Commentary: does use of food supplements influence the twin rate? New evidence from a randomized controlled trial.

Ovulation-inducing drugs are known to substantially increase the rate, but this is likely to explain only a minority of all occurrences. A number of other conditions have been found to be associated with multiple birth, including seasonality at birth, maternal age, social class, and diet. 1 However, the evidence relies nearly exclusively on observational studies, since very few randomized interventions have been undertaken in the peri-conceptional period which are large enough to enable the study of a rare outcome such as multiple birth. Until recently there was only evidence available from one such trial with a dietary intervention which was conducted in Hungary and published in 1992. 2 More than 4000 women were randomized to receive either a supplement containing different vitamins, minerals and trace elements or a supplement containing the trace elements only plus minute amounts of vitamin C, from at least one month prior to conception until the second missed menstrual period or later. The authors found a reduced occurrence of neural tube defects in the vitamin/mineral group (0 versus 6 cases), a finding which has contributed to current policies for preventing neural tube defects. A few years later, and more relevant to this commentary, the authors published results showing a significantly higher rate of twins in the group receiving vitamins and minerals, 3.8% versus 2.7%, which was an unexpected finding. 3,4 The statistical power to differentiate between effects on dizygotic and monozygotic twin rates was limited, but both twin types tended to be increased in the intervention group. These results prompted an analysis based on five observational data sets collected in the US, where peri-conceptional use of dietary supplements had been assessed retrospectively after birth. In this exploratory analysis, published in 1997, 5 four of the five studies exhibited a positive association between the twin rate and use of multivitamins from 1 month before to 3 months after conception. When exposure was defined as only comprising post-conceptional supplement use, the associations got weaker or disappeared, a pattern which was interpreted to reflect an effect on twinning rather than an effect on viability of twin pregnancies. Katz's et al.'s report 6 in this issue of the journal is based on a mega-trial conducted in a poor rural area of Nepal. The main finding of this randomized trial was a lower incidence of maternal deaths within 12 weeks of delivery in the two groups receiving vitamin A and beta-carotene …