1134

Azithromycin versus Ceftriaxone for the Treatment of Uncomplicated Typhoid Fever in Children
Robert W. Frenck, Jr.,1 Isabelle Nakhla,1 Yehia Sultan,2 Samir B. Bassily,1 Youssef F. Girgis,1 John David,1 Thomas C. Butler,4 Nabil I. Girgis,1 and Mosaad Morsy3
1

US Naval Medical Research Unit #3, 2Abbassia Fever Hospital, and 3Pzer Pharmaceuticals, Cairo, Egypt; and 4Texas Tech University School of Medicine, Lubbock

A total of 108 children aged 417 years were randomized to receive 7 days of azithromycin (10 mg/kg/day; maximum, 500 mg/day) or ceftriaxone (75 mg/kg/day; maximum, 2.5 g/day), to assess the efcacy of the agents for the treatment of uncomplicated typhoid fever. Salmonella typhi was isolated from the initial cultures of blood samples from 64 patients. A total of 31 (91%) of the 34 patients treated with azithromycin and 29 (97%) of the 30 patients treated with ceftriaxone were cured (P 1 .05). All 64 isolates were susceptible to azithromycin and ceftriaxone. Of the patients treated with ceftriaxone, 4 subsequently had relapse of their infection. No serious side effects occurred in any study subject. Oral azithromycin administered once daily appears to be effective for the treatment of uncomplicated typhoid fever in children. If these results are conrmed, the agent could be a convenient alternative for the treatment of typhoid fever, especially in individuals in developing countries where medical resources are scarce.

Downloaded from http://cid.oxfordjournals.org/ by guest on December 18, 2012

Typhoid fever, a systemic infection caused by Salmonella typhi and Salmonella paratyphi, is a common and sometimes fatal infection among children living in developing countries, especially such countries in Asia and Africa [1]. For decades, chloramphenicol has been highly effective against S. typhi and S. paratyphi, and it often remains the antibiotic of choice for the treatment of typhoid fever [2, 3]. However, the widespread emergence of multidrug-resistant S. typhi has necessitated the
Received 17 September 1999; revised 28 February 2000; electronically published 6 November 2000. Some of the authors of this manuscript are military service members or employees of the US Government. This work was prepared as part of their ofcial duties. Title 17 U.S.C. 105 provides that Copyright protection under this title is not available for any work of the United States Government. Title 18 U.S.C. 101 denes a US Government work as a work prepared by a military service member or employee of the United States Government as part of that persons ofcial duties. The study was reviewed and approved by the Egyptian Ministry of Health as well as the US Navy Bureau of Medicine and Surgery Committee for the Protection of Human Subjects. This research was conducted in compliance with all US Federal Regulations governing the protection of human subjects in research. The opinions and assertions contained herein are the private ones of the authors and are not to be construed as ofcial or as reecting the views of the US Navy Department, US Department of Defense, US Government, or Egyptian Ministry of Health. Financial support: This work was supported by Pzer Pharmaceuticals (Cairo, Egypt) and the US Naval Medical Research and Development Command, Naval Medical Center, National Capitol Region (Bethesda, MD). Correspondence: Dr. Robert W. Frenck, Jr., Commanding Ofcer, Head, Clinical Investigations, US Naval Medical Research Unit #3, PSC 452, Box 121 (Attention Code 101C), FPO AE 09835-0007 (FrenckR@namru3 .med.navy.mil). Reprints: Research Publication Branch, US Naval Medical Research Unit #3, PSC 452, Box 5000, FPO AE 09835-0007.
Clinical Infectious Diseases 2000; 31:11348 2000 by the Infectious Diseases Society of America. All rights reserved. 1058-4838/2000/3105-0004$03.00

search for other therapeutic options [4]. Fluoroquinolones have proven to be effective; however, to date, they are restricted from routine use in children, and quinolone-resistant strains of S. typhi have begun to be reported [5, 6]. Ceftriaxone, a thirdgeneration cephalosporin, is highly effective against S. typhi and has become the standard of care for the treatment of typhoid fever in many parts of the world [7]. However, because parenteral administration of ceftriaxone is required, the antibiotic is a less-than-ideal treatment alternative. The recent availability of the azalide class of antibiotics has provided another potential option for the treatment of typhoid fever. Azithromycin, the rst azalide evaluated, has in vitro activity against many enteric intracellular pathogens, including S. typhi [810]. Animal models have demonstrated that azithromycin is highly effective against both Salmonella enteritidis and Salmonella typhimurium, with drug efcacy related to the tissue concentration, rather than the serum concentration, of the antibiotic [11, 12]. Studies of human volunteers have shown that neutrophil concentrations of azithromycin are 1100 times the serum concentration of the antibiotic [13]. Five days after a 3-day course of azithromycin was completed, neutrophil concentrations of the drug still exceeded the typical MIC for S. typhi by 120 times, whereas the drug was unmeasurable in the serum [13]. These encouraging results led us to initiate a trial of azithromycin treatment in humans. Initially, azithromycin was demonstrated, in an open-labeled, nonrandomized trial, to be effective in the treatment of adults with uncomplicated typhoid fever [14]. A subsequent randomized trial demonstrated that azithromycin was as effective as ciprooxacin for the treatment of uncomplicated typhoid fever in adults [15]. The results of

CID 2000;31 (November)

Azithromycin vs. Ceftriaxone for Typhoid Fever

1135

these studies prompted the present study of azithromycin suspension versus ceftriaxone for the treatment of uncomplicated typhoid fever in children.

Patients and Methods

Study site. The Abbassia Fever Hospital is a 1500-bed infectious disease hospital that serves as both the primary infectious disease hospital in Cairo and a referral center for patients with infectious diseases from throughout Egypt. Study population. Patients coming to the hospital were initially screened in the reception department. During the study period, all children of 417 years of age who, according to the reception department physician, had a clinical diagnosis of typhoid fever were admitted to a single ward in the hospital. After admission to the ward, a study physician reevaluated the patients to determine whether they were eligible for enrollment in the study. Eligibility for enrollment required that a subject have a documented fever (temperature, 38.5 C) and a history of fever for at least 4 days plus at least 2 of the following criteria: abdominal tenderness, hepatomegaly, splenomegaly, and/or rose spots. Subjects with the following conditions were excluded from the study: allergy to ceftriaxone or erythromycin (or to other macrolides), major complications of typhoid fever (e.g., pneumonia, intestinal hemorrhage or perforation, shock, or coma), inability to swallow oral medication, signicant underlying illness (e.g., heart disease, asthma requiring chronic medications, or immunodeciencies), or treatment within the past 4 days with either study medication or chloramphenicol, trimethoprim-sulfamethoxazole (TMP-SMZ), or ampicillin. Subjects who might be pregnant or who were lactating were also excluded from the study. Parents of children meeting eligibility requirements were asked to have their child enroll in the study, and if they agreed, informed consent was obtained before randomization of the study drug. Sample size requirements. The study was designed to detect a 50% difference in clinical cure rates between the 2 treatment groups, under the assumption that 80% of the subjects treated with ceftriaxone would respond to therapy, with response dened as the patient becoming afebrile within 5 days of starting treatment. Using a type I error rate of 0.05 and a type II error rate of 0.2, it was projected that 30 evaluable subjects (those with positive blood cultures) would be needed in each treatment arm [16]. Historically, 50% of patients who are thought, as a result of ndings of clinical examination, to have typhoid fever have S. typhi or S. paratyphi isolated from their blood or stool. Thus, enrollment of 60 subjects in each treatment arm was planned for the trial. Randomization and treatment. To control for age bias, patients were stratied into 3 age groups (47 years, 812 years, and 1317 years) before randomization. After study eligibility was determined and informed consent was obtained, patients were randomized to 1 of the 2 treatment groups. Before randomization, sequentially numbered sealed envelopes containing the name of the study agent to be used were created by block randomization that was based on a random list of numbers. At the time of randomization, neither the subject nor the study physician was aware of which study agent would be administered. Treatment assignments were made at en-

rollment by opening the lowest numbered envelope that had yet to be used for the age group of the child. After assignment, subjects were treated, in an open-label format, with either oral azithromycin suspension (10 mg/kg/day; maximum dose, 500 mg/day) administered once daily for 7 days or im ceftriaxone (75 mg/kg/day; maximum dose, 2.5 g/day) administered once daily for 7 days. Azithromycin powder was reconstituted with sterile water (according to package guidelines) in a bottle, and the bottle was labeled with the unique identication number of the study subject. After reconstitution, azithromycin was stored at room temperature until it was administered to the study subject. Ceftriaxone powder, which was mixed with 1% lidocaine solution provided by the manufacturer, was administered immediately after reconstitution by means of deep im injection. All medications were administered in the hospital by the nursing staff. Procedures. Patients enrolled in the study were hospitalized for the entire treatment period and for 3 days after therapy was completed. Before initiation of antibiotic therapy, samples of blood, stool, and urine were cultured. Repeated blood cultures were performed for all subjects 4 and 10 days after treatment was initiated. Repeated stool and urine cultures were performed on day 10 if the initial culture result was positive. All subjects had a stool culture performed 1 month after completion of therapy. A blood sample was obtained for determination of the complete blood cell count and for serum chemistry analysis at baseline and on day 10. Cultures of blood and stool were processed by means of standard clinical methods. All blood cultures were blindly subcultured after 1, 7, and 14 days of incubation. Subcultures were performed at other times if the broth appeared cloudy. Identication of Salmonella isolates was performed by use of standard methods [17]. Stool specimens were plated on MacConkey agar and SalmonellaShigella agar, and colonies suggestive of Salmonella species were further evaluated by means of standard methods to determine exact identication [17]. All isolates of S. typhi or S. paratyphi were tested for susceptibility to azithromycin, ceftriaxone, ciprooxacin, chloramphenicol, ampicillin, and TMP-SMZ by means of agar disk diffusion analysis done according to the method of Bauer et al. [18]. During the period of hospitalization, vital signs (including body temperature) were measured every 8 h, and clinical examination was performed daily. During clinical examination, the general condition of the patient as well as the presence of coated tongue, abdominal tenderness, splenomegaly, hepatomegaly, or rash were specically noted. In addition, a structured questionnaire was administered daily during hospitalization to record symptomatology (including fever, headache, rash, abdominal pain, constipation, diarrhea, and/or anorexia) and possible adverse events. Subjects were routinely checked at 2 and 4 weeks after completion of treatment and were asked to return immediately if they became sick before a scheduled follow-up visit. Data analysis. Responses to treatment were classied as clinical cure or failure and as microbiological cure or failure. Clinical cure was dened as resolution of all typhoid-related symptoms or signs by the end of 7 days of therapy. For response to therapy, fever was dened as 1 rectal temperature 138.4 C during a 24h period. Clinical failure was dened either as persistence of 1 typhoid-related symptoms or signs present at study entry, or as

Downloaded from http://cid.oxfordjournals.org/ by guest on December 18, 2012

1136

Frenck et al.

CID 2000;31 (November)

development of a typhoid-related complication (including pneumonia, intestinal hemorrhage or perforation, shock, or coma) after at least 4 days of therapy. Microbiological cure was dened as a sterile blood culture on days 4 and 10 of therapy. Relapse was dened as recurrence of fever with signs or symptoms of typhoid fever within 4 weeks of completion of therapy along with isolation of S. typhi or S. paratyphi from the blood. The x2 test was used to determine signicant differences in cure rates between groups. For groups with !5 events in a cell, Fishers exact test was used.

Results A total of 108 patients (53 males and 55 females; mean age SD, 9.8 2.8 years) were enrolled in the study and assigned to a treatment arm. Cultures of blood specimens obtained from 64 children (34 azithromycin recipients and 30 ceftriaxone recipients) were positive for S. typhi; these subjects were included in the analysis. Demographic and pretreatment laboratory evaluation of the subjects demonstrated that there were no signicant differences between the treatment groups (table 1). Antimicrobial susceptibility testing showed that all 64 blood culture isolates of S. typhi were susceptible to azithromycin, ceftriaxone, and ciprooxacin. Responses to treatment were excellent in both groups (table 2). Patients responded quickly to therapy; the mean time to defervescence SD was 4.1 1.1 days and 3.9 1.0 days for azithromycin recipients and ceftriaxone recipients, respectively (P p NS). Clinical cure occurred in 31 (91%) of 34 patients treated with azithromycin, compared with 29 (97%) of 30 pa-

tients treated with ceftriaxone (P p NS ). Of the 3 patients with clinical failure in the azithromycin group, 2 had failure as a result of slow resolution of fever without other symptoms. All 3 subjects received ceftriaxone for an additional 57 days after they had received azithromycin for 7 days; all 3 had complete cures without any signicant consequences. Microbiological cure occurred in 33 (97%) of 34 patients treated with azithromycin versus 29 (97%) of 30 patients treated with ceftriaxone (P p NS). The patient with microbiological failure in the azithromycin group had S. typhi isolated from blood on both day 4 and day 10, whereas the patient with microbiological failure in the ceftriaxone group had S. typhi isolated from blood on day 10. Both patients with microbiological failure also had clinical failure. Another course of antibiotics (ceftriaxone for the subject who initially received azithromycin and chloramphenicol for the subject who initially received ceftriaxone) was administered to both subjects after they had received initial therapy for 7 days. For both subjects, the second course of antibiotics was clinically successful. After discharge from the hospital, 58 subjects (30 of 34 azithromycin recipients and 28 of 30 ceftriaxone recipients) who had S. typhi initially isolated from their blood returned for posttreatment follow-up evaluation. Six of the 30 subjects treated with ceftriaxone returned before the scheduled 1-month followup evaluation because of recurrence of typhoid-related symptoms; S. typhi was isolated from blood specimens from 4 of these patients. All 6 subjects were treated with a second course of antibiotics that resulted in resolution of their symptoms and sterile posttreatment blood cultures. No relapses occurred in

Downloaded from http://cid.oxfordjournals.org/ by guest on December 18, 2012

Table 1. Admission characteristics of azithromycin and ceftriaxone recipients who had typhoid fever and for whom blood cultures were positive.
Azithromycin recipients (n p 34) 9.7 (517) 20 9.7 (330) 34 0 5 7 0 Ceftriaxone recipients (n p 30) 10.1 (514) 17 9.2 (315) 30 0 6 3 0

Characteristic Age, mean y (range) Sex, male Duration of fever before admission, mean d (range) Blood culture result Salmonella typhi Salmonella paratyphi Blood culture that yielded MDR S. typhi Stool culture result S. typhi S. paratyphi Laboratory test result, mean SD (normal range) Hemoglobin level, g/dL (1118) 103) WBC count, cells/mm3 (4.510.5 Platelet count, cells/mm3 (150,000350,000) Total bilirubin level, mg/dL (0.21.0) AST level, U/L (033) Blood urea nitrogen level, mg/dL (718) Serum creatinine level, mg/dL (0.71.5)

10.4 6.2 212,000 0.4 87 10.3 0.7

1.2 2.3 75,000 0.1 48 3.1 0.1

10.8 6.3 211,000 0.5 83 11.4 0.6

1.3 1.6 96,000 0.1 70 6 0.2

NOTE. Data are no. of patients, unless otherwise indicated. AST, aspartate aminotransferase; MDR, multidrug resistant.

CID 2000;31 (November)

Azithromycin vs. Ceftriaxone for Typhoid Fever

1137

Table 2. Responses to treatment with azithromycin or ceftriaxone among patients who had typhoid fever and for whom blood cultures were positive.
Azithromycin recipients (n p 34) 31 (91) 4.1 1.1 33 (97) 1 a 1 0 0
a

Finding Clinical cure by day 7, no. (%) Duration of fever after starting therapy, mean d SD Microbiological cure, no. (%) Blood culture that yielded Salmonella on Day 4 Day 10 Stool culture that yielded Salmonella on day 10 Relapse Laboratory test result on day 10, mean SD (normal ranges) Hemoglobin level, g/dL (1118) 103) WBC count, cells/mm3 (4.510.5 Platelet count, cells/mm3 (150,000350,000) Total bilirubin level, mg/dL (0.21.0) AST level, U/L (033) Blood urea nitrogen level, mg/dL (718) Serum creatinine level, mg/dL (0.71.5)

Ceftriaxone recipients (n p 30) 29 (97) 3.9 1.0 29 (97) 0 1 0 4

9.9 6.3 434,000 0.3 50 8.9 0.7

1.4 1.6 122,000 0.1 30 3.5 0.1

10.4 7.4 431,000 0.4 62 9.5 0.6

1.0 2.2 164,000 0.2 34 2.4 0.2

Downloaded from http://cid.oxfordjournals.org/ by guest on December 18, 2012

NOTE. Data are no. of patients, unless otherwise indicated. AST, aspartate aminotransferase. a Same patient.

the group treated with azithromycin. With the exception of the 6 subjects described above, all patients were clinically well when they returned for evaluation 1 month after completion of therapy. Antimicrobial susceptibility testing demonstrated that all 8 of the S. typhi isolates from children who had relapses or for whom treatment failed remained susceptible to all antibiotics tested. No subject had a serious adverse event. Gastrointestinal symptoms were commonly reported by both groups. Vomiting occurred more frequently in subjects treated with azithromycin than in those treated with ceftriaxone; however, the symptom was mild and transient, resolving, in most cases, within 1 day of initiation of treatment with azithromycin. In no case was the symptom severe enough to require treatment or alteration of antibiotic therapy. Pain at the site of injection, the most common adverse event in the ceftriaxone group, was typically mild; however, 6 subjects complained of pain up to 24 hours after injection, despite both the mixture of lidocaine with ceftriaxone before injection and the rotation of injection sites. With the exception of 1 patient from each group who had a mild, residual elevation of alanine aminotransferase concentration, all subjects with abnormal results of pretreatment laboratory analysis had normal values at the end of therapy. Blood evaluation on day 10 revealed that 4 patients in the azithromycin group and 3 patients in the ceftriaxone group had thrombocytosis (platelet count, 1500,000 cells/mm3), but all subjects were asymptomatic. In addition, 4 subjects treated with ceftriaxone and 2 subjects treated with azithromycin developed mild, asymptomatic elevations in aspartate aminotransferase

concentrations (range, 110170 U/L) during the course of therapy.

Discussion In a comparative, randomized trial, we demonstrated that azithromycin is highly effective for the treatment of uncomplicated typhoid fever in children. In the present study, ndings of clinical cure rates 190% and microbiological cure rates 195% for subjects receiving either azithromycin or ceftriaxone compare favorably with ndings from past trials of standard antibiotics for the treatment of typhoid fever [2, 6, 1921]. It is interesting that, in the present study, 14% of subjects with bacteremia who were treated with ceftriaxone had relapses of infection within 1 month of completion of therapy. These data are consistent with relapse rates of 5%15% in other trials of ceftriaxone for the treatment of typhoid fever [2, 7, 19, 22]. Although our sample size was small, no subject who was treated with azithromycin had a relapse; this nding may be attributable to the long half-life of azithromycin within the intracellular compartment, with eradication of residual organisms after completion of therapy, as well as to its elevated concentration within the biliary system. The extremely long half-life of azithromycin in tissue, in conjunction with the success of the drug in the present trial, may warrant trials in which shorter courses of this agent are used for the treatment of typhoid fever, as have been successfully tried for quinolone antibiotics [6, 19, 23]. Clinical differences between the 2 treatment groups were few.

1138

Frenck et al.

CID 2000;31 (November)

Patients treated with ceftriaxone had a slightly shorter time to defervescence than did those treated with azithromycin (3.9 vs. 4.1 days, respectively); however, the difference was not significant, and both results were within time frames reported in previous typhoid treatment trials [3, 19, 2325]. Mild and transient gastrointestinal symptoms occurred in both treatment groups, but no adverse event was severe enough to require alteration in therapy. In conclusion, azithromycin given for 7 days at a dosage of 10 mg/kg/day (maximum dose, 500 mg/day) appears to be highly effective for the treatment of uncomplicated typhoid fever in children, with clinical cure rates comparable to those for ceftriaxone. Once-daily administration of oral azithromycin may offer a simple treatment regimen for typhoid fever caused by either susceptible or drug-resistant strains of S. typhi and may be suitable for use in areas where medical resources are limited.
References 1. Azad AK, Islam R, Salam MA, Alam AN, Butler T. Comparison of clinical features and pathologic ndings in fatal cases of typhoid fever during the initial and later stages of the disease. Am J Trop Med Hyg 1997; 56:4903. 2. Acharya G, Butler T, Ho M, et al. Treatment of typhoid fever: randomized trial of a three-day course of ceftriaxone versus a fourteen-day course of chloramphenicol. Am J Trop Med Hyg 1995; 52:1625. 3. Islam A, Butler T, Kabir I, Alam NH. Treatment of typhoid fever with ceftriaxone for 5 days or chloramphenicol for 14 days: a randomized clinical trial. Antimicrob Agents Chemother 1993; 37:15725. 4. Mirza SH, Beeching NJ, Hart CA. Multi-drug resistant typhoid fever: a global problem. J Med Microbiol 1996; 44:3179. 5. Wain J, Hoa NTT, Chinh NT, et al. Quinolone-resistant Salmonella typhi in Vietnam: molecular basis of resistance and clinical response to treatment. Clin Infect Dis 1997; 25:140410. 6. Vinh H, Wain J, Vo TN, et al. Two or three days of ooxacin treatment for uncomplicated multidrug-resistant typhoid fever in children. Antimicrob Agents Chemother 1996; 40:95861. 7. Soe GB, Overturf GD. Treatment of typhoid fever and other systemic salmonelloses with cefotaxime, ceftriaxone, cefoperazone and other newer cephalosporins. Rev Infect Dis 1987; 9:71936. 8. Jones K, Felmingham D, Ridgway G. In vitro activity of azithromycin (CP62,993), a novel macrolide, against enteric pathogens. Drugs Exp Clin Res 1988; 14:6135. 9. Metchock B. In vitro activity of azithromycin compared with other macrolides and oral antibiotics against Salmonella typhi. J Antimicrob Chemother 1990; 25(Suppl A):2931.

10. Gordillo ME, Singh KV, Murray BE. In vitro activity of azithromycin against bacterial enteric pathogens. Antimicrob Agents Chemother 1993; 37: 12035. 11. Girard AE, Girard D, Retsema JA. Correlation of the extravascular pharmacokinetics of azithromycin with in-vivo efcacy in models of localized infection. J Antimicrob Chemother 1990; 25(Suppl A):6171. 12. Butler T, Girard AE. Comparative efcacies of azithromycin and ciprooxacin against experimental Salmonella typhimurium infection in mice. J Antimicrob Chemother 1993; 31:3139. 13. Wildfeuer A, Laufen H, Zimmermann T. Distribution of orally administered azithromycin in various blood compartments. Int J Clin Pharmacol Ther 1994; 32:35660. 14. Tribble D, Girgis NI, Habib N, Butler T. Efcacy of azithromycin for typhoid fever. Clin Infect Dis 1995; 21:10456. 15. Girgis NI, Butler T, Frenck RW, Sultan Y, Brown FM, Tribble D. Azithromycin versus ciprooxacin for the treatment of uncomplicated typhoid fever that included patients with multidrug resistance. Antimicrob Agents Chemother 1999; 43:14414. 16. Young, MJ, Bresnitz EA, Strom BL. Sample size nomograms for interpreting negative clinical studies. Ann Intern Med 1983; 99:24851. 17. Gray LD. Escherichia, Salmonella, Shigella and Yersinia. In: Murray PR, Baron EJ, Pfaller MA, Tenover FC, Yolken RH, eds. Manual of clinical microbiology. 6th ed. Washington, DC: American Society for Microbiology Press, 1995:4506. 18. Bauer AW, Kirby WM, Sherris JC, Turck M. Antibiotic susceptibility testing by a standardized single disk method. Am J Clin Pathol 1966; 45:4936. 19. Smith MD, Duong NM, Hoa NT, et al. Comparison of ooxacin and ceftriaxone for short-course treatment of enteric fever. Antimicrob Agents Chemother 1994; 38:171620. 20. Wallace MR, Yousif AA, Mahroos GA, et al. Ciprooxacin versus ceftriaxone in the treatment of multiresistant typhoid fever. Eur J Clin Microbiol Infect Dis 1993; 12:90710. 21. Girgis NI, Sultan Y, Hammad O, Farid Z. Comparison of the efcacy, safety and cost of cexime, ceftriaxone and aztreonam in the treatment of multidrug resistant Salmonella typhi septicemia in children. Pediatr Infect Dis J 1995; 14:6035. 22. Lasserre R, Sangalang RP, Santiago L. Three-day treatment of typhoid fever with two different doses of ceftriaxone, compared to 14-day therapy with chloramphenicol: a randomized trial. J Antimicrob Chemother 1991; 28: 76572. 23. Cao XT, Kneen R, Nguyen TA, Truong DL, White NJ, Parry CM. A comparative study of ooxacin and cexime for treatment of typhoid fever in children. Dong Nai Pediatric Center Typhoid Study Group. Pediatr Infect Dis J 1999; 18:2458. 24. Van den Bergh ET, Gasem MH, Keuter M, Dolmans MV. Outcome in three groups of patients with typhoid fever in Indonesia between 1948 and 1990. Trop Med Int Health 1999; 4:2115. 25. Dutta P, Rasaily R, Saha MR, et al. Ciprooxacin for treatment of severe typhoid fever in children. Antimicrob Agents Chemother 1993; 37:11979.

Downloaded from http://cid.oxfordjournals.org/ by guest on December 18, 2012