Efficacy of a coordinated strategy for containment of multidrug-resistant Gram-negative bacteria carriage in a Neonatal Intensive Care Unit in the context of an active surveillance program
Abstract Background: Antimicrobial resistance in Neonatal Intensive Care Unit (NICU) patients is a threat, due to the large use of antimicrobial treatment and invasive devices in fragile babies.Since 2014 an active surveillance program of multidrug-resistant Gram-negative bacteria (MDR-GNB) carriage is in place in the five NICUs of Palermo, Italy. In 2017 an increase in the prevalence of MDR-GNB and in particular of extended-spectrum β-lactamases-producing Klebsiella pneumoniae (ESBL-KP) was observed in “Civico” hospital NICU. Aim: To estimate the impact of a coordinated intervention strategy in achieving long-lasting reduction of MDR-GNB prevalence in the NICU.Methods: Rectal swabs were obtained monthly and processed to detect MDR-GNB using standard methods. MDR-GNB were characterized by pulsed-field gel electrophoresis (PFGE). From November 2017 the following intervention measures were applied: a) two-months strengthening of sample collection; b) stakeholders’ meetings; c) improvement of prevention measures and antimicrobial policy. Prevalence of MDR-GNB carriage observed in the 12 months before and in the 24 months after intervention were compared by chi-square test. Risk factors for MDR-GNB carriage in a subgroup of patients were identified by a multivariate logistic regression model. Findings: During the strengthened microbiological surveillance MDR-GNB and ESBL-KP were detected in rectal swabs (34.8%; 23.2%), nasal swabs (24.6%; 14.5%), oral swabs (14.5%; 5.4%), milk samples (32.1%; 17.9%), soother swabs (30.8%; 17.9%) and from a sub-intensive room surface. Thirteen ESBL-KP strains isolated from clinical and environmental samples showed identical PFGE patterns. ESBL-KP was detected no more until June 2018. No MDR-GNB isolate was detected for three months. Prevalence of MDR-GNB and ESBL-KP carriage after intervention significantly decreased compared to the previous year (20.6% vs 62.2 %; p<0.001 and 11.1% vs 57.8%; p<0.001). Multivariate analysis of principal exposure variables showed that admission in post-intervention period significantly reduced the risk of MDR-GNB carriage (OR=0.15, p=0.01). Conclusions: MDR-GNB broadly circulate in NICU setting and can colonize different body sites and spread by various vehicles. A coordinated strategy of multiple interventions with active cooperation between epidemiologists and clinicians in the NICU can effectively reduce their circulation and in particular the carriage of most dangerous ESBL-KP strains.