A survey of antibiotic prescribing practices in Australian and New Zealand intensive care units

Objective: To evaluate antibiotic prescribing practices in empirical and directed treatment of severe sepsis and septic shock in Australian and New Zealand intensive care units. Design, setting and participants: Case vignette survey of intended antibiotic prescribing for ICU patients with sepsis associated with community-acquired pneumonia (CAP), intra-abdominal infection (IAI), hospital-acquired pneumonia (HAP) or an unidentified infectious cause (UIC). Eighty-four specialists and advanced trainees working in an ICU setting in Australia and New Zealand responded to a questionnaire survey conducted between February and May 2009. Main outcome measures: Empirical and directed antibiotic therapy, including mode of administration, frequency of administration, dose and duration of therapy. Results: A total of 656 antibiotics were empirically "prescribed", including 25 unique antibiotics. Combination therapy was prescribed in 82% of cases, with dual cover for CAP and triple therapy for IAI most common. Directed single-agent cover for Pseudomonas aeruginosa in HAP and flucloxacillin monotherapy for methicillin-sensitive Staphylococcus aureus bacteraemia were prescribed in 65% and 51% of cases, respectively. Supportive gentamicin therapy was commonly recommended (32% of all cases), predominantly in the form of once-daily dosing. Daily gentamicin dosage varied from 3 to 7 mg/kg (excluding one outlier), and was largely compliant with recommendations (76% of doses being >= 5 mg/kg). Main areas of non-compliance with guidelines were provision of broader cover for resistant organisms and beta-lactam underdosing. Continuous and extended infusions were uncommon (5%). Conclusions: Antibiotic prescribing was largely appropriate, but consideration of site-specific resistance profiles and avoidance of low dosing is advocated to provide appropriate upfront cover, prevent underdosing and reduce the risk of developing resistant organisms.