Puncturing the dura: a true clinical benefit or a distraction? A reply

We appreciate the opportunity to respond to the comments by Fung and Preston [1] about our study comparing the quality of labour analgesia following initiation with a combined spinal epidural (CSE) vs. a dural puncture epidural (DPE) technique [2].

Their first comment relates to the sample size calculation for the study. At the time of designing our protocol, the only study comparing the two techniques was a well-conducted randomised controlled trial by Chau et al. that reported a reduction in the need for top up interventions from 50% in patients allocated to the CSE group to 22.5% in those allocated to DPE [3]. Chau et al. had plausible explanations for this large effect and suggested that it could be related to the need for more top-ups during the transition from spinal to epidural analgesia or increased uterine contractions following CSE resulting in more analgesic needs. Those numbers were, therefore, used for our power analysis, which was based on existing valid literature and not done “arbitrarily” or “to calculate the minimum required sample size” as suggested by Fung and Preston. While we did not find such a large effect size, it is possible that our study was underpowered to detect smaller effect sizes that could be considered clinically relevant by some. Based on our findings, a future study with a sample size of 976 patients would have 80% power to detect a statistically significant difference in our primary outcome at α = 0.05.

The second comment relates to the choice of primary and secondary outcomes focusing on anaesthetists' workload and lack of patient-reported outcomes. There are currently no validated tools for assessing patient-reported outcome measures associated with labour analgesia. In the absence of such tools, we tried to capture outcomes that are important to patients and providers and that are in line with previously published studies investigating neuraxial labour analgesia. It could be argued that the more interventions that are needed reflect inadequate analgesia negatively impacting the desired pain relief by the mothers, which was recently reported to be the highest outcome preference by patients regarding labour epidural analgesia [4]. We agree that the mode of delivery is an important outcome (which we report in our table S2 [2]) but seems to have been missed by Fung and Preston. We also agree on the need for developing validated tools for capturing patient-reported outcomes associated with labour analgesia, as well as develop recommendations for a set of core measures to be included in labour analgesia studies.

The third and final comment from Fung and Preston points to the need for efforts to address disparities in access to labour analgesia. Similar to what has been reported in Scotland [5], a recent study using the 2017 natality data from the USA reported lower neuraxial labour analgesia use in non-Hispanic Black patients (75%) compared with non-Hispanic White patients (79%) [6]. We, therefore, wholeheartedly agree with Fung and Preston on the urgent need to address those disparities. At the same time, with a neuraxial labour analgesia rate of 78% in the USA [6], we argue that it is very relevant to continue fine tuning and improving an intervention received by the majority of women in labour in our practice.