Department of Pancreatic Surgery, Union Hospital, Tongji Medical College, Huazhong University of Science and Technology, Wuhan, China.
Department of Breast and Thyroid Surgery, Union Hospital, Tongji Medical College, Huazhong University of Science and Technology, Wuhan, China.
Dig Surg. 2021;38(2):166-174. doi: 10.1159/000513223. Epub 2021 Feb 1.
Postoperative complications after pancreatectomy are a challenging problem due to their high incidence and serious consequences. The majority of studies have focused on a specific complication, but data on predictors of overall postoperative complications (OPCs) are limited.
The data of patients who underwent pancreatectomy at a single institute between 2017 and 2019 were analyzed retrospectively. Univariate and multivariate logistic regression were used to investigate predictors of the outcomes of interest. The Clavien-Dindo classification and comprehensive complication index (CCI) were used to assess postoperative complications and the severity of postoperative complications. The relationship between predictors and the CCI was evaluated by linear regression.
A total of 490 patients were divided into a training group (n = 339) and a validation group (n = 151). The rate of OPCs was 44.25%. Fluid transfusion and albumin difference (AD) were predictors of OPCs. AD showed a good discrimination (AUC = 0.70) and good calibration in the validation cohort. AD was associated with complications, including pancreatic fistula, intra-abdominal hemorrhage, intra-abdominal infection, delayed gastric emptying, and re-intervention, and was positively correlated with complication severity. Intraoperative blood loss and preoperative albumin were independent predictors of AD.
AD, a variable that reflects dynamic physiological changes is a new and accessible predictor of OPCs following pancreatectomy.
胰腺切除术后的并发症是一个具有挑战性的问题,因为它们的发生率高且后果严重。大多数研究都集中在特定的并发症上,但关于总体术后并发症(OPC)预测因素的数据有限。
回顾性分析了 2017 年至 2019 年期间在一家机构接受胰腺切除术的患者的数据。使用单变量和多变量逻辑回归来研究预测感兴趣结局的因素。采用 Clavien-Dindo 分类和综合并发症指数(CCI)评估术后并发症和术后并发症的严重程度。线性回归评估预测因素与 CCI 之间的关系。
共 490 例患者分为训练组(n=339)和验证组(n=151)。OPC 的发生率为 44.25%。输液量和白蛋白差值(AD)是 OPC 的预测因素。AD 在验证队列中具有良好的区分度(AUC=0.70)和校准度。AD 与胰腺瘘、腹腔内出血、腹腔内感染、胃排空延迟和再次干预等并发症有关,且与并发症的严重程度呈正相关。术中出血量和术前白蛋白是 AD 的独立预测因素。
AD 是反映动态生理变化的一个新的且可获得的预测胰腺切除术后 OPC 的指标。