Background Evidence has pointed towards differences in the burden of arteriosclerosis according to its location and sex. Yet there is a scarcity of population-based data on aggregated sex-specific cardiovascular risk profiles, instead of single risk factors, and mortality risk according to the location of arteriosclerosis. We assessed sex-specific cardiovascular risk profiles and mortality risk associated with arteriosclerosis. Methods From the population-based Rotterdam Study, 2357 participants (mean age 69 years, 53% women) underwent non-contrast computed tomography to quantify calcification, as a proxy for arteriosclerosis, in the coronary arteries (CAC), aortic arch (AAC), extracranial (ECAC) and intracranial carotid arteries (ICAC), vertebrobasilar arteries (VBAC), and aortic valve (AVC). Principal component analysis (PCA) of eight distinct cardiovascular risk factors was performed, separately for women and men, to derive risk profiles based on the shared variance between factors. We used sex-stratified multivariable logistic regression to examine the associations between PCA-derived risk profiles and severe calcification at different locations. We investigated the associations of severe calcification with mortality risk using sex-stratified multivariable Cox regression. Results PCA identified three cardiovascular risk profiles in both sexes: (1) anthropometry, glucose, and HDL cholesterol; (2) blood pressure; and (3) smoking and total cholesterol. In women, the strongest associations were found for profile 2 with severe ECAC and ICAC (adjusted OR [95% CI] 1.32 [1.14–1.53]) and for profile 3 with severe at all locations, except AVC. In men, the strongest associations were found for profile 2 with VBAC (1.31 [1.12–1.52]) and profile 3 with severe AAC (1.28 [1.09–1.51]). ECAC and AVC in women and CAC in men showed the strongest, independent associations with cardiovascular mortality (HR [95% CI] 2.11 [1.22–3.66], 2.05 [1.21–3.49], 2.24 [1.21–3.78], respectively). Conclusions Our findings further underline the existence of sex- and location-specific differences in the etiology and consequences of arteriosclerosis. Future research should unravel which distinct pathological processes underlie differences in risk profiles for arteriosclerosis.

In this large population-based study, calcification in the vertebrobasilar arteries was prevalent in 21% of the elderly.• Calcification in the vertebrobasilar arteries was weak to moderately correlated with calcification in other arteries.• Traditional cardiovascular risk factors were associated with vertebrobasilar artery calcification.• Our findings suggest a different pathophysiology of arteriosclerosis in the posterior-compared to anterior circulation.

Aim Lipoprotein(a) [Lp(a)] is a potential causal factor in the pathogenesis of aortic valve disease. However, the relationship of Lp(a) with new onset and progression of aortic valve calcium (AVC) has not been studied. The purpose of the study was to assess whether high serum levels of Lp(a) are associated with AVC incidence and progression. Methods and results A total of 922 individuals from the population-based Rotterdam Study (mean age 66.0±4.2 years, 47.7% men), whose Lp(a) measurements were available, underwent non-enhanced cardiac computed tomography imaging at baseline and after a median follow-up of 14.0 [interquartile range (IQR) 13.9–14.2] years. New-onset AVC was defined as an AVC score >0 on the follow-up scan in the absence of AVC on the first scan. Progression was defined as the absolute difference in AVC score between the baseline and follow-up scan. Logistic and linear regression analyses were performed to evaluate the relationship of Lp(a) with baseline, new onset, and progression of AVC. All analyses were corrected for age, sex, body mass index, smoking, hypertension, dyslipidaemia, and creatinine. AVC progression was analysed conditional on baseline AVC score expressed as restricted cubic splines. Of the 702 individuals without AVC at baseline, 415 (59.1%) developed new-onset AVC on the follow-up scan. In those with baseline AVC, median annual progression was 13.5 (IQR = 5.2–37.8) Agatston units (AU). Lipoprotein(a) concentration was independently associated with baseline AVC [odds ratio (OR) 1.43 for each 50 mg/dL higher Lp(a); 95% confidence interval (CI) 1.15–1.79] and new-onset AVC (OR 1.30 for each 50 mg/dL higher Lp(a); 95% CI 1.02–1.65), but not with AVC progression (β: −71 AU for each 50 mg/dL higher Lp(a); 95% CI −117; 35). Only baseline AVC score was significantly associated with AVC progression (P < 0.001). Conclusion In the population-based Rotterdam Study, Lp(a) is robustly associated with baseline and new-onset AVC but not with AVC progression, suggesting that Lp(a)-lowering interventions may be most effective in pre-calcific stages of aortic valve disease.

Purpose Several studies have reported seasonal variation in intake of food groups and certain nutrients. However, whether this could lead to a seasonal pattern of diet quality has not been addressed. We aimed to describe the seasonality of diet quality, and to examine the contribution of the food groups included in the dietary guidelines to this seasonality. Methods Among 9701 middle-aged and elderly participants of the Rotterdam Study, a prospective population-based cohort, diet was assessed using food-frequency questionnaires (FFQ). Diet quality was measured as adherence to the Dutch dietary guidelines, and expressed in a diet quality score ranging from 0 to 14 points. The seasonality of diet quality and of the food group intake was examined using cosinor linear mixed models. Models were adjusted for sex, age, cohort, energy intake, physical activity, body mass index, comorbidities, and education. Results Diet quality had a seasonal pattern with a winter-peak (seasonal variation = 0.10 points, December-peak) especially among participants who were men, obese and of high socioeconomic level. This pattern was mostly explained by the seasonal variation in the intake of legumes (seasonal variation = 3.52 g/day, December-peak), nuts (seasonal variation = 0.78 g/ day, January-peak), sugar-containing beverages (seasonal variation = 12.96 milliliters/day, June-peak), and dairy (seasonal variation = 17.52 g/day, June-peak). Conclusions Diet quality varies seasonally with heterogeneous seasonality of food groups counteractively contributing to the seasonal pattern in diet quality. This seasonality should be considered in future research on dietary behavior. Also, seasonspecific recommendations and policies are required to improve diet quality throughout the year. Keywords Seasonality • Diet quality • Food frequency questionnaire • Food groups • Dietary guidelines Janine E. van der Toorn and Magda Cepeda contributed equally.