【作者】 李洁；

【导师】 肖炜明； 路国涛；

【作者基本信息】 扬州大学 ， 临床医学硕士（专业学位）， 2024， 硕士

【摘要】 目的:本研究旨在探究血糖与淋巴细胞比值（Glucose-to-lymphocyte ratio,GLR）对急性胰腺炎（Acute pancreatitis,AP）并发持续脏器衰竭（Persistent organ failure,POF）的早期预测价值。方法:本研究收集了2015-2020年扬州大学附属医院消化内科收治的AP患者并根据纳入及排除标准最终纳入677例AP患者进行回顾性研究。通过医院信息系统收集AP患者的人口统计学资料、入院24小时以内的实验室检查指标以及临床常用的AP评分系统包括:CT严重程度指数评分（Computed tomography severity index,CTSI）、严重程度床边指数（Bedside index for severity in acute pancreatitis,BISAP）等。根据POF诊断标准,收集的患者中602例患者未发生POF（非POF组）,75例患者并发POF（POF组）。酌情运用t检验、Kruskal-Wallis检验、卡方检验或Fisher精确检验进行患者的基线资料、实验室检查指标、常用评分系统的差异性分析比较。使用Spearman相关分析分析GLR与其他指标的相关性。将单因素分析中有差异的指标纳入多因素二元Logistic回归模型,并分析是否为AP并发POF的独立危险因素。绘制ROC曲线分析GLR对AP并发POF的预测效能,并与NLR、PLR及目前临床常用的评分系统进行预测效能比较分析。同时根据病因将AP患者分成胆源性急性胰腺炎（Acute biliary pancreatitis,ABP）亚组和高甘油三酯血症性急性胰腺炎（Hypertriglyceridemic acute pancreatitis,HTG-AP）亚组探究GLR对不同病因的AP并发POF的预测效能。结果:1、AP患者人口统计学资料与临床相关资料分析（1）本研究纳入的677例AP患者中非POF组602例,POF组75例（单发呼吸衰竭38例,单发肾衰竭9例,单发循环衰竭4例,多发脏器损伤24例）。与非POF组相比,POF组患者合并糖尿病例数显著增高,差异有统计学意义（P<0.05）。两组患者在性别、年龄、BMI、吸烟、饮酒、病因、合并冠心病、高血压病、高脂血症、脂肪肝、胆结石、胆囊炎的组间比较中未见明显统计学差异（P>0.05）。（2）两组患者入院24小时的实验室检查结果分析显示,POF组患者WBC、NEUT、MONO、AST、LDH、BUN、GLU、K、NLR、PLR、GLR显著升高（P<0.05）;而LYM、ALB、Ca²⁺显著降低（P<0.05）;两组患者的CTSI评分、BISAP评分的结果分析显示,POF组患者的CTSI评分、BISAP评分显著高于非POF组,差异具有统计学意义（P<0.05）。（3）Spearman相关性分析结果显示,GLR与WBC、NEUT、MONO、AST、BUN、LDH、GLU呈显著正相关,与Ca²⁺呈负相关（P<0.05）。（4）将单因素分析中有差异的指标纳入多因素二元Logistic回归分析,结果显示GLR为AP并发POF的独立危险因素,其OR值为1.879,95%CI:为1.378-2.563。2、GLR、NLR、PLR及临床常用评分系统对AP并发POF的预测价值分析（1）GLR预测AP并发POF的AUC为0.745（最佳截断值为0.76,敏感度为78.70%,特异度为59.80%）。与PLR相比,GLR与NLR对AP并发POF的预测效能更好。（2）GLR与CTSI评分、BISAP评分在预测AP并发POF方面没有明显差异（P>0.05）。3、GLR对不同病因的AP并发POF的预测价值（1）GLR预测ABP并发POF的AUC为0.758（最佳截断值为0.99,敏感度为76.50%,特异度为68.80%）,与NLR相比（AUC为0.747,最佳截断值为14.55,敏感度为64.70%,特异度为76.50%）无明显统计学差异（P>0.05）。（2）GLR预测HTG-AP并发POF的AUC为0.754（最佳截断值为0.76,敏感度为82.10%,特异度为57.80%）,与NLR与PLR相比,GLR在HLT-AP并发POF方面具有更好的预测效能（P<0.05）。结论:GLR为AP并发POF的独立危险因素,可作为一种较好的预测AP患者早期发生POF的新型血液学指标,其预测效能优于PLR,与NLR、CTSI评分、BISAP评分相似,并且GLR在预测HTG-AP并发POF方面更具有优势。更多还原

【Abstract】 Objective:This study sought to determine the blood glucose-to-lymphocyte ratio’s（GLR）early prognostic value in cases of acute pancreatitis（AP）exacerbated with persistent organ failure（POF）.Methods:In this study,We gathered AP patients who were admitted to Yangzhou University Hospital Department of Gastroenterology between 2015 and 2020 for this study,and 677 AP patients were ultimately included for a retrospective analysis based on the inclusion and exclusion criteria.We employed the hospital information system to collect demographic data from AP patients,laboratory test indices within 24 hours of admission,and commonly used clinical AP rating systems such as the CTSI and BISAP scores.According to the diagnostic criteria for POF,602 patients were collected without POF（non-POF group）and 75 patients with concomitant POF（POF group）.Differences in patients’baseline data,laboratory test indices,and generally used scoring systems were analyzed and compared using the t-test,Kruskal-Wallis test,chi-square test,or Fisher’s exact test,as appropriate.Using Spearman correlation analysis to analyze the correlation between GLR and other metrics Indicators with differences in the univariate analysis were included in a multifactorial binary logistic regression model and analyzed for independent risk factors for AP complicating POF.ROC curves were plotted to analyze the predictive efficacy of GLR on POF complicating AP,and the predictive efficacy was compared with that of NLR,PLR,and other scoring systems commonly used in clinical practice.At the same time,the patients with AP were divided into subgroups of biliary acute pancreatitis and hypertriglyceridemic acute pancreatitis according to the etiology to investigate the predictive efficacy of GLR for POF complicated by AP of different etiologies.Results:1、Analysis of demographic data and clinically relevant data of AP patients（1）Of the 677 AP patients included in the study,602 were in the non-POF group and 75 in the POF group（Single respiratory failure 38 cases,single renal failure 9 cases,single circulatory failure 4 cases,multiple organ damage 24 cases）.The number of cases of comorbid diabetes mellitus was significantly higher in the POF group compared to the non-POF group,and the difference was statistically significant（P<0.05）.No statistically significant differences were seen in the between-group comparison of gender,age,BMI,smoking,alcohol consumption,etiology,and comorbidities of coronary artery disease,hypertension,hyperlipidemia,fatty liver,gallstones,and cholecystitis between the two groups（P>0.05）.（2）Analysis of the laboratory findings of patients in both groups at 24 hours of admission showed that WBC,NEUT,MONO,AST,LDH,BUN,GLU,K,NLR,PLR,and GLR were significantly elevated in the patients in the POF group（P<0.05）;whereas LYM,ALB,and Ca²⁺were significantly reduced（P<0.05）;CTSI scores and BISAP scores of patients in the POF group were significantly higher than those in the non-POF group,and the difference was statistically significant（P<0.05）.（3）Spearman correlation analysis showed that GLR was significantly positively correlated with WBC,NEUT,MONO,AST,BUN,LDH,and GLU,and negatively correlated with Ca²⁺（P<0.05）.（4）Indicators that differed in the univariate analysis were included in the multifactorial binary logistic regression analysis,which showed that GLR was an independent risk factor for AP complicating POF,with an OR of 1.879 and a 95%CI:1.378-2.563.2、Analysis of the predictive value of GLR,NLR,PLR and commonly used clinical scoring systems for AP complicated by POF（1）GLR predicted AP concurrent POF with an AUC of 0.745（best cutoff value of 0.76,sensitivity of 78.70%and specificity of 59.80%）.Compared with PLR,GLR has a better predictive efficacy with NLR for AP-concomitant POF.（2）There was no significant difference between the GLR with the CTSI score and BISAP score in predicting AP complicating POF（P>0.05）.3、Predictive value of GLR for AP complicating POF in different etiologies（1）GLR predicted ABP concurrent with POF at an AUC of 0.758（optimal cutoff value of0.99,sensitivity of 76.50%,and specificity of 68.80%）,which was not statistically significantly different（P>0.05）compared to NLR（AUC of 0.747,optimal cutoff value of14.55,sensitivity of 64.70%,and specificity of 76.50%）.（2）GLR predicted HTG-AP concurrent POF at an AUC of 0.754（optimal cutoff value of0.76,sensitivity of 82.10%,and specificity of 57.80%）,which showed better predictive efficacy of GLR in HLT-AP concurrent POF compared to NLR and PLR（P<0.05）.Conclusion:In patients with AP,GLR is a potentially more useful novel hematological measure for predicting the early onset of POF as it is an independent risk factor for AP-complicated POF.Its predictive effectiveness surpasses that of PLR and is comparable to that of NLR,CTSI score,and BISAP score.Furthermore,GLR is more beneficial when HTG-AP complicates POF.更多还原