Abstract：
OBJECTIVE: Identifying a predictive biomarker for spontaneous bacterial peritonitis (SBP) is of crucial importance in cirrhotic patients with ascites. This study was designed to identify the predictive value of serum or ascitic heme oxygenase-1 (HO-1) for SBP in this population.
METHODS: In this cohort study, 60 patients with liver cirrhosis and ascites accompanied by SBP (studied cohort, n=26) or not (control cohort, n=34) were retrospectively included. HO-1 levels in the serum and ascites were detected by ELISA. The predictive performance of HO-1 was evaluated by calculating the area under the curve (AUC) in receiver operating characteristic (ROC) curve analysis.
RESULTS: The HO-1 level of SBP patients was significantly higher than that of non-SBP patients both in the serum and ascites (p<0.001). Serum, ascites HO-1, and their combination displayed an AUC of 0.897 (95% CI, 0.808-0.986), 0.825 (95% CI, 0.708-0.941), and 0.902 (95% CI, 0.817-0.986) for discriminating SBP from non-SBP patients. HO-1 level between serum and ascites had a higher correlation in patients in a Child-Pugh B stage (R=0.691, p<0.001) than in the Child-Pugh C stage (R=0.475, p=0.014). The correlation between HO-1 level and inflammatory factors or biochemical parameters was observed in SBP patients and presented in a Child-Pugh stage-dependent manner.
CONCLUSIONS: HO-1 level has the ability to predict the occurrence of SBP in cirrhotic patients with ascites which has the potential to be a biomarker for the early prediction of SBP and move into the clinical setting. However, the Child-Pugh stage should be considered when using the HO-1 as a predictive biomarker.
METHODS: In this cohort study, 60 patients with liver cirrhosis and ascites accompanied by SBP (studied cohort, n=26) or not (control cohort, n=34) were retrospectively included. HO-1 levels in the serum and ascites were detected by ELISA. The predictive performance of HO-1 was evaluated by calculating the area under the curve (AUC) in receiver operating characteristic (ROC) curve analysis.
RESULTS: The HO-1 level of SBP patients was significantly higher than that of non-SBP patients both in the serum and ascites (p<0.001). Serum, ascites HO-1, and their combination displayed an AUC of 0.897 (95% CI, 0.808-0.986), 0.825 (95% CI, 0.708-0.941), and 0.902 (95% CI, 0.817-0.986) for discriminating SBP from non-SBP patients. HO-1 level between serum and ascites had a higher correlation in patients in a Child-Pugh B stage (R=0.691, p<0.001) than in the Child-Pugh C stage (R=0.475, p=0.014). The correlation between HO-1 level and inflammatory factors or biochemical parameters was observed in SBP patients and presented in a Child-Pugh stage-dependent manner.
CONCLUSIONS: HO-1 level has the ability to predict the occurrence of SBP in cirrhotic patients with ascites which has the potential to be a biomarker for the early prediction of SBP and move into the clinical setting. However, the Child-Pugh stage should be considered when using the HO-1 as a predictive biomarker.
摘要：
目的：确定自发性细菌性腹膜炎（SBP）的预测生物标志物对于肝硬化腹水患者至关重要。本研究旨在确定该人群中血清或腹水血红素加氧酶-1（HO-1）对SBP的预测价值。
方法：在这项队列研究中，60例肝硬化腹水伴SBP患者（研究队列，n=26)或不(对照组，n=34）被回顾性纳入。ELISA法检测血清和腹水中的HO-1水平。通过计算受试者工作特征（ROC）曲线分析中的曲线下面积（AUC）来评估HO-1的预测性能。
结果：SBP患者血清和腹水中的HO-1水平均明显高于非SBP患者（p<0.001）。血清,腹水HO-1及其组合显示AUC为0.897（95％CI，0.808-0.986），0.825（95%CI，0.708-0.941），和0.902（95％CI，0.817-0.986）用于区分SBP与非SBP患者。血清和腹水之间的HO-1水平在Child-PughB期（R=0.691，p<0.001）患者中的相关性高于Child-PughC期（R=0.475，p=0.014）。在SBP患者中观察到HO-1水平与炎症因子或生化指标之间的相关性，并以Child-Pugh分期依赖性方式呈现。
结论：HO-1水平具有预测肝硬化腹水患者SBP发生的能力，有可能成为早期预测SBP的生物标志物并进入临床环境。然而,使用HO-1作为预测生物标志物时,应考虑Child-Pugh阶段.
方法：在这项队列研究中，60例肝硬化腹水伴SBP患者（研究队列，n=26)或不(对照组，n=34）被回顾性纳入。ELISA法检测血清和腹水中的HO-1水平。通过计算受试者工作特征（ROC）曲线分析中的曲线下面积（AUC）来评估HO-1的预测性能。
结果：SBP患者血清和腹水中的HO-1水平均明显高于非SBP患者（p<0.001）。血清,腹水HO-1及其组合显示AUC为0.897（95％CI，0.808-0.986），0.825（95%CI，0.708-0.941），和0.902（95％CI，0.817-0.986）用于区分SBP与非SBP患者。血清和腹水之间的HO-1水平在Child-PughB期（R=0.691，p<0.001）患者中的相关性高于Child-PughC期（R=0.475，p=0.014）。在SBP患者中观察到HO-1水平与炎症因子或生化指标之间的相关性，并以Child-Pugh分期依赖性方式呈现。
结论：HO-1水平具有预测肝硬化腹水患者SBP发生的能力，有可能成为早期预测SBP的生物标志物并进入临床环境。然而,使用HO-1作为预测生物标志物时,应考虑Child-Pugh阶段.
