Whether temporary antiretroviral treatment during primary HIV infection (PHI) lowers the viral set point or affects the subsequent CD4 count decline remains unclear. The objectives of this study were to analyze the clinical, viral, and immunological effects of temporary early HAART during PHI. This is a cohort study of patients with laboratory evidence of PHI. Independent predictors of early HAART and the viral set point were analyzed using multiple regression analysis. Plasma HIV-1 RNA (pVL) and CD4 trajectories were analyzed using linear mixed models. A total of 332 patients were included in the analysis. Sixty-four patients started HAART within 180 days of seroconversion. A higher baseline pVL was independently predictive of the start of early HAART (OR: 2.69/log10pVL, p = 0.001). Thirty-two patients who interrupted early HAART were compared with 250 patients who remained untreated for more than 180 days after seroconversion. Temporary early HAART was not significantly associated with a longer AIDS-free survival but did result in an initial, but transient lowering of the viral set point. The viral set point was initially 0.6 log copies/ml lower after interruption of early HAART (p < 0.001) and remained lower during 83 weeks of follow-up. No significant difference in the slopes of CD4 decline was detected between the groups. Temporary HAART in PHI is started more frequently in patients with a higher pVL and can transiently lower the viral set point compared to never treated patients.