CORRELATION AMONG TOTAL LYMPHOCYTE COUNT, ABSOLUTE CD4(-1-INFECTED SOUTH-AFRICAN PATIENTS() COUNT, AND CD4(+) PERCENTAGE IN A GROUP OF HIV)

Depletion of CD4(+) T cells is one of the hallmarks of progression of HIV-1 infection. However, measurement of the CD4(+) T-cell count is ex pensive and often unavailable in less developed areas. Previous studie s have suggested that the total lymphocyte count (TLC) can be used to predict a low absolute CD4(+) T-cell count. To determine the relations hip between TLC and CD4(+) T-cell count in HIV-1-infected South Africa n patients, 2777 HIV-1-seropositive patients visiting the Immunology c linic at the Pelonomi Hospital in Bloemfontein, South Africa from Apri l 1991 to April 1997 were included in the study. In total, 3237 observ ations were used to determine sensitivity, specificity, and likelihood ratios, with 95% confidence intervals, of various cutpoints of the TL C to predict an absolute CD4(+) T-cell count of <200 cells/mm(3), CD4( +) percentage <20%, and CD4(+) percentage <15%. Spearman rank correlat ions were calculated between TLC and CD4(+) T cells, CD4(+) percentage and CD8(+) T cells, as well as between CD4(+) and CD8(+) T cells. Res ults demonstrated that a TLC of 2 x 10(9)/L or less had a sensitivity of 90.3% to detect patients with a CD4(+) T-cell count of <200 cells/m m(3), but a specificity of only 53.7%. When the TLC cutoff value was l owered, specificity increased but sensitivity decreased. For the obser vations as a group, a correlation (r = 0.704) between CD4(+) T-cell co unt and TLC was demonstrated, but if the patients were divided into th ree groups according to their CD4(+) T-cell count, this correlation we akened considerably. Therefore, although TLC shows a correlation with CD4(+) T-cell count, it is not a good predictor of the CD4(+) T-cell c ount in this population and should preferably not be used in the clini cal care of HIV/AIDS patients.