SEARCH

SEARCH BY CITATION

Keywords:

  • cardiovascular diseases;
  • epidemiology;
  • inflammation;
  • myocardial infarction;
  • stroke

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Funding source
  9. Disclosures
  10. References

Abstract.  Ahlehoff O, Gislason GH, Charlot M, Jørgensen CH, Lindhardsen J, Olesen JB, Abildstrøm SZ, Skov L, Torp-Pedersen C, Hansen PR. (Copenhagen University Hospital Gentofte, Hellerup; Copenhagen University Hospital Bispebjerg, Copenhagen; National Institute of Public Health, University of Southern Denmark, Copenhagen; Copenhagen University Hospital Gentofte, Hellerup; University of Copenhagen, Copenhagen, Denmark) Psoriasis is associated with clinically significant cardiovascular risk: a Danish nationwide cohort study. J Intern Med2011; 270: 147–157.

Objective.  The magnitude of the cardiovascular risk from psoriasis and psoriatic arthritis is debated. We therefore investigated the psoriasis-related risk of adverse cardiovascular events and mortality.

Design, setting and subjects.  We conducted a cohort study of the entire Danish population aged ≥18 years followed from 1997 to 2006 by individual-level linkage of nationwide registers. Psoriasis was defined by prescription claims and classified as severe if patients received hospital-based treatment. Time-dependent Poisson regression models were applied to assess cardiovascular risk in patients with psoriasis and psoriatic arthritis.

Main outcome measures.  All-cause mortality, cardiovascular mortality and hospitalizations for myocardial infarction (MI), stroke and coronary revascularization were recorded.

Results.  A total of 34 371 patients with mild psoriasis and 2621 with severe psoriasis, including 607 with psoriatic arthritis, were identified and compared with 4 003 265 controls. The event rates and rate ratios (RRs) of all-cause mortality, cardiovascular death, MI, coronary revascularization, stroke and a composite of MI, stroke and cardiovascular death were increased in patients with psoriasis. The rate ratio increased with disease severity and decreased with age of onset. The overall RRs for the composite endpoint were 1.20 (95% confidence interval [CI] 1.14–1.25) and 1.58 (95% CI 1.36–1.82) for mild and severe psoriasis, respectively. The corresponding RRs for cardiovascular death were 1.14 (95% CI 1.06–1.22) and 1.57 (95% CI1.27–1.94). The risk was similar in patients with severe skin affection alone and those with psoriatic arthritis.

Conclusions.  Psoriasis is associated with increased risk of adverse cardiovascular events and all-cause mortality. Young age, severe skin affection and/or psoriatic arthritis carry the most risk. Patients with psoriasis may be candidates for early cardiovascular risk factor modification.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Funding source
  9. Disclosures
  10. References

Psoriasis is a chronic inflammatory disease that affects approximately 2% of the population. There is evidence that psoriasis is associated with conventional cardiovascular risk factors, including hypertension, diabetes mellitus (DM), obesity, smoking and hyperlipidaemia and that this disease, like other type 1 helper T-cell (Th1)-driven inflammatory disorders such as rheumatoid arthritis and systemic lupus erythromatosus, is an independent risk factor for atherothrombotic disease [1–10]. Psoriasis and atherosclerosis are both characterized by Th1 and T helper type 17 (Th17) activation and reduced T regulatory cell (Treg) function, and there are striking similarities between the immunoinflammatory mechanisms involved in these two diseases [11]. Psoriasis has also been associated with increased all-cause mortality and cardiovascular death [12–15] and with subclinical markers of atherosclerosis [16–18]. Furthermore, in patients with psoriasis, systemic anti-inflammatory therapy with methotrexate may reduce the incidence of cardiovascular disease and tumour necrosis factor-α inhibitors can reduce endothelial cell activation and circulating levels of inflammatory markers including C-reactive protein [19–21].

Although the association between psoriasis and cardiovascular disease has been established in hospital-based studies and population-based cohorts [6–10, 12–15], conflicting results have been reported, and data from national prospective registries have been called for to more precisely define the risk of cardiovascular disease in unselected patients with psoriasis [22–25]. In addition, the magnitude of cardiovascular risk conferred by psoriasis and psoriatic arthritis compared to traditional risk factors is unknown. Indeed, such knowledge could have considerable impact as demonstrated by the widespread increased public awareness of cardiovascular risk associated with rheumatoid arthritis following the recent observation that this chronic inflammatory condition is as important as DM as an independent risk factor for cardiovascular disease [26, 27]. We therefore used Danish nationwide registries of hospitalization and drug dispensing from pharmacies to determine the risk of cardiovascular disease in patients with psoriasis and psoriatic arthritis compared with the general population and patients with DM.

Methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Funding source
  9. Disclosures
  10. References

Study population and data sources

The study was conducted and reported in accordance with the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) recommendations [28]. In Denmark, all citizens have a unique personal civil registration number that enables individual-level linkage of information across nationwide registers. The study population comprised all Danish citizens aged ≥18 years on 1 January 1997, and participants were followed until 31 December 2006, or death.

Morbidity was obtained from the Danish National Patient Register, in which all hospital admissions, diagnoses and invasive procedures have been recorded since 1978 according to the International Classification of Diseases eighth revision (ICD-8) until 1994 and 10th revision (ICD-10) thereafter. Data regarding all medications dispensed from pharmacies were obtained from the National Prescription Registry (the Danish Registry of Medicinal Product Statistics), in which all dispensed prescriptions have been recorded since 1995. All deaths were identified from the Central Population Register, in which deaths are recorded within 2 weeks. Causes of death were obtained from the National Causes of Death Register, in which causes of death are recorded using ICD-10 codes. Data on death, comorbidity and concomitant medication were linked at the individual level and were also linked to socio-economic data. Patients with psoriasis were identified by prescription claims for vitamin-D derivatives (i.e. topical treatment used exclusively for psoriasis) [29]. Patients were included when claiming their second prescription for these agents to ensure persistent medical treatment for psoriasis. Patients with severe psoriasis were identified by hospitalizations for psoriasis (ICD-10 L40) or psoriatic arthritis (ICD-10 M070–M073) and included at the time of their third (inpatient or outpatient) diagnosis. To validate our definition of severe psoriasis, we examined the medical records of 50 randomly selected patients consecutively referred for first-time hospital-based treatment of psoriasis. The majority of patients were actively treated with topical agents when evaluated and presented with a mean psoriasis area and severity index score of 10, which was consistent with severe psoriasis (data not shown)[30].Patients with DM were identified by their first prescription claim for oral glucose-lowering drugs or insulin. Comorbidity at study entry was described by the Charlson index, as defined by 19 prespecified diagnoses at study entry and up to 1 year previously, and modified to ICD-10 [31]. Subjects with prevalent DM, psoriasis or atherosclerotic disease [including prior stroke and/or myocardial infarction (MI)] were excluded from the study. Socio-economic status was defined by average yearly gross income during a 5-year period prior to study inclusion, and patients were divided into quintiles according to their income.

Medical treatment

Because of the national healthcare reimbursement scheme for drug expenses, pharmacies in Denmark are required to register all dispensed prescriptions ensuring complete registration. Drugs are registered according to the international Anatomical Therapeutic Chemical (ATC) classification system. Prescriptions claimed for topical vitamin-D derivatives (ATC D05AX) and glucose-lowering drugs (ATC A10A) were used to define psoriasis and DM, respectively, as described previously. Treatment with beta-blockers (ATC C07), angiotensin-converting enzyme inhibitors/angiotensin-2 receptor antagonists (ATC C09), vitamin-K antagonists (ATC B01AA), loop diuretics (ATC C03C), statins (ATC C10A), platelet inhibitors (ATC B01AC) and spironolactone (C03D) was defined by prescriptions claimed up to 6 months prior to the study inclusion date.

Study endpoints

The following endpoints (ICD-10 codes) were assessed: MI (I21–I22), stroke (I60–I61, I63–I64), coronary revascularization (percutaneous coronary intervention and coronary artery bypass grafting), all-cause mortality, cardiovascular death (I00–I99) and the composite of stroke, MI and cardiovascular death. The diagnoses of MI (positive predictive value, 79.6–93.5%) and stroke (positive predictive value, 80.5–86.0%) in the Danish National Patient Register have previously been validated [32, 33].

Statistical analysis

Baseline characteristics were presented as means with standard deviations or frequencies and percentage. Differences in baseline characteristics were assessed by Student’s t-test for continuous variables and chi-square test for categorical variables. The rate ratio (RR) and 95% confidence interval (CI) of study endpoints was estimated by time-dependent Poisson regression models adjusted for confounding factors including age, calendar year, concomitant medication, comorbidity, socio-economic data and gender. Psoriasis status was included as a time-dependent variable so that patients were only considered at risk from the time they fulfilled the psoriasis criteria. Age and calendar year were also included as time-dependent variables, and comorbidity and concomitant medication on 1 January 1997 were included as fixed variables. Unadjusted event rates were summarized as events per 1000 patient-years. Using the Greedy matching macro (last accessed on 4 January 2010, at http://mayoresearch.mayo.edu/mayo/research/biostat/upload/gmatch.sas), patients with psoriasis were matched for age and gender with four controls from the general population. Using multivariate Poisson models, adjusted for baseline characteristics obtained at the time of psoriasis, we accessed the impact of changes in baseline characteristics during the study period. In addition, analyses were performed with the use of propensity score-matched models conditional on baseline covariates. A propensity score for the likelihood of receiving topical vitamin-D derivatives was quantified by multivariate logistic regression. Using the Greedy matching macro, each case was matched to one control based on the propensity score allowing for unadjusted analyses of similar study populations with regard to measured confounders. Analyses with inclusion of patients with psoriasis at their first vitamin-D prescription claim or first psoriasis-related diagnosis were conducted to assess the potential impact of high healthcare consumption inherited in the primary inclusion criteria. To further counter the risk of surveillance bias, all analyses were repeated with exclusion of participants with a history of hospitalizations up to 1 year prior to inclusion. Furthermore, cardiovascular risk was analysed in patients with severe psoriasis and psoriatic arthritis compared to patients with DM to establish a clinical context of the magnitude of cardiovascular risk conferred by psoriasis and psoriatic arthritis compared to a well-established cardiovascular risk factor. Model assumptions, including absence of interactions between model covariates, were tested and found to be valid unless otherwise indicated. We found a statistically significant interaction between psoriasis and age (P < 0.05), and therefore overall as well as age-stratified estimates are presented. We estimated the impact of an unmeasured confounder according to the ‘rule out’ approach [34]. Statistically significant differences between parameter estimates derived from the Poisson regression analyses were tested by Wald chi-square test. All statistical analyses were performed with the sas statistical software version 9.2 (SAS institute, Cary, NC, USA) and stata software version 11 (StataCorp, College St., TX, USA).

Ethics

This study was approved by the Danish Data Protection Agency (ref. 2007-41-1667), and data at the individual level were made available to us by the national registers in anonymized form. Registry studies do not require ethical approval in Denmark. The authors had full access to the data and take complete responsibility for the integrity of the data.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Funding source
  9. Disclosures
  10. References

Baseline characteristics and events rates

A total of 34 371 patients with mild psoriasis and 2621 with severe psoriasis, including 607 with psoriatic arthritis, were identified and compared with 4 003 265 controls (Fig. 1). Baseline characteristics are presented in Table 1. Psoriasis, and particularly severe psoriasis, was generally associated with higher event rates, and the number of events per 1000 patient-years is presented in Fig. 2.

image

Figure 1. Flowchart of selection of study population.

Download figure to PowerPoint

Table 1. Baseline characteristics of the study population
 Controls n = 4 003 265Mild psoriasis n = 34 371Severe psoriasis n = 2621 P-value for difference
  1. SD, standard derivation; ACE-I, angiotensin-converting enzyme inhibitor; ARB, angiotensin-2 receptor blocker.

Age, years (SD)47.3 (15.8)47.2 (15.9)46.9 (15.4)<0.001
Men (%)1 942 548 (48.5)16 998 (49.4)1352 (51.6)<0.001
Women (%)2 060 717 (51.5)17 373 (50.6)1269 (48.4) 
Mean follow-up time, years9.25.05.0 
No. of patient-years36 965 324172 22413 146 
Comorbidity (%)
 Congestive heart failure6842 (0.17)35 (0.10)4 (0.15)0.005
 Chronic obstructive pulmonary disease9750 (0.24)45 (0.13)6 (0.23)<0.001
 Cardiac dysrhythmia10 733 (0.27)81 (0.24)10 (0.38)0.63
 Renal disease1871 (0.05)9 (0.03)2 (0.08)0.23
 Cancer22 927 (0.57)158 (0.46)16 (0.61)0.02
 Rheumatological disease3419 (0.09)22 (0.06)3 (0.11)0.34
Treatment (%)
 Platelet inhibitor6842 (0.17)554 (1.61)35 (1.34)0.005
 Beta-blocker114 686 (2.86)1317 (3.83)107 (4.08)<0.001
 ACE-I/ARB90 076 (2.25)991 (2.88)74 (2.82)<0.001
 Vitamin-K antagonist15 024 (0.38)129 (0.38)7 (0.27)0.66
 Loop diuretic97 149 (2.43)710 (2.07)85 (3.24)<0.001
 Statin17 546 (0.44)230 (0.67)17 (0.65)<0.001
 Spironolactone11 694 (0.29)90 (0.26)10 (0.38)0.62
image

Figure 2. Events per 1000 patient-years stratified by age and psoriasis severity.

Download figure to PowerPoint

Multivariate time-dependent Poisson regression analysis

All-cause mortality and cardiovascular death.  As demonstrated in Table 2, psoriasis was associated with increased risk of all-cause mortality and cardiovascular death for both mild and severe disease. The overall (i.e. not stratified for age) RRs for all-cause mortality were 1.16 (95% CI 1.11–1–20) and 1.73 (95% CI 1.54–1.94) for mild and severe psoriasis, respectively. The corresponding risk estimates for cardiovascular death were 1.14 (95% CI 1.06–1.22) and 1.57 (95% CI 1.27–1.94). The risk was generally higher in younger patients with severe disease (Table 2).

Table 2. Rate ratios and 95% confidence intervals from multivariate adjusted time-dependent Poisson analyses
Age groups (years)Mild psoriasisSevere psoriasis
Overall (n = 34 371)18–50 (n = 16 150)51–70 (n = 13 714)>70 (n = 4507)Overall (n = 2621)18–50 (n = 1296)51–70 (n = 1031)>70 (n = 294)
  1. Rate ratio = 1 for controls. Composite endpoint: stroke, myocardial infarction and cardiovascular death; RR, rate ratio; CI, 95% confidence interval.

All-cause mortality1.161.261.231.131.732.872.321.24
RR (CI)(1.11–1.20)(1.08–1.47)(1.15–1.31)(1.08–1.19)(1.54–1.94)(2.04–4.02)(1.96–2.74)(1.05–1.48)
P-value<0.0010.003<0.001<0.001<0.001<0.001<0.0010.01
Cardiovascular death1.1411.21.141.572.982.221.18
RR (CI)(1.06–1.22)(0.66–1.50)(1.05–1.36)(1.06–1.24)(1.27–1.94)(1.32–6.73)(1.59–3.10)(0.89–1.57)
P-value<0.0010.990.010.001<0.0010.001<0.0010.26
Composite endpoint1.21.41.211.161.582.041.851.19
RR (CI)(1.14–1.25)(1.20–1.63)(1.12–1.29)(1.09–1.24)(1.36–1.82)(1.35–3.09)(1.51–2.26)(0.95–1.50)
P-value<0.001<0.001<0.001<0.001<0.0010.001<0.0010.13
Stroke1.251.611.221.151.711.641.871.47
RR (CI)(1.16–1.33)(1.32–1.97)(1.10–1.35)(1.05–1.20)(1.39–2.11)(0.88–3.07)(1.41–2.49)(1.07–1.26)
P-value<0.001<0.001<0.0010.004<0.0010.12<0.0010.02
Myocardial infarction1.221.171.121.31.452.321.441.00
RR (CI)(1.12–1.33)(0.89–1.54)(0.99–1.26)(1.16–1.45)(1.10–1.90)(1.19–4.50)(0.99–2.09)(0.63–1.45)
P-value<0.0010.30.06<0.0010.010.010.050.97
Coronary revascularization1.371.621.261.451.772.271.631.58
RR (CI)(1.26–1.49)(1.26–2.07)(1.13–1.40)(1.24–1.69)(1.35–2.32)(1.17–4.42)(1.16–2.27)(0.92–1.45)
P-value<0.001<0.001<0.001<0.001<0.0010.020.010.10

Composite cardiovascular endpoint.  Mild and severe psoriasis conferred increased risk of the composite of stroke, MI and cardiovascular death (Table 2). For mild psoriasis, the overall RR of the composite endpoint was 1.20 (95% CI 1.14–1.25); RR was further increased to 1.58 (95% CI 1.36–1.82) in patients with severe psoriasis. Age-stratified analyses displayed a consistent relationship between young age and high cardiovascular risk in patients with psoriasis (Table 2).

Stroke, myocardial infarction and coronary revascularization.  As shown in Table 2, the risk of stroke was increased with mild psoriasis and further increased with severe disease: RR was 1.25 (95% CI 1.16–1.33) and 1.71 (95% CI 1.39–2.11) in patients with mild and severe psoriasis, respectively. Mild psoriasis conferred a modest increase in the risk of MI (RR 1.22 [95% CI 1.12–1.33]), and an additional increase in risk was observed for patients with severe disease (RR 1.45 [95% CI 1.10–1.90]). Both mild psoriasis (RR 1.37 [95% CI 1.26–1.49]) and severe psoriasis (RR 1.77 [95% CI 1.35–2.32]) were also associated with increased risk of coronary revascularization (Table 2).

Additional analyses

Cardiovascular risk in severe psoriasis (skin affection alone) versus psoriatic arthritis.  A total of 2014 patients with severe psoriatic skin affection and 607 patients with psoriatic arthritis were compared with regard to cardiovascular risk. There were no significant differences in baseline characteristics (data not shown). The relationship between young age and high cardiovascular risk was maintained in these analyses; furthermore, the risk of all-cause mortality, cardiovascular death and of the composite cardiovascular endpoint was similar in patients with severe skin affection and psoriatic arthritis (Table 3).

Table 3. Rate ratios and 95% confidence intervals from multivariate adjusted time-dependent Poisson analyses stratified for skin vs. joint affection (psoriatic arthritis) in psoriasis
Age groups (years)Severe psoriasis (skin affection alone)Psoriatic arthritisWald Chi-square test for significant differences between overall estimates
Overall18–5051–7051–70Overall18–5051–70>70 P-value
  1. Rate ratio = 1 for controls. Composite endpoint: stroke, myocardial infarction and cardiovascular death; RR, rate ratio; CI, 95% confidence interval.

All-cause mortality1.813.332.591.271.742.231.871.430.79
RR (CI)(1.60–2.05)(2.30–4.84)(2.15–3.12)(1.05–1.54)(1.32–2.30)(1.06–4.69)(1.27–2.74)(0.88–2.34) 
P-value<0.001<0.001<0.001<0.001<0.0010.030.0010.15 
Cardiovascular death1.563.582.181.251.841.872.681.190.55
RR (CI)(1.22–1.98)(1.47–8.77)(1.45–3.26)(0.91–1.72)(1.11–3.06)(0.26–13.3)(1.40–5.16)(0.49–2.85) 
P-value<0.0010.01<0.0010.160.020.530.0030.7 
Composite endpoint1.561.771.931.241.793.271.791.200.44
RR (CI)(1.32–1.84)(1.04–3.00)(1.52–2.47)(0.96–1.60)(1.31–2.45)(1.70–6.31)(1.17–2.75)(0.62–2.30) 
P-value<0.0010.04<0.0010.1<0.001<0.0010.010.59 

Cardiovascular risk in severe psoriasis (skin affection alone) and psoriatic arthritis versus DM.  Diabetes mellitus was identified in a total of 127 449 patients. These patients were generally older (mean age 55.6 years) at baseline than those with psoriasis, and there was a slightly higher percentage of men than women (54.1%). Age-stratified event rates were comparable between patients with severe psoriasis, psoriatic arthritis and DM (data not shown). The overall RR for cardiovascular death in patients with DM was 1.59 (95% CI 1.56–1.63), which was comparable to that of patients with severe psoriasis (skin affection alone) (P = 0.72) or psoriatic arthritis (P = 0.34). For the composite cardiovascular endpoint, DM conferred an increased risk (RR 1.58; 95% CI 1.56–1.61), which was similar to the risk observed in severe psoriasis (P = 0.68) and psoriatic arthritis (P = 0.46). Comparable results were obtained for the risk of all-cause mortality (data not shown).

Sensitivity analyses

Cohort matched for patient age and gender.  The entire psoriasis population comprising 34 371 patients with mild psoriasis and 2621 patients with severe psoriasis was compared with 147 968 age- and gender-matched controls from the original nationwide cohort. In multivariate Poisson regression models, adjusted for updated baseline characteristics obtained at the time of psoriasis diagnosis for cases, and at the same time for corresponding controls, the risks of all-cause mortality and all cardiovascular endpoints were similar to the risks obtained in the primary analysis.

Propensity score-matched Poisson regression analyses.  A total of 34 371 patients with mild psoriasis and 2621 patients with severe psoriasis were matched on propensity score with 36 992 controls to establish study populations with similar baseline characteristics (Table 4). The C-statistic was 0.69, indicating a good discriminative power of the model. The analyses confirmed the results obtained with the primary adjusted time-dependent Poisson regression models and provided similar risk estimates for all study endpoints. The dose–response relationship between psoriasis severity and risk of adverse endpoints was maintained in these analyses.

Table 4. Baseline characteristics of the propensity score-matched population
 Controls n = 36 992Mild psoriasis n = 34 371Severe psoriasis n = 2621 P-value for difference
  1. SD, standard derivation; ACE-I, angiotensin-converting enzyme inhibitor; ARB, angiotensin-2 receptor blocker.

Age, years (SD)47.3 (15.8)47.2 (15.9)46.9( 15.4)0.82
Men (%)18 331 (49.6)16 998 (49.4)1352 (51.6)0.11
Women (%)18 631 (50.4)17 373 (50.6)1269 (48.4) 
Comorbidity (%)
 Congestive heart failure33 (0.09)35 (0.10)4 (0.15)0.35
 Chronic obstructive pulmonary disease66 (0.18)45 (0.13)6 (0.23)0.36
 Cardiac dysrhythmia69 (0.19)81 (0.24)10 (0.38)0.07
 Renal disease10 (0.03)9 (0.03)2 (0.08)0.50
 Cancer183 (0.49)158 (0.46)16 (0.61)0.91
 Rheumatological disease24 (0.06)22 (0.06)3 (0.11)0.62
Treatment (%)
 Platelet inhibitor586 (1.58)554 (1.61)35 (1.34)0.79
 Beta-blocker1433 (3.87)1317 (3.83)107 (4.08)0.80
 ACE-I/ARB1069 (2.89)991 (2.88)74 (2.82)0.98
 Vitamin-K antagonist119 (0.32)129 (0.38)7 (0.27)0.47
 Loop diuretic810 (2.19)710 (2.07)85 (3.24)0.36
 Statin269 (0.73)230 (0.67)17 (0.65)0.33
 Spironolactone92 (0.25)90 (0.26)10 (0.38)0.43

Cohort analyses accounting for baseline healthcare consumption and surveillance bias.  When identifying patients with psoriasis at their first prescription claim or first psoriasis diagnosis, a total of 49 504 patients with mild psoriasis and 5455 with severe psoriasis were eligible for analysis and a further 26 507 patients with prevalent psoriasis were excluded; this corresponded to a 10-year prevalence of psoriasis in Denmark of 1.8%. The risks estimates of all examined endpoints were comparable to the risks estimates obtained from the primary analyses, and in addition, the exclusion of participants with a history of hospitalizations and/or prescription claims prior to inclusion did not alter the results significantly (data not shown).

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Funding source
  9. Disclosures
  10. References

In this nationwide cohort study, psoriasis was a significant cardiovascular risk factor independent of age, gender, comorbidity, concomitant medication and socio-economic status. The risk of cardiovascular events and mortality was age-dependent and generally higher amongst younger patients with severe psoriasis. The results was corroborated by propensity score-matched analyses, additional sensitivity analyses, and by demonstration of a consistent dose–response relationship between psoriasis severity and risk of adverse endpoints, including all-cause mortality, cardiovascular death, MI, coronary revascularization, stroke and the composite of MI, stroke and cardiovascular death. The increased risk of all cardiovascular endpoints and all-cause mortality in patients with severe psoriasis, including psoriatic arthritis was generally comparable to the magnitude of cardiovascular risk reported in patients with rheumatoid arthritis [26, 27]. Furthermore, we demonstrated that the cardiovascular risk increase in patients with severe psoriasis is similar to that of patients with DM, suggesting that psoriasis is a significant contributor to cardiovascular disease.

Atherosclerosis and psoriasis are chronic inflammatory diseases characterized by Th1-/Th17-driven inflammation and reduced Treg activity, and there is a remarkable overlap of identified inflammatory markers and mediators in target lesions (i.e. atherosclerotic plaques and cutaneous psoriasis elements, respectively) [1–4, 11]. Psoriasis and psoriatic arthritis have been associated with conventional cardiovascular risk factors and surrogate markers of atherothrombotic disease, including endothelial dysfunction, carotid artery intima–media thickening and coronary artery calcification, as well as overt cardiovascular disease and death [3–10, 12–18]. Moreover, systemic anti-inflammatory therapy with methrotrexate may decrease cardiovascular risk in patients with psoriasis, and preliminary observations that statins can exert favourable effects on psoriasis activity also suggest that coincident pathological mechanisms contribute to the clinical manifestations of psoriasis and atherothrombosis [19–21, 35].

From studies of healthcare databases, such as the United Kingdom General Practice Research Database and US Veterans Affairs institutions, psoriasis has previously been linked to risk of cardiovascular disease independent of conventional risk factors [6–10, 12–15]. However, these results were recently questioned. It was suggested that increased prevalence of comorbidities (including previous cardiovascular disease) and surveillance bias may have contributed to earlier findings, in that patients with psoriasis were more likely to visit their physician and were therefore at increased risk of being diagnosed with cardiovascular diseases than the general population [22–24]. Indeed, a recent expert report has called for more definitive data from nationwide controlled registries with epidemiological follow-up of patients with different grades of psoriasis [25]. Our results based on an unselected nationwide cohort of patients with psoriasis and use of validated cardiovascular endpoints, however, are consistent with the available epidemiological data linking psoriasis to augmented risk of cardiovascular disease, and the magnitude of risk was comparable to previously reported data [6–10, 12–15]. Of importance, significant surveillance bias is unlikely due to the use of nationwide registries of hospitalizations and claimed drug prescriptions in Denmark where healthcare is essentially free of charge. The proportion of undiagnosed psoriasis cases in our study population is also likely to be small compared to the USA where undiagnosed cases are frequent and subject to important disparities (e.g. more often male, non-white and less educated) compared to diagnosed patients [36]. Our results were robust to adjustments for a range of relevant baseline characteristics, and because of the conservative risk–time calculation of the analysis, we may even have underestimated event rates in patients with psoriasis and overestimated event rates in controls, leading to attenuation of the true psoriasis-related risk of cardiovascular events. The risk was highest in younger patients possibly mirroring, in part, the age-dependent increase in overall cardiovascular risk, with attenuation of the psoriasis-related risk in the elderly.

As well as mechanistic correlates between psoriasis and atherothrombosis in terms of shared inflammatory processes, the probability of a causal relation between the two is increased by the consistent dose–response relationship between psoriasis severity and all examined cardiovascular endpoints, which is in agreement with earlier data on psoriasis-associated risk of MI, stroke and cardiovascular mortality [7, 10, 12–15]. The psoriasis severity-dependent increased risk of undergoing coronary revascularization has not been reported before and supports a disease-associated increased risk of clinically significant coronary artery disease.

Like psoriasis, rheumatoid arthritis is a chronic inflammatory disease that carries increased risk of cardiovascular disease, and data on cardiovascular disease in these two conditions are remarkably similar [1–4, 37, 38]. Indeed, rheumatoid arthritis has been found to confer an approximately twofold increase in cardiovascular risk compared to healthy controls, and the current evidence therefore suggests that severe psoriasis including psoriatic arthritis is associated with a risk comparable to that of rheumatoid arthritis [38, 39]. Specifically, the recent observation that rheumatoid arthritis is equal to DM as an independent cardiovascular risk factor has made a considerable impact on the perception of cardiovascular risk management in patients with rheumatoid arthritis [26, 27]. From this perspective, we compared the risk of cardiovascular events for patients with psoriasis and those with DM and found comparable risks for all examined endpoints, including all-cause mortality, cardiovascular death and the composite of MI, stroke and cardiovascular death, for DM and severe psoriasis. As it is generally accepted that there is no correlation between the severity of psoriasis skin affection and the presence of psoriatic arthritis, we further identified patients with psoriatic arthritis and found similar risk in these subjects and in patients with severe skin affection alone. Such comparison has not, to our knowledge, been presented previously, and the comparable risk in these two patient subsets supports the clinically accepted paradigm that psoriatic arthritis represents a severe form of the disease. Comparison of the magnitude of psoriasis-associated risk with that of established cardiovascular risk factors has been reported previously [7, 10, 12, 15]. However, our study is the first to provide a direct cohort-based comparison with epidemiological follow-up between psoriasis and an established cardiovascular risk factor. The current demonstration that the cardiovascular risk from severe psoriasis including psoriatic arthritis is of similar magnitude to that of DM therefore markedly reinforces the evolving view that cardiovascular risk factors in psoriasis should probably be treated as aggressively as in rheumatoid arthritis and DM. At present, annual cardiovascular risk assessments using established guidelines are recommended for patients with rheumatoid arthritis and psoriatic arthritis and, in addition to lifestyle interventions, these patients should be considered for statin therapy earlier than predicted by ordinary risk score models [39, 40]. Based on the data presented here, we believe that similar considerations could apply for patients with psoriasis with severe skin affection, and possibly for all patients with psoriasis.

Some further strengths and limitations of the study should be mentioned. By including the entire Danish population aged ≥18 years, we avoided selection bias related to for example gender, age, socio-economic status and labour market participation. All medications examined in the current study were dispensed only on prescription in Denmark, and the National Prescription Register, which is directly linked to the system for reimbursement of drug expenses, has been shown to be accurate [41]. We were unable to identify patients with psoriasis treated with topical corticosteroids alone (because of the broad indications for these agents) or patients who had never received vitamin-D derivates or hospital-based treatment. Topical treatment (i.e. vitamin-D derivates and corticosteroids) is generally used as first-line treatment in Denmark, but our results cannot be directly extrapolated to patients with psoriasis treated with topical corticosteroids alone. The use of hospitalizations to define severe psoriasis may have increased the presence of cardiometabolic comorbidities and decreased the threshold for detection of study outcomes in these patients (surveillance bias). Therefore, we made efforts to reduce these sources of bias by making multiple adjustments (e.g. for the Charlton comorbidity index and concomitant medication) and by performing sensitivity analyses. As we used claimed prescriptions for glucose-lowering drugs to define DM, we were unable to identify patients with DM who received treatment with dietary modification alone. Moreover, the registries employed in the study do not allow for identification of some cardiovascular risk factors (e.g. obesity and smoking), leaving room for residual confounding. An unmeasured confounder, however, must be prevalent and carry a very high risk to nullify the findings, for example for the composite endpoint in patients aged 46–55 years with severe psoriasis. We estimated that such a confounder should have a prevalence of 20% and increase RR by a factor of >7, and comparable estimates for hypothetical ‘rule out’ confounders were apparent for all endpoints and patient age-strata (data not shown), rendering its existence highly unlikely [34]. Finally, the Danish population is predominantly of Caucasian descent, and extrapolation of results to patients of other ethnicities should be carried out with caution.

Conclusion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Funding source
  9. Disclosures
  10. References

The results of this nationwide study of the cardiovascular risk associated with psoriasis confirm and extend previous results in this area of research by demonstrating that the disease is a clinically significant risk factor for all-cause mortality, cardiovascular death, MI, coronary revascularization, stroke and the composite of MI, stroke and cardiovascular death. The cardiovascular risk was independent of measured confounders and in patients with severe psoriasis including psoriatic arthritis was comparable to that of patients with DM. The results call for an increased awareness of the association between psoriasis and cardiovascular morbidity and mortality.

Funding source

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Funding source
  9. Disclosures
  10. References

This study was supported by the Department of Cardiology, Copenhagen University Hospital Gentofte, DK-2900, Denmark.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Funding source
  9. Disclosures
  10. References