Regarding ‘Diabetes mellitus increases the risk of ruptured abdominal aortic aneurysms’

We read with great interest the recent publication of Wierzba et al. ¹ reporting the association between abdominal aortic aneurysm (AAA) and diabetes mellitus from the National Health Fund in Poland. Epidemiological studies performed so far have highlighted a strongly negative association between diabetes and AAA incidence/prevalence.^2,3 Meta-analysis revealed that diabetes was significantly associated with lower AAA rupture. ³ In their publication, Wierzba et al. ¹ report the results of epidemiological data of the National Health Fund and the Central Statistical Office in Poland in 2012. ⁴ Intriguingly, they found that the incidence of AAA was significantly higher in patients with diabetes compared to non-diabetics in both men and women. In addition, the incidence of AAA rupture was higher in diabetic population. We would like to highlight some issues, hoping that these comments will shed light on these unexpected results.

First, it would be of interest if the authors explained how diabetes mellitus was defined. Indeed, the majority of epidemiological studies define diabetes based on medical records according to patients’ declaration or to the use of antidiabetic drugs. However, when assessing the glycemic status by an oral glucose tolerance test (OGTT) and by measurement of glycated haemoglobin A1c (HbA1c) according to international guidelines, some authors revealed that half of the patients with AAA and diabetes mellitus were in fact unaware that they had diabetes. ⁵ Hence, methodology used to define diabetes could have impacted results. Additionally, it would be of interest to analyse for how long the diagnostic of diabetes had been made as other studies revealed that AAA prevalence did not differ between newly diagnosed type 2 diabetic patients and non-diabetics. ⁶ We suspect that the duration of diabetes could have potentially impacted findings reported by Wierzba et al. At last, it would be interesting to report the proportion of insulin-dependent patients from non-insulin-dependent patients. Indeed, these forms of diabetes are characterized by distinct pathophysiological features, and we can hypothesize that they could differentially impact AAA development.

Additional questions relate to the impact of potential confounding factors on AAA incidence and rupture. Indeed, clinical characteristics of patients including age, sex, smoking, arterial hypertension or low-high-density lipoprotein (HDL) cholesterol levels represent factors that could impact AAA development and rupture. ⁷ It would be of interest for the readers to detail if there were significant differences between patients with diabetes and non-diabetics and to comment if these factors could contribute to explain the discrepancy of results reported by Wierzba et al. from other studies.

At last, several studies have highlighted ethnicity specificities that impact diabetes epidemiology and molecular pathways involved.^8,9 The study of Wierzba et al. ¹ reports the findings on a cohort of patients included in Poland. We wonder whether the authors believe that diabetes in Poland had specific epidemiological characteristics that could potentially explain why the incidence of AAA and rupture was higher in their cohort.

All in all, the recent publication of Wierzba et al. and Czeleko et al. revealed unexpected results on AAA incidence and risk of rupture in context of diabetes. While epidemiological studies performed so far pointed to a protective role of diabetes, their results rise a controversy and highlight a more complex role of diabetes than previously anticipated. We believe that further prospective epidemiological studies with a standardization of the methodology used would be of interest to better understand the link between diabetes and AAA. A large amount of clinical and fundamental research is needed to put together the pieces of this complicated puzzle.