Screening for Tuberculosis in HIV-infected Patients Eligible for Antiretroviral Treatment

Background: Tuberculosis (TB) remains as the leading opportunistic infection and cause of death among HIV-infected subjects in resource-limited settings (1). HIV infection is the strongest risk factor known for developing active TB, and as a consequence, the HIV epidemic in Sub-Saharan Africa has fuelled the incidence of TB (2). The interactions between HIV and TB are great challenges for health care, both with regard to diagnosis and treatment. In HIV-infected persons TB often has atypical manifestations, making correct and timely diagnosis difficult (3, 4). Recognition of TB before starting antiretroviral therapy (ART) is critical since a missed diagnosis of TB is associated with high early mortality after ART initiation (1, 5). Screening for TB is recommended for HIV-positive subjects, but a method with both satisfactory specificity and sensitivity that can be used in primary health care settings in Africa has not yet been identified. New diagnostic techniques for TB have recently been developed, such as the GenXpert PCR (6), but the optimal use of these alternative methods needs to be investigated.

In settings where both TB and HIV infection are common, a high proportion of patients are in need both of anti-tuberculosis treatment (ATT) and ART. Combined ART and ATT is associated with risks of drug interactions, overlapping side effects, immune reconstitution disease and poor adherence, but at least in patients with severe immunodeficiency, these risks are outweighed by decreased mortality (7-9). Studies performed in developed countries have shown that rates of virological suppression during ART in patients receiving concomitant ATT are comparable to those in persons not taking ATT (10, 11), but this topic has not been extensively investigated at primary health care level in Africa.

The number of HIV-infected subjects receiving ART in Sub-Saharan Africa has increased dramatically during the last years. Still, it is estimated that less than half of all patients in need of ART currently receive such treatment (12). In order to improve access, it will become necessary to further decentralize ART and to integrate this treatment into the primary health care system. Since many patients starting ART have unrecognized active TB at this time point, it is important to find a practical and reliable screening algorithm for TB among such individuals. Furthermore, the outcome of combined ATT and ART delivered in health centres must be assessed, especially with regard to HIV RNA suppression and selection of ART resistance.

This project aims to develop an algorithm with the potential to identify indicators of prevalent and incident TB among patients initiating ART by correlating baseline clinical signs and symptoms with results of culture- and PCR-based investigations for TB. Such clinical characteristics will be compared both to TB present before starting ART, as well as to TB presenting during the first year after ART initiation. Results from this comparison will be used to construct an algorithm for TB screening in Ethiopian patients eligible for ART. In addition, clinical, immunological and virological outcomes of ART in patients with concomitant ATT will be compared to those in patients only receiving ART. Specific aims are to study rates of virological suppression during ART, the development of antiretroviral drug resistance, and treatment adherence.

Work plan: HIV-infected adult patients eligible to start ART will be recruited from HIV clinics in several health centres providing integrated for HIV and TB in the Oromia region. 812 patients have been prospectively included into the cohort, and are under continued follow-up.

At baseline, detailed demographic, socio-economic and clinical data (such as presence of general and respiratory symptoms, body mass index, mid-upper arm circumference and various findings on physical examination) has been collected, including information on current medical and traditional therapies. Blood for CD4 cell count, complete blood count and HIV RNA will be obtained. In addition, plasma, stool and urine sample for analysis of immunological and inflammatory markers, and potential new biomarkers of active TB infection, has been collected and stored at -80 and -20 ° C.

Participants have undergone sampling for microbiological investigations for TB before starting ART. All patients have submitted two sputum samples for TB microscopy, culture and PCR. From patients with peripheral lymphadenopathy, lymph node aspiration for microbiological testing has been done. PCR has been performed using the GeneXpert technique. The study protocol will include definitions of TB based on microbiological results and clinical data according to WHO guidelines. Results of TB cultures are not awaited before ART initiation. Patients in whom TB is diagnosed receive ATT as soon as the diagnosis is established. For patients who have started ART at this time point, the recommendation will be to continue ART; however, the responsible clinician may decide to interrupt ART if medically indicated.

Follow up: Participants will be followed for 48 months (once monthly during the first 3 months, subsequently every 3 months for the first year of the study and then every 6 months). On follow-up visits, clinical and adherence data is collected. Incidence of death, hospitalization, TB and other opportunistic infections, signs of immune reconstitution inflammatory syndrome (IRIS), defaulting or treatment interruption is followed continuously, with tracing of defaulters to ascertain rates of mortality. Blood for CD4 cell counts and HIV RNA is obtained at 1 month after starting ART and at 3-monthly intervals for the first 6 months after study inclusion (and at 6-monthly intervals onwards). Samples obtained more than 6 months after starting ART with detectable HIV RNA will be subjected to genotypic testing to detect drug resistance mutations. In addition, blood samples for analysis of immunological markers will be obtained after 1, 3, 6 and 12 months of ART. Urine samples have been analyzed for the presence of lipoarabinomannane using a lateral flow assay.

Data analysis: Patient characteristics will be correlated with prevalent TB at baseline, to identify indicators of TB. These results will be used to construct screening algorithms that will be validated for sensitivity, specificity and predictive accuracy. Baseline predictors for incident TB during ART will also be investigated. For assessment of ART outcome, patients will be categorized according to their baseline TB status. Participants taking ATT will be compared to controls only taking ART. Incidence and time to defined events will be studied using survival curves (mortality, disease progression, HIV RNA suppression, loss to follow-up). HIV RNA and CD4 cell levels during ART, treatment interruptions, adherence, new opportunistic infections, drug side effects and TB IRIS will be compared between groups. Predictors of death and ART failure will be assessed.

Significance: Improving the management of patients co-infected with TB and HIV in highly endemic regions is critical for the success of continued scaling up of ART. Initiating ART during ATT is challenging under existing conditions in Sub-Saharan Africa, where the majority of these patients are diagnosed and treated. Case detection rates of TB remain below estimates in Ethiopia. Active TB is frequently missed in patients eligible for ART, a phenomenon likely to be important for mortality and medical complications following ART initiation. A screening algorithm based on clinical data with adequate predictive accuracy could help to identify TB in this population, and lead to improved treatment outcomes both for TB and HIV infection. Increased detection and timely treatment of TB in HIV-infected subjects would also reduce transmission in society. The outcome of combined ATT and ART administered in health centres must be assessed, especially concerning the development of drug resistance which can have serious consequences both for the infected individual and the community. Results from this study could inform health authorities on the efficacy and feasibility of such combined therapy, which is important for future HIV/TB policies in Ethiopia and other Sub-Saharan countries.

The project has received ethical clearance from Lund University, Sweden, and from the Institute of Science and Technology, Ethiopia.