We wish to comment upon the recent paper by Lim et al. [1] which compared a computer integrated-patient controlled epidural analgesia (CI-PCEA) system with demand only patient controlled epidural analgesia (PCEA) for pain relief in labour. One of the main conclusions was that CI-PCEA improved maternal satisfaction despite there being no difference in local anaesthetic consumption.

To assess maternal satisfaction the authors used a 101-point scale ranging from 0 (very dissatisfied) to 100 (extremely satisfied). The 20 patients who received CI-PCEA reported a mean (standard deviation, SD) satisfaction score of 93 (7) compared with the 20 patients in PCEA group who reported a mean satisfaction of 86 (11), with a p-value of 0.042. As labour pain is often severe, analgesic interventions are likely to be associated with high maternal satisfaction scores. There are many components which influence maternal satisfaction with labour analgesia. Experience of labour pain, feeling of control, fears and expectations and pain and physical condition after delivery are all likely to be contributory [2]. A subjective measure such as maternal satisfaction will have great interindividual variation in reported scores. These factors mean that careful consideration is needed when designing qualitative studies of satisfaction [3, 4] both in terms of sample size and study design.

We feel that the use of a 101-point scale produced a statistically significant difference that is not clinically important. The large number of points on the scale improves statistical power, but does not make the measure any more valid. Furthermore, the original power calculation for the study estimated sample sizes to show differences in local anaesthetic consumption, not to compare maternal satisfaction, where there is likely to be a greater variance in reported ratings. We therefore feel that the authors' conclusion that CI-PCEA improves maternal satisfaction is not supported by the evidence given in the paper.

Measures of maternal satisfaction with neuraxial analgesia in other studies have ranged from the use of similar numerical scales with a varying number of points, to detailed questionnaires and semistructured interviews [2, 5–7]. Unfortunately, this makes the literature regarding the effect of neuraxial blockade during labour upon maternal satisfaction rather difficult to interpret. We suggest that a consensus be reached about the most valid method of assessing satisfaction following obstetric anaesthetic procedures.


  1. Top of page
  2. References