Extract

To theEditor—We read with interest the article by Rowley on CD4 and viral load monitoring in resource-limited settings, particularly the section on pooling specimens for human immunodeficiency virus type 1 (HIV-1) RNA quantitation [1]. This strategy has also been proposed as a way of achieving efficiency savings in HIV clinics in the United Kingdom. The published literature on this topic tends to emphasize the efficiency of pooling by assessing the number of tests saved; however, we have reservations about the validity of the pooling method and its reliability at identifying treatment failure. In one study cited by Rowley, pooling had a low sensitivity of 60% at identifying virologic failure >500 copies/mL, despite the samples having a high median viral load of 60 000 copies/mL on individual testing [2]. In another study, which also used a virologic failure threshold of 500 copies/mL, the negative predictive value for testing minipools was just 82% [3].

You do not currently have access to this article.