We compared the therapeutic efficacy of a World Health Organization standard bicarbonate-based oral rehydration salt solution (BBORS) with a citrate-based oral rehydration solution (CBORS) in a randomized, double-blind, controlled trial in 130 dehydrated patients with cholera aged three to 82 years. On admission the 70 patients who received CBORS and the 60 who received BBORS were similar except that the serum CO2 content (mmol/liter) was significantly lower in the CBORS group (10.8 +/- 3.6 vs. 12.5 +/- 5.3). The incidence of vomiting postadmission (41% vs. 62%, respectively), the stool output during the first 24 hr (4,252 +/- 3,900 ml vs. 6,025 +/- 4,389 ml, respectively), and the time until the patients' conditions were considered normal (38.9 +/- 14.5 hr vs. 46.3 +/- 22.7 hr, respectively) were all significantly less in the CBORS group. The serum CO2 content increased more rapidly during the first 48 hr in the CBORS group (87% +/- 74% vs. 61% +/- 68% for the BBORS group); 23% of the patients receiving CBORS and 35% of the patients receiving BBORS were considered oral-therapy treatment failures. The results indicate that CBORS was superior to BBORS for rehydration and maintenance therapy of hospitalized cholera patients in Jakarta.