Risk of peripheral arterial occlusive
disease in patients with
rheumatoid arthritis
A nationwide population-based cohort study
Ya-Wen Chuang1; Mei-Ching Yu2; Cheng-Li Lin3,4; Tung-Min Yu1,5; Kuo-Hsiung
Shu1; Shih-Ting Huang1,5; Chia-Hung Kao5,6
Introduction
Rheumatoid arthritis (RA), a systemic inflammatory autoimmune disease, frequently shows complications of joint destruction and
various extra-articular manifestations. Previous studies have suggested that chronic inflammation is associated with endothelial
dysfunction and atherosclerosis (1-4). RA per se is considered a nontraditional risk factor for cardiovascular diseases (CVD) (4-6). The risk of CVD in RA is similar to the cardiovascular risk in diabetes mellitus (7-9). Previous studies have suggested that RA is associated with a higher risk of cardiovascular morbidity and mortality
than a general population (4, 5, 8-16). However, most studies have focused on the risk of cardiovascular events such as coronary artery disease (CAD), stroke, and heart failure (11-13, 15, 16). Peripheral arterial occlusive disease (PAOD) is a manifestation of atherosclerosis and is considered a cardiovascular risk equivalent to that of CAD. The prevalence of asymptomatic PAOD is
1.6 %–41.1 % in the global population (17-21). The risk of PAOD is higher in males, older people, those with a history of smoking, and those with diabetes mellitus, hypertension, hyperlipidaemia, chronic kidney disease, and chronic obstructive pulmonary disease (COPD) (17-20). Patients with RA occasionally show an impaired peripheral arterial function and are potentially associated
with an increased risk of PAOD (22-26). However, previous studies have had limited sample sizes and some of have examined only patients
with RA. Clinically, PAOD may be associated with leg ulcers,
leg amputation, and CVD. Thus, the early detection of a peripheral arterial dysfunction is necessary to prevent the progression of the
condition to arterial occlusion and subsequent complications. This nationwide cohort design study was conducted to investigate the risk of PAOD in patients with RA.
Methods
Data source
Taiwan implemented the compulsory National Health Insurance (NHI) program in 1995. The program currently covers more than 99% of the 23.75 million residents of Taiwan (http: //www.nhi.gov. tw/english/index.aspx).The National Health Research Institute (NHRI) audits and releases the National Health Insurance Research Database (NHIRD) for health care research. To ensure the
confidentiality of beneficiaries, the NHIRD, which contains registration files and the original medical claims data of all beneficiaries,
is encrypted using unique personal identifiers before being
released to the public. The diagnoses and procedures recorded in the NHIRD are coded according to the International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM).
This study was evaluated and approved by the Institutional Review Board of China Medical University and Hospital
(CMUH104-REC2–115).
Study patients
Patients newly diagnosed with RA (ICD-9-CM code 714) from 2000 to 2011 were identified from the Registry for Catastrophic Illnesses Patients, a subset of the NHIRD. The NHI classifies RA
as a catastrophic disease. Rheumatologists can apply for a catastrophic illness card for patients with RA who meet at least four
1987 American College of Rheumatology criteria. Patients with
RA who have a catastrophic illness card are exempt from the copayment of RA-related medical care. The index date was set as
the date of RA diagnosis. Patients aged < 20 years and those diagnosed with PAOD (ICD-9-CM codes 440.2, 440.3, 440.8,
440.9, 443, 444.22, 444.8, 447.8, and 447.9) before the index date were excluded. The non-RA cohort was randomly selected from among all NHI beneficiaries aged _ 20 years and who had no history of RA. The controls were frequency-matched at a ratio of 1: 1 according to age (in 5-year bands), sex, and index year of RA diagnosis. The same exclusion criteria were applied when selecting the non-RA cohort.
Taiwan launched a national health insurance (NHI) in 1995,
operated by a single-buyer, the government. Medical reimbursement specialists and peer review should scrutinise all insurance
claims. The diagnoses of PAOD were based on the ICD-9 codes
which were judged and determined by related specialists and physicians according to the standard clinical criteria. If these doctors or
hospitals make wrong diagnoses or coding, they will be punished with a lot of penalties. Therefore, the diagnoses and codes for acute pancreatitis and PAOD used in this study should be correct and reliable (27–29).
Outcome measures
The follow-up period for each participant was calculated as person-years from the year of RA diagnosis to the year that the participant was diagnosed with PAOD, withdrew from the NHI, died,
or was lost to follow-up, or until the end of 2011.
Comorbidities
Pre-existing comorbidities considered for each participant were diabetes mellitus 9-CM code 250), hypertension (ICD-9-CM codes 401–405), hyperlipidaemia (ICD-(ICD-9-CM code 272), COPD 9-CM codes 491, 492, and 496), heart failure (ICD-9-CM code 428), CAD (ICD-(ICD-9-CM codes 410–414), and stroke (ICD-9-CM codes 430–438).
Statistical analysis
The Chi-squared test was used to determine the differences in the distribution of sex, age, and comorbidities between the two cohorts, and the Student t-test was used to calculate and analyse the
mean age and follow-up period of both cohorts. We used the Kaplan-Meier method to estimate the cumulative incidence of PAOD
between the RA and non-RA cohorts, and assessed these differences using a log-rank test. The incidence densities of PAOD were
calculated in both cohorts. Univariate and multivariate Cox proportional hazards regression models were used to calculate hazard
ratios (HRs) and 95% confidence intervals (CIs) with stratification
based on sex, age, comorbidities, and follow-up period. The multivariate models were simultaneously adjusted for age, sex, and comorbidities, namely diabetes mellitus, hypertension, hyperlipidaemia,
COPD, heart failure, CAD, and stroke. Additional data was
risk factors on the risk of PAOD. All analyses were performed using SAS statistical software (Version 9.3, SAS Institute, Cary, NC, USA). The two-sided significance level was set at p <0.05.
Results
_
Table 1 shows the distributions of the demographic characteristics and comorbidities of the two cohorts. We enrolled 30,769 patients in the RA cohort and 30,741 patients in the non-RA cohort,and the groups exhibited similar age and sex distributions. Female patients were dominant (77.4 %) and 39.3% of the patients were aged < 49 years. The mean ages of the RA and non-RA cohorts were 53.7 ± 14.0 and 53.2 ± 14.4 years, respectively. Compared with the non-RA cohort, the RA cohort had a significantly higher prevalence of hypertension, hyperlipidaemia, COPD, heart failure, and CAD (all p <0.05).
The mean follow-up period was 5.45 ± 3.38 years in the RA cohort
and 5.56 ± 3.39 years in the non-RA cohort. The results of the Kaplan-Meier analysis showed that the RA cohort had a 2.38%
higher cumulative incidence of PAOD compared with the non-RA cohort (log-rank test, p <0.001) (
_
Figure 1).The overall incidence of PAOD in the RA cohort was 6.27 per 1,000 person-years and 1.69-fold higher than that in the non-RA cohort (3.71 per 1,000 person-years), with an adjusted HR of 1.73 (95 % CI = 1.57–1.91) (
_
Table 2).The sex-specific relative risk of PAOD was significantly higher for both women (adjusted HR = 1.80, 95% CI = 1.61–2.02) and men (adjusted HR = 1.49, 95% CI = 1.21–1.83). The PAOD incidence increased with age in both cohorts, but, the age-specific RA
to non-RA relative risk was the greatest for the aged _ 49 group (adjusted HR=3.23, 95% CI=2.55–4.09). Regardless of comorbidities, the RA cohort showed a higher adjusted HR of PAOD than
did the non-RA cohort.
Compared with the non-RA cohort patients without comorbidity, the RA cohort patients with diabetes mellitus exhibited a
higher risk of PAOD (adjusted HR = 4.04, 95% CI = 2.27–3.30), as did the RA cohort patients with hypertension (adjusted HR = 2.55, 1.93–3.36). Furthermore, compared with the non-RA cohort patients without comorbidity, the RA cohort patients with any six or
(adjusted HR = 10.1, 95% CI = 5.09–20.0), followed by those with any five comorbidities (adjusted HR = 8.23, 95% CI = 5.59–12.1), those with any four comorbidities (adjusted HR = 5.01, 95% CI = 3.72–6.75), those with any three comorbidities (adjusted HR =
4.93, 95% CI = 3.85–6.31), those with any two comorbidities (adjusted HR = 3.72, 95% CI = 2.96–4.68), and those with any one comorbidity (adjusted HR = 3.13, 95% CI = 2.54–3.87) (
_
Table 3).A stratified analysis of the follow-up period revealed that the risk of PAOD for the RA cohort compared with the non-RA cohort peaked during the first follow-up year (adjusted HR = 2.42, 95% CI = 1.92–3.05) and declined over 4-5 years of follow-up (adjusted HR = 1.45, 95% CI = 1.17–1.81). The risk of PAOD remained even beyond five years of follow-up (
_
Table 4).Discussion
This is the first nationwide population-based study to determine the incidence and risk factors for PAOD in patients with RA. Our findings reveal a 1.73-fold increased risk of PAOD in the RA cohort compared with the non-RA cohort. A higher risk of PAOD
was observed in the RA cohort, particularly in those at a young age and those without comorbidity. In addition, a markedly increased risk was observed in the patients with RA and cardiovascular risk factors.
In the present study, the RA cohort exhibited a significantly higher risk of PAOD compared with the non-RA cohort. The results of numerous previous studies support our findings. In one
cohort study of RA, the 30-year cumulative incidence of peripheral artery disease was found to be 16.1% (24). Another study suggested that the incidence of CVD in patients with RA was
3.17-fold higher than that in participants without RA (5). Moreover, some studies have demonstrated that the prevalence of
PAOD in patients with RA as exceeding that in a general population (22, 27).
Our results showed a higher risk of PAOD in younger patients
with RA, particularly those aged < 49 years and those without comorbidity. Previous studies have reported that patients with RA
who are aged 25-54 years had a higher incidence rate of CVD compared with those aged > 54 years (5). Solomon et al. found that the highest incidence rate of CVD in patients with RA was
among those aged 18-49 years (31). Our findings are consistent with these data and suggest that RA itself may play a major role in the development of PAOD.
In our study, both the men and women with RA had a significant risk of PAOD; women had a higher risk compared with men. One study showed that women have a higher incidence rate of CVD than men did (31). Another study suggested that both men and women are at an increased risk of congestive heart failure, with a higher incidence in women than in men (16). Gonzalez et al. indicated that men with RA are at a lower risk of CVD compared with men without RA (12). However, some studies have suggested that men are at a higher risk of CVD than women (5, 6, 28,
32).Our results showed that the risk of PAOD in patients with RA
was the highest in the first year after RA diagnosis and declined afterward. RA is a systemic inflammatory disease, with early stages
of the disease leading to atherosclerotic changes (4, 29, 30). Georgiadis et al. suggested that atherosclerotic changes in patients with
RA improved after a one-year treatment of disease-modifying antirheumatic drugs (31). Numerous studies have demonstrated that the treatment for RA reduces cardiovascular disease risk (28, 31-34). We hypothesised that the treatment and control of RA activity may lower the risk of atherosclerosis and CVD.
Furthermore, our results revealed a markedly high risk of
PAOD in patients with RA and traditional cardiovascular risk factors such as diabetes mellitus, hyperlipidaemia, and hypertension. Arida et al. reported that RA was able to lead to atheromatosis in patients with no traditional CVD risk factors (35) and independently increased the incidence of CV events (5). The present study
demonstrated that coexistence with other traditional risk factors, such as diabetes mellitus, hypertension, hyperlipidaemia, COPD, heart failure, CAD, or stroke, increased the risk of PAOD in the
patients with RA. Several studies support our finding that an increased risk of CVD in patients with RA is attributable to traditional
CVD risk factors such as hypertension, diabetes mellitus,
smoking, and hyperlipidaemia, which are more frequent and severe in patients with RA than in the general population (13, 14).
Kremers et al. suggested that the more traditional risk factors comorbid with RA, the higher the risk of CVD (40).
The present study has several strengths. First, this is the first population-based study to analyse the association between RA and PAOD. A large sample size with age- and sex-matched controls
and a long follow-up period increased the validity of the results. Second, we compared the incidence and risk factors for PAOD in
patients with RA against general population controls and analyzed the presence of any trend. Third, we adjusted for confounding factors, namely diabetes mellitus, hypertension, hyperlipidaemia,
COPD, heart failure, CAD, and stroke. Fourth, we analysed synergistic effects of RA with multiple comorbidities. These results precisely indicated the risk of PAOD in patients with RA and multiple comorbidities.
Despite these strengths, the study has some limitations. The study data were retrieved from the NHIRD and lacked information on critical cardiovascular risk factors such as daily activities,
obesity, and smoking and dietary habits. Thus, we could not adjust for these confounding factors. In the general population, smoking is a powerful risk factor for PAOD (41). However, previous studies
have suggested that the prevalence of smoking did not differ significantly between patients with RA and controls (7, 36). Some
studies have shown that patients with RA have an increased prevalence of smoking compared with patients without RA; however,
the risk of CVD did not differ significantly between them (14). Another study suggested that patients with RA who smoke were at a lower risk of CVD compared with people without RA (12). In the present study, smoking-related disease, including COPD, CAD, and stroke, were adjusted to minimise the potential influence of smoking, an approach applied in a previous study (3, 37). Several studies have shown that the prevalence and risk of CVD did not differ significantly between patients with RA and controls depending on the presence of obesity, inactivity, or metabolic syndrome
(8, 9, 11, 12, 32, 36). We used hypertension, diabetes mellitus, and hyperlipidaemia for indicating obesity and metabolic syndrome. Second, we could not access the listed drug and disease activity of the patients with RA. A study indicated that adding markers for RA severity could improve the accuracy of CVD risk prediction (6). Several studies have demonstrated that antirheumatic drug therapy attenuates the risk of CVD (31-34). We hypothesised that
the incidence of PAOD in patients with RA may be underestimated and that the significant consequence may intensify the association between RA and PAOD.
In conclusion, patients with RA are at an increased risk of PAOD, and RA per se is an independent risk factor for CVD. We recommend that rheumatologists pay more attention to the risk of PAOD in patients with RA. Additional studies on screening and
early intervention are required to prevent subsequent complications of PAOD in patients with RA.
