Dissertation committee:
Σπυρίδων Πουρνάρας, Καθηγητής, Ιατρική Σχολή, ΕΚΠΑ
Αθανάσιος Τσακρής, Καθηγητής, Ιατρική Σχολή, ΕΚΠΑ
Γεωργία Βρυώνη, Καθηγήτρια, Ιατρική Σχολή, ΕΚΠΑ
Σωτήριος Τσιόδρας, Καθηγητής, Ιατρική Σχολή, ΕΚΠΑ
Μιχαήλ Σαμάρκος, Καθηγητής, Ιατρική Σχολή, ΕΚΠΑ
Γαρυφαλλιά Πουλάκου, Αναπληρώτρια Καθηγήτρια, Ιατρική Σχολή, ΕΚΠΑ
Δάφνη Καϊτελίδου, Καθηγήτρια, Νοσηλευτική Σχολή, ΕΚΠΑ
Summary:
The aim of the study was the evaluation of an infection control (IC) bundle on the incidence of bloodstream infections (BSIs) caused by carbapenem resistant Gram-negative bacteria (CRGN), in a tertiary care hospital. The impact of these BSIs on hospital costs, mortality and length of stay (LOS) was also estimated for Intensive Care Unit (ICU) patients.
Between January 2010 and December 2015, all BSIs episodes caused by Carbapenem-Resistant Klebsiella pneumoniae (CRKP), Carbapenem-Resistant Acinetobacter baumannii (CRAB), and Carbapenem-Resistant Pseudomonas aeruginosa (CRPA) were recorded. The incidence for all patients and separately for patients hospitalized in the ICU was recorded.
An IC bundle was implemented in January 2012 by the Infection Control Committee of the Hospital. The impact of the interventions on BSIs rates between the pre-intervention (2010–2011) and intervention (2012–2013) periods, using an interrupted time-series model was evaluated. The effect of several factors that appear to have gradually weakened the effectiveness of the interventions was also estimated.
During the study period, 351 patients with BSIs were recorded, with a total of 538 episodes; the majority (83.6%) occurred in the ICU. The BSIs incidence rate per year during 2010–2015 for all hospitalized patients was 0.91, 1.04, 0.87, 0.78, 1.05, 1.26 and for ICU patients was 21.03, 19.63, 17.32, 14.45, 22.85, 25.02 per 1000 patient-days, respectively. The main analysis was restricted to the ICU, due to the majority of BSIs. The marginally statistically significant reduction in the incidence (p=0.05), which was noted immediately after the start of the interventions, indicates an immediate short-term effectiveness of the interventions.
From 2014, while interventions were still applied, BSIs incidence was gradually increased. For this reason, several factors possibly contributing to this increase for the years 2012–2015, which was considered as the intervention/follow-up period, were evaluated using a Pearson correlation and a linear mixed effects model. The reduced compliance rate with hand hygiene (p<0.0001) and the reduced participation in educational courses (p=0.003) can be considered as the most important and influential factors for the observed increase in ICU BSIs incidence.
Subsequently, the effectiveness of strengthening the IC program by an additional IC nurse and implementation of an antibiotic stewardship (AMS) program from January 2016 was evaluated. The annual incidence of CRGN BSIs per 1000 patient-days for all patients decreased from 1.05, 1.26 to 0.93, 0.66 (p=0.0007) and for ICU patients decreased from 22.85, 25.02 to 19.59, 11.77 (p = 0.0003) between 2014, 2015 and 2016, 2017 respectively.
The impact of CRGN BSIs on length of stay, costs and mortality of ICU patients over three years (2015-2017) was also investigated by a retrospective case-control study. All patients hospitalized for ≥3 days in the ICU were included in the study. A total of 419 patients (142 cases/277 controls) were analyzed. The overall mortality was 49.3% vs 25.6%, the median LOS was 30.0 vs. 12.0 days and the total cost was 20, 359.1€ vs. 8,509.3€, respectively, for patients with/without CRGN BSIs. After adjusting for baseline demographics, underlying disease severity and patients’ specialties (medical or surgical patients), CRGN BSIs remained an independent risk factor for mortality (odds ratio 2.9; p <0.001). Additionally, CRGN BSIs seem to result in significantly prolonged LOS and extra costs per infected patient (p <0.001).
Carbapenem-resistant Gram-negative bacteria are endemic in Greek hospitals. In the tertiary general hospital where the present study was conducted, approximately 1,000 strains of CRKP, CRAB and CRPA are isolated per year from cultures of clinical specimens and active surveillance cultures, causing serious infections, particularly in the ICU. The present study highlighted the effectiveness of an IC bundle as well as the negative impact of human and material resources shortages. In addition, the burden of CRGN BSIs on hospital costs, mortality and LOS for ICU patients is underlined.
The consistent implementation of a multifaceted IC program, also considering the cost-efficacy of interventions, is crucial to control the expansion of CRGN infections. However, financial support for workforce and material sufficiency is required to ensure long-term sustainability, particularly regarding resource allocation. In endemic settings, such as Greece, multimodal strategies and strong national actions are required to avert CRGN expansion levels.
