摘要
目的 描述急性胰腺炎(AP)早期淋巴细胞绝对计数(ALC)的发展轨迹,探讨其与并发脓毒症及其相关临床结果之间的关系.方法 回顾性筛查2019年1月至2023年10月徐州医科大学附属医院收治的所有AP患者,最终纳入210例,按脓毒症的诊断标准分为脓毒症组(n=54)和非脓毒症组(n=156).收集两组患者入院后前5 d的ALC指标,使用群组轨迹模型探索描述ALC随时间变化的轨迹,建立5组轨迹模型,将患者分为低ALC水平组(n=42)、中ALC水平组(n=69)、高ALC水平组(n=45)、ALC下降组(n=38)和ALC正常水平组(n=16).收集两组患者一般临床资料,包括性别、年龄、急性生理学和慢性健康状况评价Ⅱ(APACHE Ⅱ)、格拉斯哥昏迷量表(GCS)评分、病因、既往史、个人史、实验室指标[白细胞计数(WBC)、超敏C反应蛋白(hs-CRP)、中性粒细胞绝对计数(ANC)、单核细胞绝对计数(AMC)]、临床特征等.比较各轨迹组随时间的变化与相关临床预后指标的关系.脓毒症的危险因素经Cox比例风险模型筛选.采用Kaplan-Meier图和Log-rank检验来比较5个轨迹组脓毒症的发生率.结果 脓毒症组和非脓毒症组患者性别、病因学、2型糖尿病病史、AP病史、个人史、hs-CRP、AMC比较,差异均无统计学意义(P＞0.05);两组患者年龄、APACHE Ⅱ评分、GCS评分、高脂血症性AP占比、WBC、ANC、ALC比较,差异均有统计学意义(P＜0.05).不同ALC轨迹组患者的年龄、APACHE Ⅱ评分、GCS评分、ARDS、休克、AKI的发生率、机械通气患者比率、连续性肾脏替代治疗患者比率比较,差异均有统计学意义(P＜0.05);低ALC水平组患者的ICU住院时间、28 d病死率显著高于其他组,差异均有统计学意义(P＜0.05).低ALC水平组的患者相较于其他组发生脓毒症的风险更高(HR=5.277;95％CI:2.397～11.616;P＜0.001).在调整患者特征(年龄、病因)后,低ALC水平组的脓毒症发生率仍显著增高(HHRR=5.535;95％CI:2.420～12.660;P＜0.001).经Log-rank检验,不同轨迹分组脓毒症发生率差异有统计学意义(P＜0.001).结论 急性胰腺炎早期绝对淋巴细胞计数的变化轨迹与脓毒症的发生密切相关,且对早期诊断及鉴别脓毒症患者有深远的意义.
Abstract
Objective To describe the development trajectory of absolute lymphocyte count(ALC)in the early stage of acute pancreatitis(AP),and to explore its relationship with sepsis and its related clinical outcomes.Methods A retrospective screening was conducted on all AP patients admitted to Affiliated Hospital of Xuzhou Medical University from January 2019 to October 2023,and a total of 210 cases were included.According to the diagnostic criteria for sepsis,they were divided into the sepsis group(n=54)and the non sepsis group(n=156).The ALC indexes of the two groups of patients in the first 5 days after admission were collected.The group trajectory model was used to explore the trajectory of ALC over time,and five groups of trajectory models were established.The patients were divided into low ALC level group(n=42),medium ALC level group(n=69),high ALC level group(n=45),ALC decline group(n=38)and ALC normal level group(n=16).The general clinical data,gender,age,acute physiology and chronic health evaluation Ⅱ(APACHE Ⅱ)score,Glasgow coma scale(GCS)score,etiology,past history,personal history,laboratory indexes[white blood cell count(WBC),hypersensitivity C-reactive protein(hs-CRP),absolute neu-trophil count(ANC),absolute monocyte count(AMC)]and clinical characteristics of the two groups were collected.The relationship between the temporal changes of each trajectory group and relevant clinical prognostic indicators was compared.The risk factors for sepsis were screened u-sing Cox proportional hazards model.The incidence of sepsis among five trajectory groups was compared using Kaplan-Meier plot and Log rank test.Results There were no statistically significant differences in gender,etiology,history of type 2 diabetes,history of AP,personal history,hs-CRP and AMC between the two groups(P＞0.05).There were statistically significant differences in age,APACHE Ⅱ score,GCS score,proportion of hyperlipidemic AP,WBC,ANC and ALC between the two groups(P＜0.05).There were significant differences in age,APACHEⅡ score,GCS score,incidence of ARDS,shock,AKI,ratio of patients with mechanical ventilation,and ratio of patients treated with continuous renal replacement therapy among different ALC trajectory groups(P＜0.05).The ICU hospitalization time and 28-day fatality rate of patients in the low ALC level group were significantly higher than those in the other groups,and the differences were statistically significant(P＜0.05).Patients in the low ALC level group had a higher risk of sepsis than other groups(HR=5.277;95％CI:2.397-11.616;P＜0.001).After adjusting for patient characteristics(age,etiology),the incidence of sepsis in the low ALC group was still significantly higher(HR=5.535;95％CI:2.420-12.660;P＜0.001).By Log-rank test,the incidence of sepsis in different trajectory groups was statistically significant(P＜0.001).Conclusion The trajectory of absolute lymphocyte count in the early acute pancreatitis is closely related to the occurrence of sepsis,and has profound significance for early diagnosis and differentiation of sepsis patients.
维普
万方数据