ClinicalRadiology(1992)46, 304-310

Computed Tomography in Blunt Abdominal Trauma: An Analysis of Clinical Management and Radiological Findings A. R. P A D H A N I , C. J. E. WATSON*, R. Y. CALNE* and A. K. D I X O N

Departments of Radiology and *Surgery, Addenbrooke's Hospital and University of Cambridge, Cambridge We have retrospectively assessed the computed tomography (CT) findings in 92 patients suffering severe blunt abdominal trauma. Surgical findings and clinical follow-up were correlated with the CT findings. In nine patients CT was first used after emergency surgery and provided baseline data which was useful for further management. In two patients CT did not demonstrate small hepatic lacerations seen during previous surgery. No deaths were recorded. In 16 patients surgery followed CT within 24 h; there was good correlation between the CT and operative findings in 10 patients. However, CT failed to detect significant solid organ injury in five patients and was misleading (false positive) in another patient. There were two deaths amongst these 16 patients. Sixty-seven haemodynamically stable patients were initially managed non-operatively. Fifteen of these 67 patients had normal CT examinations; only one had subsequent laparotomy (for reasons unconnected with the trauma) where no injury was detected; there were no deaths. O f the 52 patients with an abnormal CT examination, 43 were successfully managed nonoperatively. There were three deaths, including one where an injury missed at CT contributed to the demise of the patient. After an initial trial of non-operative management, the remaining six patients went to surgery where there was good concordance with the CT findings except for one missed renal injury. Active non-operative management of blunt abdominal trauma is widely accepted in haemodynamically stable patients and this report shows how CT supports this policy of surgical restraint in such cases. However, on review CT missed 13 injuries in nine patients overall; stricter attention to technique and better equipment may lead to improved results in the future. Padhani, A.R., Watson, C.J.E., Calne, R.Y. & Dixon, A.K. (1992). Clinical Radiology 46, 304-310. Computed Tomography in Blunt Abdominal Trauma: An Analysis of Clinical Management and Radiological Findings

Accepted for Publication 11 June 1992

There is a growing trend towards non-operative management of visceral trauma in both adults and children with regard to splenic, renal and latterly hepatic injuries [1-3]. Computed tomography (CT) has shown high accuracy in the assessment of haemodynamically stable patients [4]; it can also demonstrate and quantify the amount of intraperitoneal haemorrhage. Recent studies [5,6] have developed CT-based classifications of the degree of solid organ injury (spleen, liver) and correlated these with the use of CT in directing treatment and predicting prognosis. The role of CT also extends to the detection of subsequent complications and documentation of healing. The use of CT in evaluating suspected intra-abdominal injury has not been well documented in this country, particularly with reference to the correlation of CT with surgical findings. To this end, we have retrospectively reviewed CT examinations done in our institution for blunt abdominal trauma in the light of the subsequent clinical findings and management. MATERIALS AND M E T H O D S We retrospectively reviewed all CT studies obtained over a period of 8 years (January 1982 to December 1989) Correspondence to: Dr A. K. Dixon, Department of Radiology, Addenbrooke's Hospital, Hills Road, Cambridge CB2 2QQ.

in patients who had sustained blunt abdominal trauma exclusively or in association with other injuries. Only the first CT study for any patient was evaluated; follow-up examinations were excluded. We identified 122 such patients at Addenbrooke's Hospital, Cambridge, a large teaching hospital which is recognized for hepatic surgery [3,7]. All CT studies were performed on one of two machines: a Siemens Somatom II or a General Electric 9000 system. The data acquisition time and the slice thickness used were 5 s, 8 mm and 10 s, 10 mm respectively. During the time of the study it was our standard policy in injured patients to obtain enhanced images at 1.5 cm intervals through the upper abdomen and at 2 cm intervals through the rest of the abdomen and pelvis [8]. Ideally such enhancement is given so that a dynamic series of images is obtained through the liver and spleen during parenchymal opacification. However, for a variety of reasons, discussed elsewhere [9], enhanced images were obtained in only 90/122 (74%) patients. In 1990 the hard copy images of the CT examinations were reviewed (AKD) without access to the clinical findings or original report. In those patients with hepatic or splenic injury, the extent of the injury was graded in line with classifications which have recently been described [5,6] (Tables 1 and 2). The presence and quantity o f intraperitoneal haemorrhage was determined by the

COMPUTED TOMOGRAPHYIN BLUNT ABDOMINALTRAUMA Table 1 - CT based grading of blunt hepatic trauma Grade I

II III IV V

Criteria

Superficial laceration(s), subcapsular haematoma < 1 cm thick, periportal haemorrhage (tracking) only Laceration(s) 1 3 cm deep, central/subcapsular haematoma 1-3 cm diam. Laceration(s) > 3 cm deep, central/subcapsular haematoma(s) > 3 cm diam. Massive central/subcapsular haematoma > 10 cm, lobar tissue destruction or devascularization Bilobar destruction (maceration) or devascularization

Adapted from the classification proposed by Mirvis et al. [5].

305

(b) those where C T (review) revealed evidence o f substantial abdominal or retroperitoneal trauma. The quality o f the C T examinations (satisfactory/ unsatisfactory), which has been the subject o f a separate study [9], was also noted at review to see if errors were more c o m m o n in those studies o f unsatisfactory quality and to see whether this aspect had any bearing on management/outcome. O f the 122 C T examinations performed, adequate clinico-radiological correlation was only available for 95 patients. Three patients were excluded f r o m analysis because o f the time delay between the first C T examination and the date of the injury (greater than 2 months). Thus 92 patients form the basis o f the data presented below.

Table 2 - CT based grading of blunt splenic trauma

RESULTS Grade I

II llI IV

Criteria

Capsular avulsion, superficial laceration(s), or subcapsular haematoma < 1 cm Parenchymal laceration(s) 1-3 cm deep, central/subcapsular haematoma(s) < 3 cm Laceration(s) > 3 cm deep, central/subcapsular haematoma(s) >3 cm Fragmentation of three or more sections, devaseularization (nonenhanced) spleen

Adapted from the classification proposed by Mirvis et al. [6].

Table 3 - Grading of intraperitoneal haemorrhage Grade

Criteria

II III

None detected Confined to one anatomical space (Morrison's pouch, perihepatie, perisplenic space, etc) In two or more anatomical spaces Occupying entire pelvis

Modified from Federle and Jeffrey [10]. semiquantitative m e t h o d o f Federle and Jeffrey [10] (Table 3). The temporal relationship o f any peritoneal lavage to the C T examination was noted. A n y disagreements between the review and the initial issued C T report were noted. The medical records o f each patient were then reviewed ( A R P and C JEW). A n assessment o f the contribution o f CT to the m a n a g e m e n t o f the patients and to their eventual o u t c o m e was made. The indications for, and the results of, l a p a r o t o m y were noted. Where possible, surgical and pathological descriptions o f the injuries were c o m p a r e d with the C T findings. The patients were divided into three main clinical groups, adapted from Mirvis e t al. [5,6] and Federle e t al.

[11]: 1 Those who had immediate surgery and subsequently had post-operative C T for further evaluation. 2 Those where surgery followed within 24 h o f the C T examination. 3 Those who were initially treated non-surgically. These patients were further divided into: (a) those where CT (review) indicated no significant abdominal injury, and

Sixty-seven o f the 92 patients were male. The age range was 5-79 years (mean age 32 years). There was an increased usage o f C T with time; in the last 3 years o f the study there were approximately 20 studies per annum. Seventy-one studies were performed on patients referred from the accident and emergency department at A d d e n b r o o k e ' s hospital; 21 patients were referred f r o m other hospitals. The majority (63/92, 68%) o f C T examinations were performed within 24 h of the injury. The mechanisms o f injury included m o t o r vehicle accidents (59 patients), falls (15), vehicle-pedestrian accidents (7), sports related injuries (5), crush injuries (3), horse kick (2) and blast injury (1). M o s t patients (77/92) also sustained extra-abdominal injuries: 57 suffered thoracic trauma; injuries to the head and face (49), limbs (34), spine (16) and pelvis (13) were also c o m m o n . Peritoneal tavage had been performed in nine patients before CT. In these patients the degree o f intraperitoneal fluid was not graded. The principal indications for performing the abdominal C T examination are outlined in Table 4. The findings in the various clinical groups will be discussed in turn. Table 5 summarizes those patients where there were discordant features between either the initial C T report, the review C T report a n d / o r the clinical findings/management.

Table 4 - Indications for CT examinations

Abnormal abdominal physical examination(s) Haematuria from suspected renal cause Severe blunt trauma with multiple injuries Unreliable examination due to altered level of consciousness or spinal injury Unexplained hypotension Post-operative evaluation Abnormal lower chest or abdominal radiographs Falling haemoglobin Positive/equivocal lavage Delayed abdominal symptoms Abnormal ultrasound examination

67 19 15 13 9 9 6 3 3 3 1

Modified from Walsh [4]. G r o u p 1. U r g e n t S u r g e r y F o l l o w e d b y C T

Nine patients had urgent surgery followed by C T in the post-operative period. Indications for urgent surgery

306

CLINICAL RADIOLOGY

Table 5 - A chart of patients in whom there were discordant features between the initial CT report, the review CT findings and/or the clinical state

Patient

Age

Sex

Date

IV Quality Adverse role Contrast (satisfactory/ medium unsatisfactory) (yes/no)

Consequence

A

30

M

1982

No

U

Nil

B

43

F

1988

Yes

s

C

10

F

1988

No-

U

D

19

M

1984

Yes

S

E

53

M

1989

No

U

F

51

F

1983

Yes

U

G

44

F

1988

Yes

S

H

35

F

1986

Yes

S

I

35

M

1988

Yes

S

J

18

F

1985

Yes

U

K

13

M

1988

Yes

S

L

68

F

1989

Yes

S

M

24

M

1989

Yes

S

N

20

M

1988

Yes

U

O

77

M

1983

No

U

P

79

F

1985

No

U

Q

20

M

1984

Yes

S

Missed lesion

Comments

Urgent splenectomy and right hemicolectomy. Liver laceration noted. CT (3 days after splenectomy) failed to detect small hepatic injury. Review CT also considered normal. Missed lesion Nil CT (18 h after laparotomy) missed known small injury anterior border of liver. Correct documentation of other injuries; hepatic lesion not seen on review CT. Misleading Surgery probably Original C T misleading, suggesting probable liver (false unnecessary laceration. This plus clinical findings led to surgery: positive) normal liver, pelvic h a e m a t o m a + + . Review CT report: normal liver; pelvis not examined. Missed lesion Delay to surgery C T underestimated splenic and missed a hepatic tear. Surgery less than 6 h after CT performed for hypotension: shallow tear (7 cm long) of liver plus splenic rupture. On CT review some perisplenic fluid noted; liver seemed normal. Missed lesions ?Delay to Initial CT missed liver tear and haemoperitoneum. surgery Surgery 7 h later showed liver tear plus 4 litres haemoperitoneum. Even on review CT no liver injury or free blood was seen. Missed lesions Nil C T missed liver laceration, diaphragmatic tear and intraperitoneal blood. Surgery 16 h after CT revealed ruptured diaphragm and superficial tear with a little intraperitoneal blood. Review C T also normal considered. Missed lesion Nil CT correctly diagnosed splenic tear. Splenectomy. Also correctly suggested torn hemidiaphragm. But small subcapsular hepatic h a e m a t o m a missed on both initial and review CT. Missed lesions Nil Both initial and review CT correctly diagnosed splenic tear. But missed small hepatic laceration and small bowel tear. Nil Initial CT suggested liver contusion; normal on Misleading review CT. No other corroborative further information. Patient recovered. Missed lesion Nil Initial CT normal. On review CT Grade I haemoperitoneum noted. Patient recovered. No further information. Nil Initial CT missed retroperitoneal blood which was Missed lesion diagnosed on review CT. No effect on management. Nil Initial CT normal; review C T probable Grade I Missed lesion splenic trauma. No effect on management. ?Led to surgery Initial C T equivocal re splenic trauma which Unhelpful being withheld encouraged conservative management. Grade III splenic trauma on review CT. This finding might have led to operative management. Possible missed Nil CT (initial and review) showed Grade II hepatic trauma. Conservative management. Late rupture of lesion post-traumatic hepatic artery aneurysm. Missed lesion Death Initial CT showed extensive (Grade III) haemoperitoneum but missed severe hepatic trauma which contributed to death. At post m o r t e m a laceration of the inferior surface of the liver with a tear of the inferior vena cava was found in this elderly patient with coronary artery disease. Even on review no hepatic trauma was seen. Initial CT suggested subcapsular hepatic injury. Misleading Probably none Review CT also showed haemoperitoneum (Grade II) but no hepatic injury. No effect on management. Patient died on first day due to ventricular cardiac arrhythmias. N o post mortem. Missed lesion Probably none Initial C T indicated splenic trauma with normal left kidney. Surgery confirmed splenic tear but also showed h a e m a t o m a around the left kidney (seen on review CT).

307

COMPUTED TOMOGRAPHY IN BLUNT ABDOMINAL TRAUMA Table 6 - The major CT findings at review in Groups 2 and 3

CT abnormalities of

Group 2 (16. pat&nts)

Grades of injury

Group 3 (67 patients)

Grades of injury

Haemoperitoneum Spleen Kidney Diaphragm Liver Retroperitoneum

11 5 4 4 2 2

Grade I : 3, II: 1, III: 7 Grade I: 1, II: 1, III: 1, IV: 2

15 7 9 0 11 I1

Grade I : 5, II: 6, III: 4 Grade I: 1, II : 4, III: 1, IV: 1

Grade III: 1, IV: 1

were: abnormal abdominal examinations (all nine patients), hypotension (5), positive peritoneal lavage (2) and for exploration of blast wound (1). CT examinations were performed to evaluate the postoperative appearances as a baseline for further management, and to look for subsequent complications. Five patients required a second laparotomy (removal of liver packs in four; drainage of the duct of a transected

Grade 1I:3, 111:7, IV: 1

pancreas into the stomach in one). There was good concordance of CT findings with surgical appearances in seven patients, in addition, CT documented retroperitoneal injury in one patient where this was not recorded in the surgical notes. CT provided an anatomical outline of the injuries sustained and aided in encouraging noninterventional management in the four patients who did not proceed to a second laparotomy. However, in two patients (Table 5, patients A and B) CT was deemed unhelpful in so far that small hepatic lacerations, identified at the earlier laparotomy, were not observed. No deaths were recorded in this group.

Group 2. Urgent CT Followed by Early Surgery

Fig. 1 - This M-year-old female had a fall. She also had injuries to the chest (haemopneumothorax, rib fractures). This study performed 16 days following the accident shows a large (Grade 3) subcapsular h a e m a t o m a with a laceration on the posterior aspect of the spleen. These were confirmed at surgery.

Fig. 2 - Patient E. This 53-year-old male had a road traffic accident. He also sustained injuries to the head (contusion), neck (paraplegia) and chest (lung contusion, p n e u m o t h o r a x and mediastinal blood). The C T study performed 5 h after the injury concentrated primarily on the head, neck and chest. A few cuts of the upper abdomen were also performed without intravenous contrast medium. No blood or liver abnormality was seen. Surgery 7 h later revealed 4 1 of blood and a small liver laceration.

Sixteen patients underwent early laparotomy (within 24 h of the CT). The range of abdominal injuries demonstrated on review of the CT images in these patients is outlined in Table 6 and Fig. 1. These patients proceeded to laparotomy for a variety of one or more of the following indications: abnormal physical abdominal examination (16 patients), abnormal CT study (11), unexplained hypotension (4) and for a ruptured diaphragm (4), Surgical and pathological observations in these patients revealed that: In 10 of these 16 patients there was good surgical correlation with the CT findings. CT demonstrated retroperitoneal trauma in four patients where this was not obvious at surgery. In one patient (Table 5, patient C) the initial report was misleading, suggesting a liver tear and this, with abnormal clinical findings, prompted surgery which showed a normal liver but noted a retroperitoneal pelvic haematoma. The latter problem had not been detected by CT as the pelvis had not been fully examined. On retrospective CT review, however, the liver was considered normal; artifacts on this study of poor quality were considered to have led to the original false positive report. In five patients the CT examination was considered unhelpful as it failed to detect significant solid organ injury which was subsequently detected at surgery. Details are given in Table 5 (patients D, E (Fig. 2), F, G, H). Only in one patient (D) was there any substantial disagreement between initial and review CT findings. Two patients died in this group, one of associated head injuries and another at a spinal rehabilitation unit (reason unknown).

Group 3. Initial Non-Surgical Management Sixty-seven patients were initially treated conservatively. Fifteen had a normal abdominal CT examination at

308

CLINICAL RADIOLOGY

Fig. 3 - Patient O. This 77-year-old male had a road traffic accident. He also sustained injuries to the chest and limbs. The CT study performed in 1983 was performed without intravenous contrast medium 5 h after the injury. A rib fracture and considerable blood around the liver and spleen are noted. No source of the haemorrhage is evident and no blood is seen around the inferior vena cava (IVC). At post mortem a small laceration of the inferior surface of the liver with extension into the IVC was seen.

review; 52 patients had evidence of abdominal inj ury. The abdominal injuries sustained are outlined in Table 6.

Group 3a. Normal Abdominal CT Examinations Of the 15 patients who had normal abdominal CT examinations at review, conservative management was initiated as all patients were haemodynamically stable. A normal CT examination contributed to the decision not to operate in 10 patients. In four patients CT confirmed the clinical impression that no significant abdominal injury was present. These normal CT findings were considered confirmed by an uneventful subsequent clinical course (14 patients) or by findings at surgery (one patient). This latter patient underwent laparotomy 180 h after trauma and 96 h after CT for pyrexia and abdominal pain. At surgery there was no evidence of injury but adhesions, jejunal diverticuli and a scarred duodenum were noted. The initial and review reports were in complete agreement in 14 patients. However, in one patient (Table 5, patient I) the original CT report had suggested an hepatic contusion; on review the CT appearances were considered normal. No other corroborative studies were performed, the clinical management was not altered and the patient made an uneventful recovery. No deaths were recorded in this group.

Group 3b. Abnormal Abdominal CT Examination Fifty-two patients had abnormal findings on review of the CT images. Most (46 patients) were managed conservatively throughout their hospital stay; three deaths occurred in this group. Six patients had late surgery ( > 24 h after CT) with no deaths.

Group 3b.1. Non-Surgical Management (46 Patients) In these patients, who were all haemodynamically stable, the review CT findings although abnormal, were

sufficiently favourable to allow conservative management. This group included eight patients with liver injury (Grade II: 3, Grade III: 5), five patients withsplenic injury (Grade I: 1, Grade II: 2, Grade III: 1, Grade IV: 1), six patients with known renal injury and eight patients with pancreatic and/or retroperitoneal injury. These patients were treated non-surgically despite there being haemoperitoneum in 10 patients (Grade I: 2, Grade II: 5, Grade III: 3). Forty-three were thereby successfully managed. In .38 patients the initial CT report, the review CT report and the clinical course were in complete agreement. The issued report was considered to have underestimated the degree of intra-abdominal injury in four patients when compared with the revised CT findings. A small amount of intraperitoneal haemorrhage went undetected in one patient (Table 5, patient J); retroperitoneal haemorrhage had not been recorded in another (K), while in a third patient (L), a Grade I splenic injury was not noted originally. However, with a fourth patient (M) there was a more serious disagreement with the issued report. Although a splenic abnormality was noted at the outset, it was not fully characterized. On review a Grade III splenic injury was present. Had this been recognized it might have led to operative, rather than conservative management. However, in none of these four patients was the clinical outcome adversely affected. In one further patient (N), a favourable CT result contributed to the successful conservative management of Grade II hepatic trauma. This patient developed a late rupture (5 months) of a post-traumatic hepatic artery aneurysm, implying that an arterial injury may have been missed on the initial dynamic enhanced CT examination. Three deaths were recorded in this group: l Patient O (Fig. 3). Died on the tenth day of hepatic injuries and renal failure. CT, which showed extensive intraperitoneal haemorrhage, had failed to reveal a significant hepatic injury which contributed to the demise of the patient. At post mortem, a laceration of the inferior surface of the liver with a tear of the inferior vena cava, intraperitoneal haemorrhage, pulmonary oedema and coronary thrombosis were noted. Even on retrospective CT review no hepatic injury could be identified; however no intravenous contrast medium had been administered. 2 Patient P. Died of disseminated intravascular coagulopathy and ventricular arrhythmias during the first 24 h. The original report suggested a subcapsular haematoma which, in retrospect, probably represented free fluid. However, neither the management of the patient, nor the clinical outcome, was considered to have been adversely affected. A post mortem was not carried out. 3 A third patient died of cerebral trauma, adult respiratory distress syndrome and adrenal infarction. Post mortem confirmed the initial CT findings of absent hepatic injury; in this case the review CT report wrongly suggested a Grade III hepatic injury.

Group 3b.2. Late Surgical Management Six patients with abnormal CT examinations proceeded to late surgery ( > 24 h after CT study). The range of abdominal injuries on retrospective CT review included: hepatic injury in three patients (Grade III: 2 patients, Grade IV: 1), splenic injury (two patients,

COMPUTED T O M O G R A P H Y IN B L U N T ABDOMINAL TRAUMA

both Grade II); renal injury (three patients) and haemoperitoneum (five patients, Grade I: 3, Grade II: 1, Grade III: 1). All six patients were initially haemodynamically stable. Clinical deterioration precipitated laparotomy. The principal CT findings were a large subcapsular haematoma of the liver (2), expanding haematoma of the right kidney (1), splenic rupture (2) and a liver abscess (1). There was good correlation between the CT reports and the surgical findings in five of the patients. However, in one patient, who was found to have continued bleeding from the spleen and left kidney (Q), the initial CT report, although correctly predicting the splenic injury, had not commented on the abnormal left kidney. Review of the CT images, however, revealed the renal trauma. No deaths were recorded. Quality of the CT Examination

It was only possible to judge the quality of the CT examination in 91 of the patients; 45 were judged to be unsatisfactory (49%) compared with 46 (51%) satisfactory. There were no differences in the distribution of unsatisfactory examinations amongst patients of different age, sex, mechanisms of injury or indications for the CT study. There were fewer unsatisfactory examinations amongst those with no extra-abdominal injuries (3/15) compared with those with substantial extra-abdominal injuries (42/ 76). There were no differences in the number of patients initially treated conservatively or proceeding to laparotomy between the satisfactory and unsatisfactory groups. The mean length of hospital stay was also similar. There were four deaths in the group with unsatisfactory CT studies compared with only one in those with a satisfactory examination.

DISCUSSION Historically, peritoneal lavage and laparotomy have long been the mainstays of surgical management of blunt abdominal. However, the potential for surgical restraint is becoming increasingly realized, both in children and adults, for hepatic and splenic trauma [1-3], as well as for renal trauma where a conservative approach for unilateral trauma has long been advocated. Such a conservative approach has been considerably influenced by the advent of CT which is recognized as a valuable tool for the assessment of both the early and late effects of trauma. Indeed CT based classification of the extent of trauma and criteria for surgical and non-surgical management have now been established [5,6]. Furthermore, a decrease in the amount of intra-abdominal fluid on serial CT has been suggested as an indicator of a likely successful outcome of conservative treatment [12]. Our results are in broad agreement with previous series as to the usefulness of CT in complementing the allimportant clinical assessment. These are probably more typical of U K District General Hospitals than the data presented from major specialized trauma units in the USA. Some areas where the CT findings were inaccurate are worthy of further discussion. It is reassuring that the initial report as issued in the emergency situation did not vary all that much from the review report. Only in three cases (C, D and Q) might this

309

have affected management. This gives us some confidence that the present system, where the study is performed by the consultant or senior registrar on duty (who seek advice where appropriate), is satisfactory. Of course there were several other minor disagreements between the initial report and the review report but in most instances there is no hard evidence as to which report was correct. It is disappointing that so many lesions were missed by CT. Some were of a serious nature. The most serious was the patient where an hepatic lesion extending to the inferior vena cava was missed on both the initial and review reports (Patient O; Fig. 3). This was judged to be an unsatisfactory CT with no intravenous enhancement in an elderly patient (77 years) performed quite early on in our experience (1983). This was the only patient in any group where a CT error contributed to the patient's demise. There are several other patients where there was major variance between the CT and surgical findings. In patient E (Fig. 2) extensive bleeding must have occurred in the interval (7 h) before surgery. Two other serious hepatic tears were missed (D and F), along with several other less important hepatic lesions (A, B, G, H and ?N). Interestingly there were few errors as regards splenic trauma. In two patients the initial CT report underestimated the extent of the splenic damage (D and M) but only in the former did this affect management (possibly delay before surgery). It is possible that the radiologist went all out for excellent images of the spleen. Certainly, on the equipment which was available to us at the time of this study, only a few images could be obtained at peak parenchymal opacification. The spleen may have been preferentially demonstrated. Obviously it is important to achieve the best possible CT study in every patient. However this may not always be possible in these distressed patients in the emergency situation. The examination is often performed 'out of hours', maybe with a less experienced team; the clinician may not wish the patient to spend too long in the CT suite; the patient may not be cooperative. Nevertheless the attention to technique is of great importance and this has been addressed elsewhere [9]. Correct intravenous contrast enhancement technique is a key factor. Although there were more deaths in those with unsatisfactory CT examinations, this difference did not reach statistical significance. In any case it is very likely that those most likely to have a poor outcome were amongst the most distressed at the time of the CT study. Thus there is little evidence to link poor quality CT examinations with poor outcome. Only in one patient (patient O) might a better CT study have 'saved' a life. Our data correlate well with previous studies [11,12], underlying the value of CT in the management of injured patients with abdominal trauma. Its role in the postoperative phase serves as a baseline for further management and to look for complications. It helped in the decision to perform early laparotomy in 11 of 16 patients. It also assisted in allowing 67 patients to be initially managed non-operatively, including those with haemoperitoneum and severe hepatic and splenic injuries. Active non-operative management of blunt abdominal trauma is becoming widely accepted and this report supports our evolving policy of surgical restraint in such cases. However, it is important to remember that in a number of patients CT was misleading, unhelpful or missed injuries (see Table 5). Overall 17 injuries were missed in 13 patients on the initial report; even on review 13 injuries in

310

CLINICAL RADIOLOGY

nine patients went undetected. Better equipment (we now have a more modern CT system) and stricter attention to technique (a printed protocol for these patients) should improve results obtained during the 1990s. Acknowledgements. We are grateful to all the clinical, radiographic and radiological colleagues who helped with these patients. The Addenbrooke's Hospital Cancer Scanner Appeal funded the initial body computed tomographic system. Karin Frans prepared the manuscript.

REFERENCES 1 Meyer AA, Crass RA, Lin RC, Jeffrey RB, Federle MP, Trunkey DD. Selective non-operative management of blunt liver injury using computed tomography. Archives of Surgery 1985;120:550-554. 2 Kakkasseril JS, Stewart D, Cox JA, Gelfand M. Changing treatment of paediatric splenic trauma. Archives of Surgery 1982;1117:758762. 3 Watson CJE, Padhani AR, Dixon AK, Calne RY. Surgical restraint in the management of liver trauma. British Journal of Surgery 1991;78:1071 1075.

4 Walsh JW. Blunt abdominal trauma. In: Husband JES, ed. C.T. review. London, Churchill Livingstone, 1989:165-177. 5 Mirvis SE, Whitley NO, Vainwright JR, Gens DR. Blunt hepatic trauma in adults: CT-cased classification and correlation with prognosis and treatment. Radiology 1989; 171:27 32. 6 Mirvis SE, Whitley NO, Gens DR. Blunt splenic trauma in adults: CT-based classification and correlation with prognosis and treatment. Radiology 1989; 171:33-39. 7 Calne RY, Wells FC, Forty J. Twenty-six cases of liver trauma. British Journal of Surgery 1982;69:365 368. 8 Dixon AK. Body CT: a handbook. Edinburgh, Churchill Livingstone, 1983. 9 Padhani AR, Watson CJE, Clements L, Calne RY, Dixon AK. Computed tomography in abdominal trauma: an audit of usage ~nd image quality. British Journal of Radiology 1992;65:397-402. 10 Federle MP, Jeffrey RB. Haemoperitoneum studied by computed tomography. Radiology 1983;148:187 192. 11 Federle MP, Goldberg HI, Kaiser JA, Moss AA, Jeffrey RB, Mall JC. Evaluation of abdominal trauma by computed tomography. Radiology 1981;138:637 644. 12 Foley WD, Cates JD, Kellman GM, Langdon T, Aprahamian C, Lawson TL et al. Treatment of blunt hepatic injuries: role of CT. Radiology 1987;164:635 638.

Computed tomography in blunt abdominal trauma: an analysis of clinical management and radiological findings.

We have retrospectively assessed the computed tomography (CT) findings in 92 patients suffering severe blunt abdominal trauma. Surgical findings and c...
2MB Sizes 0 Downloads 0 Views