Epidemiology of acute traumatic aortic injuries in Galicia, Spaindevelopment and validation of a score for early diagnosis of aortic injuries in major blunt chest trauma patients and optimization of management algorithm

Resumo

Acute traumatic aortic injury (ATAI) usually occurs in patients with major blunt chest trauma and has devastating consequences. An early diagnosis with risk stratification, control of blood pressure and cardiac contractility is the cornerstone of the approach to aortic injuries in these patients. Optimization of diagnostic resources and a specific management is mandatory to both prevent potentially lethal aortic-related complications in high-risk major trauma patients, and to avoid unnecessary costs and radiation exposure in low-risk trauma patients. This thesis describes two clinical studies designed to characterise the epidemiology and clinical profile of major trauma patients with ATAI among overall major blunt chest trauma patients in our Spanish region, Galicia, whose population is more than 2,700,000 inhabitants, and to develop and further validate a predictive score of the probability of presenting an ATAI in major trauma patients, which was named Traumatic Aortic Injury Score (TRAINS). In the first instance a retrospective descriptive study was undertaken in all major trauma patients with blunt chest trauma admitted to a public hospital in Galicia from 2006 to 2010 (1,760 patients), followed by a comparison of the distinct clinical profile between trauma patients with (44 patients) and without an associated ATAI (1,716 patients). This epidemiological part of the research showed that overall importance in terms of percentage of ATAIs among major trauma patients with blunt chest trauma between 2006 and 2010 in our institution and region was 4.4% and 2.5% respectively. Likewise, the yearly proportion of ATAIs among major trauma patients in Galicia was 1.1%-3.3%, whereas it ranged from 2.5% to 7.8% in our institution. After estimation of the number of major trauma patients with ATAI who die at the site of accident or during transportation, the actual resulting incidence of ATAI in Galicia would be 3 to 6.1 cases per 100,000 inhabitants per year. Our research confirmed that in Galicia major trauma patients with associated ATAIs present a different epidemiological profile, prognosis, initial clinical presentation and the number and distribution of associated injuries. While major trauma patients without aortic injury distribute fairly homogeneously through the whole age span, ATAIs concentrate among 16 to 35 years old trauma patients, where 54.6% of overall ATAI occurs. The motor vehicle collision is the commonest cause of accident among ATAI victims in Galicia, but we have found a higher proportion of motorcycle collisions, falls and crush traumas and a lower proportion of auto-pedestrian accidents among patients with ATAI compared to those reported in ATAI victims by other authors. The proportion of "atypical or non-isthmal" aortic injuries (43.2%) is significantly higher than the 7% to 15% reported in other clinical series, also highlighting a remarkably higher proportion of injuries at the aortic arch (25%). Major trauma patients with ATAIs presented with haemodynamic instability on admission more frequently than major trauma patients without aortic injury, and they also had a higher proportion of severe extra-thoracic injuries, which justify their worse prognosis. The second part of the research was a cross-sectional study of a diagnostic procedure. The development of a predictive score for the risk of ATAI in major trauma patients with blunt chest trauma was carried out in two stages. In the first stage, we analysed the clinical and radiological characteristics of major blunt chest trauma patients in order to develop the screening tool for ATAIs among major trauma patients. In the second stage, we validated the predictive score in an independent external population of major trauma patients. In the predictive score development, the overall study population consisted of 640 major trauma patients (all of them with blunt chest trauma) divided into two datasets: a score dataset provided only by the Complejo Hospitalario Universitario de A Coruña (76 consecutive major trauma patients with ATAI and 304 without aortic injury), and an independent validation dataset provided by other three different institutions: Complejo Hospitalario Universitario de Vigo, Complejo Hospitalario Universitario de Santiago de Compostela, and Hospital Clínic Universitari de Valencia (52 consecutive major trauma patients with ATAI and 208 without aortic injury). Bivariate analysis identified variables of potential influence in presenting aortic injury among major blunt chest trauma patients. Subsequently, confirmed variables by stepwise forward logistic regression were assigned a score according to their corresponding beta coefficient which was rounded to the closest integer value (1-4). The predictors of aortic injury that we identified included: widened mediastinum (OR: 30.82; CI: 12.05-78.81); hypotension <90 mmHg (OR: 5.85; CI: 2.26-15.15); long bone fracture (OR: 8.60; CI: 2.15-34.31); pulmonary contusion (OR: 4.12; CI: 1.11- 15.20); left scapula fracture (OR: 3.81; CI: 1.24-11.69); hemothorax (OR: 3.47; CI: 1.19-10.09), and pelvic fracture (OR: 2.96; CI: 1.15-7.60). We developed a contemporaneous multivariate prediction model for traumatic aortic injury after major trauma with a sensitivity >93% and a specificity >85%. The small number of ATAI patients who were misdiagnosed by TRAINS (false negatives) presented low-degree aortic injuries. The scoring method also proved its accuracy in both an internal and an external validation process. The TRAINS demonstrated its ability to diagnose ATAIs among major trauma patients with a sensitivity >92% and a specificity >85% in a current multicentre population of major trauma patients, thus confirming its applicability at the current time and in different geographical areas. We demonstrated that the conventional trauma risk scores fail to show statistical relationship between the severity of the trauma and the degree of severity of the aortic injury. Thus, the conventional trauma severity scores are useless to cast suspicion on the diagnosis of ATAI. On the contrary, TRAINS score has proven to be related with the severity of the aortic injury and to be useful even in the diagnosis of low-degree ATAI. In summary, the TRAINS and associated algorithm have been designed to be used in daily practice to easily and rapidly identify major trauma patients who are at risk of aortic injury, thus avoiding unnecessary costs and radiation exposure in low-risk trauma patients. On the other hand, the TRAINS may raise suspicion of ATAI even in the cases of low-degree aortic injuries, which require a close imaging surveillance to determine whether or not those patients will need a further intervention. The TRAINS algorithm is also useful for resource allocation planning, enabling clinicians to refer patients at high risk of traumatic aortic injuries to specialized units, providing the ability to rapidly diagnose and therapeutically manage this critical subset of trauma patients in order to avoid potentially lethal aortic-related complications.