Scholarly article on topic 'Multi-detector computed tomography imaging of blunt chest trauma'

Multi-detector computed tomography imaging of blunt chest trauma Academic research paper on "Clinical medicine"

CC BY-NC-ND
0
0
Share paper
Keywords
{MDCT / Chest / "Blunt trauma" / "Chest X-ray"}

Abstract of research paper on Clinical medicine, author of scientific article — Naglaa L. Dabees, Alsiagy A. Salama, Samar Abd Elhamid, Mohab M. Sabry

Abstract Background and purpose Chest trauma is a significant cause of mortality and morbidity, especially in the younger population. The purpose of this study was to evaluate the role of multi-detector computed tomography (MDCT) in the assessment of patients with blunt chest trauma. Patients and methods A prospective study was conducted on thirty (30) patients with blunt chest trauma (21 males and 9 females, aged from 6 to 62years) and 29 control patients presented with any trauma other than blunt chest trauma (23 males and 6 females, aged from 10 to 68years) at the Emergency Department, Tanta University Hospital, from January 2013 to February 2014. Cases were subjected to clinical evaluation and radiological assessment of the chest using conventional chest X-ray (CXR) and multi-detector computed tomography. Results The most common mode of injury was motor vehicle accidents (56.7%). On MDCT scan, the frequency of chest injuries were; chest wall injuries (86.7%), pleural injuries (80%), parenchymal injuries (56.7%), mediastinal injuries (30%) and finally the dorsal spine injuries (16.7%). MDCT is more sensitive, specific, and accurate than CXR in the assessment of blunt chest trauma and management of patients. Conclusion MDCT is the modality of choice for rapid assessment of emergency chest trauma patients, when chest X-ray was inconclusive.

Academic research paper on topic "Multi-detector computed tomography imaging of blunt chest trauma"

The Egyptian Journal of Radiology and Nuclear Medicine (2014) 45, 1105-1113

Egyptian Society of Radiology and Nuclear Medicine The Egyptian Journal of Radiology and Nuclear Medicine

www.elsevier.com/locate/ejrnm www.sciencedirect.com

ORIGINAL ARTICLE

Multi-detector computed tomography imaging c^Ma*

of blunt chest trauma

Naglaa L. Dabees a, Alsiagy A. Salama a'*, Samar Abd Elhamid a, Mohab M. Sabry b

a Radiodiagnosis Department, Faculty of Medicine, Tanta University, Egypt b Cardiothoracic Surgery Department, Faculty of Medicine, Tanta University, Egypt

Received 1 April 2014; accepted 4 August 2014 Available online 18 September 2014

KEYWORDS

Chest;

Blunt trauma; Chest X-ray

Abstract Background and purpose: Chest trauma is a significant cause of mortality and morbidity, especially in the younger population. The purpose of this study was to evaluate the role of multi-detector computed tomography (MDCT) in the assessment of patients with blunt chest trauma. Patients and methods: A prospective study was conducted on thirty (30) patients with blunt chest trauma (21 males and 9 females, aged from 6 to 62 years) and 29 control patients presented with any trauma other than blunt chest trauma (23 males and 6 females, aged from 10 to 68 years) at the Emergency Department, Tanta University Hospital, from January 2013 to February 2014. Cases were subjected to clinical evaluation and radiological assessment of the chest using conventional chest X-ray (CXR) and multi-detector computed tomography.

Results: The most common mode of injury was motor vehicle accidents (56.7%). On MDCT scan, the frequency of chest injuries were; chest wall injuries (86.7%), pleural injuries (80%), parenchymal injuries (56.7%), mediastinal injuries (30%) and finally the dorsal spine injuries (16.7%). MDCT is more sensitive, specific, and accurate than CXR in the assessment of blunt chest trauma and management of patients.

Conclusion: MDCT is the modality of choice for rapid assessment of emergency chest trauma patients, when chest X-ray was inconclusive.

© 2014 The Egyptian Society of Radiology and Nuclear Medicine. Production and hosting by Elsevier

B.V. All rights reserved.

Abbreviations: MDCT, multi-detector computed tomography; CXR, chest X-ray; x, mean; SD, standard deviation; MPR, multiplanar reformation; PPV, positive predictive value; NPV, negative predictive value; ATLS, Advanced Trauma Life Support.

* Corresponding author. Address: Radiodiagnosis Department, Tanta

Faculty of Medicine, Tanta, Gharbiya, Egypt. Mobile: +20

1113558503; fax: +20 40 3407734.

E-mail address: siagyali33@yahoo.com (A.A. Salama).

Peer review under responsibility of Egyptian Society of Radiology and

Nuclear Medicine.

1. Introduction

Chest trauma is a significant cause of mortality and morbidity, especially in the younger population (1).

Injuries to the thorax are the third most common injuries in trauma patients, next to injuries to the head and extremities. Thoracic trauma has an overall fatality rate of 15-25%, which is the highest in patients with cardiac or tracheobronchial-esophageal

http://dx.doi.org/10.1016/j.ejrnm.2014.08.006

0378-603X © 2014 The Egyptian Society of Radiology and Nuclear Medicine. Production and hosting by Elsevier B.V. All rights reserved.

injuries. Furthermore, the presence of thoracic injuries in the setting of multi-systemic trauma can significantly increase patient mortality. Injuries such as flail chest, lung contusion, hemothorax, and pneumothorax can complicate overall case management (2,3).

More than two-thirds of cases of blunt thoracic trauma in developed countries are caused by motor vehicle collisions. The remaining cases are the result of falls from height or of blows from blunt objects (4).

Imaging plays an important role in the diagnosis of blunt thoracic trauma. The conventional radiography remains the initial study for assessing patients sustaining blunt trauma to the chest, however, in severely injured patients. The ideal upright, full inspiratory PA chest radiography cannot be obtained. Portable supine radiographs are suffering from poor positioning, poor inspiration, or artifact from an underlying backboard or overlying monitoring equipment and many injuries may be difficult to detect in these suboptimal studies (5).

Multidetector-row CT (MDCT) has been recognized and accepted as an effective and fast imaging tool in severely injured trauma patients (6).

MDCT scanners are available in almost all trauma centers. The fast scanning time of MDCT allows for single breath-hold scanning, fewer motion artifacts, and improved contrast bolus imaging. Additionally, thinner collimation provides isotropic voxels, allowing multi-planar reformations while maintaining spatial resolution (5).

Studies have shown that MDCT may demonstrate significant injury (e.g., thoracic aortic injury) in patients with normal initial radiographs (7). Furthermore, MDCT has been credited with changing management in up to 20% of chest trauma patients with abnormal initial radiographs (8).

MDCT is more accurate than radiography for the evaluation of pulmonary contusion, thereby allowing early prediction of respiratory compromise (9). It is also valuable in the diagnosis of fractures of the thoracic spine, especially at the cervico-thoracic junction, which is difficult to evaluate with conventional radiography. In addition, MDCT has helped to exclude thoracic aortic injury, thereby limiting the number of catheter aortographic examinations (10).

2. Aim of the work

The aim of this study was to evaluate the role of multi-detector computed tomography in the assessment of patients with blunt chest trauma.

3. Patients and methods

3.1. Participants

This study was conducted according to the guidelines of the ethics committee of our university and was approved by our institutional review board. Informed written consents were obtained from relatives of all participants in this study.

A prospective study was conducted on 30 blunt chest trauma patients (21 males and 9 females, aged from 6 to 62 years with mean of 32.7 ± 14.3 years) and 29 control trauma patients presented with any trauma other than blunt chest trauma (23 males and 6 females, aged from 10 to 68 years with mean of 34.3 ± 13.2 years) at the Emergency Department

of Tanta University hospital, over a period of one year starting from January 2013 to February 2014 with the following inclusion & exclusion criteria:

- Inclusion criteria: All cases with blunt chest trauma either as a sole presentation or as a part of poly-traumatic insults were included in the study as patients.

While cases with any trauma other than blunt chest trauma were included as controls. No age predilection.

- Exclusion criteria: The following groups of patients were excluded:

1. Patients in need of emergency transfer to surgery.

2. Patients who were hemodynamically unstable.

3. Lactating and pregnant females.

4. Patients known to had sensitivity to the contrast medium.

3.2. Methodology

In this prospective study, all participants were subjected to:

3.2.1. History taking & clinical assessment

3.2.2. Plain chest X-ray

AP (supine) views were taken for 23 patients, five of them with portable radiograph & PA (standing) views could be done for 7 patients.

3.2.3. Multi-detector CT of the chest

All patients underwent MDCT of chest on Siemens Emotion 6 MDCT.

Patients are examined in the supine position and the field of view was adjusted to obtain complete anatomical imaging of the chest.

Thin axial section images (1.25 mm slice thickness). On multi-detector CT (MDCT) scanners volumetric acquisition of high-resolution CT datasets was acquired in the cranio-cau-dal direction from the base of the neck to the level of the renal arteries.

Intravenous contrast media injection: volume (80 ml), concentration (350-400 mg/ml), rate (2.5-3 ml/s) and scanning delay (30-40 s).

3.2.4. Virtual bronchoscopy

Three dimensional reconstruction based on surface and volume rendering, was done for three patients & was used as a gold standard for those patients.

3.2.5. Operative & interventional findings

Operative & interventional findings relevant to chest trauma were obtained & used as a gold standard with clinical findings & follow up in cases underwent conservative treatment.

3.3. Data analysis

The collected data were tabulated and statistically analyzed using SPSS (statistical package for social science) version 16 on Personal Computer. The level of significance was adopted at P < 0.05. Two types of statistics were done: (a) Descriptive statistics included percentage (%), mean (x) and standard devi-

ation (SD), and (b) Analytic statistics included Chi-squared test (v2) to study the association between two qualitative variables and Student's t-test a test of significance used for comparison between two groups having quantitative variables. Sensitivity, specificity, positive predictive value, negative predictive value and accuracy are used for diagnostic test evaluation.

4. Results

Demographic characteristics of studied cases are shown in Table 1, the difference between patients and controls as regards gender and means of ages was statistically non-significant (P > 0.05).

Out of the thirty patients, 70% (n = 21) were males with age ranging from 6 to 62 years (mean = 32.7 years) and most of the patients were in the age group 20-40 years (60%). The most common mechanism of blunt chest trauma was as a result of motor vehicle accidents (56.7%, n = 17) followed by fall from height (23.3%, n = 7) and pedestrian injury (13.3%, n = 4) (Tables 1 and 2).

Eight patients presented (26.7%) with mild to severe degree of coma. Three patients (10%) arrived intubated or were intu-bated in the Emergency Department. Five patients (16.7%) had a chest tube placed prior to MDCT scan (Table 2).

The most common clinical presentations were chest pain (76.7%), local chest tenderness (66.7%), dyspnoea (60%), coma (26.7%) and haemoptysis (16.7%). More than one presentation was also encountered in the same patient (Table 2).

Positive radiological findings among patients in order of frequency were chest wall injuries (86.7%), pleural injuries (80%), parenchymal injuries (56.7%), mediastinal injuries (33.3%) and finally the dorsal spine injuries (16.7%) (Table 3).

Out of a total of thirty patients, rib fractures were detected on chest radiograph in 11 patients (36.7%) and on MDCT scan in 17 patients (56.7%). Sternal fracture was detected in one patient on MDCT, which was not detected on radiography. Clavicular fractures were detected in three patients on chest radiograph and MDCT scan. Scapular fractures were detected in one patient on chest radiograph and in two patients on MDCT scan. Anterior sterno-clavicular dislocation was detected in one patient on chest radiograph and MDCT scan. Subcutaneous emphysema was detected in one patient on chest

Table 2 Admission characteristics of the 30 patients.

Characteristic Patients no. (%) n = 30

Mechanism of injuries

Motor vehicle accidents 17 (56.7)

Fall from height 7 (23.3)

Pedestrian hit by car 4 (13.3)

Assault 2 (6.7)

Glasgow coma scale score

13-15 25 (83.3)

12-9 2 (6.7)

3-8 3 (10.0)

Visible evidence of trauma

Head and neck 19 (63.3)

Chest 11 (36.7)

Extremities 23 (76.7)

Back 7 (23.3)

Tube position

Chest tube 5 (16.7)

Endotracheal tube 3 (10.0)

Clinical presentations

Chest pain 23 (76.7)

Dyspnoea 18 (60.0)

Local tenderness 20 (66.7)

Coma 8 (26.7)

Hemoptysis 5 (16.7)

N.B.: Visible evidences of trauma were defined as any abrasions,

ecchymosis, hematomas, and deformities of extremities.

radiograph and in two patients on MDCT scan, the diagnostic difference between chest radiograph and MDCT scan was statistically significant (P < 0.05) (Table 3).

Regarding pleural injuries, pneumothoraces (simple and tension) were detected in 9 patients (30%) on chest radiograph and in 17 patients (56.7%) on MDCT scan. Hemothorax and hemo-pneumothorax were detected in 3 patients (10%) on chest radiograph and in 7 patients (23.3%) on MDCT scan, the diagnostic difference between chest radiograph and MDCT scan was statistically significant (P < 0.05) (Table 3).

Pulmonary parenchymal injuries were the most common lesions next to pleural injuries. Lung contusions were detected in 9 patients (30%) on chest radiograph and in 15 patients (50%) on MDCT scan. Lung lacerations were detected in 2 patients on MDCT scan, which were not detected on chest radiograph, the diagnostic difference between chest radiograph and MDCT scan was statistically significant (P < 0.05) (Table 3).

Mediastinal injuries were detected in 10 patients (33.3%), mediastinal hematomas were detected in 3 patients (10%) on MDCT scan, which were not detected on chest radiograph. Pneumo-mediastinum was detected in two patients (6.7%) on chest radiograph and in 4 patients (13.3%) on MDCT scan. Tracheal tear was detected and diagnosed in one patient (3.3%) by bronchoscope, which was not detected on either chest radiograph or MDCT scan. Bronchial abnormal findings (Injury & FB) were detected in one patient (3.3%) on chest radiograph and in two patients (6.7%) on MDCT scan, one of the later was due to complete avulsion of left main-stem bronchus and the second one was due to foreign body aspiration (tooth) in the left posterior basal bronchus with post

Table 1 Demographic characteristics of studied cases.

Characteristic Control No. Patients No. P-value

(%) n = 29 (%) n = 30

Gender

Male 23 (79.3) 21 (70.0) 0.412

Female 6 (20.7) 9 (30.0)

Age in years

<10 0 (0.0) 1 (3.3) 0.59

10- 3 (10.0) 4 (13.3)

20- 7 (23.3) 9 (30.0)

30-40 14 (48.3) 9 (30.0)

>40 5 (16.7) 7 (23.3)

Mean of age ± SD 34.3 ± 13.2 32.7 ± 14.3 0.66

Table 3 Number and % of positive radiological findings among 30 patients.

Radiological findings Total n = 30 Chest X-ray MDCT scan P-value

Chest wall injuries 26 (86.7) 17 (56.7) 26 (86.7) 0.0099*

Rib fracture 17 (56.7) 11 (36.7) 17 (56.7)

Fracture sternum 1 (3.3) 0 (0.0) 1 (3.3)

Fracture clavicle 3 (10.0) 3 (10.0) 3 (10.0)

Fracture scapula 2 (6.7) 1 (3.3) 2 (6.7)

Anterior sterno-clavicular dislocation 1 (3.3) 1 (3.3) 1 (3.3)

Subcutaneous emphysema 2 (6.7) 1 (3.3) 2 (6.7)

Pleural injuries 24 (80.0) 12 (40.0) 24 (80.0) 0.0016*

Simple pneumothorax 15 (50.0) 7 (23.3) 15 (50.0)

Tension pneumothorax 2 (6.7) 2 (6.7) 2 (6.7)

Hemothorax 4 (13.3) 2 (6.7) 4 (13.3)

Hemo-pneumothorax 3 (10.0) 1 (3.3) 3 (10.0)

Parenchymal injuries 17 (56.7) 9 (30.0) 17 (56.7) 0.037*

Contusion 15 (50.0) 9 (30.0) 15 (50.0)

Laceration 2 (6.7) 0 (0.0) 2 (6.7)

Mediastinal injuries 10 (33.3) 3 (10.0) 9 (30.0) 0.018*

Mediastinal hematoma 3 (10.0) 0 (0.0) 3 (10.0)

Pneumomediastinum 4 (13.3) 2 (6.7) 4 (13.3)

Tracheal injury 1 (3.3) 0 (0.0) 0 (0.0)

Bronchial injuries 2 (6.7) 1 (3.3) 2 (6.7)

Dorsal vertebral injuries 5 (16.7) 1 (3.3) 5 (16.7) 0.085

Vertebral body fracture 3 (10.0) 1 (3.3) 3 (10.0)

Neural arch fracture 1 (3.3) 0 (0.0) 1 (3.3)

Epidural emphysema 1 (3.3) 0 (0.0) 1 (3.3)

N.B.: more than one finding was encountered in the same patient. = statistically significant difference.

Table 4 Sensitivity, specificity, PPV, NPV and accuracy of chest X-ray (CXR) and MDCT scan findings among 30 patients.

Type of injuries Sensitivity (%) Specificity (%) PPV (%) NPV (%) Accuracy (%)

CXR MDCT CXR MDCT CXR MDCT CXR MDCT CXR MDCT

Chest wall injuries 65.4 100 96.2 100 94.4 100 73.5 100 80.8 100

Pleural injuries 50 100 91.7 100 85.7 100 64.7 100 70.8 100

Parenchymal injuries 52.9 100 88.3 100 81.8 100 62.2 100 70.6 100

Mediastinal injuries 28.6 89 85.7 100 66.7 100 54.6 90 57.2 92.9

Dorsal vertebral injuries 20 100 80 100 50 100 50 100 50 100

obstructive atelectasis, which was not detected on chest radiograph, the diagnostic difference between chest radiograph and MDCT scan was statistically significant (P < 0.05) (Table 3).

Dorsal spine injuries were the least frequent lesions and were detected in 5 patients (16.7%); vertebral body fracture was detected in 1 patient (3.3%) on chest radiograph and in 3 patients (10%) on MDCT scan. Neural arch fracture was detected in 1 patient (3.3%) on MDCT scan, which was not detected on chest radiograph and epidural emphysema secondary to pneumomediastinum was detected in 1 patient on MDCT scan, which was not detected on chest radiograph, the diagnostic difference between chest radiograph and MDCT scan was statistically non-significant, this could be explained by a small number of patients enrolled in this group of injuries (P > 0.05) (Table 3).

Sensitivity, specificity, PPV, NPV and accuracy of chest X-ray and MDCT are shown in Table 4. Sensitivity of chest X-ray for chest wall injuries, pleural injuries, pulmonary

parenchymal injuries, mediastinal injuries and dorsal spine injuries were 65.4%, 50%, 52.9%, 28.6%, and 20% respectively, compared with 100%, 100%, 100%, 89%, and 100% by MDCT scan respectively.

MDCT findings were correlated with surgical and clinical findings. As regards the seventeen patients with rib fractures, 13 of them underwent intercostal tube insertion for pneumo-& hemo-pneumothorax, and four passed under conservative measures. Patients with parenchymal injuries (n = 17); twelve of them underwent surgical intervention and five passed under conservative measures. Regarding the five patients with spinal injuries, three of them passed under conservative measures and two underwent surgical intervention (Figs. 1-7).

5. Discussion

Radiology plays a major role in evaluation of trauma patients. The Advanced Trauma Life Support (ATLS 2004) course

Fig. 1 A 21 years old male presented with severe dyspnea and decreased breath sounds mainly on the RT side. (A) Chest X-ray showing diffuse inhomogeneous opacities in both lung fields with subcutaneous emphysema on the right side. (B) Axial CT section, lung window image showing bilateral diffuse consolidations with air bronchogram more on the right side with diffuse parenchymal contusions. (C) Coronal MPR image of the same patient showing the cranio-caudal extension of the parenchymal lung contusions.

Fig. 2 A 58 years old male. (A) X-ray on supine position showing bilateral pneumothorax, subcutaneous emphysema and bilateral rib fractures. (B) CT image of the same patient showing bilateral pneumothorax, subcutaneous emphysema, and pneumomediastinum (red arrow) which are not seen in the X-ray (Occult pneumomediastinum).

Fig. 3 A 40 years old male patient with suspected vascular injury after motor vehicle accident. (A) Axial image, lung window showing fracture of the RT 1st rib at the sterno-chondral junction (red arrow), patchy area of lung contusion in the right upper lung lobe and subcutaneous emphysema. (B) Axial image, mediastinal window, showing retrosternal hematoma (green arrow), traces of bilateral pleural effusion. (C) Sagittal MPR image created with MIP showing the clear fat plane between the retro-sternal hematoma and the intact aorta (green arrow).

recommended performing the plain film radiography of the chest, abdomen, and cervical spine in all the blunt trauma patients. Nowadays, Chest computed tomography (CCT) is being used with increasing frequency in the evaluation of blunt chest trauma. Multi-detector computed tomography frequently detects injuries not seen on routine initial chest X-ray (occult findings) (11).

In our study, the chest wall injuries were the most common findings, they were seen in 26 patients (86.7%), with rib

fractures being the most frequent (56.7%). MDCT was the most sensitive (100%) technique for imaging rib fractures, and chest radiography had limited sensitivity (65.4%). This result coincided with Primak and Collins (12) who reported that rib fractures were the most common findings after blunt chest trauma with an incidence reported up to 40%. Chest radiography is routinely used to assist in the diagnosis of rib fractures, even though it has limited sensitivity. It is even more insensitive in showing costochondral fractures. MDCT is the

Fig. 4 A 25 years old male presented with multi-trauma. Clinically presented with chest pain, dyspnoea and local tenderness. (A) Sagittal MPR image showing comminuted fracture of right scapula (red arrows), right lung contusion and laceration. (B) Axial CT section at the apex of the lung, lung widow showing comminuted fracture of the right scapula, traces of right pleural effusion and small amount of left pneumothorax. (C) Coronal MPR image showing clearly the comminuted fracture of the right scapula.

Fig. 5 A 27 years old male patient presented with marked dyspnea and absent breath sounds on the LT side. (A) Axial image mediastinal window showing collapsed left lung without air bronchogram and fallen into the dependent lateral position (fallen lung sign) signs of complete transection of the main stem bronchus. (B) Coronal MPR image, lung window demonstrating the abrupt cessation of the left main stem bronchus (also a large RT side laceration surrounded by small areas of contusions is noted). (C) Virtual bronchoscopy showing the left main stem bronchus tear and patent right bronchus.

Fig. 6 A 61 years old male patient presented by coma after motor vehicle accident. (A) Coronal reformatted image showing aspirated tooth in the left posterior basal bronchus with post obstructive atelectasis. (B) Virtual bronchoscopy showing the obstruction of the bronchus by the aspirated tooth.

most sensitive technique for imaging rib fractures, since it can help in determining the site and number of fractures and, more importantly, provides information regarding any associated injuries.

Also our results were in agreement with Kerns and Gay (13) who stated that rib fractures occur in 56% of patients with major blunt chest trauma but many of these fractures are

missed on chest radiographs possibly due to difficulties in obtaining good radiographic posterior views.

In this study, sternal fracture was found in one patient (3.3%). This finding was consistent with the findings reported in a study by Athanassiadi et al. (14) who reported that sternal fractures occur in approximately 3-8% of patients who experience blunt chest trauma, and are seen most commonly in

Fig. 7 A 27-years old male patient presented by multi-trauma after motor vehicle accident. (A) Three dimensional reconstruction CT image showing multiple right rib fractures including 5th, 6th & 7th ribs posteriorly and 9th rib laterally.

deceleration injuries or direct blows to the anterior chest wall. Also Crestanello et al. (15) found that fractures of sternum are occurring in 1.5-4% of blunt chest trauma.

In the present study, scapular fractures were detected in two patients (6.7%) on MDCT scan and in one patient (3.3%) on chest radiograph. It is near to the percent reported by Weening et al. (16) who reported that fractures of the scapula are uncommon, accounting for 3-5% of all shoulder girdle fractures and occurring in 3.7% of patients with multiple injuries. Also Kerns and Gay (13) reported that scapular fractures are overlooked or obscured on chest radiograph in as many as 35% of patients.

The two cases of scapular factures which were found in our study were associated with other findings which were in agreement with the explanation made by Veysi et al. (17) and Lunsjo et al. (18) who reported that scapular fractures indicate highforce trauma because the scapula is enveloped and protected by the large muscle masses of the posterior thorax. Isolated fractures are rare. Typically, scapular fractures are seen in a patient who has a severe chest trauma as the result of a motor vehicle accident or a fall from height. They are commonly associated with other injuries including pneumothorax, hemo-thorax, pulmonary injuries, and spinal injuries.

In the present study, we reported 3 patients (10%) with clavicular fractures and one patient (3.3%) with anterior sterno-clavicular dislocation which were detected on both chest radiography and MDCT scan, these results coincided with Shanm-uganathan and Mirvis (19) who reported that clavicular fractures from blunt chest trauma account for 3-11%, with anterior dislocations being the most common and usually without clinical significance.

This study included 5 patients (16.7%) with thoracic spine fractures with 100 percent sensitivity by MDCT scan compared to 20% by chest radiography. This finding was similar to that described by Denis (20) who reported that thoracic spine fractures account for 13-30% of all spine fractures and the thoracic region of the spine has a relatively high stability because of the stabilizing effects of the ribs and the rib cage so injuries that result in fracture are usually caused by high energy. Most thoracic spine injuries occur in flexion and axial loading because rotation in the upper thoracic spine is limited by the rib cage.

Also our results coincided with Meyer (21) who reported that spine fractures are usually difficult to detect on routine chest radiographs, especially those located in the upper portion and MDCT is much more sensitive for diagnosing thoracic spine fractures and is the imaging modality of choice.

In the present study, subcutaneous emphysema was found in 2 patients (6.7%) with 100 percent sensitivity by MDCT scan compared to 50% by chest radiography. These findings were in agreement with Criner and D'Alonzo (22) who stated that on a chest radiograph, subcutaneous emphysema may be seen as radiolucent striations which may interfere with radiography of the chest, potentially obscuring serious conditions such as pneumothorax.

Also Wicky et al. (23) reported that subcutaneous emphysema can also be seen on MDCT scans, with the air pockets appearing as dark areas. MDCT scanning is so sensitive that it commonly makes it possible to find the exact spot from which air is entering the soft tissues.

In this study, MDCT chest scanning was significantly more effective in detecting pneumothoraces and hemopneumotho-races, lung contusions, pneumomediastinum and mediastinal hematomas compared with a chest X-ray. This was in accordance with several studies that have shown a greater sensitivity for a MDCT chest scan for detecting intrathoracic injuries (24,25).

In our study, pleural injuries were the most frequent findings following chest wall injuries. Simple pneumothorax was detected in 15 patients (50%), tension pneumothorax in two patients (6.7%) and hemothorax in 4 patients (13.3%) with 100 percent sensitivity by MDCT scan compared to 50% by chest radiography, this was in agreement with Tocino et al. (26), who stated that the most common cause of pneumothorax in blunt chest trauma is a rib fracture that lacerates the lung, but it may also be caused by ruptured alveoli due to a sudden increase in intra-thoracic pressure or to blunt crushing force or deceleration force to the chest.

Also the diagnosis of pneumothorax is usually made by chest radiography. However, De Moya et al. (27) reported that 10-50% of pneumothoraces from blunt trauma are not visualized on chest radiography performed in supine patients as the air in the pleural space accumulates anteriorly and medially but can be seen on MDCT. This type of pneumothorax is called occult pneumothorax.

In our study, hemothorax was found in 4 patients (13.3%), two of them were associated with pneumothorax (hemo-pneumothorax). Hemothorax can originate from injury to the pleura, chest wall, lung, diaphragm, or mediastinum. This was in agreement with Shanmuganathan and Mirvis (19) who stated that MDCT is highly sensitive in detecting a small hemothorax. In addition, the Hounsfield unit (HU) measurement of fluid in the pleural space can be used to identify the origin of the fluid.

In the present study pulmonary contusions were the most common parenchymal injury detected. It was found in 15 patients (50%), with 100 percent sensitivity by MDCT scan. This is accepted by Cohn (28) who reported that pulmonary contusion is the most common lung injury from blunt chest trauma, with a prevalence of 17-70% and chest MDCT is highly sensitive in identifying pulmonary contusion and may help in predicting the need for mechanical ventilation.

In the present study, parenchymal lacerations were found in 2 patients (6.7%) and MDCT scan was highly sensitive in

detecting lung lacerations compared to poor sensitivity by chest radiography. This was in accordance with several studies that have shown that pulmonary lacerations were considered an uncommon injury before the widespread use of MDCT in trauma patients as these were not frequently identified on chest radiographs (29).

In the present study, there were four patients with pneumo-mediastinum (13.3%). Anastasia and Panos (30) reported that pneumomediastinum occurs in 10% of patients with blunt chest trauma.

In the present study, tracheal tear was diagnosed by bron-choscope in one patient (3.3%), which was not detected on radiography or MDCT scan that might be due to small lesion size with a sensitivity of 89% of MDCT in detecting tracheal tear. This result was in accordance with Jen et al. (31) who reported that CT has a sensitivity of 85% in detecting tracheal tears.

Bronchial injuries were detected in two patients (6.7%) on MDCT scan, and in one patient (3.3%) on chest radiograph. This was in agreement with Rathachai et al. (4) who reported that tracheobronchial injuries are rare in clinical practice because most patients die before arriving at the Emergency Department, from either associated injuries to vital structures, hemorrhage, tension pneumothorax, or respiratory insufficiency or from an airway injury. In clinical series, blunt tra-cheobronchial trauma has been reported as accounting for 0.2-8% of all cases of blunt chest trauma.

As regards radiation exposure particularly in young trauma patients, Catherine et al. (32) stated that multiple injured trauma patients receive a substantial dose of radiation. Radiation exposure is cumulative. The low individual risk of cancer becomes a greater public health issue when multiplied by a large number of examinations. Though CT scans are becoming more easily accessible, they should not replace careful clinical examination and should be used only in appropriate patients. In our study we followed a similar protocol, we utilized imaging appropriately to minimize radiation exposure by selecting patients necessary in need of CT study, using organ specific protocols and shielding for other vulnerable body regions from scattered radiation.

Reported limitations of this study were bad quality of supine radiographs due to poor positioning, poor inspiration, or artifact from an underlying backboard or overlying monitoring equipment and many injuries might be difficult to detect in these suboptimal studies.

6. Conclusion

Chest radiograph remains the initial screening modality of trauma patients. However multi-detector computed tomography (MDCT) is the modality of choice for rapid assessment of emergency chest trauma patients, as chest X-ray in these cases has limited sensitivity with missed findings possibly due to difficulties in obtaining good radiographic views, particularly in cases with subcutaneous emphysema, mediastinal injuries, costo-chondral, scapular & parenchymal lung injuries.

Multi-planner capability and 3D reconstruction images are sensitive in the evaluation of skeletal injuries and essential for optimal surgical approach. Its high resolution provides more sensitivity in the evaluation of lung parenchymal lesions. MDCT is more accurate and sensitive in the diagnosis, locali-

zation and characterization of different types of pleural and

mediastinal injuries.

Conflict of interest

None to declare.

References

(1) Mirka Hynek, Ferda Jiri, Baxa Jan. MDCT of blunt chest trauma: indications, technique and interpretation. Insights Imaging 2012;3:433-49.

(2) Peters S, Nicolas V, Heyer CM. MDCT-spectrum of blunt chest wall and lung injuries in polytraumatized patients. Clin Radiol 2010;665:333-8.

(3) Clark DE, Fantus RJ. National Trauma Data Bank (NTDB), American College of Surgeons Annual Report. Chicago. American College of Surgeons 2007; 1-64.

(4) Kaewlai Rathachai, Avery Laura L, Asrani Ashwin V, Novelline Robert A. Multidetector CT of blunt thoracic trauma. Radiographics 2008;28:1555-70.

(5) Euathrongchit J, Thoongsuwan N, Stern EJ. Nonvascular mediastinal trauma. Radiol Clin North Am 2006;44:251-8.

(6) Demehri Shadpour, Rybicki Frank J, Desjardins Benoit, et al. Blunt chest trauma-suspected aortic injury. ACR Appropriateness criteria. Emerg Radiol 2012;19(4):287-92.

(7) Exadaktylos AK, Sclabas G, Schmid SW, Schaller B, Zimmermann H. Do we really need routine computed tomographic scanning in the primary evaluation of blunt chest trauma in patients with "normal" chest radiograph? J Trauma 2001;51: 1173-6.

(8) Omert L, Yeaney WW, Protetch J. Efficacy of thoracic computerized tomography in blunt chest trauma. Am Surg 2001;67: 660-4.

(9) Livingston DH, Haurer CJ. Trauma to the chest wall and lung. In: Moore EE, Feliciano DV, Mattox KL, editors. Trauma. Philadelphia, Pa: McGraw-Hill; 2004. p. 507-37.

(10) Downing SW, Sperling J, Mirvis SE, et al. Experience with spiral CT scanning as the sole diagnostic method for traumatic aortic rupture. Ann Thorac Surg 2001;72:495-501.

(11) Shorr RM, Crittenden M, Indeck M, Hartunian SL, Rodriguez A. Blunt thoracic trauma. Analysis of 515 patients. Ann Surg 2000;206(2):200-5.

(12) Primak SL, Collins J. Blunt nonaortic chest trauma: radiographic and CT findings. Emerg Radiol 2002;9:5-12.

(13) Kerns SR, Gay SB. CT of blunt chest trauma. AJR Am J Roentgenol 1990;154:5-60.

(14) Athanassiadi K, Gerazounis M, Moustardas M, et al. Sternal fractures: retrospective analysis of 100 cases. World J Surg 2004;26:1243-6.

(15) Crestanello JA, Samuels LE, Kaufman MS. Sternal fracture with mediastinal hematoma: delayed cardiopulmonary sequelae. J Trauma 1999;47:161-4.

(16) Weening B, Walton C, Cole PA, Alanezi K, Hanson BP, Bhandari M. Lower mortality in patients with scapular fractures. J Trauma 2010;59:1477-81.

(17) Veysi VT, Mittal R, Agarwal S, et al. Multiple trauma and scapular fractures: so what? J Trauma 2009;55(6):1145-7.

(18) Lunsjo K, Tadros A, Czechowski J, Abu-Zidan FM. Scapular fractures and associated injuries in blunt trauma: a prospective study [abstr]. J Bone J Surg Br 2006;88:141-b.

(19) Shanmuganathan K, Mirvis SE. Imaging diagnosis of non aortic thoracic injury. Rdiol Clin N Am 1999;37:533-51.

(20) Denis F. The three column spine and its significance in the classification of acute thoracolumbar spinal injuries. Spine 2003;8(8):817-31.

(21) Meyer S. Thoracic spine trauma. Semin Roentgenol 1992;27: 254-61.

(22) Criner GJ, D'Alonzo GE. Critical Care Study Guide: Text and Review. Berlin: Springer; 2002, pp. 169.

(23) Wicky S, Wintermark M, Schnyder P, Capasso P, Denys A. Imaging of blunt chest trauma. Eur Radiol 2000;10(10):1524-38.

(24) Marts B, Durham R, Shapiro M, et al. Computed tomography in the diagnosis of blunt thoracic injury. Am Surg 1994;168(6): 688-92.

(25) Wilson D, Voystock JF, Sariego J, et al. Role of computed tomography scan in evaluating the widened mediastinum. Am Surg 1994;60:421-3.

(26) Tocino IM, Miller MH, Fairfax W. Distribution of pneumothorax in the supine and semi recumbent critically ill adult. Am J Roentgenol 1995;144:901-5.

(27) De Moya MA, Seaver C, Spaniolas K, et al. Occult pneumothorax in trauma patients: development of an objective scoring system. J Trauma 2007;63:13-7.

(28) Cohn SM. Pulmonary contusion: review of the clinical entity. J Trauma 1997;42:973-9.

(29) Rivas LA, Fishman JE, Munera F, Bajayo DE. Multislice CT in thoracic trauma. Radiol Clin N Am 2003;41(599-616):7.

(30) Oikonomou Anastasia, Prassopoulos Panos. CT imaging of blunt chest trauma. Insights Imaging 2011;2:281-95.

(31) Jen DC, Kathirkamanathan S, Mirvis Stuart E, Killeen Karen L, Dutton Richard P. Using CT to diagnose tracheal rupture. Am. J. Roentgenol. 2001;176(5).

(32) Hui Catherine M, MacGregor John H, Tien Homer C, Kortbeek John B. Radiation dose from initial trauma assessment and resuscitation. Can. J. Surg. 2009;52(2):147-52.