FAILURE OF SHORT TERM ANTIMICROBIAL THERAPY IN CHILDHOOD BRUCELLOSIS

Abstract

Human brucellosis continues to pose a worldwide health problem,1 and its treatment, especially during childhood, remains controversial. In 1986 the joint FAO/WHO Expert Committee on Brucellosis recommended a combination of doxycycline/rifampin for 45 days as the treatment of choice for this disease.2 However, later studies demonstrated that this therapeutic regimen is less effective than the standard treatment with doxycycline-streptomycin.3-6 During childhood this option presents important problems because tetracyclines are not recommended for children younger than 8 years of age to avoid damage to the tooth enamel. Streptomycin can cause ear and kidney damage and requires intramuscular administration, best avoided in children. These problems, together with the belief that brucellosis has a better prognosis in children than in adults, have led to the testing of short therapeutic regimens or the use of exclusively oral administration.7-9 We report our experience with a short term therapeutic regimen consisting of doxycycline/trimethoprim-sulfamethoxazole (selecting one or the other according to the patient's age) for 21 days, combined with gentamicin for 5 days. Material and methods. The study included 37 children younger than 14 years of age diagnosed at the Pediatric Infectious Diseases Unit at the Malaga Regional Hospital, Spain, or the Pediatric Departments of two dependent district hospitals, during a 2-year period (March, 1993, to February, 1995). Brucellosis was diagnosed by one of the following two criteria: either isolation of Brucella spp. in blood or any other fluid or tissue sample or a compatible clinical picture together with the presence of specific antibodies at significant titers; or seroconversion of one of the serologic tests with a 4-fold increase or more in the initial titer in two serum samples, taken within 2 or 3 weeks of each other. Significant titers were considered to be ≥1:160 for Wright's seroagglutination and ≥1:320 for the Coombs anti-Brucella test. The dosage and route of antibiotic administration were based on age: 8 years old or over (Group 1), doxycycline (4 mg/kg/day orally, divided into two doses, for 21 days) plus gentamicin (5 mg/kg/day intramuscularly, divided into two doses, for 5 days); 7 years old or less (Group 2), trimethoprimsulfamethoxazole (TMP-SMX) (8-10 mg/kg/day of the TMP component divided into two doses, for 21 days) plus gentamicin (5 mg/kg/day intramuscularly, divided into two doses, for 5 days). In children with osteoarticular complications doxycycline or TMP-SMX was maintained for 45 days. All children were enrolled in a prospective study that included a complete history and physical examination, complete blood cell count, erythrocyte sedimentation rate, blood chemistry profile, creatinine, aspartate aminotransferase, alanine aminotransferase, adenosine deaminase, C-reactive protein, Brucella serology and blood culture. Serologic tests included a Rose Bengal plate agglutination test, Wright's seroagglutination and a Coomb's test. A Brucella abortus 99 Weybridge strain suspension (BioMerieux, Charbonnaires les Bains, France) was used as antigen for serologic tests. For blood culture 5 ml of blood were inoculated in a Ruiz Castañeda biphasic medium (Materiales y Reactivos, Madrid, Spain) and incubated for 1 month. The patients were followed clinically, analytically and serologically at the end of the treatment period and after 45 days, 3 months and 6 months, or as required. Blood culture was performed only when the clinical manifestations were suggestive of a relapse. Relapse was defined as the reappearance of symptoms together with isolation of Brucella spp. or a serologic increase after the end of treatment. The study protocol was approved by the Ethics Committees of the hospitals concerned. Results. Of the 37 patients entered in the study 3 (8.1%) were excluded, 1 for failing to comply with the treatment and 2 were lost to follow-up. Of the 34 patients evaluated 33 had a primary infection and 1 had a relapse from a previous episode. Blood cultures were processed correctly in 26, of which 20 (76.9%) were positive. The remaining 14 patients were diagnosed by clinical and serologic evidence. The main clinical manifestations and the outcome of treatment in the 34 patients who completed the study are summarized in Table 1. Relapse occurred in approximately one-fourth of patients. Tolerance was good in both regimens. Abdominal pain and vomiting occurred in only 2 cases, neither of which required suspension of treatment.TABLE 1: Clinical manifestations, treatment and its outcome Discussion. To date few studies have compared the efficacy and tolerance of the various therapeutic regimens available for treatment of brucellosis in children. This is mainly because of the difficulty in collecting a sufficient number of cases from a childhood population. The general recommendations therefore are based on studies in adults, which mostly concur in recommending streptomycin (14 to 21 days) plus doxycycline (45 days) as the most effective therapeutic regimen.3-6 Based on the belief that childhood brucellosis seems to be less serious than cases in adults (a lower incidence of focal forms and an assumed lower tendency for relapse), shorter therapeutic regimens have been tested, with varying results. Lubani et al.7 studied 1100 cases treated with different antibiotic combinations for 3-, 5- and 8-week periods. Excluding monotherapy regimens all the others showed similar results irrespective of their duration, and it was concluded that 3 weeks of therapy were equally effective to longer treatment periods in uncomplicated childhood brucellosis. They recommended a combination of doxycycline or TMP-SMX for 21 days (depending on the age of the patient) plus gentamicin for 5 days, regimens administered to 38 of their patients who had no therapeutic failures or relapses. On the other hand Al-Eissa et al.8 compared different combinations of TMP-SMX or tetracyclines with streptomycin or rifampin in a sample of 102 cases and found that 3-week regimens presented an unacceptable relapse rate (12 of 14, 85.7%); they therefore recommended that treatment be prolonged to at least 45 days.5 Since 1992 we have tested the short term regimen proposed by Lubani (gentamicin for 5 days plus doxycycline or TMP-SMX for 21 days) for several reasons: (1) the shorter duration facilitates compliance; (2) it requires fewer intramuscular injections; and (3) this was the first therapeutic study to include an exclusively childhood population and cover a sufficiently large sample. The percentage of unfavorable results (therapeutic failure or relapse) in our study, 23.8 and 30% in Treatment Groups 1 and 2, respectively, contrasts with the spectacular results of Lubani et al.7 and supports Al-Eissa's8 opinion on the questionable efficacy of short term regimens in childhood brucellosis. In conclusion given the high incidence of therapeutic failures, short term regimens with gentamicin cannot be considered suitable treatment for brucellosis in children. Acknowledgments. We thank all the doctors from Malaga province for sending us the cases they diagnosed, especially the Emergency Pediatric department at the Maternity and Pediatric Hospital, and A. J. Macmichael and I. K. Johnstone for help in translating this article. Tomás Sánchez-Tamayo, M.D. Juan D. Colmenero, M.D. Francisco Martínez-Cortés, M.D. Abelardo Moreiras, M.D. Juan C. Ramos-Díaz, M.D. Francisco J García-Martín, M.D. Antonio Martínez Valverde, Ph.D. Pediatric Department; Hospital Materno-Infantil "Carlos Haya" (TST, FJGM, AMV) Infectious Diseases Unit; Internal Medicine Department; Hospital Regional "Carlos Haya" (JDC) Pediatric Section; Hospital Comarcal de Antequera (FMC, JCRD) Pediatric Section; Hospital Comarcal de Ronda (AM) Málaga, Spain