Objective:To assess the usefulness of total lymphocyte count (TLC) for monitoring HIV-infected patients receiving highly active antiretroviral therapy.Design:Observational cohort study.Methods:Correlation between difference (Δ) from baseline at week 4, 8, 12 and 48 in TLC, CD4 cell count and viral load was determined in patients initiating HAART in phase III clinical trials between 1995 and 2001 at the HIV Clinical Research Unit, Somerset Hospital, Cape Town.Results:The study included 266 patients. At weeks 4, 8, 12 and 48, median increase in TLC was 30, 52, 139 and 219 cells × 106/l, median increase in CD4 cell count was 8, 48, 88, and 145 cells × 106/l, and median decrease in viral load was −1.6, −2.2, −2.5 and −2.7 log10copies/ml, respectively. The correlation between all pairs of ΔTLC and ΔCD4 cell counts was significant (r, 0.61;P< 0.0001), but between ΔTLC and Δ viral load it was not (r, −0.014;P= 0.73). However, the correlation between median viral load reduction and median increase in both ΔCD4 cell count (r, −0.96;P< 0.0001) and ΔTLC (r, −0.89;P< 0.0001) was significant. The slope of ΔCD4 cell count was [52.493 + 0.14(ΔTLC)]. Sensitivity and specificity of an increase or decrease from baseline in TLC for similar trend in CD4 cell count during follow-up were 83.4% and 87.3% respectively.Conclusion:TLC correlated well with changes in CD4 cell count and at a group level with viral load changes. TLC may have a role in inexpensive monitoring of the immunological response to highly active antiretroviral therapy in a resource-constrained setting.