Author’s Reply

  1. Top of page
  2. Author’s Reply
  3. References


Foley considers that caesarean section rate is the correct measure of efficacy of any method of induction.1 Although it might more accurately be considered a measure of ineffectiveness, its usefulness is limited by the fact that it is heavily influenced by safety concerns, which may or may not be related to the method of induction, including changes in fetal and maternal wellbeing, as well as obstetrician preference. Our study was of open-label design and vaginal delivery within 24 hours was selected as the primary outcome as it is less liable to bias.2 Vaginal delivery within 24 hours and caesarean section rates are not interchangeable and we observed no consistent relationship between these parameters in our study; that is, those sites achieving higher 24-hour vaginal deliveries did not necessarily have lower caesarean section rates. It is therefore valid that caesarean section rates are considered separately from vaginal delivery rates.

Foley extrapolates the caesarean section rates for nullipara and multipara at the National Maternity Hospital to our study.1 This approach should be treated with some caution as caesarean section rates are well recognised to vary quite markedly between hospitals, let alone between countries. For example, while data for English hospitals in 2006–2007 show an overall caesarean section rate of 24.3%, there is a wide range of 12–33%.3

Our results cannot be expressed for multipara with and without a uterine scar as previous uterine surgery, including a previous caesarean section, was an exclusion criterion in the study. Results for women without a uterine scar can be broken down as shown in Table 1.

Table 1.  Effect of parity on 24-hour vaginal delivery and caesarean section rates
 Misoprostol (%)Dinoprostone (%)Difference between treatments
Vaginal delivery within 24 hours
Nullipara (n= 358)2931Espectives 1.81%; 95% CI (−7.74, 11.35)
Multipara (n= 265)6369Espectives 6.49%; 95% CI (−4.91, 17.89)
Caesarean section
Nullipara (n= 358)4232P= 0.06
Multipara (n= 265)118P= 0.41

The differences in rates for vaginal delivery within 24 hours and for caesarean section between nullipara and multipara in both the misoprostol and dinoprostone groups reflects current clinical knowledge, that caesarean section rates are much higher in nullipara. Indeed, our results are comparable with those of earlier studies, notably Gregson et al.4 (Table 2) and van Gemund et al.5 (Table 3), in which caesarean section rates were much higher for nullipara for both misoprostol and dinoprostone. Both studies concluded that misoprostol was a safe and effective method of labour induction.

Table 2.  Effect of parity on caesarean section rates from Gregson et al.4
 Misoprostol (%)Dinoprostone (%) Relative risk (95% CI)
Caesarean section
Nullipara (n= 152)42391.06 (0.72–1.57)
Multipara (n= 115)350.67 (0.12–3.84)
Table 3.  Effect of parity on caesarean section rate from van Gemund et al.5
 Misoprostol (%)Dinoprostone (%)Relative risk (95% CI)
Caesarean section
Nullipara (n= 400)20270.7 (0.5–1.1)
Multipara (n= 281)9110.8 (0.4–1.5)

A post hoc, covariate analysis of the caesarean deliveries in our study has suggested that there were more women with risk factors for caesarean delivery in the misoprostol group than in the dinoprostone group. Taking this analysis into account, the caesarean section rates for both nullipara and multipara are comparable to those seen in routine clinical practice, figures from 2000 for England and Wales have caesarean section rates following induced labour for nullipara as 27.9% and for multipara as 8.5%.6

We repeat the assertion that 25 mcg misoprostol is both safe and effective for the induction of labour.


  1. Top of page
  2. Author’s Reply
  3. References