• Keita Shibahashi, Hiroki Matsunaga, Takuto Ishida, Kazuhiro Sugiyama, and Yuichi Hamabe.
• Tertiary Emergency Medical Center, Tokyo Metropolitan Bokutoh Hospital, 4-23-15, Kotobashi, Sumida-ku, Tokyo, 130-8575, Japan. kshibahashi@yahoo.co.jp.
• Eur J Trauma Emerg Surg. 2022 Dec 1; 48 (6): 460746144607-4614.

PurposeEarly identification of blunt thoracic aortic injury is vital for preventing subsequent aortic rupture. However, risk factors for blunt thoracic aortic injury remain unclear, and a prediction rule remains to be established. We developed and internally validated a new nomogram-based screening model that allows clinicians to quantify blunt thoracic aortic injury risk.MethodsAdult patients (age ≥ 18 years) with blunt injury were selected from a nationwide Japanese database (January 2004-May 2019). Patients were randomly divided into training and test cohorts. A new nomogram-based blunt thoracic aortic injury-screening model was constructed using multivariate logistic regression analysis to quantify the association of potential predictive factors with blunt thoracic aortic injury in the training cohort.ResultsOverall, 305,141 patients (training cohort, n = 152,570; test cohort, n = 152,571) were eligible for analysis. Median patient age was 65 years, and 60.9% were men. Multivariate analysis in the training cohort revealed that 13 factors (positive association: age ≥ 55 years, male sex, high-energy impact, hypotension on hospital arrival, Glasgow Coma Scale score < 9 on hospital arrival, diaphragmatic injuries, hepatic injuries, pulmonary injuries, cardiac injuries, renal injuries, sternum fractures, multiple rib fractures, and pelvic fractures) were significantly associated with blunt thoracic aortic injury and included in the screening model. In the test cohort, the new screening model had an area under the curve of 0.87.ConclusionsOur novel nomogram-based screening model aids in the quantitative assessment of blunt thoracic aortic injury risk. This model may improve tailored decision-making for each patient.© 2022. The Author(s), under exclusive licence to Springer-Verlag GmbH Germany.

20 keywords...

hide…