Concordance between epidemiological evaluation of probability of transmission and whole genome sequence relatedness among hospitalized patients acquiring Klebsiella pneumoniae carbapenemase-producing Klebsiella pneumoniae.

OBJECTIVES: We aimed to evaluate the concordance between epidemiologically determined transmission and genetic linkage of Klebsiella pneumoniae carbapenemase (KPC)-producing Klebsiella pneumoniae (KPC-Kp). METHODS: We included consecutive KPC-Kp carriers between December 2016 and April 2017 in a hospital endemic for KPC-Kp. We assessed epidemiological relatedness between patients by prospective investigations by the infection control team. The probability of epidemiological relatedness was classified into four groups: no suspected transmission, low, moderate and high probability of transmission. Whole-genome sequencing of isolates was performed. Genetic linkage between KPC-Kp isolates was expressed by distance between isolates in single nucleotide polymorphisms (SNPs). We established an SNP cut-off defining a different strain based on the reconstructed phylogenetic tree. We compared the epidemiological and genetic linkage of all isolates from all patients. RESULTS: The study included 25 KPC-Kp carriers with 49 isolates. SNP variance was available for 1129 crossed patient-isolate pairs. Genomic linkage, based on a cut-off of 80 SNPs to define related isolates, was found in 115/708 (16.2%) of isolates with no transmission suspected epidemiologically, 27/319 (8.5%) of low, 11/26 (42.3%) of moderate and 64/76 (84.2%) of high epidemiological transmission risk determination (p < 0.001 for trend). Similar results and significant trends were shown on sensitivity analyses using a lower SNP cut-off (six SNPs) and patient-isolate unique pairs, analysing the first isolate from each patient. CONCLUSIONS: While significant concordance between epidemiological and genomic transmission patterns was found, epidemiological investigations of transmission are limited by the possibility of unidentified transmissions or over-estimation of associations. Genetic linkage analysis is an important aid to epidemiological transmission assessment.

The association between infection control interventions and carbapenem-resistant Enterobacteriaceae incidence in an endemic hospital.

BACKGROUND: Israel experienced a national outbreak of carbapenem-resistant Enterobacteriaceae (CRE) starting in 2006. AIM: To assess the association between infection control (IC) interventions implemented in a referral hospital in Israel and CRE incidence. METHODS: Retrospective quasi-experimental study of prospectively collected data. CRE incidence, defined as the number of patients newly acquiring CRE in surveillance or clinical samples per 100,000 hospital-days, was plotted quarterly between 2005 and 2016. IC interventions were applied at different time-points throughout this period. Data were collected on IC staffing, number of rectal surveillance cultures, and carbapenem consumption. Autocorrelated segmented linear regression analysis was used to assess the time-points at which a significant change in the CRE incidence trend occurred, and the association between the timing of IC intervention implementation and observed CRE trends was assessed. Trends between time-points were expressed as quarterly percent change (QPC) with 95% confidence intervals (CIs). FINDINGS: Between 2005 and 2008, CRE incidence increased significantly (QPC: 19.7%; CI: 11.5-28.4), reaching a peak of 186.6 new acquisitions per 100,000 hospital-days. From mid-2011 until the end of follow-up, there was a significantly decreasing incidence trend (QPC: -4.5; CI: -6.4 to -2.5). Cohorting of patients, screening of contacts and high-risk patients on admission were insufficient to control the epidemic. Improved hand hygiene compliance, cohorting with dedicated nursing staff, addition of regular screening in high-risk departments, and carbapenem restriction were required. Decreasing CRE incidence was observed with an infectious diseases/IC staffing of 1.2-1.5 per 100 beds and 20,000-36,000 yearly CRE surveillance samples. CONCLUSION: A multi-faceted hospital-wide intervention programme is required to control CRE in hospital settings.

Predominance of Gram-negative bacilli among patients with catheter-related bloodstream infections.

We evaluated changes in the epidemiology of catheter-related bloodstream infections (CRBSIs) between 1996 and 2012 in a tertiary care centre in Israel. The cohort included 1754 episodes of CRBSI. The incidence of CRBSIs decreased throughout the study period, whereas 30-day mortality following bacteraemia increased. There was a linear shift toward predominance of Gram-negative bacilli throughout the study period (p for trend<0.001). In 1996, 68% (68/100) of CRBSIs were caused by Gram-positive cocci, whereas in 2012 77.8% (28/26) were caused by Gram-negative bacilli. The shift towards Gram-negative CRBSIs and the associated mortality mandates that empirical treatment for CRBSIs be directed by local epidemiology.

Vancomycin versus cefazolin prophylaxis for cardiac surgery in the setting of a high prevalence of methicillin-resistant staphylococcal infections.

OBJECTIVE: This study was undertaken to compare the efficacy of vancomycin prophylaxis with that of cefazolin in preventing surgical site infections in a tertiary medical center with a high prevalence of methicillin-resistant staphylococcal infections. METHODS: All adult patients (> or = 18 years) scheduled for cardiac surgery requiring sternotomy were randomly assigned to receive vancomycin (1 g every 12 hours) or cefazolin (1 g every 8 hours). Prophylaxis was started during the induction of anesthesia and continued for only 24 hours. Patients were followed up for at least 30 days (1 year for those receiving a cardiac implant). Surgical site infections were stratified according to the National Nosocomial Infections Surveillance System risk index. RESULTS: Of the 885 patients included in the study, 452 received vancomycin and 433 received cefazolin. The overall surgical site infection rates were similar in the two groups (43 cases in the vancomycin group, 9.5%, vs 39 cases in the cefazolin group, 9.0%, P =.8). Superficial and deep incisional surgical site infection rates were also similar in the two groups. There was a trend toward more frequent organ-space infections and infections with beta-lactam-resistant organisms among patients receiving cefazolin, but this trend did not reach statistical significance. In contrast, surgical site infections caused by methicillin-susceptible staphylococci were significantly more common in the vancomycin group (17 cases, 3.7%, vs 6 cases, 1.3%, P =.04). The durations of postoperative hospitalization and the mortalities were similar in the two groups. CONCLUSIONS: This trial suggests that vancomycin and cefazolin have similar efficacy in preventing surgical site infections in cardiac surgery.

Device-associated, device-day infection rates in an Israeli adult general intensive care unit.

Surveillance is an essential element of hospital infection control programs. Previous studies have shown that interhospital comparison of intensive care unit (ICU) nosocomial infections (NI) may be best made by comparing ICU-type-specific, device-associated infection rates and that these adjusted rates vary by ICU type. The aim of this study was to evaluate whether significant structural improvements introduced in an adult general ICU were associated with changes in the NI rates in this unit. In addition, we compared these rates with those of ICUs reported by the National Nosocomial Infections Surveillance (NNIS) System of the Centers for Diseases Control and Prevention. During a 12-month period 337 patients were surveyed. There were 20 ventilator-associated pneumonias (VAP)/1000 ventilator (VEN)-days, 12 bloodstream infections (BSI)/1000 central vascular catheter (CVC)-days and 14 urinary tract infection (UTI)/1000 indwelling urinary catheter (IUC)-days. Structural changes and reduction in device utilization ratios were not followed by change in NI rates in this unit. VAP and BSI rates were comparable to those reported for neurosurgical and burn ICUs, respectively, in the NNIS System, despite a much higher device utilization ratios. The present study provides specific surveillance data for further interhospital comparison with similar types of ICUs.