Abstract
Background
The World Health Organization strongly recommends regular screening for tuberculosis (TB) in HIV-positive individuals.
Objective
To compare the outcome of anti-tuberculosis treatment (ATT) in HIV-positive adults diagnosed with TB through active case-finding (ACF) or passive case-finding (PCF).
Design
Antiretroviral therapy (ART)-naïve adults diagnosed with TB were included from two prospective cohort studies conducted in Ethiopia between September 2010 and March 2013. The PCF cohort was based at out-patient TB clinics, whereas participants in the ACF cohort were actively screened for TB by bacteriological sputum testing (smear microscopy, Xpert MTB/RIF assay, and liquid culture) without pre-selection on the basis of symptoms and signs. Outcomes of ATT were compared between participants in the two cohorts; characteristics at diagnosis and predictors of adverse outcomes were analysed.
Results
Among 439 TB/HIV co-infected participants, 307 and 132 belonged to PCF and ACF cohorts, respectively. Compared with the ACF participants, hemoptysis, conjunctival pallor, bedridden status, and low mid upper-arm circumference (MUAC) were significantly more common in participants identified through PCF. Sputum smear-positivity rates among pulmonary TB cases were 44.2% and 21.1% in the PCF and ACF cohorts, respectively (p<0.001). Treatment success was ascertained in 247 (80.5%) of the participants in the PCF cohort and 102 (77.2%) of the participants in the ACF cohorts (p=0.223). Low MUAC (p=0.001) independently predicted mortality in the participants in both cohorts.
Conclusion
Although patients identified through ACF had less advanced TB disease, ATT outcome was similar to the patients identified through PCF. To achieve a better outcome, case management in ACF strategy should be strengthened through enhanced patient-centred counselling and adherence support.
Co-infection with tuberculosis (TB) and HIV is common in low-income countries. Collaborative TB/HIV activities are needed to reduce the burden of TB in people living with HIV (PLHIV) (Citation1). For individuals with TB/HIV co-infection, there are two routes of entry into care: via either TB or HIV clinics. Whereas TB clinics detect TB among patients who seek medical care with symptoms (passive case-finding; PCF), active case-finding (ACF) entails screening for active TB for certain population groups regardless of symptoms or clinical suspicion. With PCF strategy, a substantial proportion of active TB patients might die before seeking care or be missed even after they reach health facilities (Citation2). Many TB patients also become contagious before diagnosis. Conversely, ACF could detect TB cases early and is consequently recommended by WHO for PLHIV (Citation3, Citation4).
Although HIV testing has been efficiently implemented in TB clinics, TB case-finding among PLHIV is suboptimal. Most persons who receive their HIV diagnosis when presenting with TB have advanced disease at presentation, with a high risk of death (Citation5). A challenge for TB/HIV integration in high-burden countries is the lack of effective diagnostic algorithms and tools for TB detection in PLHIV (Citation5–Citation7).
Sputum smear microscopy detects the most contagious cases; however, a majority of HIV-associated TB is missed by this method. The WHO TB symptom screening algorithm can be used to identify subsets of PLHIV in need of further TB investigations (Citation8). Still, in routine care 20–28% of active TB patients were missed at initiation of ART (Citation9, Citation10). Therefore, TB screening using more sensitive diagnostic tests prior to ART initiation has been recommended (Citation11–Citation13).
A systematic review of intensified TB screening among PLHIV in low-resource, high-burden countries has reported high rates of TB detection (Citation14). Also, an ACF study conducted in Ethiopia at ART clinics at health centres found 17.9% previously undetected TB cases (Citation15).
Early identification of TB in PLHIV would be expected to lead to improved survival, and reduced risk of TB transmission (Citation16). Although facility-based ACF for TB among PLHIV detects early cases and may consequently improve survival (Citation17–Citation19), empirical evidence on the benefit of ACF for treatment outcome is scarce. A recent systematic review has shown similar treatment outcome between cases identified through PCF and ACF strategies (Citation20). In this study, we hypothesised that early identification of TB in PLHIV through ACF would lead to lower rates of adverse treatment outcomes. We compared anti-TB treatment (ATT) outcome among HIV-positive adults receiving TB care at out-patient TB clinics (PCF cohort) to that of PLHIV diagnosed with TB through intensified case-finding (ACF cohort) at HIV clinics.
Design
Characteristics of study participants
Participants were identified from two cohort studies (Citation15, Citation21) conducted in public health facilities providing integrated care for TB and HIV in Oromia region, Ethiopia.
PCF cohort
Between September 2010 and September 2012, we recruited participants consecutively at eight out-patient TB clinics (two hospitals and six health centres; n=1,116). These patients had been diagnosed with TB at in- or out-patient clinics at these health facilities, or at private clinics after self-presentation, and were referred for ATT to the study TB clinics. Diagnosis of TB was based on sputum smear microscopy, clinical criteria, and chest radiography in accordance with Ethiopian National Guidelines (Citation22). The following inclusion criteria were applied: age 18 years or greater, residence in the clinic uptake area, written informed consent, and consent to HIV testing. We excluded persons who had received ATT for more than 2 weeks or had been treated for TB within the preceding 6 months, and all ART experienced patients.
At baseline, trained nurses evaluated participants using a structured questionnaire with details on socio-demographic characteristics, self-reported symptoms, physical findings, and basic laboratory tests including complete blood count (CBC) and CD4 cell count. Detailed description of the cohort was presented recently (Citation21). For the current study, we included 307 TB/HIV co-infected participants.
ACF cohort
We recruited HIV-positive participants in a prospective cohort between October 2011 and March 2013 at five health centre HIV clinics. At baseline, participants underwent testing for TB irrespective of symptoms. Consecutive individuals (both in HIV care and new cases) were assessed for eligibility and included individuals (n=812) were screened for TB. The ACF cohort had the following inclusion criteria: provision of written informed consent, age 18 years or older, CD4 count ≤350 cells/mm3 and/or WHO clinical stage 4, and ART-naïve. Trained nurses collected information on social, demographic, and clinical characteristics, and blood for CBC and CD4 cell count was obtained. The participants submitted two spontaneously expectorated morning sputum samples for bacteriological testing (including liquid culture, Xpert MTB/RIF, and smear microscopy; reported previously) (Citation15). From participants with peripheral lymphadenopathy, fine-needle aspirates were obtained for culture and Xpert MTB/RIF assay. A total of 132 TB/HIV co-infected participants were included in the ACF cohort.
Definitions and determination of ATT outcome
A TB case was defined as a patient with a positive bacteriological test result for TB and/or who fulfilled Ethiopian national criteria for clinical TB (Citation22), and who was prescribed ATT using WHO's Directly Observed Treatment Short course (DOTS) strategy.
Prospective evaluations were performed 2 and 6 or 8 months after enrolment for all participants. Treatment outcome was assessed at the end of therapy (6 or 8 months). Treatment success was defined as cure and/or treatment completion (Citation23). Loss to follow-up, treatment failure, and death were categorised as adverse outcome. Participants transferred to other health facilities for care could not be evaluated for treatment outcome and were consequently categorised as having neither treatment success nor adverse treatment outcome.
Statistical analysis
Anonymised data were entered into Excel files continuously and crosschecked with original hard copies. Data analysis was performed using IBM SPSS 20.0 (SPSS Inc., Chicago, IL, USA). Using Pearson's Chi-Square test, categorical data on social, demographic, clinical, and laboratory information were compared between the two groups. Comparisons of medians for scale variables without normal distribution were performed using non-parametric Mann-Whitney's U-test. Similar tests were performed to compare outcomes of ATT between participants in the two cohorts. The level of statistical significance was set at p<0.05.
Ethical considerations
Both studies were approved by the National Research Ethics Review Committee at the Ministry of Science and Technology of Ethiopia and the Regional Ethical Review Board at Lund University, Sweden. Written informed consent was obtained from all participants; in the presence of a witness in the case of illiterate participants.
Results
Baseline characteristics of the study participants
Characteristics of the participants are shown in . Among the 307 PCF participants, the median age was 32 years, 151 (49.2%) were female, and the median CD4 cell count was 173 cells/mm3. Among the 132 ACF participants, the median age was 35 years, 64 (48.5%) were female, and the median CD4 cell count was 169 cells/mm3. Rural residence was more common in the ACF cohort (21.7% vs 10.5%; p=0.03).
Table 1 Comparison of baseline characteristics, signs, and symptoms stratified by tuberculosis case-finding
A total of 124 (94%) study participants in the ACF cohort were culture and/or Xpert MTB/RIF positive, whereas 8 (6%) were clinically diagnosed. In the PCF cohort, 208 (67.8%) were clinically diagnosed TB cases (p<0.001). In the PCF and ACF cohorts, 7 (2.3%) and 5 (3.8%), respectively, had a previous episode of TB.
The smear-positivity rate among pulmonary TB cases was higher in the PCF cohort than in the ACF cohort (44.2% vs 21.1%; p<0.001). The distribution of symptoms and physical findings showed some differences between participants in the two cohorts. Specifically, fever (82.4% vs 65.6%; p<0.001) and night sweats (83.4% vs 72.0%; p<0.01) were significantly more common among patients recruited through PCF than in the ACF cohort. Similarly, markers of disease severity like blood stained sputum (15.4% vs 4.5%; p=0.001), conjunctival pallor (38.4% vs 26.0%; p=0.012), bedridden state (26.4% vs 9.9%; p<0.001), and lower MUAC (p=0.033) were more common among participants in the PCF cohort than in the ACF cohort ().
ART initiation
A total of 236 (68.4%) and 75 (52.4%) participants in the PCF and ACF cohorts started ART during the course of ATT, respectively. Additionally, 34 (23.4%) participants in the ACF cohort started ART before ATT initiation. No PCF participant started ART before ATT.
Outcomes of ATT
In the PCF cohort, 247 (80.5%) participants had treatment success, whereas 37 (12%) had adverse outcomes [20 (6.5%) were lost to follow-up and 17 (5.5%) died]. Treatment outcome in the PCF cohort could not be evaluated in 23 (7.5%) participants as they were transferred to other health facilities. Likewise, 102 (77.2%) participants in the ACF cohort had treatment success and 22 (16.7%) had adverse treatment outcomes [12 (9.1%) were lost to follow-up and 10 (7.6%) died]. Eight (6.1%) participants in the ACF were transferred to other health facilities (). Of seven previously treated participants in the PCF cohort, four had treatment success, whereas two participants died and one was lost to follow-up. Among five previously treated ACF participants, four had treatment success and one participant was transferred out.
Table 2 Comparison of treatment outcomes between TB cases detected through active and passive case-finding
A sub-analysis of participants with adverse outcomes in the ACF cohort showed that 10/22 (45.5%) TB patients never started ATT (all culture-positive). When these participants were tentatively included in the analysis of treatment outcome, there was no significant difference between the participants in the two cohorts (p=0.223). An analysis restricted to participants in the ACF cohort that initiated ART also showed no significant difference in ATT outcome with regard to the sequence of TB and HIV treatment initiation.
Factors associated with adverse treatment outcomes
Adverse treatment outcomes in TB cases detected through PCF were associated with lower body mass index (BMI) (median 16.9 vs 17.6 kg/m2; p=0.015) and lower MUAC (median 19 vs 21 cm; p=0.001). Median baseline CD4 cell count ≤100 cells/mm3 (p=0.023), lower BMI (median 16.3 vs 17.8 kg/m2; p=0.036) and lower MUAC (20 vs 22 cm; p=0.042) were associated with adverse treatment outcomes among subjects in the ACF cohort ().
Table 3 Characteristics of TB cases detected through active and passive case-finding stratified by treatment outcomes
Discussion
We compared outcomes of ATT among HIV-positive ART-naïve adults treated at TB clinics in Ethiopia with regard to entry into care – either through active TB case-finding in HIV clinics, or passively referred for ATT. Although participants diagnosed through ACF had characteristics of less advanced TB, the outcome of ATT was similar to that of passively detected cases. This underscores both the importance and the challenge of ACF particularly for achieving optimal treatment success.
Evaluations of ACF are inherently difficult to perform, since randomisation of participants into ACF or PCF strategy is impossible. A feasibility study in South Africa showed high treatment success rate among TB patients identified through ACF using a mobile HIV/TB clinic (Citation18). Nevertheless, another comparative study conducted in a similar setting and a recent systematic review failed to show improvement in outcome of ACF strategy (Citation20, Citation24).
The rationale for ACF for TB among PLHIV in endemic regions is obvious due to the high prevalence of active TB and the severe consequences of unrecognised TB (Citation17, Citation18). In this study, we did not find a difference in ATT outcomes between participants in the two cohorts. Our data indicate the importance of linking patients to care after submission of samples for testing. Although we did not measure adherence, other studies have suggested worse ATT adherence among patients diagnosed through ACF (Citation14). In fact, the ACF cohort comprised more patients with rural residence, a factor that has been associated with lower adherence in other studies (Citation25).
In the ACF cohort, a set of bacteriological methods in HIV-positive individuals without pre-selection on the basis of symptoms identified TB in 132 of 812 (16.3%) participants, confirming a high yield of ACF among PLHIV (Citation15). Ten (7.6%) of the TB patients in our ACF cohort never started ATT. Failure to start ATT was largely due to early loss to follow-up or death which occurred between submissions of samples and result delivery. The higher rate of early loss to follow-up in the ACF cohort was in concordance with the findings from community-based ACF in Cambodia (Citation26). Consistent with these findings, it is possible that rates of loss to follow-up are higher in patients recruited through ACF (Citation14, Citation19). On the contrary, TB patients identified through PCF strategy were immediately linked to treatment. It is also likely that a large proportion of TB/HIV co-infected persons in the community are not recognised as TB cases; hence, our PCF cohort only included a subset of the real population of such individuals.
Less advanced disease characteristics were recorded among TB cases detected actively compared with PCF cases. Biometric measurements indicating disease severity, like BMI and MUAC, were lower in the PCF study participants. This finding is consistent with a similar report from South Africa, showing that ACF identified subjects with minor or early symptoms and signs of TB (Citation24). Further, 78.9% of pulmonary TB cases in the ACF cohort were smear-negative. It is likely that some of those cases would have progressed to smear-positivity; which suggests that ACF could help reducing the burden of contagious TB in the community (Citation27). The advanced disease characteristics in the PCF study participants also implies that TB diagnosis in this subgroup of patients generally occurs later, compared with those who enter into care via HIV clinics (Citation28).
The WHO 3Is policy (Intensified TB case-finding; Isoniazid preventive therapy; and Improved TB infection control) has been vital in identifying a substantial proportion of TB cases among PLHIV (Citation29); yet its implementation has not been uniform (Citation6, Citation7). A report from South Africa showed that 87% of TB cases occurring during the first year on ART could have been detected at baseline using sputum culture (Citation16).
In the ACF cohort, 75 (52.4%) of the TB patients initiated ART during the course of TB treatment. Of particular note, 34 (23.4%) started ART prior to TB treatment in the ACF cohort, mainly due to onerous culture test result delivery. However, ART initiation before ATT did not significantly increase the risk of mortality in our population.
Adverse TB treatment outcome in actively detected cases was associated with CD4 count ≤100 cells/mm3 and low BMI and MUAC. Likewise, low BMI and MUAC were also associated with adverse treatment outcome in PCF cases. Our results support the utility of MUAC as a predictor for mortality in HIV-positive individuals with TB (Citation30).
Participants for this study were recruited from two cohorts of HIV-positive individuals with overlapping periods of enrolment, using similar recording of baseline characteristics and follow-up. Further, patient management, including ART initiation, was at the discretion of the attending clinicians for both cohorts.
This study has some limitations. Whereas the ACF cohort had eligibility criteria of baseline CD4 count ≤350 cells/mm3 and/or WHO stage 4, all ART-naïve TB patients irrespective of CD4 count and WHO stage were recruited in the PCF cohort. Consequently, 73/307 (23.8%) of the PCF cohort had CD4 count >350 cells/mm3. Additionally, methods for TB diagnosis were different in the two cohorts. In the ACF cohort, all participants underwent bacteriological TB testing, whereas national guidelines were used for TB diagnosis in the PCF cohort. Hence, it is possible that some of the clinically diagnosed PCF cases may have had other conditions than TB (Citation31). Whereas a proportion of participants in the PCF cohort were diagnosed by physicians, the ACF participants were largely identified by non-physician clinicians which might have contributed to the higher proportion of extrapulmonary cases in the PCF cohort. Finally, although the two cohorts were recruited from similar settings, potential differences in socio-demographic characteristics might have obscured any difference in ATT outcomes between the two groups.
Conclusion
Although ACF in PLHIV led to detection of TB cases with less advanced clinical characteristics, ATT outcome was similar to that in PCF subjects. A proportion of patients identified through ACF never started ATT. As the principal goals of early diagnosis are early treatment initiation and successful outcome, targeted ACF using effective diagnostic tools should be coupled with intensive patient-centred counselling. We recommend further epidemiological studies that investigate effectiveness and costs associated with ACF at health facilities and communities with high burden of TB and HIV.
Authors' contributions
Conceived and designed the experiments: PB, ES, NW, SS, TTB. Performed the experiments: TTB, SS, PB, ES, NW. Wrote the paper: TTB, PB, NW, ES, SS.
Conflict of interest and funding
No conflicts of interest declared. We received funding for this study from the Swedish Civil Contingency Agency (MSB), the Swedish International Development Cooperation Agency (SIDA) and the Swedish Medical Association.
Acknowledgements
We would like to thank the study participants in both cohorts. We also appreciate the support from the study health centres, Adama Regional Laboratory and International Clinical Laboratories. Finally, our gratitude goes to our data management team specially Gadissa Merga.
Related Research Data
References
- World Health Organization. Policy on collaborative TB/HIV activities: guidelines for national programmes and other stakeholders. 2012; Geneva: World Health Organization.
- Dowdy DW, Basu S, Andrews JR. Is passive diagnosis enough? The impact of subclinical disease on diagnostic strategies for tuberculosis. Am J Respir Crit Care Med. 2013; 187: 543–51.
- Godfrey-Faussett P, Ayles H. Can we control tuberculosis in high HIV prevalence settings?. Tuberculosis. 2003; 83: 68–76.
- World Health Organization. Systematic screening for active tuberculosis. 2013; Geneva: World Health Organization.
- Harris JB, Hatwiinda SM, Randels KM, Chi BH, Kancheya NG, Jham MA, etal. Early lessons from the integration of tuberculosis and HIV services in primary care centers in Lusaka, Zambia. Int J Tuberc Lung Dis. 2008; 12: 773–9.
- Howard AA, El-Sadr WM. Integration of tuberculosis and HIV services in sub-Saharan Africa: lessons learned. Clin Infect Dis. 2010; 50(Suppl 3): S238–44.
- Denegetu AW, Dolamo BL. Tuberculosis case finding and isoniazid preventive therapy among people living with HIV at public health facilities of Addis Ababa, Ethiopia: a cross-sectional facility based study. BMC Public Health. 2014; 14: 52.
- Balcha TT, Skogmar S, Sturegård E, Schön T, Winqvist N, Reepalu A, etal. A clinical scoring algorithm for determination of the risk of tuberculosis (TB) in HIV-positive adults – a cohort study performed at Ethiopian health centers. Open Forum Infect Dis. 2014; 1: 1–26.
- Hermans SM, van Leth F, Kiragga AN, Hoepelman AIM, Lange JMA, Manabe YC. Unrecognised tuberculosis at antiretroviral therapy initiation is associated with lower CD4+ T cell recovery. Trop Med Int Health. 2012; 17: 1527–33.
- Bassett IV, Wang B, Chetty S, Giddy J, Losina E, Mazibuko M, etal. Intensive tuberculosis screening for HIV-infected patients starting antiretroviral therapy in Durban, South Africa. Clin Infect Dis. 2010; 51: 823–9.
- Swindells S, Komarow L, Tripathy S, Cain KP, MacGregor RR, Achkar JM, etal. Screening for pulmonary tuberculosis in HIV-infected individuals: AIDS Clinical Trials Group Protocol A5253. Int J Tuberc Lung Dis. 2013; 17: 532–9.
- Reid MJ, Shah NS. Approaches to tuberculosis screening and diagnosis in people with HIV in resource-limited settings. Lancet Infect Dis. 2009; 9: 173–84.
- World Health Organization. Roadmap for rolling out Xpert MTB/RIF for rapid diagnosis of TB and MDR TB. 2010; Geneva: World Health Organization. 1–12.
- Kranzer K, Houben RM, Glynn JR, Bekker L-G, Wood R, Lawn SD. Yield of HIV-associated tuberculosis during intensified case finding in resource-limited settings: a systematic review and meta-analysis. Lancet Infect Dis. 2010; 10: 93–102.
- Balcha TT, Sturegård E, Winqvist N, Skogmar S, Reepalu A, Jemal ZH, etal. Intensified tuberculosis case-finding in HIV-positive adults managed at Ethiopian health centers: diagnostic yield of Xpert MTB/RIF compared with smear microscopy and liquid culture. PLoS One. 2014; 9: e85478.
- Lawn SD, Kranzer K, Edwards DJ, McNally M, Bekker L-G, Wood R. Tuberculosis during the first year of antiretroviral therapy in a South African cohort using an intensive pretreatment screening strategy. AIDS. 2010; 24: 1323–8.
- Schmaltz CAS, Santoro-Lopes G, Lourenço MC, Morgado MG, Velasque LDS, Rolla VC. Factors impacting early mortality in tuberculosis/HIV patients: differences between subjects naïve to and previously started on HAART. PLoS One. 2012; 7: e45704.
- Kranzer K, Lawn SD, Meyer-Rath G, Vassall A, Raditlhalo E, Govindasamy D, etal. Feasibility, yield, and cost of active tuberculosis case finding linked to a mobile HIV service in Cape Town, South Africa: a cross-sectional study. PLoS Med. 2012; 9: e1001281.
- Golub JE, Mohan CI, Comstock GW, Chaisson RE. Active case finding of tuberculosis: historical perspective and future prospects. Int J Tuberc Lung Dis. 2005; 9: 1183–203.
- Kranzer K, Afnan-Holmes H, Tomlin K, Golub JE, Shapiro AE, Corbett EL, etal. The benefits to communities and individuals of screening for active tuberculosis disease: a systematic review. Int J Tuberc Lung Dis. 2013; 17: 432–46.
- Skogmar S, Balcha TT, Jemal ZH, Bjork J, Deressa W, Schön T, etal. Development of a clinical scoring system for assessment of immunosuppression in patients with tuberculosis and HIV infection without access to CD4 cell testing results from a cross-sectional study in Ethiopia. Glob Health Action. 2014; 1: 1–10.
- Ministry of Health of Ethiopia. Guidelines for clinical and programmatic management of TB, leprosy and TB/HIV in Ethiopia. Available from: http://www.etharc.org/resources/download/finish/33/709 [cited 19 December 2014]. 5th ed.
- World Health Organization. definitions and reporting framework for tuberculosis. 2013; Geneva: World Health Organization. 1–40.
- Den Boon S, Verver S, Lombard CJ, Bateman ED, Irusen EM, Enarson DA, etal. Comparison of symptoms and treatment outcomes between actively and passively detected tuberculosis cases: the additional value of active case finding. Epidemiol Infect. 2008; 136: 1342–9.
- Shargie EB, Lindtjørn B. Determinants of treatment adherence among smear-positive pulmonary tuberculosis patients in Southern Ethiopia. PLoS Med. 2007; 4: e37.
- Lorent N, Choun K, Thai S, Kim T, Huy S, Pe R, etal. Community-based active tuberculosis case finding in poor urban settlements of Phnom Penh, Cambodia: a feasible and effective strategy. PLoS One. 2014; 9: e92754.
- Murray CJ, Salomon JA. Expanding the WHO tuberculosis control strategy: rethinking the role of active case-finding. Int J Tuberc Lung Dis. 1998; 29(9 Suppl 1): S9–15.
- Wachira J, Kimaiyo S, Ndege S, Mamlin J, Braitstein P. What is the impact of home-based HIV counseling and testing on the clinical status of newly enrolled adults in a large HIV care program in Western Kenya?. Clin Infect Dis. 2012; 54: 275–81.
- Martinson NA, Hoffmann CJ, Chaisson RE. Epidemiology of tuberculosis and HIV: recent advances in understanding and responses. Proc Am Thorac Soc. 2011; 8: 288–93.
- Gustafson P, Gomes VF, Vieira CS, Samb B, Nauclér A, Aaby E, etal. Clinical predictors for death in HIV-positive and HIV-negative tuberculosis patients in Guinea-Bissau. Infection. 2007; 35: 69–80.
- Siddiqi K, Lambert M-L, Walley J. Clinical diagnosis of smear-negative pulmonary tuberculosis in low-income countries: the current evidence. Lancet Infect Dis. 2003; 3: 288–96.