艱難梭狀桿菌感染之實驗室診斷流程優化

Optimizing Workflow for Laboratory Diagnosis of Clostridioides difficile Infection

當臨床實驗室只使用傳統培養或艱難梭狀桿菌毒性分子檢測法作為艱難梭狀桿菌感染（Clostridioides difficile infection，CDI）的唯一診斷工具時，就可能會發生過度診斷的問題。本研究利用兩步驟演算法（谷胺酸去氫酶測試[Glutamate dehydrogenase test]和艱難梭狀桿菌毒素測試[C. difficile toxin A/B test]）、分子檢測法（Xpert C.difficile/Epi）與傳統培養法等三種方式的比較，藉以評估兩步驟法的效能。當三種檢驗方式結果不一致時，再進行病歷審查來判斷結果。在這項研究中，我們收集了300個布里斯托糞便量表（Bristol stool scale）分類為5-7的糞便樣本進行評估。其中培養陽性的檢體共93個（31.0%），細分後第5型的培養陽性為13個（44.8%），第6型為56個（43.4 %），第7型為24個（16.9%）。分子檢測結果陽性有87個（29.0%），而兩步驟演算法陽性結果為79個（2 6.3%）。這8個差異結果中，有6個根據病歷判斷為CDI。實驗結果發現通過兩步驟法和分子測試獲得的結果高度吻合。我們建議在檢驗之前，嚴格選擇糞便樣本搭配兩步驟法，將可減少過度診斷的機會。

Over-diagnosis of Clostridioides difficile infection(CDI)is of great concern when molecular methods were used as the only diagnostic tools. In current study, we evaluated the performance of two-step algorithm(glutamate dehydrogenase [GDH] test + Toxin A/B assay)by comparing the results with Xpert C.difficile/Epi and traditional cultures. The discrepant results were determined by reviewing medical records. In this study we collected 300 stool samples with Bristol stool scale types 5-7 for evaluation. A total of 93 samples(31.0%)were culture-positive for C. difficile with 13(44.8%)in type 5, 56(43.4%)in type 6, and 24(16.9%)in type 7. Eighty- seven samples(29.0%)were considered positive by the Xpert C.difficile/Epi test while 79(26.3%)by the two-step algorithm. Of the eight discrepant results, six cases had CDI judged from the medical records. Our results showed high agreement of the results obtained by two-step algorithm and molecular test. Strict selection of stool samples before analysis is crucial to reduce the over-diagnosis.