Cardiovascular disease event rates in patients with severe psoriasis treated with systemic anti-inflammatory drugs: a Danish real-world cohort study
Psoriasis is a chronic inflammatory disorder associated with cardiovascular morbidity and mortality. Systemic anti-inflammatory drugs, including biological agents, are widely used in the treatment of patients with moderate to severe psoriasis and may attenuate the risk of cardiovascular disease events. We therefore examined the rate of cardiovascular disease events in patients with severe psoriasis treated with systemic anti-inflammatory drugs.
Design, setting and participants
Individual-level linkage of nationwide administrative databases was used to assess the event rates associated with use of biological agents, methotrexate or other therapies, including retinoids, cyclosporine and phototherapy, in Denmark from 2007 to 2009.
Main outcome measure
Death, myocardial infarction and stroke.
A total of 2400 patients with severe psoriasis, including 693 patients treated with biological agents and 799 treated with methotrexate, were identified. Incidence rates per 1000 patient-years and 95% confidence intervals (CIs) for the composite endpoint were 6.0 (95% CI 2.7–13.4), 17.3 (95% CI 12.3–24.3) and 44.5 (95% CI 34.6–57.0) for patients treated with biological agents, methotrexate and other therapies, respectively. Age- and sex-adjusted hazard ratios (HRs) were 0.28 (95% CI 0.12–0.64) and 0.65 (95% CI 0.42–1.00) for patients treated with biological agents and methotrexate, respectively, using other therapies as the reference cohort. Corresponding HRs for a secondary composite endpoint of cardiovascular death, myocardial infarction and stroke were 0.48 (95% CI 0.17–1.38) and 0.50 (95% CI 0.26–0.97).
In this nationwide study of patients with severe psoriasis, systemic anti-inflammatory treatment with biological agents or methotrexate was associated with lower cardiovascular disease event rates compared to patients treated with other anti-psoriatic therapies.
Inflammation plays a pivotal role in atherosclerosis, and targeted anti-inflammatory therapy holds considerable promise in the treatment of atherosclerotic disease [1-4]. Psoriasis is also a common chronic inflammatory disease which affects approximately 125 million people worldwide . Like other inflammatory diseases, such as rheumatoid arthritis, psoriasis is associated with increased risk of cardiovascular mortality and morbidity possibly due, in part, to shared immune–inflammatory mechanisms and higher prevalence of classical cardiovascular disease risk factors in patients with psoriasis [6-9]. Systemic anti-inflammatory therapy with, for example, methotrexate and biological agents is well established in the treatment of patients with moderate to severe psoriasis . Although evidence suggests a beneficial effect of systemic anti-inflammatory treatments on the risk of cardiovascular disease, conflicting results have been reported and limited data are available in patients with psoriasis [11-19]. The objective of this study was therefore to investigate rates of cardiovascular disease events in patients with severe psoriasis treated with biological agents, methotrexate and other nonbiological systemic therapies in a real-world setting with use of Danish nationwide registries.
The study was a retrospective longitudinal cohort study using Danish nationwide prospectively recorded registries. The study was completed and reported in accordance with the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) recommendations .
The study was approved by The Danish Data Protection Agency and the DERMBIO (see below) steering committee. Data at the individual case level were made available by the national registers in an anonymized form. Registry studies do not require ethical approval in Denmark.
Data sources and study participants
The Danish Civil Registration System, where a unique personal identifier is assigned to all Danes at birth and residents at immigration, allows for linkage of information on an individual level across prospectively recorded registries. This study population comprised all subjects from the Danish population identified with psoriasis [International Classification of Diseases, 10th revision (ICD-10) L40] classified as severe disease (primary diagnosis) requiring hospitalization or outpatient hospital care between 2007 and 2009 [21, 22]. Patients treated with biological agents were identified in a mandatory (since 2007) nationwide Danish registry (DERMBIO) . Patients who required hospitalization or outpatient hospital visits for psoriasis prior to the start of the study were excluded, to ensure homogenous follow-up of patients in the treatment groups. Patients with severe psoriasis were divided into three treatment groups: (i) biological agents, (ii) methotrexate and (iii) other therapies, including retinoids, cyclosporine, phototherapy and/or climate therapy. Patients were followed until 31 December 2009, emigration or the occurrence of an end-point. Data on all medications dispensed from pharmacies were obtained from the Danish Registry of Medicinal Product Statistics, in which all dispensed prescriptions since 1995 are registered. Morbidity data were obtained from the Danish National Patient Register in which hospital admissions, procedures and diagnoses have been recorded since 1978 using ICD codes. Cause of death data was obtained from the National Causes of Death Register, in which causes of death are recorded using ICD codes. Socio-economic status was defined as average yearly income in a period of up to 5 years prior to inclusion and divided into quintiles.
Medical treatment and comorbidity
Pharmacological treatment for cardiovascular diseases and cardiovascular disease risk factors, including medically managed hypertension, dyslipidaemia and diabetes mellitus, was identified by prescriptions for platelet inhibitors [anatomical therapeutic chemical classification (ATC) code: B01AC],vitamin K antagonists (B01AA), beta-blockers (C07), angiotensin-converting enzyme inhibitors/angiotensin 2 receptor antagonists (C09), calcium antagonists (C08), loop diuretics (C03C), thiazide diuretics (C03A), spironolactone (C03D), cholesterol-lowering drugs (C10A) and glucose-lowering drugs (A10). Treatment with biological agents approved for psoriasis [L04AB, L04AC (from DERMBIO)], methotrexate (L01BA01, L04AX03), retinoids (D05BB) and cyclosporine (L04AD01) was also recorded, as was the use of nonsteroidal anti-inflammatory drugs (NSAIDs; MA01A) and antidepressants (N06A). The use of dermatological phototherapy (procedural code: BNG) was also recorded. Comorbidity was defined by hospitalization up to 12 months prior to inclusion according to the previously validated Charlson comorbidity index as adapted to ICD-10 [24, 25]. Information was linked on an individual level across nationwide registries, including information on use of biological agents, prescription claims, inpatient and outpatient visits and procedures.
The primary study end-point was a composite of death, myocardial infarction (ICD-10 I21 and I22) and stroke (ICD-10 I60, I61, I63 and I64). A cardiovascular end-point of cardiovascular death, myocardial infarction and stroke was examined as the secondary end-point. Previous studies have demonstrated that the diagnoses of myocardial infarction and stroke in the Danish registries are valid [26, 27].
Baseline characteristics were summarized and presented as means and numbers and percentages. Differences in baseline characteristics were assessed by chi-squared test for categorical covariates and either two-sided t-test or analyses of variance for continuous variables, as appropriate. Incidence rates (IRs) for study end-points were reported as events per 1000 patient-years. To ensure accurate allocation of the time at risk and account for changes in treatment, information on systemic antipsoriatic therapy according to the three treatment groups was included as time-dependent variables. Cox regression models controlling for confounding factors including patient age, sex, baseline use of medication, comorbidities and socio-economic status were used to estimate hazard ratios (HRs) and corresponding 95% confidence intervals (CIs).
A sensitivity analysis with exclusion of subjects with a history of hospitalization and/or use of cardiovascular pharmacotherapy up to 12 months prior to inclusion in the study was made to address potential selection bias related to differences in comorbidity between patients with severe psoriasis treated with biological agents or other therapies. Model assumptions, including absence of interaction between model covariates and the proportional hazards assumption, were tested and found to be valid. Appropriateness of the statistical models was tested by the goodness of fit test. A two-sided P < 0.05 was considered statistically significant. All statistical analyses were performed with sas statistical software version 9.2 (SAS Institute Inc., Cary, NC, USA) or stata software version 11 (Statacorp, College St, TX, USA).
A total of 2400 patients with severe psoriasis, including 693 patients treated with biological agents and 799 treated with methotrexate, were identified. Baseline characteristics of the study population are presented in Table 1. Patients treated with biological agents were younger at baseline and included a higher percentage of men compared with those treated with nonbiological therapies. Amongst patients treated with biological agents, the majority (>80%) received tumour necrosis factor α inhibitors, and no events were related to use of interleukin (IL)-12/IL-23 inhibitors. In general, patients treated with biological agents had less comorbidity and were less often treated with cardiovascular pharmacotherapy, NSAIDs or antidepressants. The maximum follow-up was 3 years with a mean follow-up of approximately 18 months. There were no significant differences in follow-up time between groups and no losses to follow-up.
Table 1. Baseline characteristics of the study population
|Women (%)||238 (34.3)||436 (54.6)||461 (50.8)||<0.001|
|Mean follow-up time, years||1.4||1.5||1.4||0.87|
|Number of patient-years||1001.4||1238.1||1328.4|| |
|Mean Charlson comorbidity index||0.06||0.21||0.29||<0.001|
|Myocardial infarction||1 (0.1)||2 (0.25)||5 (0.6)||0.33|
|Cerebrovascular disease||0 (0.0)||17 (2.1)||24 (2.6)||<0.001|
|Peripheral atherosclerosis||3 (0.4)||4 (0.5)||6 (0.7)||0.81|
|Congestive heart failure||2 (0.3)||5 (0.6)||11 (1.2)||0.09|
|Chronic pulmonary disease||2 (0.3)||15 (1.9)||30 (3.3)||<0.001|
|DM with chronic complications||3 (0.4)||10 (1.3)||16 (1.8)||0.054|
|Renal disease||3 (0.4)||5 (0.6)||10 (1.1)||0.27|
|Peptic ulcer||4 (0.9)||8 (1.0)||9 (1.0)||0.61|
|Pharmacological treatment (%)|
|NSAID||53 (7.7)||129 (16.2)||84 (9.3)||<0.001|
|Antidepressant||75 (10.8)||110 (13.8)||131 (14.4)||0.09|
|Platelet inhibitor||44 (6.4)||83 (10.4)||134 (14.8)||<0.001|
|Beta-blocker||36 (5.2)||77 (9.6)||109 (12.0)||<0.001|
|ACE-I/ARB||97 (14.0)||127 (15.9)||180 (19.8)||0.01|
|Calcium antagonist||53 (7.7)||75 (9.4)||98 (10.8)||0.10|
|Vitamin K antagonist||4 (0.6)||22 (2.8)||25 (2.8)||0.004|
|Thiazide diuretic||36 (5.2)||83 (10.4)||88 (9.7)||0.001|
|Loop diuretic||24 (3.5)||53 (6.6)||85 (9.4)||<0.001|
|Spironolactone||9 (1.3)||9 (1.1)||22 (2.4)||0.08|
|Statin||76 (11.0)||112 (14.0)||133 (14.7)||0.08|
|Glucose-lowering drug||40 (5.8)||60 (7.5)||70 (7.7)||0.28|
Primary composite end-point
A total of 105 individual primary outcome events were recorded, and of these, six and 33 events occurred in patients treated with biological agents and methotrexate, respectively. IRs per 1000 patient-years for the composite end-point of death, myocardial infarction and stroke were 6.0 (95% CI 2.7–13.4), 17.3 (95% CI 12.3–24.3) and 44.5 (95% CI 34.6–57.0) for patients treated with biological agents, methotrexate and other therapies, respectively. HRs adjusted for age and sex were 0.28 (95% CI 0.12–0.64) and 0.65 (95% CI 0.42–1.00) for patients treated with biological agents and methotrexate, respectively, with other therapies as the reference cohort. For the subset of patients treated with both methotrexate and biological agents, the risk reduction was similar to that seen in patients treated with biological agents alone. An overview of the IR values and the crude and adjusted HRs is presented in Table 2.
Table 2. Incidence rates and hazard ratios (HRs) of events
|Death, myocardial infarction and stroke|
|Incidence rate (95% CI)||6.0 (2.7–13.4)||17.3 (12.3–24.3)||44.5 (34.7–57.0)|
|HR (95% CI)||0.20 (0.09–0.47)||0.57 (0.37–0.87)||Ref.|
|HR adjusted for age and sex (95% CI)||0.28 (0.12–0.64)||0.65 (0.42–1.00)||Ref.|
|HR adjusted for age, sex, medication, comorbidity and socio-economic status (95% CI)||0.29 (0.12–0.68)||0.73 (0.47–1.14)||Ref.|
|Cardiovascular death, myocardial infarction and stroke|
|Incidence rate (95% CI)||4.0 (1.5–10.7)||6.8 (3.9–11.7)||22.2 (15.6–31.6)|
|HR (95% CI)||0.32 (0.11–0.92)||0.41 (0.21–0.80)||Ref.|
|HR adjusted for age and sex (95% CI)||0.48 (0.17–1.38)||0.50 (0.26–0.97)||Ref.|
|HR adjusted for age, sex, medication, comorbidity and socio-economic status (95% CI)||0.52 (0.17–1.53)||0.58 (0.29–1.15)||Ref.|
Secondary composite end-point
A total of 54 individual secondary outcome events were recorded in the study period, including four and 15 events in patients treated with biological agents and methotrexate, respectively. Systemic anti-inflammatory treatment with biological agents or methotrexate was associated with an attenuated risk of adverse cardiovascular disease events. IR values per 1000 patient-years for the composite of cardiovascular death, myocardial infarction and stroke were 4.0 (95% CI 1.5–10.7), 6.8 (95% CI 3.9–11.7) and 22.2 (95% CI 15.6–31.6) for patients treated with biological agents, methotrexate and other therapies, respectively. HRs were 0.48 (95% CI 0.17–1.38) and 0.50 (95% CI 0.26–0.97) for patients treated with biological agents and methotrexate, respectively. In contrast to the primary analysis, patients treated with both methotrexate and biological agents had the lowest risk estimate (HR 0.30; 95% CI 0.07–1.26).
The lower risk associated with the use of biological agents and methotrexate was confirmed after exclusion of subjects with a history of hospitalization and/or cardiovascular drug use. In this analysis, IRs for the primary composite end-point were 2.6 (95% CI 0.65–10.4), 3.6 (95% CI 1.5–8.7) and 23.9 (95% CI 15.6–36.7) for patients treated with biological agents, methotrexate and other therapies, respectively. HRs were 0.17 (95% CI 0.03–0.91) and 0.30 (95% CI 0.11–0.81) for patients treated with biological agents and methotrexate, respectively.
In this nationwide study of patients with severe psoriasis treated with systemic anti-inflammatory therapies, the use of biological agents and methotrexate was associated with lower rates of death, myocardial infarction and stroke compared with patients treated with other therapies.
Psoriasis and atherosclerosis are chronic inflammatory diseases with a considerable overlap of inflammatory mechanisms [8, 9], and psoriasis is associated with cardiovascular morbidity and mortality [21, 22, 28-34]. In theory, anti-inflammatory therapy could influence several important steps in atherogenesis and atherothrombosis from insulin resistance, endothelial dysfunction and fatty streak formation to plaque rupture and thrombus formation [35-37]. Indeed, anti-inflammatory treatment, for example, with very low doses of methotrexate for patients with atherosclerotic disease has been proposed and could have a major clinical impact . Accordingly, there has been considerable interest in the cardiovascular effects of systemic anti-inflammatory treatment in patients with other chronic inflammatory diseases. Evidence has suggested beneficial effects of systemic anti-inflammatory treatments including methotrexate and biological agents on the risk of cardiovascular disease, for example, in patients with rheumatoid arthritis, but conflicting results have been reported and data on patients with psoriasis are generally lacking [11-19]. Also, the recent withdrawal of the application for market approval of the IL-12/IL-23 inhibitor briakinumab, presumably due to increased occurrence of cardiovascular disease events, and a general lack of postmarketing cardiovascular safety data with biological agents have led to concern and debate over the impact on cardiovascular outcomes of these very expensive therapies [38, 39]. In a recent meta-analysis, no significant differences in cardiovascular disease event rates were demonstrated between patients treated with biological agents, including IL-12/IL-23 inhibitors, and those receiving placebo. However, the conclusions of this study were limited by the absence of access to individual-level data, low event rates and a very short follow-up period . Recently, in a study to compare cardiovascular outcomes in patients with psoriasis, no overall beneficial role of systemic anti-inflammatory treatment was demonstrated, but there was a trend towards a beneficial effect in younger patients . By contrast, a recently published meta-analysis of the effects of methotrexate in chronic inflammatory disorders, including psoriasis, showed that this agent was associated with lower risk of cardiovascular disease events compared with other treatments . Furthermore, treatment with tumour necrosis factor-α inhibitors may ameliorate the risk of diabetes mellitus and markers of cardiovascular disease, such as endothelial dysfunction, compared with nonbiological systemic anti-inflammatory drugs in patients with psoriasis, psoriatic arthritis or rheumatoid arthritis [16, 40, 41].
Our results indicate that treatment with biological agents or methotrexate is associated with reduced risk of death and cardiovascular disease events in patients with severe psoriasis. It is interesting that the demonstration of an additional beneficial effect of combining biological agents and methotrexate on the risk of cardiovascular disease events (secondary study end-point) appears to support the notion that systemic inflammation is an important link between psoriasis and cardiovascular disease. The results, therefore, also support the hypothesis that targeted anti-inflammatory treatment may be a viable option to reduce cardiovascular morbidity and mortality in patients with chronic low-grade inflammation in the absence of evident disease [1-4]. It could be argued that the present results were driven by selection bias as patients selected for treatment with biological agents were younger and had less comorbidity. Indeed, initially biological agents were rarely used for treatment of psoriatic patients with cardiovascular disease. However, the results were not qualitatively affected by multivariable adjustment for baseline comorbidity, use of various medications and socio-economic status. More importantly, a sensitivity analysis with exclusion of patients with a history of hospitalization and/or cardiovascular drug use did not significantly alter the results. Whilst the results of randomized trials are awaited to draw definite conclusions regarding the cardiovascular safety and efficacy of targeted anti-inflammatory treatment, analyses using real-world databases can provide important insight.
The following limitations should be acknowledged. The Danish population predominantly comprises individuals of Caucasian descent, and the results should, therefore, be extrapolated to other ethnic groups with caution. The registries examined do not hold information on the rationale for any given treatment, and thus, bias related to confounding by indication cannot be excluded. In addition, these registries do not hold information regarding some potentially important confounders, such as obesity, blood pressure, lipid levels, plasma glucose levels, circulating inflammatory markers and smoking. Also, our study was not powered to address the impact of individual biological agents, and the relatively short follow-up period and low numbers of events are important limitations. As discussed in detail above, selection bias may have affected the results. However, effort was made to address this important source of bias, and the sensitivity analysis supports the integrity of our results.
Important strengths of the study include the completeness of follow-up, the nationwide coverage, the contemporary real-world clinical setting and the use of validated information on exposure, comorbidity and outcome measures. In addition, the use of nationwide registries of hospitalization data and prescription claims from all pharmacies in Denmark minimized selection bias related to sex, age, socio-economic status, healthcare insurance and employment status.
In conclusion, in this nationwide study of patients with severe psoriasis, systemic anti-inflammatory treatment with biological agents or methotrexate was associated with reduced cardiovascular disease event rates. Further studies are needed to confirm these findings and evaluate their clinical implications.
Conflict of interest statement
The sponsors had no influence on data collection, no access to the data and no influence on research originating from the database. The database is administered by an independent DERMBIO steering committee of specialists in dermatology representing hospital dermatology departments and private practices. OA, LS, RG, LI and TND have received consultancy and/or speaker honoraria from Abbott, Pfizer and Janssen-Cilag; in addition, LS has received honoraria from MSD and Leo Pharma. RG from MSD and Cellgene. LI from MSD and TND from LEO Pharma.
The study was financially supported by unrestricted grants from the Axel Muusfeldts Foundation, DERMBIO and the Danish Psoriasis Association. DERMBIO database maintenance expenditure is covered by unrestricted equal grants from the manufacturers of biological agents (Abbott, Janssen-Cilag, Merck-Serono [2007–2009], MSD and Pfizer). Data were extracted from the DERMBIO database by Mikkel Abildtoft, Zitelab Aps.