RANDOMIZED COMPARISON OF AZITHROMYCIN VERSUS CEFIXIME FOR TREATMENT OF SHIGELLOSIS IN CHILDREN

To evaluate the efficacy of azithromycin for treatment of shigellosis, 75 children with dysenteric diarrhea in whom Shigella was isolated were randomized to receive either azithromycin or cefixime orally for 5 days. Treatment with azithromycin led to a significantly higher bacteriologic eradication rate (P < 0.01) and a trend toward better clinical efficacy (P = 0.1) than with cefixime therapy. It is estimated that 140 million cases of shigellosis occur annually, causing almost 600 000 deaths in children younger than 5 years old in developing countries. 1 In Paraguay Shigella spp. are the etiologic agents most commonly isolated in children with bloody diarrhea. In a previous study we isolated Shigella spp. in 40% of cases of bloody diarrhea in children younger than 2 years old. 2 Many experts believe that effective antimicrobial therapy for shigellosis, especially in the pediatric population, shortens the duration of illness, achieves a rapid bacteriologic cure and may also have a significant positive impact on growth and nutritional status of the affected children, especially in developing countries. 3, 4 Multiple resistance of Shigella spp. to commonly used antibiotics such as ampicillin and trimethoprim-sulfamethoxazole (TMP-SMX) has been reported from many countries. 5, 6 In Paraguay at least 80% of the strains are currently resistant to ampicillin and TMP-SMX. 7, 8 Although nalidixic acid can be effective in the treatment of shigellosis, its introduction has been followed by fast development of resistance. 9 Options for antimicrobial therapy for multiply resistant Shigella infections are limited actually, and current drugs are often not widely available in the developing world (as ceftriaxone) or their use in children remains controversial (as ciprofloxacin). 3, 10 Azithromycin, an azalide antibiotic, has significant activity against Gram-negative bacilli in vitro. 11 It is more potent than erythromycin against members of the Enterobacteriaceae family, including Shigella spp. The intracellular concentrations of azithromycin achieved in colonic cells and leukocytes exceed serum concentrations by 100-fold or more. 12, 13 These properties suggest that azithromycin could be an option in the treatment of enteral infections caused by Shigella spp. In this sense one study that had included only adult patients with diarrhea showed that azithromycin exhibits an acceptable clinical and bacteriologic efficacy for shigellosis. 14 The third generation cephalosporin cefixime has been reported to be effective in the treatment of shigellosis in children, 15 although one study in adults with shigellosis found cefixime to be ineffective. 16 Optimal treatment of multidrug-resistant shigellosis in children remains uncertain, given that experts are divided about the efficacy of cefixime for treatment of these infections. 17 The objectives of the present study were to evaluate the clinical and bacteriologic efficacy of azithromycin treatment of shigellosis in children and to reassess the clinical and bacteriologic efficacy of cefixime for this infection. Materials and methods. Study site. The Instituto de Medicina Tropical serves as the primary infectious diseases center in Asuncion, the capital of Paraguay, and represents a principal referral center for patients with infectious diseases in the area of the Gran Asuncion (an area with 1 500 000 inhabitants). Study design. This open study was prospective, randomized and comparative and included children ages 6 months to 5 years with dysenteric diarrhea who were referred to the Instituto de Medicina Tropical from May 1998 through January 2000. Patients coming to the hospital with diarrhea were initially screened in the admission department. One physician member of the infectious diseases staff evaluated each patient to determine eligibility for enrollment in the study. Dysenteric diarrhea was defined as two or more bloody diarrheal stools in the 24 h before enrollment and/or the presence of >20 leukocytes per high power field on microscopic examination of stool, accompanied by a temperature ≥38.5°C, abdominal pain and/or tenesmus. A proven case of shigellosis was defined by the occurrence of diarrhea and a positive stool culture for Shigella spp. Exclusion criteria included antibiotic therapy in the 72 h before enrollment, severe vomiting precluding intake of oral medications, presence of shock, abdominal distension, severe malnutrition, primary or secondary immunodeficiency and known allergy to cephalosporins and/or macrolides. Patients whose stool cultures did not yield Shigella spp. were excluded from the study. Parents of children who met the eligibility requirements were asked to have their child enrolled in the study, and if they agreed informed written consent was obtained before study entry. The study protocol was approved by the institutional review board of the Instituto de Medicina Tropical. Treatment groups and analysis. After a stool sample was obtained for culture and microscopic examination for fecal leukocytes and red blood cells, patients were randomized (according to a computer-generated randomization scheme) to receive cefixime 8 mg/kg/day (maximum, 400 mg) once a day for 5 days or azithromycin 12 mg/kg for the first day (maximum, 500 mg) and then 6 mg/kg/dose (maximum, 250 mg) for the next 4 days. Azithromycin and cefixime powder were reconstituted with sterile water (according to package guidelines) and stored at temperature <25°C until administration to the study subject. All medications were administered by the nursing staff. Stool cultures were obtained at enrollment, at Day 3 and at Day 7. Stool samples were cultured in Salmonella-Shigella agar, xylose-lysine-deoxycholate and sorbitol-MacConkey agar for identification of Shigella, Salmonella and pathogenic Escherichia coli strains by standard methods and grouped serologically by slide agglutination with specific antisera. The susceptibility of isolates of Shigella isolates to ampicillin, TMP-SMX, cefixime and azithromycin was determined by the disk diffusion method (Kirby-Bauer) according to National Committee for Clinical Laboratory Standards guidelines. 18 For azithromycin disks containing 15 μg of the drug were used. A clear zone of ≥17 mm for azithromycin was used to indicate susceptibility. 19 Evaluation and follow-up. Patients were considered evaluable if: (1) stool cultures of specimens obtained before enrollment were positive for Shigella spp.; (2) the patient completed 5 consecutive days of treatment (unless defined as clinical failure); and (3) no other antimicrobial agent was used during the study period. Clinical and bacteriologic responses to therapy were assessed. Clinical cure was defined as resolution or significant improvement of signs and symptoms within 72 h of treatment. Clinical failure was defined as the persistence of clinical signs and symptoms (bloody diarrhea, fever and/or abdominal pain) after 72 h of treatment. Bacteriologic outcome was assessed on Days 3 and 7 of treatment, according to the following definitions: (1) bacteriologic cure: culture negative for Shigella spp. in the stool sample taken at Day 3; (2) bacteriologic failure: culture positive for Shigella spp. in the stool sample taken at Day 3. Relapse was assessed during 1 week after the end of treatment. Clinical relapse was defined as a recurrence of diarrhea, with or without fever, in patients who had been considered cured or improved at the end of treatment. Bacteriologic relapse was defined as a positive stool culture result after therapy was stopped in patients who had negative culture result on Day 3. Sample size. The study was designed to detect a 30% difference in response rates between the 2 treatment groups, assuming that 80% of the subjects treated with cefixime would respond to therapy. With a type I error rate of 0.05 and a type II error rate of 80%, it was projected that 30 valuable subjects would be needed in each treatment arm. 20 Statistical analysis. The significance of differences in proportions was determined by chi square test or Fisher’s exact test when appropriate. The significance of differences in continuous variables was determined by a two sample Student’s t test. Two tailed tests were applied. Results. Characteristics of patients. From May 1998 to January 2000, we evaluated 182 children with bloody diarrhea. Seventy-five patients had Shigella spp. isolated in stool samples and were enrolled in the study: 36 in the azithromycin group; and 39 in the cefixime group. Thirteen children (6 in the azithromycin group and 7 in the cefixime group) were excluded from the analysis because they either did not return for the follow-up at 3 days (n = 10) or did not tolerate the antibiotics (2 in the azithromycin group and 1 in the cefixime group). Of the 62 evaluable patients, 30 were treated with azithromycin and 32 were treated with cefixime. The treatment groups were well-matched for clinical and laboratory features on the day of enrollment. The age of the children enrolled ranged from 6 months to 5 years. In the cefixime group 69% of the patients were male, compared with 53% of the patients who were treated with azithromycin (difference not significant). Bloody diarrhea occurred in 94% of the patients, and abdominal pain was present in 71%. Sixty-three percents of the children had a temperature of ≥38°C. Clinical outcome. Treatment was clinically successful in 28 (93%) patients treated with azithromycin and in 25 (78%) patients treated with cefixime (P = 0.1). Clinical failure was observed in 9 patients (2 in the azithromycin group and 7 in the cefixime group) at Day 3 of the study. In patients in whom therapy was determined to have failed on Day 3, the treatment was switched to nalidixic acid (n = 8) or to cefotaxime (n = 1) on Study Day 4. The diarrhea persisted after enrollment for a mean of 2.5 ± 2.3 days in the azithromycin-treated patients vs. 3.9 ± 1.9 days in the cefixime group (P < 0.1). Clinical relapse was infrequent and was observed in only 1 of the 27 patients treated with azithromycin considered as cured at the end of the treatment. Potentially drug-related adverse events occurred in 5 of 62 patients (2 and 3 patients in the azithromycin and cefixime groups, respectively): maculopapular rash, 3 (1 patient in azithromycin group and 2 in cefixime group); vomiting, 2 (1 in each group). No case was considered serious adverse effects. Bacteriologic outcome.Shigella flexneri was the most frequent serotype isolated, accounting for 73 and 87% of the strains isolated in the azithromycin and cefixime groups, respectively. The rates of resistance to ampicillin and TMP-SMX of the Shigella spp. isolates were 63 and 77%, respectively, in the azithromycin group and 91 and 94%, respectively, in the cefixime group. Overall 85 and 79% of the strains were resistant to ampicillin and TMP-SMX, respectively. All Shigella spp. isolates were susceptible to cefixime and azithromycin. Eradication of Shigella was observed on Day 3 in 19 (59%) of the children treated with cefixime and in 28 (93%) of the children treated with azithromycin (P < 0.01) (95% confidence interval for the difference, 1.73 to 69.7). In the follow-up stool cultures performed on the 7th day after treatment, 2 children in the cefixime group and 1 child in the azithromycin group had positive cultures for Shigella. Discussion. Options for antimicrobial therapy of shigellosis are limited in developing countries because of the predominance of strains that exhibit resistance to ampicillin, TMP-SMX and nalidixic acid. 5, 6 In the present study 85 and 79% of the Shigella isolates were resistant to ampicillin and TMT-SMX, respectively. Increasing antimicrobial resistance and the observation that optimal antibiotics for shigellosis must reach high intracellular concentrations have limited the therapeutic options for shigellosis. 3, 10, 14, 15 Azithromycin is active in vitro against common bacterial enteral pathogens. 11 The MIC90 of azithromycin for diarrheagenic E. coli, Campylobacter spp., Salmonella spp., Shigella spp. and Vibrio cholerae is below the concentrations that this antibiotic attains in several tissues, including the intestinal mucosa. 12 However, experience with the use of azithromycin in the treatment of enteral infections is limited. Khan et al. 14 evaluated the efficacy of azithromycin in adults with diarrhea caused by Shigella spp. and found clinical and bacteriologic efficacies of 82 and 92%, respectively. However, there were no studies published about children. In the present study done in children, a clinical success rate of 98% was observed for azithromycin compared with 78% for cefixime (P = 0.1). When bacteriologic efficacy was examined, azithromycin was clearly superior to cefixime (P < 0.01). These clinical and bacteriologic cure rates with azithromycin are similar to those of the recommended antimicrobial agents for treatment of antibiotic-susceptible shigellosis, including ampicillin, TMP-SMX and nalidixic acid. 9, 15, 21 The greater efficacy of azithromycin in treating shigellosis despite low serum concentrations probably results from the fact that azithromycin (unlike cefixime) attains high intracellular concentrations and that Shigella organisms invade colonic epithelium and multiply intracellularly. 4, 13 Although Ashkenazi et al. 15 reported a clinical efficacy of 89% with the use of cefixime for the treatment of shigellosis in children, we found that cefixime was inferior to azithromycin in terms of bacteriologic response, and we also observed a trend toward reduced clinical efficacy of this agent. This finding is not surprising because the clinical experience with the use of first and second generation cephalosporins in the treatment of shigellosis has been disappointing (despite excellent in vitro activity). 21, 22 Furthermore in the study of Ashkenazi et al. 15 the majority of the children (81%) were infected with Shigella sonnei, which usually produces a much less severe and more frequently self-limited diseases than that caused by S. flexneri, 4 which was responsible of the 87% of the infections in our patients treated with cefixime. Our results are consistent with the observations of Salam et al., 16 who in a randomized, double blind clinical trial, including only adult patients with shigellosis, found that cefixime therapy failed in 47% (7 of 15) patients. The high level of eradication of Shigella reached with azithromycin can have important implications in the epidemiologic setting. Because of the low infectious dose, shigellosis is usually spread by person-to-person transmission. 4 Treatment of Shigella diarrhea with azithromycin can reduce the infectivity, although this must be confirmed. Acknowledgments. We thank Dr. Keith English and Dr. Steven C. Buckingham, University of Tennessee Health Science Center and Le Bonheur Children’s Medical Center, Memphis, TN, for their helpful suggestions and for reviewing the manuscript; Gustavo Chamorro and Ana Campos for technical assistance; and Dr. Ivan Allende and Valeria Moreira for their help in compiling the data.