目的：流感嗜血杆菌是一种可以引起严重感染的条件致病菌,故快速、准确地报告其药敏试验结果尤为重要。本研究将对某商品奈瑟菌及嗜血杆菌试剂盒（NH-AST）及相应自动分析系统检测流感嗜血杆菌的药敏结果进行评估。方法：收集101株临床标本分离得到的流感嗜血杆菌,以美国临床和实验室标准协会（Clinical and Laboratory Standards Institute,CLSI）M07-A10文件中推荐的微量肉汤稀释法作为对照方法,评估过程按照ISO20776-2标准分析NH-AST试剂盒与对照方法比较的分类一致率（category agreement,CA）、基本符合率（essential agreement,EA）、较大偏差（major discrepancy,MD）和重大偏差（very major discrepancy,VDM）,并通过Kappa分析和McNemar检验分析NH-AST试剂盒与对照方法之间的一致性。结果：本研究共评估了18种抗菌药物,除美罗培南外,其余抗菌药物的药敏试验结果Kappa值均大于0.75,P值除氨苄西林外均大于0.05。CA、EA、MD和VDM总体值分别为97.80%、95.38%、2.07%、3.29%。除美罗培南、阿奇霉素和环丙沙星的EA略小于90%（分别为88.12%、84.16%、88.12%）外,其余抗菌药物全部大于90%,且全部抗菌药物CA均超过90%。结论：NH-AST商品试剂盒及其相应的检测系统简便易用,可用于流感嗜血杆菌的临床常规药敏试验。β内酰胺酶阴性但氨苄西林耐药的流感嗜血杆菌需要使用其他方法确认氨苄西林药敏结果,并慎重报告。

Objective: Haemophilus influenzae is a pathogen that can cause severe infections. Accurate determination of H. influenzae susceptibility is important to ensure treatment efficiency. This study aimed to evaluatethe performance of TDR NH-AST kit for testing the antimicrobial susceptibility of Haemophilus influenzae. Methods: A total of 101 isolates were collected from clinical laboratories. Broth microdilution method was performed following the Clinical and Laboratory Standards Institute guideline M07-A10 as the reference method. The category agreement (CA), essential agreement (EA), major discrepancy (MD) and very major discrepancy (VDM) were used to evaluate the agreement between the TDR NH-AST kit and reference method according to ISO 20776-2. Kappa value was calculated and McNemar test was performed to analyze the consistency between the two methods. Results: A total of 18 antimicrobial drugs were involved. Apart from meropenem, the Kappa values of the other antimicrobial drug susceptibility test were all higher than 0.75, and the P va-lues were all higher than 0.05 except ampicillin. The overall values of CA, EA, MD and VMD were 97.80%, 95.38%, 2.07% and 3.29%, respectively. Although EAs of meropenem, azithromycin and ciprofloxacin were slightly less than 90% (88.12%, 84.16% and 88.12%, respectively), all the other antibacterial drugs were more than 90%, and CAs of all antibacterial drugs were more than 90%. Conclusions: TDR NH-AST has a good consistency, accuracy and reliability in the detection of antimicrobial susceptibility of H. influenzae, and could meet the requirements of routine clinical antimicrobial susceptibility test of H. influenzae.