate, such a strategy can only be implemented when it does not compromise perinatal outcome. To calculate the upper limit of the 95% confidence interval (non-inferiority, alpha 0.05, beta 0.20) we assumed a 5% incidence of composite poor perinatal outcome.
Analysis will follow the intention to treat principle. Treatment effect will be described as relative risk with 95% confidence intervals.
Prof. Dr. F. Vandenbussche, UMC St Radboud, Nijmegen.
Prof. Dr. F.K. Lotgering, UMC St Radboud, Nijmegen.