Pre-Existing Gestational Diabetes not tied to increased risk of maternal Thromboembolism: Study

Diabetes causes physiological changes such as altered concentrations of coagulatory proteins, elevated levels of triglycerides and low levels of high-density lipoprotein. These physiological changes can contribute to hypercoagulability and hence an increase in the risk of cardiovascular diseases such as VTE. Obesity is linked to diabetes and therefore a plausible reason as to why obesity is a risk factor for maternal VTE. As such, the presence of diabetes could be considered a separate criterion for identifying women to participate in clinical trials. However, the degree to which diabetes increases the risk of VTE or whether any association is due to confounding by other variables which directly influence maternal VTE is not fully understood.

In this systematic review, authors synthesised population-based research on the association between diabetes and maternal VTE, primarily exploring whether there were differences between pre-existing and gestational diabetes and whether the association differed for VTE occurring during pregnancy or postpartum.

To comprehensively review literature on the extent to which pre-existing or gestational diabetes influences the risk of VTE in both pregnancy and postpartum authors used Medline, Embase and Google Scholar to identify observational studies published up to 2 November 2023. Studies which quantified the relationship between diabetes on antepartum and/or postpartum VTE, and which provide separate data for pre-existing and gestational diabetes.

Twenty one studies from Europe, the United States and Asia were included. There was an increased risk of antepartum VTE in women with gestational diabetes (RR=2.48, 95% CI 1.47 – 4.16, I2=45%, 4 studies) but not pre-existing diabetes (RR=1.71, 0.43 – 6.77, I2=68%, 2 studies). For postpartum VTE, there was no clear association with either pre-existing (RR=1.28, 0.73 – 2.24, I2=73%, 6 studies) or gestational (RR=1.39, 0.77 – 2.51, I2=70%, 10 studies) diabetes.

In a comprehensive review of literature using observational designs, authors found no clear evidence of a link between diabetes and risk of VTE either during pregnancy or following childbirth. Individual studies reported higher risks of VTE during pregnancy specifically for both pre-existing and gestational diabetes. However, heterogeneity between results was high. This would be expected based on differences in geography, study design, data source, methods of ascertaining VTE, differences in methodological quality and variables adjusted for in the analysis. Study addressed the last of these by restricting analyses to studies which adjusted for a measure of BMI or obesity. From this there was no obvious suggestion that there was an association between pre-existing or gestational diabetes and VTE risk which was mediated through BMI.

In a comprehensive systematic review, study found no clear evidence of a link between either pre-existing or gestational diabetes and the risk of maternal VTE, with high heterogeneity in results between studies. These results should enable healthcare workers to provide a degree of re-assurance when counselling pregnant women with diabetes about their likely risk of VTE, giving attention to other risk factors which may co-exist with this. However, the wide confidence intervals surrounding pooled effect estimates in this review will limit this degree of re-assurance. The UK RCOG thromboprophylaxis guidelines include Type 1 diabetes with nephropathy as a risk factor which would make them eligible for both antepartum and postpartum VTE prophylaxis. Another clinical guideline from Canada includes gestational diabetes in its guideline for postpartum thromboprophylaxis if it occurs alongside at least one other risk factor, while the American College of Chest Physicians pregnancy and VTE guidelines make no specific mention of diabetes. Data on Type 1 pre-existing diabetes specifically was only provided for three of the studies in this review. With registry based studies in particular rarely making this distinction because of how diabetes is coded, it may be difficult even for future studies to look at diabetes types separately to inform future guideline updates. Furthermore, studies included in this review did not consider the role of nephropathy, however, proteinuria has been found to be higher in women with a history of diabetes as well as being associated with VTE in pregnancy and postpartum. This would therefore be a plausible explanation as to why VTE risk could be higher in women with a history of diabetes.

Future research on this topic should shed light on the heterogeneity of findings reported to date, in particular whether characteristics particular to a small number of studies which observed a large association between a reporting of diabetes and VTE may give more precise clues as to the nature of any such association. It is essential that future cohort studies on this topic are based on a causal inference framework which considers that postpartum VTE in particular is influenced by a large number of often highly correlated factors.

In this case, the value will be understanding the degree to which either pre-existing or gestational diabetes may increase the risk of other risk factors, especially related to delivery which may themselves increase the risk of VTE. Finally, observational studies can only go so far in helping us settle controversies in this area. It is important that future clinical trials and value of information studies are conducted focussing on thromboprophylaxis targeted towards women at highest risk. Those at highest risk are where the benefits of receiving prophylaxis are considered to outweigh the harms according to various international guidelines. In this review, authors found insufficient evidence to advise that diabetes should be considered independently of obesity when identifying such women.

Source: Molly Orrin, Emilia Barber, Matthew J. Grainge; BJOG: An International Journal of Obstetrics & Gynaecology, 2024; 0:1–10 https://doi.org/10.1111/1471-0528.18043