University of California, San Francisco Logo

University of California, San Francisco | About UCSF | Search UCSF | UCSF Medical Center

Switching to second-line antiretroviral therapy in resource-limited settings: comparison of programmes with and without viral load monitoring
Global Health Sciences Literature Digest
Published May 10, 2010
Journal Article

The ART-LINC of IeDEA Study Group. Switching to second-line antiretroviral therapy in resource-limited settings: comparison of programmes with and without viral load monitoring. AIDS. 2009 Sept10;23(14):1867-1874.

In Context

In high-income countries, viral-load monitoring is used to identify when treatment has failed and second- or third-line drugs are needed. In resource-limited settings, however, viral-load monitoring is usually not available. Because viral failure occurs earlier than immunological failure (based on CD4 counts), lack of viral-load testing could result in treatment being switched late, increasing the risk of drug resistance. On the other hand, switching too early could reduce the amount of time a patient is on effective treatment, as drugs beyond second-line regimens are not widely available. In resource-poor settings, it is not well known how important viral-load monitoring is for the management of highly active antiretroviral therapy (HAART) in adults.


To compare rates of, time to, and determinants of switching to second-line ART regimes in programs that use viral load versus CD4 monitoring.


The Antiretroviral Therapy in Lower Income Countries (ART-LINC) is a collaboration of the International Epidemiological Databases to Evaluate AIDS (IeDEA). For this study, data from 17 HIV treatment programs from 14 countries in Africa, South America, and southern and Southeast Asia were evaluated.

Study Design

Longitudinal multi-cohort study


All patients who were ≥16 years of age, were ART naïve, and had at least six months of follow-up were evaluated. Sites considered to have routine viral-load monitoring (VLM) were those that provided at least one viral-load measurement between three and nine months after ART initiation for at least 50% of patients. All sites with VLM also had immunologic (CD4) monitoring.


Patients' characteristics and time to and rates of switching to second-line therapy were compared between sites with and without VLM. Time to switching was measured from six months after treatment was started. Maximum follow-up was 42 months. Among patients who switched regimens, whether immunological or virologic failure occurred in the three months prior to switching was determined. Immunologic failure was defined as a drop in CD4 count to baseline or below, or persistent CD4 count <100 cell/µL after six months of treatment. Virologic failure was defined as a plasma HIV viral load of >10,000 copies/mL.


Data from 20,113 patients were analyzed; 6369 (32%) were from programs that had routine VLM; 16,591 (83%) patients were from programs in sub-Saharan Africa. Median CD4 cell counts at the start of ART were lower at programs with VLM than without VLM (97 cells/µL vs. 129 cells/µL, respectively; P<0.001). The most common first-line regimen was 3TC/d4T/EFV (55% -71%). At VLM sites, a viral load six months after starting treatment was performed on most (58%-99%) but not all patients. A total of 576 patients (3%) switched to second-line drugs (236 at VLM sites and 340 at non-VLM sites); the majority (55%) changed both NRTIs. Only 402/576 (69.8%) patients had a CD4 count at the time of switching. Among those who did, the median CD4 count at the time of regimen change was higher (161 cells/µL vs. 102 cells/µL; P<0.001) at VLM compared to non-VLM programs. The rate of switching also was greater (3.2 per 100 person-years [p100py] vs. 2.0 p100py; P<0.001], and the median time to switching was earlier (16.3 months vs. 21.8 months; P<0.001). Reasons for switching were available for 241 (42%) of patients and included treatment failure (74%) and toxicity (10%). Among 399 patients, 263 (66%) had immunological, failure, virologic failure, or both. Independent predictors of switching included being female, being on ART for a longer period of time, and lower baseline CD4 count. Clinical stage, age, drug regimen used, and number of anti-retroviral drugs available at the program were not predictors of switching.


Patients at sites with compared with no available VLM tend to switch regimens earlier and at higher CD4 counts. Other studies are needed to determine the consequences of early or late switching on clinical outcomes.

Quality Rating

This was a good quality study. Sites participating in the ART-LINC collaboration are not necessarily representative of all sites providing ART because they have electronic medical record systems and access to CD4 cell counts and to second-line regimens. The study was unable to or did not report on some key issues, which limited interpretation of the findings. For example, there was no information on viral or immunologic failure for 30% of patients although their regimens were switched; any cases of inappropriate switching were not identified. Stated reasons for switching were unavailable for more than half the patients. It was not reported whether a correlation existed between reported failure as a reason for switching and documented treatment failure (either immunological or virologic). In other words, the question remains whether some patients switched unnecessarily or merited switching but did not have a regimen change.

There were no data on mortality of patients, clinical progression, or loss to follow-up.

Programmatic implications

Although CD4 counts at the time of regimen switches are higher at sites that have VLM, whether these differences result in better clinical outcomes was not addressed by this evaluation. In contrast, other studies seem to indicate that adult patients who are managed with CD4 cell count monitoring have the same survival rates as those managed with viral-load measurements, at least during the first year of treatment.(1, 2) A recent article reviewed in this series, however, found that among pediatric ART patients, viral failure and multiple resistance mutations occurred early in treatment, and recommended VLM, if possible.(3) In this study, only regimen switching, not single-drug switches, were evaluated; it is possible that single-drug switching could lead to higher levels of drug resistance. Information on drug resistance from resource-poor settings is limited, but what data are available indicate low levels of drug resistance so far.(4) Further studies and longer follow-ups are required to determine whether routine viral load monitoring would significantly improve patient outcomes, and whether it is cost effective.


  1. Braitstein P, Brinkhof MW, Dabis F, et al. Mortality of HIV-1-infected patients in the first year of antiretroviral therapy: comparison between low income and high-income countries. Lancet 2006; 367:817-24.
  2. Couthino R, Mermin J, Ekwaru J, et al. Utility of routine viral load, CD4 cell count, and clinical monitoring among HIV-infected adults in Uganda: a randomized trial [abstract]. In: Proceedings of the 15th Conference on Retroviruses and Opportunistic Infections; February 3-6 2008. Boston; 2008; abstract 125.
  3. Jittamala P, Puthanakit T, Chaiinseeard S. Predictors of virologic failure and genotypic resistance mutation patterns in Thai children receiving non-nucleoside reverse transcriptase inhibitor-based antiretroviral therapy. Pediatr Infect Dis J. 2009 Sep;28(9):826-30.
  4. Gupta RK, Pillay D. HIV resistance and the developing world. Int J Antimicrob Agents 2007; 29:510-17.