RESUMEN
STUDY OBJECTIVE: To determine whether critically ill patients receiving extended-infusion (EI) piperacillin/tazobactam would have improved clinical outcomes compared with patients receiving intermittent infusions. DESIGN: Single-center, open-label, prospective study. SETTING: Twenty-two-bed intensive care unit (ICU) in a regional hospital in Hong Kong. PATIENTS: A total of 367 adults who had a diagnosis of either bacterial infection or neutropenic fever and had received treatment with piperacillin/tazobactam for at least 48 hours between December 1, 2013, and August 31, 2015. INTERVENTION: Patients were assigned to receive piperacillin/tazobactam as either a 4-hour EI (182 patients [EI group]) or a 30-minute intermittent infusion (185 patients [non-extended infusion (NEI) group]). MEASUREMENTS AND MAIN RESULTS: All patients were followed for at least 14 days after treatment assignment. The primary outcome was the 14-day mortality rate after initiation of piperacillin/tazobactam. Secondary outcomes included in-hospital mortality rate, time to defervescence, duration of mechanical ventilatory support, length of ICU stay, and duration of hospital stay. Both groups demonstrated similar 14-day mortality (11.5% in the EI group vs 15.7% in the NEI group, p=0.29). The mean time to defervescence was significantly reduced in the EI group (4 days in the EI group vs 6 days in the NEI group, p=0.01); no significant differences between groups were noted in the other secondary outcomes. An Acute Physiology and Chronic Health Evaluation II score of 29.5 or higher was found to strongly predict 14-day mortality (p=0.03) by Classification and Regression Tree analysis. In the post hoc analyses, a 14-day mortality benefit was demonstrated in patients in the EI group in whom infectious organisms were identified (mortality rate 9.3% in the EI group vs 22.4% in the NEI group, p=0.01) and in whom respiratory tract infection was diagnosed (mortality rate 8.9% in the EI group vs 18.7% in the NEI group, p=0.02). CONCLUSION: Both the EI and NEI groups demonstrated similar 14-day mortality. Post hoc subgroup analysis revealed a mortality benefit in patients in the EI group who had infectious organisms identified or were diagnosed with respiratory tract infections.