Abstract
Rationale: Although nontuberculous mycobacteria (NTM) are widely documented as a cause of illness among HIV-infected people in the developed world, studies describing the prevalence of NTM disease among HIV-infected people in most resource-limited settings are rare.
Objectives: To evaluate the prevalence of mycobacterial disease in HIV-infected patients in Southeast Asia.
Methods: We enrolled people with HIV from three countries in Southeast Asia and collected pulmonary and extrapulmonary specimens to evaluate the prevalence of mycobacterial disease. We adapted American Thoracic Society/Infectious Disease Society of America guidelines to classify patients into NTM pulmonary disease, NTM pulmonary disease suspects, NTM disseminated disease, and no NTM categories.
Measurements and Main Results: In Cambodia, where solid media alone was used, NTM was rare. Of 1,060 patients enrolled in Thailand and Vietnam, where liquid culture was performed, 124 (12%) had tuberculosis and 218 (21%) had NTM. Of 218 patients with NTM, 66 (30%) were classified as NTM pulmonary disease suspects, 9 (4%) with NTM pulmonary disease, and 10 (5%) with NTM disseminated disease. The prevalence of NTM disease was 2% (19 of 1,060). Of 51 patients receiving antiretroviral therapy (ART), none had NTM disease compared with 19 (2%) of 1,009 not receiving ART.
Conclusions: Although people with HIV frequently have sputum cultures positive for NTM, few meet a strict case definition for NTM disease. Consistent with previous studies, ART was associated with lower odds of having NTM disease. Further studies of NTM in HIV-infected individuals in tuberculosis-endemic countries are needed to develop and validate case definitions.
The prevalence of pulmonary and disseminated disease due to nontuberculous mycobacteria (NTM) in people with HIV has been well described in industrialized countries. Less is known about the prevalence of NTM disease in developing countries with a high burden of tuberculosis (TB) and HIV.
When broth-based culture for mycobacteria was used in people with HIV in Southeast Asia, isolation of NTM was very common. Many of the NTM isolates were known and a small proportion of patients had NTM isolated from a sterile site or met the American Thoracic Society/Infectious Disease Society of America (ATS/IDSA) guidelines for NTM pulmonary disease. Our data suggest that NTM contribute to clinical disease in people with HIV. Further studies are required to refine the diagnostic and treatment guidelines for NTM disease in this population.
Nontuberculous mycobacteria (NTM) are free-living, saprophytic organisms that were first reported as an important cause of disease among people living with HIV/AIDS in the 1980s (1–3). Diagnosis of NTM disease is challenging in this population due to the high prevalence of Mycobacterium tuberculosis (MTB) disease, which has similar microbiologic and clinical characteristics. Both appear as acid-fast bacilli (AFB) by microscopy, and differentiation requires molecular, biochemical, or other tests (4, 5). In HIV-infected people, both NTM and MTB disease can present with cavitary or noncavitary lung disease as well as disseminated disease, accompanied by symptoms such as fever, night sweats, and weight loss, and both diseases can produce severe illness and death. In people with HIV, TB is typically diagnosed across a broad range of CD4 cell levels, whereas disseminated disease due to NTM occurs when CD4 cell levels are less than 100 cells/μl (6). In industrialized, low TB–burden countries, disseminated NTM disease in HIV-infected people has been well documented (7, 8). Given the low percentage of NTM pulmonary disease observed in resource-limited countries with a high burden of TB and HIV when compared with the higher percentage of mycobacterial pulmonary disease due to NTM in low TB–burden countries (9), it is generally assumed that HIV-infected people in TB-endemic settings with cough and either AFB-positive sputum, cavitary lung disease, or febrile wasting illness have TB. Few high-quality studies, however, have assessed the prevalence of NTM in this population in these countries (10, 11–16).
Because NTM live freely in the environment, particularly in soil and water, NTM found in nonsterile patient specimens may be attributable to environmental contamination, colonization, or disease. To address this diagnostic challenge in the United States, the American Thoracic Society (ATS) published guidelines for NTM diagnosis in 1997 and subsequently updated these guidelines with the Infectious Disease Society of America (IDSA) in 2007 (17, 18). According to the ATS/IDSA guidelines, NTM disease can be divided into four specific clinical syndromes: lymphadenitis, cutaneous disease, disseminated disease, and pulmonary disease (18). For pulmonary disease, which is the most common NTM clinical syndrome, the ATS/IDSA guidelines specify strict clinical and microbiologic criteria to determine whether NTM isolated from a respiratory specimen is clinically relevant, including: (1) pulmonary symptoms or radiographic abnormalities, including nodular or cavitary opacities on chest radiograph or multifocal bronchiectasis with multiple small nodules on high-resolution computed tomography (CT) scan; (2) appropriate exclusion of other diagnoses; and (3) NTM-positive cultures from at least two separately expectorated sputum samples or from at least one bronchial wash (18).
Although the ATS/IDSA guidelines are referred to by clinicians throughout the world, they are subject to two important limitations. First, as acknowledged in the guidelines, they represent a limited perspective of NTM disease, relying primarily on data and expert opinion from the United States. Second, the guidelines do not specifically address how to evaluate the significance of pulmonary NTM in HIV-infected individuals.
We enrolled people infected with HIV living in Cambodia, Thailand, and Vietnam to develop an evidence-based clinical algorithm in this population for diagnosing tuberculosis in resource-limited settings (19). The estimated TB incidence during the study period was 495/100,000 in Cambodia, 142/100,000 in Thailand, and 171/100,000 in Vietnam (20). We further analyzed data from our study to improve understanding about the prevalence and risk factors of NTM in countries with a high burden of TB and HIV (19).
Details of the patient enrollment, specimen collection, and laboratory procedures for this study have been published elsewhere (19, 21, 22). Briefly, from September 2006 to July 2008, we recruited newly enrolled people with HIV from outpatient facilities that perform HIV counseling, testing, and clinical care, including four clinics in Cambodia (two in Banteay Meanchey province, one in Battambang, and one in Phnom Penh), one in Bangkok, Thailand, and three in Ho Chi Minh City, Vietnam. All persons 7 years of age and older presenting to the clinics during the enrollment period were screened for study eligibility and, if eligible, offered enrollment. In the parent study, patients on antiretroviral therapy (ART) were excluded from analysis (19); in the present NTM study these patients were included.
After providing written informed consent, patients underwent a standardized interview and physical examination, chest radiography, and blood testing for complete blood cell count and CD4+ T lymphocytes (CD4 count). Patients also provided three sputum specimens and one specimen each of urine, blood, and stool. If patients had enlarged peripheral lymph nodes (>1 cm for noninguinal nodes, >2 cm for inguinal nodes), the largest palpable node was aspirated.
In Thailand and Vietnam, specimens were inoculated into broth-based culture media (Mycobacterial Growth Indicator Tube × 1 [500 μl] [BD, Franklin Lakes, NJ]) and on solid media (Lowenstein Jensen × 2 [150–200 μl]); in Cambodia, solid media only was used. Blood specimens were inoculated directly into Myco-F-Lytic bottles (BD, Franklin Lakes, NJ) and placed into an automated blood culture instrument (BACTEC 9050/9120/9240 system, BD).
AFB-positive cultures were identified using niacin accumulation and nitrate reduction (23, 24); high-performance liquid chromatography (25), the Genotype CM/AS assay (Hain Lifescience, Nehren, Germany), and the Accuprobe MTB assay (GenProbe, San Diego, CA).
Given the use of solid culture media only in Cambodia and the marked increase in NTM recovery when liquid culture methods are applied, we believed that we could not reliably categorize patients from Cambodia with respect to NTM status. Therefore, the Cambodia data were only used to evaluate the difference in NTM prevalence across countries and to explain why that difference in yield may have occurred. Patients from Cambodia were excluded from all analyses evaluating characteristics associated with NTM disease. Patients from Thailand and Vietnam were grouped into five categories: pulmonary TB disease, NTM pulmonary disease, NTM pulmonary disease suspect, NTM disseminated disease, and no pulmonary mycobacterial disease.
Pulmonary TB disease was defined as having any sputum specimen culture positive for MTB (including patients with only a sputum culture positive for MTB and patients with both sputum and nonsputum specimens positive for MTB), regardless of the growth of NTM. This approach was used because the definition of NTM pulmonary disease requires appropriate exclusion of other diagnoses. Patients with only extrapulmonary TB were eligible to be included in one of the NTM categories if they met criteria for one of them, because they did not, by definition, have pulmonary TB.
We modified the ATS/IDSA microbiologic and clinical criteria for classification of NTM pulmonary disease, given the high risk of mycobacterial morbidity and mortality associated with HIV infection and given the design of the study, which only collected sputum specimens at one point in time, thereby not affording the opportunity to have a positive culture for NTM at multiple time points. Although ATS/IDSA criteria require isolation of NTM from at least two separate sputum specimens or isolation of NTM from at least one bronchial wash or lavage, we classified isolation of NTM from at least one sputum specimen as sufficient in this population. We did not perform chest high-resolution CT scan but did use chest radiography results to further categorize patients. Our definitions of NTM pulmonary disease, NTM pulmonary disease suspects, and NTM disseminated disease are provided in Table 1. We analyzed clinical characteristics of patients, stratified by these diagnostic categories. Among patients classified as NTM suspects or NTM disease, we performed similar analyses stratifying by NTM species.
| NTM pulmonary disease: |
| One or two sputum cultures positive for NTM, and |
| Pulmonary symptoms (cough or hemoptysis or difficulty breathing), andChest radiograph abnormality (cavity or mass/nodule or infiltrate), andNo sputum specimen culture positive for MTB |
| NTM pulmonary disease suspect: |
| One or two sputum cultures positive for NTM, and |
| No sputum specimen culture positive for MTB, and |
| Pulmonary symptoms (cough or hemoptysis or difficulty breathing),orChest radiograph abnormality (cavity or mass/nodule or infiltrate) |
| NTM disseminated disease: |
| Blood, urine, or lymph node aspirate culture positive for NTM |
For all analyses, we compared proportions (categorical variables) using chi-square and medians (continuous variables) using the Wilcoxon rank-sum test. Statistical significance was defined as a two-sided P less than 0.05. To identify predictors of NTM disease, we conducted multivariable analysis comparing patients with NTM disease (pulmonary or disseminated) to patients who did not have NTM disease, were not NTM pulmonary disease suspects, and did not have pulmonary TB. In multivariable analysis, we first assessed for colinearity and then included variables that had P less than 0.2 in bivariate analysis. No variables that directly affect the NTM pulmonary disease definition were included. When multiple variables of a single type (e.g., duration of cough for 2, 3, or 4 wk) had P less than 0.2, the variable with the lowest P value was used. Manual stepwise selection was used with P < 0.10 required for variables to be retained.
The study was approved by human subjects review committees at U.S. Centers for Disease Control and Prevention and collaborating institutions in each country.
Sexual transmission and injection drug use were the primary HIV transmission categories in each country during the study period (26). Of 2,009 people with HIV screened and enrolled across the three countries, 21 patients could not be accurately classified as having or not having TB due to insufficient data and were therefore excluded from further analysis. Of the remaining 1,988 patients with sufficient data, MTB was detected in 249 (13%), and 220 (11%) patients had at least one culture positive for NTM (Figure 1). Of 630 patients enrolled in Thailand, 187 (30%) had at least one culture positive for NTM compared with 31 (7%) of 430 in Vietnam and two (< 1%) of 928 in Cambodia (P < 0.01 for both comparisons).
Figure 1. Patient enrollment and results. * Seven patients included in these groups were culture positive for both tuberculosis (TB) and nontuberculous mycobacteria (NTM).
[More] [Minimize]Of 1,060 patients from Thailand and Vietnam included in our analysis, 342 (32%) had at least one positive culture for mycobacterium: 124 (36%) MTB, 218 (64%) NTM; 7 (2%) of these patients were positive for both TB and NTM. Of the 218 patients in whom at least one culture grew NTM, 190 (87%) were culture positive for NTM in liquid media only, and 28 (13%) were positive in liquid and solid media. We performed Accuprobe on 188 mycobacteria-positive liquid cultures (including repeat Accuprobe on two patients) from 187 patients and found three patients who were originally confirmed as NTM to also be positive for TB.
Of the 218 patients from Thailand and Vietnam with at least one NTM-positive culture, 66 (30%) were classified as NTM pulmonary disease suspects, nine (4%) as having NTM pulmonary disease, and 10 (5%) as having NTM disseminated disease. The remaining 133 (61%) patients did not meet our criteria for any of the NTM categories (Table 2). Of the nine patients categorized as having NTM pulmonary disease, five (56%) were positive from a single sputum specimen; the remaining four (44%) patients had two or more sputum specimens that were positive. Of the 66 patients classified as NTM pulmonary disease suspects, 42 (64%) were positive from a single sputum specimen, and the remaining 24 (36%) had two or more sputum specimens that were positive (data not shown). The overall prevalence of NTM disease in this population was 2% (19/1,060).
| No TB, Not an NTM Suspect or Case (N = 851) | NTM Pulmonary Disease Suspect (N = 66) | NTM Pulmonary Disease (N = 9) | NTM Disseminated Disease (N = 10) | TB (N = 124) | |
| Symptoms | |||||
| Any cough | 308 (36) | 51 (77) | 8 (89) | 5 (50) | 87 (70) |
| Hemoptysis | 23 (3) | 4 (6) | 1 (11) | 0 | 4 (3) |
| Daily cough | 254 (30) | 46 (70) | 8 (89) | 5 (50) | 66 (53) |
| Cough with sputum | 235 (28) | 41 (62) | 7 (78) | 4 (40) | 66 (53) |
| Cough ≥ 2 wk | 138 (16) | 17 (26) | 6 (67) | 3 (30) | 39 (32) |
| Cough ≥ 3 wk | 96 (11) | 11 (17) | 6 (67) | 3 (30) | 25 (20) |
| Cough ≥ 4 wk | 79 (9) | 10 (15) | 6 (67) | 3 (30) | 21 (17) |
| Cough with insomnia | 71 (8) | 13 (20) | 3 (33) | 1 (10) | 26 (21) |
| Difficulty breathing | 174 (20) | 25 (38) | 5 (56) | 6 (60) | 54 (44) |
| Respiratory symptoms | 381 (44) | 62 (94) | 9 (100) | 8 (80) | 99 (80) |
| Any fever | 228 (27) | 34 (52) | 6 (67) | 4 (40) | 74 (60) |
| Fever ≥ 2 wk | 82 (10) | 8 (12) | 6 (67) | 3 (30) | 39 (32) |
| Fever ≥ 3 wk | 52 (6) | 8 (12) | 6 (67) | 2 (20) | 27 (22) |
| Fever ≥ 4 wk | 39 (5) | 8 (12) | 6 (67) | 2 (20) | 22 (18) |
| Any night sweats | 80 (9) | 9 (14) | 2 (22) | 4 (40) | 38 (31) |
| Night sweats ≥ 2 wk | 44 (5) | 3 (5) | 2 (22) | 2 (22) | 17 (14) |
| Night sweats ≥ 3 wk | 28 (3) | 3 (5) | 2 (22) | 2 (20) | 14 (11) |
| Night sweats ≥ 4 wk | 22 (3) | 2 (3) | 2 (22) | 1 (10) | 11 (9) |
| Weight loss | 245 (29) | 22 (33) | 6 (67) | 4 (40) | 70 (56) |
| Headache | 373 (44) | 35 (53) | 4 (44) | 6 (60) | 64 (52) |
| Abdominal pain | 151 (18) | 15 (23) | 2 (22) | 2 (20) | 38 (31) |
| Difficulty sleeping | 322 (38) | 31 (47) | 5 (56) | 3 (30) | 60 (48) |
| Dizzy or light-headed | 321 (38) | 37 (56) | 5 (56) | 2 (20) | 75 (60) |
| Fatigue | 334 (39) | 39 (59) | 5 (56) | 6 (60) | 78 (63) |
| Loss of appetite | 232 (27) | 25 (38) | 7 (78) | 3 (30) | 67 (54) |
| Muscle weakness | 159 (19) | 18 (27) | 4 (44) | 4 (40) | 38 (31) |
| Nausea or vomiting | 141 (17) | 17 (26) | 2 (22) | 2 (20) | 32 (26) |
| Pain in muscles or joints | 333 (39) | 29 (44) | 5 (56) | 8 (80) | 49 (40) |
| Pain when swallowing | 165 (19) | 21 (32) | 2 (22) | 5 (50) | 35 (28) |
| Shaking chills | 134 (16) | 18 (27) | 7 (78) | 3 (30) | 42 (34) |
| Medications | |||||
| Antiretroviral therapy ≥ 2 wk | 46 (6) | 4 (6) | 0 | 0 | 1 (2) |
| Other history | |||||
| Ever treated for TB | 15 (2) | 2 (3) | 1 (11) | 0 | 1 (1) |
| Treated for TB in the past yr | 0 | 0 | 0 | 0 | 0 |
| Ever injected drug | 167 (20) | 10 (15) | 1 (11) | 6 (60) | 63 (51) |
| TB exposure | |||||
| Anyone at home currently treated for TB | 48 (6) | 1 (2) | 1 (11) | 0 | 5 (4) |
| Anyone at home treated for TB in the past 2 yr | 104 (12) | 5 (8) | 2 (22) | 0 | 19 (15) |
| Know anyone currently treated for TB | 43 (5) | 1 (2) | 0 | 0 | 14 (11) |
| Know anyone treated for TB in the past 2 yr | 68 (8) | 1 (2) | 0 | 0 | 19 (15) |
Both respiratory and nonrespiratory symptoms were common in all patients, regardless of their disease classification (Table 2). Abnormalities (nodule, cavity, or infiltrate) were observed on chest radiography in four (6%) of 66 patients classified as NTM pulmonary disease suspects, compared with all nine (100%) patients with NTM pulmonary disease (Table 3). The most frequently isolated Mycobacterium species was Mycobacterium kansasii, which was isolated in 6 (67%) of 9 patients with NTM pulmonary disease and 17 (26%) of 66 NTM pulmonary disease suspects. Mycobacterium avium, Mycobacterium intracellulare (MAC), and Mycobacterium spp. unidentified were the most frequently isolated from patients with NTM disseminated disease (Table 4).
| No TB, Not an NTM Suspect or Case (N = 851) | NTM Pulmonary Disease Suspect (N = 66) | NTM Pulmonary Disease (N = 9) | NTM Disseminated Disease (N = 10) | TB (N = 124) | |
| Physical examination | |||||
| BMI < 18.5 | 175 (21) | 12 (18) | 3 (33) | 4 (40) | 50 (40) |
| Age, median (IQR) | 30 (26–36) | 30 (27–39) | 32 (29–35) | 31.5 (29–33) | 29 (26–35) |
| BMI, median (IQR) | 20.6 (18.8–22.7) | 21.6 (19.3–23.8) | 20.0 (18.0–20.8) | 19.2 (17.6–22.9) | 19.1 (17.6–20.4) |
| CD4, median (IQR) | 300 (157–453) | 261 (90–396) | 89 (14–275) | 173 (3–357) | 124 (43–318) |
| Hemoglobin, median (IQR) | 13.4 (12.3–14.5) | 13.1 (11.6–14.2) | 10.9 (10.2–13.3) | 11.1 (10.8–12.2) | 11.7 (10.2–13.4) |
| Neutrophil count, median (IQR) | 3.2 (2.4–4.2) | 3.1 (2.2–4.5) | 3.6 (3.1–3.9) | 3.9 (2.8–4.5) | 4.5 (3.0–6.1) |
| Platelets, median (IQR) | 230 (189–274) | 230 (184–279) | 233 (166–270) | 290 (245–548) | 259 (205–320) |
| Temperature ≥ 38°C | 8 (1) | 2 (3) | 0 | 0 | 15 (12) |
| BCG scar | 589 (69) | 49 (74) | 7 (78) | 7 (70) | 90 (9) |
| Heart rate > 100/min | 49 (6) | 2 (3) | 1 (11) | 1 (10) | 43 (35) |
| Jaundice | 2 (0) | 1 (2) | 0 | 0 | 0 |
| Karnofsky score < 70 | 7 (1) | 2 (3) | 1 (11) | 1 (10) | 2 (2) |
| Needle tracks | 21 (2) | 3 (5) | 0 | 0 | 14 (11) |
| Any lymphadenopathy | 80 (9) | 9 (14) | 1 (11) | 2 (20) | 36 (29) |
| Chest radiography findings | |||||
| Nodule | 10 (1) | 1 (2) | 4 (44) | 1 (10) | 8 (6) |
| Cavity | 3 (0) | 0 | 2 (22) | 0 | 7 (6) |
| Infiltrate | 49 (6) | 3 (5) | 6 (67) | 2 (20) | 52 (42) |
| Nodule, cavity, or infiltrate | 56 (7) | 4 (6) | 9 (100) | 3 (30) | 57 (46) |
| Normal chest radiograph | 769 (87) | 59 (89) | 0 | 6 (60) | 52 (42) |
| No TB, Not an NTM Suspect or Case (N = 851) | NTM Pulmonary Disease Suspect (N = 66) | NTM Pulmonary Disease (N = 9) | NTM Disseminated Disease (N = 10) | TB (N = 124) | |
| Microbiology | |||||
| ≥1 Positive sputum smear, n (%) | 8 (1) | 3 (5) | 0 | 2 (20) | 45 (36) |
| NTM species | |||||
| Mycobacterium abscessus | 21 | 8 | 1 | 1 | 0 |
| Mycobacterium kansasii | 26 | 17 | 6 | 0 | 0 |
| Mycobacterium fortuitum | 13 | 7 | 0 | 1 | 0 |
| Mycobacterium avium | 1 | 1 | 0 | 2 | 0 |
| Mycobacterium intracellulare | 7 | 5 | 0 | 1 | 0 |
| Mycobacterium simiae | 2 | 2 | 0 | 1 | 0 |
| Mycobacterium fortuitum/ peregrinum | 1 | 0 | 0 | 0 | 0 |
| Mycobacterium chelonae/abscessus | 0 | 1 | 0 | 0 | 0 |
| Mycobacterium szulgai | 5 | 0 | 0 | 0 | 0 |
| Mycobacterium scrofulaceum | 7 | 2 | 0 | 0 | 1 |
| Mycobacterium lentiflavum | 0 | 2 | 0 | 0 | 0 |
| Mycobacterium gordonae | 7 | 1 | 0 | 0 | 0 |
| Mycobacterium interjectum | 0 | 1 | 0 | 0 | 0 |
| Mycobacterium asiaticum | 1 | 1 | 0 | 0 | 0 |
| Mycobacterium haemophilum | 1 | 0 | 0 | 0 | 0 |
| Mixed (2 NTM species) | 5 | 5 | 2 | 1 | 0 |
| Unidentified | 21 | 13 | 0 | 3 | 2 |
In bivariate analysis, patients with pulmonary or disseminated NTM disease were more likely than patients without NTM and without TB to have fever greater than 4 weeks, shaking chills, night sweats greater than 3 weeks, and pain when swallowing. An association between receipt of ART and absence of NTM disease was observed; of 51 patients receiving ART for 2 weeks or more, none had NTM disease compared with 19 (2%) of 1,009 that were not receiving ART. In multivariable analysis, fever lasting greater than 4 weeks, cough lasting greater than 4 weeks, muscle or joint pain, and history of injection drug use were independently associated with having NTM disease (includes pulmonary and disseminated NTM disease) (Table 5). CD4 levels were not significantly associated with NTM disease. Because inclusion of ART as a variable in the multivariable analysis made the model unstable, it was not included in the final model. Except for the presence of an abnormal chest radiograph, which was part of the case definition, clinical characteristics and CD4 levels of patients with NTM pulmonary disease were similar to patients with NTM disseminated disease.
| Characteristic | Pulmonary, Disseminated NTM Disease (N = 19) | No NTM Disease (N = 851) | OR (95% CI) Bivariate | Adjusted OR (95% CI) | P Value |
| Fever ≥ 4 wk | 8 (42) | 39 (5) | 15.1 (5.8–39.7) | 6.8 (2.3–19.9) | <0.001 |
| Cough ≥ 4 wk | 9 (47) | 79 (9) | 8.8 (3.4–22.2) | 4.6 (1.6–13.1) | 0.004 |
| Pain muscles/joints | 13 (68) | 333 (39) | 3.2 (1.2–8.0) | 3.1 (1.1–8.6) | 0.032 |
| Ever injected drug | 7 (37) | 167 (20) | 2.4 (0.93–6.2) | 2.4 (0.87–6.6) | 0.091 |
Compared with patients classified as NTM pulmonary disease suspects, NTM pulmonary disease patients were more likely to have lower body mass index (BMI) (P = 0.019), fever lasting greater than 2 weeks (P < 0.001), shaking chills (P = 0.005), and cough lasting more than 4 weeks (P = 0.002). The median CD4 level was lower for pulmonary disease patients than for suspects, but the comparison was not significant (P = 0.11) (Table 6).
| Characteristic | NTM Pulmonary Disease (N = 9) | NTM Pulmonary Disease Suspects (N = 66) | P Value |
| CD4, median (IQR) | 89 (14–275) | 261 (90–396) | 0.11 |
| BMI, median (IQR) | 20.0 (18.0–20.8) | 21.6 (19.3–23.8) | 0.019 |
| Fever > 2 wk, n (%) | 6 (67) | 8 (12) | <0.001 |
| Cough > 4 wk, n (%) | 6 (67) | 10 (15) | 0.002 |
| Shaking chills, n (%) | 7 (78) | 18 (27) | 0.005 |
In multivariable analysis comparing patients classified as NTM pulmonary disease suspects to patients without NTM, characteristics independently associated with being an NTM pulmonary disease suspect included daily cough, fever, knowing someone in the past 2 years who had TB disease, and dizziness/light-headedness (Table 7). Of the 66 patients classified as NTM pulmonary disease suspects, many grew NTM species that are known pathogens in people with HIV infection. Among NTM pulmonary disease suspects with these species, only those with MAC had a significantly lower CD4 count (78 cells/μl, IQR 10–104) than patients without TB or NTM disease (300 cells/μl, IQR 157–453). This was lower than the CD4 count for suspects with other species of NTM as well (Table 8).
| Characteristic | NTM Pulmonary Disease Suspects (N = 66) | No NTM Disease (N = 851) | OR (95% CI) Bivariate | Adjusted OR (95% CI) | P Value |
| Cough daily | 46 (70) | 254 (30) | 5.4 (3.1–9.3) | 4.6 (2.6–8.1) | <0.001 |
| Fever | 34 (2) | 228 (27) | 2.9 (1.8–4.8) | 1.7 (0.97–2.9) | 0.065 |
| Know anyone with TB, past 2 yr | 1 (2) | 68 (8) | 0.18 (0.02–1.3) | 0.15 (0.02–1.1) | 0.63 |
| Dizzy or light-headed | 37 (56) | 321 (38) | 2.1 (1.3–3.5) | 1.7 (1.0–3.0) | 0.04 |
| NTM Pulmonary Disease Suspects (N = 66) | ||||||
| Mycobacterium kansasii (N = 17) | Mycobacterium chelonae/abscessus (N = 9) | Mycobacterium avium/intracellulare (N = 6) | Mycobacterium fortuitum (N = 7) | Other NTM (N = 27) | No NTM or TB (N = 851) | |
| CD4 | 215 (101–426), P = 0.25 | 200 (154–220), P = 0.021 | 78 (10–104), P = 0.0019 | 313 (251–369), P = 0.46 | 290 (83–462), P = 0.37 | 300 (157–453) |
| BMI | 20.6 (19.6–21.7), P = 0.37 | 18.1 (17.8–21.6), P = 0.080 | 18.3 (17.6–19.6), P = 0.040 | 21.7 (20.6–25.4), P = 0.075 | 22.6 (21.7–24.1), P < 0.001 | 20.6 (18.8–22.7) |
Although the prevalence of MTB in HIV-infected people living in Southeast Asia has been reported previously (19, 27–29), studies describing the extent of NTM disease in this region are limited. Even though HIV-infected people living in Southeast Asia frequently have sputum cultures positive for NTM, we found that only approximately 1% met a strict case definition for pulmonary NTM disease. A similar proportion had disseminated NTM disease.
A key finding of this study was that sensitive detection of NTM required the use of liquid culture media in this population. In Cambodia, where only solid culture media was used, only 2 of 928 (<1%) patients enrolled had at least one culture positive for NTM; in Thailand and Vietnam, where liquid and solid culture were used, 187 of 630 (30%) and 31 of 430 (7%) patients, respectively, had at least one culture positive for NTM. That only 13% of all patients with NTM in liquid media had a positive result on solid media further supports this point. We believe that this finding has three key clinical implications: (1) when seeking to detect NTM in a patient with HIV, use of liquid culture media is essential; (2) as the use of broth-based culture systems increases in high–TB burden settings, laboratories will need to implement identification methods that identify NTM to species level (30), and clinicians will need to differentiate NTM infection from NTM disease ; and (3) patients with mixed MTB and NTM infections may be erroneously diagnosed as NTM only. Laboratories seeking to diagnose TB in such populations with liquid culture media will need to have available techniques that detect MTB and NTM in mixed cultures (31). By performing Accuprobe on liquid cultures that grew NTM, we detected three patients with TB that would not previously have been detected by performing subcultures and biochemical identification methods alone.
We are unaware of an evidence-based classification system for pulmonary NTM disease for people infected with HIV. Determining the clinical relevance of NTM and need for treatment in this population requires following patients with additional diagnostic and microbiologic evaluations. Therefore, we relied on ATS/IDSA definitions, with minor modifications, to divide patients with positive cultures into suspects and those with disease. We found that patients classified as having NTM pulmonary disease more frequently reported systemic symptoms and signs—such as fever, chills, and low BMI—and had a lower CD4 count than patients classified as NTM pulmonary disease suspects or as having no TB or NTM. M. kansasii, a well-established pathogen in HIV-infected people causing chronic, pulmonary disease clinically resembling TB (32, 33), was isolated in six (67%) of nine patients with NTM pulmonary disease. This pathogen has been reported in HIV-infected individuals in other high TB–burden settings, including South Africa (12), as well as previous studies from Thailand (13). Three patients classified with NTM pulmonary disease in our study were positive for M. kansasii from a single sputum culture; previous investigators have also reported the clinical significance of M. kansasii from a single isolate (34). These findings suggest that reliance solely on ATS/IDSA case definitions may not be sufficient for a population that is severely ill, possibly from NTM.
Although these definitions may be specific, we do not know how sensitive they are. NTM pulmonary disease suspects were more likely than patients with no TB or NTM to have cough, other respiratory symptoms, and low BMI. Although we do not have enough clinical data to determine whether NTM in these patients represents the primary pathogen, a co-pathogen, or a commensal, we believe that a subset of these patients may have eventually progressed to disease (35). Specifically, NTM pulmonary disease suspects with isolation of MAC had lower CD4 count and lower BMI when compared with pulmonary disease suspects with other species of NTM.
No person with HIV who was on ART was found to have NTM disease. Although the design of this study does not allow us to determine whether or not ART was truly protective, this finding is consistent with data from previous studies in the United States, which has seen a dramatic decline in NTM bloodstream infections after the introduction of ART (36). In the present study, 4 of 10 patients with disseminated NTM disease had positive blood cultures: two with M. avium, one with M. intracellulare, and one with Mycobacterium simiae. The CD4 count in these patients was less than 100 cells/μl, which is similar to previous studies reporting the high frequency of disseminated MAC in HIV-infected individuals when the CD4 count is low (7, 37). Although prophylaxis with macrolide antibiotics can help reduce the incidence of disseminated MAC in this group, ART is considered the most effective prophylaxis, consistent with the observations in our study (38, 39). The remaining six (60%) patients categorized with NTM disseminated disease had positive urine cultures. One patient in this group had a CD4 count of 112 cells/μl; the CD4 count of the remaining five patients was greater than 200 cells/μl.
We found the isolation of NTM to be significantly higher in Thailand than in Vietnam, despite using identical methods. We analyzed the media and reagents used in the Thailand laboratory and did not find a laboratory source of contamination. Mycobacterium chelonae/abscessus was isolated however, from eight water samples collected at different intervals from the specimen collection site in Thailand. M. abscessus was isolated from 27 of 187 patients in Thailand with at least one culture positive for NTM; one patient with this species met our criteria for having pulmonary disease, and one patient was classified as having disseminated disease. Although we did not perform molecular studies to determine whether the patient and environmental isolates were genetically related, we believe the recovery of M. abscessus from patients in Thailand is likely related to contaminated municipal water supplies. Given the standardization of collection procedures across all sites, including mouth rinsing prior to expectoration, we suspect the increased recovery of M. abscessus and other NTM species reflects a difference in geographic distribution or exposure to NTM in Thailand.
Our study is subject to important limitations. First, we evaluated patients at one point in time; serial clinical, radiographic, and microbiological monitoring may have helped elucidate whether those classified as NTM pulmonary disease suspects had or would have progressed to disease. Second, our study design limits our ability to conclude that ART is protective against NTM disease in this population, as patients were not randomly assigned to receive or not receive ART. However, the association that we found is consistent with data from other settings that demonstrate a protective effect of ART.
Although the overall number of patients with at least one NTM isolated in culture was similar to the total number of patients confirmed with TB, the prevalence of NTM pulmonary and disseminated disease among people living with HIV in Southeast Asia is low. Previous studies have suggested that the low prevalence of NTM disease observed in resource-limited settings may be related to broad mycobacterial immunity from prior infection with TB, bacillus Calmette-Guérin, or exposure to environmental mycobacteria (14, 15). Although access to ART may also reduce the incidence of NTM infections in these settings, the identification of NTM infections may, paradoxically, increase as diagnostic capacity to detect and identify these organisms improves. Research in settings with limited resources and/or high rates of TB and HIV is needed to develop and validate NTM case definitions that can be used for further epidemiologic studies and clinical care of people living with HIV. Such studies should include more extensive longitudinal clinical evaluations, perhaps including CT scan of the lungs, serial sputum cultures, and assessment of response to treatment.
The authors thank the following members of the study team for their contributions to patient care, data collection, and laboratory testing: From Cambodia: Dr. Mao Tan Eang, Dr. Mean Chi Vun, Dr. Chheng Phalkun, Dr. Sopheak Thai, Dr. Pe Reaksmey, Mr. Sophanna Song, Ms. Poda Sar, Mr. Sambo Boy, Dr. Chhum Vannarith and staff of the Banteay Meanchey Provincial Health Department, Dr. Ngek Bunchhup and staff of the Battambang Provincial Health Department, Sereysophon Referral Hospital, Mongkul Borey Referral Hospital, Battambang Referral Hospital, and Sihanouk Hospital Center of Hope. From Thailand: Dr. Praphan Phanuphak, Dr. Tippawan Pankam, Dr. Nittaya Phanuphak, Dr. Channawaong Burpat, Ms. Apiratee Kanphukiew, Ms. Chalinthorn Sinthuwattanawibool, and the staff of the Bangkok Metropolitan Authority mycobacteriology laboratory, Chulalongkorn University mycobacteriology laboratory, and Chest Institute mycobacteriology laboratory. Vietnam: Dr. Thai Le, Dr. Trinh Thanh Thuy, Dr. Hoang Thi Quy, Dr. Nguyen Huu Minh, Dr. Nguyen Hong Duc, Dr. Nguyen Tuan Tai, Mrs. Le Thi Ngoc Bich and staff of the Pham Ngoc Thach HIV outpatient clinic, Mrs. Dai Viet Hoa and staff of the Pham Ngoc Thach microbiology laboratory, Dr. Le Truuong Giang and staff of the People's AIDS Committee of Ho Chi Minh City, District 1 and District 2 HIV outpatient clinics.
| 1. | Falkinham JO. The changing pattern of nontuberculous mycobacterial disease. Can J Infect Dis Med Microbiol 2003;14:281–286. |
| 2. | Collins FM. Mycobacterial disease, immunosuppression, and acquired immunodeficiency syndrome. Clin Microbiol Rev 1989;2:360–377. |
| 3. | Young LS. Mycobacterial diseases and the compromised host. Clin Infect Dis 1993;17:S436–S441. |
| 4. | Glassroth J. Pulmonary disease due to nontuberculous mycobacteria. Chest 2008;133:243–251. |
| 5. | Somoskovi A, Mester J, Hale YM, Parsons LM, Salfinger M. Laboratory diagnosis of nontuberculous mycobacteria. Clin Chest Med 2002;23:585–597. |
| 6. | Porter JD. Mycobacteriosis and HIV infection: the new public health challenge. J Antimicrob Chemother 1996;37 Suppl B:113–120. |
| 7. | Nightingale SD, Byrd. LT, Southern PM, Jockusch JD, Cal SX, Wynne BA. Incidence of Mycobacterium avium-intracellulare complex bacteremia in human immunodeficiency virus-positive patients. J Infect Dis 1992;165:1082–1085. |
| 8. | Horsburgh CR, Selik RM. The epidemiology of disseminated nontuberculous mycobacterial infection in the acquired immunodeficiency syndrome (AIDS). Am Rev Respir Dis 1989;139:4–7. |
| 9. | Zumla AI, Grange J. Non-tuberculous mycobacterial pulmonary infections. Clin Chest Med 2002;23:369–376. |
| 10. | Hatherill M, Hawkridge T, Whitelaw A, Tameris M, Mahomed H, Moyo S, Hanekom W, Hussey G. Isolation of non-tuberculous mycobacteria in children investigated for pulmonary tuberculosis. PLoS ONE 2006;1:e21. |
| 11. | Buijtels PC, van-der-Sande MA, de-Graaff CS, Parkinson S, Verbrugh HA, Petit PL, van-Soolingen D. Nontuberculous mycobacteria, Zambia. Emerg Infect Dis 2009;15:242–249. |
| 12. | Corbett EL, Blumberg L, Churchyard GJ, Moloi N, Mallory K, Clayton T, Williams BG, Chaisson RE, Hayes RJ, De Cock KM. Nontuberculous mycobacteria: defining disease in a prospective cohort of South African miners. Am J Respir Crit Care Med 1999;160:15–21. |
| 13. | Saritsiri S, Udomsantisook N, Suankratay C. Nontuberculous mycobacterial infections in King Chulalongkorn memorial hospital. J Med Assoc Thai 2006;89:2035–2046. |
| 14. | Morrissey AB, Aisu TO, Falkinham JO, Eriki PP, Ellner JJ, Daniel TM. Absence of Mycobacterium avium complex disease in patients with AIDS in Uganda. J Acquir Immune Defic Syndr 1992;5:477–478. |
| 15. | von Reyn CF, Arbeit RD, Tosteson AN, Ristola MA, Barber TW, Waddell R, Sox CH, Brindle RJ, Gilks CF, Ranki A, et al.. The international epidemiology of disseminated Mycobacterium avium complex infection in AIDS. International MAC study group. AIDS 1996;10:1025–1032. |
| 16. | Gilks CF, Brindle RJ, Mwachari C, Batchelor B, Bwayo J, Kimari J, Arbeit RD, von Reyn CF. Disseminated Mycobacterium avium infection among HIV-infected patients in Kenya. J Acquir Immune Defic Syndr Hum Retrovirol 1995;8:195–198. |
| 17. | Wallace RJ, Glassroth J, Griffith DE, Olivier KN, Cook JL, Gordin F. Supplement: American Thoracic Society diagnosis and treatment of disease caused by nontuberculous mycobacteria. Am J Respir Crit Care Med 1997;156:S1–S25. |
| 18. | Griffith DE, Aksamit T, Brown-Elliott BA, Catanzaro A, Daley C, Gordin F, Holland SM, Horsburgh R, Huitt G, Iademarco MF, et al.. An official ATS/IDSA statement: diagnosis, treatment, and prevention of nontuberculous mycobacterial diseases. Am J Respir Crit Care Med 2007;175:367–416. |
| 19. | Cain KP, McCarthy KD, Heilig CM, Monkongdee P, Tasaneeyapan T, Kanara N, Kimerling ME, Chheng P, Thai S, Sar B, et al.. An algorithm for tuberculosis screening and diagnosis in people with HIV. N Engl J Med 2010;362:707–716. |
| 20. | WHO. Global tuberculosis control—epidemiology, strategy, financing: WHO report 2009. Geneva, Switzerland: World Health Organization; 2009. Report No.:WHO/HTM/TB/2009.411. |
| 21. | Monkongdee P, McCarthy KD, Cain KP, Tasaneeyapan T, Dung NH, Lan NT, Yen NT, Teeratakulpisarn N, Udomsantisuk N, Heilig C, et al.. Yield of acid-fast smear and mycobacterial culture for tuberculosis diagnosis in people with HIV. Am J Respir Crit Care Med 2009;180:903–908. |
| 22. | Varma JK, McCarthy KD, Tasaneeyapan T, Monkongdee P, Kimerling ME, Buntheoun E, Sculier D, Keo C, Phanuphak P, Teeratakulpisarn N, et al.. Bloodstream infections among HIV-infected outpatients, Southeast Asia. Emerg Infect Dis 2010;16:1569–1575. |
| 23. | Kent P, Kubica G. Public health mycobacteriology: guide for the level III laboratory. Atlanta, GA: U.S. Department of Health and Human Services, Centers for Disease Control; 1985. |
| 24. | World Health Organization. Laboratory services in tuberculosis control. Culture part III. Geneva, Switzerland: World Health Organization; 1998. |
| 25. | Butler WR, Floyd MM, Silcox V, Cage G, Desmond E, Duffey PS, Guthertz LS, Gross WM, Jost KC, Ramos LS, et al.. Mycolic acid pattern standards for HPLC identification of mycobacteria. Atlanta, GA: Centers for Disease Control and Prevention; 1996. |
| 26. | Asian Development Bank. Practice guidelines for harmonizing HIV prevention initiatives in the infrastructure sector Greater Mekong subregion [Internet]. c2010 [accessed 2011 Sept]. Available from: http://www.adb.org/sites/default/files/pub/2010/harmonizing-hiv-prevention-gms.pdf |
| 27. | Cain KP, Anekthananon T, Burapat C, Akksilp S, Mankhatitham W, Srinak C, Nateniyom S, Sattayawuthipong W, Tasaneeyapan T, Varma JK. Causes of death in HIV-infected persons who have tuberculosis, Thailand. Emerg Infect Dis 2009;15:258–264. |
| 28. | Cain KP, Kanara N, Laserson KF, Vannarith C, Sameourn K, Samnang K, Qualls ML, Wells CD, Varma JK. The epidemiology of HIV-associated tuberculosis in rural Cambodia. Int J Tuberc Lung Dis 2007;11:1008–1013. |
| 29. | Quy HT, Cobelens FG, Lan NT, Buu TN, Lambregts CS, Borgdorff MW. Treatment outcomes by drug resistance and HIV status among tuberculosis patients in Ho Chi Minh City, Vietnam. Int J Tuberc Lung Dis 2006;10:45–51. |
| 30. | Daley P, Petrich A, May K, Luinstra K, Rutherford C, Chedore P, Jamieson F, Smieja M. Comparison of in-house and commercial 16s RNA sequencing with high-performance liquid chromatography and Genotype AS and CM for identification of nontuberculous mycobacteria. Diagn Microbiol Infect Dis 2008;61:284–293. |
| 31. | Ngamlert K, Sinthuwattanawibool C, McCarthy KD, Sohn H, Starks A, Kanjanamongkolsiri P, Anek-vorapong R, Tasaneeyapan T, Monkongdee P, Diem L, et al.. Diagnostic performance and costs of Capilia TB for Mycobacterium tuberculosis complex identification from broth-based culture in Bangkok, Thailand. Trop Med Int Health 2009;14:748–753. |
| 32. | Whitson BD, George RB. Fatal mycobacterial infection in a 32-year-old man with AIDS. Chest 1994;106:1577–1579. |
| 33. | Gadkowski LB, Stout JE. Cavitary pulmonary disease. Clin Microbiol Rev 2008;21:305–333. |
| 34. | Witzig RS, Fazal BA, Mera RM, Mushatt DM, Dejace PM, Greer DL, Hyslop NE. Clinical manifestations and implications of coinfection with Mycobacterium kansasii and human immunodeficiency virus type 1. Clin Infect Dis 1995;21:77–85. |
| 35. | Chin DP, Hopewell PC, Yajko DM, Vittinghoff E, Horsburgh CR, Hadley WK, Stone EN, Nassos PS, Ostroff SM, Jacobson MA, et al.. Mycobacterium avium complex in the respiratory or gastrointestinal tract and the risk of M. avium complex bacteremia in patients with human immunodeficiency virus infection. J Infect Dis 1994;169:289–295. |
| 36. | Kaplan JE, Hanson D, Dworkin MS, Frederick T, Bertolli J, Lindegren ML, Holmberg S, Jones JL. Epidemiology of human immunodeficiency virus-associated opportunistic infections in the United States in the era of highly active antiretroviral therapy. Clin Infect Dis 2000;30:S5–S14. |
| 37. | Horsburgh CR, Havlik JA, Ellis DA, Kennedy E, Fann SA, Dubois RE, Thompson SE. Survival of patients with acquired immune deficiency syndrome and disseminated Mycobacterium avium complex infection with and without antimycobacterial chemotherapy. Am Rev Respir Dis 1991;144:557–559. |
| 38. | Jones D, Havlir DV. Nontuberculous mycobacteria in the HIV infected patient. Clin Chest Med 2002;23:665–674. |
| 39. | Phillips P, Chan K, Hogg R, Bessuille E, Black W, Talbot J, O'Shaughnessy M, Montaner J. Azithromycin prophylaxis for Mycobacterium avium complex during the era of highly active antiretroviral therapy: evaluation of a provincial program. Clin Infect Dis 2002;34:371–378. |
Originally Published in Press as DOI: 10.1164/rccm.201107-1327OC on February 16, 2012
Author disclosures
