Increased risk of acute coronary syndrome in patients with bronchiectasis: A population-based cohort study

Data source

The National Health Insurance (NHI) programme was launched in Taiwan. Taiwan has abundant medical resources and convenient transportation, and thus people in Taiwan can easily visit a clinic or hospital and incur only low medical expenditure. According to the National Health Insurance Administration (NHIA) of the Ministry of Health and Welfare, more than 99.5% of residents and more than 97% of healthcare institutions in Taiwan are part of the NHI programme (http://www.nhi.gov.tw/english/index.aspx). The Taiwan National Health Research Institutes established the Longitudinal Health Insurance Database (LHID), which contains the claims data of outpatient, inpatient, emergency and dental care, as well as prescription drugs. The NHIA encrypted the patient identities in the LHID before it was released for public use and research. We used the LHID2000, comprising data of 1 million randomly selected beneficiaries of the NHI programme in 2000. No statistically significant differences were observed in the distribution for age, sex and health care costs between the LHID2000 and NHI beneficiaries (http://nhird.nhri.org.tw/en/Data_Subsets.html#S3). This database has been used reliably in our previous studies.22, 23 Patient diagnoses in the LHID2000 are based on the International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM) codes. This cohort study was approved by the Ethics Review Board of China Medical University Hospital (CTU103-1/CMUH-REC-103-REC1-088).

Study patients

This study was a retrospective population-based cohort study. We identified 7156 patients diagnosed with bronchiectasis (ICD-9-CM code 494; for more than two times at health care institutions) from 2000 to 2010. Among the 7156 patients with bronchiectasis diagnosis, 3521 patients received further HRCT diagnosis to constitute the bronchiectasis cohort. The first bronchiectasis diagnosis date was defined as the entry date. Furthermore, we excluded patients below 18 years and above 90 years; those with incomplete information regarding sex or date of birth; and those diagnosed with bronchiectasis and ACS before the index date. In addition, we randomly selected 14 084 people without diagnosis of bronchiectasis from the general population and frequency matched them according to sex, age and entry year in a 4:1 ratio, thus forming the comparison cohort. The exclusion criteria for the comparison cohort were identical to those of the bronchiectasis cohort.

Outcome variables

The primary outcome was an ACS event (ICD-9-CM codes 410 indicating acute myocardial infarction and 411.1 indicating unstable angina). In Taiwan, the physicians made the diagnosis of ACS based on clinical symptoms, history, serial examinations of electrocardiogram and cardiac enzymes, cardiac echo and cardiac catheterization if available. We determined the follow-up person-years by calculating the interval from the entry date to the date on which any of the following events first occurred: ACS diagnosis, withdrawal from the NHI programme, death or December 31, 2010.

Covariates

Age was stratified into ≤34 years, 35–49 years, 50–64 years, 65–74 years and ≥75 years. The baseline comorbidities were hypertension (ICD-9-CM codes 401–405), diabetes (ICD-9-CM code 250), hyperlipidemia (ICD-9-CM code 272), stroke (ICD-9-CM codes 430–438), coronary heart disease (ICD-9-CM code 428), chronic obstructive pulmonary disease (COPD, ICD-9-CM codes 491, 492 and 496), primary ciliary dyskinesia (ICD-9-CM code 759.3), allergic bronchopulmonary aspergillosis (ICD-9-CM code 518.6) and non-tuberculosis mycobacterium (NTM, ICD-9-CM code 031). Furthermore, we stratified the study participants by the annual instances of respiratory infection-related emergency room (ER) visits and hospitalization to evaluate their effect on the risk of ACS.

Statistical analysis

All statistical analyses were conducted using SPSS 17.0 (SPSS Inc., Chicago, IL, USA). The mean age of patients in both cohorts were compared and tested using the Student t-test, and the distribution of demographics and comorbidities in both cohorts were measured and tested using the chi-square test, The overall age-, sex- and comorbidity-specific ACS incidence for both cohorts were evaluated. Univariable and multivariable Cox proportional hazard regression models were employed to measure and compare the crude and adjusted hazard ratios (HR) with 95% CI) of ACS in the cohorts.

The Kaplan–Meier analysis and log-rank test were used for assessing the differences in the ACS-free probability in both cohorts. A two-sided P value <0.05 was considered statistically significant.

Demographic characteristics and comorbidities in both cohorts

Figure 1 illustrates the categorization of the study participants. The current study included a total of 3521 patients with bronchiectasis and 14 084 comparison patients. Both cohorts exhibited similar sex and age distributions because of frequency matching. The mean age (±SD) was the same (63.63 ± 14.71 years) in both cohorts. The prevalence of stroke, CHD, COPD and NTM was higher in the bronchiectasis cohort than in the comparison cohort (27.2% vs 20.4%, 31.2% vs 23.3%, 65.1% vs 21.3% and1.5% vs 0%, respectively; Table 1).

ACS incidence and hazard ratio stratified by sex, age and comorbidities

The overall ACS incidence was higher in the bronchiectasis cohort than in the comparison cohort (13.49 vs 9.07 per 1000 person-years). After adjustment for sex, age and comorbidities, patients with bronchiectasis had a 1.40-fold higher risk of ACS than those without bronchiectasis (95% CI: 1.20–1.62). The ACS incidence was higher in men than in women in both cohorts. After we controlled potential covariates, the risk of ACS was a 1.31-fold risk higher in men than in women. The ACS incidence increased with age in both cohorts. After adjustment for covariates, the risk of ACS was the highest in adults aged 75 years and above (adjusted hazard ratio (HR) = 7.57, 95% CI: 3.35–17.13), followed by adults aged 65–74 years (adjusted HR = 6.02, 95% CI: 2.66–13.61) and adults aged 50–64 years (adjusted HR = 3.34, 95% CI: 1.47–7.57) compared with that in adults aged 34 years and below. The risk of ACS was significantly higher in patients with comorbidities than in those without comorbidities (adjusted HR = 1.99, 95% CI: 1.56–2.55; Table 2).

ACS risk in joint effect of bronchiectasis and comorbidity

Compared with non-bronchiectasis patients without comorbidity, the bronchiectasis patients without comorbidity exhibited a higher risk of subsequent ACS but did not reach statistical significance. The risk of ACS was higher in bronchiectasis patients with comorbidities than in those non-bronchiectasis patients without comorbidities (adjusted HR = 2.82, 95% CI: 2.15–3.72; Table 3).

ACS risk and annual instances of respiratory infection-related ER visits and hospitalization

Compared with the comparison cohort comprising patients with <1 instance of ER visit in a year, the risk of ACS was substantially increased in patients with bronchiectasis ≥3 instances of respiratory infection-related ER visits per year (adjusted HR = 5.46, 95% CI: 4.29–6.96). The patients with bronchiectasis who experienced ≥3 instances of respiratory infection-related hospitalization per year exhibited a 8.15-fold higher risk of ACS development than did the comparison cohort hospitalized <1 instance per year (adjusted HR = 8.15, 95% CI: 6.27–10. 61; Table 4).

ACS-free probability during follow-up periods

The cumulative ACS probability was significantly higher in the bronchiectasis cohort than in the comparison cohort (log-rank test, P < 0.0001; Fig. 2).