Type 2 diabetes-related sex differences in cardiovascular risk: reasons, ramifications, and clinical realities

This editorial refers to ‘Does type 2 diabetes confer higher relative rates of cardiovascular events in women compared with men?’†, by M. Malmborg et al., on page 1346. The average man is at greater cardiovascular risk than his female counterpart, developing their first cardiovascular (CV) event when ∼5–10 years younger. Many theories for this well-established fact have been proposed, centring mostly on effects of female sex hormones to either directly or indirectly lessen vascular risks. Remarkably, in the 1960s, researchers randomized men to differing levels of conjugated oestrogen vs. placebo to test effects on CV outcomes.1 A significant early increase in myocardial infarction rates with higher doses of conjugated oestrogen led to premature cessation of this arm. In doing so, this trial emphasized that sex differences in CV risk are more complex than appreciated and not simply ‘resolved’ by addition of sex hormones. On the other side of the equation, type 2 diabetes (T2DM) is generally considered to enhance CV risk more in women than in men.2 However, further high-quality data addressing this topic are needed as not all recent data showed such consistent patterns.3 , 4 In this issue of the European Heart Journal, Malmborg and colleagues have taken advantage of their excellent Danish hospital databases to ask several relevant questions.5 They sought to determine not only whether relative risks for a combined major adverse cardiovascular event and incident heart failure (MACE-HF) endpoint increase in women more than men in the context of T2DM, but also whether such patterns differ by age. They also asked to what extent sex differences in CV risk changes were evident for recurrent events, i.e. after a patient’s first event. Their main findings support a 15% greater relative risk increase in MACE-HF due to T2DM in women vs. men. This T2DM-induced excess relative risk increase in women was also evident across all ages but highest between ages 50 and 60 for the combined MACE-HF endpoint. However, for the separate heart failure and stroke endpoints, relative risk escalations were greatest for women at younger ages and less obvious at older ages. Finally, they observed no clear sex difference linked to T2DM for recurrent vascular events.5 The present study has several strengths, including its large size, minimal loss to follow-up (due to the excellent set up of the Danish Registries), and excellent presentation of key findings. Their ability to track outcomes with generally excellent sensitivity and specificity added further strength. On the down side, the investigators could not address to what extent risk factor differences explained sex-associated differences in CV risk, and nor did they include diet-only-treated patients as diabetes drug allocations were needed to make the diagnosis. They also did not look at people younger than 40 years of age and may have included some women with conditions [e.g. polycystic ovary syndrome) other than T2DM. For these reasons, readers should be less concerned about the accuracy of all the findings but rather concentrate on the big picture and how the key findings fit with or extend what is already known. More importantly, given many relevant papers on this topic, it is timely to now ask whether a reliable answer is apparent and, if so, what this means for practice and policy. The most important takeaways from this Danish study fit broadly to the best previous relevant data emanating from high-income countries,6 including very recent findings from neighbouring Sweden.7 Taking all such findings into account, one can reasonably conclude that T2DM imposes a greater relative increase in CV risk in women compared with men, with such differences being greatest at younger ages and least or negligible at older ages. However, this greater relative increase in CV risk does not mean women with T2DM are at greater absolute CV risk than their male counterparts. They are not. Rather, increments in absolute risk due to diabetes appear either broadly similar in men and women, or with a slightly greater increase in women, and, as men start with far higher absolute risks, they remain so even in the face of T2DM. This difference in relative vs. absolute risks is important and can create confusion if not properly explained. Notably, T2DM appears to be one of the few established CV risk factors which does increase relative risk more in women than in men, with the others being type 1 diabetes and atrial fibrillation, where female relative risks appear to be even more markedly elevated than in males.8 In contrast, an increase in relative CV risks due to the three other major CV risk factors, namely dyslipidaemia, hypertension, and smoking, appear broadly similar in men and women, at least as reported in QRISK3.8 So why might the relative risk for CV outcomes increase more in women than in men when T2DM manifests? The answers appear not to be due major sex differences in the use of preventative treatment.6 Rather, they are more likely to be due to differences in body composition between the sexes, with greater subcutaneous fat stores and lower liver fat in women. In fact, prior research suggests that adjustment for waist to hip ratio may best explain sex differences in CV risk, at least as assessed by carotid intima media thickness.9 Due to such body compositional differences, the average adult woman is more insulin sensitive than the average man and, to develop T2DM in the first place, women need to put on more weight than men to overcome this better insulin sensitivity,10 particularly at younger adult ages; the picture in childhood and adolescence is somewhat different.10 These observations explain why T2DM prevalence is lower in adult women compared with men, and why when women transit to T2DM, some of their other risk factors linked to obesity/insulin resistance may undergo greater relative changes than in their male counterparts.11 They also explain why relative risks for CV disease and heart failure increase more at younger ages in women when they develop T2DM (i.e. the group with the highest average body mass index). A clear picture now emerges whereby women, due in large part to more favourable body composition, have lower risks not only for CV outcomes but also for T2DM, with the transition to T2DM narrowing relative risks for CV outcomes between the sexes due to the need for greater weight gain and associated relative metabolic changes in women (Figure 1). That the current Danish study observed no clear sex difference in the risk for recurrent CV events with T2DM5 also seems to make sense as once a hard CV event has occurred, individuals, whether men or women, have shown themselves to be at comparable event risk. Patterns of T2DM and CV risk amongst men and women, potential reasons, and clinical ramifications. One must now ask what the aforementioned findings mean for clinical practice. As T2DM accelerates CV risk in both men and women with broadly similar or marginally greater increments in absolute risk in women, both men and women with T2DM deserve similar aggressive management. The only time period for any difference in management is the need to be careful about use of certain drugs in women with T2DM of child-bearing age. Otherwise, one should treat men and women with T2DM identically. This conclusion then leads to another important question. Is there a strong need for further research on sex differences in T2DM? Perhaps not, at least in high-income populations of mainly white ethnicity. It would, however, be interesting to conduct more relevant research in other ethnicities particularly given recent studies from China and Mexico3 , 12 where sex-related differences for key outcomes were not clearly apparent. Finally, away from a focus on sex differences, a much more important simple risk stratifier in T2DM is age of onset. As we recently reported, those who develop T2DM under 30–40 years of age may lose over a decade of life compared with their non-diabetes counterparts, whereas, once T2DM is diagnosed at >80 years of age, life expectancy losses appear negligible.7 This means that if the onset of T2DM in those at risk can be delayed, perhaps by a few years or a decade or more, then substantial benefits will ensue. We must therefore use any opportunity to identify men or women at elevated T2DM risk and help/encourage sustainable lifestyle changes to lose modest amounts of weight and, ideally to take part in more activity, to delay their conversion for as long as possible. Such changes would improve quality of life and lower future CV risks.13 Fortunately, we have an earlier warning signal of gestational diabetes in some women, though there is much to do to improve post-natal screening and preventative practice. However, we should not neglect targeting and screening men at higher risks. Men are at greater absolute risks for T2DM and CV disease than women, and die younger. This means we need better ways to entice men to take their health more seriously and get screened. Finally, the recently reported preliminary HbA1c and weight change data of the English National Health Service Diabetes Prevention Program (NHS DPP), the largest prevention programme globally, were encouraging, but suggest that we need better ways to reach some groups at highest risk.14 This includes in particular younger and more deprived individuals; the groups with the highest lifetime risks. Thus, whilst sex differences in the effects of T2DM are of interest, other areas of diabetes research and care should concern us more. By shifting our priorities to these areas, there is much to do but also much to gain. The opinions expressed in this article are not necessarily those of the Editors of the European Heart Journal or of the European Society of Cardiology. Thanks to Liz Coyle (University of Glasgow) for her technical assistance and with formatting of the figure.