Reference ranges and sources of variability of CD4 counts in HIV-seronegative women and men

BACKGROUND: CD4 lymphocyte counts are used to monitor immune status in HIV disease. An understanding of the variability of CD4 counts which occurs in the absence of HIV infection is essential to their interpretation. The sources and degree of such variability have not been extensively studied. OBJECTIVES: To establish reference ranges for CD4 counts in HIV-seronegative women and heterosexual men attending a genitourinary medicine (GUM) clinic, and to identify possible differences according to gender and cigarette smoking and, in women, any effect of the menstrual cycle, oral contraceptive use and cigarette smoking. DESIGN: Female and heterosexual male patients attending a GUM clinic and requesting an HIV-antibody test were recruited prospectively. Results from an earlier study of CD4 counts in homosexual men were available for comparison. METHODS: Lymphocyte subpopulation analysis on whole blood by flow cytometry. RESULTS: The absolute CD4 count and percentage of CD4 cells (CD4%) were significantly higher in women (n = 195) than heterosexual men (n = 91) [difference between the means 111 x 106/1 (95% CI 41, 180) and 3.1% (1.30, 4.88)]. The absolute CD4 count and CD4% were also significantly higher in smokers (n = 143) than non-smokers (n = 140) [difference 143 (79, 207) and 2.1% (0.43, 3.81)]. Reference ranges for absolute CD4 counts (geometric mean +/- 2SD) were calculated on log transformed data as follows; female smokers 490-1610, female non-smokers 430-1350, heterosexual male smokers 380-1600, heterosexual male non-smokers 330-1280. Among other variables examined, combined oral contraceptive pill use was associated with a trend towards a lower absolute CD4 count. Changes were seen in CD4% with the menstrual cycle. CD4 counts and CD4% did not differ significantly between heterosexual men and homosexual men (n = 45). CONCLUSION: There is a significant gender and smoking effect on CD4 counts. The effects of oral contraceptive use and the menstrual cycle warrant further investigation.