研究目的：輪狀病毒和諾羅病毒是引起兒童急性病毒性胃腸炎最常見的病原。本研究旨在探討住院兒童的輪狀病毒和諾羅病毒性胃腸炎其血清介白素-6 (IL-6)和介白素-10 (IL-10)的之濃度與其臨床意義。
研究方法：本前瞻性研究共納入168位因急性胃腸炎至本校附設醫院就醫並住院接受治療之兒童。其中發生腹瀉未達72小時者收集其周邊血液檢測血清IL-6和IL-10的含量。臨床試驗的診斷力估計採用ROC曲線(Receiver Operating Characteristic Curve)分析。二元羅吉斯迴歸模型(Binary Logistic Regression Model)進行試驗的迴歸分析與預測。
研究結果：共計輪狀病毒感染30人、諾羅病毒感染25人、沙門氏菌感染26人組和11名健康對照組進行血清IL-6和IL-10測定。輪狀病毒感染者與諾羅病毒感染者相較之下有顯著高的疾病嚴重程度與血清IL-10的含量。迴歸分析顯示，絕對嗜中性白血球(absolute neutrophil count; ANC)數與最高體溫為區別輪狀病毒和諾羅病毒性胃腸炎的臨床預測因子。ROC曲線評估此迴歸模型的準確性所得曲線下的面積(Area Under Curve; AUC)為0.847(95％CI：0.741–0.952，p < 0.001)。
結?與建議：血清IL-10具有鑑別輪狀病毒和諾羅病毒感染之能力。臨床上，將ANC與最高體溫合併之預測模型運用在兒童之病毒性胃腸炎，可以幫助兒科醫師來區分輪狀病毒和諾羅病毒感染。
Objectives: This study aimed to explore the association of serum interleukin-6 (IL-6) and interleukin-10 (IL-10) levels and the clinical features in children hospitalized with rotavirus and norovirus gastroenteritis.
Methods and Materials: This prospective study enrolled 168 acute gastroenteritis patients admitted to a tertiary care center. Peripheral blood samples were collected for IL-6 and IL-10 assays within the first 72 hours of illness. The diagnostic performance of clinical tests was evaluated using the receiver operating characteristic (ROC) analysis. Binary logistic regression modeling was performed to examine the predictive variables.
Results: Serum IL-6 and IL-10 were measured in children with rotavirus infection (n = 30), norovirus infection (n = 25), Salmonella infection (n = 26), and in 11 healthy controls. There were significant higher degrees of severity of illness and levels of IL-10 in the rotavirus group as compared with the norovirus group. The binary logistic regression analysis revealed that both the ANC and maximum body temperature were significant clinical predictors for discriminating rotavirus and norovirus gastroenteritis. The ROC curve to evaluate the accuracy of logistic regression model had an AUC of 0.847 (95% CI: 0.741–0.952, p < 0.001).
Conclusions and Suggestions: IL-10 shows a significant discriminating ability between rotavirus and norovirus infection. A model incorporating maximum BT and ANC can help pediatricians to distinguish between rotavirus and norovirus in children with a suspected viral gastroenteritis.
