Inflammatory Bowel Disease Increases the
Risk of Peripheral
Arterial Disease
A Nationwide Cohort Study
Te-Yu Lin, MD, Yu-Guang Chen, MD, Cheng-Li Lin, MSc, Wen-Sheng Huang,
MD,
and Chia-Hung Kao, MD
INTRODUCTION
I
nflammatory bowel disease (IBD) is a group of inflammatory intestinal disorders mainly comprising 2 types of disease: Crohn’s disease (CD) and ulcerative colitis (UC).1 Theincidence and prevalence rates of IBD vary among countries and regions. Traditionally, the incidence of this disease is relatively stable in Western countries; however, the IBD incidence has been increasing in Asian countries in the past few years.2–6
Peripheral arterial disease (PAD) is the narrowing of
arteries other than those that supply blood to brain or heart, causing a considerable burden on health care systems worldwide.
7 The prevalence of peripheral vascular disease in the
general population is 12% to 14%, affecting up to 20% of people >70 years.8 According to the Taiwan National Health Insurance
(NHI) data analysis, the incidence of invasive PAD treatment is increasing.9 Traditional risk factors for PAD are an older age,
male sex, hypertension, diabetes, hyperlipidemia, obesity, smoking, and family history of vascular disease.10–12
Patients with IBD are vulnerable to vascular complications.
13 Chung et al observed that IBD patients exhibited a
high risk of developing deep vein thrombosis and pulmonary
embolism.14 In addition, recent studies have revealed an association
between IBD and arterial thromboembolic events.15–17
the relationship between IBD and the development of PAD are
scant. Therefore, we conducted a nationwide population-based cohort study to investigate whether IBD increases the risk of
developing PAD.
METHODS
Data Source
The National Health Insurance program was implemented in Taiwan on March 1, 1995 and covered >99% of the 25 million residents in Taiwan. In this study, data were accessed from the National Health Insurance Research Database
(NHIRD), which is managed and released for research purposes by the National Health Research Institutes. The Bureau of National Health Insurance (BNHI) scrambled the identification of insurants in the NHIRD to protect their personal information before releasing the data to researchers. A committee of the BNHI is responsible for randomly selecting and verifying the accuracy of claims. The International Classification of Disease, Ninth Revision, Clinical Modification (ICD-9-CM) was used to code diagnoses. This study was approved to fulfill the condition for exemption by the Institutional Review Board (IRB) of China Medical University (CMUH-104-REC2-115). The IRB also specifically waived the consent requirement.
Study Participants
Patients in the inpatient claims with newly diagnosed IBD, including Crohn’s disease (ICD-9-CM codes 555) and ulcerative colitis (ICD-9-CM codes 556) from 2000 to 2010 were
selected as the IBD cohort. The IBD diagnosis is made based on clinical history, physical examination, and objective findings from endoscopic, radiological, laboratory, and histological studies. The index date for patients with IBD was the date of first
hospitalization. The comparison cohort was randomly selected from inpatient claims without the history of IBD. For each IBD case, _4 comparisons were identified, frequency matched by sex, age (within 5 years), index year of patients in the IBD cohort, and comorbidities including diabetes, hypertension, hyperlipidemia, COPD, heart failure, CAD, stroke, obesity, chronic kidney disease, autoimmune disease, surgery of the intestine, laparotomy, and colorectal cancer. The control-to-case
ratio was set at _4:1 to enhance the power of statistical tests. In both cohorts, the patients who were diagnosed with PAD (ICD-9-CM codes 440.2, 440.3, 440.8, 440.9, 443, 444.22, 444.8, 447.8, and 447.9) and deep vein thrombosis (ICD-9-CM codes 453.8) before the index date, histories of coagulopathies (ICD-9-CM codes 286, 289.8), sickle-cell disease (ICD-(ICD-9-CM codes 282.6), age <20 years, or missing information, were excluded.
Outcome and Comorbidities
The main outcome was hospitalization with a new diagnosis of PAD during follow-up. Person-years of the follow-up
were calculated for each patient until hospitalization for PAD or censored for loss to follow-up, withdrawal from the insurance program, or the end of 2011.
Comorbidities were identified according to hospital admissions prior to the index date as potential confounding factors. These comorbidities included diabetes (ICD-9-CM code 250), hypertension (ICD-9-CM codes 401- 405), hyperlipidemia (ICD-9-CM code 272), chronic obstructive pulmonary disease (COPD) (ICD-9-CM codes 491, 492, and 496), heart failure CM code 428), coronary artery disease (CAD) (ICD-9-CM codes 410–414), stroke (ICD-9-(ICD-9-CM codes 430–438), obesity (ICD-9-CM code 278.0), chronic kidney disease CM codes 580–589), autoimmune disease (ICD-9-CM codes 710.0, 710.1, 710.2, 714), surgery of the intestine (ICD-9-CM procedure codes 45.0–45.9, 46.0–46.9, 48.0– 48.9), laparotomy (ICD-9-CM procedure codes 54.11, 54.19), and colorectal cancer (ICD-9-CM codes 153–154).
Statistical Analysis
Chi-squared tests were used to examine the differences in the distribution of sex, age, and comorbidities between the IBD and comparison cohorts. Student’s t test was used to examine the mean ages and mean follow-up period between both cohorts. The sex-, age-, and comorbidity-specific incidence density rates of PAD per 10,000 person-years of follow-up were calculated for each cohort. Univariable and multivariable Cox proportional hazard models were applied to measure the hazard ratios (HRs) and 95% confidence intervals (CIs) for PAD in the IBD cohort compared with those in the comparison cohort. The multivariable
Cox models were simultaneously adjusted for sex,
age, and comorbidities of diabetes, hypertension, hyperlipidemia, COPD, heart failure, CAD, stroke, obesity, chronic kidney
disease, autoimmune disease, surgery of the intestine, laparotomy, and colorecal cancer. Further analysis was performed to
assess the dose–response effect of IBD on the risk of PAD according to the average frequency of hospitalizations per year for IBD. The cumulative incidence curves of PAD were determined using the Kaplan–Meier method between the IBD cohort
and the comparison cohort. Their differences were estimated using the log-rank test. All analyses were performed using SAS, Version 9.3, computer software (SAS Institute Inc, Cary, NC). A 2-tailed P<0.05 was statistically significant.
RESULTS
Our data set included 54,832 participants including both the IBD patient and comparison cohorts. The distributions of gender, age, and comorbidity were similar in the both cohorts, except colorectal cancer. Both cohorts comprised more men than women (54.2% vs 45.8%). The mean age of the comparison cohort was 53 years and that of the IBD cohort was 53.1
years, with _37% of the patients in the age group of 40 to 64 years (Table 1). As shown in Figure 1, the cumulative incidence of PAD was higher in the IBD cohort than in the comparison cohort by the end of follow-up. Overall, the incidence of PAD was 1.24-fold higher in the IBD cohort compared with the comparison cohort (23 vs 18.6 per 10,000 person-years). After age, sex, and comorbidities were adjusted, the adjusted HR for developing PAD was 1.29-fold higher (95% CI¼1.22–1.37) for the IBD patients compared with the comparison cohort. The sex-specific adjusted HRs for PAD in the IBD cohort were statistically significant for both the women (HR: 1.25, 95% CI¼1.14–1.36) and men (HR: 1.35, 95% CI¼1.24–1.46) relative to those of the comparison cohort. The age-specific
incidence of PAD increased with age in both cohorts. The agespecific adjusted HRs for PAD in the IBD cohort were significantly
higher for all age groups, relative to those of the
comparison cohort. A comorbidity-specific analysis revealed a high incidence of PAD for patients with comorbidities in both
cohorts (Table 2).
Table 3 presents the risk of PAD according to annual mean
hospitalizations for the IBD patients, compared with the comparison cohort. The IBD patients presenting with 2 or more
hospitalizations per year exhibited a 27.5-fold increased risk of PAD development compared with the comparison cohort (95% CI¼18.7–40.4). Table 4 shows the incidence and HRs for PAD
associated with ulcerative colitis and Crohn’s disease in the IBD patients. The patients with UC exhibited a 32% higher risk of
PAD compared with the comparison cohort (adjusted
HR¼1.32, 95% CI¼1.22–1.43). The CD patients also showed a 27% higher risk of PAD compared with the comparison cohort (adjusted HR¼1.27, 95% CI¼1.18–1.38), which
was significant.
DISCUSSION
This is the first study to investigate the long-term risk of PAD in patients with IBD based on nationwide data. We
adjusted for several classical PAD risk factors in this population-based cohort study and determined that IBD is an independent risk factor for PAD. We observed that adults with IBD
had 1.29-fold risks of developing PAD compared with the comparison cohort. In addition, by using the frequency of
hospitalization as a surrogate index of IBD activity, we determined that IBD severity was significantly associated with the
risk of PAD (Table 3).
Although the patients with IBD in this study exhibited a higher prevalence of comorbidities and coexistent conditions associated with the development of PAD than the comparison cohort, IBD remained an independent risk factor for developing PAD after covariates were adjusted. The exactmechanisms predisposing patients with IBD to PAD are multifactorial. Long-term inflammation is known to predispose patients to vascular endothelial dysfunction and atherosclerosis.18,19 IBD
is characterized by an unusual chronic inflammation of the
colon and small intestine. Similarities between the pathophysiological processes in the colonic wall of IBD patients and the
process in the arterial wall exist during the progression of atherosclerosis, subsequently resulting in atherosclerotic plaque
rupture and thrombotic vascular events. Several reports have indicated that patients with IBD tend to exhibit lipid
profile derangement, which promotes atherosclerosis.20–23 The findings of these studies are
consistent with our epidemiological results.
Most of the IBD patients in this study were men, and PAD risks increased in both sexes with IBD. The age-stratified effect of IBD on PAD development was the highest in the patients<40 years. However, the incidence of PAD increased with age in the patients with IBD and those in the non-IBD cohorts, which is consistent with previous studies.8 As people age, they become
less active, and the function of the cardiopulmonary system deteriorates. Recent studies have shown that frailty, which is commonly associated with aging, can cause subclinical peripheral vascular disease to develop.24 In our study, study cohort
with comorbidities were shown to further increase the risk of PAD compared with the comparison group (adjusted
HR¼1.45). Among the comorbidities, there were several
well-known traditional risk factors for PAD such as hypertension, diabetes, and hyperlipidemia. Recent studies have shown that inflammatory diseases might increase risk of vascular complications.25–26
This study revealed a strong association between an
increased number of annual IBD-related medical admissions and the risk of developing PAD. IBD patients with high admission rates may represent increasing IBD flare-ups and severity of the active periods. The results showed that the systemic inflammatory burden with IBD patients may be a key determinant of atherosclerosis risk. Thus, aggressively controlling the inflammatory activity in IBD patients may
reduce the vascular complications. This study had potential limitations. First, the insurance data do not contain detailed information on smoking habits,
body mass index, or family history of PAD, all of which might be confounding factors in this study. Consequently, as with previous publications, we adjusted for several comorbidities of smoking-related diseases (including CAD, COPD, and stroke) to make the potential confounding effect of smoking on our study results minimally. These adjust models had been utilized
in previous studies. In addition, to adjust for the influence of body mass index, we included several traditional metabolic syndrome comorbidities such as hypertension, diabetes, and hyperlipidemia. Second, the evidence derived from a retrospective cohort study is generally lower in statistical quality
than that from randomized trials because of potential biases related to adjustments for confounding variables. Despite our meticulous study design and control measures for confounding factors, bias resulting from unknown confounders might have affected our results.
In conclusion, patients hospitalized for IBD exhibited an increased risk of PAD compared with the comparison cohort. We recommend that physicians carefully monitor vascular complication when caring for patients with IBD.
