To the Editor—Gumá et al.[1] reported interesting findings regarding the influence of human cytomegalovirus (HCMV) infection on the distribution of NK cell subsets bearing inhibitory CD94/NKG2A and activating CD94/NKG2C C-type lectin-like receptors in HIV-1–infected individuals, compared with HIV-1–uninfected individuals. Although their results are in agreement with those of our previous article [2] showing that the proportion of CD56+ NK cells expressing NKG2C+ cells is higher overall in HIV-1–infected individuals than in HIV-1–uninfected individuals, they found no difference in the proportion of these cells between HIV-1–infected and HIV-1–uninfected individuals who were seropositive for anti-HCMV IgG [1]. They concluded that HCMV infection is the major determining factor influencing the distribution of these NK cell subsets in both HIV-1–infected and–uninfected individuals. Although we are in broad agreement with these conclusions, we would like to point out additional observations from our work that are relevant to the interaction between HCMV and HIV-1 infections and their influence on NK cell subsets. The mean percentage for NKG2C+ cells and the range in the HIV-1–uninfected control cohort reported by Gumá et al. was much greater than that described in our article [1, 2]. This means that the fold increase in the proportion of NKG2C+ cells in HIV-1–infected individuals (either treatment naive or receiving highly active antiretroviral therapy [HAART] with plasma virus RNA loads! 50 copies/mL of blood [below the limit of detection]), compared with that in HIV-1–uninfected individuals, was much greater in our study. There are 2 main possibilities for this discrepancy. First, our strategy for flow-cytometric analysis differed from that of Gumá et al., who gated on CD3JCD56+ cells. Second, the higher level of NKG2C+ cells in the HIV-1–uninfected cohort of Gumá et al., compared with our control group, could have resulted from known geographical variation in exposure to HCMV. For example, the seroprevalence of HCMV in adults in Madrid has been reported to be 78% and that in London to be 48%[3, 4]. To address these questions, we used the same flow-cytometric analysis as Gumá et al., to compare the proportions of NKG2C+ and NKG2A+ cells in 48 patients (92% male) receiving HAART and whose HIV-1 loads were below the limit of detection for 11 year; we compared these patients with 20 healthy control subjects (85% male)(table 1). Comparisons were also done after individuals were grouped according HCMV serostatus using an anti-CMV IgG ELISA (Biokit). In agreement with the data of Gumá et al. and with our previous findings, we observed an overall increase in the proportion of NKG2C+ NK cells in HIV-1–infected individuals, compared with HIV-1–uninfected individuals (table 1). However, we observed a concomitant decrease in the proportion of NKG2A+ cells, as report-