Huang Dongya, Lu Zipeng, Li Qiang, Jiang Kuirong, Wu Junli, Gao Wentao, Miao Yi
Pancreas Center, The First Affiliated Hospital of Nanjing Medical University, Nanjing, China.
Pancreas Center, The Affiliated BenQ Hospital of Nanjing Medical University, Nanjing, China.
J Gastrointest Surg. 2023 Oct;27(10):2145-2154. doi: 10.1007/s11605-023-05772-z. Epub 2023 Jul 24.
For infected necrotizing pancreatitis (INP), percutaneous catheter drainage (PCD) is now widely acknowledged as the initial intervention in a step-up approach, followed, if necessary, by minimally invasive necrosectomy or even open pancreatic necrosectomy. However, an overemphasis on PCD may cause a patient's condition to deteriorate, leading to missed surgical opportunities or even death. This study aimed to develop a simple and convenient scoring tool for assessing the need for surgery in INP patients who received PCD procedures.
In an observational study conducted between April 2015 and December 2020, PCD was utilized as the initial step to treat 143 consecutive INP patients. A surgical necrosectomy was performed when the patient failed to respond. Risk factors of PCD failure (i.e., need for surgical necrosectomy) were identified by multivariate logistic regression models. An integer-based risk scoring tool was developed using the β coefficients derived from the logistic regression model.
In 62 (43.4%) patients, PCD was successful, while the remaining 81 (56.6%) individuals required subsequent surgical necrosectomy. In the multivariate model, organ failure, percentage of pancreatic necrosis, extrapancreatic necrosis volume, and mean CT density of extrapancreatic necrosis volume were associated with a need for surgical necrosectomy. A predictive scoring tool based on these four factors demonstrated an area under the receiver operating characteristic curve (AUC) of 0.893. Under the scoring tool, a total score of 4 or more indicates a high possibility of surgical necrosectomy being required (at least 80%). Using the coordinates of the receiver operating characteristic curve (ROC), the sensitivity and specificity at this threshold are 0.802 and 0.903, respectively.
A risk score model integrating organ failure, percentage of pancreatic necrosis, extrapancreatic necrosis volume, and mean CT density of extrapancreatic necrosis volume can identify INP patients at high risk for necrosectomy. The straightforward risk assessment tool assists clinicians in stratifying INP patients and making more judicious medical decisions.
对于感染性坏死性胰腺炎(INP),经皮导管引流(PCD)目前已被广泛认可为逐步治疗方法中的初始干预措施,如有必要,随后可进行微创坏死组织清除术甚至开放性胰腺坏死组织清除术。然而,过度强调PCD可能导致患者病情恶化,导致错失手术时机甚至死亡。本研究旨在开发一种简单便捷的评分工具,用于评估接受PCD治疗的INP患者的手术需求。
在2015年4月至2020年12月进行的一项观察性研究中,PCD被用作治疗143例连续INP患者的初始步骤。当患者无反应时进行手术坏死组织清除术。通过多因素逻辑回归模型确定PCD失败(即需要进行手术坏死组织清除术)的危险因素。使用从逻辑回归模型得出的β系数开发了一种基于整数的风险评分工具。
62例(43.4%)患者PCD成功,其余81例(56.6%)患者随后需要进行手术坏死组织清除术。在多因素模型中,器官功能衰竭、胰腺坏死百分比、胰腺外坏死体积以及胰腺外坏死体积的平均CT密度与需要进行手术坏死组织清除术相关。基于这四个因素的预测评分工具的受试者操作特征曲线(AUC)下面积为0.893。在该评分工具下,总分4分及以上表明极有可能需要进行手术坏死组织清除术(至少80%)。使用受试者操作特征曲线(ROC)的坐标,该阈值下的敏感性和特异性分别为0.802和0.903。
整合器官功能衰竭、胰腺坏死百分比、胰腺外坏死体积以及胰腺外坏死体积的平均CT密度的风险评分模型可以识别出有坏死组织清除术高风险的INP患者。这种简单的风险评估工具有助于临床医生对INP患者进行分层,并做出更明智的医疗决策。