The synergistic action of hydroxyurea with some other antiretroviral drugs led us to evaluate the effect of therapy with the combination of didanosine and hydroxyurea in HIV-1-infected patients. We aimed to assess the anti-HIV activity of therapy with this combination by measuring variations in viral load and in CD4 cell counts. We also evaluated the potential side effects of this drug combination in HIV-1-positive patients with advanced disease. A total of 15 HIV-1-seropositive homosexual men with a mean baseline CD4 cell count of 149 cells/mm3 (range: 1-430 cells/mm3) were recruited to the study, and received didanosine (200 mg) plus hydroxyurea (500 mg) twice daily for 12 weeks. Ten patients were didanosine naive and five had previously received didanosine (for > 3 months). The combination therapy was well tolerated, although grade 2-3 alopecia appeared in four patients who had very low CD4 cell counts (< 50 cells/mm3). No significant variation in renal, hepatic and pancreatic functions occurred. A significant reduction in the plasma HIV-1 RNA (> 0.5 logs) was observed in seven of ten patients naive to didanosine after weeks 4 and 12 of the study; five of these patients had a decrease in plasma HIV-1 RNA of > 1.5 logs, with two having a decrease of > 2.0 logs. The viral load became undetectable (below 200 copies/ml) in three patients. The patients whose plasma HIV-1 RNA levels were not significantly reduced by the combination therapy had a higher baseline viral load. CD4 cell counts did not increase significantly in most patients. We observed a better response in those patients who had virus of the non-syncytium-inducing phenotype. In conclusion, hydroxyurea in combination with didanosine was well tolerated and led to a reduction in viral load mainly in patients who were initially naive to didanosine.