January 5, 2016

expert reaction to study investigating oral medication for thrush in pregnancy and risk of miscarriage

An association between use during pregnancy of a specific type of orally-administered medication for thrush and risk of miscarriage is the subject of a paper published in the Journal of the American Medical Association, in which the authors recommend cautious prescribing of the medication and further investigation. In the UK the medicine concerned is already not recommended for use during pregnancy.

Dr Sarah Branch, deputy director of MHRA’s Vigilance and Risk Management of Medicines division, said:

“Fluconazole is an important and widely used medicine for the treatment of fungal infections, including vaginal candidiasis (thrush). The product information accompanying fluconazole states that it should not be used during pregnancy unless clearly necessary. In the UK, alternative antifungal agents (topical clotrimazole and miconazole) are recommended for the treatment of thrush in pregnant women.

“If people have any questions they should consult their pharmacist or doctor for further advice.

“As with all medicines, the MHRA continually monitors the safety and efficacy of fluconazole, and will take action to protect patient safety as appropriate. The MHRA is aware of the publication in JAMA by Mølgaard-Nielsen et al. and will consider whether the findings of the study have any implications for healthcare professionals and patients.”

Prof. Kevin McConway, Professor of Applied Statistics, The Open University, said:

“Although this is an observational study, the fact the authors used propensity score matching allows us to be slightly more confident that the association they found really did have something to do with fluconazole (though the paper does, properly, avoid making a causal interpretation of the association). However, the fact that they used propensity score matching also makes it a bit hard to look at the size of the increase in risk. The authors used almost all the eligible pregnancies where the mother was exposed to fluconazole during the appropriate time window, but they compared them only with unexposed pregnancies that they were matched to. This meant that they didn’t include most of the unexposed pregnancies that were in the register. For example, for spontaneous abortion they included about 13,000 unexposed pregnancies but they left out about 1.4 million such pregnancies because they weren’t needed for the matching. So the adverse event rate in the unexposed pregnancies that they did include might not be typical of the adverse event rate in all unexposed pregnancies, if there is something unusual about women who are prescribed fluconazole compared to the general pattern in Danish pregnant women – and in fact the fluconazole takers do look different in several respects (if you look at the numbers in tables 1 and 2 before matching).  So you can’t say that the adverse effect risk for all Danish pregnancies is the same as what it is for the pregnancies in this study.

“The interpretation in terms of risk is made more difficult by the fact they fitted proportional hazards models and found hazard ratios rather than, say, odds ratios for the outcomes. It was correct to do that, particularly for spontaneous abortion, because the exposure time comes into it – if an antifungal did indeed increase the chance of a miscarriage, it perhaps wouldn’t be so likely to do so if it was taken just before the end of the relevant window (22 weeks) as if it was taken at the start (7 weeks), and also because spontaneous abortion can occur at any time within quite a long window. But hazard ratios are a lot less intuitive to interpret. Just because a hazard ratio is, say, 1.5, that doesn’t mean that someone who is exposed to fluconazole will have a 50% higher chance of a spontaneous abortion than someone who isn’t exposed.

“It’s worth remembering that for women in a country like Denmark with very good health care, the chance of a pregnancy ending in a miscarriage between weeks 4 and 22 is small, and the chance of a stillbirth is smaller still. If fluconazole is associated with an increase of risk of the kind of amount observed in this study, then in absolute terms, that’s a very small increase to what’s already a small risk.”

‘Association between use of oral fluconazole during pregnancy and risk of spontaneous abortion and stillbirth’ by Ditte Mølgaard-Nielsen et al. published in JAMA on Tuesday 5 January 2016. 

Declared interests

Prof. Kevin McConway declares that he has no relevant conflicts of interest.

No others received