Influenza Vaccination and Stillbirth Prevention in High-Income Countries: Is It Really That Effective?

To the Editor—As a multidisciplinary group of professionals working in the fields of prevention, pharmacoepidemiology, and stillbirth research, we read with interest the article by Regan et al published in the March issue of Clinical Infectious Diseases [1]. The authors reported that the adjusted risk of stillbirth was 51% lower among women vaccinated against seasonal influenza compared with unvaccinated women. We believe that any effort aimed at clarifying the possible causes of stillbirth should be welcomed; nevertheless, we think that the estimated risk reduction might be exaggerated: the data seem rather too good to be true, and we suggest that they should be reanalyzed. A proper time-dependent Cox regression analysis was carried out, with vaccination status as the time-dependent exposure. This approach is important to avoid the maligned immortal time bias [2]. However, the underlying time variable of follow-up started at 20 weeks of pregnancy, with 38% of vaccinated women having received the vaccine prior to week 20. This may have introduced bias from depletion of susceptibles [3]. Indeed, some women who received the vaccine in the first half of pregnancy could have experienced a miscarriage after vaccination and before week 20. These women were inherently not included in the study cohort since they did not make it to week 20, whereas women who did not experience a miscarriage in the first part of the pregnancy, who may represent a lower-risk group, were included in the post-20 week study cohort (Figure 1). Indeed, miscarriage and stillbirth may share the same etiology [4], and having had previous losses has been shown to be an independent risk factor for stillbirth in subsequent pregnancies [5]. The hazard rate for vaccinated women compared with unvaccinated women could have been reduced by selection bias if some vaccinated women had a miscarriage before week 20 and were selected out of the study cohort; such a phenomenon may have overestimated the protective effect associated with the vaccination. Consequently, to avoid such potential form of selection bias, it would be worthwhile to restrict the analysis to women immunized after 20 weeks of pregnancy and compare their risk of pregnancy loss to that of unvaccinated women. Estimated risk of pregnancy loss by gestational age in high-income countries and depletion of susceptibles. Unfortunately, also in countries with a relatively low stillbirth rate, each stillborn baby per se represents a tragic life event for families. Suggesting that more than half stillborn babies could have been saved through a simple intervention such as influenza vaccination during pregnancy, in the absence of firmly established data, could be misleading and put an unnecessary burden of guilt on stillbirth mothers who did not undergo vaccination. Although we recognize influenza vaccination during pregnancy as a public health intervention of paramount importance to prevent severe disease in pregnant women, as well as in their newborns, and we support the scientific societies’ recommendation to immunize all pregnant women without contraindications to vaccination [6], we also suggest that its efficacy in preventing stillbirth should not be overemphasized, so as to provide parents and professionals with clear, unbiased, and affordable data on the real extent of such a benefit. Potential conflicts of interest. All authors: No potential conflicts. All authors have submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest. Conflicts that the editors consider relevant to the content of the manuscript have been disclosed.

Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.