Background:
The advanced HIV disease (AHD) care package reduces mortality in AHD but is barely implemented. We assessed the feasibility of its implementation, including a novel point-of-care CD4-test.
Methods:
A mixed-methods study, embedded in the TB TRIAGE+ study (ClinicalTrials.gov: NCT04666311), in patients with TB symptoms presenting at two rural facilities in Lesotho and South Africa. HIV-positive participants received VISITECT CD4 lateral flow assay (CD4 LFA), urine Alere Determine TB-lipoarabinomannan antigen (TB-LAM) LFA, Immy cryptococcal antigen (CrAg) LFA on plasma in case of CD4≤200cells/µL, and treatment referral. We assessed compliance with the testing sequence, observed procedural completeness (Likert-scale) and timing, and assessed acceptability among implementers (questionnaires, group discussions).
Results:
In Lesotho and South Africa, respectively, 47.9%(335/700) and 49.3%(341/692) of study participants were HIV-positive. Among those eligible, 99.1%(332/335) and 99.7%(340/341) received CD4 LFA, 100.0%(335/335) and 99.4%(339/341) TB-LAM and 98.7(150/152) and 100.0%(50/50) CrAg (Table 1). Among tested females and males, respectively, 20.2%(70/347) and 40.6(132/325) had CD4≤200cells/µL, TB-LAM was detected in 14.1%(49/348) and 12.3%(40/326), and CrAg in 5.7%(4/70) and 3.1%(4/130). Study procedures were rated complete (n=16 Likert scales). The three tests were performed in 73 [median, IQR:68-85] minutes (n=5 observations). Reasons to not perform a test were patient refusal after HIV diagnosis (1 CD4 LFA and TB-LAM), failure to produce a sample (1 TB-LAM), and expired tests or reagents (3 CD4 LFA, 2 CrAg). Implementers estimated the risk of making mistakes low but found timing of procedural steps, especially for CD4 LFA, initially challenging. Waiting time was reduced by running CD4 LFA and TB-LAM in parallel. Implementers agreed that the package is acceptable for them and patients, easy to perform after a short learning curve, and can be performed anywhere, by any cadre with training. They were in favor of community implementation, where the need was perceived as higher.
Conclusions:
We found a high prevalence of AHD, especially in Lesotho, including among people with known HIV. Implementing the AHD care package was feasible at facility level and anticipated to be feasible in the community. Enablers include uninterrupted supplies of tests and good time management. The AHD care package can and should be implemented to reduce AIDS-related mortality in these settings.
Table 1: Test results of advanced HIV disease care package ![]()