High death rate in Xpert-negative people with HIV in Uganda

Tuberculosis (TB) and HIV remain closely linked in sub-Saharan Africa, and diagnosing TB in people living with HIV (PLHIV) can be difficult because common sputum tests miss some infections. To better understand outcomes when sputum molecular tests are negative, corresponding author Lydia Nakiyingi and colleagues carried out a prospective study of adults living with HIV who were suspected of having TB but had a negative Xpert result. The team recruited participants from Mulago Referral Hospital and Kisenyi Health Centre-IV in Kampala, Uganda between November 2017 and December 2020. Most of the people in the study were already on antiretroviral therapy (ART), but clinicians sometimes started TB treatment on clinical grounds despite the negative Xpert — so-called empiric TB treatment. The researchers followed 300 Xpert-negative, presumptive TB adult PLHIV for 12 months, recording clinical information, laboratory tests and whether doctors chose to start empiric TB treatment. By tracking outcomes over a year, the study aimed to compare survival in those who did and did not receive empiric TB therapy and to look for other infections or causes of illness that might explain deaths.

The study collected detailed clinical data — TB symptoms, chest X-ray findings and the treating clinician’s decision about empiric TB treatment — and ran several laboratory tests. Investigations included CD4 cell count, serum cryptococcal antigen (CrAg), urine TB-lipoarabinomannan (TB-LAM), microbiological blood cultures, and sputum mycobacterial growth indicator tube (MGIT) cultures. Participants were followed monthly for 12 months to determine vital status. Of 300 enrolled adults, 61.3% were inpatients, 55.7% were female, median age was 37 years, 82.3% were on ART, and median CD4 was 206 cells/mm³. Clinicians started empiric TB treatment in 68 participants (22.7%), most of whom (53) were inpatients. Overall 12-month mortality was 31.0% (93/300); most deaths were among inpatients and 72% of deaths occurred within three months of enrollment. Mortality rates were higher among those who received empiric TB treatment (51.5 vs. 30.2 per 1,000 person-months; p=0.013). TB cultures were positive in 5.0% (15/300), seven of those were also TB-LAM positive; CrAg was positive in 12.3% and 3.7% had positive blood culture.

The findings show a troubling level of death among Xpert-negative PLHIV even where ART use is common. Cryptococcal antigenemia and bacteremia were not uncommon in this group and could help explain some of the illnesses and deaths. Importantly, patients who received empiric TB treatment had higher mortality, a signal that either sicker patients were being treated empirically or that TB was not the only or main cause of their illness. Based strictly on these results, the authors conclude that when a person living with HIV has a negative Xpert test, clinicians should not stop with that single result. Instead, the abstract recommends performing extensive laboratory evaluations to identify possible comorbidities or alternative non-TB diagnoses so that treatment can be better targeted and other life-threatening infections are not missed.