重症急性胰腺炎患者血浆致动脉粥样硬化指数的动态变化及其与预后的关系

Dynamic changes of the atherogenic index of plasma in patients with severe acute pancreatitis and its association with prognosis

摘要:

目的]分析血浆致动脉粥样硬化指数(AIP)在重症急性胰腺炎(SAP)患者病程中的动态变化及其对患者预后的预测效能。 [方法]收集并分析2020年6月—2024年6月三六三医院150例SAP患者的临床数据,且按照患者入院28天随访期内的预后进行分组,其中112例纳入生存组,38例纳入死亡组。采用重复测量方差分析,探究两组患者在治疗前(T0)、治疗后第5天(T5)、治疗后第10天(T10)的甘油三酯(TG)、高密度脂蛋白胆固醇(HDLC)、AIP及炎症指标的变化情况；多元线性回归分析不同时间点AIP与炎症指标的关系；Logistic回归模型分析不同时间点AIP与SAP患者死亡风险的独立相关性；受试者工作特征(ROC)曲线分析不同时间点AIP、TG、HDLC及炎症指标与SAP患者死亡风险的关系；限制性立方样条模型(RCS)分析AIP与SAP患者死亡风险的关系；多维度分层分析AIP水平与SAP患者死亡风险的关联性；Kaplan-Meier生存曲线分析SAP患者生存率；亚组分析不同病因患者T5、T10时AIP与SAP患者死亡风险的关联；采用DeLong检验进一步分析剔除脂源性患者后AIP对SAP的预测效能。 [结果]死亡组患者的年龄、多器官功能衰竭、白细胞计数(WBC)、急性生理与慢性健康评分Ⅱ(APACHE Ⅱ)及序贯器官衰竭评分(SOFA)均高于生存组；白蛋白(ALB)、总胆固醇(TC)、低密度脂蛋白胆固醇(LDLC)水平均低于生存组,差异有统计学意义(均P＜0.05)。生存组TG、AIP、C反应蛋白(CRP)、降钙素原(PCT)及白细胞介素6(IL-6)在T0、T5、T10依次下降,而死亡组AIP和CRP、PCT、IL-6呈先升高、后下降趋势；生存组HDLC上升,而死亡组呈先下降后上升趋势,差异有统计学意义(均P＜0.05)。重复测量方差分析对两组患者的TG、HDLC、AIP、CRP、PCT、IL-6进行统计检验,结果显示两组患者在时间、组间和时间×组间上的差异均有统计学意义(P＜0.05)。T0、T5、T10时AIP分别与CRP、PCT、IL-6呈正向线性回归关系(P＜0.001)。Logistic回归分析表明T5、T10时AIP与SAP患者死亡风险均具有独立相关性(P＜0.05)。ROC曲线分析结果显示,AIP在T5(AUC:0.922,95%CI:0.873～0.965)、T10(AUC:0.951,95%CI:0.914～0.993)预测效能均佳,且T10时预测效能最高,灵敏度(99.5%)、精确度(88.7%)最高。RCS分析结果表明T5、T10时AIP与SAP患者死亡风险呈非线性剂量-反应关系(P＜0.05)。多维度分层分析结果显示T5、T10时,AIP与SAP患者死亡风险在APACHEⅡ≥20分、SOFA≥8分、CRP≥200 mg/L、PCT≥10 mg/L及IL-6≥300 ng/L亚组内存在关联性,且SAP患者伴有器官衰竭时死亡风险更高。生存曲线分析显示,T5时AIP＞0.20组与AIP≤0.20组比较,差异具有统计学意义(χ²＝6.437,P＜0.001)；T10时AIP＞0.10组与AIP≤0.10组比较,差异具有统计学意义(χ²＝5.831,P＜0.001)。亚组分析显示不同病因中T5、T10时AIP与死亡风险均存在独立相关性,且高AIP水平组相关性更明显,且均具有良好的区分能力；剔除脂源性患者后,ROC结果显示AIP在非脂源性SAP患者中仍具有预测效能,且T10时预测效能较高(AUC:0.928,95%CI:0.880～0.971)。 [结论]AIP在SAP患者病程中呈动态变化,T5、T10时AIP与死亡风险独立相关,且与CRP、PCT、IL-6呈线性相关；T10时AIP预测效能最高,且剔除脂源性患者后AIP仍具有良好的预测效能。将AIP的动态监测纳入临床评估体系,有助于早期识别高危患者,改善预后评估效能。

Abstract:

Aim To analyze the dynamic changes of the atherogenic index of plasma (AIP) during the course of severe acute pancreatitis (SAP) and its predictive efficacy for patient prognosis. Methods We collected and analyzed clinical data from 150 patients with SAP at 363 Hospital from June 2020 to June 2024, and patients were grouped based on their prognosis outcomes during the 28-day follow-up period after admission, with 112 patients assigned to the survival group and 38 patients to the mortality group. Repeated-measures analysis of variance (ANOVA) was used to compare changes in triglyceride (TG), high density lipoprotein cholesterol (HDLC), AIP and inflammatory markers between the two groups at baseline (T0), day 5 post-treatment (T5), and day 10 post-treatment (T10). Multiple linear regression analysis was performed to examine the relationship between AIP and inflammatory markers at different time points; Logistic regression models were used to analyze the independent association between AIP and SAP patient mortality risk at different time points; Receiver operating characteristic (ROC) curve analysis was used to evaluate the relationship between AIP, TG, HDLC, inflammatory markers and SAP patient mortality risk at different time points;Restricted cubic spline (RCS) analysis was used to analyze the relationship between AIP and SAP patient mortality risk; Multidimensional stratified analysis was used to analyze the association between AIP levels and SAP patient mortality risk; Kaplan-Meier survival curve analysis was used to analyze SAP patients' survival rates; Subgroup analysis was performed to examine the association between AIP and SAP mortality risk at T5 and T10 in patients with different etiologies. DeLong's test was further used to analyze the predictive efficacy of AIP for SAP after excluding lipidogenic patients. Results The age, multiple organ failure, white blood cell count (WBC), acute physiological and chronic health score Ⅱ (APACHEⅡ), and sequential organ failure assessment (SOFA) scores of patients in the death group were significantly higher than those in the survival group.Conversely, albumin (ALB), total cholesterol (TC), and low density lipoprotein cholesterol (LDLC) levels were significantly lower in the death group than those in the survival group (P＜0.05). In the survival group, TG, AIP, C-reactive protein (CRP), procalcitonin (PCT), and interleukin-6 (IL-6) levels decreased sequentially at time points T0, T5, and T10, whereas in the death group, AIP, CRP, PCT, and IL-6 exhibited an initial increase followed by a decline. HDLC levels increased in the survival group but decreased first and then increased in the death group, with statistically significant differences (P＜0.05). Repeated measures ANOVA on TG, HDLC, AIP, CRP, PCT, and IL-6 in both groups revealed statistically significant differences in time, between groups, and time×group interaction (P＜0.05). At T0, T5, and T10, AIP showed a positive linear correlation with CRP, PCT and IL-6 (P＜0.05). Logistic regression analysis indicated that AIP at stages T5 and T10 was independently associated with mortality risk in SAP patients (P＜0.05). ROC curve analysis revealed that AIP exhibited optimal predictive efficacy at T5 (AUC:0.2,5%CI:0.873～0.965) and T10 (AUC:0.1,5%CI:0.914～0.993), with the highest predictive efficacy at T10, accompanied by the highest sensitivity (99.5%) and accuracy (88.7%). RCS analysis indicated a nonlinear dose-response relationship between AIP and SAP mortality risk at T5 and T10 (P＜0.05). Multidimensional stratified analysis revealed associations between AIP and SAP mortality risk at T5 and T10 in subgroups with APACHEⅡ≥20, SOFA≥8, CRP≥200 mg/L, PCT≥10 mg/L, and IL-6≥300 ng/L, with higher mortality risk observed in SAP patients with organ failure. Survival curve analysis revealed statistically significant differences between the AIP＞0.20 group and the AIP≤0.20 group at T5 (χ²＝6.437, P＜0.001) and between the AIP＞0.10 group and the AIP≤0.10 group at T10 (χ²＝5.831, P＜0.001). Subgroup analysis revealed independent associations between AIP and mortality risk at T5 and T10 in different etiologies, with more pronounce correlations observed in the high AIP level group, and both demonstrated good discriminatory ability. After excluding lipid-derived patients, ROC curve analysis indicated that AIP still retained predictive efficacy in non-lipid-derived SAP patients, with higher predictive performance at T10 (AUC:0.8,5%CI:0.880～0.971). Conclusions AIP exhibits dynamic changes during the disease course in SAP patients, At T5 and T10, AIP shows independent correlation with mortality risk, and exhibits linear correlations with CRP, PCT, and IL-6. The predictive efficacy of AIP is highest at T10, and it maintains robust predictive performance even after excluding lipid-derived patients. Incorporating dynamic AIP monitoring into clinical evaluation systems helps identify high-risk patients at an early stage and improves the efficacy of prognosis assessment.