One Healthy Baby Part 2a: Do IVF twins do better than spontaneous twins?

I felt the need to add a 2a section to the "One Healthy Baby" series after a recent Twitter follower asked if there was a difference in perinatal outcomes between spontaneous twins (natural occuring twins) and those resulting from IVF. Some of my colleagues have published research about differences among singletons and using their work I found an article that addresses this question.

Initial study suggested that singletons resulting from IVF had lower birth weights and were more likely to deliver preterm, but those initial studies were fraught with poor study design. For example, increased maternal age can lead to higher rates of preterm delivery and since infertile patient cohorts tend to be older - is it the IVF or the maternal age to blame for the higher rate of preterm delivery. 


Better designed, follow up studies that controlled for age, did confirm these findings, but some actually showed the opposite: IVF singletons do better. 

Eitherway, while data remains mixed, surprisingly, there is some consensus that IVF twins actually do "better" than spontaneous occurring twins. This paper (a meta analysis) from the British Medical Journal looked at 25 studies that examined outcomes among singletons and twins both born naturally and via IVF. Of those 25 studies, 17 used matched controls (good study design) and 8 did not (not as good a study design). 

The frequency of very preterm twins (<32 weeks) was examined in three matched studies (the better study design) in IVF twins was 7-10.5% and 4.9-10.7% in spontaneously occurring twins. Although the spontaneous twins have a range with a lower frequency (4.9%), overall the frequency does not differ between the two. Birth weights were not significantly different between the ART twins and spontaneous twins, either.

Rates of c-section were approximately 21% higher among mothers of IVF twins  compared to those of spontaneous twins.

Perinatal mortality (death) was actually lower in IVF twins compared to spontaneous twins but hard to rely on since there was a higher than average mortality rate in the control group in one study used by the authors in their meta-analysis.

This is a good example of the limitations of these types of studies (meta-analyses). As one mentor of mine likes to point out: if you put "junk" into a meta-analysis, you will get "junk" out - essentially, if you combine poorly designed studies to increase the number of people included to enhance a result, your conclusions may be flawed.

While the differences between IVF twins and spontaneously conceived twins do not appear to be markedly divergent, the lower risk of perinatal mortality among IVF twins remains puzzling. Obviously chorionicity matters (mono-chorionic twins are at higher risk for poor outcomes than dichorionic) and in studies that controlled for chorionicity, for example, dizygotic spontaneous twins had a lower perinatal mortality rate than dizyogtic IVF twins.

Ultimately, to answer the reader's question, there is evidence to support that IVF twins "do better" than spontaneously conceived twins, but the data is conflicting in the literature and the truth probably lies somewhere in between. Eitherway, twins are a high risk pregnancy that requires close attention from both the patient and the provider.