To objectively evaluated the usefulness of the standard formulation of cefaclor (CCL) and the long-acting formulation of cefaclor (S6472) in noncatheterized complicated urinary tract infection (UTI), a double-blind comparison study was carried out using cephalexin (CEX) as a control. Patients were orally treated with either 500 mg of CCL 3 times/day, 750 mg of S6472 2 times/day, or 500 mg of CEX 4 times/day for 14 days. Overall clinical effect was evaluated on days 5 and 14 in accordance with the UTI therapeutic evaluation standard, with check for recurrence on day 21. There was no demographic difference between the groups. There was no difference in the effective rate on day 5 among the 3 treatment groups: 58.1% in S6472 group, 66.0% in CCL group and 61.9% in CEX group. Nor on day 14, was there any significant difference in the effective rate among the 3 groups: 70.8% in S6472 group, 63.4% in CCL group and 61.8% in CEX group. Stratification analyses (by UTI group, infection site, in- or out-patient, time of starting treatment, pretreatment severity of pyuria, total number of bacteria before treatment) revealed no significant difference among the 3 groups. Therapeutic effect evaluated by physicians in charge was not significantly different among the 3 groups on day 5 or day 14. In terms of overall therapeutic effect, all 3 products were very effective in patients infected with sensitive bacteria: On day 5, 85.4% in S6472 group, 84.4% in CCL group, and 83.7% in CEX group. There was no significant difference among the 3 groups on either day. The incidence of side effects was not significantly different among the 3 groups: 4 out of 129 patients treated with S6472 (3.1%), 2 of 131 treated with CCL (1.5%) and 2 of 128 with CEX (1.6%). Clinical laboratory tests revealed 4 abnormal findings in 4 patients treated with S6472, 6 findings in 4 treated with CCL, and 4 findings in 2 treated with CEX, showing no significant difference in incidence among the 3 groups. Both side effects and abnormal clinical laboratory findings were mild and reversible. Physicians in charge judged the usefulness of the 3 drugs on days 5 and 14, taking efficacy and safety into consideration. Significant difference was not observed. The presence of recurrence was examined 7 days after drug withdrawal in patients regarded as remarkable responders to 14-day treatment by overall therapeutic effect evaluation.(ABSTRACT TRUNCATED AT 400 WORDS)