Women and men face different heart attack risks with diabetes

A study spanning four decades reveals how cholesterol abnormalities and high blood pressure help explain differences in diabetes-related heart attack risk between women and men, with warning signs emerging years before diabetes is diagnosed.

Study: Sex differences in the diabetes-related risk of incident myocardial infarction and relationship to established CHD risk factors. The Tromsø Study (1979-2016). Image credit: PeopleImages/Shutterstock.com

A study published in European Heart Journal Open reported that the relative increase in myocardial infarction, or heart attack, risk associated with diabetes was similar in women and men after adjustment for conventional cardiovascular risk factors. The initially observed sex difference narrowed after adjustment for these factors, whose combined contribution was greater in women than in men.

Why diabetes may affect heart attack risk differently by sex

Diabetes is an established risk factor for cardiovascular disease, including myocardial infarction (MI). Previous research has suggested that diabetes may confer a greater relative increase in cardiovascular risk in women than in men. However, the reasons for this difference, and the extent to which conventional risk factors for coronary heart disease (CHD) account for it, remain unclear.

Diabetes often occurs alongside such risk factors, including dyslipidemia, high blood pressure, smoking, and excess body weight. Diabetic dyslipidemia is characterized by elevated triglycerides and low levels of high-density lipoprotein (HDL) cholesterol, with greater differences among women.

Diabetes in women is also more often accompanied by hypertension, which also affects women’s MI risk more than in men. These abnormalities may independently contribute to cardiovascular risk, and may also help explain the association between diabetes and MI.

The researchers aimed to determine whether the diabetes-related risk of a first MI differed between women and men, and to quantify the contribution of established CHD risk factors to this association.

Tromsø Study examines diabetes and heart attack incidence over time

The prospective analysis included 42,907 participants in the Tromsø Study, a population-based study in Tromsø, Norway. Women accounted for 51.5% of participants. The analysis covered ages 35–104 years and used data collected between 1979 and 2016, with follow-up for heart attacks continuing until the end of 2019.

During follow-up, 3,942 participants had a new heart attack, of which 357 occurred in people with diabetes. The researchers used Poisson regression to estimate incidence rate ratios (IRRs), which compare the incidence rates between groups while accounting for differences in the follow-up duration.

Diabetes-related MI risk in women and men

The researchers first examined the association between diabetes and MI after adjusting for age and birth cohort. They then assessed how this association changed after accounting for established CHD risk factors, and quantified the contribution of each factor to the association.

Diabetes became overtly associated with a higher MI risk from approximately age 45 in men and age 55 in women. After adjustment for age and birth cohort, the IRR for MI associated with diabetes was 2.26 in women and 1.84 in men. That is, the incidence of MI among women and men with diabetes was 2.26 and 1.84 times that among women and men without diabetes, respectively.

However, the sex difference was not statistically significant. Notably, age-stratified analyses showed that men with diabetes were at the highest absolute risk for MI over most of their lives. Women with diabetes had an almost similar risk to that in men without diabetes.

Effects of CHD risk factors

The researchers next examined how individual risk factors affected the diabetes–MI association. The triglyceride-to-HDL cholesterol ratio was the major contributor in both men and women.

When adjusted for this measure, the association between MI and diabetes weakened in both sexes. The finding highlights the importance of diabetic dyslipidemia in explaining the increased MI risk associated with diabetes.

In models accounting for multiple cardiovascular risk factors, adjustment for systolic blood pressure weakened the MI-diabetes association in women by 10.3%. Conversely, it strengthened the association in men by 2.2%. This difference suggests that elevated systolic blood pressure contributed more to the diabetes-related MI risk observed in women.

Adjustment for smoking and total cholesterol strengthened the diabetes association in both sexes, but more in men than in women. In the combined model, adjustment for smoking strengthened the association by 23.1% in men versus 7.1% in women. Thus, accounting for smoking revealed a stronger diabetes–MI association, particularly among men.

Body mass index (BMI) affected risk similarly in both sexes, with no major impact.

Combined impact greater in women

When the researchers considered the combined impact of the CHD risk factors, they found that it reduced the strength of the association more in women, by 33.6%, compared with 15.9% in men.

Joint adjustment thus narrowed the initial difference between the sexes. The adjusted IRR was 1.72 in women and 1.67 in men. The results indicate that established CHD risk factors accounted for a substantial part of the initially observed difference in the diabetes-related MI risk estimates.

The effect of several factors in the joint adjustment was weaker than with individual adjustment. For instance, though the triglyceride-to-HDL-C ratio remained the single largest contributor to weakening the diabetes–MI association, its effect was smaller after combined adjustment, from -29.6% to -17.1% in men, and from -28% to -19% in women.

When stratified by age, the triglyceride-to-HDL-C ratio had the largest negative impact on the MI-diabetes association in older individuals, both men and women, at or above 75 years, but more in men than in women. Between 55 and 74 years, the association in women was weakened to a similar extent by adjustment for this ratio and systolic blood pressure.

Adjustment for smoking became the major contributor to strengthening the diabetes–MI association in men, with its impact increasing from +11% to as high as 23.1%. For women, accounting for the stronger impact of systolic blood pressure on MI risk attenuated the association by 9%-10%.

Risk factors before diabetes develops

The researchers found that adverse lipid levels and increasing systolic blood pressure were already evident during the prediabetic phase. Lipid levels were generally worse among individuals who later developed diabetes, and stabilized or improved soon after diabetes onset. Among those who had an MI later, lipid levels were more unfavorable at and after diabetes onset.

The researchers hypothesize that dyslipidemia in people with diabetes is driven by insulin resistance, and that it mediates part of the MI risk in this group.

Among women who later developed diabetes and had an MI, systolic blood pressure was higher before diabetes onset, and continued to rise afterward in correlation with the duration of diabetes. This was less obvious among men with diabetes. According to the researchers, elevated systolic pressure may mediate some of the increase in MI risk.

Similarly, BMI increased in both sexes, reaching obesity levels 5–10 years before diabetes onset, before declining afterward. Women with diabetes who subsequently experienced an MI had notably higher BMI before diabetes onset than those who did not, followed by a more pronounced decline after diagnosis.

This observation suggests that cardiovascular risk factors are already established before diabetes is diagnosed. It also emphasizes the importance of considering the broader cardiovascular risk profile of people with prediabetes, rather than focusing exclusively on blood glucose or diabetes status.

The changes in risk estimates following statistical adjustment do not indicate that the individual risk factors cause MI in people with diabetes.

The findings also do not show that diabetes-related MI incidence is similar in women and men, in absolute terms. Rather, they describe relative risks in men and women.

Uncertain diabetes diagnosis timing may affect risk estimates

This was an observational study, precluding causal inferences. Age at survey entry was used if age at diabetes onset was unknown, potentially introducing inaccuracies. Residual confounding may have persisted, with unknown impact on the results. The study did not distinguish between type 1 and type 2 diabetes.

Cardiovascular warning signs emerge years before diabetes diagnosis

The Tromsø Study found that diabetes was associated with a higher incidence of MI in both women and men. After adjusting for age and birth cohort, the adjusted relative risk estimate was higher in women, but the difference was not statistically significant. Adjustment for established CHD risk factors, particularly the triglyceride-to-HDL cholesterol ratio and systolic blood pressure, weakened the associations, but more in women. The final IRRs were almost identical, at 1.72 in women and 1.67 in men.

In particular, lipid abnormalities and elevated systolic blood pressure were present in the prediabetic phase. The findings underscore the contribution of conventional cardiovascular risk factors to diabetes-related MI risk and the importance of identifying adverse lipid profiles and elevated blood pressure, including during the prediabetic phase.

These factors are potentially important targets for cardiovascular risk assessment in people with prediabetes and after diabetes onset, including women.