The number of clinical symptoms correlated significantly with both the duration and peak values ofCD8+ lymphocytosis (P < .05 for both comparisons in the Spearman nonparametric rank correlation test). CD8+ T cell numbers were moderately elevated in 5 of II cohort participants and within a normal range in the 6 others (data not shown). Although 8 retrospectively had some evidence ofacute HIV-l infection, the number and severity of symptoms were less than in patients 1-7 and did not prompt them to seek medical attention. To investigate the activation state of the CD8+ T cell population, in patients 1-6 and 8 the CD38+/CD8+ and HLA-DR+/ CD8+ subpopulations were determined. In patients 1-5 elevated levels were observed for ~2 months after seroconversion, again most pronounced in the patients with the most severe clinical symptoms ofacute HIV-I infection. In patients 6 and 8, however, only transient and moderate increases of this subpopulation were found. Similar results were obtained with anti-HLA-DR monoclonal antibodies (data not shown).
Discussion
In this study we analyzed immunologic parameters and the biologic phenotype of HIV-l during primary HIV-l infection in 19 individuals. In agreement with other studies, we observed a transient CD4+ and CD8+ T lymphocytopenia about the time ofseroconversion in 6 of8 individuals from whom frequent PBMC samples were available directly after infection. A similar dip in T lymphocyte numbers occurred during acute influenza virus infection (unpublished data). After seroconversion in 10 of 19 individuals, CD8+ T cell numbers increased. Transient rises in numbers of CD8+ T cells and activated CD8+ T cells have also been seen in acute infection with other viruses (e.g., cytomegalovirus, Epstein-Barr virus, influenza, and rubella [unpublished data]. Therefore, the CD8+ T lymphocytosis and increased numbers of activated CD8+ T cells in HIV-1 infection most likely reflect an antiviral immune response rather than immunopathologic events typical for HIV-1 infection. In a previous longitudinal study on HIV-I-infected individuals, we observed only NSI isolates in stable asymptomatic persons. In about half of the individuals progressing to AIDS, SI isolates were detected and this detection was associated with rapid CD4+ T cell decline. Similarly, in this study, NSI isolates were found exclusively in 16 of 19 individuals with primary HIV-1 infection. In 15 of 16 of these individuals, normal CD4+ T cell numbers were observed until the end of follow-up (mean, 391 days ± 259). The occurrence of Kaposi’s sarcoma with normal CD4+ T cell numbers in two of these individuals is in agreement with our previous findings that this AIDS indicator diagnosis is observed more frequently in patients with NSI isolates. In the present study rapid CD4+ T cell decline, however, was only observed in the three persons with SI isolates, resulting in AIDS with opportunistic infections in two of them, 6 and 19 months after seroconversion.