Background

Epidural analgesia is often used for pain relief during labour and childbirth, and involves administration of local anaesthetics (LA) into the epidural space resulting in sensory blockade of the abdomen, pelvis, and perineum. Epidural opioids are often co‐administered to improve analgesia. Administration of epidural medications can be accomplished by basal infusion (BI) or automated mandatory bolus (AMB). With BI, medications are administered continuously, while AMB involves injecting medications at set time intervals. Patient‐controlled epidural analgesia (PCEA) on top of AMB or BI enables patients to initiate additional boluses of epidural medications. 

The superior method of delivering epidural medications would result in lower incidence of pain requiring anaesthesiologist intervention (breakthrough pain). Also, it should be associated with lower incidence of epidural‐related adverse effects including caesarean delivery, instrumental delivery (use of forceps or vacuum devices), prolonged duration of labour analgesia, and LA consumption. However, clear evidence of the superiority of one technique over the other is lacking. Also, differences in the initiation of epidural analgesia such as combined spinal‐epidural (CSE) (medications given into the intrathecal space in addition to the epidural space) compared to epidural only, and medications used (types and doses of LA or opioids) may not have been accounted for in previous reviews. 

Our prior systematic review suggested that AMB reduces the incidence of breakthrough pain compared to BI with no significant difference in the incidence of caesarean delivery or instrumental delivery, duration of labour analgesia, and LA consumption. However, several studies comparing AMB and BI have been performed since then, and inclusion of their data may improve the precision of our effect estimates.

Objectives

To assess the benefits and harms of AMB versus BI for maintaining labour epidural analgesia in women at term.

Search methods

We searched CENTRAL, Wiley Cochrane Library), MEDLINE, (National Library of Medicine), Embase(Elseiver), Web of Science (Clarivate), the WHO‐ICTRP (World Health Organization) and ClinicalTrials.gov (National Library of Medicine) on 31 December 2022. Additionally, we screened the reference lists of relevant trials and reviews for eligible citations, and we contacted authors of included studies to identify unpublished research and ongoing trials.

Selection criteria

We included all randomised controlled studies that compared bolus dosing AMB with continuous BI during epidural analgesia. We excluded studies of women in preterm labour, with multiple pregnancies, with fetal malposition, intrathecal catheters, those that did not use automated delivery of medications, and those where AMB and BI were combined.

Data collection and analysis

We used standard methodology for systematic review and meta‐analysis described by Cochrane. Primary outcomes included: incidence of breakthrough pain requiring anaesthesiologist intervention; incidence of caesarean delivery; and incidence of instrumental delivery. Secondly, we assessed the duration of labour; hourly LA consumption in bupivacaine equivalents, maternal satisfaction after fetal delivery, and neonatal Apgar scores. 

The following subgroup analyses were chosen a priori: epidural alone versus CSE technique; regimens that used PCEA versus those that did not; and nulliparous versus combination of nulli‐ and multi‐parous women.

We used the GRADE system to assess the certainty of evidence associated with our outcome measures.

Main results

We included 18 studies of 4590 women, of which 13 enrolled healthy nulliparous women and five included healthy nulli‐ and multiparous women. All studies excluded women with preterm or complicated pregnancies. Techniques used to initiate epidural analgesia differed between the studies: seven used combined spinal epidural, 10 used epidural, and one used dural puncture epidural (DPE). There was also variation in analgesics used. Eight studies utilised ropivacaine with fentanyl, three used ropivacaine with sufentanil, two utilised levobupivacaine with sufentanil, one used levobupivacaine with fentanyl, and four utilised bupivacaine with fentanyl. Most of the studies were assessed to have low risk of randomisation, blinding, attrition, and reporting biases, except for allocation concealment where eight studies were assessed to have uncertain risk and three with high risk.

Our results showed that AMB was associated with lower incidence of breakthrough pain compared to BI (risk ratio (RR) 0.71; 95% confidence interval (CI) 0.55 to 0.91; I² = 57%) (16 studies, 1528 participants), and lower hourly LA consumption in bupivacaine equivalents (mean difference (MD) ‐0.84 mg/h; 95% CI ‐1.29 to ‐0.38, I² = 87%) (16 studies, 1642 participants), both with moderate certainty. AMB was associated with an estimated reduction in breakthrough pain incidence of 29.1% (incidence 202 per 1000, 95% CI 157 to 259), and was therefore considered clinically significant.

The incidence of caesarean delivery (RR 0.85; 95% CI 0.69 to 1.06; I² = 0%) (16 studies, 1735 participants) and instrumental delivery (RR 0.85; 95% CI 0.71 to 1.01; I² = 0%) (17 studies, 4550 participants) were not significantly, both with moderate certainty. There was no significant difference in duration of labour analgesia (MD ‐8.81 min; 95% CI ‐19.38 to 1.77; I² = 50%) (17 studies, 4544 participants) with moderate certainty. Due to differences in the methods and timing of outcome measurements, we did not pool data for maternal satisfaction and Apgar scores. Results reported narratively suggest AMB may be associated with increased maternal satisfaction (eight studies reported increased satisfaction and six reported no difference), and all studies showed no difference in Apgar scores.

WIth the exception of epidural alone versus CSE which found significant subgroup differences in LA consumption between AMB and BI, no significant differences were detected in the remaining subgroup analyses.

Authors' conclusions

Overall, AMB is associated with lower incidence of breakthrough pain, reduced LA consumption, and may improve maternal satisfaction. There were no significant differences between AMB and BI in the incidence of caesarean delivery, instrumental delivery, duration of labour analgesia, and Apgar scores. Larger studies assessing the incidence of caesarean and instrumental delivery are required.