Background: Tuberculosis (TB) and HIV are leading causes of disease and death in Subsaharan Africa. Antiretroviral therapy (ART) dramatically improves prognosis in HIV infection, but TB is still a common complication in HIV-infected subjects. Management of TB-HIV co-infection is complex, both with regard to diagnosis and treatment. Since scaling-up of ART requires management of most patients in primary health care, it is critical to achieve better strategies for TB-HIV co-infection at peripheral levels of the health care system in endemic regions. This includes development of new methods to assess the severity of HIV disease and the need to start ART during TB treatment in this population.

Aims: To compare a scoring system for clinical signs of immunosuppression with CD4 cell counts to assess HIV disease severity and indications for ART initiation, and to correlate immunosuppression status with treatment outcome.

Workplan: CD4 cell levels and results of clinical scoring has been compared in 1100 patients with TB, using HIV-negative subjects with TB treatment for control. Inclusion was closed in February 2012, and follow-up of participants completed in August 2012. Plasma levels of immune activation and inflammatory markers will be correlated with the degree of immunosuppression.

Significance: TB is the most significant clinical challenge to the successful scaling-up of ART in Africa. In order to improve management in primary health care it is necessary to find robust and reliable techniques for determining disease severity and identification of patients who need to start ART during TB treatment. This study may contribute to increased knowledge in this field and help to modify guidelines for management of TB-HIV co-infection in Ethiopia as well as in other resource-limited settings.

Tuberculosis (TB) remains as the leading opportunistic infection and cause of death among HIV-infected subjects in resource-limited settings (1). The management of TB-HIV co-infection constitutes a great challenge for health care systems. Often, HIV infection is first detected when patients are diagnosed with active TB, and many such patients have advanced immunosuppression (2, 3). However, TB can also occur in individuals with relatively well preserved CD4 cell counts, especially in areas with high TB endemicity (4). In settings where both TB and HIV infection are common, a high proportion of patients are in need both of anti-TB treatment (ATT) and antiretroviral therapy (ART) at diagnosis. Combined ART and ATT is associated with risks of drug-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 (5-7).

Although the optimal timing for when to start ART in patients receiving ATT is not completely understood, CD4 cell count thresholds are currently often used to guide clinicians on when to start ART in co-infected patients (8). Briefly, initiation of ART during ATT is indicated for patients with CD4 counts lower than 350 cells/mm3, whereas it is usually recommended to defer ART for patients without other clinical signs of immunodeficiency and CD4 cell counts higher than 350 cells/mm3. In a recently published WHO document, initiation of ART is proposed for all patients with active TB, but timing of ART remains uncertain, as well as the need for ART in subjects with CD4 cell counts exceeding 350 cells/mm3 (9). Studies aiming at defining optimal time points for ART initiation in patients receiving ATT are ongoing, and might lead to revised recommendations with regard to co-administration of these therapies.

The number of HIV-infected subjects initiating ART has increased impressively over recent years, especially in Subsaharan Africa. Still, it is estimated that less than half of all patients in need of ART currently receive such treatment (10). In order to improve access, it is necessary to decentralize treatment and integrate ART into the primary health care system. In such settings, CD4 cell count analysis is rarely available, and patients are usually managed by health care workers with limited knowledge and experience of HIV-TB co-infection. Finding simple alternative tools to identify immunosuppressed patients in these circumstances could therefore lead to significant improvements in care.

The investigators hypothesize that a structured scoring system based on clinical symptoms and signs could be used to assess the degree of immunosuppression in HIV-infected patients with TB, and thus help to categorize patients to determine when ART should be initiated in settings with limited access to laboratory resources. The principal aim of the project is to investigate how clinical scoring correlates with CD4 cell levels and specific CD4 cell cut-off points that are used to identify patients who fulfil criteria for starting ART during ATT (specifically for thresholds of 200, 350 and 500 cells/mm3), and to compare these results with scoring in TB patients without HIV co-infection. The study will also evaluate whether scoring results can predict short term mortality and treatment outcome. In addition, the investigators aim to investigate a panel of alternative immunological surrogate markers that could be used to measure HIV-related immunosuppression in patients co-infected with TB.

Work plan Methods: Patients diagnosed with active tuberculosis in health care facilities in the Oromia region, Ethiopia are eligible for inclusion. The study has been performed in the district hospital of Bishofto and in 4 health centres providing integrated care for TB and HIV (Adama, Welenchity, Mojo and Dukam health centres). Approximately 1 112 patients have been included. In these regions, approximately 40% of patients newly diagnosed with TB are HIV seropositive.

Inclusion criteria are: age over 18 years, written informed consent to participate, consent to HIV testing, residence in the catchment area of the study site, no previous treatment with ART (except for treatment for prevention of mother-to-child transmission), and diagnosis of active TB (either pulmonary or extrapulmonary according to current national Ethiopian guidelines based on WHO definitions). Patients having received ATT within the preceding 6 months and those who have received ATT for more than two weeks for their current episode of TB have not been included.

The scoring system contains 15 items, including reported symptoms (history of dysphagia, fever or diarrhoea of >1 month's duration) and clinical findings (anaemia, low body mass index and mid-upper arm circumference, oral candidiasis, oral hairy leukoplakia, gingivitis, dermatitis, Kaposi's sarcoma, herpes zoster scar).

Scoring has been performed at the time of initiation of ATT (irrespective of HIV status) by either a trained nurse or a health officer. HIV testing has been done at the same time according to local routine. Blood for CD4 cell counts and immunological markers have been obtained from all consenting participants, irrespective of HIV status (in total, 10 ml of blood). During follow-up, scoring has been repeated by study staff. CD4 cell counts as well as haemoglobin measurements for validation of clinical evaluation of anaemia in the score has been determined using local laboratory equipment already in place at the study sites. Plasma for further analyses of immune activation markers is stored at -80 ° C.

Clinicians in charge of the patients determine when ART should be started following existing national criteria (based on CD4 cell counts and WHO clinical staging). The result of the scoring under investigation has thus not been used for determining the time point for starting ART in this study.

Follow up: On initiation of ATT and at HIV diagnosis, patients have been evaluated according to the clinical score. Treatment outcome according to the WHO (cure, treatment completion, default, death, treatment failure) will be assessed for all patients. Defaulters will be traced to ascertain rates of mortality.

All patients have been longitudinally followed up with repeated clinical scoring and CD4 count after two months of TB treatment and at the end of TB treatment, with recording of mortality, new AIDS-defining conditions, paradoxical reactions occurring during ATT, and start of ART.

Data analysis: Results of the scoring system has been correlated with CD4 cell counts. Specifically, scoring results are compared with recommended CD4 cell thresholds for when ART should be started in relation to ATT, in order to investigate whether scoring cut-offs may be defined to identify patients correctly. Results of scoring in HIV-negative TB patients have been used to assess how specific total results and different included items are for HIV-related immunosuppression among co-infected individuals.

In addition, results of the scoring system will be compared to the clinician's decision with regard to the patient's indication for starting ART or not. All data will be managed under code for study purposes. Immune activation markers (neopterin and C-reactive protein) have been analyzed after completion of patient inclusion in reference laboratories in Ethiopia.

For statistical evaluation, logistic and linear regression analysis of each separate item in the scoring system will initially be performed to determine its relative importance, existence of interactions with other variables and to define relevant cut-offs for continuous variables. Different models of weighing separate scoring items will be evaluated to assess positive and negative predictive values for the respective outcomes.

For comparison of inflammatory markers in the general population living in the geographical study area, demographic data, CD4 cell counts and plasma samples for storage has been obtained from 400 healthy adults at Mojo health centre. These persons have been recruited from a voluntary counselling and testing service, with negative HIV test result and no clinical signs of active TB. Persons in this comparison group have provided written consent to participation.

Preliminary results and public health significance 1112 TB patients have been included in this study (307 co-infected with HIV). Inclusion of new participants was terminated in February 2012, with continued follow-up until the end of August 2012. Analysis of baseline data for the primary outcome has been completed and the resulting scoring system has been published in 2014. Testing of baseline samples for inflammatory markers (neopterin and C-reactive protein) was completed in November 2012.

No similar system has been studied. It is of critical importance to improve the management of patients co-infected with TB and HIV in resource-limited countries, especially in settings with access to ART. Initiating ART during ATT is challenging under existing conditions in Subsaharan Africa, where the majority of these patients are diagnosed and treated. If the clinical scoring system can provide a reasonably reliable substitute for CD4 cell counts to estimate the degree of immunosuppression in co-infected patients, and can be used to define cut-off points for optimal time points for initiation of ART, it could become a robust and simple tool to be used at peripheral levels of health care systems. This would help to identify and monitor immunosuppressed patients who need to start ART in parallel to ATT, and could lead to improved treatment outcomes and decreased mortality.

Ethical approval from the review boards of Lund University, the Oromia Regional Health Bureau and the National Research Ethics Review Committee of Ethiopia has been received.

References

1. Lawn SD, Harries AD, Anglaret X, et al. Early mortality among adults accessing antiretroviral treatment programmes in sub-Saharan Africa. AIDS 2008; 22: 1897-908
2. Ackah AN, Coulibaly D, Digbeu H, et al. Response to treatment, mortality, and CD4 lymphocyte counts in HIV-infected persons with tuberculosis in Abidjan, Côte d'Ivoire. Lancet 1995; 345: 607-10
3. Morris L, Martin DJ, Bredell H, et al. Human immunodeficiency virus-1 RNA levels and CD4 lymphocyte counts, during treatment for active tuberculosis, in South African patients. J Infect Dis 2003; 187: 1967-71
4. Mukadi Y, Perriens JH, St Louis ME, et al. Spectrum of immunodeficiency in HIV-1-infected patients with pulmonary tuberculosis in Zaire. Lancet 1993 ; 342 : 143-6
5. Dean GL, Edwards SG, Ives NJ, et al. Treatment of tuberculosis in HIV-infected persons in the era of highly active antiretroviral therapy. AIDS 2002; 16: 75-83
6. Karim SA, Naidoo K, Grobler A, et al. Timing of initiation of antiretroviral drugs during tuberculosis therapy. N Engl J Med 2010; 362: 697-706
7. Lawn SD, Myer L, Bekker LG, et al. Tuberculosis-associated immune reconstitution disease: incidence, risk factors and impact in an antiretroviral treatment service in South Africa. AIDS 2007; 21: 335-41
8. Tuberculosis care with TB-HIV co-management. WHO manual, 2007; WHO/htm/HIV/2007-01
9. Rapid advice. Antiretroviral therapy for HIV infection in adults and adolescents. WHO 2009
10. Towards universal access: scaling up priority HIV/AIDS interventions in the health sector. September 2009 progress report. WHO 2009

Inclusion Criteria:

• Diagnosis of tuberculosis
• Not having received TB treatment for more than two weeks
• Consent to HIV testing

Exclusion Criteria:

• Previous or ongoing antiretroviral therapy
• TB treatment within the preceding six months

Patients initiating treatment for tuberculosis in primary health centres providing integrated TB-HIV care in Oromia region, Ethiopia