Background: Continuous epidural infusion (CEI) is commonly used for labour analgesia, but concerns over potential motor block, second-stage labour complications, and ineffective analgesia in late labour have prompted examining intermittent epidural bolus (IEB) as an alternative. However, evidence comparing these modalities is conflicting. The meta-analysis evaluates the analgesic efficacy of CEI vs IEB. Methods: Databases were searched for trials comparing CEI to IEB for labour analgesia. The two co-primary outcomes were risk of breakthrough pain and difference in area under the curve (AUC) for pain scores during the first 4 h postepidural initiation. Local anaesthetic consumption, maternal outcomes (i.e. delivery mode, labour duration, and maternal satisfaction), and side-effects of epidural analgesia were also evaluated. Results were pooled using randomeffects modelling. Trial sequential analysis (TSA) was used to evaluate evidence reliability. Results: Twenty-seven studies (3133 patients) were analysed. Compared with CEI, IEB decreased risk of breakthrough pain by 38% (risk ratio [95% confidence interval {CI}] of 0.62 [0.48, 0.81]; P¼0.0004; I 2 ¼47%; 1164 patients) and reduced AUC of pain during the 4 h interval by 32.9% (mean difference [95% CI] of e16.7 mm h À1 [e18.9, e14.4]; P<0.0001; 1638 patients). Intermittent epidural bolus enhanced maternal satisfaction, shortened labour duration, decreased motor block, and reduced local anaesthetic consumption. The difference between the two groups was not statistically significant for epidural side-effects or mode of delivery. The TSA indicated adequate power for reliable inferences. Conclusions: Intermittent epidural bolus provides improved labour pain control during the first 4 h after epidural initiation with less breakthrough pain. Moderate-to high-quality evidence of intermittent epidural bolus superiority support its use as a safe and effective continuous epidural infusion alternative for labour analgesia.

BMI, body mass index; ASA, American Society of Anesthesiologists; NA, not applicable. Data are mean6SD, n (%), or median (range) unless otherwise specified. * Two patients were missing BMI. † Administered uterotonics included methylergonovine, carboprost, and misoprostol.