In 2009, there were an estimated 9.4 million new cases of tuberculosis (TB) reported globally with 13% of the cases occurring among HIV-infected persons.1 In the same year, South Africa reported an estimated 490,000 new TB cases of whom 49% knew their HIV status and 58% were HIV infected.2 The increased risk of TB associated with HIV occurs early in the course of infection and continues to increase as CD4+ T-cell counts decline over time.3–5
To reduce the burden of TB among HIV-infected persons, the World Health Organization has endorsed and promoted the 3I's strategy: intensified case finding for TB (ICF), isoniazid preventive therapy (IPT) and TB infection control (IC). ICF is defined as the systematic screening of HIV-infected patients for TB symptoms with prompt evaluation and investigation of identified TB suspects. ICF is recommended at HIV diagnosis, before initiating IPT or antiretroviral therapy (ART), and at every visit to a health facility.6,7 As the number of HIV-infected patients receiving ART continues to grow and ART programs are integrated with TB services, improved implementation of TB IC measures is essential to minimize TB transmission within HIV clinics. Globally, the implementation of the 3I's has been poor with only 5% of all HIV-infected patients screened for TB and <1% receiving IPT in 2009.1 Challenges in delivering the 3I's include staff shortages, lack of a sensitive point-of-care test for TB, uncertainty regarding the optimum method to screen for TB, and which subpopulation of HIV-infected persons to target for IPT in resource-limited settings.6
ART reduces the risk of incident TB by 70%–90%8–11 and mortality from HIV-associated TB by at least 50%,12 but the incidence of TB remains higher among HIV-infected patients on ART than among HIV-uninfected persons.13 The burden of undiagnosed TB among HIV clinic patients being evaluated before initiating ART in South Africa has ranged from 19% to 35%14–18 and has been associated with CD4+ T-cell counts ≤100 cells per microliter, body mass index (BMI) ≤18 kg/m2 and previous TB treatment.17–19 There is little reported data regarding undiagnosed TB among HIV-infected clinic patients stratified by receipt of ART and the consequent implications for ICF, IPT, and TB IC. The aim of this study was to describe factors associated with undiagnosed TB among HIV-infected adults attending a primary health clinic.
MATERIALS AND METHODS
Participants and Setting
Between October 2009 and May 2010, a convenience sample of HIV-infected adult patients waiting to receive care at an ambulatory HIV clinic housed within a primary health care centre located in an urban resource-limited setting in Gauteng Province, South Africa, were invited to participate in a study to determine the performance of urine lipoarabinomannan assay, the results of which have been previously reported.20 This secondary analysis examines determinants of undiagnosed TB in this population.
The HIV clinic offered cotrimoxazole preventive therapy and ART according to South African National Guidelines for initiating ART, which at that time were CD4+ T-cell counts <200 cells per microliter and/or World Health Organization Stage IV disease.21 TB diagnostic services were available at the clinic and the standard of care for diagnosis of TB in symptomatic HIV-infected patients (defined as those having 1 or more of cough >2 weeks, fever >2 weeks, night sweats >2 weeks, and weight loss >1.5 kg in the past month) was 2 sputum specimens (1 spot and 1 early morning) with a third specimen for culture if both sputum smears were negative. IPT was not standard of care at primary care clinics at the time the study was conducted. To be eligible for enrollment, participants had to be HIV-infected, ≥18 years of age, and live or work in the catchment area of the clinic. Participants were ineligible if they were on TB treatment, had completed TB treatment <3 months before enrolment, were acutely ill, institutionalized, or incarcerated.
Data Collection and Laboratory Investigations
A trained nurse administered a questionnaire, which collected sociodemographic data and information regarding the presence and duration of TB symptoms, history of prior or current TB treatment, IPT, and ART to all participants. All participants had weight and height measured to calculate the BMI. Participants were also asked to submit one spot sputum specimen for smear microscopy and mycobacterial culture. Participants who were unable to expectorate spontaneously were nebulized with hypertonic (0.9%) saline using a disposable nebulization masks for up to 15 minutes. Participants who were known to be asthmatic were not nebulized.
Participants who had enlarged and accessible lymph nodes (ie, cervical, supraclavicular, axillary, inguinal, and epitrochlear) underwent fine needle aspiration for microbiologic and histological examinations. All participants provided blood for mycobacterial culture and CD4+ T-cell count measurement. In addition, a urine specimen for Clearview enzyme-linked immunosorbent assay (urine lipoarabinomannan) test was collected for the parent study. Pregnancy was excluded using urine pregnancy tests, and nonpregnant participants were asked to have a chest radiograph.
Participants were considered to be on ART if they reported being on ART for more than a month before enrollment to allow time for some immune restoration to occur. Participants were considered to be TB suspects if they reported any of the following symptoms, irrespective of duration: cough, fever, night sweats, or unintentional weight loss. Participants were classified as having bacteriologically confirmed TB if they had a sputum or blood culture positive for Mycobacterium tuberculosis. Participants were classified as having probable TB if they had 1 positive sputum smear or compatible histological features on fine needle aspiration or compatible radiological features, in the absence of a positive sputum or blood culture. Chest radiographs were read by a radiologist using a standardized tool adapted from the chest radiograph reading and recording system for TB epidemiological surveys,22 assigning each radiograph a probability of having TB based on his overall impression.
Data were entered into a Microsoft Access 2007 database (Seattle, WA) and exported into STATA 11.0 (College Station, TX) for analysis. Characteristics of participants on ART and those not on ART were compared using the χ2 test for categorical data and the t test for continuous data. Characteristics of bacteriologically confirmed TB and probable TB were compared using Wilcoxon rank sum test for continuous data and χ2 test for categorical data (Fisher exact test where some observations were less than 5). Logistic regression was used to determine the factors associated with having undiagnosed prevalent TB at enrollment. Because of the small number of outcomes observed in the study, a maximum of three variables could be included in the multivariable analysis of factors associated with undiagnosed TB in the overall population. Variables with the smallest P values <0.05 in the univariable analysis were included in the multivariable analysis. The likelihood ratio tests were used to determine the presence of a linear trend between CD4+ T-cell categories ≤100 cells per microliter, 101–200 cells per microliter, and ≥200 cells per microliter and log odds of undiagnosed TB.
The study was approved by ethics committees of the University of Witwatersrand, the Johns Hopkins University and the London School of Hygiene and Tropical Medicine. The study was also approved by the Ekurhuleni Metropolitan Municipal Research Committee. Study procedures were explained to potential participants and individual written informed consent was obtained from willing and eligible participants. Participants who were unable to read or write gave witnessed verbal consent. The study was conducted according to the International Conference on Harmonization/Good Clinical Practice guidelines.
Screening and Characteristics of Participants Enrolled
Four hundred forty-three patients were screened for eligibility, of whom 21 were ineligible [HIV uninfected (8), being treated for TB (12), and completed TB treatment <3 months before screening (1)] (Fig. 1). Of the 422 participants enrolled, 143 (34%) were male, median age was 37 years [interquartile range (IQR): 31–44 years] and the median CD4+ T-cell count was 215 cells per microliter (IQR: 107–347 cells/μL). One-hundred ninety-six (46.5%) participants were on ART for a median duration of 8 months (IQR: 3.3-23.8 months). Participants who were on ART were more likely to be female, to have a prior history of TB, and to have higher BMI, hemoglobin, and CD4+ T-cell count measurements at enrollment (Table 1).
TB Suspects and Cases
Three hundred sixty-one participants (85.6%) were identified as TB suspects. There was no difference in the proportion of participants who were TB suspects among participants on ART compared with those not on ART (82.7% vs. 88%, χ2P = 0.116), although participants on ART were less likely to report unintentional weight loss when compared with those not on ART (50.1% vs. 70.2%, χ2P < 0.001). The proportion of participants who were TB suspects decreased with increasing CD4+ T-cell counts [91.8%, 88.9%, and 81.3% among those with CD4+ T-cell counts ≤100 cells/μL, 101–200 cells/μL, and >200 cells/μL, respectively, (χ2P = 0.027)].
Fifty participants met the criteria for TB. Twenty-seven TB cases were bacteriologically confirmed, of which 20 were sputum culture positive and 7 were blood culture positive. Twenty-three TB cases were defined as probable (21 diagnosed by chest radiograph alone, 1 by a positive sputum smear, and 1 by both fine needle aspiration and chest radiograph). Participants with probable TB were more likely to have higher CD4+ T-cell counts and pleural effusions on chest radiograph (P = 0.025) but were less likely to have cavitations on chest radiograph (P = 0.043) compared with those with bacteriologically confirmed TB as shown in Table 2.
The proportion of participants with all forms of undiagnosed TB (bacteriologically confirmed and probable TB) was 11.8% [95% confidence interval (CI): 8.9 to 15.3%]. There was also no difference in the proportion of participants with all forms of TB among those on ART compared with those not on ART [10.2% (20 of 196) vs. 13.3% (30 of 226) P = 0.332]. One or more TB symptoms were reported by 49 of the 50 participants (98%) with any form of TB.
The proportion of participants with undiagnosed bacteriologically confirmed TB was 6.4% [95% CI: 4.3 to 9.2%]. All participants with bacteriologically confirmed TB had at least 1 symptom at enrollment. The proportion of participants with bacteriologically confirmed TB was lower among those on ART compared with those who were not [4.1% (8 of 196) vs. 8.4% (19 of 226)] although the differences did not reach statistical significance [odds ratio (OR) = 0.46 (95% CI: 0.20 to 1.08), P value = 0.076]. There was no difference in the proportion of TB cases who were sputum smear or culture positive, and thus potentially infectious, between those on ART compared with those who were not on ART [45% (9 of 20) vs. 50% (15 of 30), χ2P = 0.729].
Factors Associated With Undiagnosed TB
In univariable analyses (Table 3), bacteriologically confirmed TB was associated with low CD4+ T-cell counts [OR: 7.33 (95% CI: 2.56 to 20.99) among participants with CD4+ T-cell counts ≤100 cells/μL and OR: 3.87 (95% CI: 1.23 to 12.14) among participants with CD4+ T-cell counts of 100–200 cells/μL compared with those with CD4+ T-cell counts >200 cells/μL], low BMI [OR: 3.56 (95% CI: 1.57 to 8.10) for BMI <18.5kg/m2 compared with higher], and low hemoglobin levels [OR: 4.25 (95% CI: 1.87 to 9.64) for hemoglobin <10 g/dL]. These associations were also observed for all forms of TB, although the association with CD4+ T-cell counts was weaker (Table 3).
In multivariable models that adjusted for the effect of CD4+ T-cell counts at enrollment, low Hb and BMI; participants with CD4+ T-cell counts ≤100 cells/μL had increased odds of having bacteriologically confirmed TB at enrolment when compared with those with CD4+ T-cell counts >200 cells/μL [OR: 5.05 (95% CI: 1.69 to 15.12)]. These participants also had increased odds of having all forms of TB [OR: 2.35 (95% 1.07 to 5.17)] compared with those with CD4+ T-cell counts >200 cells/μL. Participants with CD4+ T-cell counts of 101–200 cells per microliter had increased odds of having any bacteriologically confirmed TB and all of forms of TB compared with those with CD4+ T-cell counts >200 cells/μL [OR: 2.24 (95% CI: 0.65 to 7.74) and OR: 2.23 (95% CI: 1.01 to 4.96), respectively]. There was an association between having low Hb and having undiagnosed TB [OR: 3.12 (95% CI: 1.26 to 7.72) for bacteriologically confirmed TB and OR: 2.25 (95% CI: 1.08 to 4.69) for all form of TB]. Low BMI was associated with increased odds of having any form of TB [OR: 2.70 (95% CI: 1.39 to 5.26)], an association which was weaker for bacteriologically confirmed TB [OR: 2.02 (95% CI: 0.83 to 4.90), P = 0.123] (Table 4).
Eligibility for IPT
Participants who were not TB suspects at enrollment would have been eligible for IPT provided there were no contraindications to isoniazid. Sixty-one participants (14.4%) did not report any of the symptoms in the screen and were by definition not TB suspects. These participants would have been immediately eligible for IPT and an additional 312 (73.9%) would have been eligible after active TB had been excluded. The proportion of participants who were immediately eligible for IPT was highest among participants with CD4+ T-cell counts >200 cells/μL [18.7% (42 of 225), 11.1% (11 of 99), and 8.2% (8 of 98) among participants with CD4+ + T-cell counts >200, 101–200, and ≤100 cells/μL, respectively, (P = 0.001)] and was slightly higher among those receiving ART compared with those not receiving ART [17.4% (27 of 226) vs. 12% (34 of 196), P = 0.057]. One of 61 asymptomatic participants (1.6%) who would have been immediately eligible for IPT turned out to have TB at enrollment.
TB symptoms and previously undiagnosed TB were common in this group of ambulatory HIV clinic patients. CD4+ T-cell count was the dominant factor associated with having undiagnosed TB, regardless of ART status.
ART use is associated with decreased risk of incident TB.13,23 Increased use of ART in a Cape Town community was associated with a reduction in the prevalence of undiagnosed TB among HIV-infected persons from 5.2% to 1.3% in community-based surveys conducted 5 years apart.24 A Tanzanian study conducted in settings similar to ours found that 8.5% of HIV-infected patients in regular care and 6.7% of those on ART for an average 6.8 months25 had undiagnosed TB. These rates are comparable to our findings. A number of factors may account for the high burden of undiagnosed TB among participants receiving ART in our study. It is possible that some patients were not appropriately screened for TB before ART initiation or during routine visits and that some TB cases were missed during this screening process. Prior studies have found that as many as half of the TB cases occurring in the first year after initiating ART occur in the first 3–6 months, and most are prevalent cases that could have been detected before ART initiation through ICF.17,26,27 Undiagnosed TB during the first 6 months of ART may also be due to reactivation of latent TB infection or unmasking of TB as part of an immune reconstitution syndrome,28 whereas TB among patients on ART for more than 6 months could be due to reinfection and rapid progression. Because the study sample was a convenience rather than random sample, we may have overestimated the burden of undiagnosed TB through self-selection by participants who were symptomatic and more likely to have TB. However, because the study was done in a routine clinical setting, the high burden among this convenience sample of patients waiting to access care is worrying and raises concerns about TB screening and IC in HIV clinics.
In our study, a high proportion of participants on ART were female. High levels of female unemployment in excess of 60%29 resulting in more women available to attend clinics during weekdays and higher numbers of women in ART programs compared with men, most likely as a result of better retention in care30 may account for this disparity.
A very high proportion of participants (85.6%) were symptomatic. In settings where large proportions of patients are symptomatic, TB symptom screening alone may not be useful. Algorithms which may include BMI, Hb levels, and CD4+ T-cell count testing to better target individuals more likely to have undiagnosed TB are needed. The performance and cost effectiveness of such algorithms should be investigated. The yield from blood mycobacterial cultures was low (7 of 422). This was not surprising for this ambulant population. Blood culture would be more useful in very ill patients suspected of having extrapulmonary TB. The use of fine needle aspiration in this primary care setting did not detect any TB cases additional to those detected with blood/ sputum mycobacterial culture or chest radiographs. However, the use of fine needle aspiration in this population may be useful in identifying comorbid conditions such as lymphomas.
Using the currently recommended 4 symptom screening tool alone, only 14.4% of participants would have been eligible for IPT without further evaluation, with an additional 73.8% eligible for IPT after sputum smear microscopy, sputum and/or blood mycobacterial culture, fine needle aspiration from enlarged lymph nodes, and chest radiograph. The higher proportion of participants eligible for IPT when screened at higher CD4+ T-cell counts may suggest that assessment for IPT eligibility is more easily done at higher CD4+ T-cell counts. However, patients with lower CD4+ T-cell counts, the group in which excluding TB is most difficult, stand to benefit the most from IPT and from having active TB diagnosed and treated in the process of assessing for eligibility for IPT. There may be a case for offering IPT to patients with higher CD4+ T-cell counts, among whom the prevalence of undiagnosed TB is lower, on the basis of the symptom screen alone, and to patients with lower CD4+ T-cell counts after symptom screen and further investigations to minimize the number of TB suspects who require additional investigations while maximizing the potential benefit of IPT.
Almost half of all the TB cases (48%) in our study were sputum smear or culture positive (Fig. 1), indicating that there was a risk of nosocomial transmission of TB in the clinic. We cannot overstate the need to strengthen implementation of administrative and environmental IC measures. HIV clinic attendees should be screened for a cough as they enter health facilities, and those with a cough triaged and fast tracked through the clinics to minimize time spent potentially infecting other patients and staff. Approximately, 35% of participants who were on ART had a prior history of TB. For this group of patients, the opportunity to use ART as a tool to prevent the first episode of TB had been missed, although recurrence of TB could be prevented with secondary isoniazid prophylaxis. Earlier initiation of ART is encouraged to prevent TB and reduce mortality among those who do develop TB.
1. World Health Organization. Global Tuberculosis Control Report. Geneva, Switzerland: World Health Organization; 2010.
3. van Asten L, Langendam M, Zangerle R, et al.. Tuberculosis risk varies with the duration of HIV infection: a prospective study of European drug users with known date of HIV seroconversion. AIDS. 2003;17:1201–1208.
4. Sonnenberg P, Glynn JR, Fielding K, et al.. How soon after infection with HIV does the risk of tuberculosis start to increase? A retrospective cohort study in South African gold miners. J Infect Dis. 2005;191:150–158.
5. Glynn JR, Murray J, Bester A, et al.. Effects of duration of HIV infection and secondary tuberculosis transmission on tuberculosis incidence in the South African gold mines. AIDS. 2008;22:1859–1867.
6. World Health Organization. Priority Research Questions for Tuberculosis/Human Immunodeficiency Virus (TB/HIV) in HIV-Prevalent and Resource-Limited Settings. Geneva, Switzerland: World Health Organization; 2010.
7. World Health Organization. Guidelines for Intensified Tuberculosis Case-Finding and Isoniazid Preventive Therapy for People Living with HIV in Resource Constrained Settings. Geneva, Switzerland: World Health Organization;2011.
8. Golub JE, Pronyk P, Mohapi L, et al.. Isoniazid preventive therapy, HAART and tuberculosis risk in HIV-infected adults in South Africa: a prospective cohort. AIDS. 2009;23:631–636.
9. Golub JE, Saraceni V, Cavalcante SC, et al.. The impact of antiretroviral therapy and isoniazid preventive therapy on tuberculosis incidence in HIV-infected patients in Rio de Janeiro, Brazil. AIDS. 2007;21:1441–1448.
10. Jones JL, Hanson DL, Dworkin MS, et al.. HIV-associated tuberculosis in the era of highly active antiretroviral therapy. The Adult/Adolescent Spectrum of HIV Disease Group. Int J Tuberc Lung Dis. 2000;4:1026–1031.
11. Girardi E, Antonucci G, Vanacore P, et al.. Impact of combination antiretroviral therapy on the risk of tuberculosis among persons with HIV infection. AIDS. 2000;14:1985–1991.
12. Dheda K, Lampe FC, Johnson MA, et al.. Outcome of HIV-associated tuberculosis in the era of highly active antiretroviral therapy. J Infect Dis. 2004;190:1670–1676.
13. Lawn SD, Harries AD, Williams BG, et al.. Antiretroviral therapy and the control of HIV-associated tuberculosis. Will ART do it? Int J Tuberc Lung Dis. 2011;15:571–581.
14. Houlihan CF, Mutevedzi PC, Lessells RJ, et al.. The tuberculosis challenge in a rural South African HIV programme. BMC Infect Dis. 2010;10:23.
15. Lawn SD, Myer L, Edwards D, et al.. Short-term and long-term risk of tuberculosis associated with CD4 cell recovery during antiretroviral therapy in South Africa. AIDS. 2009;23:1717–1725.
16. Bassett IV, Wang B, Chetty S, et al.. Intensive tuberculosis screening for HIV-infected patients starting antiretroviral therapy in Durban, South Africa. Clin Infect Dis. 2010;51:823–829.
17. Lawn SD, Kranzer K, Edwards DJ, et al.. Tuberculosis during the first year of antiretroviral therapy in a South African cohort using an intensive pretreatment screening strategy. AIDS. 2010;24:1323–1328.
18. Lawn SD, Myer L, Bekker LG, et al.. Burden of tuberculosis in an antiretroviral treatment programme in sub-Saharan Africa: impact on treatment outcomes and implications for tuberculosis control. AIDS. 2006;20:1605–1612.
19. Moore D, Liechty C, Ekwaru P, et al.. Prevalence, incidence and mortality associated with tuberculosis in HIV-infected patients initiating antiretroviral therapy in rural Uganda. AIDS. 2007;21:713–719.
20. Gounder CR, Kufa T, Wada NI, et al.. Diagnostic accuracy of a urine lipoarabinomannan enzyme-linked immunosorbent assay for screening ambulatory HIV-infected persons for tuberculosis. J Acquir Immune Defic Syndr. 2011;58:219–223.
21. National Department of Health South Africa. National Antiretroviral Treatment Guidelines. Cape Town, South Africa: National Department of Health South Africa; 2004.
22. Den Boon S, Bateman ED, Enarson DA, et al.. Development and evaluation of a new chest radiograph reading and recording system for epidemiological surveys of tuberculosis and lung disease. Int J Tuberc Lung Dis. 2005;9:1088–1096.
23. Lawn SD, Kranzer K, Wood R. Antiretroviral therapy for control of the HIV-associated tuberculosis epidemic in resource-limited settings. Clin Chest Med. 2009;30:685–699; viii.
24. Middelkoop K, Bekker LG, Myer L, et al.. Antiretroviral program associated with reduction in untreated prevalent tuberculosis in a South African township. Am J Respir Crit Care Med. 2010;182:1080–1085.
25. Ngowi BJ, Mfinanga SG, Bruun JN, et al.. Pulmonary tuberculosis among people living with HIV/AIDS attending care and treatment in rural northern Tanzania. BMC Public Health. 2008;8:341.
26. Rajasekaran S, Raja K, Jeyaseelan L, et al.. Post-HAART tuberculosis in adults and adolescents with HIV in India: incidence, clinical and immunological profile. Indian J Tuberc. 2009;56:69–76.
27. Grant AD, Bansi L, Ainsworth J, et al.. Tuberculosis among people with HIV infection in the United Kingdom: opportunities for prevention? AIDS. 2009;23:2507–2515.
28. Manabe YC, Breen R, Perti T, et al.. Unmasked tuberculosis and tuberculosis immune reconstitution inflammatory disease: a disease spectrum after initiation of antiretroviral therapy. J Infect Dis. 2009;199:437–444.
29. Statistics South Africa. Republic of South Africa. Millenium Development Goals, Country Report 2010. Pretoria, South Africa: Statistics South Africa; 2010.
30. Fox MP, Rosen S. Patient retention in antiretroviral therapy programs up to three years on treatment in sub-Saharan Africa, 2007–2009: systematic review. Trop Med Int Health. 2010;15(suppl 1):1–15.
Keywords:© 2012 Lippincott Williams & Wilkins, Inc.
tuberculosis; HIV; antiretroviral therapy; screening; case finding; epidemiology