The objective of this prospective pilot study was to evaluate the response of HIV-infected patients with asymptomatic syphilis to one of two intensive antibiotic treatment regimens. Thirty-one HIV-infected patients with serum rapid plasma reagin titre > or =1:4 and no clinical findings of syphilis were randomized to receive daily intramuscular injections of ceftriaxone or procaine penicillin (plus oral probenecid) for 15 days; 24 returned for follow-up study. Seven of 10 (70%) procaine penicillin-treated patients and 10 of 14 (71%) ceftriaxone-treated patients had a > or =4-fold decline in RPR (P=0.94); two penicillin-treated and one ceftriaxone-treated patient relapsed. Two patients failed ceftriaxone therapy. Three penicillin-treated, and two ceftriaxone-treated patients were serofast. Serological responses were similar in those patients with and without asymptomatic neurosyphilis. There was no difference in the serologic response to daily treatment with ceftriaxone vs that with procaine penicillin plus probenecid; both treatments were associated with comparatively high rates of serological non-response and relapse.