Comparison of fluoroquinolones with other antibiotics in the treatment of typhoid fever: results of a meta-analysis
The purpose of a meta-analysis by Pakistani scientists was to assess the existing evidence for the use of fluoroquinolones as first-line drugs for the treatment of typhoid and paratyphoid fever. The meta-analysis included randomized controlled trials, which were searched in the Cochrane Controlled Research Database, the Cochrane Infectious Diseases Group Specialized Registry, CENTRAL, Medline (1966-2007), Embase (1974-2007), LILACS (1982-2007), conference proceedings on the problems of infectious diseases and antimicrobial therapy, bibliography in articles on relevant subjects, as well as in the current register of clinical trials (November 2007).
A meta-analysis included studies comparing fluoroquinolones to chloramphenicol, cephalosporins or azithromycin in culturally proven cases of typhoid fever. The selection and assessment of the methodological quality of the publications were carried out by two independent experts. Studies involving more than 60% of children were analyzed separately from studies involving adult patients. The main outcomes evaluated were the ineffectiveness of clinical and microbiological treatment and the rate of recurrence.
The meta-analysis included 20 studies, all of the studies being small and in most cases of poor methodological quality. Only 10 studies used the technique of "allocation concealment" (masking the procedure for assigning test participants to one or the other group) and only 3 studies were "blinded".
In adult studies, fluoroquinolones did not differ statistically significantly from chloramphenicol in the frequency of clinical or microbiological treatment failures, however, the use of fluoroquinolones in 6 clinical studies with a total of 467 participants reduced the frequency of clinical relapses (odds ratio 0.14, 95% confidence interval (CI) 0.04-0.50). Azithromycin was comparable to fluoroquinolones in terms of clinical and microbiological efficacy according to 2 studies. According to 3 clinical studies, fluoroquinolones compared to ceftriaxone reduce the incidence of clinical treatment failure (OR 0.08, 95% CI 0.01-0.45, n = 120), but do not affect failure of microbiological treatment and the occurrence of relapses. Compared to cefixime, fluoroquinolones reduce the incidence of clinical ineffectiveness (OR 0.05, 95% CI 0.01-0.24, 2 clinical studies, n = 238) and relapses (OR 0.18, CI 95% 0.03-0.91, 2 clinical studies, n = 218).
In studies of children infected with strains of salmonella resistant to nalidixic acid, there were more cases of clinical ineffectiveness compared to azithromycin than to azithromycin (OS 2.67, 95% CI 1.16-6.11, one study, n = 125), however, these differences were not observed when using gatifloxacin (one study, n = 285). In children, there was no difference between fluoroquinolones and cefixime (a clinical study, n = 82).
Thus, in adult patients, fluoroquinolones are more effective than chloramphenicol in preventing the occurrence of clinical relapses of typhoid fever, but there are currently limited data on other comparison medicines, particularly in the treatment of typhoid fever in pediatrics.