Evaluation of the antibiotic prescription in a paediatric intensive care unit

The antibiotic prescription in intensive care units is frequent using often broad-spectrum antibiotics; its quality has never been evaluated in paediatric intensive care units. To describe the modalities of antibiotic prescriptions in a paediatric intensive care unit and confront them to the literature guidelines and bacteriological data. From January 1st to March 31st 2005, 52 consecutive prescriptions regarding 45 children, with a total of 47 hospitalisations were prospectively analysed. Confirmed diagnosis of bacterial infection was retained for 50 of the 52 patients: community acquired infection in 35 cases (70%) and a nosocomial infection in 15 cases. Ten children died during the antibiotic treatment (22%), with 5 deaths related to the infection (11%). Monotherapy represented 56% of the prescriptions of antibiotics. The initial antibiotic treatment was empirical in 42 of 52 cases (81%). The empirical prescriptions were documented afterward in 48% of cases. One or more microorganisms were isolated for 60% of the initial prescriptions. Misuses in antibiotic doses (in excess [10%] or by insufficiency [13%]), number of daily administration (4%), and way of administration and/or length of treatment were observed. Seventy-seven percent of the initial prescriptions seemed to be adapted to the identified or suspected bacteria, but only 63% adequate to recommendations. Almost 2/3rd of the antibiotic prescriptions were adequate to the recommendations. The implementation of standardized and specific protocols should contribute to improve the quality of these prescriptions.