- Email: [email protected]
Nation-based case-control study investigating the relationship between oral corticosteroids use and pulmonary tuberculosis
EJINME-03560; No of Pages 5 European Journal of Internal Medicine xxx (2017) xxx–xxx
Contents lists available at ScienceDirect
European Journal of Internal Medicine journal homepage: www.elsevier.com/locate/ejim
Nation-based case-control study investigating the relationship between oral corticosteroids use and pulmonary tuberculosis Shih-Wei Lai a,b, Cheng-Li Lin a,c, Kuan-Fu Liao d,e,f,⁎ a
College of Medicine, China Medical University, Taichung, Taiwan Department of Family Medicine, China Medical University Hospital, Taichung, Taiwan Management Office for Health Data, China Medical University Hospital, Taichung, Taiwan d Graduate Institute of Integrated Medicine, China Medical University, Taichung, Taiwan e College of Medicine, Tzu Chi University, Hualien, Taiwan f Department of Internal Medicine, Taichung Tzu Chi General Hospital, Taichung, Taiwan b c
a r t i c l e
i n f o
Article history: Received 18 April 2017 Received in revised form 20 May 2017 Accepted 22 May 2017 Available online xxxx Keywords: Oral corticosteroids Pulmonary tuberculosis Case-control study Taiwan National Health Insurance Program
a b s t r a c t Background and objective: No published formal study focuses on the relationship between oral corticosteroids use and pulmonary tuberculosis in Taiwan. The aim of the study was to investigate whether oral corticosteroids use was associated with increased odds of pulmonary tuberculosis. Methods: The retrospective, nation-based, case-control study was conducted to analyze the database of the Taiwan National Health Insurance Program between 2000 and 2013. Subjects aged 20 to 84 years with newly diagnosed pulmonary tuberculosis were defined as the cases (n = 6021). Randomly selected subjects without pulmonary tuberculosis aged 20 to 84 years were defined as the controls (n = 6021). Subjects who never had a prescription for oral corticosteroids were defined as never use. Subjects who ever had a prescription for oral corticosteroids were defined as ever use. Results: After adjustment for confounders, the adjusted odds ratio of pulmonary tuberculosis was 1.91 for subjects with even use of oral corticosteroids (95% confidence interval 1.73, 2.11), comparing with never use. The adjusted odds ratio of pulmonary tuberculosis was 1.03 for subjects with increasing cumulative dose of oral corticosteroids for per milligram (95% confidence interval 1.02, 1.03), comparing with never use. In addition, the adjusted odds ratios of pulmonary tuberculosis were 1.83 (95% CI 1.65, 2.02) for subjects with cumulative duration of oral corticosteroids b 3 months, and 2.12 (95% CI 1.87, 2.40) for subjects with cumulative duration ≥ 3 months, comparing with never use. Conclusion: Though the finding is not unexpected, use of oral corticosteroids is significantly associated with 1.9fold increased odds of pulmonary tuberculosis, independent of other risk comorbidities. There are dose-dependent and duration-dependent effects of oral corticosteroids use on the risk of pulmonary tuberculosis. © 2017 Published by Elsevier B.V. on behalf of European Federation of Internal Medicine.
1. Introduction Tuberculosis remains one of the deadliest infectious diseases worldwide. WHO reported that there were about 9.0 million people developing tuberculosis and 1.5 million people dying from this disease in 2013 [1]. Current evidence indicates that numerous risk factors are found to be associated with development of pulmonary tuberculosis, including smoking, diabetes mellitus, malnutrition, alcoholism, human immunodeficiency virus infection, and living or working in crowded space [2–4].
⁎ Corresponding author at: Department of Internal Medicine, Taichung Tzu Chi General Hospital, No.66, Sec. 1, Fongsing Road, Tanzi District, Taichung City 427, Taiwan. E-mail address: [email protected] (K.-F. Liao).
Current evidence indicates that use of corticosteroids is associated with the increased susceptibility to tuberculosis due to its inhibition of immune functions [5–7]. In Liao et al.'s study in Taiwan, the overall incidence of tuberculosis was 68 per 100,000 population from 2004 to 2008 [8]. To date, no published formal study focuses on the relationship between use of oral corticosteroids and pulmonary tuberculosis in Taiwan. Given that tuberculosis remains a high prevalence in Taiwan, we make a rational hypothesis that use of oral corticosteroids may be associated with increased risk of pulmonary tuberculosis in Taiwan. If the relationship is positive, physicians should be aware of the potential risk of pulmonary tuberculosis in patients with use of oral corticosteroids. Therefore, a nation-based, retrospective, case–control study was conducted to analyze the database of the Taiwan National Health Insurance Program to investigate whether (1) use of oral corticosteroids was
http://dx.doi.org/10.1016/j.ejim.2017.05.020 0953-6205/© 2017 Published by Elsevier B.V. on behalf of European Federation of Internal Medicine.
Please cite this article as: Lai S-W, et al, Nation-based case-control study investigating the relationship between oral corticosteroids use and pulmonary tuberculosis, Eur J Intern Med (2017), http://dx.doi.org/10.1016/j.ejim.2017.05.020
2
S.-W. Lai et al. / European Journal of Internal Medicine xxx (2017) xxx–xxx
associated with increased odds of pulmonary tuberculosis; (2) there were dose-dependent and duration-dependent effects of oral corticosteroids use on the risk of pulmonary tuberculosis? 2. Methods 2.1. Data source and study design Taiwan is an independent country with N23 million persons. The Taiwan National Health Insurance Program launched in March 1995, and now it has covered around 99% of the persons living in Taiwan [9]. The details of the program have been written in previous studies [10–14]. A nation-based, retrospective, case–control study was conducted to analyze the database of the Taiwan National Health Insurance Program. The Research Ethics Committee of China Medical University and Hospital in Taiwan approved the study (CMUH-104-REC2-115). 2.2. Selection of cases and controls We selected subjects aged 20 to 84 years with newly diagnosed pulmonary tuberculosis (the International Classification of Diseases (ICD) 9th Revision, ICD-9 codes 010, 011, 012, and 018) between 2000 and 2013 as the cases with pulmonary tuberculosis. The index date was defined as the date of cases being diagnosed with pulmonary tuberculosis. We randomly selected subjects without pulmonary tuberculosis aged 20 to 84 years from the same database as the controls. Both cases and controls were matched with sex, age (every 5-year interval), comorbidities, and the year of index date (Fig. 1). 2.3. Potential confounders Some medical conditions which could be related to pulmonary tuberculosis were included as follows: alcohol-related disease, chronic kidney disease, chronic obstructive pulmonary disease, diabetes mellitus, human immunodeficiency virus infection, gastrectomy, pneumoconiosis, as well as chronic liver disease including cirrhosis, hepatitis
B infection, hepatitis C infection, and other chronic hepatitis. To increase the validity of analysis, subjects who had the same diagnosis for 3 consecutive clinical visits in the ambulatory care and/or one episode of hospitalization diagnosis could be included in the study. Principal diagnosis and secondary diagnosis were used equally. Therefore, pulmonary tuberculosis and other medical conditions were documented for 3 or more records in the ambulatory care and/or one record during hospitalization. All comorbidities were diagnosed according to ICD-9 codes. The accuracy of ICD-9 codes has been validated in previous studies [10–20]. 2.4. Assessment of corticosteroids exposure It is relatively difficult to calculate the doses of topical corticosteroids use or inhaled corticosteroids use. It is also very rare for patients with long-term use of injected corticosteroids at outpatient department. In order to quantify the dosage, only oral corticosteroids were included for detailed analysis. Topical, inhaled, and injected corticosteroids were collected together as other forms of corticosteroids for adjustment. Oral corticosteroids in Taiwan were available as follows: cortisone, dexamethasone, fludrocortisone, methylprednisolone, prednisolone, and triamcinolone. To investigate the relationship between oral corticosteroids use and pulmonary tuberculosis, prescription history of oral corticosteroids before the index date was included. To diminish the biased results, subjects whose final prescriptions for oral corticosteroids were filled N 12 months before the index date were excluded from the study. Therefore, only subjects whose final prescriptions for oral corticosteroids were filled within 12 months before the index date were included. Subjects who never had a prescription for oral corticosteroids were defined as never use. Those who ever had a prescription for oral corticosteroids were defined as ever use. 2.5. Statistical analysis Distributions of sex, age, oral corticosteroids use, other forms of corticosteroids use, and comorbidities between the cases and controls were compared by the Chi-square test for categorized variables, and the t-test
Fig. 1. Flow diagram of the study revealing the selection of cases with pulmonary tuberculosis and controls.
Please cite this article as: Lai S-W, et al, Nation-based case-control study investigating the relationship between oral corticosteroids use and pulmonary tuberculosis, Eur J Intern Med (2017), http://dx.doi.org/10.1016/j.ejim.2017.05.020
S.-W. Lai et al. / European Journal of Internal Medicine xxx (2017) xxx–xxx Table 1 Information and comorbidities between pulmonary tuberculosis cases and controls. Non-tuberculosis Tuberculosis N = 6021 N = 6021 Characteristic Sex Female Male Age group (years) 20–39 40–64 65–84 Age (years), mean (standard deviation)† Duration of exposure to oral corticosteroids (days), mean (standard deviation)† Ever use of oral corticosteroids Ever use of other forms of corticosteroids Comorbidities Alcohol-related disease Chronic kidney disease Chronic liver disease Chronic obstructive pulmonary disease Diabetes mellitus Human immunodeficiency virus infection Gastrectomy Pneumoconiosis
n
(%)
n
(%)
P value⁎ 0.34
1709 4312
(28.4) (71.6)
1756 4265
(29.2) (70.8) 0.38
876 2154 2991 60.6
(14.6) (35.7) (49.7) (16.7)
894 2212 2915 60.4
(14.9) (36.7) (48.4) (16.8)
67.8
(288.0)
102.5 (375.8) b0.001
2332 2274
(38.7) (37.8)
3230 1781
(53.7) (29.6)
b0.001 b0.001
439 311 942 2076
(7.29) (5.17) (15.7) (34.5)
487 367 1016 2145
(8.09) (6.10) (16.9) (35.6)
0.10 0.03 0.07 0.19
864 16
(14.4) (0.27)
984 29
(16.3) (0.48)
0.002 0.05
22 173
(0.37) (2.87)
59 220
(0.95) (3.65)
b0.001 0.02
0.40
Data are presented as the number of subjects in each group with percentages given in parentheses, or mean with standard deviation given in parentheses. ⁎Chi-square test, and †t-test comparing subjects with and without pulmonary tuberculosis.
for continuous variables. Initially, all variables were included in the univariable unconditional logistic regression model. Variables found to be statistically significant in the univariable model were further examined in the multivariable unconditional logistic regression model. The odds ratio (OR) and 95% confidence interval (CI) were used to investigate the correlation of oral corticosteroids use with pulmonary tuberculosis. We further conducted the analysis to investigate whether there were dose-dependent and duration-dependent effects of oral corticosteroids use on the risk of pulmonary tuberculosis. All data processing and statistical analyses were performed with the SAS software version
3
9.2 (SAS Institute, Inc., Cary, North Carolina, USA). A two-tailed P value b 0.05 was considered statistically significant. 3. Results 3.1. Characteristics of the study population Table 1 reveals the distributions of sex, age, oral corticosteroids use, other forms of corticosteroids use, and comorbidities between the cases and controls. The study included 6021 cases with pulmonary tuberculosis and 6021 controls, with similar distributions of sex and age. The mean ages (standard deviation) of the study subjects were 60.4 (16.8) years for the cases and 60.6 (16.7) years for the controls (t-test, P = 0.40), without statistic significance. The mean durations (standard deviation) of exposure to oral corticosteroids were 102.5 (375.8) days for the cases and 67.8 (288.0) days for the controls (t-test, P b 0.001), with statistic significance. The cases had a statistically higher proportion of ever use of oral corticosteroids than the controls (53.7% vs. 38.7%, Chisquare test, P b 0.001). The cases had statistically higher proportions of chronic kidney disease, diabetes mellitus, gastrectomy, and pneumoconiosis than the controls (Chi-square test, P b 0.05). 3.2. Association of pulmonary tuberculosis with oral corticosteroids use and comorbidities After adjustment for other forms of corticosteroids use, chronic kidney disease, diabetes mellitus, gastrectomy, and pneumoconiosis, the multivariable unconditional logistic regression model revealed that the adjusted OR of pulmonary tuberculosis was 1.91 for subjects with even use of oral corticosteroids (95% CI 1.73, 2.11), comparing with never use (Table 2). In addition, the adjusted ORs of pulmonary tuberculosis were 1.17 (95%CI 1.06, 1.29) for subjects with history of diabetes mellitus, and 2.55 (95%CI 1.56, 4.19) for subjects with history of gastrectomy, respectively, comparing with those without such a history (Table 2).
3.3. Association of pulmonary tuberculosis with cumulative dosage of oral corticosteroids use We conducted an analysis for the dose-dependent effect of oral corticosteroids on the risk of pulmonary tuberculosis. After adjustment for confounders, the adjusted OR of pulmonary tuberculosis was 1.03 for subjects with increasing cumulative dose of oral corticosteroids for per milligram (95% CI 1.02, 1.03), comparing with never use (Table 3).
Table 2 Odds ratio and 95% confidence interval of pulmonary tuberculosis associated with corticosteroids use, other forms of corticosteroids use, and comorbidities. Adjusteda
Crude Variable
OR
(95%CI)
OR
(95%CI)
Sex (male vs. female) Age (per one year) Oral corticosteroids (never use as a reference) Ever use Other forms of corticosteroids (never use as a reference) Ever use Comorbidities (yes vs. no) Alcohol-related disease Chronic kidney disease Chronic liver disease Chronic obstructive pulmonary disease Diabetes mellitus Human immunodeficiency virus infection Gastrectomy Pneumoconiosis
0.96 1.00
(0.89, 1.04) (0.99, 1.00)
– –
– –
1.83
(1.70, 1.97)
1.91
(1.73, 2.11)
0.69
(0.64, 0.75)
1.08
(0.97, 1.19)
1.12 1.19 1.09 1.05 1.17 1.82 2.70 1.28
(0.98, 1.28) (1.02, 1.39) (0.99, 1.21) (0.98, 1.13) (1.06, 1.29) (0.99, 3.35) (1.65, 4.40) (1.05, 1.57)
– 1.07 – – 1.17 – 2.55 1.16
– (0.92, 1.26) – – (1.06, 1.29) – (1.56, 4.19) (0.95, 1.43)
a Variables found to be statistically significant in the univariable model were further examined in the multivariable model. Adjusted for other forms of corticosteroids use, chronic kidney disease, diabetes mellitus, gastrectomy, and pneumoconiosis.
Please cite this article as: Lai S-W, et al, Nation-based case-control study investigating the relationship between oral corticosteroids use and pulmonary tuberculosis, Eur J Intern Med (2017), http://dx.doi.org/10.1016/j.ejim.2017.05.020
4
S.-W. Lai et al. / European Journal of Internal Medicine xxx (2017) xxx–xxx
Table 3 Association of pulmonary tuberculosis with cumulative dosage of oral corticosteroids use. Variable
Case number/control number
Crude OR
(95% CI)
Adjusted ORa
(95% CI)
Never use of oral corticosteroids as a reference Cumulative dosage of oral corticosteroids use (increase in dosage for per mg)
2791/3689 3230/2332
1.00 1.03
(reference) (1.02, 1.03)
1.00 1.03
(reference) (1.02, 1.03)
a Variables found to be statistically significant in the univariable model were further examined in the multivariable model. Adjusted for other forms of corticosteroids use, chronic kidney disease, diabetes mellitus, gastrectomy, and pneumoconiosis.
3.4. Association of pulmonary tuberculosis with cumulative duration of oral corticosteroids use We conducted an analysis for the duration-dependent effect of oral corticosteroids on the risk of pulmonary tuberculosis. After adjustment for confounders, the adjusted ORs of pulmonary tuberculosis were 1.83 (95% CI 1.65, 2.02) for subjects with cumulative duration of oral corticosteroids b3 months, and 2.12 (95% CI 1.87, 2.40) for subjects with cumulative duration of oral corticosteroids ≥3 months, comparing with never use (Table 4). 3.5. Interaction effects on the risk of pulmonary tuberculosis between oral corticosteroids use and comorbidities Table 5 reveals the risk of pulmonary tuberculosis stratified by oral corticosteroids use and comorbidities. When comparing with those with never use of oral corticosteroids and without comorbidities including chronic kidney disease, diabetes mellitus, gastrectomy, and pneumoconiosis, the adjusted OR of pulmonary tuberculosis was 1.98 (95% CI 1.79, 2.20) among subjects with use of oral corticosteroids alone and without comorbidities. The adjusted OR increased to 2.04 (95% CI 1.78, 2.34) among subjects with ever use of oral corticosteroids and with any comorbidity. 4. Discussion In this study, a strong relationship was noted between oral corticosteroids use and the risk of pulmonary tuberculosis (adjusted OR 1.91, Table 2). This finding is compatible with Jick et al.'s case-control study in the United Kingdom showing that current use of corticosteroids was associated with increased odds of tuberculosis (adjusted OR 4.9, 95%CI 2.9–8.3) [21]. We also found that the risk of pulmonary tuberculosis was strongly related to the cumulative dosage of oral corticosteroids use, with higher risk in patients using higher dose (adjusted OR1.03, Table 3). This finding is compatible with previous studies showing that high cumulative dose of corticosteroids use was an important risk factor for developing tuberculosis [22,23]. We think that there seems to be a dose-dependent effect of oral corticosteroids use on the risk of pulmonary tuberculosis. We also found that the risk of pulmonary tuberculosis was strongly related to the cumulative duration of oral corticosteroids use, with higher risk in patients using a longer period (adjusted OR 1.83 for b3 months vs. adjusted OR 2.12 for ≥3 months, Table 4). This finding is
compatible with Chan et al.'s suggestion showing that patients on prolonged corticosteroids use might be at risk for tuberculosis [24]. We think that there seems to be a duration-dependent effect of oral corticosteroids use on the risk of pulmonary tuberculosis. That is, the longer use of oral corticosteroids, the higher the risk of pulmonary tuberculosis. The cases and the controls were matched with comorbidities, but the cases remained to have statistically higher proportions of chronic kidney disease, diabetes mellitus, gastrectomy, and pneumoconiosis than the controls. To diminish the confounding effects caused by comorbidities, we made a further analysis stratified by oral corticosteroids use and comorbidities. We found that subjects with use of oral corticosteroids alone and without comorbidities were associated with 1.98-fold increased odds of pulmonary tuberculosis, comparing with those with never use of oral corticosteroids and without comorbidities (Table 5). This finding indicates that the risk of pulmonary tuberculosis associated with oral corticosteroids use is independent of other risk comorbidities. Oral corticosteroids use has a unique effect on the risk of pulmonary tuberculosis. Diabetes mellitus and gastrectomy have been found to be associated with increased risk of pulmonary tuberculosis. That is why we included diabetes mellitus and gastrectomy for adjustment [14,25]. We found that the adjusted ORs of pulmonary tuberculosis were 1.17 for subjects with diabetes mellitus, and 2.55 for subjects with gastrectomy. In addition, impairment of immune functions in diabetes mellitus makes diabetic patients more susceptible to various infections [26]. Gastrectomy-related poor nutrition can result in secondary immunodeficiency which makes gastrectomized patients also more susceptible to various infections [27]. Similarly, long-term use of oral corticosteroids can lead to inhibition of immune functions that correlates with an increased susceptibility to various infections maybe including pulmonary tuberculosis [5–7]. Some limitations of the present study deserve comments. First, due to the natural limitation of this database, there was no specific record showing the reasons of the prescription for oral corticosteroids. The underlying diseases for oral corticosteroids therapy could be associated with the risk of pulmonary tuberculosis and could play a role in the onset of active pulmonary tuberculosis. Furthermore, it indicates a future research direction about the relationship between the underlying diseases for oral corticosteroids therapy and pulmonary tuberculosis. Second, alcohol consumption and smoking have been found as two risk factors for pulmonary tuberculosis [28]. Due to the same limitation, the histories of alcohol consumption and smoking were not recorded. Therefore, alcohol-related disease was included instead of alcohol
Table 4 Association of pulmonary tuberculosis with cumulative duration of oral corticosteroids use. Variable
Case number/control number
Crude OR
(95% CI)
Adjusted ORa
(95% CI)
Never use of oral corticosteroids as a reference Cumulative duration of oral corticosteroids use b3 months ≥3 months
2791/3689
1.00
(reference)
1.00
(reference)
2133/1624 1097/708
1.74 2.05
(1.60, 1.88) (1.84, 2.28)
1.83 2.12
(1.65, 2.02) (1.87, 2.40)
a Variables found to be statistically significant in the univariable model were further examined in the multivariable model. Adjusted for other forms of corticosteroids use, chronic kidney disease, diabetes mellitus, gastrectomy, and pneumoconiosis.
Please cite this article as: Lai S-W, et al, Nation-based case-control study investigating the relationship between oral corticosteroids use and pulmonary tuberculosis, Eur J Intern Med (2017), http://dx.doi.org/10.1016/j.ejim.2017.05.020
S.-W. Lai et al. / European Journal of Internal Medicine xxx (2017) xxx–xxx Table 5 Interaction effects on the risk of pulmonary tuberculosis between oral corticosteroids use and comorbidities. Oral corticosteroids
Any comorbiditya
Case number/control number
Adjusted (95% CI) ORb
Never use Never use Ever use Ever use
No Yes No Yes
2147/2961 644/728 2449/1780 781/552
1.00 1.21 1.98 2.04
(Reference) (1.07, 1.36) (1.79, 2.20) (1.78, 2.34)
a Comorbidities including chronic kidney disease, diabetes mellitus, gastrectomy, and pneumoconiosis. b Adjusted for other forms of corticosteroids use.
consumption and chronic obstructive pulmonary disease was included instead of smoking, which has been mentioned in previous studies [18,29]. Despite the above limitations, some strengths should be addressed. To the best of our knowledge, this is the first nation-based study focusing on oral corticosteroids use and pulmonary tuberculosis in Taiwan. The study is methodologically adequate, with information that can have important therapeutic implications. The study results add the updated evidence to the literature on the theme, especially confirming what has already been described in international medical literature, including the relationship between dosage and duration of oral corticosteroids use and the increase in the likelihood of pulmonary tuberculosis. We conclude that use of oral corticosteroids is significantly associated with 1.9-fold increased odds of pulmonary tuberculosis, independent of other risk comorbidities. The study consistently reveals dose-dependent and duration-dependent increases in risk for pulmonary tuberculosis. Physicians should consider the possibility of pulmonary tuberculosis when cough, hemoptysis, fever, night sweats, or unintentional body weight loss develops among patients with oral corticosteroids use [30], particularly on high-dose use or long-term use. Whether routine screening for latent pulmonary tuberculosis is needed for patients with long-term oral corticosteroids therapy requires much prospective research to confirm. Specific author contributions Shih-Wei Lai planned and conducted this study. He substantially contributed to the conception of the article, initiated the draft of the article, and critically revised the article. Cheng-Li Lin conducted the data analysis and critically revised the article. Kuan-Fu Liao planned and conducted this study. They participated in the data interpretation, and also critically revised the article. Conflict of interest statement The authors disclose no conflicts of interest. Acknowledgement This study was supported in part by Taiwan Ministry of Health and Welfare Clinical Trial Center (MOHW106-TDU-B-212-113004), Academia Sinica Taiwan Biobank Stroke Biosignature Project (BM10501010037), and National Research Program for Biopharmaceuticals (NRPB) Stroke Clinical Trial Consortium (MOST 105-2325-B-039-003), These funding agencies did not influence the study design, data collection and analysis, decision to publish, or preparation of the manuscript.
5
References [1] Global TB Report 2014 WHO, Geneva. [2] Ferrara G, Murray M, Winthrop K, Centis R, Sotgiu G, Migliori GB, et al. Risk factors associated with pulmonary tuberculosis: smoking, diabetes and anti-TNFalpha drugs. Curr Opin Pulm Med 2012;18:233–40. [3] Lonnroth K, Jaramillo E, Williams BG, Dye C, Raviglione M. Drivers of tuberculosis epidemics: the role of risk factors and social determinants. Soc Sci Med 2009;68: 2240–6. [4] Lonnroth K, Raviglione M. Global epidemiology of tuberculosis: prospects for control. Semin Respir Crit Care Med 2008;29:481–91. [5] Cline JC, Davis SM. Risks of infection or reactivation of tuberculosis associated with chronic corticosteroid therapy. Ann Pharmacother 1997;31:775–6. [6] Stanbury RM, Graham EM. Systemic corticosteroid therapy—side effects and their management. Br J Ophthalmol 1998;82:704–8. [7] Kobashi Y, Matsushima T. Clinical analysis of pulmonary tuberculosis in association with corticosteroid therapy. Intern Med 2002;41:1103–10. [8] Liao CM, Hsieh NH, Huang TL, Cheng YH, Lin YJ, Chio CP, et al. Assessing trends and predictors of tuberculosis in Taiwan. BMC Public Health 2012;12:29. [9] National Health Insurance Research Database. Taiwan. http://nhird.nhri.org.tw/en/ index.html. [cited on April 1, 2017, English version]. [10] Liao KF, Cheng KC, Lin CL, Lai SW. Etodolac and the risk of acute pancreatitis. BioMedicine-Taiwan 2017;7:25–9. [11] Liao KF, Huang PT, Lin CC, Lin CL, Lai SW. Fluvastatin use and risk of acute pancreatitis:a population-based case-control study in Taiwan. BioMedicine-Taiwan 2017;7:1–5. [12] Lai SW, Wang IK, Lin CL, Chen HJ, Liao KF. Splenectomy correlates with increased risk of pulmonary tuberculosis: a case-control study in Taiwan. Clin Microbiol Infect 2014;20:764–7. [13] Lai SW, Lin CL, Liao KF, Tsai SM. Increased risk of pulmonary tuberculosis among patients with appendectomy in Taiwan. Eur J Clin Microbiol Infect Dis 2014;33: 1573–7. [14] Lin HF, Liao KF, Chang CM, Lai SW, Tsai PY, Sung FC. Anti-diabetic medication reduces risk of pulmonary tuberculosis in diabetic patients: a population-based cohort study in Taiwan. Kuwait Medical Journal 2017;49:22–8. [15] Shen ML, Liao KF, Tsai SM, Lin CL, Lai SW. Herpes zoster correlates with pyogenic liver abscesses in Taiwan. BioMedicine-Taiwan 2016;6:24–9. [16] Lin CM, Liao KF, Lin CL, Lai SW. Use of simvastatin and risk of acute pancreatitis: a nationwide case-control study in Taiwan. J Clin Pharmacol 2017;16:881. [17] Liao KF, Cheng KC, Lin CL, Lai SW. Statin use correlates with reduced risk of pyogenic liver abscess: a population-based case-control study. Basic Clin Pharmacol Toxicol 2017;8:12777. [18] Lai S-W, Lin H-F, Lin C-L, Liao K-F. No association between losartan use and acute pancreatitis in hypertensive patients. Eur J Hosp Pharm 2017;24:120–3. [19] Hung SC, Lin CH, Hung HC, Lin CL, Lai SW. Use of selective serotonin reuptake inhibitors and risk of hip fracture in the elderly: a case-control study in Taiwan. J Am Med Dir Assoc 2017;18:350–4. [20] Hsu F-G, Sheu M-J, Lin C-L, Hsieh Y-W, Lai S-W. Use of zolpidem and risk of acute pyelonephritis in women: a population-based case-control study in Taiwan. J Clin Pharmacol 2017;57:376–81. [21] Jick SS, Lieberman ES, Rahman MU, Choi HK. Glucocorticoid use, other associated factors, and the risk of tuberculosis. Arthritis Rheum 2006;55:19–26. [22] Kim HA, Yoo CD, Baek HJ, Lee EB, Ahn C, Han JS, et al. Mycobacterium tuberculosis infection in a corticosteroid-treated rheumatic disease patient population. Clin Exp Rheumatol 1998;16:9–13. [23] Tam LS, Li EK, Wong SM, Szeto CC. Risk factors and clinical features for tuberculosis among patients with systemic lupus erythematosus in Hong Kong. Scand J Rheumatol 2002;31:296–300. [24] Chan YC, Yosipovitch G. Suggested guidelines for screening and management of tuberculosis in patients taking oral glucocorticoids–an important but often neglected issue. J Am Acad Dermatol 2003;49:91–5. [25] Yagi T, Yamagishi F, Sasaki Y, Itakura M, Fujikawa A, Kuga M, et al. A study on cases developed pulmonary tuberculosis after receiving gastrectomy. Kekkaku 2004;79: 355–9. [26] Moutschen MP, Scheen AJ, Lefebvre PJ. Impaired immune responses in diabetes mellitus: analysis of the factors and mechanisms involved. Relevance to the increased susceptibility of diabetic patients to specific infections. Diabete Metab 1992;18:187–201. [27] Gupta KB, Gupta R, Atreja A, Verma M, Vishvkarma S. Tuberculosis and nutrition. Lung India 2009;26:9–16. [28] Rao VG, Bhat J, Yadav R, Muniyandi M, Bhondeley MK, Wares DF. Smoking and alcohol consumption: risk factors for pulmonary tuberculosis among the tribal community in central India. Indian J Tuberc 2017;64:40–3. [29] Lai SW, Lin CL, Liao KF. Atrial fibrillation associated with acute pancreatitis: a retrospective cohort study in Taiwan. J Hepatobiliary Pancreat Sci 2016;23:242–7. [30] den Boon S, White NW, van Lill SW, Borgdorff MW, Verver S, Lombard CJ, et al. An evaluation of symptom and chest radiographic screening in tuberculosis prevalence surveys. Int J Tuberc Lung Dis 2006;10:876–82.
Please cite this article as: Lai S-W, et al, Nation-based case-control study investigating the relationship between oral corticosteroids use and pulmonary tuberculosis, Eur J Intern Med (2017), http://dx.doi.org/10.1016/j.ejim.2017.05.020
Contents lists available at ScienceDirect
European Journal of Internal Medicine journal homepage: www.elsevier.com/locate/ejim
Nation-based case-control study investigating the relationship between oral corticosteroids use and pulmonary tuberculosis Shih-Wei Lai a,b, Cheng-Li Lin a,c, Kuan-Fu Liao d,e,f,⁎ a
College of Medicine, China Medical University, Taichung, Taiwan Department of Family Medicine, China Medical University Hospital, Taichung, Taiwan Management Office for Health Data, China Medical University Hospital, Taichung, Taiwan d Graduate Institute of Integrated Medicine, China Medical University, Taichung, Taiwan e College of Medicine, Tzu Chi University, Hualien, Taiwan f Department of Internal Medicine, Taichung Tzu Chi General Hospital, Taichung, Taiwan b c
a r t i c l e
i n f o
Article history: Received 18 April 2017 Received in revised form 20 May 2017 Accepted 22 May 2017 Available online xxxx Keywords: Oral corticosteroids Pulmonary tuberculosis Case-control study Taiwan National Health Insurance Program
a b s t r a c t Background and objective: No published formal study focuses on the relationship between oral corticosteroids use and pulmonary tuberculosis in Taiwan. The aim of the study was to investigate whether oral corticosteroids use was associated with increased odds of pulmonary tuberculosis. Methods: The retrospective, nation-based, case-control study was conducted to analyze the database of the Taiwan National Health Insurance Program between 2000 and 2013. Subjects aged 20 to 84 years with newly diagnosed pulmonary tuberculosis were defined as the cases (n = 6021). Randomly selected subjects without pulmonary tuberculosis aged 20 to 84 years were defined as the controls (n = 6021). Subjects who never had a prescription for oral corticosteroids were defined as never use. Subjects who ever had a prescription for oral corticosteroids were defined as ever use. Results: After adjustment for confounders, the adjusted odds ratio of pulmonary tuberculosis was 1.91 for subjects with even use of oral corticosteroids (95% confidence interval 1.73, 2.11), comparing with never use. The adjusted odds ratio of pulmonary tuberculosis was 1.03 for subjects with increasing cumulative dose of oral corticosteroids for per milligram (95% confidence interval 1.02, 1.03), comparing with never use. In addition, the adjusted odds ratios of pulmonary tuberculosis were 1.83 (95% CI 1.65, 2.02) for subjects with cumulative duration of oral corticosteroids b 3 months, and 2.12 (95% CI 1.87, 2.40) for subjects with cumulative duration ≥ 3 months, comparing with never use. Conclusion: Though the finding is not unexpected, use of oral corticosteroids is significantly associated with 1.9fold increased odds of pulmonary tuberculosis, independent of other risk comorbidities. There are dose-dependent and duration-dependent effects of oral corticosteroids use on the risk of pulmonary tuberculosis. © 2017 Published by Elsevier B.V. on behalf of European Federation of Internal Medicine.
1. Introduction Tuberculosis remains one of the deadliest infectious diseases worldwide. WHO reported that there were about 9.0 million people developing tuberculosis and 1.5 million people dying from this disease in 2013 [1]. Current evidence indicates that numerous risk factors are found to be associated with development of pulmonary tuberculosis, including smoking, diabetes mellitus, malnutrition, alcoholism, human immunodeficiency virus infection, and living or working in crowded space [2–4].
⁎ Corresponding author at: Department of Internal Medicine, Taichung Tzu Chi General Hospital, No.66, Sec. 1, Fongsing Road, Tanzi District, Taichung City 427, Taiwan. E-mail address: [email protected] (K.-F. Liao).
Current evidence indicates that use of corticosteroids is associated with the increased susceptibility to tuberculosis due to its inhibition of immune functions [5–7]. In Liao et al.'s study in Taiwan, the overall incidence of tuberculosis was 68 per 100,000 population from 2004 to 2008 [8]. To date, no published formal study focuses on the relationship between use of oral corticosteroids and pulmonary tuberculosis in Taiwan. Given that tuberculosis remains a high prevalence in Taiwan, we make a rational hypothesis that use of oral corticosteroids may be associated with increased risk of pulmonary tuberculosis in Taiwan. If the relationship is positive, physicians should be aware of the potential risk of pulmonary tuberculosis in patients with use of oral corticosteroids. Therefore, a nation-based, retrospective, case–control study was conducted to analyze the database of the Taiwan National Health Insurance Program to investigate whether (1) use of oral corticosteroids was
http://dx.doi.org/10.1016/j.ejim.2017.05.020 0953-6205/© 2017 Published by Elsevier B.V. on behalf of European Federation of Internal Medicine.
Please cite this article as: Lai S-W, et al, Nation-based case-control study investigating the relationship between oral corticosteroids use and pulmonary tuberculosis, Eur J Intern Med (2017), http://dx.doi.org/10.1016/j.ejim.2017.05.020
2
S.-W. Lai et al. / European Journal of Internal Medicine xxx (2017) xxx–xxx
associated with increased odds of pulmonary tuberculosis; (2) there were dose-dependent and duration-dependent effects of oral corticosteroids use on the risk of pulmonary tuberculosis? 2. Methods 2.1. Data source and study design Taiwan is an independent country with N23 million persons. The Taiwan National Health Insurance Program launched in March 1995, and now it has covered around 99% of the persons living in Taiwan [9]. The details of the program have been written in previous studies [10–14]. A nation-based, retrospective, case–control study was conducted to analyze the database of the Taiwan National Health Insurance Program. The Research Ethics Committee of China Medical University and Hospital in Taiwan approved the study (CMUH-104-REC2-115). 2.2. Selection of cases and controls We selected subjects aged 20 to 84 years with newly diagnosed pulmonary tuberculosis (the International Classification of Diseases (ICD) 9th Revision, ICD-9 codes 010, 011, 012, and 018) between 2000 and 2013 as the cases with pulmonary tuberculosis. The index date was defined as the date of cases being diagnosed with pulmonary tuberculosis. We randomly selected subjects without pulmonary tuberculosis aged 20 to 84 years from the same database as the controls. Both cases and controls were matched with sex, age (every 5-year interval), comorbidities, and the year of index date (Fig. 1). 2.3. Potential confounders Some medical conditions which could be related to pulmonary tuberculosis were included as follows: alcohol-related disease, chronic kidney disease, chronic obstructive pulmonary disease, diabetes mellitus, human immunodeficiency virus infection, gastrectomy, pneumoconiosis, as well as chronic liver disease including cirrhosis, hepatitis
B infection, hepatitis C infection, and other chronic hepatitis. To increase the validity of analysis, subjects who had the same diagnosis for 3 consecutive clinical visits in the ambulatory care and/or one episode of hospitalization diagnosis could be included in the study. Principal diagnosis and secondary diagnosis were used equally. Therefore, pulmonary tuberculosis and other medical conditions were documented for 3 or more records in the ambulatory care and/or one record during hospitalization. All comorbidities were diagnosed according to ICD-9 codes. The accuracy of ICD-9 codes has been validated in previous studies [10–20]. 2.4. Assessment of corticosteroids exposure It is relatively difficult to calculate the doses of topical corticosteroids use or inhaled corticosteroids use. It is also very rare for patients with long-term use of injected corticosteroids at outpatient department. In order to quantify the dosage, only oral corticosteroids were included for detailed analysis. Topical, inhaled, and injected corticosteroids were collected together as other forms of corticosteroids for adjustment. Oral corticosteroids in Taiwan were available as follows: cortisone, dexamethasone, fludrocortisone, methylprednisolone, prednisolone, and triamcinolone. To investigate the relationship between oral corticosteroids use and pulmonary tuberculosis, prescription history of oral corticosteroids before the index date was included. To diminish the biased results, subjects whose final prescriptions for oral corticosteroids were filled N 12 months before the index date were excluded from the study. Therefore, only subjects whose final prescriptions for oral corticosteroids were filled within 12 months before the index date were included. Subjects who never had a prescription for oral corticosteroids were defined as never use. Those who ever had a prescription for oral corticosteroids were defined as ever use. 2.5. Statistical analysis Distributions of sex, age, oral corticosteroids use, other forms of corticosteroids use, and comorbidities between the cases and controls were compared by the Chi-square test for categorized variables, and the t-test
Fig. 1. Flow diagram of the study revealing the selection of cases with pulmonary tuberculosis and controls.
Please cite this article as: Lai S-W, et al, Nation-based case-control study investigating the relationship between oral corticosteroids use and pulmonary tuberculosis, Eur J Intern Med (2017), http://dx.doi.org/10.1016/j.ejim.2017.05.020
S.-W. Lai et al. / European Journal of Internal Medicine xxx (2017) xxx–xxx Table 1 Information and comorbidities between pulmonary tuberculosis cases and controls. Non-tuberculosis Tuberculosis N = 6021 N = 6021 Characteristic Sex Female Male Age group (years) 20–39 40–64 65–84 Age (years), mean (standard deviation)† Duration of exposure to oral corticosteroids (days), mean (standard deviation)† Ever use of oral corticosteroids Ever use of other forms of corticosteroids Comorbidities Alcohol-related disease Chronic kidney disease Chronic liver disease Chronic obstructive pulmonary disease Diabetes mellitus Human immunodeficiency virus infection Gastrectomy Pneumoconiosis
n
(%)
n
(%)
P value⁎ 0.34
1709 4312
(28.4) (71.6)
1756 4265
(29.2) (70.8) 0.38
876 2154 2991 60.6
(14.6) (35.7) (49.7) (16.7)
894 2212 2915 60.4
(14.9) (36.7) (48.4) (16.8)
67.8
(288.0)
102.5 (375.8) b0.001
2332 2274
(38.7) (37.8)
3230 1781
(53.7) (29.6)
b0.001 b0.001
439 311 942 2076
(7.29) (5.17) (15.7) (34.5)
487 367 1016 2145
(8.09) (6.10) (16.9) (35.6)
0.10 0.03 0.07 0.19
864 16
(14.4) (0.27)
984 29
(16.3) (0.48)
0.002 0.05
22 173
(0.37) (2.87)
59 220
(0.95) (3.65)
b0.001 0.02
0.40
Data are presented as the number of subjects in each group with percentages given in parentheses, or mean with standard deviation given in parentheses. ⁎Chi-square test, and †t-test comparing subjects with and without pulmonary tuberculosis.
for continuous variables. Initially, all variables were included in the univariable unconditional logistic regression model. Variables found to be statistically significant in the univariable model were further examined in the multivariable unconditional logistic regression model. The odds ratio (OR) and 95% confidence interval (CI) were used to investigate the correlation of oral corticosteroids use with pulmonary tuberculosis. We further conducted the analysis to investigate whether there were dose-dependent and duration-dependent effects of oral corticosteroids use on the risk of pulmonary tuberculosis. All data processing and statistical analyses were performed with the SAS software version
3
9.2 (SAS Institute, Inc., Cary, North Carolina, USA). A two-tailed P value b 0.05 was considered statistically significant. 3. Results 3.1. Characteristics of the study population Table 1 reveals the distributions of sex, age, oral corticosteroids use, other forms of corticosteroids use, and comorbidities between the cases and controls. The study included 6021 cases with pulmonary tuberculosis and 6021 controls, with similar distributions of sex and age. The mean ages (standard deviation) of the study subjects were 60.4 (16.8) years for the cases and 60.6 (16.7) years for the controls (t-test, P = 0.40), without statistic significance. The mean durations (standard deviation) of exposure to oral corticosteroids were 102.5 (375.8) days for the cases and 67.8 (288.0) days for the controls (t-test, P b 0.001), with statistic significance. The cases had a statistically higher proportion of ever use of oral corticosteroids than the controls (53.7% vs. 38.7%, Chisquare test, P b 0.001). The cases had statistically higher proportions of chronic kidney disease, diabetes mellitus, gastrectomy, and pneumoconiosis than the controls (Chi-square test, P b 0.05). 3.2. Association of pulmonary tuberculosis with oral corticosteroids use and comorbidities After adjustment for other forms of corticosteroids use, chronic kidney disease, diabetes mellitus, gastrectomy, and pneumoconiosis, the multivariable unconditional logistic regression model revealed that the adjusted OR of pulmonary tuberculosis was 1.91 for subjects with even use of oral corticosteroids (95% CI 1.73, 2.11), comparing with never use (Table 2). In addition, the adjusted ORs of pulmonary tuberculosis were 1.17 (95%CI 1.06, 1.29) for subjects with history of diabetes mellitus, and 2.55 (95%CI 1.56, 4.19) for subjects with history of gastrectomy, respectively, comparing with those without such a history (Table 2).
3.3. Association of pulmonary tuberculosis with cumulative dosage of oral corticosteroids use We conducted an analysis for the dose-dependent effect of oral corticosteroids on the risk of pulmonary tuberculosis. After adjustment for confounders, the adjusted OR of pulmonary tuberculosis was 1.03 for subjects with increasing cumulative dose of oral corticosteroids for per milligram (95% CI 1.02, 1.03), comparing with never use (Table 3).
Table 2 Odds ratio and 95% confidence interval of pulmonary tuberculosis associated with corticosteroids use, other forms of corticosteroids use, and comorbidities. Adjusteda
Crude Variable
OR
(95%CI)
OR
(95%CI)
Sex (male vs. female) Age (per one year) Oral corticosteroids (never use as a reference) Ever use Other forms of corticosteroids (never use as a reference) Ever use Comorbidities (yes vs. no) Alcohol-related disease Chronic kidney disease Chronic liver disease Chronic obstructive pulmonary disease Diabetes mellitus Human immunodeficiency virus infection Gastrectomy Pneumoconiosis
0.96 1.00
(0.89, 1.04) (0.99, 1.00)
– –
– –
1.83
(1.70, 1.97)
1.91
(1.73, 2.11)
0.69
(0.64, 0.75)
1.08
(0.97, 1.19)
1.12 1.19 1.09 1.05 1.17 1.82 2.70 1.28
(0.98, 1.28) (1.02, 1.39) (0.99, 1.21) (0.98, 1.13) (1.06, 1.29) (0.99, 3.35) (1.65, 4.40) (1.05, 1.57)
– 1.07 – – 1.17 – 2.55 1.16
– (0.92, 1.26) – – (1.06, 1.29) – (1.56, 4.19) (0.95, 1.43)
a Variables found to be statistically significant in the univariable model were further examined in the multivariable model. Adjusted for other forms of corticosteroids use, chronic kidney disease, diabetes mellitus, gastrectomy, and pneumoconiosis.
Please cite this article as: Lai S-W, et al, Nation-based case-control study investigating the relationship between oral corticosteroids use and pulmonary tuberculosis, Eur J Intern Med (2017), http://dx.doi.org/10.1016/j.ejim.2017.05.020
4
S.-W. Lai et al. / European Journal of Internal Medicine xxx (2017) xxx–xxx
Table 3 Association of pulmonary tuberculosis with cumulative dosage of oral corticosteroids use. Variable
Case number/control number
Crude OR
(95% CI)
Adjusted ORa
(95% CI)
Never use of oral corticosteroids as a reference Cumulative dosage of oral corticosteroids use (increase in dosage for per mg)
2791/3689 3230/2332
1.00 1.03
(reference) (1.02, 1.03)
1.00 1.03
(reference) (1.02, 1.03)
a Variables found to be statistically significant in the univariable model were further examined in the multivariable model. Adjusted for other forms of corticosteroids use, chronic kidney disease, diabetes mellitus, gastrectomy, and pneumoconiosis.
3.4. Association of pulmonary tuberculosis with cumulative duration of oral corticosteroids use We conducted an analysis for the duration-dependent effect of oral corticosteroids on the risk of pulmonary tuberculosis. After adjustment for confounders, the adjusted ORs of pulmonary tuberculosis were 1.83 (95% CI 1.65, 2.02) for subjects with cumulative duration of oral corticosteroids b3 months, and 2.12 (95% CI 1.87, 2.40) for subjects with cumulative duration of oral corticosteroids ≥3 months, comparing with never use (Table 4). 3.5. Interaction effects on the risk of pulmonary tuberculosis between oral corticosteroids use and comorbidities Table 5 reveals the risk of pulmonary tuberculosis stratified by oral corticosteroids use and comorbidities. When comparing with those with never use of oral corticosteroids and without comorbidities including chronic kidney disease, diabetes mellitus, gastrectomy, and pneumoconiosis, the adjusted OR of pulmonary tuberculosis was 1.98 (95% CI 1.79, 2.20) among subjects with use of oral corticosteroids alone and without comorbidities. The adjusted OR increased to 2.04 (95% CI 1.78, 2.34) among subjects with ever use of oral corticosteroids and with any comorbidity. 4. Discussion In this study, a strong relationship was noted between oral corticosteroids use and the risk of pulmonary tuberculosis (adjusted OR 1.91, Table 2). This finding is compatible with Jick et al.'s case-control study in the United Kingdom showing that current use of corticosteroids was associated with increased odds of tuberculosis (adjusted OR 4.9, 95%CI 2.9–8.3) [21]. We also found that the risk of pulmonary tuberculosis was strongly related to the cumulative dosage of oral corticosteroids use, with higher risk in patients using higher dose (adjusted OR1.03, Table 3). This finding is compatible with previous studies showing that high cumulative dose of corticosteroids use was an important risk factor for developing tuberculosis [22,23]. We think that there seems to be a dose-dependent effect of oral corticosteroids use on the risk of pulmonary tuberculosis. We also found that the risk of pulmonary tuberculosis was strongly related to the cumulative duration of oral corticosteroids use, with higher risk in patients using a longer period (adjusted OR 1.83 for b3 months vs. adjusted OR 2.12 for ≥3 months, Table 4). This finding is
compatible with Chan et al.'s suggestion showing that patients on prolonged corticosteroids use might be at risk for tuberculosis [24]. We think that there seems to be a duration-dependent effect of oral corticosteroids use on the risk of pulmonary tuberculosis. That is, the longer use of oral corticosteroids, the higher the risk of pulmonary tuberculosis. The cases and the controls were matched with comorbidities, but the cases remained to have statistically higher proportions of chronic kidney disease, diabetes mellitus, gastrectomy, and pneumoconiosis than the controls. To diminish the confounding effects caused by comorbidities, we made a further analysis stratified by oral corticosteroids use and comorbidities. We found that subjects with use of oral corticosteroids alone and without comorbidities were associated with 1.98-fold increased odds of pulmonary tuberculosis, comparing with those with never use of oral corticosteroids and without comorbidities (Table 5). This finding indicates that the risk of pulmonary tuberculosis associated with oral corticosteroids use is independent of other risk comorbidities. Oral corticosteroids use has a unique effect on the risk of pulmonary tuberculosis. Diabetes mellitus and gastrectomy have been found to be associated with increased risk of pulmonary tuberculosis. That is why we included diabetes mellitus and gastrectomy for adjustment [14,25]. We found that the adjusted ORs of pulmonary tuberculosis were 1.17 for subjects with diabetes mellitus, and 2.55 for subjects with gastrectomy. In addition, impairment of immune functions in diabetes mellitus makes diabetic patients more susceptible to various infections [26]. Gastrectomy-related poor nutrition can result in secondary immunodeficiency which makes gastrectomized patients also more susceptible to various infections [27]. Similarly, long-term use of oral corticosteroids can lead to inhibition of immune functions that correlates with an increased susceptibility to various infections maybe including pulmonary tuberculosis [5–7]. Some limitations of the present study deserve comments. First, due to the natural limitation of this database, there was no specific record showing the reasons of the prescription for oral corticosteroids. The underlying diseases for oral corticosteroids therapy could be associated with the risk of pulmonary tuberculosis and could play a role in the onset of active pulmonary tuberculosis. Furthermore, it indicates a future research direction about the relationship between the underlying diseases for oral corticosteroids therapy and pulmonary tuberculosis. Second, alcohol consumption and smoking have been found as two risk factors for pulmonary tuberculosis [28]. Due to the same limitation, the histories of alcohol consumption and smoking were not recorded. Therefore, alcohol-related disease was included instead of alcohol
Table 4 Association of pulmonary tuberculosis with cumulative duration of oral corticosteroids use. Variable
Case number/control number
Crude OR
(95% CI)
Adjusted ORa
(95% CI)
Never use of oral corticosteroids as a reference Cumulative duration of oral corticosteroids use b3 months ≥3 months
2791/3689
1.00
(reference)
1.00
(reference)
2133/1624 1097/708
1.74 2.05
(1.60, 1.88) (1.84, 2.28)
1.83 2.12
(1.65, 2.02) (1.87, 2.40)
a Variables found to be statistically significant in the univariable model were further examined in the multivariable model. Adjusted for other forms of corticosteroids use, chronic kidney disease, diabetes mellitus, gastrectomy, and pneumoconiosis.
Please cite this article as: Lai S-W, et al, Nation-based case-control study investigating the relationship between oral corticosteroids use and pulmonary tuberculosis, Eur J Intern Med (2017), http://dx.doi.org/10.1016/j.ejim.2017.05.020
S.-W. Lai et al. / European Journal of Internal Medicine xxx (2017) xxx–xxx Table 5 Interaction effects on the risk of pulmonary tuberculosis between oral corticosteroids use and comorbidities. Oral corticosteroids
Any comorbiditya
Case number/control number
Adjusted (95% CI) ORb
Never use Never use Ever use Ever use
No Yes No Yes
2147/2961 644/728 2449/1780 781/552
1.00 1.21 1.98 2.04
(Reference) (1.07, 1.36) (1.79, 2.20) (1.78, 2.34)
a Comorbidities including chronic kidney disease, diabetes mellitus, gastrectomy, and pneumoconiosis. b Adjusted for other forms of corticosteroids use.
consumption and chronic obstructive pulmonary disease was included instead of smoking, which has been mentioned in previous studies [18,29]. Despite the above limitations, some strengths should be addressed. To the best of our knowledge, this is the first nation-based study focusing on oral corticosteroids use and pulmonary tuberculosis in Taiwan. The study is methodologically adequate, with information that can have important therapeutic implications. The study results add the updated evidence to the literature on the theme, especially confirming what has already been described in international medical literature, including the relationship between dosage and duration of oral corticosteroids use and the increase in the likelihood of pulmonary tuberculosis. We conclude that use of oral corticosteroids is significantly associated with 1.9-fold increased odds of pulmonary tuberculosis, independent of other risk comorbidities. The study consistently reveals dose-dependent and duration-dependent increases in risk for pulmonary tuberculosis. Physicians should consider the possibility of pulmonary tuberculosis when cough, hemoptysis, fever, night sweats, or unintentional body weight loss develops among patients with oral corticosteroids use [30], particularly on high-dose use or long-term use. Whether routine screening for latent pulmonary tuberculosis is needed for patients with long-term oral corticosteroids therapy requires much prospective research to confirm. Specific author contributions Shih-Wei Lai planned and conducted this study. He substantially contributed to the conception of the article, initiated the draft of the article, and critically revised the article. Cheng-Li Lin conducted the data analysis and critically revised the article. Kuan-Fu Liao planned and conducted this study. They participated in the data interpretation, and also critically revised the article. Conflict of interest statement The authors disclose no conflicts of interest. Acknowledgement This study was supported in part by Taiwan Ministry of Health and Welfare Clinical Trial Center (MOHW106-TDU-B-212-113004), Academia Sinica Taiwan Biobank Stroke Biosignature Project (BM10501010037), and National Research Program for Biopharmaceuticals (NRPB) Stroke Clinical Trial Consortium (MOST 105-2325-B-039-003), These funding agencies did not influence the study design, data collection and analysis, decision to publish, or preparation of the manuscript.
5
References [1] Global TB Report 2014 WHO, Geneva. [2] Ferrara G, Murray M, Winthrop K, Centis R, Sotgiu G, Migliori GB, et al. Risk factors associated with pulmonary tuberculosis: smoking, diabetes and anti-TNFalpha drugs. Curr Opin Pulm Med 2012;18:233–40. [3] Lonnroth K, Jaramillo E, Williams BG, Dye C, Raviglione M. Drivers of tuberculosis epidemics: the role of risk factors and social determinants. Soc Sci Med 2009;68: 2240–6. [4] Lonnroth K, Raviglione M. Global epidemiology of tuberculosis: prospects for control. Semin Respir Crit Care Med 2008;29:481–91. [5] Cline JC, Davis SM. Risks of infection or reactivation of tuberculosis associated with chronic corticosteroid therapy. Ann Pharmacother 1997;31:775–6. [6] Stanbury RM, Graham EM. Systemic corticosteroid therapy—side effects and their management. Br J Ophthalmol 1998;82:704–8. [7] Kobashi Y, Matsushima T. Clinical analysis of pulmonary tuberculosis in association with corticosteroid therapy. Intern Med 2002;41:1103–10. [8] Liao CM, Hsieh NH, Huang TL, Cheng YH, Lin YJ, Chio CP, et al. Assessing trends and predictors of tuberculosis in Taiwan. BMC Public Health 2012;12:29. [9] National Health Insurance Research Database. Taiwan. http://nhird.nhri.org.tw/en/ index.html. [cited on April 1, 2017, English version]. [10] Liao KF, Cheng KC, Lin CL, Lai SW. Etodolac and the risk of acute pancreatitis. BioMedicine-Taiwan 2017;7:25–9. [11] Liao KF, Huang PT, Lin CC, Lin CL, Lai SW. Fluvastatin use and risk of acute pancreatitis:a population-based case-control study in Taiwan. BioMedicine-Taiwan 2017;7:1–5. [12] Lai SW, Wang IK, Lin CL, Chen HJ, Liao KF. Splenectomy correlates with increased risk of pulmonary tuberculosis: a case-control study in Taiwan. Clin Microbiol Infect 2014;20:764–7. [13] Lai SW, Lin CL, Liao KF, Tsai SM. Increased risk of pulmonary tuberculosis among patients with appendectomy in Taiwan. Eur J Clin Microbiol Infect Dis 2014;33: 1573–7. [14] Lin HF, Liao KF, Chang CM, Lai SW, Tsai PY, Sung FC. Anti-diabetic medication reduces risk of pulmonary tuberculosis in diabetic patients: a population-based cohort study in Taiwan. Kuwait Medical Journal 2017;49:22–8. [15] Shen ML, Liao KF, Tsai SM, Lin CL, Lai SW. Herpes zoster correlates with pyogenic liver abscesses in Taiwan. BioMedicine-Taiwan 2016;6:24–9. [16] Lin CM, Liao KF, Lin CL, Lai SW. Use of simvastatin and risk of acute pancreatitis: a nationwide case-control study in Taiwan. J Clin Pharmacol 2017;16:881. [17] Liao KF, Cheng KC, Lin CL, Lai SW. Statin use correlates with reduced risk of pyogenic liver abscess: a population-based case-control study. Basic Clin Pharmacol Toxicol 2017;8:12777. [18] Lai S-W, Lin H-F, Lin C-L, Liao K-F. No association between losartan use and acute pancreatitis in hypertensive patients. Eur J Hosp Pharm 2017;24:120–3. [19] Hung SC, Lin CH, Hung HC, Lin CL, Lai SW. Use of selective serotonin reuptake inhibitors and risk of hip fracture in the elderly: a case-control study in Taiwan. J Am Med Dir Assoc 2017;18:350–4. [20] Hsu F-G, Sheu M-J, Lin C-L, Hsieh Y-W, Lai S-W. Use of zolpidem and risk of acute pyelonephritis in women: a population-based case-control study in Taiwan. J Clin Pharmacol 2017;57:376–81. [21] Jick SS, Lieberman ES, Rahman MU, Choi HK. Glucocorticoid use, other associated factors, and the risk of tuberculosis. Arthritis Rheum 2006;55:19–26. [22] Kim HA, Yoo CD, Baek HJ, Lee EB, Ahn C, Han JS, et al. Mycobacterium tuberculosis infection in a corticosteroid-treated rheumatic disease patient population. Clin Exp Rheumatol 1998;16:9–13. [23] Tam LS, Li EK, Wong SM, Szeto CC. Risk factors and clinical features for tuberculosis among patients with systemic lupus erythematosus in Hong Kong. Scand J Rheumatol 2002;31:296–300. [24] Chan YC, Yosipovitch G. Suggested guidelines for screening and management of tuberculosis in patients taking oral glucocorticoids–an important but often neglected issue. J Am Acad Dermatol 2003;49:91–5. [25] Yagi T, Yamagishi F, Sasaki Y, Itakura M, Fujikawa A, Kuga M, et al. A study on cases developed pulmonary tuberculosis after receiving gastrectomy. Kekkaku 2004;79: 355–9. [26] Moutschen MP, Scheen AJ, Lefebvre PJ. Impaired immune responses in diabetes mellitus: analysis of the factors and mechanisms involved. Relevance to the increased susceptibility of diabetic patients to specific infections. Diabete Metab 1992;18:187–201. [27] Gupta KB, Gupta R, Atreja A, Verma M, Vishvkarma S. Tuberculosis and nutrition. Lung India 2009;26:9–16. [28] Rao VG, Bhat J, Yadav R, Muniyandi M, Bhondeley MK, Wares DF. Smoking and alcohol consumption: risk factors for pulmonary tuberculosis among the tribal community in central India. Indian J Tuberc 2017;64:40–3. [29] Lai SW, Lin CL, Liao KF. Atrial fibrillation associated with acute pancreatitis: a retrospective cohort study in Taiwan. J Hepatobiliary Pancreat Sci 2016;23:242–7. [30] den Boon S, White NW, van Lill SW, Borgdorff MW, Verver S, Lombard CJ, et al. An evaluation of symptom and chest radiographic screening in tuberculosis prevalence surveys. Int J Tuberc Lung Dis 2006;10:876–82.
Please cite this article as: Lai S-W, et al, Nation-based case-control study investigating the relationship between oral corticosteroids use and pulmonary tuberculosis, Eur J Intern Med (2017), http://dx.doi.org/10.1016/j.ejim.2017.05.020