We randomly allocated 80 children with suspected multidrug-resistant tyhpoid fever to therapy with either cefixime or ceftriaxone. Of these, an alternative diagnosis was subsequently made in 10 children and another 10 were excluded because cultures were negative. In 9 cases the typhoidal organisms isolated were susceptible to first-line drugs. In all, 50 children were randomly allocated to receive therapy with either intravenous ceftriaxone (65 mg/kg/day once daily, Group A, n = 25) or oral cefixime (10 mg/kg/day divided every 12 hours, Group B, n = 25) for 14 days. The two groups were comparable in their clinical characteristics, duration and severity of illness at the time of admission. The time to defervescence was comparable in both groups (8.3 Φ 3.7vs.8.0 Φ 4.1 days,P= not significant). An equal number (3 in each group) failed to respond and underwent a change in therapy. Three children in Group A and one in Group B relapsed. No adverse effects were seen in either group during the course of therapy. Our data suggest that oral cefixime can be used as effectively as parenter-ally administered ceftriaxone for management of typhoid fever in children.