Background: The performance of point-of-care ultrasound (PoCUS) to diagnose HIV-associated tuberculosis has not been evaluated in large prospective studies. We determined the diagnostic accuracy of individual PoCUS features, performed an external validation of the focused assessment with sonography for HIV/TB (FASH) protocol, and determined independent PoCUS predictors of HIV-associated tuberculosis appropriate for use by emergency center practitioners.
Setting: A cross-sectional diagnostic study was performed at the emergency center of Khayelitsha Hospital (Cape Town, South Africa).
Methods: HIV-positive adults with the suspicion of having tuberculosis were prospectively enrolled. PoCUS was performed according to a standardized protocol. Reference standard was the detection of Mycobacterium tuberculosis using Xpert MTB/RIF or culture.
Results: We enrolled 414 participants: 243 female, median age 36 years, median CD4 cell count 86/mm, and 172 (42%) had tuberculosis. Sensitivity and specificity were ≥1 individual PoCUS feature [73% (95% CI: 65 to 79), 54% (95% CI: 47 to 60)], FASH protocol [71% (95% CI: 64 to 78), 57% (95% CI: 50 to 63)]. Independent PoCUS predictors identified were intra-abdominal lymphadenopathy of any size (aDOR 3.7 (95% CI: 2.0 to 6.7)], ascites [aDOR 3.0 (95% CI: 1.5 to 5.7)], and pericardial effusion of any size [aDOR 1.9 (95% CI: 1.2 to 3.0)]. The c-statistic for the derivation model was 0.680 (95% CI: 0.631 to 0.729), compared with 0.630 (95% CI: 0.576 to 0.684) of the FASH protocol. Two or more independent PoCUS predictors had 91% (95% CI: 86 to 94) specificity.
Conclusion: The presence of 2 or more independent PoCUS predictors (intra-abdominal lymphadenopathy, ascites, and pericardial effusion) had moderate discrimination for HIV-associated tuberculosis in patients presenting to the emergency center.