Idelevich Evgeny A, Grunewald Camilla M, Wüllenweber Jörg, Becker Karsten
Institute of Medical Microbiology, University Hospital Münster, Münster, Germany.
PLoS One. 2014 Dec 9;9(12):e114834. doi: 10.1371/journal.pone.0114834. eCollection 2014.
Fungaemia is associated with high mortality rates and early appropriate antifungal therapy is essential for patient management. However, classical diagnostic workflow takes up to several days due to the slow growth of yeasts. Therefore, an approach for direct species identification and direct antifungal susceptibility testing (AFST) without prior time-consuming sub-culturing of yeasts from positive blood cultures (BCs) is urgently needed. Yeast cell pellets prepared using Sepsityper kit were used for direct identification by MALDI-TOF mass spectrometry (MS) and for direct inoculation of Vitek 2 AST-YS07 card for AFST. For comparison, MALDI-TOF MS and Vitek 2 testing were performed from yeast subculture. A total of twenty four positive BCs including twelve C. glabrata, nine C. albicans, two C. dubliniensis and one C. krusei isolate were processed. Applying modified thresholds for species identification (score ≥ 1.5 with two identical consecutive propositions), 62.5% of BCs were identified by direct MALDI-TOF MS. AFST results were generated for 72.7% of BCs directly tested by Vitek 2 and for 100% of standardized suspensions from 24 h cultures. Thus, AFST comparison was possible for 70 isolate-antifungal combinations. Essential agreement (minimum inhibitory concentration difference ≤ 1 double dilution step) was 88.6%. Very major errors (VMEs) (false-susceptibility), major errors (false-resistance) and minor errors (false categorization involving intermediate result) amounted to 33.3% (of resistant isolates), 1.9% (of susceptible isolates) and 1.4% providing 90.0% categorical agreement. All VMEs were due to fluconazole or voriconazole. This direct method saved on average 23.5 h for identification and 15.1 h for AFST, compared to routine procedures. However, performance for azole susceptibility testing was suboptimal and testing from subculture remains indispensable to validate the direct finding.
真菌血症与高死亡率相关，早期进行适当的抗真菌治疗对于患者管理至关重要。然而，由于酵母生长缓慢，传统的诊断流程需要长达数天的时间。因此，迫切需要一种无需事先从阳性血培养物（BCs）中对酵母进行耗时的传代培养即可直接进行菌种鉴定和直接抗真菌药敏试验（AFST）的方法。使用Sepsityper试剂盒制备的酵母细胞沉淀用于通过基质辅助激光解吸电离飞行时间质谱（MS）进行直接鉴定，并直接接种到Vitek 2 AST-YS07卡上进行AFST。为作比较，从酵母传代培养物中进行MALDI-TOF MS和Vitek 2检测。共处理了24份阳性BCs，包括12株光滑念珠菌、9株白色念珠菌、2株都柏林念珠菌和1株克鲁斯念珠菌分离株。应用修改后的菌种鉴定阈值（得分≥1.5且有两个相同的连续判断），62.5%的BCs通过直接MALDI-TOF MS鉴定。对于直接用Vitek 2检测的BCs，72.7%获得了AFST结果，对于来自24小时培养物的标准化悬液，100%获得了AFST结果。因此，对于70种分离株-抗真菌药物组合可以进行AFST比较。基本一致率（最低抑菌浓度差异≤1个双倍稀释步长）为88.6%。非常重大错误（VMEs）（假敏感）、重大错误（假耐药）和次要错误（涉及中间结果的假分类）分别占耐药分离株的33.3%、敏感分离株的1.9%和1.4%，分类一致率为90.0%。所有VMEs均由氟康唑或伏立康唑引起。与常规程序相比，这种直接方法平均节省了23.5小时的鉴定时间和15.1小时的AFST时间。然而，唑类药敏试验的性能并不理想，传代培养物检测对于验证直接检测结果仍然不可或缺。
