Friday, August 15, 2003

Aspirin, ibuprofen and miscarriage

The British Medical Journal reports today that women using non-steroidal anti-inflammatory drugs and aspirin are 80 percent more likely to suffer miscarriage than those who do not, or who take paracetamol.

The study is based on 1055 pregnant women in the San Francisco area. 162 suffered miscarriage. Of these, 149 had not used aspirin or NSAIDs. 13 had used them. The overall hazard ratio was 1.8, but the 95 percent confidence interval ranged from 1.0 to 3.2. If the 1.0 figure is correct, then we can say there is no effect whatsoever.

Moreover, how many extra miscarriages were there compared to what would have been expected? About 18 percent of non-users suffered miscarriage. If the same rate had been true for NSAID users, then there would have been 9.5 miscarriages. Therefore, the study has found an extra three or four miscarriages for NSAID users. For aspirin, the number of extra miscarriages compared to what would have been expected was... one.

Based on such small numbers, it is hard to see how any firm conclusions can be drawn at all. What is worse, the study was actually investigating the effect of magnetic fields on miscarriage rates. Only after hearing about an admittedly flawed Danish study did the researchers look at the effect of NSAID and aspirin usage. Could this be a case of finding something in the results that is statistically significant but not real?

The strongest conclusion that can be drawn is that more research might be warranted.

Exposure to non-steroidal anti-inflammatory drugs during pregnancy and risk of miscarriage: population based cohort study, British Medical Journal, 16 August 2003


Post a Comment

<< Home