OBJECTIVE: Carbapenemase-producing carbapenem-resistantEnterobacteriaceae(CP-CRE) are extremely drug-resistant pathogens. Screening of contacts of newly identified CP-CRE patients is an important step to limit further transmission. We aimed to determine the risk factors for CP-CRE acquisition among patients exposed to a CP-CRE index patient.METHODS: A matched case-control study was performed in a tertiary care hospital in Israel. The study population was comprised of patients who underwent rectal screening for CP-CRE following close contact with a newly identified CP-CRE index patient. Cases were defined as positive tests for CP-CRE. For each case patient, 2 matched controls were randomly selected from the pool of contacts who tested negative for CP-CRE following exposure to the same index case. Bivariate and multivariate analyses were conducted using conditional logistic regression.RESULTS: In total, 53 positive contacts were identified in 40 unique investigations (896 tests performed on 735 contacts) between October 6, 2008, and June 7, 2012.bla_KPCwas the only carbapenemase identified. In multivariate analysis, risk factors for CP-CRE acquisition among contacts were (1) contact with an index patient for ≥3 days (odds ratio [OR], 9.8; 95% confidence interval [CI], 2.0–48.9), (2) mechanical ventilation (OR, 4.1; 95% CI, 1.4–11.9), and (3) carriage or infection with another multidrug-resistant organism (MDRO; OR, 2.6; 95% CI, 1.0–7.1). Among patients who received antibiotics, cephalosporins were associated with a lower risk of acquisition.CONCLUSIONS: Patient characteristics (ventilation and carriage of another MDRO) as well as duration of contact are risk factors for CP-CRE acquisition among contacts. The role of cephalosporins requires further study.Infect Control Hosp Epidemiol2016;1–7

Limited data exist on long-term consequences of bloodstream infections (BSIs). We aimed to examine incidence, 1-year mortality, and years of potential life lost (YPLL) following BSI. We estimated the relative contribution of hospital-onset BSI (HO-BSI) and antibiotic-resistant BSI to incidence, mortality and YPLL. We used data from Israel's national BSI surveillance system (covering eight sentinel bacteria, comprising 70% of all BSIs) and the national death registry. Adults with BSI between January 2018 and December 2019 were included. The outcomes were all-cause 30-day and 1-year mortality, with no adjustment for co-morbidities. We calculated the age-standardized mortality rate and YPLL using the Global Burden of Disease reference population and life expectancy tables. In total, 25,376 BSIs occurred over 2 years (mean adult population: 6,068,580). The annual incidence was 209·1 BSIs (95% CI 206·5–211·7) per 100,000 population. The case fatality rate was 25·6% (95% CI 25·0-26·2) at 30 days and 46·4% (95% CI 45·5-47·2) at 1 year. The hazard of death increased by 30% for each decade of age (HR=1·3 [95% CI 1·2-1·3]). The annual age-standardized mortality rate and YPLL per 100,000 were 50·8 (95% CI 49·7-51·9) and 1,012·6 (95% CI 986·9-1,038·3), respectively. HO-BSI (6,962 events) represented 27·4% (95% CI 26·9-28·0) of BSIs, 33·9% (95% CI 32·6-35·0) of deaths and 39·9% (95% CI 39·5-40·2) of YPLL. HO-BSI by drug-resistant bacteria (3,072 events) represented 12·1% (95% CI 11·7-12·5) of BSIs, 15·6% (95% CI 14·7-16·5) of deaths, and 18·4% (95% CI 18·1-18·7) of YPLL. One-year mortality following BSI is high. The burden of BSI is similar to that of ischemic stroke. HO-BSI and drug-resistant BSI contribute disproportionately to BSI mortality and YPLL. None.

Objective:We evaluated environmental contamination by carbapenem-resistant Acinetobacter baumannii (CRAB), the effectiveness of cleaning practices, the performance of aerosolized hydrogen-peroxide (aHP) technology, and the correlation between measures of cleaning and environmental contamination.Design:Serial testing of environmental contamination during a 7-month period.Setting:Single-patient rooms in intensive care units (ICUs) and multipatient step-up and regular rooms in internal medicine wards in a tertiary-care hospital with endemic CRAB.Methods:CRAB environmental contamination was determined semiquantitatively using sponge sampling.Results:In step-up rooms, 91% of patient units (56% of objects) were contaminated, and half of them were heavily contaminated. In regular rooms, only 21% of patient units (3% of objects) were contaminated. In ICUs, 76% of single-patient rooms (24% of objects) were contaminated. Cleaning did not reduce the number of contaminated objects or patient units in step-up rooms. After refresher training, cleaning reduced the proportion of contaminated objects by 2-fold (P = .001), but almost all patient units remained contaminated. Using aerosolized hydrogen peroxide (aHP) disinfection after discharge of a known CRAB-carrier decreased room contamination by 78%, similar to the reduction achieved by manual hypochloride cleaning. Measuring cleaning efficacy using fluorescent gel did not correlate with recovery of CRAB by sponge cultures.Conclusions:In step-up rooms, the high number of objects contaminated combined with poor efficacy of cleaning resulted in failure to eliminate CRAB in patient units. Fluorescent gel is a poor detector of CRAB contamination. The role of aHP is still unclear. However, its use in multipatient rooms is limited because it can only be used in unoccupied rooms.

Infections caused by extended-spectrum beta-lactamase-producing Klebsiella pneumoniae (ESBL-KP) are on a constant rise and are a noted cause of outbreaks in neonatal intensive care units (NICUs). We used whole genome sequencing (WGS) to investigate the epidemiology of consecutive and overlapping outbreaks caused by ESBL-KP in NICUs in three hospitals in close proximity. Clonality of 43 ESBL-KP isolates from 40 patients was determined by BOX-PCR. Short-read sequencing was performed on representative isolates from each clone. The dominant clones from each NICU were sequenced using long-read sequencing. Bioinformatics methods were used to define multilocus sequence type (MLST), analyze plasmid content, resistomes, and virulence factors. In each NICU, we found a unique dominant clone (ST985, ST37, and ST35), each belonging to a distinct sequence type (ST), as well as satellite clones. A satellite strain in NICU-2 (ST35) was the dominant strain in NICU-3, where it was isolated four weeks later, suggesting transmission. NICU-1- and NICU-2-dominant strains had bla_CTX-M-15 carried on a similar transposable element (Tn3-ISEcp1) but at different locations: on a plasmid and on the chromosome, respectively. We concluded that the overlapping ESBL-KP outbreaks were a combination of clonal transmission within NICUs, possible transposable element transmission between NICUs, and repeated importation of ESBL-KP from the community.

Management of partially-treated, community-acquired bacterial meningitis (PCBM) is commonly compromised by lack of microbiological diagnosis. We aimed to analyze the impact of FilmArray Meningitis-Encephalitis (FA-ME) PCR on the management of PCBM. Comparison of treatment variables of PCBM cases between two periods, before (6.5 years, control group) and after (2 years, study group) the application of FA-ME PCR assay. The total duration of antimicrobial treatment in the study group (n = 8) was significantly shorter than the control group (n = 23) (9.5 ± 3.7 days vs. 15.2 ± 5 days, p = 0.007). The percentage of narrow-spectrum regimens was significantly higher in the study group (78 ± 11% vs. 40 ± 9%, p = 0.03). There was a significant difference in implementation of antimicrobial chemoprophylaxis for close contacts (4/8 (50%) vs. 1/23 (4%), p = 0.01). The use of FA-ME PCR provides significant benefits in the management of PCBM by shortening duration of antibiotic treatment, increasing the use of narrow-spectrum regimens, and allowing proper administration of antimicrobial chemoprophylaxis. The study was approved and retrospectively registered by the Tel-Aviv Sourasky Medical Center ( 0378–17-TLV , 10/17/2017) and Rabin Medical Center ( 0270–18-RMC , 11/11/2018) Ethics committees and conforms to recognized standards.

Nationwide studies on hospital-onset bloodstream infections (HO-BSIs) are scarce. To describe incidence, mortality and antimicrobial resistance (AMR) of HO-BSI caused by eight sentinel bacteria in Israel, we used laboratory-based BSI surveillance data from 1 January 2018 to 31 December 2019. All hospitals reported positive blood cultures growing Escherichia coli, Klebsiella pneumoniae, Pseudomonas aeruginosa, Acinetobacter baumannii, Streptococcus pneumoniae, Staphylococcus aureus, Enterococcus faecalis and Enterococcus faecium. We calculated HO-BSI incidence and 14-day, 30-day and 1-year mortality in adults. We performed multivariable logistic regression to identify predictors of 30-day mortality. The study included 6752 HO-BSI events: K. pneumoniae (1659, 22.1%), E. coli (1491, 19.8%), S. aureus (1315, 17.5%), P. aeruginosa (1175, 15.6%), E. faecalis (778, 10.4%), A. baumannii (654, 8.7%), E. faecium (405, 5.4%) and S. pneumoniae (43, 0.6%). Overall incidence was 2.84/1000 admissions (95% CI: 2.77–2.91) and 6.88/10,000 patient-days (95% CI: 6.72–7.05). AMR isolates accounted for 44.2% of events. Fourteen-day, thirty-day and one-year mortality were 30.6% (95% CI: 28.5%–32.8%), 40.2% (95% CI: 38.2%–42.1%) and 66.5% (95% CI: 64.7%–68.3%), respectively. Organisms with highest risk for 30-day mortality (compared with E. coli) were A. baumannii (OR 2.85; 95% CI: 2.3–3.55), E. faecium (OR 2.16; 95% CI: 1.66–2.79) and S. pneumoniae (OR 2.36; 95% CI: 1.21–4.59). Mortality was higher in AMR isolates (OR 1.57; 95% CI: 1.4–1.77). This study highlights the incidence, associated high mortality and important role of antibiotic resistance in HO-BSI.