Abstract
Antioxidants may protect against oxidative stress, which is associated with tuberculosis (TB) disease. However, direct evidence for a protective association between dietary antioxidants and TB incidence in humans has been lacking. The relationship between intake of antioxidant vitamins (vitamins A, C, D, and E) and individual carotenoids (α-carotene, β-carotene, β-cryptoxanthin, lycopene, and lutein) and TB incidence was examined in the Singapore Chinese Health Study, a prospective cohort study of 63,257 adults aged 45–74 years enrolled during 1993–1998. Baseline intake of these antioxidants was estimated using a validated semiquantitative food frequency questionnaire including questions on use of dietary supplements. After an average of 16.9 years of follow-up, 1,186 incident active TB cases were identified among cohort participants. Compared with the lowest quartile, reduced risk of active TB was observed for the highest quartile of vitamin A intake (hazard ratio = 0.71, 95% confidence interval: 0.59, 0.85; P-trend < 0.01) and β-carotene intake (hazard ratio = 0.76, 95% confidence interval: 0.63, 0.91; P-trend < 0.01), regardless of smoking status. Lower TB risk was seen for vitamin C intake among current smokers only. Other vitamins and carotenoids were not associated with TB risk. These results suggest that vitamin C may reduce TB risk among current smokers by ameliorating oxidative stress, while vitamin A and β-carotene may have additional antimycobacterial properties.
Keywords: antioxidants, carotenoids, retinol, tuberculosis, vitamins
Tuberculosis (TB) is a leading cause of death globally, and an estimated 2–3 billion people in the world are infected with the Mycobacterium tuberculosis bacterium (1). Malnutrition is known to impair immune function (2–4), and poor nutritional status has been identified as an important risk factor for progression from TB infection to active TB disease (5, 6). However, most studies examining the relationship between malnutrition and TB risk have focused on protein-energy malnutrition (5, 7), and there is a paucity of studies examining the relationship between specific micronutrients and TB disease. Micronutrients are required for the immune system to function efficiently (8, 9), and antioxidant nutrients such as vitamins and carotenoids can counteract the potential damage caused by free radicals (9, 10), levels of which are elevated during TB infection (11, 12). It has been shown experimentally that defective host antioxidant defenses contribute to increased oxidative stress during TB infection and that restoration of oxidative stress levels with antioxidants is potentially beneficial in the prevention of TB disease (13).
Low levels of micronutrients are commonly observed among TB patients (14–18), and TB patients have been reported to have increased levels of oxidative stress on top of low circulating concentrations of antioxidant vitamins (19, 20). However, most of the epidemiologic evidence assessing the relationship between micronutrients and TB disease has been based on cross-sectional and case-control studies, which were unable to establish a cause-effect (causal) relationship, as the acute-phase response during TB infection can lead to changes in nutrient status (5, 21). Among the prospective cohort studies that have investigated the relationship between micronutrient status and incidence of active TB, lower blood levels of vitamins A and C have been reported to be associated with higher TB incidence (22), while another study did not find low serum vitamin A level to be significantly associated with increased risk of TB (23). However, those investigators did not adjust for cigarette smoking, a strong risk factor for TB, in their analyses (22, 23). Higher intake of fruits and vegetables was reported to be associated with lower TB incidence more than a decade ago (24), but the role of specific dietary nutrients in these food sources in TB risk has not been well studied. It is unclear whether dietary intake of antioxidant vitamins and carotenoids prior to the onset of disease plays a role in modulating the development of active TB and whether the associations of different vitamins and carotenoids with TB risk differ.
We examined the relationship between dietary intake of antioxidant vitamins (vitamins A, C, D, and E) and carotenoids (α-carotene, β-carotene, β-cryptoxanthin, lycopene, and lutein) and risk of developing active TB disease, using data from a prospective population-based cohort study of middle-aged and elderly adults in Singapore. Members of this cohort went through periods when TB was highly prevalent in the country a few decades ago, and those who had acquired latent TB infection in those early years would be at risk of disease reactivation at an advanced age (25).
METHODS
Study population
This study was conducted using data from the Singapore Chinese Health Study, which consisted of 63,257 Chinese adults aged 45–74 years recruited between April 1993 and December 1998 (26). Enrolled participants were citizens or permanent residents of Singapore residing in government housing estates, which was where the majority (86%) of the resident Singapore population lived during the period of recruitment. The study participants were restricted to the 2 major dialect groups of Chinese in Singapore—the Hokkiens, who originated from the southern part of Fujian Province, and the Cantonese, who came from the central region of Guangdong Province (26). Recruitment was initiated using a posted letter to invite potential participants to take part in the study. Study staff went door to door 5–7 days later to recruit participants for the study, if they met the inclusion criteria based on their ethnicity, age, dialect group, and residency status. Approximately 85% of the eligible subjects invited agreed to participate. This study was approved by the Institutional Review Board of the National University of Singapore, and all study participants gave informed consent.
Assessment of dietary exposure
A face-to-face interview was conducted during recruitment using a structured questionnaire, and information collected included participant demographic characteristics, lifetime use of tobacco, alcohol consumption, and physician-diagnosed medical history, such as history of diabetes and cancer. A 165-item semiquantitative food frequency questionnaire specifically developed for this study population was used to assess the usual diet of the participant over the past year. This questionnaire has been validated using two 24-hour dietary recall interviews and repeat administration of the food frequency questionnaire in a subset of 810 cohort participants, and most mean pair values for energy and nutrients were within 10% of each other according to the 2 different methods (26). Dietary intake of specific nutrients for each study participant was computed on the basis of the Singapore Food Composition Database, which was created specifically for this study (26).
Case ascertainment and follow-up
Identification of active TB cases in the study cohort was conducted via linkage analysis of the cohort database with the National TB Notification Registry (27) through December 31, 2014. Notification of TB cases in Singapore is mandatory by law, and all cases of suspected and confirmed TB must be reported to the Ministry of Health within 72 hours. Follow-up of the cohort was made by regular linkage to the Singapore Registry of Births and Deaths to update information on the vital status of cohort members. As of December 31, 2014, only 52 participants were known to be lost to follow-up, due to migration out of Singapore or for other reasons.
Statistical analysis
We excluded participants with a history of TB before recruitment (n = 3,012), as recorded in the National TB Notification Registry. We also excluded participants with a baseline history of cancer (n = 1,807) due to possible changes in dietary habits after cancer diagnosis. Participants who reported extreme energy intakes (<700 kcal/day or >3,700 kcal/day for men and <600 kcal/day or >3,000 kcal/day for women; n = 967) were additionally excluded, so the final analyses included 57,471 participants. Baseline characteristics of participants who developed TB and those who remained free of TB at the end of the study were compared using χ2 tests for categorical variables and Student's t tests for the continuous variables that followed a normal distribution. Person-years of follow-up for each participant were calculated from the date of recruitment to the date of diagnosis of active TB, death, loss to follow-up, or December 31, 2014, whichever occurred earlier. Dietary intakes of vitamins and carotenoids were energy-adjusted using the nutrient density model (28), and intakes were expressed per 1,000 kcal per day. Quartile levels of the nutrient density of each vitamin and carotenoid intake were derived from their respective distributions among all of the study participants in the cohort.
Cox proportional hazards regression models were used to examine the relationship between the nutrient density of vitamins and carotenoids and the risk of developing active TB. We included total energy intake in the model to control for potential confounding by total energy intake and to examine the association between dietary composition and disease risk. The analysis also included adjustment for factors that have been shown to affect TB risk in our cohort or others (29–32). The first model adjusted for age at recruitment, year of baseline interview, sex, dialect group, level of education, and total energy intake. The highest level of education attained by the study participant was our surrogate for the participant's socioeconomic status. The second model additionally adjusted for body mass index (weight (kg)/height (m)2), baseline history of diabetes, smoking intensity, alcohol intake, tea intake, and energy-adjusted intakes of protein, cholesterol, marine ω-3 fatty acids, and ω-6 fatty acids. The strength of a given association was measured using the hazard ratio and its corresponding 95% confidence interval. The median intake values of the quartiles were entered into the statistical models as a continuous variable to examine linear trend. There was no violation of Cox proportional hazards assumptions for our variables of interest.
To examine whether vitamins and carotenoids modulate the risk of active TB via their antioxidant properties, we stratified the analysis by the participant's smoking status (never, former, or current smoker), as cigarette smoking is a great contributor to in vivo oxidative damage (33, 34). Because responses to questions on nutrient intake could be affected by age and sex, we also performed analyses stratified by these factors. The heterogeneity of the associations according to the different factors was tested by including an interaction term (the product of quartile nutrient intake and the factor of interest) in the Cox model.
Analyses were conducted using SAS statistical software, version 9.3 (SAS Institute, Inc., Cary, North Carolina). All P values quoted are 2-sided, and P < 0.05 was considered statistically significant.
RESULTS
As of December 31, 2014, a total of 1,186 incident cases of active TB were identified in our cohort. The mean duration of follow-up was 9.1 years for participants who developed active TB (cases) and 17.1 years for participants who remained free of active TB. The mean age at TB diagnosis was 68.6 years. The incidence rates of active TB, adjusted to the age structure of our cohort, were 221 per 100,000 person-years in men and 55 per 100,000 person-years in women. Compared with participants who did not develop active TB, TB cases were significantly older and had a lower body mass index (Table 1). TB cases were more likely to have a history of smoking or physician-diagnosed diabetes, a lower level of education, a higher frequency of alcohol intake, and lower intakes of protein, marine ω-3 fatty acids, and ω-6 fatty acids (Table 1). Vegetables were the main dietary sources of vitamin A (66.7%), α-carotene (87.8%), β-carotene (78.3%), and lutein (84.2%) in our study population, while fruits were the biggest contributor to dietary vitamin C (54.5%), β-cryptoxanthin (99.1%), and lycopene (49.8%) intake. The main dietary sources of vitamin D were fish and shellfish (42.7%) and dairy products (16.1%). Vitamin E was mostly obtained from fats/oils (29.3%), vegetables (17.2%), and grain products (14.0%).
Table 1.
Baseline Characteristics of Cohort Members Who Developed Active Tuberculosis and Those Who Remained Free of Tuberculosis, Singapore Chinese Health Study, 1993–2014
| Characteristic | TB Cases (n = 1,186) | Noncases (n = 56,285) | P Valuea | ||||
|---|---|---|---|---|---|---|---|
| Mean (SD) | No. of Persons | % | Mean (SD) | No. of Persons | % | ||
| Age at recruitment, years | 59.1 (7.9) | 56.2 (8.0) | <0.01 | ||||
| Body mass indexb | 22.3 (3.5) | 23.2 (3.2) | <0.01 | ||||
| Male sex | 872 | 73.5 | 24,044 | 42.7 | <0.01 | ||
| Dialect group | 0.01 | ||||||
| Cantonese | 508 | 42.8 | 26,191 | 46.5 | |||
| Hokkien | 678 | 57.2 | 30,094 | 53.5 | |||
| Level of education | <0.01 | ||||||
| No formal education | 286 | 24.1 | 15,321 | 27.2 | |||
| Primary school (1–6 years) | 636 | 53.6 | 24,768 | 44.0 | |||
| Secondary school and above | 264 | 22.3 | 16,196 | 28.8 | |||
| Smoking statusc | <0.01 | ||||||
| Never smoker | 502 | 42.3 | 39,994 | 71.1 | |||
| Former smoker | 164 | 13.8 | 5,909 | 10.5 | |||
| Current smoker | 520 | 43.8 | 10,382 | 18.5 | |||
| History of diabetes | 165 | 13.9 | 4,919 | 8.7 | <0.01 | ||
| Alcohol intakec | <0.01 | ||||||
| None | 984 | 83.0 | 49,901 | 88.7 | |||
| Weekly | 111 | 9.4 | 4,563 | 8.1 | |||
| Daily | 91 | 7.7 | 1,821 | 3.2 | |||
| Tea intake | 0.36 | ||||||
| None | 500 | 42.2 | 23,131 | 41.1 | |||
| Monthly | 150 | 12.6 | 6,806 | 12.1 | |||
| Weekly | 264 | 22.3 | 13,796 | 24.5 | |||
| Daily | 272 | 22.9 | 12,552 | 22.3 | |||
| Energy intake, kcal/day | 1,598 (541) | 1,545 (519) | <0.01 | ||||
| Daily dietary intaked | |||||||
| Protein, g | 57.6 (10.4) | 59.2 (9.81) | <0.01 | ||||
| Cholesterol, mg | 174.9 (84.4) | 172.8 (73.5) | 0.40 | ||||
| Marine ω-3 fatty acids, g | 0.30 (0.16) | 0.32 (0.16) | <0.01 | ||||
| ω-6 fatty acids, g | 7.3 (3.1) | 7.9 (3.1) | <0.01 | ||||
Abbreviations: SD, standard deviation; TB, tuberculosis.
aP values were derived by Student's t test for continuous variables and χ2 test for categorical variables.
b Weight (kg)/height (m)2.
c Percentages do not sum to 100 because of rounding.
d Adjusted for energy intake.
Table 2 shows the association between dietary intake of vitamins and risk of active TB in our study cohort. Higher intakes of vitamins A, C, and E were associated with significantly reduced risk of developing active TB in a dose-dependent manner (all P's for trend < 0.01), while the association with dietary vitamin D was not significant (P-trend = 0.90) (Table 2). Adjustment for multiple risk factors did not alter the inverse association between vitamin A intake and TB risk (P-trend < 0.01). The multivariable-adjusted hazard ratio for the highest quartile of vitamin A intake was 0.71 (95% confidence interval (CI): 0.59, 0.85) compared with the lowest quartile of intake. The adjustment for other risk factors materially altered the associations between intake of vitamins C and E and risk of active TB, producing nonsignificant results (Table 2).
Table 2.
Risk of Active Tuberculosis According to Dietary Vitamin Intake, Singapore Chinese Health Study, 1993–2014
| Quartile of Energy-Adjusted Vitamin Intake (Range) | No. of Cases | Person-Years of Follow-up | Model 1a | Model 2b | ||
|---|---|---|---|---|---|---|
| HR | 95% CI | HR | 95% CI | |||
| Vitamin A, IU/1,000 kcal/day | ||||||
| Q1 (<2,085) | 439 | 230,063 | 1.00 | Referent | 1.00 | Referent |
| Q2 (2,085–2,900) | 304 | 244,403 | 0.79 | 0.69, 0.92 | 0.88 | 0.76, 1.03 |
| Q3 (2,901–4,087) | 268 | 248,777 | 0.78 | 0.67, 0.91 | 0.91 | 0.78, 1.07 |
| Q4 (≥4,088) | 175 | 250,828 | 0.60 | 0.50, 0.72 | 0.71 | 0.59, 0.85 |
| P-trend | <0.01 | <0.01 | ||||
| Vitamin C, mg/1,000 kcal/day | ||||||
| Q1 (<31) | 388 | 229,363 | 1.00 | Referent | 1.00 | Referent |
| Q2 (31–48) | 329 | 244,420 | 0.89 | 0.77, 1.04 | 1.03 | 0.89, 1.20 |
| Q3 (49–75) | 244 | 248,564 | 0.71 | 0.60, 0.83 | 0.88 | 0.75, 1.04 |
| Q4 (≥76) | 225 | 251,724 | 0.72 | 0.61, 0.85 | 0.92 | 0.77, 1.10 |
| P-trend | <0.01 | 0.21 | ||||
| Vitamin D, IU/1,000 kcal/day | ||||||
| Q1 (<38) | 318 | 243,906 | 1.00 | Referent | 1.00 | Referent |
| Q2 (38–57) | 327 | 244,816 | 1.11 | 0.95, 1.29 | 1.10 | 0.94, 1.29 |
| Q3 (58–86) | 275 | 244,430 | 0.97 | 0.82, 1.14 | 0.99 | 0.83, 1.17 |
| Q4 (≥87) | 266 | 240,919 | 1.05 | 0.89, 1.24 | 1.06 | 0.88, 1.27 |
| P-trend | 0.90 | 0.81 | ||||
| Vitamin E, mg/1,000 kcal/day | ||||||
| Q1 (<3.1) | 431 | 228,951 | 1.00 | Referent | 1.00 | Referent |
| Q2 (3.1–3.6) | 297 | 242,825 | 0.81 | 0.70, 0.94 | 0.90 | 0.77, 1.06 |
| Q3 (3.7–4.4) | 254 | 250,095 | 0.79 | 0.68, 0.93 | 0.94 | 0.79, 1.13 |
| Q4 (≥4.5) | 204 | 252,200 | 0.74 | 0.62, 0.87 | 0.90 | 0.74, 1.10 |
| P-trend | <0.01 | 0.40 | ||||
Abbreviations: CI, confidence interval; HR, hazard ratio; Q, quartile.
a Model 1 adjusted for age at recruitment (years), year of baseline interview (1993–1998), sex, dialect group (Hokkien, Cantonese), level of education (no formal education, primary school, secondary school or higher), and total energy intake (kcal/day).
b Model 2 further adjusted for body mass index (weight (kg)/height (m)2), baseline history of diabetes (yes, no), smoking status and intensity (never smoker, former smoker of <12 cigarettes/day, former smoker of 13–22 cigarettes/day, former smoker of ≥23 cigarettes/day, current smoker of <12 cigarettes/day, current smoker of 13–22 cigarettes/day, current smoker of ≥23 cigarettes/day), alcohol intake (none, weekly, daily), tea intake (none, monthly, weekly, daily), and energy-adjusted intakes of protein, cholesterol, marine ω-3 fatty acids, and ω-6 fatty acids (g/day, in quartiles).
High intake of all individual carotenoids except lutein was significantly associated with reduced risk of active TB (all P’s for trend < 0.01) (Table 3). Adjustment for other risk factors significantly attenuated these associations, except that for β-carotene, which remained statistically significant (P-trend < 0.01). Compared with the lowest quartile, the highest quartile of β-carotene intake was significantly associated with reduced risk of active TB (hazard ratio (HR) = 0.76, 95% CI: 0.63, 0.91).
Table 3.
Risk of Active Tuberculosis According to Dietary Carotenoid Intake, Singapore Chinese Health Study, 1993–2014
| Quartile of Energy-Adjusted Carotenoid Intake (Range) | No. of Cases | Person-Years of Follow-up | Model 1a | Model 2b | ||
|---|---|---|---|---|---|---|
| HR | 95% CI | HR | 95% CI | |||
| α-carotene, μg/1,000 kcal/day | ||||||
| Q1 (<60) | 390 | 230,297 | 1.00 | Referent | 1.00 | Referent |
| Q2 (60–113) | 320 | 246,896 | 0.88 | 0.76, 1.02 | 0.98 | 0.84, 1.14 |
| Q3 (114–211) | 271 | 248,195 | 0.82 | 0.70, 0.95 | 0.95 | 0.81, 1.11 |
| Q4 (≥212) | 205 | 248,684 | 0.70 | 0.59, 0.83 | 0.85 | 0.71, 1.01 |
| P-trend | <0.01 | 0.06 | ||||
| β-carotene, μg/1,000 kcal/day | ||||||
| Q1 (<850) | 432 | 229,787 | 1.00 | Referent | 1.00 | Referent |
| Q2 (850–1,234) | 301 | 245,239 | 0.79 | 0.68, 0.92 | 0.88 | 0.76, 1.03 |
| Q3 (1,235–1,771) | 272 | 248,226 | 0.80 | 0.69, 0.93 | 0.93 | 0.79, 1.09 |
| Q4 (≥1,772) | 181 | 250,819 | 0.62 | 0.52, 0.75 | 0.76 | 0.63, 0.91 |
| P-trend | <0.01 | <0.01 | ||||
| β-cryptoxanthin, μg/1,000 kcal/day | ||||||
| Q1 (<43) | 349 | 230,164 | 1.00 | Referent | 1.00 | Referent |
| Q2 (43–102) | 315 | 243,636 | 0.90 | 0.77, 1.04 | 1.01 | 0.87, 1.18 |
| Q3 (103–199) | 265 | 248,152 | 0.80 | 0.68, 0.94 | 0.96 | 0.82, 1.14 |
| Q4 (≥200) | 257 | 252,119 | 0.78 | 0.66, 0.92 | 0.95 | 0.80, 1.12 |
| P-trend | <0.01 | 0.46 | ||||
| Lycopene, μg/1,000 kcal/day | ||||||
| Q1 (<192) | 328 | 231,134 | 1.00 | Referent | 1.00 | Referent |
| Q2 (192–451) | 327 | 243,273 | 1.02 | 0.87, 1.19 | 1.07 | 0.92, 1.26 |
| Q3 (452–857) | 312 | 249,540 | 0.96 | 0.82, 1.12 | 1.06 | 0.90, 1.24 |
| Q4 (≥858) | 219 | 250,124 | 0.75 | 0.63, 0.89 | 0.87 | 0.73, 1.04 |
| P-trend | <0.01 | 0.06 | ||||
| Lutein, μg/1,000 kcal/day | ||||||
| Q1 (<782) | 412 | 231,318 | 1.00 | Referent | 1.00 | Referent |
| Q2 (782–1,103) | 288 | 245,708 | 0.79 | 0.68, 0.92 | 0.85 | 0.73, 1.00 |
| Q3 (1,104–1,543) | 278 | 247,163 | 0.85 | 0.73, 0.99 | 0.96 | 0.82, 1.13 |
| Q4 (≥1,544) | 208 | 249,882 | 0.73 | 0.61, 0.86 | 0.85 | 0.71, 1.01 |
| P-trend | 0.24 | 0.67 | ||||
Abbreviations: CI, confidence interval; HR, hazard ratio; Q, quartile.
a Model 1 adjusted for age at recruitment (years), year of baseline interview (1993–1998), sex, dialect group (Hokkien, Cantonese), level of education (no formal education, primary school, secondary school or higher), and total energy intake (kcal/day).
b Model 2 further adjusted for body mass index (weight (kg)/height (m)2), baseline history of diabetes (yes, no), smoking status and intensity (never smoker, former smoker of <12 cigarettes/day, former smoker of 13–22 cigarettes/day, former smoker of ≥23 cigarettes/day, current smoker of <12 cigarettes/day, current smoker of 13–22 cigarettes/day, current smoker of ≥23 cigarettes/day), alcohol intake (none, weekly, daily), tea intake (none, monthly, weekly, daily), and energy-adjusted intakes of protein, cholesterol, marine ω-3 fatty acids, and ω-6 fatty acids (g/day, in quartiles).
Because vegetables were the main dietary source of vitamin A and β-carotene, we also performed further analyses to ensure that the reduction in TB incidence was due to these nutrients and not to other factors associated with increased vegetable intake. The highest quartile of vegetable intake was found to be inversely associated with risk of active TB (HR = 0.82, 95% CI: 0.68, 0.99; P-trend = 0.02), but this association was attenuated when the model further adjusted for intake of vitamin A (HR = 1.00, 95% CI: 0.78, 1.28; P-trend = 0.92) or β-carotene (HR = 0.94, 95% CI: 0.72, 1.23; P-trend = 0.58) (data not shown). Conversely, additional adjustment for vegetable intake did not materially change the association between vitamin A or β-carotene and risk of active TB.
The inverse associations between intake of vitamin A or β-carotene and risk of active TB were similar irrespective of smoking status (P-interaction > 0.43) (Tables 4 and 5). Although the associations between vitamin A or β-carotene and risk of active TB among former smokers were not statistically significant, we observed a reduction in TB risk (albeit not statistically significant) when comparing the highest quartile of intake with the lowest quartile for vitamin A and β-carotene among former smokers, and this was consistent with the results in never and current smokers. On the other hand, a statistically significant stepwise reduction in TB risk was observed with increased vitamin C intake among current smokers (P-trend = 0.02; Table 4) after adjustment for other risk factors, and the P value for interaction between smoking status (never vs. current) and vitamin C intake was of borderline significance (P-interaction = 0.06). Even though we found a significant interaction between smoking status (never vs. current) and vitamin E intake in risk of TB (P-interaction < 0.01), the association of reduced TB risk among current smokers was not statistically significant. The associations between all of the vitamins and carotenoids studied and TB risk were similar when we stratified the participants by sex or age group (≤55 years vs. >55 years) for the analysis (results not shown).
Table 4.
Risk of Active Tuberculosis According to Dietary Vitamin Intake and Smoking Status, Singapore Chinese Health Study, 1993–2014
| Quartilea of Energy-Adjusted Vitamin Intake, per 1,000 kcal/day | Never Smokers (n = 40,496) | Former Smokers (n = 6,073) | Current Smokers (n = 10,902) | ||||||
|---|---|---|---|---|---|---|---|---|---|
| No. of Cases | HRb | 95% CI | No. of Cases | HR | 95% CI | No. of Cases | HR | 95% CI | |
| Vitamin A | |||||||||
| Q1 | 133 | 1.00 | Referent | 53 | 1.00 | Referent | 253 | 1.00 | Referent |
| Q2 | 138 | 0.94 | 0.74, 1.20 | 38 | 0.82 | 0.54, 1.25 | 128 | 0.86 | 0.69, 1.07 |
| Q3 | 135 | 0.95 | 0.74, 1.22 | 52 | 1.26 | 0.85, 1.87 | 81 | 0.73 | 0.57, 0.95 |
| Q4 | 96 | 0.71 | 0.54, 0.94 | 21 | 0.64 | 0.38, 1.08 | 58 | 0.75 | 0.56, 1.01 |
| P-trend | 0.02 | 0.26 | 0.02 | ||||||
| P-interactionc | 0.61 | 0.44 | |||||||
| Vitamin C | |||||||||
| Q1 | 119 | 1.00 | Referent | 43 | 1.00 | Referent | 226 | 1.00 | Referent |
| Q2 | 140 | 1.08 | 0.84, 1.39 | 44 | 1.09 | 0.71, 1.67 | 145 | 1.00 | 0.80, 1.23 |
| Q3 | 116 | 0.92 | 0.71, 1.21 | 33 | 0.82 | 0.51, 1.31 | 95 | 0.88 | 0.69, 1.13 |
| Q4 | 127 | 1.00 | 0.77, 1.30 | 44 | 1.19 | 0.76, 1.86 | 54 | 0.71 | 0.52, 0.97 |
| P-trend | 0.75 | 0.53 | 0.02 | ||||||
| P-interaction | 0.39 | 0.06 | |||||||
| Vitamin D | |||||||||
| Q1 | 135 | 1.00 | Referent | 40 | 1.00 | Referent | 143 | 1.00 | Referent |
| Q2 | 130 | 1.10 | 0.85, 1.41 | 43 | 1.19 | 0.76, 1.86 | 154 | 1.11 | 0.88, 1.40 |
| Q3 | 113 | 0.97 | 0.74, 1.27 | 42 | 1.16 | 0.72, 1.85 | 120 | 0.97 | 0.75, 1.27 |
| Q4 | 124 | 0.93 | 0.71, 1.22 | 39 | 1.08 | 0.66, 1.78 | 103 | 1.21 | 0.91, 1.61 |
| P-trend | 0.40 | 0.90 | 0.30 | ||||||
| P-interaction | 0.37 | 0.68 | |||||||
| Vitamin E | |||||||||
| Q1 | 130 | 1.00 | Referent | 53 | 1.00 | Referent | 248 | 1.00 | Referent |
| Q2 | 111 | 0.88 | 0.67, 1.15 | 43 | 0.91 | 0.59, 1.40 | 143 | 0.94 | 0.75, 1.18 |
| Q3 | 137 | 1.13 | 0.85, 1.50 | 41 | 0.96 | 0.60, 1.54 | 76 | 0.76 | 0.57, 1.02 |
| Q4 | 124 | 1.05 | 0.77, 1.42 | 27 | 0.77 | 0.44, 1.34 | 53 | 0.81 | 0.57, 1.14 |
| P-trend | 0.50 | 0.39 | 0.11 | ||||||
| P-interaction | 0.74 | <0.01 | |||||||
Abbreviations: CI, confidence interval; HR, hazard ratio; Q, quartile.
a For quartile cutpoints, see Table 2.
b HRs were adjusted for age at recruitment (years), year of baseline interview (1993–1998), sex, dialect group (Hokkien, Cantonese), level of education (no formal education, primary school, secondary school or higher), body mass index (weight (kg)/height (m)2), total energy intake (kcal/day), baseline history of diabetes (yes, no), alcohol intake (none, monthly, weekly, daily), tea intake (none, monthly, weekly, daily), and energy-adjusted intakes of protein, cholesterol, marine ω-3 fatty acids, and ω-6 fatty acids (g/day, in quartiles).
c Interaction with never smokers.
Table 5.
Risk of Active Tuberculosis According to Dietary Carotenoid Intake and Smoking Status, Singapore Chinese Health Study, 1993–2014
| Quartilea of Energy-Adjusted Carotenoid Intake, per 1,000 kcal/day | Never Smokers (n = 40,496) | Former Smokers (n = 6,073) | Current Smokers (n = 10,902) | ||||||
|---|---|---|---|---|---|---|---|---|---|
| No. of Cases | HRb | 95% CI | No. of Cases | HR | 95% CI | No. of Cases | HR | 95% CI | |
| α-carotene | |||||||||
| Q1 | 128 | 1.00 | Referent | 51 | 1.00 | Referent | 211 | 1.00 | Referent |
| Q2 | 129 | 0.86 | 0.67, 1.10 | 34 | 0.77 | 0.49, 1.19 | 157 | 1.14 | 0.92, 1.41 |
| Q3 | 143 | 1.00 | 0.78, 1.27 | 45 | 1.09 | 0.73, 1.64 | 83 | 0.78 | 0.60, 1.01 |
| Q4 | 102 | 0.75 | 0.57, 0.98 | 34 | 0.98 | 0.63, 1.53 | 69 | 0.91 | 0.69, 1.20 |
| P-trend | 0.06 | 0.75 | 0.20 | ||||||
| P-interactionc | 0.19 | 0.93 | |||||||
| β-carotene | |||||||||
| Q1 | 137 | 1.00 | Referent | 51 | 1.00 | Referent | 244 | 1.00 | Referent |
| Q2 | 132 | 0.87 | 0.68, 1.11 | 39 | 0.93 | 0.61, 1.43 | 130 | 0.89 | 0.71, 1.10 |
| Q3 | 134 | 0.94 | 0.73, 1.20 | 53 | 1.37 | 0.92, 2.04 | 85 | 0.75 | 0.58, 0.97 |
| Q4 | 99 | 0.75 | 0.57, 0.99 | 21 | 0.71 | 0.42, 1.20 | 61 | 0.80 | 0.60, 1.08 |
| P-trend | 0.06 | 0.47 | 0.05 | ||||||
| P-interaction | 0.56 | 0.48 | |||||||
| β-cryptoxanthin | |||||||||
| Q1 | 124 | 1.00 | Referent | 46 | 1.00 | Referent | 179 | 1.00 | Referent |
| Q2 | 132 | 0.98 | 0.76, 1.25 | 25 | 0.54 | 0.33, 0.88 | 158 | 1.18 | 0.95, 1.47 |
| Q3 | 115 | 0.84 | 0.65, 1.09 | 50 | 1.13 | 0.75, 1.70 | 100 | 0.99 | 0.77, 1.27 |
| Q4 | 131 | 0.91 | 0.71, 1.18 | 43 | 0.91 | 0.59, 1.41 | 83 | 0.95 | 0.72, 1.24 |
| P-trend | 0.47 | 0.65 | 0.42 | ||||||
| P-interaction | 0.41 | 0.73 | |||||||
| Lycopene | |||||||||
| Q1 | 140 | 1.00 | Referent | 46 | 1.00 | Referent | 142 | 1.00 | Referent |
| Q2 | 129 | 0.98 | 0.77, 1.25 | 41 | 0.94 | 0.61, 1.44 | 157 | 1.20 | 0.95, 1.52 |
| Q3 | 146 | 1.12 | 0.88, 1.42 | 41 | 0.97 | 0.63, 1.50 | 125 | 1.01 | 0.79, 1.29 |
| Q4 | 87 | 0.75 | 0.57, 1.00 | 36 | 0.98 | 0.62, 1.55 | 96 | 0.96 | 0.73, 1.25 |
| P-trend | 0.05 | 0.98 | 0.36 | ||||||
| P-interaction | 0.32 | 0.34 | |||||||
| Lutein | |||||||||
| Q1 | 135 | 1.00 | Referent | 56 | 1.00 | Referent | 221 | 1.00 | Referent |
| Q2 | 116 | 0.81 | 0.63, 1.05 | 40 | 0.80 | 0.53, 1.22 | 132 | 0.90 | 0.73, 1.13 |
| Q3 | 126 | 0.94 | 0.73, 1.20 | 47 | 1.07 | 0.71, 1.61 | 105 | 0.92 | 0.73, 1.17 |
| Q4 | 125 | 0.99 | 0.76, 1.29 | 21 | 0.56 | 0.33, 0.94 | 62 | 0.76 | 0.57, 1.02 |
| P-trend | 0.18 | 0.15 | 0.21 | ||||||
| P-interaction | 0.09 | 0.04 | |||||||
Abbreviations: CI, confidence interval; HR, hazard ratio; Q, quartile.
a For quartile cutpoints, see Table 3.
b HRs were adjusted for age at recruitment (years), year of baseline interview (1993–1998), sex, dialect group (Hokkien, Cantonese), level of education (no formal education, primary school, secondary school or higher), body mass index (weight (kg)/height (m)2), total energy intake (kcal/day), baseline history of diabetes (yes, no), alcohol intake (none, monthly, weekly, daily), tea intake (none, monthly, weekly, daily), and energy-adjusted intakes of protein, cholesterol, marine ω-3 fatty acids, and ω-6 fatty acids (g/day, in quartiles).
c Interaction with never smokers.
Because there may be a lag period between onset of disease and diagnosis of the disease, there may have been undiagnosed TB cases at the time of recruitment, and the dietary habits of the participants may have been influenced by the subclinical conditions of TB. To overcome this potential bias, we performed the same analyses after excluding all participants with less than 2 years of follow-up. The results were essentially the same (data not shown).
DISCUSSION
In our prospective population-based cohort study, we found increased dietary intake of vitamin A and its precursor β-carotene to be associated with reduced risk of active TB, and this association was significant among both never smokers and current smokers. Stratification by smoking status revealed a protective association between vitamin C intake and TB risk among current smokers but not among never or former smokers.
The pathogenesis of TB is determined by the balance between the generation of free radicals to kill intracellular mycobacteria and the antioxidant mechanisms of the host to prevent oxidative tissue damage (35), and many studies have found increased levels of oxidative stress markers, as well as reduced circulating concentrations of antioxidants, in TB patients (19, 20). Smoking is an important risk factor for the development of active TB (29, 36). Exposure to cigarette smoke can induce oxidative damage in the lungs (34), which may further elevate levels of oxidative stress on top of the free radicals produced by the host in response to TB infection (11). Antioxidant vitamins and carotenoids are able to counteract the potential damage caused by uncontrolled free radical production (9, 10), and vitamins C and E and β-carotene have been shown to serve as the major nonenzymatic antioxidants of the lungs in the defense against oxidants (37). Hence, amelioration of increased levels of oxidative stress may be a possible explanation for the reduction in TB risk with increased vitamin C intake that was observed only among current smokers. The reduction in risk of active TB with increased intakes of vitamin A and β-carotene among both never and current smokers suggests the possibility of additional antimycobacterial effects on top of their antioxidant properties.
Vitamin A is known to affect susceptibility to infections by regulating the expression and differentiation of many immune cells (38), and it has also been shown to inhibit the growth of mycobacteria in macrophages (39). A recent study suggested that reduction of cellular cholesterol via regulation of Niemann-Pick disease type C2 gene (NPC2) expression may be a possible mechanism for its antimycobacterial effect (40). Cholesterol is an essential molecule for phagocytic entry and intracellular survival of mycobacteria (41), and the reduction in cellular cholesterol content has been linked to increased phagosome maturation, which is required to initiate antigenic presentation and generate adaptive immune responses (41, 42). Vitamin A can also act synergistically with vitamin D to enhance maturation of the mycobacterial phagosome by limiting the expression of the tryptophan-aspartate-containing coat protein gene (TACO) (43). Although vitamin A has been shown to have antimycobacterial properties, its precursor, β-carotene, did not show any inhibitory effect on the growth of mycobacteria in culture (44). Hence, our observation of a protective association between dietary intake of β-carotene and risk of active TB may be a result of its conversion into vitamin A in vivo.
Although increased intake of fruits and vegetables has been reported to reduce risk of TB disease (24), the dietary components accounting for this protective association have not been well examined. To our knowledge, this is the first prospective cohort study that has comprehensively examined the relationship between specific antioxidant vitamins and carotenoids and the risk of developing active TB, and we have identified vitamin A and β-carotene to be the potential nutrients that reduce risk of active TB. The attenuation of the association between vegetable intake and TB risk after inclusion of vitamin A or β-carotene in the same model suggests that these nutrients mediated the protective association between vegetable intake and risk of active TB.
The prospective study design allowed us to establish the temporality of the associations, and recall bias in the dietary data is unlikely, since the data were collected before the participants were diagnosed with the disease. The food frequency questionnaire used was specifically created for this population and has been well validated (26). The age-adjusted mean daily intake nutrient densities for vitamins A, C, and E were comparable between this cohort and respondents to the Third National Health and Nutrition Examination Survey in the United States (26). Even though Singapore currently has an intermediate TB incidence rate of about 40 per 100,000 population, our study cohort consisted of older residents who probably resided in the country many decades ago, when TB incidence was as high as 300 per 100,000 population (45). Hence, many of our study participants could have acquired latent infection in those early years when TB was far more rampant (25), and this made our cohort suitable to study factors associated with latent TB reactivation. We were also able to adjust for other risk factors for TB (29, 32) in our statistical models, such as smoking, body mass index, history of diabetes, alcohol intake, total energy intake, and protein intake. The rate of loss to follow-up in our study cohort was negligible, and the ascertainment of active TB cases via linkage with the nationwide registry was virtually complete, as notification of TB cases is mandatory by law in Singapore.
Our study had some limitations, including the use of baseline dietary data and smoking status for the analysis. The dietary and smoking habits of the participants may have changed since their baseline reporting. Nonetheless, any subsequent changes would have led to nondifferential misclassification and potential underestimation of the association between dietary vitamin and carotenoid intakes and risk of active TB. Because vitamin D status could be affected by genetic and environmental factors (46), the use of dietary vitamin D in our analysis may also have been inadequate to study the relationship between vitamin D and TB risk. Because most of the TB cases in our cohort were likely to be cases of disease reactivation, our results may not be generalized to primary TB infection. We also did not have information about the use of immunosuppressants or human immunodeficiency virus status, which can increase the risk of active TB, but the use of immunosuppressants is likely to be very low in a population-based cohort, and Singapore has a relatively low human immunodeficiency virus infection rate (45). The adjustment for socioeconomic status by using highest level of education as a surrogate was also not optimal. We lacked statistical power to examine the associations in former smokers due to the low numbers of former smokers and TB cases among them in our cohort. Finally, as in any observational study, residual confounding could not be completely ruled out, although we adjusted for all identified potentially confounding factors in the statistical analyses.
In conclusion, we found increased dietary intake of vitamin A and its precursor, β-carotene, to be associated with reduced risk of developing active TB in a prospective cohort study of middle-aged and elderly adults in Singapore. Vitamin C intake was associated with reduced risk of active TB only among current smokers. The reduction in TB risk associated with vitamin A and β-carotene irrespective of the participant's smoking status suggests that these nutrients may exert anti-TB effects through mechanisms other than their antioxidant properties. Further research is needed to define the mechanisms of operation for the antimycobacterial effects of these compounds and to investigate the use of vitamin A and β-carotene as chemopreventive agents for active TB disease.
ACKNOWLEDGMENTS
Author affiliations: Epidemiology Domain, Saw Swee Hock School of Public Health, National University of Singapore, Singapore, Republic of Singapore (Avril Z. Soh, Woon-Puay Koh); Singapore Tuberculosis Control Unit, Tan Tock Seng Hospital, Singapore, Republic of Singapore (Cynthia B. E. Chee, Yee-Tang Wang); Division of Cancer Control and Population Sciences, University of Pittsburgh Cancer Institute, Pittsburgh, Pennsylvania (Jian-Min Yuan); Department of Epidemiology, Graduate School of Public Health, University of Pittsburgh, Pittsburgh, Pennsylvania (Jian-Min Yuan); and Office of Clinical Sciences, Duke-NUS Medical School, Singapore, Republic of Singapore (Woon-Puay Koh).
This work was supported by the National Cancer Institute, US National Institutes of Health (grants UM1 CA182876 and R01 CA144034). W.-P.K. was supported by the National Medical Research Council, Republic of Singapore (grant NMRC/CSA/0055/2013).
We thank Siew-Hong Low of the National University of Singapore for supervising the field work of the Singapore Chinese Health Study and Dr. Renwei Wang for maintenance of the cohort study database. We also thank Dr. Jeffrey Cutter of the Singapore Ministry of Health and Dr. Kyi-Win Khin Mar of the National Tuberculosis Notification Registry for assistance with the identification of tuberculosis cases in the cohort.
Conflict of interest: none declared.
REFERENCES
- 1. World Health Organization Global Tuberculosis Report 2015. Geneva, Switzerland: World Health Organization; 2015. [Google Scholar]
- 2. Ambrus JL Sr, Ambrus JL Jr. Nutrition and infectious diseases in developing countries and problems of acquired immunodeficiency syndrome. Exp Biol Med (Maywood). 2004;229(6):464–472. [DOI] [PubMed] [Google Scholar]
- 3. Chandra RK, Kumari S. Nutrition and immunity: an overview. J Nutr. 1994;124(8 suppl):1433S–1435S. [DOI] [PubMed] [Google Scholar]
- 4. Katona P, Katona-Apte J. The interaction between nutrition and infection. Clin Infect Dis. 2008;46(10):1582–1588. [DOI] [PubMed] [Google Scholar]
- 5. Cegielski JP, McMurray DN. The relationship between malnutrition and tuberculosis: evidence from studies in humans and experimental animals. Int J Tuberc Lung Dis. 2004;8(3):286–298. [PubMed] [Google Scholar]
- 6. World Health Organization Guideline: Nutritional Care and Support for Patients With Tuberculosis. Geneva, Switzerland: World Health Organization; 2013. [PubMed] [Google Scholar]
- 7. Lonnroth K, Williams BG, Cegielski P, et al. A consistent log-linear relationship between tuberculosis incidence and body mass index. Int J Epidemiol. 2010;39(1):149–155. [DOI] [PubMed] [Google Scholar]
- 8. Maggini S, Wintergerst ES, Beveridge S, et al. Selected vitamins and trace elements support immune function by strengthening epithelial barriers and cellular and humoral immune responses. Br J Nutr. 2007;98(suppl 1):S29–S35. [DOI] [PubMed] [Google Scholar]
- 9. Wintergerst ES, Maggini S, Hornig DH. Contribution of selected vitamins and trace elements to immune function. Ann Nutr Metab. 2007;51(4):301–323. [DOI] [PubMed] [Google Scholar]
- 10. Amir Aslani B, Ghobadi S. Studies on oxidants and antioxidants with a brief glance at their relevance to the immune system. Life Sci. 2016;146:163–173. [DOI] [PubMed] [Google Scholar]
- 11. Kwiatkowska S, Piasecka G, Zieba M, et al. Increased serum concentrations of conjugated diens and malondialdehyde in patients with pulmonary tuberculosis. Respir Med. 1999;93(4):272–276. [DOI] [PubMed] [Google Scholar]
- 12. Wiid I, Seaman T, Hoal EG, et al. Total antioxidant levels are low during active TB and rise with anti-tuberculosis therapy. IUBMB Life. 2004;56(2):101–106. [DOI] [PubMed] [Google Scholar]
- 13. Palanisamy GS, Kirk NM, Ackart DF, et al. Evidence for oxidative stress and defective antioxidant response in guinea pigs with tuberculosis. PLoS One. 2011;6(10):e26254. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Dou Y, Liang H, Wang Q, et al. Vitamin A, vitamin E and beta-carotene nutritional status and antioxidase level analysis among tuberculosis patients. Wei Sheng Yan Jiu. 2013;42(3):364–368. [PubMed] [Google Scholar]
- 15. Pakasi TA, Karyadi E, Wibowo Y, et al. Vitamin A deficiency and other factors associated with severe tuberculosis in Timor and Rote Islands, East Nusa Tenggara Province, Indonesia. Eur J Clin Nutr. 2009;63(9):1130–1135. [DOI] [PubMed] [Google Scholar]
- 16. Karyadi E, Schultink W, Nelwan RH, et al. Poor micronutrient status of active pulmonary tuberculosis patients in Indonesia. J Nutr. 2000;130(12):2953–2958. [DOI] [PubMed] [Google Scholar]
- 17. Ramachandran G, Santha T, Garg R, et al. Vitamin A levels in sputum-positive pulmonary tuberculosis patients in comparison with household contacts and healthy ‘normals’. Int J Tuberc Lung Dis. 2004;8(9):1130–1133. [PubMed] [Google Scholar]
- 18. Srinivasan A, Syal K, Banerjee D, et al. Low plasma levels of cholecalciferol and 13-cis-retinoic acid in tuberculosis: implications in host-based chemotherapy. Nutrition. 2013;29(10):1245–1251. [DOI] [PubMed] [Google Scholar]
- 19. Vijayamalini M, Manoharan S. Lipid peroxidation, vitamins C, E and reduced glutathione levels in patients with pulmonary tuberculosis. Cell Biochem Funct. 2004;22(1):19–22. [DOI] [PubMed] [Google Scholar]
- 20. Madebo T, Lindtjorn B, Aukrust P, et al. Circulating antioxidants and lipid peroxidation products in untreated tuberculosis patients in Ethiopia. Am J Clin Nutr. 2003;78(1):117–122. [DOI] [PubMed] [Google Scholar]
- 21. Bresnahan KA, Tanumihardjo SA. Undernutrition, the acute phase response to infection, and its effects on micronutrient status indicators. Adv Nutr. 2014;5(6):702–711. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22. Getz HR, Long ER, Henderson HJ. A study of the relation of nutrition to the development of tuberculosis; influence of ascorbic acid and vitamin A. Am Rev Tuberc. 1951;64(4):381–393. [DOI] [PubMed] [Google Scholar]
- 23. Cegielski JP, Arab L, Cornoni-Huntley J. Nutritional risk factors for tuberculosis among adults in the United States, 1971–1992. Am J Epidemiol. 2012;176(5):409–422. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24. Hemila H, Kaprio J, Pietinen P, et al. Vitamin C and other compounds in vitamin C rich food in relation to risk of tuberculosis in male smokers. Am J Epidemiol. 1999;150(6):632–641. [DOI] [PubMed] [Google Scholar]
- 25. Singapore Ministry of Health Enhancing public health measures against tuberculosis. Singapore, Republic of Singapore: Ministry of Health; 2008. https://www.moh.gov.sg/content/moh_web/home/pressRoom/pressRoomItemRelease/2008/enhancing_public_health_measures_against_tuberculosis.html. Updated November 28, 2008. Accessed August 13, 2015. [Google Scholar]
- 26. Hankin JH, Stram DO, Arakawa K, et al. Singapore Chinese Health Study: development, validation, and calibration of the quantitative food frequency questionnaire. Nutr Cancer. 2001;39(2):187–195. [DOI] [PubMed] [Google Scholar]
- 27. Chee CB, James L. The Singapore Tuberculosis Elimination Programme: the first five years. Bull World Health Organ. 2003;81(3):217–221. [PMC free article] [PubMed] [Google Scholar]
- 28. Willett WC, Howe GR, Kushi LH. Adjustment for total energy intake in epidemiologic studies. Am J Clin Nutr. 1997;65(4 suppl):1220S–1228S. [DOI] [PubMed] [Google Scholar]
- 29. Patra J, Jha P, Rehm J, et al. Tobacco smoking, alcohol drinking, diabetes, low body mass index and the risk of self-reported symptoms of active tuberculosis: individual participant data (IPD) meta-analyses of 72,684 individuals in 14 high tuberculosis burden countries. PLoS One. 2014;9(5):e96433. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30. Soh AZ, Chee CB, Wang YT, et al. Dietary cholesterol increases the risk whereas PUFAs reduce the risk of active tuberculosis in Singapore Chinese. J Nutr. 2016;146(5):1093–1100. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31. Chen M, Deng J, Li W, et al. Impact of tea drinking upon tuberculosis: a neglected issue. BMC Public Health 2015;15:515. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32. Gupta KB, Gupta R, Atreja A, et al. Tuberculosis and nutrition. Lung India. 2009;26(1):9–16. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33. Wooten JB, Chouchane S, McGrath TE. Tobacco smoke constituents affecting oxidative stress In: Halliwell BB, Poulsen HE, eds. Cigarette Smoke and Oxidative Stress. Berlin, Germany: Springer-Verlag GmbH; 2006:5–46. [Google Scholar]
- 34. Asami S, Manabe H, Miyake J, et al. Cigarette smoking induces an increase in oxidative DNA damage, 8-hydroxydeoxyguanosine, in a central site of the human lung. Carcinogenesis. 1997;18(9):1763–1766. [DOI] [PubMed] [Google Scholar]
- 35. Verma I, Jindal SK, Ganguly NK. Oxidative stress in tuberculosis In: Ganguly KN, Jindal KS, Biswal S, et al., eds. Studies on Respiratory Disorders. New York, NY: Springer New York; 2014:101–114. [Google Scholar]
- 36. Bates MN, Khalakdina A, Pai M, et al. Risk of tuberculosis from exposure to tobacco smoke: a systematic review and meta-analysis. Arch Intern Med. 2007;167(4):335–342. [DOI] [PubMed] [Google Scholar]
- 37. Rahman I, Biswas SK, Kode A. Oxidant and antioxidant balance in the airways and airway diseases. Eur J Pharmacol. 2006;533(1–3):222–239. [DOI] [PubMed] [Google Scholar]
- 38. Kim CH. Retinoic acid, immunity, and inflammation. Vitam Horm. 2011;86:83–101. [DOI] [PubMed] [Google Scholar]
- 39. Crowle AJ, Ross EJ. Inhibition by retinoic acid of multiplication of virulent tubercle bacilli in cultured human macrophages. Infect Immun. 1989;57(3):840–844. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 40. Wheelwright M, Kim EW, Inkeles MS, et al. All-trans retinoic acid-triggered antimicrobial activity against Mycobacterium tuberculosis is dependent on NPC2. J Immunol. 2014;192(5):2280–2290. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 41. de Chastellier C, Thilo L. Cholesterol depletion in Mycobacterium avium-infected macrophages overcomes the block in phagosome maturation and leads to the reversible sequestration of viable mycobacteria in phagolysosome-derived autophagic vacuoles. Cell Microbiol. 2006;8(2):242–256. [DOI] [PubMed] [Google Scholar]
- 42. Hestvik AL, Hmama Z, Av-Gay Y. Mycobacterial manipulation of the host cell. FEMS Microbiol Rev. 2005;29(5):1041–1050. [DOI] [PubMed] [Google Scholar]
- 43. Anand PK, Kaul D, Sharma M. Synergistic action of vitamin D and retinoic acid restricts invasion of macrophages by pathogenic mycobacteria. J Microbiol Immunol Infect. 2008;41(1):17–25. [PubMed] [Google Scholar]
- 44. Greenstein RJ, Su L, Brown ST. Vitamins A & D inhibit the growth of mycobacteria in radiometric culture. PLoS One. 2012;7(1):e29631. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 45. Singapore Ministry of Health Communicable Diseases Surveillance in Singapore 2014 Singapore, Republic of Singapore: Communicable Diseases Division, Ministry of Health; 2014:120–145. [Google Scholar]
- 46. Robien K, Butler LM, Wang R, et al. Genetic and environmental predictors of serum 25-hydroxyvitamin D concentrations among middle-aged and elderly Chinese in Singapore. Br J Nutr. 2013;109(3):493–502. [DOI] [PMC free article] [PubMed] [Google Scholar]