中国实用外科杂志

中国实用外科杂志 ›› 2024, Vol. 44 ›› Issue (05): 564-570.DOI: 10.19538/j.cjps.issn1005-2208.2024.05.16

• 论著 • 上一篇    下一篇

感染坏死性胰腺炎胰腺外坏死区域和体积与死亡风险的关联分析

黄东亚，李    强，苗    毅，蒋奎荣，陆子鹏，吴峻立，高文涛，肖    斌   

  1. 南京医科大学第一附属医院胰腺中心，江苏南京210029
  • 出版日期:2024-05-01 发布日期:2024-05-17

  • Online:2024-05-01 Published:2024-05-17

摘要：  目的    探讨感染坏死性胰腺炎（INP）胰腺外坏死区域（ENA）和体积（ENV）与死亡风险的关联性。方法    回顾性分析2015年3月至2020年12月南京医科大学第一附属医院胰腺中心收治的150例INP病人的临床资料，按是否死亡分为存活组（136例）与死亡组（14例）。使用logistic回归模型、广义线性模型（GLM）和广义加性模型（GAM）来探讨ENA和ENV与死亡风险之间的线性和非线性关系，通过两分段回归模型来计算饱和效应。结果    病死率为9.3%（14/150），病人平均年龄为（50.5±15.1）岁，其中64.7%（97/150）为男性。单因素分析结果显示胰腺实质坏死程度、ENV、持续衰竭器官≥2个、扩大经皮穿刺置管引流术（PCD）管径以及胰腺外坏死（EN）平均CT密度与死亡风险有关（P＜0.1）。通过引进或剔除协变量对回归系数影响>10%或单因素分析中P＜0.1的方法来筛选协变量。在对协变量进行充分调整后发现，ENA与死亡风险无关，ENV与死亡风险呈非线性关系。两分段回归模型显示ENV拐点为1250 mL。在拐点左侧，ENV每增加100 mL，病人死亡风险增加40%（OR=1.4，95% CI 1.0~2.0，P=0.042）。拐点右侧的效应大小无统计学意义（R=0.7，95% CI 0.4~1.1，P=0.120）。结论    INP病人ENA与死亡风险无关，ENV与死亡风险之间呈非线性关系。当ENV＜1250 mL时，死亡风险随ENV增大而升高，当ENV达到1250 mL时呈饱和效应。

关键词: 胰腺炎, 感染坏死性, 胰腺外坏死区域, 胰腺外坏死体积, 死亡

Abstract: Association analysis of the areas and volume of extrapancreatic necrosis in infected necrotizing pancreatitis with the risk of death        HUANG Dong-ya, LI Qiang, MIAO Yi, et al. Department of Pancreas Center, the First Affiliated Hospital of Nanjing Medical University, Nanjing 210029, China
Corresponding author: LI Qiang, E-mail：liqiang020202@163.com
Abstract    Objective    This study aimed to investigate the association between extrapancreatic necrotic areas (ENA) and extrapancreatic necrotic volume (ENV) in infected necrotizing pancreatitis (INP) patients and their mortality rates. Methods    A retrospective analysis was conducted on clinical data of 150 patients with INP treated at the Pancreas Center of the First Affiliated Hospital of Nanjing Medical University from March 2015 to December 2020. Patients were divided into survival (136 cases) and mortality (14 cases) groups. Logistic regression models, generalized linear models (GLM), and generalized additive models (GAM) were employed to explore the linear and nonlinear relationships between ENA, ENV, and mortality risk. The saturation effect was calculated using a two-piecewise regression model. Results  The overall mortality rate was 9.3% (14/150), with a mean patient age of (50.5 ± 15.1) years, and 64.7% (97/150) were male. Univariate analysis showed that the degree of pancreatic parenchymal necrosis, ENV, number of persistent organ failures ≥2 , upsizing percutaneous catheter drainage (PCD) diameter, and mean computed tomography (CT) density of extrapancreatic necrosis (EN) were positively correlated with mortality risk (P<0.1). After adjusting for confounding variables by introducing or removing covariates that changed the regression coefficient by >10% or had P < 0.1 in the univariate analysis, ENA was not associated with mortality risk, while ENV exhibited a nonlinear relationship with mortality risk. The two-piecewise regression analysis identified a turning point at 1250 mL for ENV. On the left side of this threshold, a 100 ml increase in ENV was associated with a 40% higher mortality rate (OR=1.4, 95%CI 1.0-2.0, P=0.042). No statistically significant effect was detected on the right side of the turning point (OR=0.7, 95%CI 0.4-1.1, P=0.120). Conclusion    ENA in INP patients does not correlate with the risk of death, while a nonlinear relationship exists between ENV and mortality rate. Mortality increased with the increase of ENV when ENV is less than 1250 mL. A saturation effect is observed when ENV reaches 1250 mL.

Key words: infected necrotizing pancreatitis, extrapancreatic necrotic areas, extrapancreatic necrotic volume, death