Abstract
Disseminated tuberculosis (TB) remains the leading cause of mortality among severely immunocompromised individuals living with HIV, particularly in hospital settings where sputum-based diagnostics often fail due to patient debility. The urine lipoarabinomannan (LF-LAM) lateral flow assay provides a rapid, point-of-care diagnostic method that bypasses the requirement for respiratory specimens. This prospective quasi-experimental implementation study evaluated the clinical impact of integrating routine urine LF-LAM testing into emergency diagnostic protocols for severely immunocompromised HIV-infected medical inpatients. Over an 18-month period across three tertiary hospitals, 412 hospitalized patients with CD4 counts ≤100 cells/µL were enrolled across two sequential phases: a pre-implementation phase using standard diagnostics (n = 204) and a post-implementation phase integrating bedside LF-LAM testing (n = 208). Routine urine LF-LAM integration reduced the median time from admission to anti-TB treatment initiation from 4.2 days (IQR 2.1–6.8) to 0.5 days (IQR 0.2–1.1; p < 0.001). Overall diagnostic yield for confirmed or probable active TB rose from 31.4% to 48.1% (p < 0.001). Importantly, 30-day all-cause mortality was significantly lower in the LF-LAM implementation cohort compared to the baseline cohort (14.9% vs. 23.5%; adjusted hazard ratio 0.61, 95% CI: 0.39–0.94; p = 0.024). Bedside urine LF-LAM testing substantially accelerates treatment initiation and reduces short-term mortality, reinforcing its mandatory inclusion in routine diagnostic bundles for acute HIV admissions.
