The findings are consistent with previous studies directly linking HHV-6 endometrial infection to miscarriage.
Globally, 24 million cases of miscarriage occur annually, amounting to about 15% of pregnancies. While women experiencing a miscarriage typically have no lasting physical injury, the loss of a hoped-for newborn can be very upsetting for the parents and other family members. Past studies have linked various viral infections with some cases of miscarriage.
Over the past 20 years—with the advent of large genome-wide association studies (GWAS) and whole genome-sequencing (WGS) studies—enthusiasm has grown for conducting Mendelian randomization studies. In these studies, polymorphisms that are linked to specific exposures (like infections with specific viruses) are then searched for in people with specific outcomes (like miscarriage). The idea is that since polymorphisms occur at random, the chance of a specious association between an exposure and an outcome is reduced.
A multi-institutional team from China examined three large European databases containing genome-wide association study (GWAS) data as well as serological data for a large number of viruses. Using the complex statistical methods of Mendelian randomization studies, they found a likely causal association between miscarriage and infections by several human herpesviruses (HHV-1, HHV-3, HHV-4, HHV-6, and HHV-7), polyomaviruses (BK, JC, and Merkel cell polyomaviruses), and Chlamydia trachomatis. The strongest signal was for Merkel cell polyomavirus.
Mendelian randomization studies are conceptually attractive, since they reduce the risk of confounding (specious associations) between an exposure (like a viral infection) and an outcome (like miscarriage). However, they are most useful when specific polymorphisms have been unequivocally linked to the exposure. There was no clear evidence of such an unequivocal linkage in the text of this report.
Multiple previous studies have linked HHV-6A to primary infertility and to preeclampsia (which increases the risk of miscarriage), as summarized in a recent review (Komaroff 2021). Thus, this study’s conclusion that infection with HHV-6 may be risk factor for miscarriage is plausible. However, to this reader, the result of this Mendelian randomization study involving thousands of people is less persuasive than direct laboratory studies such as those showing HHV-6A in the endometrium of 43% of women with primary unexplained infertility vs. 0% of women without this condition (Marci 2016), and showing that HHV-6A infection affects the local immune response and trophoblast invasion (Caselli 2017; Bortolotti 2019; Pegoraro 2020).
Read the full text: Zhou 2024