Optimizing the diagnosis and management of tuberculosis and assessing the burden of hypertension among HIV patients in a high TB and HIV burden setting

• Main Author: Owolabi, Rotimi Samuel
• Other Authors: Cuevas, Luis ; Dacombe, Russell
• Published: University of Liverpool 2018
• Online Access: https://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.762800

Background: TB is the commonest cause of death among people living with HIV (PLHIV) in Nigeria. It is difficult to diagnose TB among PLHIV and diagnosed patients often experience poor treatment outcome. Although, PLHIV are living longer because of increased access to antiretroviral therapy (ART), hypertension is emerging as a major co-morbidity in this population. This study explored better means of diagnosing TB among individuals with and without HIV, described TB treatment outcomes and their determinants, and assessed the burden of hypertension among PLHIV. Methods: This thesis comprises four studies conducted in University of Abuja Teaching Hospital (UATH): one cross-sectional study of PLHIV to assess the performance of the WHO TB symptom screening algorithm; a prospective study to evaluate whether C-reactive protein (CRP) and Interferon gamma-inducible protein 10 (IP-10) could be used to screen individuals for TB; a retrospective study to describe the TB treatment outcome of PLHIV and HIV-negative patients and the risk factors for poor treatment outcome and a retrospective study to describe the prevalence and incidence of hypertension and its determinants among PLHIV registered over a period of 3 years. Results: 202 PLHIV were screened for TB and 72.3% had symptoms of TB. However, only 3% and 6.5% had culture or culture plus smear confirmed TB, respectively. The WHO algorithm had 83.3% sensitivity, 29.1% specificity and 98.2% negative predictive value. CRP and IP-10 were measured in 408 patients with TB symptoms, of which, 21% had culture-confirmed TB. CRP had 91.4% and 33.2% sensitivity/specificity among all participants, 95.3% and 42.6% among HIV-negative patients, and 84.8% and 22.1% sensitivity/specificity among PLHIV, respectively. IP-10 had 87.3% and 40.9% sensitivity/specificity among all participants and 87.5% and 50.3% among HIV-negative and 79.4% and 47.2% among PLHIV, respectively. 998 patients were treated for TB by the hospital in the last 5 years. Of these, 62% had treatment success, 8.4% died and 18% were lost-to-follow up (LTFU). TB/HIV-coinfection rate was 44.3%. Treatment success was 52.3% among PLHIV and 70% among HIV-negative patients (p = 0.001) with higher deaths (15.5% versus 2.9%, p = 0.001) and LTFU (22.5% versus 14.3%, p = 0.001) among PLHIV. Poor treatment outcome and LTFU were more frequent among older individuals, PLHIV, those without sputum smear results and low body weight. Low weight and not receiving Co-trimoxazole were associated with poor outcome and LTFU among PLHIV. 12.8% of 883 PLHIV had hypertension at enrolment and 11.2% developed hypertension 12 months after enrolment. Hypertension was associated with older age, higher BMI, hepatitis B and higher CD4 counts. Patients with incident hypertension had higher systolic and diastolic blood pressure on enrolment. Conclusion: The WHO TB screening algorithm performed as reported by WHO. However, most patients had symptoms of TB and required further tests, which limits its efficiency in high burden settings. CRP performed better than IP-10 and both markers performed better in HIV-negative than among PLHIV. TB treatment outcome was poor and was worse among PLHIV. Hypertension is a common problem among PLHIV. There is a need for better screening tools for TB among HIV-positive patients and to develop interventions to improve the outcome of patients with TB, especially among the elderly, PLHIV and those presenting with underweight. The high burden of hypertension among HIV patients signals the need to integrate care of hypertension and other noncommunicable diseases into HIV care programs.