Poor Micronutrient Status of Active Pulmonary Tuberculosis Patients in ...

Community and International Nutrition

Poor Micronutrient Status of Active Pulmonary Tuberculosis Patients in Indonesia1 Elvina Karyadi,*† Werner Schultink,**,2 Ronald H. H. Nelwan,‡ Rainer Gross,** Zulkifli Amin,‡ Wil M. V. Dolmans,†† Jos W. M. van der Meer,†† Joseph G.A.J Hautvast† and Clive E. West†‡‡3 *SEAMEO-TROPMED Regional Center for Community Nutrition University of Indonesia, Indonesia; **Gesellschaft fu¨r Technische Zusammenarbeit (GTZ) GmbH, Eschborn, Germany; ‡Department of General Internal Medicine, Faculty of Medicine, University of Indonesia, Indonesia; ††Departments of General Internal Medicine and ‡‡Gastroenterology, University Medical Centre, Nijmegen, The Netherlands; and †Division of Human Nutrition and Epidemiology, Wageningen University, The Netherlands ABSTRACT Malnutrition is observed frequently in patients with pulmonary tuberculosis (TB), but their nutritional status, especially of micronutrients, is still poorly documented. The objective of this study was to investigate the nutritional status of patients with active TB compared with that of healthy controls in Jakarta, Indonesia. In a case-control study, 41 out-patients aged 15–55 y with untreated active pulmonary TB were compared with 41 healthy controls selected from neighbors of the patients and matched for age and sex. Cases had clinical and radiographic abnormalities consistent with pulmonary TB and at least two sputum specimens showing acid-fast bacilli. Anthropometric and micronutrient status data were collected. Compared with the controls, TB patients had significantly lower body mass index, skinfold thicknesses (triceps, biceps, subscapular, suprailiac), mid-upper arm circumference, proportion of fat, and concentrations of serum albumin, blood hemoglobin, plasma retinol and plasma zinc, whereas plasma zinc protoporphyrin concentration, as a measure of free erythrocyte protoporphyrin concentration, was greater. When patients and controls were subdivided on the basis of nutritional status, concentrations of serum albumin, blood hemoglobin, and zinc and retinol in plasma were lower in malnourished TB patients than in well-nourished healthy controls, well-nourished TB patients and malnourished healthy controls. In conclusion, the nutritional status of patients with active pulmonary TB was poor compared with healthy subjects, i.e., significantly more patients were anemic and more had low plasma concentrations of retinol and zinc. Low concentrations of hemoglobin, and of retinol and zinc in plasma were more pronounced in malnourished TB patients. J. Nutr. 130: 2953–2958, 2000. KEY WORDS:

●

malnutrition

●

tuberculosis

●

vitamin A

Tuberculosis (TB)4 is on the increase throughout the world and is one of the most important causes of death among adults in developing countries. In 1993, the WHO declared TB to be a “global health emergency” (Reichman 1996). In Indonesia, TB is one of the most important public health problems with a prevalence of 0.29%, and 5.6% of the world’s 7.5 million new cases of TB in 1990. TB ranks second among the leading causes of death in Indonesia (Household Health Survey 1995). Pulmonary TB, a chronic infectious disease caused by Myco-

●

zinc

●

micronutrient

●

humans

bacterium tuberculosis, is characterized by prolonged cough, hemoptysis, chest pain and dyspnea. Systemic manifestations of the disease include fever, malaise, anorexia, weight loss, weakness and night sweats (Hopewell 1994). Although malnutrition has been described in TB patients previously (Onwubalili 1988, Saha and Rao 1989, Tsukaguchi et al. 1991), contrary to what is commonly believed, little is known about nutritional status with respect to the micronutrients vitamin A, zinc and iron. Low concentrations of these nutrients may affect host defense. Vitamin A deficiency was found to be common among adults with TB and human immunodeficiency virus (HIV) infection in Rwanda (Rwanganbwoba et al. 1998). In the prechemotherapeutic era, cod liver oil rich in vitamins A and D was used regularly for the treatment of TB in an attempt to strengthen host defenses (Goldberg 1946). More recently, in vitro studies have shown that retinoic acid can inhibit multiplication of mycobacterium in macrophages (Crowle and Ross 1989). In addition, vitamin A has a vital role in lymphocyte proliferation and in maintaining the function of epithelial tissues (Chandra 1991). Zinc

1 Supported by grants from Gesellschaft fu¨r Technische Zusammenarbeit (GTZ) GmbH, Eschborn, Germany, Nutrition Research and Development Center of the Department of Health Republic of Indonesia, Bogor, Indonesia and PT Kimia Farma, Jakarta, Indonesia. PT Indo Farma, Jakarta, Indonesia provided standard anti-TB drugs. 2 Current address: UNICEF Organization, 3 United Nations Plaza, New York, NY. 3 To whom correspondence should be addressed. 4 Abbreviations used: ALAT, alanine amino transferase; ASAT, aspartate amino transferase; BMI, body mass index; CRP, C-reactive protein; ESR, erythrocyte sedimentation rate; HIV, human immunodeficiency virus; TB, tuberculosis; ZPP, zinc-protoporphyrin.

0022-3166/00 $3.00 © 2000 American Society for Nutritional Sciences. Manuscript received 25 April 2000. Initial review completed 31 May 2000. Revision accepted 4 August 2000. 2953

KARYADI ET AL.

2954

has been shown to be essential in vitamin A metabolism because it is required to mobilize vitamin A from the liver (Smith et al. 1973). Zinc deficiency also affects host defense in a variety of ways. It results in decreased phagocytosis and leads to a reduced number of circulating T cells and reduced tuberculin (purified protein derivative) reactivity, at least in animals (McMurray et al. 1990). Iron deficiency anemia has been reported in patients with pulmonary TB, as indicated by low hemoglobin concentrations, serum iron and total iron-binding capacity (Saha and Rao 1989), but how iron deficiency affects host defense against M. tuberculosis is unclear. Because of the limited data available on the relationship between nutritional status and TB, especially with respect to multiple micronutrient deficiencies, and because of the increasing incidence of TB, particularly in Southeast Asia, we decided to compare nutritional status of adult active pulmonary TB patients with that of matched healthy controls in Indonesia and investigate the concentrations of micronutrients in malnourished TB patients. SUBJECTS AND METHODS Subjects. Cases were out-patients with untreated active pulmonary TB admitted to the Rumah Sakit Umum Nasional Cipto Mangunkusumo, which is a major general public hospital and the national referral hospital in Central Jakarta, Indonesia. Controls were healthy subjects with no history of pulmonary TB, matched with cases for sex and age, and selected randomly from nonfamily neighbors of the patients in the smallest administrative unit in Indonesia (rukun tetangga), usually comprising ⬃20 houses and 100 persons. Field workers asked the head of the administrative unit (rukun tetangga) for a list of healthy subjects of the same sex and age (within ⫾ 2 y) as the patients. One person was selected at random as a control from the list of 3–7 persons proposed. Selection of cases was based on the following criteria: age 15–55 y; at least two sputum specimens positive for acid-fast bacilli by microscopy; and clinical and radiographic abnormalities consistent with pulmonary TB. Exclusion criteria for cases and controls were as follows: previous anti-TB treatment; pregnancy; lactation; use of corticosteroids or supplements containing vitamin A, zinc or iron during the previous month; moderate-tosevere injury or surgery during the last month; and diseases such as abnormal liver function as measured by elevated serum levels of aspartate amino transferase (ASAT) and alanine amino transferase (ALAT), diabetes mellitus as measured by elevated fasting blood glucose levels, neoplasm as determined by clinical examination, chronic renal failure as determined by elevated serum levels of urea and creatinine, and congestive heart failure. Study design. The study was designed as a case-control study. The sample size was based on the ability to determine a difference with ␣ ⫽ 0.05 and 1-␤ ⫽ 0.95 using a one-tailed test for concentrations of serum retinol and zinc and of blood hemoglobin. Because serum zinc concentration was the variable requiring the largest sample size, we calculated that with a sample size of 35 in each group, a between-group difference of 0.46 ␮mol/L in Zn (Narang et al. 1995) could be detected. We recruited 45 subjects for each group because we assumed that 25% of patients might not meet the inclusion criteria. Data collection. Potential cases and controls were interviewed using a structured questionnaire requesting information related to the inclusion and exclusion criteria. Those apparently eligible were then screened clinically including a chest X-ray by one of the coauthors (Z.A.), a pulmonologist at the University Hospital Cipto Mangunkusumo, Medical Faculty, University of Indonesia. All patients had evidence based on a chest X-ray of lung infiltration indicating active TB at the time of data collection. From those with evidence of TB, three specimens of early morning sputum were examined by direct microscopy after Kinyoun-Gabbett staining, which is a simplification of the Ziehl-Nielsen method (Chadwick 1982); specimens were cultured in Kudoh medium (Chadwick 1982). Subjects were weighed without shoes using an electronic platform model weighing scale (SECA 770 alpha; SECA, Hamburg, Germany)

and weight recorded to the nearest 0.1 kg; height was recorded to the nearest 0.1 cm using a microtoise. Body mass index (BMI) was calculated as body weight divided by height squared (kg/m2). Subjects were regarded as being malnourished if BMI ⬍ 18.5 kg/m2 (James et al. 1988). Biceps, triceps, suprailiac and subscapular skinfolds on the left side of the body were measured to the nearest 0.2 mm three times at each site using a Holtain skinfold caliper (Holtain, Crosswell, Crymych, Dyfed, UK) (Gibson 1990). Calculations of proportion of total body fat and fat-free mass were based on anthropometric data using the equations of Durnin and Womersley (1974). Mid-upper arm circumference was measured with a flexible steel tape (Gibson 1990). Two field workers were trained and standardized by one of the authors (E.K.) to take all of the anthropometric measurements. At the end of the standardization period, the technical error of the measurements was determined. The mean technical error, expressed as a standard deviation (SD ⫽ d2/2n, where d is the the difference between paired measurements and n is the the number of subjects) was 0.13 cm for mid-upper arm circumference, 0.35 mm for biceps skinfold, 0.22 mm for triceps skinfold, 0.31 mm for subscapular skinfold and 0.26 mm for suprailiac skinfold. Blood samples (15 mL) were collected from fasting subjects via venipuncture to determine total white blood cell count, hematocrit, erythrocyte sedimentation rate (ESR), albumin and hemoglobin in blood, zinc-protoporphyrin (ZPP) in erythrocytes, and to prepare plasma by centrifugation at 750 ⫻ g for 10 min at room temperature. All biochemical tests above were carried out on the same day. Plasma was stored at ⫺20°C until analysis of C-reactive protein (CRP), retinol, ␣-tocopherol and zinc. Hemoglobin concentration, hematocrit, white blood cells, ASAT and ALAT were measured directly using an automatic analyzer (Sysmex Microdilutor F-800, Kobe, Japan). The intra- and interassay CV for hemoglobin were ⬍5%. ESR was determined directly using the Westergreen technique (Kohli et al. 1975). Albumin was determined by the bromcresol green method (Dumas et al. 1997). Hemoglobin, hematocrit, white blood cells, ESR and albumin were analyzed in Multilab Laboratory, Jakarta, which collaborates with the pulmonary clinic. ZPP as a measure of free erythrocyte protoporphyrin was measured in duplicate using the portable front-face hematofluorometer (AVIV Biomedical, (Lakewood, NJ) at the SEAMEO-TROPMED laboratory at the University of Indonesia (Hastka et al. 1992). Variability based on these duplicate ZPP measurements was 1.6%. CRP was measured at the University Medical Centre, Nijmegen, using an immunoturbidimetric (Behringwerke, Marburg, West Germany) method (Metzmann 1985). Variability based on analysis of 15 samples was 2.6%. Plasma retinol and ␣-tocopherol were measured using HPLC with a C-18 column (Bondapak, Waters, Milford, MA); a UV detector (model SPD-6AV, Shimadzu, Tokyo); and methanol/water (95:5, v/v) as mobile phase at the Nutrition Research and Development Center of the Department of Health in Bogor, Indonesia (Arroyave et al. 1982) using standards from Sigma (St. Louis, MO). Plasma zinc was measured using atomic absorption spectrometry (Prasad et al. 1965) in the laboratory of Clinical Chemistry and Hematology, University of Bonn, Germany with values of a quality control analyzed with each set of determinations within 3% of certified values. Food intake was assessed on the basis of two consecutive 24-h recalls (Bingham et al. 1988) to estimate the intake of energy, protein, fat, carbohydrate, vitamin A, zinc, iron and vitamin E. The two recalls were conducted on two consecutive weekdays. The 24-h diet recall was collected by two interviewers trained by one of the authors (E.K.). Each 24-h recall was conducted using a standardized four-stage protocol (Gibson 1993). First, a complete list of all food and beverages consumed during the previous day was obtained. Second, detailed descriptions of all of the food and beverages consumed, including the cooking methods and brand names were recorded, together with the time and place of consumption. Third, estimates of the amounts of all foods and beverages consumed were recorded by referring to two- and three-dimensional models, household measuring and serving utensils (e.g., spoons, plates or cups), and food packages. Finally, the food recall was reviewed to ensure that all items had been recorded correctly. Part of the training session consisted of determining the differences between the amount estimated by each trainee and the actual weight of the food. An acceptable training level was

MICRONUTRIENT STATUS IN ACTIVE PULMONARY TUBERCULOSIS PATIENTS

considered to have been achieved when the average difference between the trainee’s estimate and the actual food weight was ⱕ 5 g. A pilot study was conducted by observing five patients and five healthy subjects while they ate a meal; the next day, a trained dietary interviewer had these subjects recall what and how much they had eaten at that meal. In general, the subjects accurately described what they had eaten. The calculation of nutrient intake from dietary recalls was done using World Food (Version 2.0, University of California, Berkeley CA), in which the Indonesian food composition tables had been incorporated. Ethical considerations. The ethical guidelines of the Council for International Organizations of Medical Sciences (1991) were followed and the study was approved by the Committee on Health Research Ethics, Faculty of Medicine, University of Indonesia, Jakarta. Informed consent was obtained from each subject before the start of the study. Statistical analysis. A one-sample Kolmogorov-Smirnov test was used to check whether data were normally distributed. Mean and standard deviation (SD) are used for reporting normally distributed data, and median and 25th–75th percentiles are used for reporting nonnormally distributed data. An independent sample t test was used to assess the differences between patients and controls for normally distributed parameters, whereas differences in nonnormally distributed parameters were tested using the Mann-Whitney test. A multiple stepwise regression analysis was performed to predict concentrations of plasma retinol and zinc by using age, sex, BMI, body temperature, presence of pulmonary cavity, white blood cell count, ESR and concentrations of CRP, and albumin as independent variables. Differences in prevalence were tested with a ␹2 test. The SPSS software package (Windows version 7.5.2, SPSS, Chicago, IL) was used for all statistical analyses and a P-value ⬍ 0.05 was considered significant.

RESULTS Four patients were withdrawn because they had severe hemoptysis during data collection and required intensive treatment. Thus, 41 (25 men and 16 women) active pulmonary TB patients (cases) and 41 healthy control subjects (25 men and 16 women) aged 28 ⫾ 9 y (mean ⫾ SD) were included in the study. Thirty-four patients (83%) compared with 36 controls (88%) had a BCG-scar on clinical examination. Symptoms and signs of patients were presented as follows: fever (ⱖ38°C) (54%), cough ⬎ 1 mo (93%), night sweats (61%), hemoptysis (51%), dyspnea (68%), chest pain (63%) and loss of appetite (76%). Of the cases, 26 (63%) had three positive smears and a remaining 15 (37%) had two positive smears for acid-fast bacilli, whereas 24 (59%) of cases had a positive sputum

2955

culture. The radiographic signs of patients were as follows: all patients had lung infiltration, 14 had pulmonary cavities, one had miliary disease and one had pleural effusion. Because the prevalence of HIV infection in this area was low (from the available data, possibly ⬍2%), no testing for HIV was carried out. The mean BMI in all patients was 20% lower than in controls (P ⬍ 0.001), and the mean proportion of fat in all patients (17.7%) was lower than in controls (21.9%) (P ⬍ 0.05). The number of patients with BMI ⬍ 18.5 kg/m2 (66%) was more than sixfold that of the healthy controls (10%) (P ⬍ 0.001). The mean body weight, BMI, skinfold thickness, mid-upper arm circumference, proportion of fat, fat mass and fat free mass in male patients were significantly lower than in male controls, whereas all of these variables except biceps and suprailiac skinfold thickness were significantly different between female patients and controls (Table 1). Serum albumin concentration was 10% lower in TB patients than in controls (Fig. 1). Serum albumin concentration was lower in malnourished TB patients than in well-nourished healthy controls, malnourished healthy controls and well-nourished TB patients (P ⬍ 0.05 for all comparisons) (Table 2). Compared with controls, TB patients had 13% lower mean concentrations of hemoglobin (Fig. 1) and 11% lower median hematocrit (P ⬍ 0.001). In malnourished TB patients, the mean concentrations of hemoglobin were 16% lower than in well-nourished controls, and 11% lower than in malnourished controls (Table 2). The median CRP concentration in TB patients was significantly higher than in controls (Fig. 1). Of the TB patients, 24 had hemoglobin concentrations indicating anemia, whereas only 9 controls had hemoglobin concentrations below normal (Fig. 2). The median concentrations of erythrocyte ZPP in TB patients were significantly higher than in controls (P ⬍ 0.001) with 32 patients and 14 controls having concentrations above normal (⬎40 ␮mol/mol heme). The mean white blood cell count (8626 ⫾ 3207 vs. 6434 ⫾ 1712 cells/mm3) and the median ESR were significantly higher in patients (52, interquartile range: 25– 82 mm/h) than in controls (22, interquartile range: 12–31 mm/h) (P ⬍ 0.001 for both variables). The mean plasma retinol concentration in patients was significantly lower than in controls with 10 (33%) patients and 4 controls (13%) having plasma retinol concentrations ⬍ 0.70 ␮mol/L, indicating marginal vitamin A deficiency (Fig.

TABLE 1 Nutritional status using anthropometric measurements in active pulmonary tuberculosis patients and healthy controls1 Patients

n Weight, kg Height, cm Body mass index, kg/m2 Biceps skinfold, mm Triceps skinfold, mm Subscapular skinfold, mm Suprailiac skinfold, mm Mid-upper arm circumference, cm Proportion of fat, g/100 g body Fat mass, kg Fat-free mass, kg 1 Values are mean ⫾ sample t test).

SD.

Controls

Male

Female

Male

Female

25 50.6 ⫾ 9.6 165.1 ⫾ 6.3 18.5 ⫾ 3.2 5.5 ⫾ 2.8 7.0 ⫾ 3.7 8.5 ⫾ 3.8 9.2 ⫾ 5.2 24.0 ⫾ 3.4 13.5 ⫾ 6.3 7.3 ⫾ 5.0 43.4 ⫾ 5.6

16 40.8 ⫾ 6.5 151.5 ⫾ 5.3 17.8 ⫾ 3.1 6.3 ⫾ 3.8 12.1 ⫾ 5.4 11.0 ⫾ 5.1 11.6 ⫾ 5.4 22.3 ⫾ 3.2 23.0 ⫾ 6.2 9.6 ⫾ 3.8 31.2 ⫾ 3.8

25 58.1 ⫾ 7.3** 163.0 ⫾ 5.1 21.9 ⫾ 2.8** 7.3 ⫾ 2.6* 9.2 ⫾ 3.2* 11.5 ⫾ 3.9* 13.4 ⫾ 5.5* 28.4 ⫾ 2.5*** 17.3 ⫾ 5.1* 10.4 ⫾ 3.9* 47.8 ⫾ 4.4**

16 50.5 ⫾ 9.9** 151.5 ⫾ 4.8 21.9 ⫾ 3.5** 8.9 ⫾ 4.4 17.1 ⫾ 5.5* 14.2 ⫾ 3.9* 14.0 ⫾ 5.0 26.6 ⫾ 3.5*** 27.9 ⫾ 4.5* 14.4 ⫾ 5.0** 36.0 ⫾ 5.2**

Asterisks indicate significantly different from same sex patients, * P ⬍ 0.05; ** P ⬍ 0.01; *** P ⬍ 0.001 (independent

KARYADI ET AL.

2956

FIGURE 2 The prevalence of low hemoglobin, retinol and zinc in active tuberculosis patients (n ⫽ 41, 30 and 38, respectively) and healthy controls (n ⫽ 41, 30 and 39, respectively) in Jakarta, Indonesia; f ⫽ female; m ⫽ male. Asterisks indicate significant difference between patients and controls, *P ⬍ 0.05; **P ⬍ 0.01 (␹2 test).

FIGURE 1 Distribution of biochemical variables in active tuberculosis patients and healthy controls in Jakarta, Indonesia. (A) Blood hemoglobin concentration; (B) plasma C-reactive protein (CRP); (C) serum albumin concentration; (D) plasma retinol; (E ) plasma zinc; (F ) plasma ␣-tocopherol. Each dot represents one individual. For CRP, the bars represent median and interquartile range; for all other measurements, the bars represent mean ⫾ SD (n in parentheses). Significance of differences was tested with the independent sample t test.

2). If TB patients with plasma CRP concentrations ⱖ 10 mg/L are excluded, the proportion with marginal vitamin A deficiency was 25%. No controls had elevated plasma CRP levels. In malnourished TB patients, the mean plasma retinol con-

centration was 32% lower than in well-nourished healthy controls (P ⬍ 0.05), 27% lower than in well-nourished TB patients and 18% lower than in malnourished healthy controls (Table 2). Similarly plasma zinc concentrations in patients were significantly lower than those in controls with eight patients and two controls having plasma zinc concentrations ⬍ 10.7 ␮mol/L (Fig. 2). The mean plasma zinc concentration in malnourished TB patients was 13% lower than in wellnourished healthy controls (P ⬍ 0.05) and was 7% lower than in well-nourished TB patients. Compared with malnourished healthy controls, malnourished TB patients had 10% lower mean plasma zinc concentration (Table 2). The mean plasma ␣-tocopherol concentration in patients was not significantly different from that in controls. However, 16 patients and 10 controls had serum ␣-tocopherol concentrations below normal (⬍11.5 ␮mol/L) (Fig. 1). In a stepwise multiple regression analysis, plasma retinol concentration in patients was significantly associated with BMI (␤ ⫽ 0.672, P ⫽ 0.008) and age (␤ ⫽ ⫺0.476, P ⫽ 0.032, R2 ⫽ 0.230 for both BMI and age). Sex, body temperature, presence of cavity, white blood cell count, ESR and concentrations of CRP and albumin were included in the model but did not contribute significantly to the prediction of plasma retinol concentration. Plasma zinc concentration was significantly associated with ESR (␤ ⫽ ⫺0.517, P ⫽ 0.005, R2 ⫽ 0.239), but not with CRP (␤ ⫽ 0.015, P ⫽ 0.949) and the other factors mentioned above. ZPP concentration was not associated with CRP (␤ ⫽ 0.090, P ⫽ 0.602) nor ESR (␤ ⫽ 0.204, P ⫽ 0.249).

TABLE 2 Comparison of biochemical variables in tuberculosis (TB) patients and healthy controls on the basis of their nutritional status1 Albumin (n)

Hemoglobin (n)

Retinol (n)

␮mol/L

g/L BMI ⬎ 18.5 kg/m2 Healthy controls TB patients BMI ⬍ 18.5 kg/m2 Healthy controls TB patients

Zinc (n)

47.7 ⫾ 6.0a (30) 47.0 ⫾ 7.0a (11)

136.6 ⫾ 13.5a (37) 129.4 ⫾ 19.3a (14)

1.32 ⫾ 0.6a (26) 1.22 ⫾ 0.5ab (9)

13.8 ⫾ 0.4a (35) 12.9 ⫾ 0.7ab (13)

47.6 ⫾ 5.0a (3) 41.7 ⫾ 6.0b (21)

128.5 ⫾ 8.5ab (4) 114.8 ⫾ 19.2b (27)

1.09 ⫾ 0.6ab (4) 0.89 ⫾ 0.4b (21)

13.4 ⫾ 1.3ab (4) 12.0 ⫾ 2.4b (25)

1 Values are mean ⫾ SD. Values in the same column without a common superscript are different P ⬍ 0.05, using least significant differences multiple comparisons test.

MICRONUTRIENT STATUS IN ACTIVE PULMONARY TUBERCULOSIS PATIENTS

Intakes of energy, carbohydrate, fat, protein, vitamin A and iron tended to be lower (P ⫽ 0.20 – 0.93) in patients than in controls (data not shown). DISCUSSION We demonstrated that the nutritional status assessed by measuring weight, height, mid-upper arm circumference, skinfold thicknesses and serum albumin was significantly lower in patients with active pulmonary TB compared with healthy controls. These data corroborate those found in England (Onwubalili 1988), India (Saha and Rao 1989) and Japan (Tsukaguchi et al. 1991). As in our study, TB patients had significantly lower BMI, skinfold thickness and serum albumin concentration than healthy controls. Malnutrition per se had a more pronounced effect on serum albumin concentration in TB patients. As a result, serum albumin concentration in malnourished patients was lower than that in well-nourished healthy controls, well-nourished patients and malnourished healthy controls. Lower values of transthyretin (previously referred to as prealbumin) and retinol-binding protein were reported in the Indian and Japanese studies. The poorer nutritional status of patients with pulmonary TB may be due to anorexia (Hopewell 1994), impaired absorption of nutrients or increased catabolism. Energy and nutrient intake tended to be lower in TB patients than in controls, but the differences were not significant. Two 24-h recalls are not sufficient to determine whether patients have a lower energy intake than controls (Bingham et al. 1988). Because the number of subjects or number of days required to obtain significant differences are much higher, it may be preferable to use another method to determine food intake. On the other hand, patients and controls may have similar food habits and food intakes because their socioeconomic background and living conditions are similar. Thus, infectious disease such as TB may led to impaired absorption and increased rates of metabolism (Ginzburg and Dadamukhamedov 1990, Ulijaszek 1997). The diseaseinduced production of cytokines such as interleukin-6 and tumor necrosis factor-␣ may induce fever, hepatic synthesis of acute phase reactant proteins, inhibit production of serum albumin and cause dramatic shifts in plasma concentration of certain essential micronutrients (Beisel 1998). TB is probably associated with more severe malnutrition than other chronic illnesses; in the Indian study referred to above (Saha and Rao 1989), the nutritional status of the patients with TB was worse than that of those with leprosy. Concentrations of selected micronutrients tested in our TB patients were significantly lower than in controls. Low concentrations of hemoglobin and of retinol and zinc in plasma in malnourished patients were more pronounced than in healthy controls and well-nourished patients. Furthermore, the prevalence of low concentrations of vitamin A and zinc and of anemia was higher in patients than in controls. Low concentrations of retinol in plasma can be due to a number of factors, including reduced intake or reduced absorption of fat. In addition, the infection itself can compromise vitamin A status in a number of ways. It can increase urinary excretion of vitamin A as has been shown in patients with fever, e.g., due to pneumonia and shigellosis (Mitra et al. 1998b, Stephensen et al. 1994). During the acute phase response, leakage of transthyretin (prealbumin) and albumin through the vascular endothelium occurs, and production of retinol-binding protein and transthyretin by the liver is reduced (Fleck and Myers 1985). Finally, low serum retinol levels can also result from increased utilization of retinol by tissues (Fleck and Myers

2957

1985). It is likely that a combination of mechanisms is operative in TB patients. In our study, however, low plasma retinol concentration did not correlate significantly with acute phase markers (CRP concentration and ESR). By contrast, a study in children with shigellosis showed that the serum concentration of CRP was negatively correlated with that of serum retinol (Mitra et al. 1998a). CRP is an acute phase protein whose concentration changes rapidly as a result of infection. Thus, CRP was probably not the best choice of protein to control for the acute phase changes in plasma micronutrients during a chronic illness such as TB. Therefore, it is not really surprising that CRP did not correlate with micronutrient measures. In addition, this finding may suggest that low plasma retinol is a result of a primary deficiency. Plasma zinc concentrations were significantly lower in TB patients than in controls, in agreement with a study in India (Taneja 1990). This was likely due to redistribution of zinc from plasma to other tissues (Filteau and Tomkins 1994) or reduction of the hepatic production of the zinc-carrier protein ␣2-macroglobulin and to a rise in the production of metallothionein, a protein that transports zinc to the liver (Gabay and Kushner 1999). This agrees with our finding that ESR was negatively correlated with plasma zinc concentration although not with CRP. In TB patients in this study, concentrations of hemoglobin were significantly lower and those of ZPP were significantly higher than in controls. Elevated concentrations of ZPP, a measure of free erythrocyte protoporphyrin, are indicative of iron-deficient erythropoiesis (Hastka et al. 1992). These results are not affected by the acute phase response as shown here, i.e., ZPP did not significantly correlate with CRP levels. Low iron status, as measured by low serum iron concentrations and total iron-binding capacity, has also been reported in pulmonary TB patients in England (Onwubalili 1988). There are two explanations for the association of low iron status with infection. One is that anemia results from chronic infection. The other, which is more speculative, is that iron deficiency would increase susceptibility to an infection such as TB. In this context, it is relevant that cell-mediated immunity is compromised in iron deficiency (Dallman 1987) before anemia becomes apparent. In conclusion, this study shows that the nutritional status of patients with active pulmonary TB was poor compared with healthy controls. The prevalence of anemia and low concentrations of plasma retinol and zinc was significantly higher in patients than in controls. The low concentrations of hemoglobin, and of retinol and zinc in plasma were more pronounced in malnourished TB patients. Further studies are required to establish the role of these low concentrations in host defense against TB. LITERATURE CITED Arroyave, G., Chichester, C. O., Flores, H., Glover, J., Mejia, L. A., Olson, J. A., Simpson, K. L. & Underwood, B. A. (1982) Biochemical methodology for the assessment of vitamin A status: a report of the International Vitamin A Consultative Group. Nutrition Foundation, Washington, DC. Beisel, W. R. (1998) Metabolic responses of the host to infection. In: Textbook of Pediatric Diseases (Feigin, R. D. & Cherry, J. D., eds.), pp. 54 – 69. W. B. Saunders, Philadelphia, PA. Bingham, S. A., Nelson, M., Paul, A. A., Haraldsdottir, J., Loken, E. B. & Van Staveren, W. A. (1988) Method for data collection at an individual level. In: Manual on Methodology for Food Consumption Studies (Cameron, M. E. & Van Staveren, W. A., eds.), pp. 83– 88. Oxford University Press, New York, NY. Chadwick, M. V. (1982) Mycobacteria, pp. 27– 64. Stonebridge Press, London, UK. Chandra, R. K. (1991) 1990 McCollum Award lecture. Nutrition and immunity:

2958

KARYADI ET AL.

lessons from the past and new insights into the future. Am. J. Clin. Nutr. 53: 1087–1101. Council for International Organization of Medical Sciences (1991) International guidelines for ethical review of epidemiological studies. Geneva, Switzerland. Crowle, A. J. & Ross, E. J. (1989) Inhibition by retinoic acid of multiplication of virulent tubercle bacilli in cultured human macrophages. Infect. Immun. 57: 840 – 844. Dallman, P. R. (1987) Iron deficiency and the immune response. Am. J. Clin. Nutr. 46: 329 –334. Dumas, B. T., Watson, W. A. & Biggs, H. G. (1997) Albumin standards and the measurement of serum albumin with bromcresol green. Clin. Chim. Acta 258: 21–30. Durnin, J. V. & Womersley, J. (1974) Body fat assessed from total body density and its estimation from skinfold thickness: measurements on 481 men and women aged from 16 to 72 years. Br. J. Nutr. 32: 77–97. Filteau, S. M. & Tomkins, A. M. (1994) Micronutrients and tropical infections. Trans. R. Soc. Trop. Med. Hyg. 88: 1–3. Fleck, A. & Myers, M. A. (1985) Diagnostic and prognostic significance of the acute phase proteins. In: The Acute Phase Response to Injury and Infection (Gordon, A. H. & Koj, A., eds.), pp. 249 –271. Elsevier, Amsterdam, The Netherlands. Gabay, C. & Kushner, I. (1999) Acute-phase proteins and other systemic responses to inflammation. N. Engl. J. Med. 340: 448 – 454. Gibson, R. S. (1990) Principles of Nutritional Assessment, pp. 187–208. Oxford University Press, New York, NY. Gibson, R. S (1993) Nutritional Assessment. Oxford University Press, New York, NY. Ginzburg, V. S. & Dadamukhamedov, A. A. (1990) Absorption of nutrients in patients with pulmonary tuberculosis. Probl. Tuberk. 10: 44 – 46 (abs.). Goldberg, G. (1946) Clinical Tuberculosis, pp. E3–E45. F. A. Davis, Philadelphia, PA. Hastka J., Lasserre J. J., Schwarzbeck, A., Strauch, M. & Hehlmann, R. (1992) Washing erythrocytes to remove interferents in measurements of zinc protoporphyrin by front-face hematofluorometry. Clin. Chem. 38: 2184 –2189. Hopewell, P. C. (1994) Overview of clinical tuberculosis. In: Tuberculosis: Pathogenesis, Protection and Control (Bloom, B. R., ed.), pp. 25–36. ASM Press, Washington, DC. Household Health Survey (1995) Department of Health RI, Jakarta, Indonesia. James, W.P.T., Ferro-Luzzi, A. & Waterlow, J. C. (1988) Definition of chronic energy deficiency in adults. Eur. J. Clin. Nutr. 42: 969 –981. Kohli, R. N., Singh, S. & Singh M. (1975) Studies on erythrocyte sedimentation rate in buffaloes. Evaluation of various techniques. Indian Vet. J. 52: 12, 915–918.

McMurray, D. N., Bartow, R. A., Mintzer, C. L. & Hernandez, F. E. (1990) Micronutrient status and immune function in tuberculosis. Ann. N.Y. Acad. Sci. 587: 59 – 69. Metzmann, E. (1985) Protein quantitation on both branches of the Heidelberger curve by monitoring the kinetic of immunoprecipitation. Behring Inst. Mitt. 78: 167–175. Mitra, A. K., Alvarez, J. O., Wahed, M. A., Fuchs, G. J. & Stephensen, C. B. (1998a) Predictors of serum retinol in children with shigellosis. Am. J. Clin. Nutr. 68: 1088 –1094. Mitra, A. K., Alvarez, J. O., Woodford, L. G., Fuchs, G. J., Wahed, M. A. & Stephensen, C. B. (1998b) Urinary retinol excretion and kidney function in children with shigellosis. Am. J. Clin. Nutr. 68: 1095–1103. Narang, A.P.S., Whig, J. & Mahajan, R. (1995) Serum copper and zinc levels in patients with pulmonary tuberculosis. Trace Elem. Electrolytes 12: 74 –75. Onwubalili, J. K. (1988) Malnutrition among tuberculosis patients in Harrow, England. Eur. J. Clin. Nutr. 42: 363–366. Prasad, A. S., Oberleas, D. & Holsted, J. A. (1965) Determination of zinc in biological fluids by atomic absorption spectrophotometry in normal and cirrhotic subjects. J. Lab. Clin. Med. 508 –516. Reichman, L. B. (1996) How to ensure the continued resurgence of tuberculosis. Lancet 347: 175–177. Rwangabwoba, J. M., Fischman, H. & Semba, R. D. (1998) Serum vitamin A levels during tuberculosis and human immunodeficiency virus infection. Int. J. Tuberc. Lung Dis. 2: 771–773. Saha, K. & Rao, K. N. (1989) Undernutrition in lepromatous leprosy. V. Severe nutritional deficit in lepromatous patients co-infected with pulmonary tuberculosis. Eur. J. Clin. Nutr. 43: 117–128. Smith, J. C., McDaniel, E. G., Fan, F. F. & Halsted, J. A. (1973) Zinc: a trace element essential in vitamin A metabolism. Science (Washington, DC) 181: 954 –955. Stephensen, C. B., Alvarez, J. O., Kohatsu, J., Hardmeier, R., Kennedy, J. I. & Gammon, R. B. (1994) Vitamin A is excreted in the urine during acute infection. Am. J. Clin. Nutr. 60: 388 –392. Taneja, D. P. (1990) Observations on serum zinc in patients of pulmonary tuberculosis. J. Indian Med. Assoc. 88: 280 –281. 275 (abs.). Tsukaguchi, K., Yoneda, T., Yoshikawa, M., Narita, N., Enoki, Y., Miyazaki, R., Shirai, F., Kitamura, H. & Tsukaguchi, M. (1991) Interaction between interleukin-1 and tumor necrosis factor productions by peripheral blood monocytes and nutritional disturbance in active pulmonary tuberculosis. Kekkaku 66: 447– 484 (abs.). Ulijaszek, S. (1997) Transdisciplinarity in the study of undernutrition-infection interactions. Coll. Antropol. 21: 3–15.