Search

Objective To investigate the clinical features and risk factors of infections caused by cefepime- and ertapenem-heteroresistant Klebsiella pneumoniae (FEP-HR-KP and ETP-HR-KP, respectively). Methods We conducted a retrospective case-control study of 108 Klebsiella pneumoniae strains isolated from the body fluids of the inpatients in our hospital from January, 2014 to December, 2015. The heteroresistance of the isolated strains was detected using Kirby-Bauer disk diffusion test (K-B test) and E-test, and the results were further confirmed by population analysis profile (PAP) method. The risk factors for infections by FEP-HR-KP and ETP-HR-KP were analyzed using Chi-square test and logistic regression analysis. Results Of the total of 108 non-duplicate Klebsiella pneumoniae strains isolated, 41 (40.0%) were heteroresistant to cefepime, 63(58.3%) were heteroresistant to ertapenem, and 38 (35.2%) showed heteroresistance to both. The annual detection rate of FEP-HR-KP increased from 41.7% (20/48) in 2014 to 51.7% (31/60) in 2015, and that of ETP-HR-KP increased from 54.1% (26/48) in 2014 to 61.7% (37/60) in 2015. The strains were isolated mainly from blood specimens (25/66, 37.9%), and more than half (47/66, 71.2%) of the strains were isolated from the department of surgery. Multivariate logistic regression analysis revealed that soft tissue infection (OR=1.92, 95%CI: 0.91~3.91, P=0.049), mechanical ventilation (OR=2.45, 95%CI: 1.31~4.47, P=0.013), and use of cephalosporins (OR=2.10, 95%CI: 1.18~3.13, P=0.020) were independent risk factors for infections by FEP-HR-KP; use of cephalosporins (OR=2.08, 95%CI: 0.96~4.20, P=0.033), use of carbapenems (OR=2.52, 95%CI: 1.08~3.86, P=0.023) and drainage tube placement (OR=2.76, 95%CI: 1.49~4.90, P=0.018) were independent risk factors for ETP-HR-KP infections. Conclusion The Klebsiella pneumoniae strains isolated in our hospital show high rates of heteroresistance to cefepime and ertapenem. Avoiding unnecessary invasive operation and optimizing the use of antibiotics can help to reduce the generation of heteroresistant strains and improve the treatment efficacy for FEP-HR-KP and ETP-HR-KP infections.

The continuous technology advance into the VDSM regime has brought the noise and signal integrity issues into the spotlight. It seems to be more important to estimate the power supply noise efficiently and accurately than other design metrics such as area, timing and power. At the same time, the capacitive and inductive effect should no longer be ignored in the noise analysis process. In the work, a GA-based statistical modeling technique is presented to find the worst-case time-domain voltage variation in VLSI power distribution network including inductive and resistive effects. And, the statistical power supply noise is also mapped into perturbation of the macroblock delay. Experimental results are put together with SPICE simulations, and show the validity of the method.