| 刘李,杨烈,李俊川,等.联合多项实验室指标的预测模型在早期诊断老年急性胰腺炎中的价值[J].安徽医药,2022,26(5):1000-1003. |
| 联合多项实验室指标的预测模型在早期诊断老年急性胰腺炎中的价值 |
| Diagnostic value of multiple early indicators in elderly patients with acute pancreatitis |
| |
| DOI：10.3969/j.issn.1009-6469.2022.05.035 |
| 中文关键词: 胰腺炎 淀粉酶类 logistic回归 ROC曲线 老年人 诊断价值 |
| 英文关键词: Pancreatitis Amylases Logistic regression ROC curve Elderly Diagnostic value |
| 基金项目: |
|
| 摘要点击次数: 967 |
| 全文下载次数: 343 |
| 中文摘要: |
| 目的探讨联合多项血液及尿液指标的预测模型在早期诊断老年急性胰腺炎(AP)中的价值。方法纳入2014年8月至2019年8月就诊于资阳市第一人民医院的老年急腹症病人160例作为研究对象，其中80例急性胰腺炎病人为AP组，80例非急性胰腺炎病人为非AP组；应用logistic回归分析筛选出早期诊断老年急性胰腺炎的指标，建立标志物的预测模型；最后采用ROC曲线评价该模型预测效能。结果AP组病人血清淀粉酶(S-AMY)、尿淀粉酶(U-AMY)、胰脂肪酶(LPS)、胰蛋白酶原激活肽(TAP)、降钙素原(PCT)水平均显著高于非AP组(P<0.05)，而C反应蛋白(CRP)、白细胞(WBC)、肿瘤坏死因子α(TNF-α)、白细胞介素6(IL-6)等指标差异无统计学意义(P>0.05)；logistic回归分析结果显示:S-AMY、LPS、TAP及U-AMY的相关性及95%CI 分别为1.019(1.011～1.028)、1.023(1.012～1.034)、1.548(1.244～1.925)及1.291(0.983～1.695)，均差异有统计学意义(P<0.05)。早期诊断老年急性胰腺炎的模型为P=1／［1+e(-12.540+0.019SAMY+0.023LPS+0.437TAP+0.255UAMY)］，C-index为0.941，提示该模型区分度良好，校准曲线提示模型一致性良好，模型的AUC 为0.941(95%CI:0.903～0.979)，提示该模型预测效率高。结论联合S-AMY、LPS、TAP及U-AMY的预测模型有助于提高对于老年急性胰腺炎的早期诊断能力，有重要临床意义。 |
| 英文摘要: |
| Objective To explore the diagnostic value of multiple early indicators in elderly patients with acute pancreatitis.Meth?ods A total of 160 elderly patients with acute abdomen who were treated in the First People's Hospital of Ziyang from August 2014 to August 2019 were included as the research subjects. Among them, 80 patients with acute pancreatitis belonged to the AP group, and 80 patients with nonacute pancreatitis belonged to the non-AP group. Logistic regression analysis was used to screen out the indicators for the early diagnosis of acute pancreatitis in the elderly, and a predictive model of the markers was established. Finally, the receiver oper?ator characteristic (ROC) curve was used to evaluate the predictive performance of the model.Results The levels of serum amylase (SAMY),urinary amylase (U-AMY), pancreatic lipase (LPS), trypsinogen-activated peptide (TAP) and procalcitonin (PCT) in the AP group were significantly higher than those in the non-AP group (P < 0.05), but there was no significant difference in C-reactive protein (CRP), white blood cell (WBC), tumor necrosis factor α (TNF- α), or interleukin 6 (IL-6) (P > 0.05). The results of logistic regression analysis showed that the correlation and 95% confidence interval (95%CI) of S-AMY, LPS, TAP and U-AMY were 1.019 (1.011-1.028),1.023 (1.012-1.034) and 1.548(1.244-1.925) and 1.291 (0.983-1.695), respectively, and the difference was statistically significant (P<0.05). The model for the early diagnosis of acute pancreatitis was P = 1/[1+e(-12.540+0.019SAMY+0.023LPS+0.437TAP+0.255UAMY)], and the C-index is 0.941,indicating that the model had a good degree of discrimination. The calibration curve indicated that the model had good consistency, and the area under the curve (AUC) of the model was 0.941 (95%CI:0.903-0.979), indicating that the model had high prediction efficiency.Conclusion The combined prediction model of S-AMY, LPS, TAP and U-AMY can help improve the early diagnostic ability of acute pancreatitis in the elderly individuals, which has important clinical significance. |
|
查看全文
查看/发表评论 下载PDF阅读器 |
| 关闭 |
|
|
|
