Unfortunately there has been much misinformation spreading about the ARRIVE trial and what it means such that I think I now need to add my two pence worth.
This blog has been triggered by this story in today’s news titled “INDUCE TO CUT THE STILLBIRTH RISK” here are some quotes from it:
DOCTORS should induce labour in expectant mothers who reach 40 weeks to reduce the risk of stillbirths, according to one of WA’s most experienced obstetricians….
Dr Gannon said a 40-week delivery policy would require a “massive cultural change” for public hospitals. “But surely reducing the stillbirth rate is a worthy cause,” the immediate past president of the Australian Medical Association said.
Dr Gannon said research in Canada and the United States showed the benefits of inducing at 39 to 40 weeks included reduced stillbirths and caesareans.
So let me summarise what Dr Gannon seems to be saying. If we induce at 40 weeks instead of 40 weeks and 10 days, evidence shows that we will reduce both stillbirth and caesarean sections. That we should do this because it “worthy.”
So lets have a look at the “evidence” in the ARRIVE trial itself and what the authors themselves said about their findings
If you go to the journal there is this cute little video that sums the trial up quite nicely:
The reason for the trial was that traditionally induction of labour (IOL) between 39 weeks and 41 weeks has been avoided due to concerns that this might increase risk of caesarean.
That in their randomised controlled trail of more than 6,000 women who have not had a baby before, that 3062 were randomly assigned to have their baby induced between 39 weeks and 39+4 and the other group were randomly assigned to expectant management until their pregnancy reached at least 40 +5.
They compared “perinatal mortality and severe neonatal morbidity” and found no apparent differences between their groups. They did report that the caesarean section rate was slightly less in their induction group than in their expectant management group”
Their conclusion “For low risk women having their first baby that if you induce at 39 weeks you get “similar” rates of adverse perinatal outcomes and less frequency of caesarean births”
So lets have a look at this...directly from the source.
There are a few things to say about this table:
- You can see that the perinatal deaths were 2 in the IOL group and 3 in the expectant management group and that this was not significantly different. So people who are saying that this study showed reduced risk of stillbirth as a result of induction of labour are incorrect
- I think that it is disappointing that a study that set out to look at the “primary outcome” of perinatal mortality as well as perinatal morbidity was so underpowered to actually not be able to look at mortality.
- Perinatal death covers both stillbirth and neonatal death. It would be important to know when the 5 deaths occurred.
- I also think it is disappointing that they lumped all these outcomes in together and reported them as a composite…without comment. Don’t get me wrong of course it is completely fine to report the apparent lack of differences in the total score for rare outcomes but I have questions about the individual items making up this composite that are not answered anywhere in the paper and they are:
- When did the perinatal deaths occur?
- Is there any way the deaths could be linked to the intervention, for example did the 2 baby’s in the IOL group die as a result of a ruptured vasa praevia or cord prolapse? Did the 3 babies in the expectant management group die as a result of being in that group eg did their mothers present at 39 weeks and 2 days with altered fetal activity and there was a reluctance to induce because they were in the wrong arm of the trial?
- 11 babies in the IOL group had “seizures” versus only 4 in the expectant group. What were the seizures from ? and could these be linked in any way to the IOL?