Chest x-ray as a screening tool for blunt thoracic trauma in children

Clinical Predictors of Major Intrathoracic Injury in Pediatric Blunt Trauma

OBJECTIVES

Blunt trauma in pediatric patients accounts for a significant proportion of pediatric death from traumatic injury. Currently, there are no clinical decision-making tools available to guide imaging choice in the evaluation of pediatric patients with blunt thoracic trauma (BTT). This study aimed to analyze the rates of missed major intrathoracic injuries on chest x-ray (CXR) and identify clinical risk factors associated with major intrathoracic injuries to formulate a clinical decision-making tool for computed tomography (CT) use in pediatric patients with BTT.

METHODS

We performed a retrospective single-center study using an institutional trauma database of pediatric patients. Inclusion criteria included age, blunt trauma, and patients who received a CXR and thoracic CT within 24 hours of presentation. Thoracic CT findings were graded as major, minor, or none, and comparison CXR was used to determine the rate of missed thoracic injuries. Eighty-four patient variables were then collected, and clinically relevant variables associated with major intrathoracic injuries were placed in a logistic regression model to determine the best predictors of major injury in pediatric BTT patients.

RESULTS

A total of 180 patients (48.3%) had CXR that missed an injury that was seen on thoracic CT. In our cohort, 20 patients (5.4%) had major injuries that were missed on CXR. Characteristics correlating with major thoracic injuries were older age (odds ratio [OR], 1.125; 95% confidence interval [CI], 1.015-1.247), chest pain (OR, 4.907; 95% CI, 2.173-11.083), abnormal chest auscultation (OR, 3.564; 95% CI, 1.406-9.035), and tachycardia (OR, 2.876; 95% CI, 1.256-6.586). Using these 4 variables, receiver operating characteristic analysis revealed an area under the curve of 0.7903.

CONCLUSIONS

Pediatric BTT patients older than 15 years with tachycardia, chest pain, or abnormal chest auscultation are at increased risk for major intrathoracic injuries and may benefit from thoracic CT.

Comparison of Computed Tomography Use and Mortality in Severe Pediatric Blunt Trauma at Pediatric Level I Trauma Centers Versus Adult Level 1 and 2 or Pediatric Level 2 Trauma Centers

INTRODUCTION

Computed tomography (CT) is the criterion standard for identifying blunt trauma injuries in pediatric patients, but there are long-term risks of CT exposure. In pediatric blunt trauma, multiple studies have shown that increased CT usage does not necessarily equate to improvements in mortality. The aim of this study was to compare CT usage between level 1 pediatric trauma centers versus level 2 pediatric centers and adult level 1 and 2 centers.

METHODS

We performed a retrospective, multicenter analysis of National Trauma Data Bank patient records from the single admission year of 2015. Eligible subjects were defined as younger than 18 years with abdominal or thoracic blunt trauma, had an Injury Severity Scale score of greater than 15, and were treated at a level 1 or 2 trauma center. Data were then compared between children treated at level 1 pediatric trauma centers (PTC group) versus level 2 PTCs or adult level 1/2 trauma centers (ATC group). The primary outcomes measured were rates of head, thoracic, abdominal CT, and mortality. Data from ATC and PTC groups were propensity matched for age, sex, race, and Glasgow Coma Scale.

RESULTS

There were 6242 patients after exclusion criteria. Because of differences in patient demographics, we propensity matched 2 groups of 1395 patients. Of these patients, 39.6% of PTC patients received abdominal CT versus 45.5% of ATC patients (P = 0.0017). Similarly, 21.9% of PTC patients received thoracic CT versus 34.7% of ATC patients (P < 0.0001). There was no difference in head CT usage between PTC and ATC groups (P = 1.0000). There was no significant difference in mortality between patients treated in the PTC versus ATC groups (P = 0.1198).

CONCLUSIONS

Among children with severe blunt trauma, patients treated at level 1 PTCs were less likely to receive thoracic and abdominal CTs than those treated at level 2 pediatric or adult trauma level 1/2 centers, with no significant differences in mortality. These findings support the use of selective imaging in severe blunt pediatric trauma.

Chest x-ray vs. computed tomography of the chest in pediatric blunt trauma

INTRODUCTION

Chest x-ray (CXR) has been shown to be an effective detection tool for clinically significant trauma. We evaluated differences in findings between CXR and computed tomography of the chest (CCT), their impact on clinical management and the performance of the CXR.

METHODS

This retrospective study examined children (less than 18 years) who received a CXR and CCT between 2009 and 2015. We compared characteristics of children by conducting univariate analysis, reporting the proportion of additional diagnoses captured by CCT, and using it to evaluate the sensitivity and specificity of the CXR. Outcome variables were diagnoses made by CCT as well as the ensuing changes in the clinical management attributable to the diagnoses reported by the CCT and not observed by the CXR.

RESULTS

In 1235 children, CCT was associated with diagnosing higher proportions of contusion or atelectasis (60% vs 31%; p < .0001), pneumothorax (23% vs 9%; p < .0001), rib fracture (18% vs 7%; p < .0001), other fracture (20% vs 10%; p < .0001), diaphragm rupture (0.2% vs 0.1%; p = .002), and incidental findings (7% vs 2%; p < .0001) as compared to CXR. CCT findings changed the management of 107 children (8.7%) with 32 (2.6%) of the changes being surgical procedures. The overall sensitivity and specificity of the CXR were 57.9% (95% CI: 54.5-61.2) and 90.2% (95% CI: 86.8-93.1), respectively. The positive predictive value and negative predictive value were 93.1% and 48.6%, respectively.

CONCLUSION

CXR is a useful initial screening tool to evaluate pediatric trauma patients along with clinical presentation in the Emergency Department in children.

LEVEL OF EVIDENCE

Level III, diagnostic test.

Blunt traumatic scapular fractures are associated with great vessel injuries in children

BACKGROUND

Patients with stable blunt great vessel injury (GVI) can have poor outcomes if the injury is not identified early. With current pediatric trauma radiation reduction efforts, these injuries may be missed. As a known association between scapular fracture and GVI exists in adult blunt trauma patients, we examined whether that same association existed in pediatric blunt trauma patients.

METHODS

Bluntly injured patients younger than 18 years old were identified from 2012 to 2014 in the National Trauma Data Bank. Great vessel injury included all major thoracic vessels and carotid/jugular. Demographics of patients with and without scapular fracture were compared with descriptive statistics. The χ test was used to examine this association using SAS Version 9.4 (SAS Institute, Inc, Cary, NC).

RESULTS

We found a significant association between pediatric scapular fracture and GVI. Of 291,632 children identified, 1,960 had scapular fractures. Children with scapular fracture were 10 times more likely to have GVI (1.2%) compared to those without (0.12%, p < 0.0001). Most common GVI seen were carotid artery, thoracic aorta, and brachiocephalic or subclavian artery or vein. Children with both scapular fracture and GVI were most commonly injured by motor vehicles (57% collision, 26% struck).

CONCLUSIONS

Injured children with blunt scapular fracture have a 10-fold greater risk of having a GVI when compared to children without scapular fracture. Presence of blunt traumatic scapular fracture should have appropriate index of suspicion for a significant GVI in pediatric trauma patients.

LEVEL OF EVIDENCE

Epidemiologic and prognostic study, level III; Therapeutic, level IV.

Predicting Thoracic Injury in Children With Multitrauma

OBJECTIVES

Previous pediatric trauma studies focused on predictors of abnormal chest radiographs or included patients with low injury severity. This study identified predictors of thoracic injury (TI) diagnoses in a high-risk population and determined TI rate without predictors.

METHODS

This study was a retrospective trauma registry analysis of previously healthy children aged 0 to 17 years with multisystem blunt trauma requiring trauma team activation and chest radiography who were divided into those with and without TI. Plausible TI predictors included Glasgow Coma Scale score of 13 or less, abnormal thoracic symptoms/signs, abnormal chest auscultation, respiratory distress/ rate higher than the 95th percentile, oxygen saturation less than 95%, abnormal abdominal signs/symptoms, tachycardia higher than the 95th percentile, blood pressure lower than the 5th percentile, and femur fracture.

RESULTS

One hundred forty-one (29%) of 493 eligible patients had TI. Independent TI predictors include thoracic symptoms/signs (odds ratio [OR], 6.0; 95% confidence interval [CI], 3.6-10.1), abnormal chest auscultation (OR, 3.5; 95% CI, 2.0-6.2), saturation less than 95% (OR, 3.1; 95% CI, 1.8-5.5), blood pressure lower than the 5th percentile (OR, 3.7; 95% CI, 1.1-12.2), and femur fracture (OR, 2.5; 95% CI, 1.2-5.4). Six (5%) of 119 children (95% CI, 0.01-0.09) without predictors had TI.

CONCLUSIONS

Predictors of TI include thoracic symptoms/signs, abnormal chest auscultation, saturation less than 95%, blood pressure lower than the 5th percentile, and femur fracture. Because an important portion of children without predictors had TI, chest radiography should remain part of pediatric trauma resuscitation.

Does the incidence of thoracic aortic injury warrant the routine use of chest computed tomography in children?

BACKGROUND

Thoracic aortic injury is a potentially life-threatening injury associated with rapid deceleration mechanisms. Diagnosis is made by chest computed tomography (CT), which is associated with a risk of radiation-induced malignancy. We sought to determine the incidence of aortic injuries in the pediatric population to weigh against the risk of CT imaging.

METHODS

The Pediatric Health Information Systems was queried for children ≤18 years with discharge diagnosis code of thoracic aortic injury (901.0) between December 2004 and 2014. Data abstracted included patient age, gender, diagnosis and procedure codes, and discharge disposition, where available. We also queried for imaging codes to determine what type of chest imaging the child received.

RESULTS

Between December 2004 and 2014, 311,850 children were admitted to Pediatric Health Information Systems hospitals with traumatic injury. Of these patients, 46 (0.015%) were coded with a thoracic aortic injury and an accompanying E-code. Twenty-seven patients (58.7%) were male, and the median age was 13 years. The most common mechanism of injury was motor vehicle collision (63%, n = 29). Eighteen hospitals (41.9%) had no patients with a thoracic aortic injury in the 10-year period. In children with a thoracic aortic injury, the mortality rate was 11% (n = 5) and 22 (47.8%) underwent a chest CT during their hospitalization. Forty percent (124,909) of all trauma patients underwent chest CT, with a positive rate for aortic injury of 1.8/10,000. The reported estimated cancer risk from a chest CT scan is 25/10,000 for girls and 7.5/10, 000 in boys, greater than the positive CT rate.

CONCLUSION

Thoracic aortic injuries are rare in children in the United States. The risk of cancer associated with screening chest CT is greater than the likelihood of identifying an aortic injury. Therefore, screening chest CT scans are unwarranted in injured children.

LEVEL OF EVIDENCE

Therapeutic/Care Management, level IV.

Review of Interventions and Radiation Exposure from Chest Computed Tomography in Children with Blunt Trauma

OBJECTIVE

To determine the radiation risk to a child undergoing trauma evaluation with chest computed tomography (CCT) for every clinically actionable injury identified.

STUDY DESIGN

This observational, cross-sectional study included all blunt trauma patients under 18 years of age undergoing CCT in a single urban emergency department. Via a retrospective chart review, therapeutic interventions done exclusively for chest injuries identified on CCT scan were identified. Effective radiation from each CCT was calculated and averaged and the dose required to diagnose 1 management-changing chest injury was determined.

RESULTS

Of 209 children undergoing CCT over a 19-month period, 168 were victims of blunt trauma. Ten required an intervention specifically for a chest injury identified on CCT (suggesting development of 1 malignancy per 37 actionable injures identified). None required an intervention for an injury exclusively noted on CCT, as all 10 actionable injuries were apparent via other modalities (radiograph, ultrasound examination, clinical examination).

CONCLUSION

Although 10 uniquely actionable injuries were identified on CCT, none were found only on CCT. Because CCTs rarely modified management, the amount of radiation administered per management change was sufficiently high to recommend reconsideration of current imaging practice in this single-center study.

Pediatric Thoracic Trauma: Recognition and Management

Thoracic injuries account for less than one-tenth of all pediatric trauma-related injuries but comprise 14% of pediatric trauma-related deaths. Thoracic trauma includes injuries to the lungs, heart, aorta and great vessels, esophagus, tracheobronchial tree, and structures of the chest wall. Children have unique anatomic features that change the patterns of observed injury compared with adults. This review article outlines the clinical presentation, diagnostic testing, and management principles required to successfully manage injured children with thoracic trauma.

Pediatric thoracic trauma: Current trends

Pediatric thoracic trauma is relatively uncommon but results in disproportionately high levels of morbidity and mortality when compared with other traumatic injuries. These injuries are often more devastating due to differences in children׳s anatomy and physiology relative to adult patients. A high index of suspicion is of utmost importance at the time of presentation because many significant thoracic injuries will have no external signs of injury. With proper recognition and management of these injuries, there is an associated improved long-term outcome. This article reviews the current literature and discusses the initial evaluation, current management practices, and future directions in pediatric thoracic trauma.

Predictors of Intrathoracic Injury after Blunt Torso Trauma in Children Presenting to an Emergency Department as Trauma Activations

BACKGROUND

Thoracic injuries are a major cause of death associated with blunt trauma in children. Screening for injury with chest x-ray study, compared with chest computed tomography (CT) scan, has been controversial, weighing the benefits of specificity with the detriment of radiation exposure.

OBJECTIVE

To identify predictors of thoracic injury in children presenting as trauma activations to a Level I trauma center after blunt torso trauma, and to compare these predictors with those previously reported in the literature.

METHODS

We performed a retrospective chart review of pediatric patients (<18 years of age) who presented to the Emergency Department of a Level I trauma center between June 2010 and June 2013 as a trauma activation after sustaining a blunt torso trauma and who received diagnostic imaging of the chest as part of their initial evaluation.

RESULTS

Data analysis was performed on 166 patients. There were 33 patients (20%) with 45 abnormalities detected on diagnostic imaging of the chest, with the most common abnormalities being lung contusion (36%), pneumothorax (22%), and rib fracture (13%). Statistically significant predictors of abnormal diagnostic imaging of the chest included Glasgow Coma Scale score (GCS) < 15 (27% with abnormality vs. 13% without abnormality), hypoxia (22% vs. 5%), syncope/loss of consciousness (55% vs. 35%), cervical spine tenderness (12% vs. 3%), thoraco-lumbar-sacral spine tenderness (41% vs. 17%), and abdominal/pelvic tenderness (12% vs. 3%).

CONCLUSIONS

Based on our data, predictors of thoracic injury in children after blunt torso trauma include GCS < 15, hypoxia, syncope/dizziness, cervical spine tenderness, thoraco-lumbar-sacral spine tenderness, and abdominal/pelvic tenderness.

The use of whole body computed tomography scans in pediatric trauma patients: Are there differences among adults and pediatric centers?

INTRODUCTION

Whole body CT (WBCT) scan is known to be associated with significant radiation risk especially in pediatric trauma patients. The aim of this study was to assess the use WBCT scan across trauma centers for the management of pediatric trauma patients.

METHODS

We performed a two year (2011-2012) retrospective analysis of the National Trauma Data Bank. Pediatric (age≤18years) trauma patients managed in level I or II adult or pediatric trauma centers with a head, neck, thoracic, or abdominal CT scan were included. WBCT scan was defined as CT scan of the head, neck, thorax, and abdomen. Patients were stratified into two groups: patients managed in adult centers and patients managed in designated pediatric centers. Outcome measure was use of WBCT. Multivariate logistic regression analysis was performed.

RESULTS

A total of 30,667 pediatric trauma patients were included of which; 38.3% (n=11,748) were managed in designated pediatric centers. 26.1% (n=8013) patients received a WBCT. The use of WBCT scan was significantly higher in adult trauma centers in comparison to pediatric centers (31.4% vs. 17.6%, p=0.001). There was no difference in mortality rate between the two groups (2.2% vs. 2.1%, p=0.37). After adjusting for all confounding factors, pediatric patients managed in adult centers were 1.8 times more likely to receive a WBCT compared to patients managed in pediatric centers (OR [95% CI]: 1.8 [1.3-2.1], p=0.001).

CONCLUSIONS

Variability exists in the use of WBCT scan across trauma centers with no difference in patient outcomes. Pediatric patients managed in adult trauma centers were more likely to be managed with WBCT, increasing their risk for radiation without a difference in outcomes. Establishing guidelines for minimizing the use of WBCT across centers is warranted.

Nuking the radiation: minimizing radiation exposure in the evaluation of pediatric blunt trauma

PURPOSE OF REVIEW

Our objective is to highlight recent literature investigating low-radiation diagnostic strategies in the evaluation of pediatric trauma.

RECENT FINDINGS

In the area of minor head injury, research has focused on implementation of validated clinical decision rules into practice to reduce unnecessary computed tomography scans. Clinical observation may also serve as an adjunct to initial assessment and a potential substitute for computed tomography imaging. Subgroups of children with special needs or severe injury mechanisms may also be safely characterized by the clinical decision rule and spared radiation exposure. Physical examination techniques may be useful in diagnosing mandibular fractures. In addition, evidence suggests that plain radiography for evaluation of blunt thoracic trauma may be sufficient in many cases, and computed tomography could be reserved for those with abnormal radiographs, high-risk mechanisms, or abnormal physical findings. Clinical decision rules are able to predict intra-abdominal injury with high sensitivity. Data suggest that skeletal surveys may be modified to limit radiation exposure in the case of suspected nonaccidental trauma.

SUMMARY

More research is needed in development of pediatric-specific clinical decision rules and risk stratification and in testing low-radiation diagnostic modalities in the pediatric trauma population.