New review identifies markers of severe tuberculosis in people with HIV

Tuberculosis remains a leading infectious cause of death and serious illness in people both with and without HIV. That makes it crucial to spot the cases that are most likely to become severe so care teams can intervene early and reduce deaths and lasting harm. In a systematic review led by Robert Akpata, researchers set out to gather what is known about the features of tuberculosis that predict severe outcomes, and to use that evidence to propose a single multifactorial severity score tailored for people living with HIV. The review highlights a gap: many existing severity scores do not include important disease characteristics such as how many bacteria are present (bacillary load), how much of the lungs are affected, or whether the infection has spread beyond the lungs. They also note that most current scores were not developed specifically for people living with HIV, a group that often faces different risks and complications. To address these problems, the team collected and reviewed published data to identify factors linked to mortality, major morbidity, treatment failure, or lasting damage from tuberculosis.

The authors followed the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) approach and registered their protocol in the Prospective Register of Systematic Reviews (PROSPERO, CRD42022323983). They searched for baseline data tied to tuberculosis severity as defined by death, severe morbidity, treatment failure or sequelae in study participants aged ≥15 years. From 117 selected articles, the review catalogued factors most often reported as associated with worse outcomes. Many were host-related or mixed factors: older age, anaemia, immunosuppression, other health conditions, low body mass index, fever, fast heart rate (tachycardia), trouble breathing (dyspnoea), low blood pressure, and social factors. The review also identified disease-specific markers: higher bacillary load shown by high smear positivity level or culture positivity; greater lung involvement such as cavitation, bilateral disease, or infiltration of more than two lobes; and signs of spread beyond the lungs including LAM+, extra pulmonary involvement, and miliary or disseminated tuberculosis. The authors plan to use these findings to develop a tuberculosis severity score for people living with HIV using the Rand Appropriateness Method.

This review brings together a wide range of clinical and laboratory signals that point to more severe tuberculosis. By showing that severity is linked not only to patient characteristics but also to objective disease features — bacillary load, the extent of lung damage, and whether the infection has disseminated — it underlines what clinicians and researchers should pay attention to when assessing risk. The work is practical as well as conceptual: the team intends to convert these gathered factors into a consensual, multifactorial severity score specifically for people living with HIV. Such a score could help guide decisions about who needs quicker or more intensive treatment, closer monitoring, or additional diagnostics. Importantly, the review also reinforces that existing scoring tools may miss key disease markers, suggesting that an HIV-focused, evidence-based score could fill a real gap in care for a high-risk population.