Adjusted incidence rate ratios (data markers) with 95% CIs in patients with hidradenitis suppurativa (HS) compared with healthy controls and patients with severe psoriasis. MACEs indicates major adverse cardiovascular events.
eTable. Identification of Alcohol Abuse
Egeberg A, Gislason GH, Hansen PR. Risk of Major Adverse Cardiovascular Events and All-Cause Mortality in Patients With Hidradenitis Suppurativa. JAMA Dermatol. 2016;152(4):429-434. doi:10.1001/jamadermatol.2015.6264
Hidradenitis suppurativa (HS) is a common inflammatory skin disease. The disease has been associated with cardiovascular (CV) risk factors, but the risk of CV disease in patients with HS is unknown.
To investigate CV risk in patients with HS.
Design, Setting, and Participants
A population-based cohort study was conducted from January 1, 1997, to December 31, 2011, using individual-level linkage of nationwide administrative registers. In a study population of 35 368 Danish individuals, 5964 patients aged 18 years or older with a hospital-based diagnosis of HS (cases) were matched 1:5 on age, sex, and calendar time with 29 404 individuals serving as controls.
Main Outcomes and Measures
Outcomes were myocardial infarction (MI), stroke, CV-associated death, major adverse CV events (MACEs), and all-cause mortality. Incidence rate ratios (IRRs) were estimated by Poisson regression.
The 5964 patients with HS had a mean (SD) age of 37.7 (11.7) years; 4346 (72.9%) were women. In this sample, a total of 62 (42 749.0 person-years) MIs, 74 (42 647.8 person-years) ischemic strokes, 63 (42 941.7 person-years) CV-associated deaths, 169 (42 463.5 person-years) MACEs, and 231 (42 941.7 person-years) all-cause deaths occurred during follow-up. Adjusted (age, sex, socioeconomic status, smoking, comorbidity, and medication) IRRs (95% CIs) were 1.57 (1.14-2.17) for MI, 1.33 (1.01-1.76) for ischemic stroke, 1.95 (1.42-2.67) for CV-associated death, 1.53 (1.27-1.86) for MACEs, and 1.35 (1.15-1.59) for all-cause mortality. When patients with severe psoriasis were used as controls, the adjusted IRRs in patients with HS were 1.00 (0.74-1.35) for MI, 0.93 (0.71-1.22) for ischemic stroke, 1.58 (1.17-2.12) for CV-associated death, 1.08 (0.90-1.29) for MACEs, and 1.09 (0.94-1.28) for all-cause mortality.
Conclusions and Relevance
Hidradenitis suppurativa was associated with a significantly increased risk of adverse CV outcomes and all-cause mortality independent of measured confounders. The risk of CV-associated death was higher in patients with HS compared with the risk in those with severe psoriasis. The results call for increased awareness of this association and for studies of its clinical consequences.
Hidradenitis suppurativa (HS) is a chronic, inflammatory skin disease, characterized by painful recurrent, suppurating, and malodorous abscesses in areas rich with apocrine sweat glands, such as the axillae and genital area.1 European studies1 have suggested an HS prevalence of 1% to 4% with a 3:1 female preponderance, and the disease usually develops in individuals in their early twenties. The psychosocial burden and negative effect of HS on quality of life is greater than that of other dermatologic conditions, and the disease is likely to be underdiagnosed and frequently associated with a 5- to 10-year delay in diagnosis.1- 5 Although the etiopathogenesis of HS is not fully understood, increased expression of inflammatory cytokines, such as interleukin 1β (IL-β), IL-10, IL-17, and tumor necrosis factor (TNF), has been demonstrated in HS lesions, and reports of elevated circulating levels of TNF are suggestive of systemic inflammatory activation.6- 9 A recent study10 found that patients with HS carry a greater systemic inflammatory load, with higher circulating leukocyte counts and C-reactive protein levels than, for example, patients with psoriasis. Evidence has also associated HS with cardiovascular (CV) risk factors, such as smoking, obesity, metabolic syndrome, and diabetes mellitus.11 Atherosclerosis is a chronic inflammatory disease, and other chronic inflammatory diseases, such as psoriasis, rheumatoid arthritis, and inflammatory bowel disease, have been linked with an increased risk of CV disease independent of CV risk factors, which may be explained, in part, by shared inflammatory mechanisms.12- 16 Scarce evidence is available on the risk of CV disease in patients with HS, and we therefore investigated this clinically important issue in a population-based cohort study.
Question: Do patients with hidradenitis suppurativa (HS) have an increased risk of adverse cardiovascular (CV) outcomes?
Findings: This population-based cohort study found a significantly increased risk of myocardial infarction (57%), ischemic stroke (33%), CV death (95%), major adverse CV events (53%), and all-cause mortality (35%) in patients with HS when compared to the controls. Notably, the risk of CV death was 58% higher in patients with HS than in patients with severe psoriasis, a significant difference.
Meaning: Patients with HS have a higher risk of major CV events—including myocardial infarction and stroke—and higher mortality, especially from CV events, compared with controls.
Study approval was obtained from the Danish Data Protection Agency; approval from an ethics committee is not required for register studies in Denmark. The Strengthening the Reporting of Observational Studies in Epidemiology recommendations were used in conducting and reporting of this study.17
All Danish citizens receive free and equal health care services, which ensure unencumbered access to general practitioners and hospitals. At birth or immigration, each individual receives a unique, 10-digit personal identification number, which allows for unambiguous linkage across the nationwide administrative registers in Denmark. All inpatient and outpatient hospital consultations are recorded in the Danish National Patient Registry.18 The register, established in 1978, used codes from the International Classification of Diseases, Eighth Revision (ICD-8) until 1994; codes from the International Statistical Classification of Diseases and Related Health Problems, Tenth Revision (ICD-10) have been used thereafter. For administrative reasons, ICD-9 was never used in Denmark. Hospital procedures, including hospital-based pharmacologic treatment (eg, biological agents), are coded in the register as treatment procedures. Since 1994, detailed and accurate information on all pharmacy-dispensed medications is registered according to the international Anatomical Therapeutic Chemical classification in the Danish Registry of Medicinal Products Statistics.19 Information on tax-reported household income is recorded by Statistics Denmark, and information on age, sex, vital (ie, updated daily to note date of death), and migration status is available from the Civil Registration System.20,21 Within 14 days of death, the numbers and causes of all deaths are registered in the national Register of Causes of Death using ICD-10 codes.22
We identified all patients aged 18 years or older with a first-time (inpatient or outpatient) hospital dermatologist diagnosis of HS (ICD-8 code 705.91 and ICD-10 code L73.2) between January 1, 1997, and December 31, 2011. The index date for cases was the date of first diagnosis, and each patient was matched on age, sex, and calendar time with 5 individuals serving as controls from the general population. Index date for the controls was the index date for the corresponding cases, and the cohort was followed until migration, death from any cause, or the occurrence of an end point, whichever came first. We excluded patients with a history of myocardial infarction (MI) or ischemic stroke prior to the study to enable examination of the temporal association between exposure and outcome.
The primary end points were a diagnosis of MI (ICD-10 codes I21-I22), ischemic stroke (ICD-10 codes I63-I64), and CV-associated death (ICD-10 codes I00-I99). Secondary end points were all-cause mortality and major adverse CV events (MACEs), which are a composite of MI, ischemic stroke, and CV-associated death. The identification of MI and ischemic stroke has been validated in the Danish National Patient Registry.23,24
Baseline treatment up to 6 months before study inclusion was defined for the following drugs: loop diuretics, platelet inhibitors, cholesterol-lowering drugs, and vitamin K antagonists. Baseline comorbidity was assessed by ICD codes up to 5 years before study inclusion for the following diagnoses: cardiac dysrhythmia, depression, diabetes mellitus, renal disease, hypertension, inflammatory bowel disease, ischemic heart disease and peripheral artery disease, venous thromboembolic disease, and heart failure. Hypertension was defined by either a hospital diagnosis or treatment within 90 days with at least 2 of the following classes of antihypertensive drugs: α-adrenergic blockers, non–loop diuretics, vasodilators, β-blockers, calcium channel blockers, and renin-angiotensin system inhibitors as previously described and validated with a positive predictive value of 80% and a specificity of 95%.25 Diabetes mellitus was defined by either a hospital diagnosis or use of glucose-lowering drugs. We tracked smoking history by a data retrieval algorithm dependent on diagnoses of smoking, tobacco use, chronic obstructive pulmonary disease, lung cancer, treatment and/or therapeutic interventions aimed at smoking cessation, and/or prescriptions of drugs used for smoking cessation.26 Alcohol abuse was defined by diagnoses of alcohol abuse or conditions strongly related to alcohol abuse (eg, alcoholic liver disease), treatment with drugs used for alcohol dependence, and treatment interventions for alcohol dependence (eTable in the Supplement). Information was continually updated during the follow-up period. We calculated an index of socioeconomic status (standardized by age) between 0 (lowest) and 4 (highest) based on the mean gross annual income during a 5-year period before study inclusion.
We described baseline characteristics with means (SDs) for continuous variables and frequencies and percentages for categorical variables. Incidence rates were summarized per 10 000 person-years, and Poisson regression analyses were performed to estimate unadjusted and fully adjusted (age, sex, socioeconomic status, smoking, medication, and comorbidity) incidence rate ratios (IRRs). Hidradenitis suppurativa is significantly associated with obesity, which is hence a potential confounder in the association between HS and CV disease.11 To address this issue, we performed sensitivity analyses in which patients with HS (cases) were compared with patients with severe psoriasis (controls). Severe psoriasis is also strongly associated with obesity, and nationwide registry data from patients receiving biological treatment have shown that Danish individuals with severe psoriasis have a mean body weight of 88.6 (21.1) kg.27 Moreover, with use of the same data set from which the HS controls for the present study were sampled, patients with psoriasis were shown14 to have a disease severity–dependent increased risk of MI, ischemic stroke, and CV-associated death compared with matched controls and that this risk was comparable to the risk of CV disease observed in patients with diabetes mellitus. With use of another obese population (individuals with severe psoriasis) with an established elevated risk of CV disease as controls, sensitivity analyses yielding no significant differences in the risk of adverse CV end points between the control group and HS (or a higher risk in patients with HS) would therefore suggest that the results of our primary analyses were not solely explained by confounding owing to obesity. Consequently, we identified all patients aged 18 years or older with a first-time diagnosis of severe psoriasis between January 1, 1997, and December 31, 2011. Patients with severe psoriasis were defined as those receiving systemic antipsoriatic therapy. This method for psoriasis identification and severity classification has been described and validated with a sensitivity of 98%.26 All statistical tests were conducted using a level of significance of P < .05, and results are reported with 95% CIs where applicable. All analyses were performed using SAS, version 9.4 (SAS Institute Inc) and Stata, version 11.0 (StataCorp) statistical software.
The study comprised a total of 5964 cases (ie, patients with HS without a history of MI or ischemic stroke prior to the study). These patients were matched with 29 820 controls from the general population, and, after exclusion of 416 individuals with a history of MI or ischemic stroke as well as patients with incomplete follow-up due to migration, the final control population comprised 29 404 individuals. Baseline characteristics are reported in Table 1. The mean age of the patients with HS was 37.7 (11.7) years; 4346 (72.9%) were women.
The mean follow-up time was 7.1 (4.4) years; among the 35 368 cases and controls combined, a total of 200 (0.57%) MIs, 282 (0.80%) ischemic strokes, 183 (0.52%) CV-associated deaths, 565 (1.60%) MACEs, and 800 (2.26%) all-cause deaths occurred during follow-up. The incidence rates of adverse CV outcomes per 10 000 person-years for HS and controls are detailed in Table 2. In patients with HS, the unadjusted IRRs (95% CIs) were 2.18 (1.61-2.94) for MI, 1.73 (1.32-2.25) for ischemic stroke, 2.54 (1.87-3.44) for CV-associated death, 2.07 (1.73-2.48) for MACEs, and 1.96 (1.68-2.29) for all-cause mortality (Table 3). As also reported in Table 3, the results remained consistent in the fully adjusted analyses.
In the sensitivity analysis in which patients with HS were compared with 13 093 patients with severe psoriasis, age- and sex-adjusted IRRs (95% CIs) in patients with HS were 1.12 (0.83-1.51; P = .47) for MI, 1.05 (0.80-1.38; P = .73) for ischemic stroke, 1.63 (1.23-2.18; P < .001) for CV-associated death, 1.20 (1.00-1.44; P < .47) for MACEs, and 1.25 (1.07-1.46; P < .004) for all-cause mortality. In fully adjusted analyses, IRRs (95% CIs) were 1.00 (0.74-1.35; P > .99) for MI, 0.93 (0.71-1.22; P = .60) for ischemic stroke, 1.58 (1.17-2.12, P < .003) for CV-associated death, 1.08 (0.90-1.29; P = .44) for MACEs, and 1.09 (0.94-1.28; P = .26) for all-cause mortality (Figure).
We found a significantly increased risk of MI, ischemic stroke, CV-associated death, MACEs, and all-cause mortality in patients with HS after adjustment for confounding factors. The risk of MI, ischemic stroke, MACEs, and all-cause mortality in patients with HS was similar to that of patients with severe psoriasis; however, compared with the latter group, the risk of CV-associated death was significantly higher in patients with HS. These novel findings suggest that HS is an independent risk factor for adverse CV outcomes.
Studies6- 8,28 have suggested that, in HS, atrophy of the sebaceous glands, follicular hyperkeratinization, and subsequent hair follicle destruction are associated with deep-seated inflammation, increased susceptibility to secondary infections, and chronic perpetuation of the inflammatory response. Augmented circulating levels of TNF have been reported9 in patients with HS, and a recent study10 of 50 Danish outpatients with HS and 250 matched controls from the same dermatology clinic, including patients with other dermatologic conditions (eg, psoriasis), found that patients with HS had a greater systemic inflammatory load, with higher circulating leukocyte counts and C-reactive protein levels. Atherosclerosis is an inflammatory disease, and increased systemic levels of C-reactive protein and TNF have been associated12,13 with an independent risk of CV disease in the general population. In view of the accumulating evidence of the association between CV disease and other chronic inflammatory diseases (eg, psoriasis, rheumatoid arthritis, and inflammatory bowel disease), there is a conspicuous absence of reports on the risk of CV disease in patients with HS; it has been suggested that academic research on HS has been somewhat neglected.29 However, our present finding of an increased risk of adverse CV outcomes in patients with HS is clearly in agreement with observations in patients with other chronic inflammatory diseases and supports the notion that shared inflammatory mechanisms contribute to this phenomenon.12,13 Diminished numbers of circulating endothelial progenitor cells have been associated with atherosclerosis and inflammation; in patients with HS, a reduced number of these cells was recently reported,30 whereas measurable changes of peripheral blood coagulation markers (eg, activated partial thromboplastin time) were not observed.31
Psoriasis is the prototypical inflammatory skin disease, and inflammatory cytokines (eg, IL-17 and TNF) play significant roles in psoriasis onset and progression.32 As indicated above, IL-17 and TNF have also been implicated in the pathogenesis of HS, and treatment with TNF inhibitors is efficacious in both conditions.7,33,34 Furthermore, patients with HS or psoriasis have an increased prevalence of CV risk factors that are also associated with systemic inflammation (eg, smoking, obesity, metabolic syndrome, and diabetes mellitus). It is tempting to speculate that, in addition to the increased risk of CV disease, HS may carry a risk of other comorbidities that have been associated with psoriasis and systemic inflammation.35,36 For example, an increased risk of spondyloarthritis, depression, and inflammatory bowel disease is relatively well established in psoriasis and has also been suggested for HS.5,35- 40 Indeed, some or all of these comorbidities are increasingly being perceived as independent risk factors for CV, and more studies are warranted to examine the prevalence of traditional and nontraditional CV risk factors and other comorbidities in patients with HS and their potential therapeutic and incremental prognostic importance.12- 16,41 Our sensitivity analyses showed comparable risk of MI, stroke, MACEs, and all-cause mortality in patients with HS vs those with severe psoriasis, indicating that obesity is unlikely to be a main determinant of the observed association. Moreover, this finding positions the risk of CV disease in HS alongside that of severe psoriasis, which is clinically well established and likely to be the same magnitude as the risk of CV disease in persons with diabetes mellitus.14,42 The apparent increased risk of CV-associated death in patients with HS compared with the risk in individuals with severe psoriasis in the absence of significant differences in the risk of first-time MI or stroke between the 2 diseases may suggest that patients with HS have increased mortality after these adverse CV events; this hypothesis should be examined in future studies.
Several limitations and strengths apply to the interpretation of the present results. Owing to the observational design of our study, we cannot determine causality. We were unable to distinguish between severities of HS, and, although patients with HS in our study were identified by hospital diagnosis, studies11 have suggested that patients with HS seen in hospital settings have more comorbidities than do those sampled from a population setting. In addition, we lacked information on body weight in patients with HS. We attempted to address this issue in sensitivity analyses that compared HS with severe psoriasis; however, we cannot exclude the possibility that residual confounding may have affected our results. A further limitation is that the Danish population is predominantly white, and extrapolation of results to patients of other races/ethnicities should be conducted with caution. Important strengths of the study include the high accuracy of the nationwide registries as well as the available information on household income, which minimized bias associated with sex, age, concurrent medication, and socioeconomic status. In addition, the statistical adjustments for a range of comorbidities for which data were continuously updated during follow-up, the length and accuracy of follow-up, the results of the sensitivity analyses, and the large number of individuals included in the analyses add credibility to our findings.
We found a significantly increased risk of adverse CV outcomes and all-cause mortality in patients with HS. The results were independent of measured confounders and were comparable to the risk of CV disease and all-cause mortality in patients with severe psoriasis, albeit with an increased risk of CV-associated death in those with HS. The results call for greater awareness of this association and for studies of its clinical consequences.
Corresponding Author: Alexander Egeberg, MD, PhD, Department of Dermato-Allergology, Herlev and Gentofte Hospital, University of Copenhagen, Kildegårdsvej 28, DK-2900 Hellerup, Denmark (firstname.lastname@example.org).
Accepted for Publication: December 17, 2015.
Published Online: February 17, 2016. doi:10.1001/jamadermatol.2015.6264.
Author Contributions: Drs Egeberg and Gislason had full access to all the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.
Study concept and design: Egeberg, Hansen.
Acquisition, analysis, or interpretation of data: All authors.
Drafting of the manuscript: Egeberg.
Critical revision of the manuscript for important intellectual content: All authors.
Statistical analysis: Egeberg.
Administrative, technical, or material support: Egeberg, Gislason.
Study supervision: All authors.
Conflict of Interest Disclosures: None reported.