Reducing Duration of Antibiotic Use for Presumed Neonatal Early-Onset Sepsis in Greek NICUs. A “Low-Hanging Fruit” Approach

Ioannis Kopsidas, Grammatiki-Christina Tsopela, Nafsika-Maria Molocha, Eleni Bouza,Elisavet Chorafa, Evangelia Chorianopoulou, Vasileios Giapros, Despoina Gkentzi, Theodoros Gkouvas, Anastasia Kapetanaki, Korina Karachristou, Georgia Karavana, Eleni Kourkouni, Georgia Kourlaba, Maria Lithoxopoulou, Vassiliki Papaevangelou, Maria Polychronaki, Emmanuel Roilides, Tania Siahanidou, Evangelia Stratiki, George A. Syrogiannopoulos, Christos Triantafyllou, Maria N. Tsolia, Emmanouela Tsouvala, Theoklis Zaoutis, Nikos Spyridis and Preventing Hospital-Acquired Infections in Greece (PHiG) Investigators

Antibiotics, 2021 Mar;Online. DOI:https://doi.org/10.3390/antibiotics10030275

Abstract

Antibiotics are commonly prescribed in Neonatal Intensive Care Units (NICU), where stewardship interventions are challenging. Lowering antibiotic consumption is desperately needed in Greece, a country with high antibiotic resistance rates. We sought to assess the effectiveness of a low-cost and -resource intervention to reduce antibiotic use in Greek NICUs implementing a “low-hanging fruit” approach. A prospective quasi-experimental study was conducted in 15/17 public NICUs in Greece (9/2016–06/2019).

The intervention selected was discontinuation of antibiotics within 5 days for neonates with gestational age ≥ 37 weeks, no documented signs or symptoms of sepsis, CRP ≤ 10 mg/L and negative cultures within 3 days of antibiotic initiation. Impact was evaluated by the percentage of discontinued regimens by day 5, length of therapy (LOT) and stay. Trends of antibiotic consumption were assessed with days of therapy (DOT) per 1000 patient-days. Overall, there was a 9% increase (p = 0.003) of antibiotic discontinuation in ≤5 days. In total, 7/13 (53.8%) units showed a ≥10% increase. Overall, 615 days on antibiotics per 1000 patients were saved. Interrupted time-series analysis established a declining trend in DOT/1000 patient-days relative to the pre-intervention trend (p = 0.002); a monthly decrease rate of 28.96 DOT/1000 patient-days (p = 0.001, 95%CI [−45.33, −12.60]). The intervention had no impact on antibiotic choice. Antibiotic use was successfully reduced in Greek NICUs using a “low-hanging fruit” approach. In resource-limited settings, similar targeted stewardship interventions can be applied.