Department of Hepatobiliary and Pancreatic Surgery, Cangzhou Central Hospital, Cangzhou, Hebei, China.
Sci Rep. 2023 Jun 9;13(1):9440. doi: 10.1038/s41598-023-36552-z.
Severe acute pancreatitis (SAP) presents with an aggressive clinical presentation and high lethality rate. Early prediction of the severity of acute pancreatitis will help physicians to further precise treatment and improve intervention. This study aims to construct a composite model that can predict SAP using inflammatory markers. 212 patients with acute pancreatitis enrolled from January 2018 to June 2020 were included in this study, basic parameters at admission and 24 h after hospitalization, and laboratory results such as inflammatory markers were collected. Pearson's test was used to analyze the correlation between heparin-binding protein (HBP), procalcitonin (PCT), and C-reactive protein (CRP). Risk factors affecting SAP were analyzed using multivariate logistic regression, inflammatory marker models were constructed, and subject operating curves were used to verify the discrimination of individual as well as inflammatory marker models and to find the optimal cut-off value based on the maximum Youden index. In the SAP group, the plasma levels of HBP, CRP, and PCT were 139.1 ± 74.8 ng/mL, 190.7 ± 106.3 mg/L and 46.3 ± 22.3 ng/mL, and 25.3 ± 16.0 ng/mL, 145.4 ± 67.9 mg/L and 27.9 ± 22.4 ng/mL in non-SAP patients, with a statistically significant difference between the two groups (P < 0.001), The Pearson correlation analysis showed a positive correlation between the three values of HBP, CRP, and PCT. The results of the multivariate logistic regression analysis showed that HBP (OR = 1.070 [1.044-1.098], P < 0.001), CRP (OR = 1.010 [1.004-1.016], P = 0.001), and PCT (OR = 1.030[1.007-1.053], P < 0.001) were risk factors for SAP, and the area under the curve of the HBP-CRP-PCT model was 0.963 (0.936-0.990). The HCP model, consisting of HBP, CRP, and PCT; is well differentiated and easy to use and can predict the risk of SAP in advance.
严重的急性胰腺炎 (SAP) 表现出侵袭性的临床特征和高死亡率。早期预测急性胰腺炎的严重程度将有助于医生进一步精确治疗并改善干预效果。本研究旨在构建一种使用炎症标志物预测 SAP 的综合模型。本研究纳入了 2018 年 1 月至 2020 年 6 月间收治的 212 例急性胰腺炎患者，收集了入院时和入院后 24 小时的基本参数以及炎症标志物等实验室结果。采用 Pearson 检验分析肝素结合蛋白 (HBP)、降钙素原 (PCT) 和 C 反应蛋白 (CRP) 之间的相关性。采用多变量 logistic 回归分析影响 SAP 的危险因素，构建炎症标志物模型，并使用受试者工作特征曲线验证单个和炎症标志物模型的区分度，并根据最大 Youden 指数找到最佳截断值。在 SAP 组中，HBP、CRP 和 PCT 的血浆水平分别为 139.1±74.8ng/mL、190.7±106.3mg/L 和 46.3±22.3ng/mL，而非 SAP 患者的 HBP、CRP 和 PCT 的水平分别为 25.3±16.0ng/mL、145.4±67.9mg/L 和 27.9±22.4ng/mL，两组间差异有统计学意义 (P<0.001)。Pearson 相关性分析显示 HBP、CRP 和 PCT 三个值之间呈正相关。多变量 logistic 回归分析结果显示，HBP (OR=1.070[1.044-1.098]，P<0.001)、CRP (OR=1.010[1.004-1.016]，P=0.001) 和 PCT (OR=1.030[1.007-1.053]，P<0.001) 是 SAP 的危险因素，HBP-CRP-PCT 模型的曲线下面积为 0.963(0.936-0.990)。由 HBP、CRP 和 PCT 组成的 HCP 模型具有良好的区分度和易用性，可以提前预测 SAP 的风险。
