PDF(Pubmed)
Abstract：
Diagnosis of relevant organ injury after blunt abdominal injury (AI) in multiple-injury/polytraumatised patients is challenging. AI can be distinguished between injuries of parenchymatous organs (POI) of the upper abdomen (liver, spleen) and bowel and mesenteric injuries (BMI). Still, such injuries may be associated with delays in diagnosis and treatment. The present study aimed to verify laboratory parameters, imaging diagnostics, physical examination and related injuries to predict intraabdominal injuries. This retrospective, single-centre study includes data from multiple-injury/polytraumatised patients between 2005 and 2017. Two main groups were defined with relevant abdominal injury (AI+) and without abdominal injury (AI-). The AI+ group was divided into three subgroups: BMI+, BMI+/POI+, and POI+. Groups were compared in a univariate analysis for significant differences. Logistic regression analysis was used to determine predictors for AI+, BMI+ and POI+. 26.3% (271 of 1032) of the included patients had an abdominal injury. Subgroups were composed of 4.7% (49 of 1032) BMI+, 4.7% (48 of 1032) BMI+/POI+ and 16.8% (174 of 1032) POI+. Pathological abdominal signs had a sensitivity of 48.7% and a specificity of 92.4% for AI+. Transaminases were significantly higher in cases of AI+. Pathological computed tomography (CT) (free fluid, parenchymal damage, Bowel Injury Prediction Score (BIPS), CT Grade > 4) was summarised and had a sensitivity of 94.8%, a specificity of 98%, positive predictive value (PPV) of 94.5% and, negative predictive value (NPV) of 98.2% for AI+. The detected predictors for AI+ were pathological abdominal findings (odds ratio (OR) 3.93), pathological multi-slice computed tomography (MSCT) (OR 668.9), alanine (ALAT) ≥ 1.23 µmol/ls (OR 2.35) and associated long bone fractures (OR 3.82). Pathological abdominal signs, pathological MSCT and lactate (LAC) levels ≥ 1.94 mmol/l could be calculated as significant risk factors for BMI+. For POI+ pathological abdominal MSCT, ASAT ≥ 1.73 µmol/ls and concomitant thoracic injuries had significant relevance. The study presents reliable risk factors for abdominal injury and its sub-entities. The predictors can be explained by the anatomy of the trunk and existing studies. Elevated transaminases predicted abdominal injury (AI+) and, specifically, the POI+. The pathological MSCT was the most reliable predictive parameter. However, it was essential to include further relevant parameters.
摘要：
在多发伤/多发伤患者中，钝性腹部损伤（AI）后相关器官损伤的诊断具有挑战性。AI可以区分上腹部实质器官（POI）的损伤（肝脏，脾）和肠和肠系膜损伤（BMI）。尽管如此，这种损伤可能与诊断和治疗的延误有关。本研究旨在验证实验室参数，成像诊断,体格检查和相关损伤预测腹内损伤。这次回顾，单中心研究包括2005年至2017年间多发伤/多发伤患者的数据.两个主要组定义为相关腹部损伤（AI）和无腹部损伤（AI-）。AI+组分为三个亚组：BMI+，BMI+/POI+,POI+。在单变量分析中比较各组的显著性差异。Logistic回归分析用于确定AI+的预测因子，BMI+和POI+。26.3％（1032例中的271例）的患者患有腹部损伤。亚组由4.7%(49/1032)BMI+组成，4.7%(1032个中的48个)BMI+/POI+和16.8%(1032个中的174个)POI+。病理腹部体征对AI+的敏感性为48.7%，特异性为92.4%。在AI+的情况下转氨酶显著较高。病理计算机断层扫描（CT）（游离液，实质损伤,肠道损伤预测评分(BIPS),CT等级>4)被总结，灵敏度为94.8%，特异性为98%，阳性预测值（PPV）为94.5%，AI+的阴性预测值（NPV）为98.2%。检测到的AI+预测因子是病理性腹部结果（比值比（OR）3.93），病理性多层螺旋CT（MSCT）（OR668.9），丙氨酸（ALAT）≥1.23µmol/ls（OR2.35）和相关长骨骨折（OR3.82）。腹部病理性体征,病理性MSCT和乳酸（LAC）水平≥1.94mmol/l可以计算为BMI的重要危险因素。对于POI+病理性腹部MSCT，ASAT≥1.73µmol/ls与伴随的胸部损伤具有显着相关性。该研究提出了腹部损伤及其亚实体的可靠危险因素。预测因素可以通过躯干的解剖结构和现有研究来解释。转氨酶升高可预测腹部损伤(AI+)和，具体来说,POI+。病理MSCT是最可靠的预测参数。然而,必须包括进一步的相关参数。
