Retrospective review of urinary lipoarabinomannan (ULAM) in the diagnosis of disseminated tuberculosis (TB) in medical inpatients at a district level hospital in Cape Town

Erwee, Christie

Retrospective review of urinary lipoarabinomannan (ULAM) in the diagnosis of disseminated tuberculosis (TB) in medical inpatients at a district level hospital in Cape Town

Thesis / Dissertation

2025

Publisher

University of Cape Town

Department

Department of Medicine

Faculty

Faculty of Health Sciences

Abstract

Background: Mycobacterium tuberculosis (MTB) and Human Immunodeficiency Virus (HIV) remain a major public health concern in South Africa (SA) with high mortality rates in patients with MTB and HIV co-infection. In 2016 Urinary Lipoarabinomannan (ULAM) became available in SA to expedite the diagnosis and treatment of Tuberculosis (TB) in this vulnerable population. Objectives To describe the experience at a district level hospital following the availability of ULAM testing for investigating HIV positive TB suspects. Firstly, to define the population and clinical outcomes in this cohort and secondly to assess if the hospital's standard operating procedure (SOP) regarding ULAM testing was followed and what additional MTB investigations were performed. Methods An observational, retrospective cohort study of patients who underwent ULAM testing at Mitchells Plain District Hospital (MPDH) was conducted from 2016 until 2018. The following data was collected: age, gender, CD4 count, viral load (VL), whether Antiretroviral Treatment (ART) naïve or interrupted and duration of ART if on treatment. Additionally, MTB diagnostic tests were performed and finally, patient clinical outcomes with regards to length of stay (LOS) and 30-day mortality and whether TB treatment was initiated. Descriptive statistics were used to analyse the data. Results 324 participants were included in the study. 54.6% of participants were female and the median age of the entire cohort was 36 (IQR 30-34). Median CD4 count was 41 (IQR 16-76) with high viral load of log 4.9 (IQR 3.9-5.4). 44.1% of participants were ART naïve, 36.1% of participants were on treatment and 19.8% had interrupted treatment. The median duration of ART was 0.6 years (IQR 0.1-2). 750 MTB specific tests were performed in the cohort, 48% of the participants initiated on TB treatment had positive microbiological results proving MTB infection. Overall 30-day mortality was 12.7%. Conclusions This study showed participants who were severely immunocompromised with little ART exposure and a mortality rate of 12.7%. Patients were frequently initiated on empiric MTB treatment with a proportion of participants not being adequately investigated, particularly in the LAM negative group. Sputum was the most frequent MTB specific test performed. Clinicians adhered to the SOP put in place to guide the use of ULAM.