Diagnostic Accuracy of Sonographic Signs of Extrapulmonary Tuberculosis and Treatment Response Monitoring in HIV-Positive and -Negative Populations

Background Diagnosis and monitoring of extrapulmonary tuberculosis (EPTB) remains challenging. Ultrasound such as the extended focused assessment for HIV-associated tuberculosis (eFASH) protocol might improve diagnosis and monitoring of treatment responses. This study determined the diagnostic accuracy of eFASH for EPTB and its value in monitoring EPTB treatment response compared with clinical signs and symptoms. Methods We performed a post-hoc analysis of a trial assessing eFASH impact on management of adults with suspected EPTB. Participants who had baseline and follow-up ultrasound examinations were included. We assessed the diagnostic accuracy of eFASH and compared the evolution of eFASH and clinical signs and symptoms in participants with definite EPTB, stratified by favorable treatment outcomes at 6 months. Results In 296 included participants (95 with definite EPTB, 201 with no definite EPTB), the most common eFASH signs were pleural effusion (47%) and pulmonary B-lines with subpleural granular artefacts (34%). Pleural effusion was the only sign that persisted beyond 6 months. eFASH had a sensitivity of 93.7% (95% CI, 86.8-97.6) and a specificity of 37.8% (95% CI, 31.1- 44.9) for definite EPTB. At 2 months, favorable outcomes were similar between participants with full and partial resolution of eFASH signs (83% versus 81%). In contrast, a higher proportion of favorable outcomes was seen in participants with full resolution of clinical signs and symptoms (90% versus 60%). Conclusion eFASH shows high sensitivity but low specificity for definite EPTB. Ultrasound can be used alongside clinical signs and symptoms to monitor treatment response in patients with EPTB.