Prognosis following first-time myocardial infarction in patients with psoriasis: a Danish nationwide cohort study

Authors


Ole Ahlehoff MD, Research fellow, Copenhagen University Hospital, Gentofte, Post 67, Niels Andersens Vej 65, 2900 Hellerup, Denmark. (fax: (+45)70201281; e-mail: olahha01@geh.regionh.dk).

Abstract

Abstract.  Ahlehoff O, Gislason GH, Lindhardsen J, Olesen JB, Charlot M, Skov L, Torp-Pedersen C, Hansen PR. (Copenhagen University Hospital Gentofte, Hellerup; Copenhagen University Hospital Roskilde, Roskilde, Denmark). Prognosis following first-time myocardial infarction in patients with psoriasis: a Danish nationwide cohort study. J Intern Med 2011; 270: 237–244.

Objectives.  The magnitude of cardiovascular risk associated with psoriasis has been debated and the prognostic impact of psoriasis following myocardial infarction (MI) is unknown. Therefore, we investigated the risk of mortality and adverse cardiovascular events in patients with psoriasis following first-time MI.

Design, setting and participants.  Cohort study of the entire Danish population including all individuals who experienced first-time MI during the period 2002–2006. Multivariable Cox regression models were used to assess the post-MI prognostic impact of psoriasis.

Main outcome measures.  All-cause mortality and a composite cardiovascular end-point of recurrent MI, stroke and cardiovascular death.

Results.  A total of 462 patients with psoriasis and 48 935 controls (mean age 69.5 and 70.6 years, respectively) were identified with first-time MI during the study period. The mean follow-up was 19.5 months [standard deviation (SD) 16.5] for patients with psoriasis and 22 .0 months (SD 18.7) for those without psoriasis. Incidence rates (IRs) per 1000 patient-years for all-cause mortality were 119.4 [95% confidence interval (CI) 117.2–138.3] and 138.3 (95% CI 114.1–167.7) for patients without and with psoriasis, respectively, and the adjusted hazard ratio (HR) associated with psoriasis was 1.18 (95% CI 0.97–1.43). For the composite end-point, the IRs were 149.7 (95% CI 147.1–152.4) and 185.6 (95% CI 155.8–221.0) for patients without and with psoriasis, respectively, with an HR of 1.26 (95% CI 1.04–1.54) for patients with psoriasis.

Conclusion.  This first study of the impact of psoriasis on prognosis after first-time MI indicated a significantly impaired prognosis in patients with psoriasis. Further studies of this novel association are warranted.

Introduction

Psoriasis, like atherosclerosis, is a common chronic inflammatory disease, and recent studies have suggested that psoriasis is an independent risk factor for adverse cardiovascular events, such as myocardial infarction (MI), possibly because of coincident inflammatory mechanisms [1–9]. The impact of psoriasis on the prognosis after MI is unknown, and such information could impact future secondary prevention strategies in patients with this condition. Furthermore, results implicating psoriasis as a cardiovascular risk factor have been conflicting, and surveillance bias because of increased healthcare consumption in patients with psoriasis has been proposed to contribute to the observed increase in risk of atherothrombotic events associated with psoriasis [10–12]. We therefore used the Danish nationwide prospectively recorded registers to examine the psoriasis-related risk of all-cause mortality and cardiovascular morbidity in patients with a first-time MI.

Methods

Population and data sources

The present study population comprised all individuals amongst the entire Danish population who experienced a first-time MI during the period 2002–2006. Patients were followed until 31 December 2006, death, emigration or an event. The study was conducted and reported in accordance with the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) recommendations [10]. In Denmark, all citizens have a unique personal civil registration number that enables individual-level linkage of information across nationwide prospectively recorded registers. Morbidity was obtained from the Danish National Patient Register in which hospital admissions and diagnoses have been recorded since 1978 using the 10th revision of the International Classification of Diseases (ICD) codes (ICD-10). All medications dispensed from pharmacies were obtained from the national prescription registry (the Danish Registry of Medicinal Product Statistics), in which all dispensed prescriptions since 1995 are registered. Deaths were identified from the Central Population Register, in which all deaths are recorded within 2 weeks. Causes of death were obtained from the National Causes of Death Register recorded using ICD codes. Socioeconomic status was defined as average yearly income in the 5 years prior to inclusion, and divided into quintiles. Prescriptions claimed for topical treatment used exclusively for psoriasis, i.e. topical vitamin D derivatives (ATC D05AX, between 1995 and the index MI), and hospitalizations (between 1995 and the index MI) for psoriasis (ICD-10 L40) or psoriatic arthritis (M070-M073) were used for identification of subjects with psoriasis, as previously described [9]. According to protocol, patients were classified as having psoriasis at the time of their second prescription claim to ensure persistent medical treatment or by their third (if not included by the vitamin D criteria) psoriasis-related diagnosis. Topical treatment, including with vitamin D derivatives, is used as first-line therapy for psoriasis. Data on death, comorbidity, concomitant medication and socioeconomic status were linked at an individual level.

Medical treatment, comorbidity and invasive coronary revascularization

Pharmacologically managed risk factors for cardiovascular disease, including hypertension, dyslipidaemia and diabetes mellitus were identified by prescriptions for platelet inhibitors (B01AC), beta-blockers (C07), angiotensin-converting enzyme (ACE) inhibitors/angiotensin 2 receptor antagonists (C09), loop diuretics (C03C), spironolactone (C03D), statins (C10A) and glucose-lowering drugs (A10) filled up to 6 months prior to study inclusion. Comorbidity was assessed according to the Ontario acute MI mortality prediction rules [11]. Invasive coronary revascularization was defined as percutaneus coronary intervention (PCI) or coronary artery bypass grafting (CABG), and classified as emergent revascularization when the procedure was carried out <48 h after the index MI.

End-points

The primary end-points of all-cause mortality and a composite of recurrent MI (ICD-10 I21-I22), stroke (I60-I61, I63-I64) and cardiovascular death (I00-I99) were assessed. Fatal index MI was defined as survival time <30 days. The diagnoses of MI (positive predictive value 93.5%) and stroke (positive predictive value 86.0%) in the Danish National Patient Register have previously been validated [12, 13]. Patients with a prior MI or stroke were identified by hospitalizations for these events from 1978 and onwards.

Statistical analysis

Baseline characteristics of the study population were summarized descriptively and presented as means (SD) and percentages. Incidence rates (IRs) of study end-points were reported as events per 1000 patient-years. The hazard ratios (HRs) and corresponding 95% confidence intervals (CIs) were estimated by Cox regression models controlling for confounding factors including age, gender, year of inclusion, concomitant medication, comorbidity and socioeconomic status. Short-term prognosis was evaluated as 30-day outcome. To address the potential impact of misclassification and increased healthcare consumption of the prespecified psoriasis inclusion criteria, we conducted sensitivity analyses with psoriasis defined as first prescription claim for vitamin D derivates or, in the case of no prescription claims, first psoriasis diagnosis. Finally, we performed analyses of nonfatal MI with start of follow-up 30 days from the index MI, and inclusion of revascularization strategy and post-MI medication in the regression model to assess the impact of differences in post-MI treatment strategy on prognosis. Adherence to pharmacotherapy was evaluated 1 year after the index MI. Model assumptions, including absence of interaction between model covariates and the proportional hazard assumption, were tested and found to be valid. A two-sided P-value <0.05 was considered to be statistically significant. All statistical analyses were performed with sas statistical software version 9.2 (Sas Institute Inc., Cary, NC, USA) and stata software version 11 (Statacorp, College Station, TX, USA).

Results

Baseline characteristics

Baseline characteristics of the study population are presented in Table 1. A total of 49 397 patients experienced a first-time MI during the study period, of which 462 patients had psoriasis at the time of MI. Patients with psoriasis had a mean age of 69.5 years, compared with 70.6 years for patients without psoriasis. There was a higher percentage of men than women amongst patients with psoriasis, compared with those without psoriasis (63.4% vs. 61.3%), and the mean of follow-up period was 19.5 months (SD 16.5) and 22 .0 months (SD 18.7), respectively. In total, <2% of patients were lost to follow-up. At baseline, patients with psoriasis had a higher rate of ischaemic heart disease other than MI (e.g. hospitalizations for stable and/or unstable angina pectoris (P = 0.01 for difference) and a higher percentage were treated with statins and ACE inhibitors/angiotensin 2 receptor blockers (P = 0.02 and P = 0.01, respectively) than patients without psoriasis. There were no other significant differences in baseline cardiovascular risk factors, pharmacotherapy or comorbidity at baseline (Table 1).

Table 1. Baseline characteristics of the study population with first-time myocardial infarction
CharacteristicPatients without psoriasis n = 48 935Patients with psoriasis n = 462 P for difference
  1. COPD, chronic obstructive pulmonary disease; ACEI, angiotensin-converting enzyme inhibitor; ARB, angiotensin 2 receptor blocker.

Age, years (SD)70.6 (13.5)69.5 (12.1)0.06
Men (%)29 987 (61.3)293 (63.4)0.35
Comorbidity (%)
 Shock1328 (2.7)13 (2.8)0.90
 Pulmonary oedema758 (1.6)2 (0.4)0.05
 Cardiac dysrhythmia6790 (13.9)62 (13.4)0.78
 Peripheral atherosclerosis1786 (3.65)12 (2.6)0.23
 Congestive heart failure7515 (15.4)76 (16.5)0.52
 COPD3836 (7.8)41 (8.9)0.41
 Acute renal failure837 (1.7)7 (1.5)0.75
 Cancer1773 (3.6)24 (5.2)0.07
Treatment (%)
 Platelet inhibitor16 673 (34.1)175 (37.9)0.09
 Beta-blocker15 016 (30.7)155 (33.6)0.18
 ACEI/ARB14 442 (29.5)161 (34.9)0.01
 Statin11 177 (22.8)127 (27.5)0.02
 Loop diuretic11 216 (22.9)118 (25.5)0.18
 Spironolactone2697 (5.5)25 (5.5)0.93
 Glucose-lowering drug5776 (11.8)61 (13.2)0.35

Pharmacological and invasive treatment following the index MI

The proportion of fatal index MIs did not differ statistically between the two groups (18.0% vs. 16.9%; P = 0.53). Patients with psoriasis surviving the index MI were more likely to receive statin therapy than those without psoriasis (75.3% vs. 70.5%, P < 0.05). Furthermore, patients with psoriasis demonstrated a nonsignificant trend towards decreased use of beta-blockers, and increased use of glucose-lowering drugs, loop diuretics and ACE inhibitors/angiotensin 2 receptor antagonists (Table 2), compared with patients without psoriasis. Significant differences in rates of invasive coronary revascularization (PCI or CABG), following the index MI, were not apparent between the two groups (Table 3). Adherence to post-MI pharmacotherapy (evaluated 1 year after the index MI) was very high regardless of psoriasis status (Table 4).

Table 2. Pharmacological treatment following nonfatal first-time myocardial infarction
Pharmacological treatment (%)Patients without psoriasis n = 40 127Patients with psoriasis n = 384 P for difference
  1. ACEI, angiotensin-converting enzyme inhibitor; ARB, angiotensin 2 receptor blocker.

Platelet inhibitor33.283 (82.9)319 (83.1)0.95
Beta-blocker31 118 (77.6)293 (76.3)0.56
ACEI/ARB19 757 (49.2)200 (52.1)0.26
Statin28 275 (70.5)289 (75.3)0.04
Loop diuretic14 222 (35.4)151 (39.3)0.11
Spironolactone3901 (9.7)36 (9.4)0.46
Glucose-lowering drug4526 (11.2)51 (13.3)0.22
Table 3. Invasive coronary revascularization following nonfatal first-time myocardial infarction (MI)
Revascularization (%)Patients without psoriasis n = 40 127Patients with psoriasis n = 384 P for difference
  1. Emergent, <48 h from index MI; PCI, percutaneus coronary intervention; CABG, coronary artery bypass grafting.

PCI16 512 (41.2)168 (43.8)0.30
Emergency PCI10 690 (26.6)101 (26.3)0.88
CABG2515 (6.3)29 (7.6)0.30
Emergency CABG273 (0.7)5 (1.3)0.14
Table 4. Pharmacological treatment assessed 1 year after nonfatal first-time myocardial infarction
Pharmacological treatment (%)Patients without psoriasis n = 29 649Patients with psoriasis n = 277 P for difference
  1. ACEI, angiotensin-converting enzyme inhibitor; ARB, angiotensin 2 receptor blocker.

Platelet inhibitor24 799 (83.6)232 (88.8)0.96
Beta-blocker23 859 (80.5)222 (80.1)0.89
ACEI/ARB14 826 (50.0)146 (52.7)0.37
Statin21 833 (73.6)219 (79.1)0.04
Loop diuretic9583 (32.3)97 (35.0)0.34
Spironolactone2619 (8.8)18 (6.5)0.17
Glucose-lowering drug3265 (11.0)37 (13.4)0.22

All-cause mortality and composite cardiovascular end-point

The incidence of the primary end-points of all-cause mortality and the composite of recurrent MI, stroke and cardiovascular death was increased in patients with psoriasis compared to those without. For all-cause mortality, the IRs per 1000 patient-years were 119.4 (95% CI 117.2–138.3) and 138.3 (95% CI 114.1–167.7) for patients without and with psoriasis, respectively. After adjustment for age, gender, year of inclusion, comorbidity, concomitant medication and socioeconomic status, the HR for all-cause mortality associated with psoriasis was 1.18 (95% CI 0.97–1.43). For the composite cardiovascular end-point, the IRs were 149.7 (95% CI 147.1–152.4) and 185.6 (95% CI 155.8–221.0) for patients without and with psoriasis, respectively, with a fully adjusted psoriasis-associated HR of 1.26 (95% CI 1.06–1.54). Analyses of the short-term prognosis yielded similar results with HRs of 1.20 (95% CI 0.99–1.46) and 1.24 (95% CI 1.04–1.49) for all-cause mortality and the composite end-point, respectively. Results of the models adjusted for year of inclusion, age and sex as well as the fully adjusted regression models are presented in Tables 5 and 6. Kaplan–Meier event-free survival curves are presented for all-cause mortality and the composite end-point in Figs 1 and 2.

Table 5. Hazard ratio and 95% confidence interval (CI) for all-cause mortality in patients with psoriasis compared with those without
End-pointPatients without psoriasisPatients with psoriasis
Complete follow-up
 Adjusted for age, sex and year of inclusion11.28 (CI 1.06–1.56)
 Adjusted for age, sex, year of inclusion, treatment, socioeconomic data and comorbidity11.18 (CI 0.97–1.43)
1-year follow-up
 Adjusted for age, sex, year of inclusion, treatment, socioeconomic data and comorbidity11.15 (CI 0.95–1.40)
30-day follow-up
 Adjusted for age, sex, year of inclusion, treatment, socioeconomic data and comorbidity11.20 (CI 0.99–1.46)
Table 6. Hazard ratio and 95% confidence interval (CI) for the composite end-point (recurrent myocardial infarction, stroke and cardiovascular death) in patients with psoriasis compared with those without
End-pointPatients without psoriasisPatients with psoriasis
Complete follow-up
 Adjusted for age, sex and year of inclusion11.30 (CI 1.07–1.58)
 Adjusted for age, sex, year of inclusion, treatment, socioeconomic data and comorbidity11.26 (CI 1.06–1.54)
1-year follow-up
 Adjusted for age, sex, year of inclusion, treatment, socioeconomic data and comorbidity11.24 (CI 1.04–1.48)
30-day follow-up
 Adjusted for age, sex, year of inclusion, treatment, socioeconomic data and comorbidity11.24 (CI 1.04–1.49)
Figure 1.

Kaplan–Meier curves for all-cause mortality.

Figure 2.

Kaplan–Meier curves for the composite end-point (recurrent myocardial infarction, stroke and cardiovascular death).

Analyses accounting for differences in post-MI treatment strategy

The risks of all-cause mortality and of the composite cardiovascular end-point remained increased in patients with psoriasis after adjustments for differences in post-MI pharmacological and invasive treatment, with HRs of 1.15 (95% CI 0.93–1.44) and 1.26 (95% 1.03–1.53), respectively.

Additional analyses

When classifying patients with psoriasis by the first vitamin D prescription claim or psoriasis diagnosis, a total of 1031 patients with psoriasis at the time of the index MI were identified, corresponding to a period prevalence of 2.1%. The IRs and HRs obtained in this analysis were fully comparable with values obtained in the primary analysis. The HRs for all-cause mortality and the composite cardiovascular end-point were 1.18 (95% CI 1.03–1.34) and 1.25 (95% CI 1.11–1.42), respectively.

Discussion

To our knowledge, this isthe first study to assess the prognosis following MI in patients with psoriasis, and to demonstrate a significantly increased risk of recurring adverse cardiovascular events and a trend for increased all-cause mortality associated with psoriasis. The results were supported by analyses accounting for differences in post-MI medical and invasive treatment. These novel findings extend the available evidence indicating that psoriasis is a clinically relevant risk factor for cardiovascular morbidity and mortality, and argue against a predominant contribution of surveillance bias to these findings. Furthermore, the demonstration of poor prognosis in the post-MI psoriasis population suggests the need for investigations to determine the value of a more aggressive approach to secondary prevention of cardiovascular disease in this group of patients.

Psoriasis and atherosclerosis are chronic inflammatory diseases characterized by T-helper (Th) 1-/Th17-driven inflammation, and there is a considerable overlap of inflammatory markers and mediators involved in these diseases [1, 3]. Indeed, psoriasis has been associated with conventional cardiovascular risk factors, surrogate markers of atherothrombotic disease and cardiovascular morbidity and mortality [4–9, 14–20]. However, results implicating psoriasis as an independent cardiovascular risk factor have recently been questioned, and it has been suggested that surveillance bias may have contributed to earlier findings, that is, patients with psoriasis may be at increased risk of being diagnosed with cardiovascular diseases by being more likely to visit their physician than the general population [21–23]. Of note, we adjusted for important confounders, including post-MI pharmacological treatment and invasive revascularization, in our cohort-based comparison between patients with psoriasis and controls following MI. In this context, it should also be emphasized that surveillance bias was inherently reduced in our study in the post-MI setting, as essentially all patients (with or without psoriasis) in Denmark are provided with extensive secondary prevention healthcare, i.e. a considerably more uniform surveillance situation than before the first occurrence of cardiovascular disease.

The results of our study also suggest some potentially important differences in post-MI treatment, that is, patients with psoriasis displayed a nonsignificant trend for increased use of ACE inhibitors/angiotensin 2 receptor blockers and loop diuretics (Table 2). Although the same trend for use of these agents was observed before the index MI, and increased prevalence of hypertension in patients with psoriasis may have contributed to the results [1, 2, 14–17], the post-MI findings could be suggestive of more post-MI left ventricular dysfunction and larger infarct size, i.e. ST-elevation MI (STEMI), in patients with psoriasis. The current Danish registries cannot differentiate between STEMI and non-STEMI, but primary PCI has been the preferred treatment for STEMI in Denmark since 2002–2003 [24] and it is notable that emergent (<48 h after admission) PCI rates after the index MI were similar between patients with psoriasis and controls (Table 3), indicating a comparable distribution of STEMI and non-STEMI in the two groups. Furthermore, patients with psoriasis were more likely to receive statins than patients without psoriasis, and showed a nonsignificant trend towards increased treatment with glucose-lowering drugs, before and after the index MI (Tables 1 and 2), supporting previous reports of an increased prevalence of hypercholesterolaemia and diabetes mellitus in patients with psoriasis compared with those without [1, 2, 14–17]. Even after adjustments for these and other measured confounders, however, we found that psoriasis was an independent risk factor for poor post-MI prognosis. This result mirrors findings in patients with rheumatoid arthritis, which is also a chronic Th1-driven inflammatory disease that has been associated with increased risk of cardiovascular disease and impaired prognosis following MI [25–27]. Taken together, these findings, including the present indication of worse post-MI prognosis in patients with psoriasis, support the notion that inflammation is an important determinant of impaired prognosis following cardiovascular events in patients with inflammatory diseases.

The following limitations of our study should be discussed. Observational studies may reveal potentially important disease associations but cannot establish causal relationships, and our findings should be interpreted in this context. The psoriasis cohort was based on prescriptions filled for topical vitamin D derivatives, and our results may not apply to patients treated with topical corticosteroids alone. Furthermore, patients with psoriasis who were not captured by the inclusion criteria would have been misclassified as controls, thus introducing an uncertainty on the presented results towards the null hypothesis. Of note, topical treatments, including vitamin D derivatives, are used as first-line psoriasis treatment, and there is no reason to suspect that there would be important differences in baseline characteristics between patients treated with topical vitamin D, corticosteroid or another first-line drug. We were unable to assess the impact of disease severity and different systemic psoriasis treatment modalities on the outcome, and studies addressing these limitations could provide further insight. Because the Danish population is predominantly of Caucasian descent, the results should only be extrapolated to other ethnicities with caution. Finally, the possibility of residual confounding is present in all observational studies, and in the present study, we were unable to account for some important cardiovascular risk factors, such as obesity and smoking. As indicated previously, psoriasis is associated with a higher prevalence of classical cardiovascular risk factors including smoking and obesity. Our lack of information on these conditions would most probably inflate the psoriasis-related impact on post MI prognosis, although some of these potential confounders are associated with covariates that were included in our adjustments, for example socioeconomic status and medical treatment, allowing us to assess some of these effects. It is evident that classical cardiovascular risk factors are present in almost all patients with recent MI regardless of psoriasis status, and it is therefore unlikely that these factors are the main determinants of the observed differences in post MI prognosis. This notion is supported by our demonstration of very similar risk estimates in the crude and fully adjusted statistical models, indicating very little impact of measured differences in baseline characteristics, but we cannot fully exclude the possibility that our findings were a result of unmeasured or residual confounding.

The strengths of the present study include the large number of participants, the nationwide coverage of prospectively recorded registries, the use of validated end-points, the contemporary clinical setting and the completeness of follow-up. The use of nationwide registries of hospitalization data and dispensed prescriptions from all pharmacies in Denmark where healthcare is readily accessible and essentially free of charge minimized selection bias related to, for example, gender, age, socioeconomic status, healthcare insurance provision and labour market association. Moreover, as indicated earlier, surveillance bias was reduced by direct comparison of patients with and without psoriasis within the post-MI population.

Conclusion

This first study in post-MI patients with psoriasis demonstrated a worse prognosis with increased risk of the composite end-point of recurrent MI, stroke and cardiovascular death and a nonsignificant trend towards increased all-cause mortality compared with patients without psoriasis. Prospective studies are needed to confirm these findings and determine the value of more aggressive secondary prevention of cardiovascular disease in patients with psoriasis.

Acknowledgements

The study was supported by the Department of Cardiology, Copenhagen University Hospital Gentofte, DK-2900, Denmark. The funding source had no influence on study design, interpretation of the results or the decision to submit the article.

Conflict of interest statement

No conflicts of interest to declare.

Ancillary