Risk of Peripheral Arterial Occlusive Disease in Patients With Systemic Lupus Erythematosus: A Nationwide Population-Based Cohort Study.

Systemic lupus erythematosus (SLE) is associated with atherosclerosis, but the relationship between SLE and peripheral arterial occlusive disease (PAOD) remains unclear. We sought to investigate this relationship by comparing cardiovascular complications in patients with and without SLE.Data on patients from 2000 to 2011 were collected from the National Health Insurance Research Database of Taiwan. The SLE cohort was frequency-matched according to age, sex, and history of diabetes mellitus (DM) with patients without SLE (control cohort). We evaluated the risk of cardiovascular complications, including hypertension, DM, stroke, chronic obstructive pulmonary disease, heart failure, coronary artery disease, and hyperlipidemia.The study included 10,144 patients with SLE and 10,144 control patients. The incidence of PAOD was 9.39-fold higher (95% confidence interval [CI] = 7.70-11.15) in the SLE cohort than in the non-SLE cohort. Moreover, SLE was an independent risk factor for PAOD. The adjusted risk of PAOD was highest in patients with SLE who were aged ≤34 years (hazard ratio = 47.6, 95% CI = 26.8-84.4). The risk of PAOD was highest during the first year of follow-up and decreased over time.Patients with SLE exhibit a higher incidence and an independently higher risk of PAOD compared with the general population. The PAOD risk is markedly elevated in patients with SLE who are young and in whom the disease is at an early stage.

INTRODUCTION

Cardiovascular disease (CVD) is a major cause of morbidity and mortality in systemic lupus erythematosus (SLE). Many studies have indicated that SLE patients have higher risks of endothelial dysfunction and atherosclerosis compared with the general population.[1-4] Numerous cohort studies have reported that the risk of myocardial infarction is increased 2 to 10-fold in patients with SLE.[5-8] Epidemiological studies have further reported that patients with SLE also exhibit an increased risk of congestive heart failure and stroke.[9] Both traditional and disease-related factors contribute to the development of CVD in SLE patients.[10-13] Koenig et al suggested that the risk of CVD in patients with SLE is similar to the cardiovascular risk in patients with type 1 diabetes mellitus (DM).[14] Moreover, SLE per se is an independent risk factor for the development of CVD.[15]

Peripheral arterial occlusive disease (PAOD) is a manifestation of atherosclerosis that is associated with a 2.5 to 6-fold increase in cardiovascular mortality.[16,17] The prevalence and incidence of PAOD have been shown to be higher in patients with SLE than in the general population.[18-20] The risk factors for PAOD in patients with SLE include age, DM, dyslipidemia, smoking, an extended duration of steroid use, a Systemic Lupus International Collaborating Clinics Damage Index, use of azathioprine or warfarin, and plasma levels of thrombotic variables.[18-22] However, the risk factors for and incidence and consequences of PAOD in patients with SLE remain poorly investigated. Thus, we aimed to determine the incidence of PAOD in patients with SLE and analyze the risk factors for PAOD in such patients.

METHODS

Data Source

Data for this study were collected from the National Health Insurance Research Database (NHIRD). The NHIRD was created by the National Health Research Institute (NHRI), and it contains the claims data from the Taiwan National Health Insurance (NHI) program. The Taiwan NHI was established in 1995 by combining 13 existing insurance programs into 1 nationwide, single-payer health insurance program. The Taiwan NHI is a compulsory insurance program for residents of Taiwan, and it covered nearly 99% of the 23 million residents in 1998. The NHIRD includes a registry for beneficiaries, historical disease records, and medical service information, and this database is renewed every year. All original personal identification information on insured people is removed, and the NHRI assigns a scrambled and anonymous identification number to each insured person for safeguarding patient privacy before releasing the database for research. This study was approved to fulfill the condition for exemption by the Institutional Review Board (IRB) of China Medical University (CMUH104-REC2-115). The IRB also specifically waived the consent requirement.

In the NHIRD, diseases are recorded based on the criteria of the International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM). In this study, the history of SLE was collected from the catastrophic illness registry and the history of comorbidities was recorded from inpatient and outpatient files. All insurance claims should be scrutinized by medical reimbursement specialists and peer review according to the standard diagnosed criteria in the study. If these doctors or hospitals make wrong diagnoses or coding, they will be punished with a lot of penalties. Therefore, the diagnoses of PAOD and SLE in this study were highly reliable. In addition, some related studies with the same diagnosed method and criteria by ICD-9 coding were already accepted for publication.[23-26]

Study Population

The SLE cohort was established using patients >20 years old with a new onset of SLE (ICD-9-CM 710.0) between January 1, 2000 and December 31, 2011. The SLE diagnosis date was used as the index date of the SLE cohort. The comparison cohort included insured patients without SLE, was 4-fold larger than the SLE cohort, and was frequency-matched according to age (per 5 years), sex, and history of DM (ICD-9-CM 250). We randomly assigned a month and day but the same year as that of matched cases as the index date for the comparison cohort. We excluded people with a history of PAOD (ICD-9-CM 440.0, 440.2, 440.3, 440.8, 440.9, 443, 444.0, 444.22, 444.8, 447.8, and 447.9). The follow-up period ended when the study patients withdrew from the insurance, at the time of PAOD event occurrence, or on December 31, 2011.

In addition to demographic factors, comorbidities were considered here as confounding factors. Comorbidities were defined as diseases that were diagnosed in the patients before the index date. The comorbidities included hypertension (ICD-9-CM 401-405), cancer (ICD-9-CM 140-208), stroke (ICD-9-CM 430-438), chronic obstructive pulmonary disease (COPD; ICD-9-CM 491, 492, and 496), heart failure (HF; ICD-9-CM 428), coronary artery disease (CAD; ICD-9-CM 410-414), hyperlipidemia (ICD-9-CM 272), smoking (ICD-9-CM 305.1), and asthma (ICD-9-CM 493). In addition, steroid use, warfarin use, and various biologic therapies (including azathioprine, cyclosporin, immunoglobulin, mycophenolate, and tacrolimus) were analyzed between the SLE patients and the controls.

Statistical Analysis

We calculated the mean and standard deviation (SD) in the case of patient age and determined the number and percentage for age group, sex, and comorbidities in order to describe the distribution of the study population. A t test was use to analyze the difference in age distribution between the study cohorts, and a χ2 test was used in the case of age group, sex, comorbidities, and medications. The PAOD incidence density was calculated as the total number of PAOD events divided by the sum of the follow-up years (per 1000 person-year) for each group. We used the Kaplan–Meier method to obtain the PAOD cumulative-incidence curves for the SLE and control cohorts, and then applied the log-rank test to evaluate the difference between these 2 incidence curves. To assess the PAOD risk in SLE patients, we calculated the hazard ratios (HRs) and 95% confidence intervals (CIs) by using single-variable and multivariable Cox proportional hazard models. We also used the Cox model to measure the risk of PAOD in patients with SLE after stratification according to distinct demographic factors, comorbidities, and medications.

SAS 9.4 software (SAS Institute, Cary, NC, USA) was used for data management and statistical analysis. The PAOD cumulative-incidence curve was drawn using R software (R Foundation for Statistical Computing, Vienna, Austria). P < 0.05 (2-sided testing) was considered statistically significant.

RESULTS

This study enrolled 10,144 patients with SLE and 10,144 patients without SLE (comparison cohort) (Table 1). Males accounted for only 13.5% of the study population, and the mean age of the 2 cohorts was 41 years. The SLE cohort and non-SLE cohort were well matched for sex, age group, comorbidities, and medications. The mean follow-up duration was 5.22 ± 3.64 years for the SLE cohort and 6.05 ± 3.19 years for the control cohort.

TABLE 1

Comparison of Demographics and Comorbidities Between Patients With Systemic Lupus Erythematosus and Controls

Table 2 shows the risk of PAOD associated with SLE, demographic factors, and selected comorbidities and medications. According to the single-variable Cox proportional hazard model, SLE, sex, CAD, and warfarin were significantly associated with an increased risk of PAOD. After adjustment for age and the comorbidities of DM, hypertension, hyperlipidemia, COPD, HF, CAD, stroke, asthma, smoking, and medication of steroid use, and warfarin, the PAOD risk inpatients with SLE was 9.39-fold higher than that for patients without SLE (HR = 9.39, 95% CI = 7.70–11.5). The PAOD cumulative-incidence curve plotted for the SLE cohort was also significantly higher than the curve plotted for the comparison cohort (Fig. 1; P < 0.001, log-rank test).

TABLE 2

Cox Model Results Presented With Hazard Ratios and 95% Confidence Intervals of Peripheral Arterial Occlusive Disease Associated With Systemic Lupus Erythematosus and Covariates

FIGURE 1

Cumulative incidence of peripheral arterial occlusive disease in patients with (dashed line) and without (solid line) systemic lupus erythematosus.

Table 3 shows the risk of PAOD in patients with SLE according to distinct demographic factors, comorbidities, steroid use, and warfarin, further indicating that the PAOD risk in SLE patients was significantly higher than that of control patients. The PAOD risk in female patients with SLE was 9.90-fold higher than that in female patients in the comparison cohort (HR = 9.90, 95% CI = 7.98–12.3), but this risk in male patients with SLE was only 5.96-fold higher than that in control male patients (HR = 5.96, 95% CI = 3.50–10.2). Moreover, the PAOD risk was 2.92-fold higher in elderly patients (aged >50 years) with SLE than in patients without SLE. By contrast, in the population group aged ≤34 years, the risk of PAOD in patients with SLE was 47.6-fold higher than that in patients without SLE (HR = 47.6, 95% CI = 26.8–84.4). Furthermore, in the study population lacking any of the examined comorbidities, SLE clearly increased the risk of PAOD (HR = 19.7, 95% CI = 14.3–27.3). Among the nonmedication subjects, SLE patient still had great risk of PAOD than control patients.

TABLE 3

Incidence and Adjusted Hazard Ratios of Peripheral Arterial Occlusive Disease Stratified According to Sex and Age and Compared Between the SLE Cohort and the Non-SLE Cohort

Table 4 presents the risk of PAOD in SLE patients according to follow-up time. The results suggest that PAOD events occurred at an increased frequency in the early stages of the follow-up period. The rate of PAOD incidence in patients with SLE was 21.9-fold higher than that in control patients when the follow-up time was <1 year (HR = 32.0, 95% CI = 18.7–54.5), but this rate dropped to 10.8-, 4.99-, and 3.59-fold higher relative to control patients when the follow-up period was 1 to 3 years (HR = 10.8, 95% CI = 7.25–16.1), 4 to 5 years (HR = 4.99, 95% CI = 3.37–7.39), and >5 years (HR = 3.59, 95% CI = 2.49–5.18), respectively.

TABLE 4

Trends of Peripheral Arterial Occlusive Disease Stratified According to Follow-Up Years

Compared with patients without SLE, SLE patients without various biologic therapies used demonstrated a significantly higher risk of PAOD (adjusted HR = 10.6, 95% CI = 8.63–13.1), followed by SLE patients with various biologic therapies (adjusted HR = 8.10, 95% CI = 6.54–10.0) (Table 5). SLE patients who received various biologic therapies treatment exhibited significantly lower risk of PAOD (adjusted HR = 0.74, 95% CI = 0.64–0.86) than SLE patients without various biologic therapies used.

TABLE 5

Incidence, Crude and Adjusted Hazard Ratio of Peripheral Arterial Occlusive Disease Compared Among SLE Patients With and Without Various Biologic Therapies Treatment and Non-SLE Controls

DISCUSSION

Based on our research, this is the first population-based cohort study to determine the incidence of and risk factors for PAOD among patients with SLE by performing appropriate comparisons with non-SLE controls. Our results revealed that the risk of PAOD in the SLE cohort was 9.38-fold higher than that in the control cohort, and that SLE was an independent risk factor for PAOD. The PAOD risk was elevated in young patients with SLE, particularly in patients aged <34 years. Moreover, the risk of PAOD in patients with SLE was highest in the first year after SLE diagnosis and then it declined.

A 2004 American study conducted by Fischer et al determined that in patients with SLE, the overall relative risk of developing myocardial infarction was 2.67-fold higher than that in patients without SLE.[6] A California study reported that patients with SLE were 1 to 3-fold more likely to be admitted to the hospital with congestive heart failure than were patients without SLE.[9] Manzi et al demonstrated that the risk of CVD in SLE patients was >50-fold higher than that in non-SLE controls.[27] Patients with SLE were also reported to be associated with a 1.65 to 5-fold higher risk of PAOD compared with control groups.[18,19] In a large, multicenter cohort study, 5.3% of 637 patients with SLE developed PAOD over a mean follow-up period of 4.4 years.[20] However, our study revealed for the first time that the incidence of PAOD is higher in patients with SLE than in the general population. Previous studies have reported that renal disease, cytokines, inflammatory mediators, antiphospholipid antibodies, and SLE treatment contribute to the development of CVD in SLE.[10-13,28] Furthermore, the prevalence of atherosclerosis was reported to be elevated among patients with SLE.[2] Shang et al demonstrated that SLE was an independent risk factor for atherosclerosis and arterial stiffness.[3] Moreover, another study showed that SLE per se was a significant predictor for the development of CVD.[15] This finding agrees with our result showing that SLE per se is an independent risk factor for PAOD.

Our results revealed that the PAOD risk was high in young patients with SLE (particularly patients aged <34 years old) and was lower than that in elderly patients. A Swedish population-based study conducted by Bengtsson et al demonstrated that the risk of myocardial infarction or stroke in the SLE population was 1.27-fold higher than that in the general population, but among patients with SLE who were aged 40 to 49 years, this risk was 8-fold higher during a 7-year follow-up period.[8] Ward determined that in18 to 44-year-old patients with SLE, the risk of stroke was 1.75 times higher than that in age-matched controls. Moreover, Manzi et al demonstrated that the incidence of myocardial infarction in 35 to 44-year-old female patients with lupus was >50 times higher than that in females of a similar age from a population-based sample.[27]

Our study showed that PAOD was also strongly associated with traditional cardiovascular risk factors such as COPD and CAD. The inflammatory milieu of SLE leads to a dysregulation of lipid metabolism pathways, thus contributing to an increased risk of atherosclerotic disease among patients with SLE.[29,30] A Framingham Heart Study cohort showed an association with hyperlipidemia, but not with other traditional cardiovascular risk factors.[27] In a Spanish study, the final model revealed that only age was an independent variable of PAOD, although other cardiovascular risk factors such as DM, hypertension, hypercholesterolemia, or current smoking also tended to be present.[22] Bhatt et al identified a relationship between PAOD and dyslipidemia, but not age, DM, or hypertension.[31] Our results showed that the PAOD risk was higher in female patients with SLE than in male patients with the disease. In 2 large cohort studies, male patients with SLE had a 4-fold higher risk of developing CVD compared with female patients[10,32]; however, in both studies, nearly 90% of the patients with SLE were female. Nevertheless, in previous studies, sex showed nostatistically significant difference between groups in relation to PAOD occurrence among patients with SLE.[18-20,22]

Our results showed that the risk of developing PAOD was highest in the first year of follow-up and then decreased over time. A nationwide study indicated that the risk of CAD among patients with SLE decreased over time and was highest in the first year of follow-up (standardized incidence ratio (SIR) = 4.94, 95% CI 4.15–5.83).[33] Furthermore, the risk of hemorrhagic stroke among patients with SLE was highest in the first year and then decreased (SIR = 8.65 in the first year, 95% CI 3.92–16.5; SIR = 2.89 after 1–5 years, 95% CI 1.53–4.95).[5] Previous studies have demonstrated that immunosuppressive treatment reduced endothelial damage and improved patient condition in atherosclerosis.[34,35] This finding agrees with our result showing that the PAOD risk was higher in SLE patients without immunosuppressive treatment than in SLE patients with various treatments. Thus, we hypothesized that suppression of inflammation and disease control might reduce cardiovascular risk in patients with SLE.

The present study has several strengths. First, this is the first population-based study to analyze the association between SLE and PAOD. Second, we examined the incidence and risk factors for PAOD in patients with SLE in comparison with controls from the general population. Third, we adjusted for these confounding factors: DM, hypertension, hyperlipidemia, COPD, heart failure, CAD, stroke, asthma, smoking, and medication of steroid use, and warfarin. Despite possessing these strengths, our study had some limitations. First, the NHIRD cannot provide the individual SLE patient's laboratory data (such as disease activity in SLE and autoantibody of lupus anticoagulant). Therefore, we cannot do the further study about the potential relationship of lupus anticoagulant and peripheral artery occlusive disease. This is one of the study limitations. Second, we could have underestimated the incidence of PAOD because patients who exhibited no symptoms or presented with mild PAOD symptoms might have been overlooked. Finally, from the NHIRD, we could not obtain data related to cardiovascular risk factors including smoking habits, dietary habits, body mass index, and the extent of daily activity. To minimize the influence from smoking, we have used an alternative way and adjusted for smoking-related diseases (including chronic obstructive pulmonary disease, coronary artery disease, stroke, asthma) in our analysis. In addition, less than 5% women are smoker in Taiwan, the PAOD risk associated with SLE was greater for women.

In conclusion, the risk of PAOD is increased in patients with SLE, and SLE per se is an independent risk factor for PAOD. Attention must be devoted to the risk of PAOD among young patients with SLE because this risk is markedly elevated. Additional studies on disease screening and early intervention are necessary to prevent subsequent complications of PAOD in patients with SLE.