Evaluation of a low-cost method, the guava EasyCD4 assay, to enumerate CD4-positive lymphocyte counts in HIV-Infected patients in the United States and Uganda

Objective: To evaluate the EasyCD4 assay, a less expensive method to enumerate CD4(+) lymphocytes, in resource-limited settings. Design: Cross-sectional study conducted in the United States and Uganda. Methods: We compared CD4(+) cell counts obtained on replicate samples from HIV-infected patients by the EasyCD4 assay, a microcapillary flow-based system, and by standard flow cytometry or FACSCount with linear regression and the Bland-Altman method. Results: Two hundred eighteen samples were analyzed (77 in the United States and 141 in Uganda). In the United States, mean +/- SD CD4 was 697 +/- 438 cells/mu L by standard flow cytometry and 688 +/- 451 cells/mu L by EasyCD4. In Uganda, the mean +/- SD CD4 was 335 +/- 331 cells/mu L by FACSCOUnt and 340 +/- 327 cells/mu L by EasyCD4. The 2 methods were highly correlated (US cohort, r(2) = 0.97, slope = 1.0, intercept = - 18; Ugandan cohort, r(2) = 0.92; slope = 0.95; intercept = 23). The mean differences ill CD4 cell counts were 9.0 and -4.6 cells/mu L for the US and Ugandan cohorts, respectively, and they were not significant ill either cohort. In the Ugandan cohort, sensitivity and specificity of the EasyCD4 for CD4 below 200 cells/ mu L were 90% and 98%, respectively. Positive predictive value was 96% negative predictive value was 93%. Conclusions: Our results Suggest that EasyCD4 may be used with high positive and negative predictive value in resource-limited settings.