Diagnosing tuberculosis (TB) in people living with HIV (PLHIV) remains challenging in part, because of its diversity of clinical manifestations, including high rates of extra-pulmonary and disseminated disease [1]. In particular, disseminated TB, involving multiple organ systems, is associated with high mortality but often presents non-specifically, which may hinder prompt diagnosis [2, 3]. Xpert MTB/RIF (Xpert; Cepheid, Sunnyvale, CA, USA), is currently recommended by the World Health Organization (WHO) as the first line assay for evaluating a subset of extra-pulmonary TB disease (EPTB) manifestations [4]. To detect specific forms of EPTB, such as pleural TB, TB meningitis or TB lymphadenitis, Xpert may require an invasive sample to be collected, which often limits its use for EPTB detection to hospitals where appropriate equipment is available and invasive sampling can be safely performed. Furthermore, even when concomitant pulmonary disease is present, it can be very difficult to obtain sputum in the sickest HIV patients to submit for Xpert testing [5, 6]. Therefore, an urgent priority for improving TB detection among PLHIV remains the development of rapid, point-of-care (POC) assays that use an easily obtainable clinical specimen, such as urine, and that have good diagnostic accuracy for both pulmonary and EPTB, including disseminated disease [7]. The commercially available Alere Determine TB LAM (AlereLAM; Abbott, Chicago, IL, USA) assay is a rapid, inexpensive, urinary POC TB test [8]. While its use is associated with a mortality benefit in severely ill and immunocompromised PLHIV [9, 10], it has only moderate sensitivity that is limited to patients with low CD4 counts, which has led to limited programmatic uptake [11]. We have previously reported on the Fujifilm SILVAMP TB LAM (FujiLAM; Fujifilm, Tokyo, Japan) POC assay that, similar to AlereLAM, detects the presence of lipoarabinomannan (LAM) in urine [12]. It offers on average 30% improved sensitivity for detecting TB (independent of whether it is PTB or EPTB) compared to AlereLAM across subgroups stratified by CD4 strata, while maintaining high specificity. Here we report the sensitivity of FujiLAM in comparison to AlereLAM specifically for detecting EPTB in the same patient cohorts. This post hoc analysis utilised data from two previously published, prospective cohort studies of adults (>18 years) living with HIV who were admitted to South African district hospitals on the outskirts of Cape Town [13, 14]. Cohort A enrolled patients without a current TB diagnosis regardless of presenting signs or symptoms, and independent of CD4 count [13]. Cohort B enrolled patients with a CD4 count <350 cells•μL −1 in whom TB was considered the most likely diagnosis on admission [14]. A third previously published cohort was not included in the present analysis as it excluded patients with exclusively extra-pulmonary TB disease [12]. Informed consent was obtained from patients who had capacity or regained capacity and all study-related activities were approved by the H...