Management of post-term pregnancies. Systematic review
Systematic review
|Published
We have carried out a systematic review of four randomised tri-als to examine the effects of inducting labour at or beyond 41 weeks (+ 0-2 days) of gestation compared with monitoring up to 42 weeks (+ 0-2 days).
Key message
There is some disagreement among health professionals on the management of post-term pregnancies. The WHO recommends offering induction of labour at or beyond 41 gestational weeks.
We have carried out a systematic review of four randomised trials to examine the effects of inducting labour at or beyond 41 weeks (+ 0-2 days) of gestation compared with monitoring up to 42 weeks (+ 0-2 days).
Low to moderate quality evidence indicates that induction at 41 weeks compared to waiting and monitoring until 42 weeks:
- probably reduces stillbirths, perinatal deaths and admissions in newborn intensive care units
- probably slightly reduces the incidence of Apgar score <7 after 5 minutes
- may give little or no difference with regard to the incidence of asphyxia
- probably makes no difference with regard to Caesarean delivery, operative vaginal delivery, or postpartum haemorrhage
- may slightly reduce the incidence of grade 3 or 4 perineal injury
We judged most results to be of moderate certainty.
Induction at 41 weeks appears to reduce the risk of complications in newborns, but does not appear to have significant implications for maternal outcomes.
Summary
Background
Post-term pregnancy (42 gestational weeks) is associated with an increased risk of birth complications and perinatal death. Induction of labour is a method to prevent post-term pregnancy and associated complications.
There is no scientific consensus with regard to the timing of induction in low-risk pregnancies. Should labour be induced at 41 weeks, or at 42 weeks if spontaneous labour has not occurred? The World Health Organization recommends induction at 41 weeks, while the Norwegian Directorate of Health recommends to induce labour by day 294".
In the period 2018-2020, there have been four systematic reviews published on the effects of induction versus waiting until 42 weeks. The most comprehensive, most updated and methodologically strongest of these reviews is a Cochrane review published in July 2020. However, as the Cochrane review does not include subgroup analyses limited to induction at 41 weeks compared to monitoring up to 42 weeks, we have chosen to conduct a review within these limits in our response to the Directorate of Health’s commission.
Objective
The objective of this systematic review was to examine the effects of induction at 41 gestational weeks (+ 0-2 days) compared with waiting until 42 weeks (+ 0-2 days).
Method
We have conducted a systematic review according to the methods described in the National Institute of Public Health's handbook for research synthesis and in a peer-reviewed protocol.
We included randomised studies that examined the effect of induction at 41 weeks (+ 0-2 days) compared with waiting and monitoring until 42 weeks (+ 0-2 days). Study participants were women with low risk pregnancies at full term up to 41 gestational weeks and with term determined by ultrasound. Our primary outcomes were stillbirth and perinatal death. Secondary outcomes were complications in mother and child. The studies should be conducted in a middle or high-income country.
We based our literature retrieval on the Cochrane review from 2020. Further, we largely relied on the Cochrane review authors' data retrieval, risk of bias assessments and methods for data synthesis. That is, calculation of relative risk (RR) with 95% confidence intervals (CI) for dichotomous outcomes (eg death), and mean difference (MD) with 95% CI for continuous outcomes (eg satisfaction).
We made our own judgements of the certainty of the evidence using the GRADE approach.
The number of included studies was too low to perform planned subgroup analyses.
Results
We found and included four randomised studies from high and middle income countries that examined the effects of induction at 41 weeks (+ 0-2 days) compared with waiting and monitoring until 42 weeks (+ 0-2 days). The studies were conducted in Norway, Sweden, the Netherlands and Turkey, respectively, in the period 2002 to 2018 and included a total of 5,669 participating women.
The effect estimates from the meta-analyses showed that induction at 41 weeks (+ 0-2 days) probably reduces stillbirths and perinatal deaths compared to waiting and monitoring until 42 weeks (+ 0-2 days) (moderate certainty of the evidence). Mortality was low (0.03% and 0.35%, respectively) in both groups.
There are probably fewer admissions in newborn intensive care units among children born after induction at 41 weeks than after waiting and monitoring (moderate certainty of the evidence). Induction at 41 weeks probably results in a small reduction in the proportion of newborns with Apgar score <7 after 5 minutes (moderate certainty of the evidence). There may be little or no difference between the interventions with regard to the incidence of asphyxia (low certainty of the evidence).
There is probably little or no difference between induction and waiting with regard to caesarean delivery (moderate certainty of the evidence), operative vaginal delivery (moderate certainty of the evidence), or postpartum haemorrhage (moderate certainty of the evidence). Induction at 41 weeks may result in a small reduction in the proportion of women with grade 3 or 4 perineal injury (low certainty of the evidence).
None of the women in the studies died, and none had uterine rupture.
In one study that examined patient satisfaction, most women preferred induction at 41 weeks.
| Outcomes | Anticipated absolute effects (95% CI) | Relative effect (95% CI) |
№ of participants (studies) |
Certainty of the evidence (GRADE) |
Comments | |
| Risk with monitoring | Risk with induction | |||||
| Stillbirth | 3 per 1 000 | 1 per 1 000 (0 to 4) |
RR 0,27 (0,05 to 1,33) |
5668 (4 studies) |
⨁⨁⨁◯ |
|
| Perinatal death | 4 per 1 000 | 1 per 1 000 (0 to 4) |
RR 0,27 (0,06 to 1,12) |
5668 (4 studies) |
⨁⨁⨁◯ |
|
| NICU admission |
44 per 1 000 |
30 per 1 000 (23 to 40) |
RR 0,69 |
5661 (4 studies) |
⨁⨁⨁◯ |
|
| Apgar < 7 | 16 per 1 000 | 13 per 1 000 (8 to 20) |
RR 0,77 (0,49 to 1,20) |
5661 (4 studies) |
⨁⨁⨁◯ MODERATE a,b,c |
|
| Asphyxia |
12 per 1 000 |
20 per 1 000 |
RR 1,67 |
508 |
⨁◯◯◯ |
|
| Caesarean delivery |
121 per 1 000 |
115 per 1 000 |
RR 0,95 |
5669 |
⨁⨁⨁◯ |
|
| Operative vaginal delivery |
89 per 1 000 |
84 per 1 000 |
RR 0,94 |
5069 |
⨁⨁⨁◯ |
|
| Severe perineal injury |
37 per 1 000 |
33 per 1 000 |
RR 0,89 |
5069 |
⨁⨁◯◯ |
|
| Postpartum hemorrhage |
99 per 1 000 |
99 per 1 000 |
RR 1,00 |
5069 |
⨁⨁⨁◯ |
|
| CI: Confidence interval; RR: Risk ratio | ||||||
|
Explanations a. Not downgraded for lack of blinding as unlikely to have influenced objective outcomes |
||||||
Discussion
The research results were generally comparable across studies. The induction methods were standard in all studies, and comparable to Norwegian practice. The women who participated in the studies were also relatively homogeneous across the studies and comparable to Norwegian women. However, in one of the studies, caesarean delivery was more widely practiced than in Norway, and in another study prenatal care after 41 weeks for women in the control group was less comprehensive than Norwegian prenatal care. In summary, we consider the generalisability of the results to be moderate when taking into account the GRADE assessments.
Although the findings in our and others' systematic reviews do not quite reach the point of assessment of confidence in the results, we doubt that more studies will help to change the conclusions.
Conclusion
Our systematic review is the fifth in a series of recent systematic reviews with fairly similar but not identical research questions. Our conclusion coincides with the conclusions in the most comprehensive, most updated and methodologically strongest of other reviews:
Induction at 41 weeks appears to reduce the risk of complications in newborns but does not appear to have significant implications for maternal outcomes.
Moreover, it seems that a majority, but not all, of the women prefer induction at 41 weeks rather than waiting until 42 weeks.
