We read with interest the results reported by Reynolds et al.. In a cohort of patients followed for a median (range) time of 20.2 (12.4–29.5) months, it was shown that the majority of patients with immunological criteria for treatment failure are virologically suppressed. The authors conclude that immunological monitoring alone could result in unnecessary switching to second-line treatment.
At the Infectious Diseases Institute (IDI) in Kampala, Uganda, viral load testing is performed only in the context of clinical trials due to cost; patients on antiretroviral therapy (ART) are clinically evaluated monthly and have CD4+ cell count testing every 6 months. In January 2008, we sought to assess the validity of the WHO immunological criteria as a surrogate for virological failure by testing HIV viral load in clinic patients fulfilling the WHO criteria for ART failure: patients with a CD4+ cell count of less than 100 cells/μl after 1 year on ART; patients with a 50% absolute CD4+ cell count fall from the peak CD4+ cell count while on ART; and CD4+ cells per microlitre of baseline or less after 1 year on ART .
From the clinic database of more than 8000 patients who had ever initiated ART at IDI, we identified all patients who had received ART for at least 1 year and who fulfilled at least one WHO criterion for immunological failure. Viral load testing was performed at the next available clinic visit. An undetectable viral load was defined as 400 copies/ml or less.
We identified 4403 patients who had been on their first-line ART regimen for at least 1 year; 70% were initiated on a nevirapine and 30% on an efavirenz-containing regimen. A subset of 140 patients fulfilled at least one WHO criterion for immunological failure. Sixty-seven patients did not have viral load testing for the following reasons: four had died, 17 were lost to follow up, four were transferred, 22 had already switched to second-line ART because they fulfilled criteria for immunological failure, and 20 had already a viral load test. The remaining 73 patients underwent viral load testing. The median time on ART was 2.5 years (1.9–3.7 years), 37 (51%) patients were women, median [interquartile range (IQR)] age in years was 36 (30–40 years), and median CD4+ cell count at evaluation was 43 cells/μl (11–113 cells/μl). The viral load results for these 73 patients are shown in Table 1. Overall, 52/73 (71%) of the patients tested had viral suppression. In the group of patients (18) that fulfilled more that one criterion for treatment failure, a substantial proportion of patients were also virologically suppressed (12, 66%).
Our results support the conclusions of Reynolds et al.  as well as previous reports from resource-limited settings (RLS) [3,4]. Overall, almost two-thirds of patients fulfilling at least one of the WHO criterion for immunological failure at a median of 2.5 years on first-line ART had an undetectable viral load. Current WHO immunological criteria alone are insensitive in the diagnoses of treatment failure and may result in unnecessary switching to second-line therapy, which is both more costly and associated with greater adverse events. As an increasing number of patients begin to fail first-line ART in RLS, affordable viral load testing should be made available to confirm virological failure before switching patients to second-line therapy.
All the authors declare that they have conflicts of interest.
1. Reynolds S, Nakigozi G, Newell K, Ndyanabo A, Galiwongo R, Boaz I, et al.Failure of immunologic criteria to appropriately identify antiretroviral treatment failure in Uganda
3. Moore DM, Mermin J, Awor A, Yip B, Hogg RS, Montaner JS. Performance of immunologic responses in predicting viral load suppression: implications for monitoring patients in resource limited settings. J Acquir Immune Defic Syndr 2006; 43:436–439.
4. Mee P, Fielding KL, Charalambous S, Churchyard GJ, Grant AD. Evaluation of the WHO criteria for antiretroviral treatment failure among adults in South Africa. AIDS 2008; 22:1971–1977.