Abstract
Brucella bacteremia is not uncommon in children living in endemic areas. Reports on brucella bacteremia, however, are scarce. Its clinical features and complications are unknown. This retrospective review describes the clinical and laboratory characteristics, the relapse rate, and response to different regimens of antimicrobials in children with brucella bacteremia over a 5-year period. Antimicrobial susceptibility testing was performed on all isolates. Data on 62 children with brucella bacteremia were collected between 1996 and 2000. All isolates were of Brucella melitensis species. Most children were between five and 10 years of age; males were twice as affected as females (66% vs 34%). Fever and arthralgia were the most common presenting symptoms, 81% and 48% respectively. Fever and arthritis were the most common physical findings, 81% and 19% respectively. Forty-five (73%) patients presented within 10 days of illness onset. Brucella titers were measured in all patients; 95% had a positive titer of 1:320 or more. Resistance to co-trimoxazole (sulfamethoxazole + trimethoprim) increased from 22% in 1996 to 66% in year 2000. Rifampicin and co-trimoxazole were the most commonly used combination in 50%, rifampicin, co-trimoxazole, supplemented with gentamicin or streptomycin in 27%. The median duration of therapy was 6 weeks. The overall relapse rate was 13% (95% CI, 4.6%-21.2%) but was higher among those with symptoms lasting >10 days (P<0.001). There was a high relapse rate among patients infected with co-trimoxazole-resistant species and treated with co-trimoxazole compared to patients infected with sensitive species who also received co-trimoxazole (22% vs. 8%), but this was not statistically significant (P = 0.16). Patients with brucella bacteremia present early in their course of illness. Their clinical features, however, did not differ from brucellosis patients who did not have bacteremia. Despite the high rate of in-vitro resistance to co-trimoxazole, this did not correlate with a significant relapse rate.