耐碳青霉烯肺炎克雷伯菌对替加环素不同体外药敏试验方法学评估

目的比较不同药敏试验方法检测替加环素对耐碳青霉烯肺炎克雷伯菌(CRKP)的敏感性。初步了解辽宁地区CRKP对替加环素的耐药情况,为临床合理用药提供依据。方法回顾性收集辽宁省内多家大型三甲医院2011年1月—2016年12月分离的CRKP共269株,采用微量肉汤稀释法、KB纸片扩散法、Vitek-2仪器法及E试验法检测替加环素对CRKP的敏感性。结果微量肉汤稀释法、E试验法、Vitek-2仪器法测定替加环素MIC50和MIC90分别为(0.5/1)、(0.25/0.5)和(0.5/4)μg/mL。按美国食品与药品监督管理局(FDA)和欧洲药敏试验委员会(EUCAST)判读标准,微量肉汤稀释法测得CRKP对替加环素的敏感率为97.4%/93.3%,中介率为2.2%/4.1%,耐药率为0.4%/2.6%。E试验法敏感率最高,为100%/98.9%,Vitek-2法耐药率最高,为8.6%/13.4%,纸片扩散法的中介率最高,为11.9%/46.5%。与微量肉汤稀释法比较,E试验法的基本一致率(EA)和分类一致率(CA)均≥90%,但存在重大误差(VME),分别为0.4%/1.9%;Vitek-2法EA仅为61.7%,CA为87.4%/72.1%,且大误差(ME)为6.3%/7.4%,无VME;纸片扩散法CA仅为85.9%/46.5%,小误差(mE)为1.9%/2.6%,ME为0.7%/5.6%。结论辽宁地区绝大多数CRKP对替加环素仍保持较高的敏感性。对于CRKP、E-试验法、Vitek-2仪器法和纸片扩散法均不适合单独检测替加环素敏感性,可考虑联合检测,若结果不一致,均应参考微量肉汤稀释法。

Evaluation of different in vitro tigecycline susceptibility tests for carbapenem-resistant Klebsiella pneumoniae

Objective To evaluate different in vitro tigecycline susceptibility test methods for carbapenem-resistant Klebsiella pneumoniae (CRKP). To understand the resistance of CRKP to tigecycline in Liaoning province, and provide evidence for rational clinical drug use. Methods A retrospective collection of 269 strains of CRKP isolated from January 2011 to December 2016 in several large hospitals in Liaoning province. Sensitivity of tigecycline to CRKP was detected by the broth microdilution method (BMD), the K-B disk diffusion method, the Vitek-2 method, and the E-test method. Results The MIC50 and MIC90 of BMD, the E-test method, and the Vitek-2 method were as follows: (0.5/1), (0.25/0.5) and (0.5/4)μg/mL. According to the US Food and Drug Administration (FDA) and the European Committee on Antimicrobial Susceptibility Testing (EUCAST), the sensitivity of CRKP to tigecycline by BMD was 97.4%/93.3%, and the intermediate rate was 2.2%/4.1%, and the resistance rate was 0.4%/2.6%. The E-test method had the highest sensitivity rate of 100%/98.9%, the Vitek-2 had the highest resistance rate of 8.6%/13.4%, and the disk diffusion method had the highest intermediate rate of 11.9%/46.5%. Compared with the BMD method,the essential agreement rates (EA) and the categorical agreement rates (CA) of the E-test method were≥90%, but there was the very major error (VME) of 0.4%/1.9%, respectively. In the Vitek-2 results, EA was only 61.7%, CA was 87.4%/72.1%, the major error (ME) was 6.3%/7.4%，and no VME was produced. The results using the disk diffusion method, CA was only 85.9%/46.5%, the minor error (mE) was 1.9%/2.6%, and the ME was 0.7%/5.6%. A total of 251 CRKP tigecycline MIC<2μg/mL was determined by the BMD method. Most of the MIC values measured by the E-test method were the same as or different from the BMD method by ±1 dilution. The MIC values measured by the Vitek-2 method were higher than those of the BMD method. 1, 2, and 3 dilutions accounted for 40.2%, 27.5% and 10.4%, respectively. Conclusion The vast majority of CRKP in Liaoning still maintains high sensitivity to tigecycline. For CRKP, the E-test method, the Vitek-2 method, and the disk diffusion method are not suitable for the detection of tigecycline sensitivity alone. Joint detection can be considered and the BMD method should be referred to when results are