Bertil Leidner1 MD, Madeleine Adiels1, Peter Aspelin3 MD, PhD,

Per Gullstrand2 MD, PhD, Staffan Wallén2 MD

Departments of Diagnostic Radiology1 and Surgery2 Oskarshamn´s Hospital, S-572 28 Oskarshamn and Department of Diagnostic Radiology³, Huddinge University Hospital, S-141 86 Huddinge, Sweden

Corresponding author, offprint requests:

Bertil Leidner, MD

Department of Diagnostic Radiology

Oskarshamn´s Hospital

Box 701,

S-572 28 Oskarshamn, Sweden

Phone: +46 491782000

Fax: +46 491782637

E-mail: bertil.leidner@alinks.se

Purpose: To develop a standardized CT protocol including head, body and proximal extremities in order to achieve a good time efficiency and diagnostic accuracy in the radiological evaluation of the multitraumatized patient.

Material and Methods: A protocol starting with contiguous 10 mm slices of the head without i.v. contrast medium injection, followed by 10 mm slices every 30 mm through thorax-abdomen-pelvis after i.v. contrast medium enhancement was used (with occasional modifications) to examine 111 blunt trauma patients. All data in the medical reports was collected and used as "end-point", and the outcome of the CT examination was compared to this final diagnosis.

Results: Mean examination time was 20 minutes (range 12–32 minutes). In total 56 head injuries, 89 thoracic injuries, 27 abdominal/pelvic injuries and 62 fractures were found. CT correctly identified the injuries except one brain stem injury, one contusion/rupture of the heart, one hepatic injury, two intestinal injuries, and 9 skeletal injuries.

Conclusion: A standardized CT examination of the head and body can be achieved in 20 min. Its diagnostic accuracy was found to be high, why we recommend it to be the method of choice as initial radiological examination of multitraumatized patients.

Keywords: Multiple trauma, CT examination, CT diagnosis, CT protocol.

In the management of trauma victims two principle concepts have gained wide acceptance, namely the Advanced Trauma Life Support (ATLS) and "The Golden Hour".

The trauma care education world-wide has the last decade evolved towards the ATLS-system, a standardized way for management of the trauma patient by a team, working with clearly defined roles and in a standardized manner.

The concept of the Golden Hour stresses the fact that the patients chances for survival increases the sooner trauma care is instituted after the injury.

To clinically evaluate the full extent of a multitraumatized patient’s injuries is extremely difficult, especially when the patient is unconscious. The radiological work-up is therefore a vital part of the evaluation.

To comply with the concepts above, the radiological examination must be fast, systematic and as complete as possible. Since moving the patient is time-consuming, we have chosen to use only one imaging method, i.e. computed tomography (CT), also because it is the method of choice in imaging of acute head injury and mandatory in an unconscious trauma patient. Data also support that scanning is necessary when a conscious trauma patient has suffered a period of unconsciousness or has amnesia for the accident [ 1, 2, 3, 4 ] . After the head scan is made, our protocol takes advantage of the fact that the patient already is placed in the scanner and the scanning time of the rest of the body adds very little time to the whole scanning procedure and gives vital information. In this way, we have structured and standardized a CT examination to include the head, body and proximal extremities, possible to perform in 20 minutes. The examination is performed as an intimate team work, ideally including a senior radiologist, an anaesthesiologist and the trauma chief surgeon.

To evaluate a standardized multitrauma CT protocol with regard to diagnostic accuracy and time efficiency.

TECHNIQUE:

CT-scanner: A third generation CT-scanner, Philips Tomoscan LX, (Philips, Holland) was used with a performance of 4-5 reconstructed scans per minute.

Protocol (Fig. 1): A lateral scanogram covering the cervical spine and head (I) is followed by contiguous 10 mm slices of the head (II) from the vertex to the skull base (after 2 years modified to include C1- C2)(Fig. 2 a). If possible, the patients arms are then placed above the head in order to minimize artifacts.

The trunk is examined with 10 mm slices every 30 mm (III) from the jugular fossa to the pubic symphysis with i.v. contrast medium administration (Omnipaqueâ 300 mg/ml or Visipaqueâ 270 mg/ml, Nycomed, Norway) in a bolus injection of 50 ml, 2 ml per second followed by an infusion of 70 ml, 0.5 ml per second (Fig. 2 b). Next, the patient is positioned into the gantry as far as possible in order to obtain two frontal scanograms, the first to visualize the skeleton, covering the proximal part of the femurs and pelvic bone (IV) (Fig. 2 c), the latter to obtain an overview of the lungs (V) (Fig. 2 d).

Patient preparation and handling: In order to increase the sensitivity for intestinal injury we routinely administer peroral contrast, either by drinking or by nasogastric tube, if this was possible to do without undue time delay in the emergency room. The patient is transported to the CT-room with the cervical spine immobilized in a collar, without prior X-ray, and handled with care as if with cervical spine injury. In the CT suite the patient is monitored by an anesthesiologist/nurse. The radiologist reads the examination on the monitor, immediately conveying the results to the trauma surgeon. Interventional procedures like thoracic drainage may be performed in the suite.

PATIENTS AND RECORDS:

111 trauma victims, with either clinical suspicion of multiple organ injuries or with a mechanism of injury capable of producing major injury to multiple organ systems, were examined according to the multitrauma CT-protocol during 1991–95, in Oskarshamn´s hospital, (a trauma level II - III hospital). The head scanning was omitted in 18 patients who had no history of unconsciousness or amnesia and lacked neurological signs of injury. I.v. contrast medium was not administered in 3 cases.

A time protocol of scanning was kept in 1991–95.The scan time was measured from the very first scanogram to the completion of the examination, i.e. the last scanogram. Patient handling time was recorded during 1995.

The diagnostic result of the CT examination was compared to the final diagnosis, which was decided on, by the clinical authors, from the sum of information gathered by all available medical records (i.e. all radiology reports, clinical records from all primary and secondary hospitals and autopsy). The patients final injuries were then classified and judged by the Injury Severity Score (ISS).

Time: The time for the trauma CT-examination was calculated for the 93 patients that completed the full examination, i.e. exclusive of the patients who did not examine the head. The mean examination time was 20.0 minutes (range 12–32 minutes) for the whole group. The median ISS was calculated to 8. In the group with ISS < 8 the mean examination time was 20.7 minutes (range 12–27 minutes) and in the group with ISS ³ 8 it amounted to 19.8 minutes (range 13–32 minutes). In the group with the ten most severely injured patients (ISS ³ 24) the mean examination time was 19.5 minutes (range 15–26 minutes).

During 1995 the time interval from patient arrival to the X-ray department until the first scan was noted for 31 patients and amounted to a mean of 9.7 minutes (range 4–22 minutes).

Head: A total of 28 intracranial injuries were recorded; one epidural and 4 subdural hematomas together with 5 subarachnoidal and 17 supratentorial parenchymal hemorrhages. Three patients had diffuse brain edema, one patient died of a brain stem injury. All injuries except the brain stem injury were diagnosed by the trauma CT exam.

All extracranial injuries (i.e. mainly skull- and facial fractures; n = 28) were primarily diagnosed by CT. A maxillary tumor with bone destruction was initially misdiagnosed as an orbital fracture.

Thorax: A total of 89 thoracic injuries were registered with the following distribution: 8 mediastinal injuries, 17 pneumothorax, 28 hemothorax together with 34 lung contusions and two sternal fractures. The record does not include registration of isolated rib fractures, since they are not considered clinically significant by themselves.

The injuries were all recognized in the CT examination, although two injuries were not accurately or completely defined. In one patient, where the autopsy revealed a rupture of both the heart muscle and a coronary artery, the CT-scan showed an injured heart, but was not able to correctly define the injury. This patient also had a rupture of the thoracic aorta, which was diagnosed retrospectively. One pulmonary laceration was only recognized as a total pneumothorax.

Abdomen and pelvis: Totally 27 injuries were detected, 7 hepatic, 3 splenic, 6 miscellaneous injuries including one diaphragmatic rupture and 11 pelvic fractures/injuries including one urinary bladder rupture. The mere recognition of blood/free fluid in the abdomen or pelvis was not separately accounted for since the possible source of bleeding was found in all our cases.

Two intestinal injuries together with one hepatic injury (shown by autopsy in a patient who died by thoracic aortic rupture) went undetected by the original examination. In addition to the 27 injuries, the dynamic information of hypovolemia in three patients were recognized by CT signs before clinical signs of shock were evident.

Spinal column: Totally 40 vertebral fractures were diagnosed, of which 8 (4 cervical and 4 thoracic) were not appreciated in the trauma CT examination. In several patients the fractures were recognized by indirect signs and the full extent of injury was established either by additional scanning or by plain films.

Axial skeleton: The frontal scanograms make it possible to diagnose fractures in the shoulder, hip and proximal femur. Fractures in other body parts not covered by the scanning have not been recorded as "missed". Twentytwo fractures or dislocations were diagnosed, of which one dislocation in the humeroscapular joint was missed in the CT examination.

Varia: One injury of the axillary artery and nerve was diagnosed by arteriography and clinical examination.

Time: Since the introduction of CT in clinical practice, the value of CT in blunt trauma imaging has been compared to clinical and other radiological methods in numerous studies [ 5, 6, 7, 8, 9] . The literature has focused on comparing the method’s value in imaging region by region in the body, and little attention has been directed to the need of imaging multiple areas by the same process in the multitraumatized patient. Little is published about time effectiveness in imaging in trauma.

With our examination technique with CT scanning, we have tried to comply with the Golden Hour concept and the principles of ATLS. With an examination time of a mean of 20 minutes, ranging from 12–32 minutes, we have shown the method to be highly time efficient. There was no difference in examination time relating to the severity of injuries measured by the ISS. The patient handling time from arrival to the X-ray department until the first scan was measured for a subgroup of patients to a mean of 9.7 minutes, which is acceptable. The longer handling or examination times occurred when something interrupted the normal procedure, like patient vomiting, loss of i.v. line or anesthesiological problems. The short examination time facilitates the decision whether the patient is hemodynamically stable enough to be transported from the emergency room to the X-ray department. It is also reasonable to assume that our method allows "less stable" patients to be more thoroughly examined radiologically than if it may take an undefined amount of time to image the patient.

The simultaneous presence of an anesthesiologist and a trauma surgeon during the examination assures adequate patient surveillance and makes immediate therapeutic intervention possible.

A standardized way of radiological examination in trauma care eliminates unnecessary time delay caused by discussion about what type of imaging that is necessary for the handling of the patient in the initial stage. It also gives training possibilities for the radiological personnel, especially essential for departments that only experience a small number of trauma cases.

Our scanning order is structured to save time and minimize patient table movements. Thus the body scanograms are used only for diagnostic purposes.

With the scanner limitation in speed (4–5 reconstructed scans per minute) and in X-ray tube heat loading, it has been necessary to scan the trunk with discontinuous slices if the examination is to be completed in a reasonable time.

Head: Our method with consecutive 10 mm thick slices from vertex to C 2, without scanogram and gantry angulation, is developed in order to save time. Our results showed also high diagnostic accuracy. Rescanning due to clinical deterioration (with angulation to the OM-line, 5 mm scans in the posterior fossa, and 10 mm scans in the rest of the skull) revealed progress of one intracerebral hematoma, development of one extracerebral hematoma and recognition of subarachnoidal blood in one case. This is expected since the original scanning merely is the status of one moment in a dynamic event. The brain stem injury was clinically diagnosed and could probably have been visualized only with MR.

Thorax: Scanning the thorax in conjunction with a trauma examination already encompassing the head or abdomen costs little extra time and yields more information than plain chest X-ray [ 10] . Studies have shown that CT is superior to plain chest X-ray in trauma in detecting all types of injuries with the exception of rib fractures. A thorax scanning protocol , with discontinuous slices (mean of 7 scans, 8 mm thick), comparable to ours, has also been shown to be superior to plain chest X-ray [ 11, 12, 13] .

Angiography is still the gold standard in detecting Traumatic Aortic Injury (TAI), since CT with conventional technique cannot rule out this injury [ 14] . Recent studies suggest that helical chest CT may offer the possibility to exclude TAI [ 15] .

CT may rule out the presence of an mediastinal hematoma and thus help to discriminate candidates for aortography [ 16, 17] . Our experience includes several cases with a widened upper mediastinum which was possible to explain by abundant fat, and thereby avoiding a thoracic aortography. As a widened mediastinum indicates the need of an angiogram, a possible flaw is that our protocol offers only a frontal scanogram of the thorax, which is of less quality than a plain X-ray. However, we believe that this is compensated for by the axial CT scans. In the Swedish trauma scene, TAI is an infrequent problem with less than 7 deaths per million inhabitants every year. Considering a 75 % initial death rate at the scene of accident, an approximate number of 2 patients per million inhabitants with TAI should come alive to the hospital. In our case with TAI that was not recognized initially, it is, however, possible to identify a mediastinal hematoma and a contrast leak in retrospect, why this case is considered as an interpreter error (Fig. 3).

In our study a pulmonary laceration was recognized by a total pneumothorax, and no more information leading to this diagnosis would have been accomplished by plain X-ray.

Abdomen and pelvis: There was no problem to discover significant hepatic or splenic injuries, even with discontinuous scanning technique, due to the of size of the organs and the frequent finding of free fluid/blood in the abdomen. It cannot be ruled out that minor injuries might have been overlooked, but if so, there was no record of clinical implication. Our undiscovered case of hepatic injury was in a patient who did not receive i.v. contrast medium, because of rapid deterioration and death, due to aortic rupture.

In several institutions CT has replaced diagnostic peritoneal lavage (DPL) as the primary method of evaluation of blunt trauma to the abdomen and pelvis [ 18] . CT examination has limitations especially in the detection of injuries to the bowel and pancreas [ 19] . These limitations also apply to our material with two undiscovered intestinal injuries found during surgical exploration because of diaphragmatic rupture and splenic injury in one case, and in urinary bladder rupture in the other case. These types of injury may be detected by DPL which preferably is made on clinical suspicion after the CT examination. We have diagnosed one evident case of diaphragmatic rupture, which is difficult to diagnose with any method and cannot be ruled out with our protocol. Helical scanning is in this regard reported to be advantageous [ 20] .

We had no renal injuries. A major advantage with the use of i.v. contrast medium is the detection of major renal artery damage manifested by lack of contrast enhancement. Another advantage is the early detection of hypovolemia characterized by vasoconstriction of the aorta and inferior vena cava, intense contrast enhancement of the kidneys and in the walls of the small intestine [ 21] . In three patients these signs were noted before the clinical signs of shock were noticeable. We have also noted extremely high contrast density in the thoracic aorta as another indicator of a large blood volume loss (Fig. 4 a, b).

Extensive pelvic fractures may produce intraperitoneal blood leakage. DPL would be positive in such a case, but CT can localize the source of bleeding, and thus avoid unnecessary explorative laparatomy.

In patients with ongoing bleeding the source may be localized by CT [ 22, 23] and haemostasis may be accomplished by radiological interventional techniques, if available.

Spinal column: The head scanning protocol was modified after we discovered a C 1 fracture with our standard scanning technique which in that patient accidentally covered the first vertebrae. We now routinely include scanning of C1-C2 in the head protocol, although we do not consider these 10 mm slices sufficient to exclude fractures. The standard multitrauma protocol is not geared to examine the spinal column, but still gives vital information, partly in the lateral scanogram of the cervical column, partly in the axial scanning of the trunk every 30 mm, where the paravertebral hematoma from a spinal fracture propagates to give an indirect sign of the injury (Fig. 5 a, b).

Since the protocol does not include scanning of the spinal column, the undiscovered fractures are not to be regarded as cases of misdiagnosis. In contrary, the vertebral fractures we diagnosed are to be considered as an additional benefit. (Ideally we supplement the trauma CT with a plain X-ray of the cervical column routinely within 24 hours, and the thoracic and lumbar spine are X-rayed, if considered necessary).

Axial skeleton: Our method only gives an orientation of fracture incidence and must be supplemented with ordinary plain films according to clinical information. One case of hip dislocation was diagnosed and corrected directly on the CT-table (Fig. 6), showing the clinical utilization of immediate information.

With a third generation CT-scanners limitations we have structured the emergency examination as a survey because of the need to keep the examination time short. We used a discontinuous scanning of the trunk. Our results show that the diagnostic sensitivity with this scanning technique is comparable to protocols with continuous scanning and has only the traditionally known limitations of CT-scanning, i. e. injuries of thoracic aorta, diaphragm, pancreas and bowel cannot be completely ruled out. Use of the new helical scanning CT technique alleviates the compromise between speed and accuracy and will probably accomplish the maximum accuracy that the CT technique may render today. The addition of a helical CT cervical spine examination to a helical examination of the trunk may be sufficient to diagnose spinal column injuries with high sensitivity.

The main importance of this study lies in the recognition of the fact that it is possible to accomplish a fast and structured examination of the acutely traumatized patient which covers the body’s vital organ systems, with high accuracy in injury detection.

We conclude that a standardized CT examination is to be recommended as the method of choice as the initial radiological evaluation of multitraumatized patients. The use of helical CT scanning may further improve the method significantly.

REFERENCES

1. Stein SC, Ross SE (1990) The value of computed tomographic scans in patients with low-risk head injuries. Neurosurgery 26: 638-640.
1. Stein SC, Ross SE (1992) Mild head injury: A plea for routine early CT scanning. J Trauma 33: 11-13.
1. Harad FT, Kerstein MD (1992) Inadequacy of bedside clinical indicators in identifying significant intracranial injury in trauma patients. J Trauma 32: 359-363.
1. Shackford SR, Wald SL, Ross SE et al (1992) The clinical utility of computed tomographic scanning and neurologic examination in the management of patients with minor head injuries. J Trauma 33: 385-394.
1. Brooks AP, Olson LK, Shackford SR (1989) Computed tomography in the diagnosis of traumatic rupture of the thoracic aorta. Clin Radiol 40: 133-138.
1. Wing VW, Federle MP, Morris J (1985) The clinical impact of CT for blunt abdominal trauma. AJR 145: 1191-1194.
1. Federle MP (1983) Computed tomography of blunt abdominal trauma. Radiol Clin North Am 21: 461-475.
1. Rizzo MJ, Federle MP, Griffiths BG (1989) Bowel and mesenteric injury following blunt abdominal trauma: Evaluation with CT. Radiology 173: 143-148.
1. Pollack HM, Wein AJ (1989) Imaging of renal trauma. Radiology 172: 297-308.
1. Karaaslan T, Meuli R, Androux R et al. (1995) Traumatic chest lesions in patients with severe head trauma: A comparative study with computed tomography and conventional chest roentgenograms. J Trauma 39: 1081-1086.

 

1. Poole GV, Morgan DB, Cranston PE (1993) Computed tomography in the management of blunt thoracic trauma. J Trauma 35: 296-302.
1. Rhea JT, Novelline RA, Lawrason J, Sacknoff R, Oser A (1989) The frequency and significance of thoracic injuries detected on abdominal CT scans of multiple trauma patients. J Trauma 29: 502-505.
1. Bahren W, Biehl C (1990) Comparison between the value of thorax tomography and thorax radiography in severely traumatized patients. Rontgenpraxis 43: 164-169.
1. Miller FB, Richardson JD, Thomas HA, Cryer HM, Willings SJ (1989) Role of CT in diagnosis of major arterial injury after blunt thoracic trauma. Surgery 106: 596-603.
1. Gavant ML, Flick P, Menke P, Gold RE (1996) CT aortography of thoracic aortic rupture. AJR 166: 955-961.
1. Raptopoulos V, Sheiman RG, Phillips DA, Davidoff A, Silva WE (1992) Traumatic aortic tear: Screening with chest CT. Radiology 182: 667-673.
1. Mirvis SE, Shanmuganathan K, Miller BH, White CS, Turney SZ (1996) Traumatic aortic injury: Diagnosis with contrast-enhanced thoracic CT - Five-year experience at a major trauma center. Radiology 200: 493-501.
1. Wolfman NT, Bechtold RE, Scharling ES, Meredith JW (1992) Blunt upper abdominal trauma: Evaluation by CT. AJR 158: 493-501.
1. Meredith JW, Ditesheim JA, Stonehouse S, Wolfman N (1992) Computed tomography and diagnostic peritoneal lavage: Complementary roles in blunt trauma. Am J Surg 58: 44-48.
1. Israel RS, McDaniel PA, Primack SL, Salmon CJ, Fountain RL et al. (1996) Diagnosis of diaphragmatic trauma with helical CT in a swine model. AJR 167: 637-641.
1. Sivit CJ, Taylor GA, Bulas DI, Kushner DC, Potter BM et al. (1992) Posttraumatic shock in children: CT findings associated with hemodynamic instability. Radiology 182: 723-726.
1. Jeffrey RB Jr, Cardoza JD, Olcott EW (1991) Detection of active intraabdominal arterial hemorrhage: Value of dynamic contrast-enhanced CT. AJR 156: 725-729.
1. Shanmuganathan K, Mirvis SE, Sover ER (1993) Value of contrast-enhanced CT in detecting active hemorrhage in patients with blunt abdominal or pelvic trauma. AJR 161: 65-69.

Figure 1 and 2 should preferably match the entire page width.

The other figures may have the width that is most favorably in the layout.

The figures with two photographs: The photos may be placed either side by side (preferably) or above and below each other.

Fig. 1. CT examination protocol, see text for legend.

Fig. 2 a - d. CT examination overview; see text for legend

Fig. 3. Thoracic Aortic Injury. Mediastinal hematoma shown as periaortic increased density, in areas 1 and 2 with HU= +9-17, compared to normal mediastinal fat in area 3 with HU= -120. Arrowhead shows injury site with aortic contrast leakage.

Fig. 4 a,b. a Increased contrast enhancement (HU= +373) in the thoracic aorta in a patient with uncompensated large blood volume loss. b corresponding image of a normal subject shows less enhancement (HU=+120). Equal parameters for contrast administration was used.

Fig. 5 a,b. Paravertebral hematoma (a) as an indirect sign of a vertebral fracture, evident in another vertebral body, visualized in a supplementary scan (b).

Fig. 6 a,b. Scanograms of right hip dislocation (a) before and (b) after reduction made during the completion of the emergency examination.