When Is Evidence Based Medicine NOT Evidence Based? Inductions for PROM at Term
[Editor's Note: I would like to give a warm welcome to Science & Sensibility's newest regular contributor, Mayri Sagady Leslie. Mayri has a brilliant mind for making sense of obstetric research and has over a decade of experience putting evidence-based principles into practice as a nurse-midwife, midwifery practice director, and midwifery instructor. You can find out more about Mayri by clicking on the Contributors tab above. And look for more regular contributors joining the Science & Sensibility ranks this fall. - AMR]
The release of ACOG Practice Bulletin No. 80, Premature Rupture of Membranes in April of 2007 should be marked as a red letter day in the downfall of evidence-based maternity care in the United States. The story of this red letter day begins back in June of 1998 with the publication of the ACOG Practice Bulletin Number 1, Premature Rupture of Membranes: Clinical Practice Guidelines for Obstetricians-Gynecologists. This is the antecedent document which Bulletin 80 replaces. Both guidelines address the issues, evidence, and suggested management involved when a mother’s water bag breaks before labor begins near, at, or beyond her estimated due date.
All ACOG bulletins address relevant clinical questions, reviewing the current available evidence pertinent to those questions and providing citations to studies referenced. At the end, specific recommendations are made and graded by letters, with “A” representing recommendations coming from the most trustworthy evidence, usually randomized, controlled trials.
The 1998 version of the bulletin on Premature Rupture of Membranes (PROM) gave the following recommendation, based on Level A evidence:
With term PROM, labor may be induced at the time of presentation or patients may be observed for up to 24-72 hours for the onset of spontaneous labor.
However, the 2007 bulletin, nine years later, again based on Level A evidence, made a significant change in their recommendation as follows:
For women with PROM at term, labor should be induced at the time of presentation, generally with oxytocin infusion, to reduce the risk chorioamnionitis.
What’s wrong with this new recommendation? One might assume nothing. Obviously, in those nine years something changed in the Level A evidence. The problem is – nothing changed. No new significant studies were published that favored induction over observation for mothers who had no medical indications to start labor. In fact, both guidelines cite the exact same study.
It gets worse. The study, published by Hannah et al in 1996, known as the TERMPROM Trial, was a large clinical trial which took place from 1992 to 1995 in 72 medical centers throughout 6 countries. The study examined the effects of induction after PROM with oxytocin or prostaglandins versus waiting up to 4 days for labor to start on its own or inducing with the same 2 agents if labor still had not begun or of other problems developed (such as a fever or non-reassuring fetal heart rate pattern). The lowest rate of maternal infection was found in mothers that were induced with oxytocin. Presumably, since the Hannah trial is the only study cited as evidence for ACOG’s Clinical Bulletin 80, this is the source of the evidence that induction “ reduce[s] chorioamnionitis”. Yet, this logic is fraught with issues:
- There was no study protocol for either screening for or treating Group B Streptococcus (GBS)
- Only 20% of GBS culture results were available at the time of labor and birth management
- The majority of GBS+ mothers did not receive treatment
- 30% of all study mothers had vaginal exams at the time of PROM
- The group with the fewest infections was also the group with the fewest vaginal exams
(For more about these and other problems with the TERMPROM trial, see Henci Goer’s critique, “When Research is Flawed: Should Labor be Induced Immediately with Term Prelabor Rupture of Membranes?”)
To bring these issues into perspective, think about what we now know and how judicious care providers work with women today if their water breaks before labor at term:
- Unless a woman declines, she is screened for GBS before term, so if her water breaks, we can factor whether she is GBS positive or negative into our recommendation of whether she may be at less risk opting for an induction or waiting.
- As indicated by guidelines from the Center for Disease Control (CDC), we treat women who are GBS positive with antibiotics prophylatically to reduce the risk of infection for her and her baby.
- For women who decline screening or whose GBS status is unknown, we follow guidelines from the CDC which suggests that we treat according to the existence of other risk factors such as the length of time the membranes are ruptured, signs of infection in the mother and baby, etc.
- In either case, we know NOT to do any vaginal exams until we know mom is in active labor and even then to minimize and avoid them until absolutely necessary because vaginal exams themselves are one of the highest risk factors for increasing infections once the water bag has ruptured.
But here’s the tragedy. Bulletin number 80 was released. The recommendation was changed. Based on data that was 11 years old at the time, from one study that had no protocol for screening or treatment regarding maternal infection and was conducted before current guidelines were available or practiced – a new recommendation, focused on preventing infection, was made that represents itself as current and evidenced-based.
Now, this powerful “guideline” – which drives industry standards, institutional policies and procedures, medical school education, and, potentially, legal judgments – allows for just one option for mothers in this normal physiological condition: induction.
What impact is this having? As numbers of inductions increase so do the numbers of technological interventions such as electronic monitoring and the mother’s need for pain medications – as these increase her ability to move freely and choose her own positions decline as well. Operative deliveries have also been found to increase with inductions in some studies and with those come increased risks for more severe perineal injuries and lifelong issues with pelvic floor disorders. While inductions, like all medical interventions have their place and time when indicated – they cannot be justified as a standard procedure for a normal physiological occurrence in a healthy full term pregnancy.
I was giving a talk on this topic recently to a group of providers and a woman’s eyes suddenly lit up with recognition. Here’s what she said:
I was admitting a client recently whose water bag was broken and the attending said we have to induce her. I said why? He said, haven’t you heard, there is new evidence. It’s dangerous now to wait. We have to induce them now. I said no, I hadn’t heard.
When is evidence based medicine NOT based on the evidence? When one can make opposing recommendations based on the same exact study publication. When one can use findings that are inappropriately outdated and not applicable to current practice to make clinical recommendations. When is evidence based medicine NOT based on the evidence? When it’s not.
American College of Obstetricians and Gynecologists. (1998). ACOG practice bulletin. Number 1, June, 1998. Premature rupture of membranes. Clinical management guidelines for obstetrician-gynecologists. International Journal of Gynaecology & Obstetrics, 63(1), 75-84.
American College of Obstetricians and Gynecologists. (2007). ACOG practice bulletin. Number 80: Premature rupture of membranes. Clinical management guidelines for obstetrician-gynecologists. Obstetrics & Gynecology, 109(4), 1007-1019.
Hannah M. E., Ohlsson A., Farine D., Hewson S. A., Hodnett E. D., Myhr T. L., et al. (1996). Induction of labor compared with expectant management for prelabor rupture of the membranes at term. TERMPROM Study Group. The New England Journal of Medicine, 334 (16), 1005-10