Pulmonary tuberculosis increases the risk
of
pulmonary thromboembolism: a
nationwide
population-based cohort study
Wei-Sheng Chung1,2,3; Cheng-Li Lin4; Yung-Fu Chen3; Wu-Huei Hsu2;
Chia-Hung Kao2,5*
Dear Sirs,
Increasing interest in the relationship between infection and pulmonary thromboembolism (PE) has reported (1–3).
However, the incidence of PE in tuberculosis (TB) patients remains unclear. Therefore, we conducted a nationwide retrospective cohort study to explore the relationship
between TB and PE in Taiwan. This nationwide population-based cohort study is
the first to examine whether TB increases the risk of PE in an Asian population. A population-based cohort study was conducted by using data from the Taiwan National Health Insurance Research Database (NHIRD) that is managed by the
National Health Research Institutes (NHRI). This study was approved by the Ethics Review Board at China Medical University (CMU-REC-101–012). We
identified a study cohort consisting of patients newly diagnosed with TB
(ICD-9-CM 011–012) aged _ 20 years from 2000 to 2010 in the NHIRD. The TB diagnosis
date was defined as the index date. For each TB case, participants from all National Health Insurance beneficiaries free from TB were randomly selected as the controls and were frequency matched according to sex, age (every 5-year span), and
index year. The control-to-case ratio was 4:1. In both cohorts, we excluded patients diagnosed with PE (ICD-9-CM 415.1) and deep vein thrombosis (ICD-9-CM 453.8) before the index date, and participants whose records contained incomplete age or sex information. The preexisting comorbidities included atrial fibrillation
(ICD-9-CM 427.31), hypertension (ICD-(ICD-9-CM
401–405), diabetes (ICD-9-CM 250), hyperlipidaemia (ICD-9-CM 272), cerebrovascular
accidents (CVA) (ICD-9-CM
430–438), heart failure (ICD-9-CM 428), lower leg fracture or surgery (ICD-9-CM 820, 821; 823; 81.51, 81.52, 81.53, 81.54) and cancer (140–208). The TB patient cohort and the control cohort were followed
from the index date to December 31, 2011, until the diagnosis of PE, or withdrawal from the NHI program, whichever came first.
The study cohort consisted of 9,985 newly diagnosed TB patients and 39,938 non-TB controls. The mean ages (± standard deviation) of the non-TB and TB cohorts were 58.8 ± 19.9 years and 59.4 ± 20.0 years, respectively. Compared with the non-TB controls, the TB patients had a greater probability of being diagnosed with
atrial fibrillation (2.09 % vs 1.55 %), hypertension (43.1 % vs 40.9 %), diabetes (23.4 %
failure (7.75 % vs 4.74 %), lower leg fracture or surgery (3.50 % vs 2.26 %), and cancer (12.0 % vs 6.00 %) (data not shown). The incidence of PE in the TB cohort
was 2.90-fold greater than that in the
non-TB cohort (2.10 vs 0.72 per 10,000 personyears). The TB patients exhibited a
2.46-fold adjusted hazard ratio (HR) of PE development compared with the comparison cohort (95 % confidence interval [CI] =
1.10–5.51). After adjusting for potential covariates, heart failure (adjusted HR = 4.31,
95 % CI = 1.51–12.3) and cancer (adjusted HR = 3.39, 95 % CI = 1.32, 8.72) showed an independent risk factor of developing PE (
_
Table 1).Clayton et al. conducted a case-control
study and determined that respiratory infection indicated a 2.5-fold adjusted odds
ratio for PE (4). Clayton et al. did not focus on specific bacterial agents that caused PE. Dentan et al. used the Premier Hospitalization Database in the United States and observed that adults with TB had a 1.55-fold
greater risk of venous thromboembolism than those without TB (5). However, the TB incidence rate is 3.2 cases per 105 people in the United States, which is lower than 55 cases per 105 people in Taiwan (6,
7). TB incidence and PE prevalence discrepancies exist among racial differences
(8).
Mycobacterium tuberculosis infections
result in chronic granulomatous inflammation, which may be associated with haemostatic changes and a hypercoagulable
status (9). Chronic infection and inflammation may modulate thrombotic responses
by upregulating the activation procoagulants, downregulating that of anticoagulants,
and suppressing fibrinolysis (10). Heart failure exhibited an independent risk factor of PE development. The increased risk of PE observed with heart failure
may be attributed to reduced flow
caused by low cardiac output and abnormalities of haemostasis, platelet function,,
and endothelial function (11). Cancer also
showed an independent risk factor of developing PE in our study. Cancer and its
treatments have been well-recognised risk factors for venous thromboembolism (12). The strength of this study is that it provides a nationwide population-based cohort study that revealed the effects of TB on the increased risk of subsequent PE events. However, some limitations must be considered when interpreting the results. First, the NHIRD does not provide detailed information, such as records of cigarette smoking, alcohol consumption, body mass index, physical activity levels, socioeconomic status, and family history, which are
all potential confounding factors for this study. However, these factors may be randomly
distributed in these two large cohorts. Second, the outcome measured was symptomatic PE; therefore, the incidence of PE may have been underestimated. Third, the healthcare claims data might contain a potential misclassification bias of primary outcomes even though the auditing mechanism of the Bureau of National
Health Insurance can help minimise diagnostic uncertainty and misclassification
hormone replacement therapy and the use of contraceptive, anticoagulant, and antiplatelet drugs, may also have influenced
the primary outcomes of this study. In summary, this study shows that TB
patients exhibit an elevated risk of PE development. Despite effective treatments
and available cures, TB remains an endemic disease and a major health problem
in Taiwan. To holistically care for TB patients, the government and clinicians
should not only focus on successful treatment but also strive to reduce PE risk factors.
