Traditional risk factors underestimate cardiovascular risk in individuals with coeliac disease in the Trondelag Health Study (HUNT4)

Background Individuals with coeliac disease (CeD), an immune-mediated enteropathy driven by dietary gluten proteins, have a moderately increased mortality, which is mainly attributed to cardiovascular disease (CVD) and malignancy despite a favourable cardiovascular risk profile(1-3). Thus, the underlying cause of an increased CVD prevalence among individuals with CeD is likely to differ from that of the general population. Both chronic inflammation, consequences of a gluten-free diet and non-traditional CVD risk factors may be potential drivers in this context(1,2). Purpose The aim of the present study was to assess both traditional and non-traditional CVD risk for individuals with CeD compared to non-CeD controls. In addition, we aimed to compare the estimated CVD risk by SCORE2 and SCORE2-OP with registry data on CVD. Methods Previously undiagnosed (new) CeD cases were identified by serological screening of 54,541 adult participants in the fourth survey of the Norwegian population-based Trøndelag Health Study (HUNT4), followed by histological confirmation. Known CeD cases were identified through linkage to health records(4). SCORE2(5) and SCORE2-OP(6) were calculated based on traditional CVD risk factors assessed in HUNT4. In addition, serum samples were analysed for lipoprotein (a), N-terminal pro B-type natriuretic peptide (NT-proBNP), interleukin 6 (IL-6), Growth Differentiation factor 15 (GDF-15), Fibroblast Growth Factor 23 (FGF-23), and soluble Suppression of Tumorigenesis 2 (sST2) in an age- and sex-matched subgroup (n=990). We used non-parametric tests for pairwise comparison of variables, and linear regression for examination of associations. Results Individuals with CeD exhibited a more favourable CVD risk profile compared to non-CeD controls based on traditional risk factors. In line with this, the 10-year estimated risk by SCORE2 of a first-time CVD event in middle-aged adults (40 to 69 years) was significantly lower (p=.017; see table) for individuals with CeD compared to controls. In contrast, the estimated CVD risk for older individuals by SCORE2-OP did not differ (p=.449; see table). Despite a lower overall estimated CVD risk (p=.004; see table), the incidence of CVD and heart failure after the first 5 years of follow-up did not differ between CeD cases and controls (p=.602 for a composite of ischaemic heart disease, stroke and death of CVD and p=.660 for heart failure in age- and sex-adjusted models; see table). Of note, GDF-15, a marker of general poor health(7) and unfavourable outcome in CVD(8), was higher in individuals with CeD (p=.049; see figure). NT-proBNP levels did not differ significantly between individuals with and without a known CeD diagnosis (p=.069; see figure). Conclusion Our findings indicate that CVD risk in individuals with CeD is underestimated when considering traditional risk factors only. Higher levels of GDF-15 in individuals with CeD support this conclusion.