Increased Risk of Aortic Valve Stenosis in Patients With Psoriasis

A Nationwide Cohort Study

Usman Khalid; Ole Ahlehoff; Gunnar Hilmar Gislason; Lone Skov; Christian Torp-Pedersen; Peter Riis Hansen


Eur Heart J. 2015;36(32):2177-2183. 

In This Article


The present study is, to the best of our knowledge, the first to assess the risk of AS in patients with psoriasis compared with the general population. The main result was that psoriasis was associated with increased risk AS independent of age, gender, comorbidity, and socioeconomic status. Importantly, the risk of AS demonstrated a dose–response relationship with respect to psoriasis severity.

Psoriasis is a T helper 1 (Th1)- and Th17 cell-driven immune-mediated disease associated with a range of comorbidities including cardio-metabolic diseases.[2–8] Indeed, previous studies have indicated that psoriasis is an independent risk factor for atherosclerotic disease and persistent systemic inflammation is likely to contribute to this association.[2,5,17,19] Although the pathogenesis of AS is poorly understood, it has traditionally been considered as a degenerative process that shares clinical risk factors with atherosclerosis.[9,11,20] However, evolving evidence has clearly suggested that AS is an active cellular process, where inflammation play an important contributory role.[9,11,12,20] For example, the presence of T lymphocytes and elevated levels of inflammatory mediators, e.g. tumour necrosis factor (TNF)-α, have been reported in stenotic aortic valves and appear to correlate with the progression rate of AS.[12,13,21] In addition, circulating levels of C-reactive protein and inflammatory adhesion molecules may be increased in patients with AS and recent studies with use of positron emission tomography have further corroborated that aortic valve inflammation is associated with AS severity.[22–24] TNF-α is also considered to be an important inflammatory mediator in pathogenesis of psoriasis adding weight to the hypothesis that coincident inflammatory mechanisms and genetic predispositions may explain, in part, the association between psoriasis and AS. Moreover, in addition to hypertension being more prevalent in patients with psoriasis, the chronic inflammatory state in psoriasis has been associated with increased arterial stiffness and endothelial dysfunction (both of which potentially amenable to treatment with anti-TNF-α agents) which may contribute to increased aortic valve haemodynamic stress leading to AS.[25–27] Large systematic cardiac imaging studies of unselected patients with psoriasis have not been reported, but in the as yet largest studies with echocardiography of patients with psoriasis without clinical evidence of cardiovascular disease, increased rates of aortic valve sclerosis or subclinical AS was not observed compared with controls, albeit that subclinical left ventricular dysfunction was more prevalent.[28,29] Increased aortic root diameter that correlated with psoriasis severity and platelet activation was also recently reported, suggesting the intriguing possibility that this may represent an early reflection of the more systemic nature of AS, e.g. with effects on both the left ventricle and the downstream systemic vasculature.[30] Importantly, we and others have reported that the risk of congestive heart failure is increased in patients with psoriasis and the contribution of AS to this finding clearly requires further study.[31–33]

Study Strengths and Limitations

The major strengths of the present study include the use of prospectively recorded nationwide data, complete follow-up, adjustment for important confounding factors, and use of validated measures of exposure and endpoints. Due to a government-financed free of charge and readily accessible health care for all Danish inhabitants, the risk of selection bias related to e.g. gender, age, health insurance, and socioeconomic status was evaded. Furthermore, inclusion of the entire adult Danish population, and adjustment for continuously updated comorbidity at baseline reduced the potential for surveillance bias. Exclusion of subjects with prevalent psoriasis and/or AS at the study start ensured a more exact allocation of time at risk for the study population. Our results were also strengthened by the psoriasis severity-dependent increased risk of AS and this association both remained statistically significant after adjustments for potential confounders and in sensitivity analyses aimed at reducing surveillance bias, respectively.

When interpreting the results, there are certain limitations to be acknowledged. We used number of hospitalizations to classify psoriasis severity, which may have increased the surveillance bias and thus decreased the threshold for detection of the study outcome in the psoriasis population. We addressed this limitation by adjusting for important confounding factors including concomitant medication and comorbidities, and by conducting sensitivity analyses, respectively, as described above. Also, the use of claimed prescriptions of vitamin D derivatives to define mild psoriasis, does not account for patients who may have received other topical therapies or ultraviolet light treatment for psoriasis. However, this method of identifying mild psoriasis has previously been validated and it was demonstrated that ~75% of the Danish population with psoriasis was prescribed with vitamin D derivatives as first-line treatment.[17] Hence, bias related to a possible misclassification is expected to be minor and would arguably favour the null hypothesis. Aortic valve stenosis is a condition that progresses slowly over time and the prevalence has been reported to be 0.2% in subjects aged 50–59 years, increasing with age up to 9.8% in individuals 80–89 years of age.[10] To determine the impact of age on the association between psoriasis and AS, we therefore performed an age-stratified analysis, and consistent with our primary analysis we here also found incidence rates of AS to be significantly increased in subjects with mild and severe psoriasis, compared with the reference population. Notably, rates of AS increased exponentially with age and were highest for patients with severe psoriasis >65 years of age. Among other limitations to be mentioned, the study was based on administrative registers and therefore did not include important clinical parameters such as echocardiographic data, blood pressure, smoking status, lipid profiles, and measures of obesity. Importantly, however, the multivariable adjustments for socioeconomic status, comorbidities, and a number of concomitant medications are likely to have covered some of these effects. Finally, the Danish population predominantly consists of Caucasians, which may limit the ability to generalize the results to other ethnicities.