Hong Kong Journal of Orthopaedic Surgery

Volume 6 Number 1, April 2002

Intraoperative frozen section for detecting active infection

Wong YC,1 Lau WT,1 Wai YL,1 Ng WF2
1Department of Orthopaedics and Traumatology; and 2Department of Pathology, Yan Chai Hospital, Hong Kong

Presence of infection may be difficult to detect preoperatively or intraoperatively, but if the infection is left undetected, the result may be disastrous, especially in joint replacement. This is a retrospective study to evaluate the reliability in determining the presence or absence of active infection with intraoperative frozen sections during revision or conversion total joint replacement. Twenty-six sessions of frozen section have been performed in 22 patients; nine were done during revision total hip replacement, nine during conversion total hip replacement and eight during revision total knee replacement. The correlation rate between frozen section and permanent section was 0.88 to 0.90. Taking polymorph count of greater than 10 per 40 times high power field as positive frozen section result, the sensitivity was 0.71, and the specificity was 0.79. When the polymorph count of greater than five was chosen as positive, the sensitivity increased to 0.86 with the specificity remaining the same. The authors concluded that intraoperative frozen section is simple, inexpensive and useful in determining the presence or absence of active infection. A polymorph count of lower than five strongly suggests the absence of active infection. (Hong Kong Journal of Orthopaedic Surgery 2002;6(1):13-18)

Key Words : Bacterial infection; Frozen section, Hip prosthesis; Knee prosthesis; Total joint replacement


黃耀忠, 劉宏德, 韋玉良, 吳維富

在術前或術中,較難確定是否存在感染。但若未能及時發現,感染可能對手術 - 尤其全關節置換術 - 造成災難性後果。本回顧性研究評估以術中冰凍切片法決定有否活躍感染的可靠性。於22名病者中進行了26次冰凍切片取樣;常中九例正進行全髖重置術,九例全髖轉換術,八例全膝重置術。冰凍切片與最終切片的相關性為0.88-0.90。如以冰凍切片樣本40倍視野高於10個多核粒細胞為陽性,則敏感性為0.71,特異性為0.79。如以高於五個多核粒細胞數為陽性,則敏感性增加至0.86,特異性保持不變。我們總結,術中冰凍切片法是決定有否活躍感染的一種簡單,不昂貴而有效的方法。多核粒細胞數低於五則高度提示活躍感染並不存在。

Back to Previous Page