- Email: [email protected]
A review of computed tomography in the diagnosis of intestinal and mesenteric injury in pediatric blunt abdominal trauma
A Review of Computed Tomography in the Diagnosis of Intestinal Mesenteric Injury in Pediatric Blunt Abdominal Trauma By John S. Graham and Andrew Calgary, Alberta 0 Objective: To determine the sensitivity, specificity, and positive and negative predictive values of the computed tomography (CT) scan in the diagnosis of clinically significant intestinal and mesenteric injury in pediatric bfunt abdominal trauma. Patients: The records of 145 children who presented to a tertiary care pediatric hospital between 1987 and 1994 were reviewed retrospectively. All had experienced single or multiple injuries and underwent CT as part of the trauma assessment. Methods: The patients were divided into two cohorts, based on the results of the initial CT scan: either positive (n = 20) or negative (n = 152) for evidence of intestinal or mesenteric injury. The two cohorts were similar with respect to age, trauma score, and timing of CT scan. The outcome of surgical (n = 23) and conservative management (n = 122) was compared with the initial CT scan results. (Some of the laparotomies were for solid-organ injury only.) Results: The sensitivity of the CT scan in the diagnosis of clinically significant intestinal and mesenteric injury is 0.93. The specificity and positive and negative predictive values are 0.95, 0.65, and 0.99, respectively. Conclusion: The CT scan is an excellent test to screen for clinically significant intestinal and mesenteric injury in pediatric patients with blunt abdominal trauma. Because of the lower positive value, other clinical and diagnostic imaging information may hefp to improve diagnostic accuracy. Most importantly, CT rarely misses a significant intestinal or mesenteric injury. Copyright o 1996 by W. B. Saunders Ckmpan y INDEX WORDS: Abdominal trauma, teric injury, computed tomography.
intestinal
injury,
mesen-
P
EDIATRIC BLUNT abdominal trauma most frequently results in injury of the spleen, liver, and kidney, and usually is managed conservatively in hemodynamicaIly stable chiIdren.1-4 Although significant intestinal and mesenteric injury occurs less frequently, the management is surgical.1,2 Computed tomography (CT) is a sensitive test for diagnosing solid-organ injury.1,2 However, controversy surrounds the use of the CT scan to screen for intestinal and mesenteric injury.1,4-11 Often, because of concurrent injuries such as those to the head, CT is the only means to assess for intraabdominal injury. The outcome of management decisions based on CT scan From the Department of Pediattic Surgery, University of Calgary, Calgary, Alberta. Presented at the 27th Annual Meeting of the Canadian Association of Paediatnc Surgeons, Montreal, Quebec, September 2-4, 1995. Address reprint requests to Dr Andrew L. Wong, Depafiment of Pediatric Surgery, AlberTa Children’s Hospital, 1820 Richmond Rd SW Calgary, Alberta, Canada T2T5C7. Copyright o 1996 by WB. Saunders Company
0022-3468/96/3106-0003$03.00l0 754
and
L. Wong
findings was reviewed to determine the suitability of the CT scan to screen for intestinal and mesenteric injuries. MATERIALS
AND METHODS
The records of 214 patients who presented to a tertiary care pediatric hospital between 1987 and 1994 were reviewed retrospectively. One hundred forty-five of them met the inclusion criteria: blunt abdominal trauma, CT scan during initial evaluation, and all necessary clinical documentation. Patient demographics, clinical status, results of investigations (including CT scans), timing between accident and CT scan, mechanism of accident, and the outcome of management (either surgical or conservative) were documented for each patient. All patients had CT scanning from the diaphragm to the symphysis pubis. All were given intravenous and enteral contrast. One patient was hemodynamically unstable during the CT scan and was successfully resuscitated with fluid. The results of each CT scan were reviewed with reference to the specific signs of intestinal and mesenteric injury and also to detect other significant findings such as solid-organ injury.1,6J2 The patients were divided into two cohorts depending on the results of their CT scan. The scans were considered either positive or negative, based on evidence of intestinal or mesenteric injury.2.3j-7 In our study, the findings denoting positive CT scans included pneumoperitoneum and/or unexplained free fluid, bowel wall thickening and/or enhancement, and retroperitoneal swelling and/or gas. Negative CT scans had none of these findings. Evidence of solid-organ injury alone was considered a negative finding. The surgical reports of patients who underwent laparotomy were reviewed, and findings were compared with the CT results. The outcome of patients managed conservatively also was compared with the CT results. Uncomplicated recovery, during a period of observation, was used as clinical evidence of an absence of significant intestinal or mesenteric injury. RESULTS
One hundred forty-five patients were included in this study, 20 of whom had CT scans that were positive for intestinal or mesenteric injury. One hundred twenty-five patients had a negative CT result. The two cohorts were similar with respect to age, trauma score, and the intervals between accident, presentation, and CT scan (Table 1). Most commonly, the abdominal examination findings in the positive-CT group were equivocal because of abdominal wall contusion; in the negative CT group, the examination results usuahy were normal. At least one third of the patients in each group were believed to have equivocal or unreliable abdominal examination findings. Journa/ofPed/atr~
Surgery,
Vol 31, No 6 (June),
1996: pp 754-756
MESENTERIC
Table
INJURY:
1. Patient
755
CT DIAGNOSIS
Characteristics
According
to CT Result
Table
3. Summary
of CT Results
and Patient
CT Result Characteristic
Mean age (yr) Mean trauma score Median time from accident presentatron (h) Median
time from
accident
CT Result
Posltlve
Negatw
Outcome
7.36 (SD, 3.97) 8.95 (SD, 2.69)
8.58 (SD, 3.59) 10.30 (SD, 2.10)
Laparotomyfor intestrnal or mesenteric injury Negative laparotomy or solid-organ injury alone
1.5
2.0
1.0
2.0
3.0
4.5
until
presentatron
CT scan (h) Median time from
Uncomplicated conservative management Intestinal or mesenteric injury managed
until
conservatively
Among the 20 patients with a positive CT result, 14 underwent laparotomy. Ten of these had significant intestinal or mesenteric injury, and four had negative laparotomy results (Table 2). Three patients with a positive CT result (unexplained free fluid) had clinically significant partial bowel obstructions, documented on contrast x-rays, that were secondary to intestinal wall or mesenteric hematomas (Tables 2 and 3). Although these obstructions resolved with conservative therapy, they were recorded as significant injuries. Finally, there were three patients with a positive CT result (small amounts of unexplained free fluid) who had observation only, and they recovered without complication (Tables 2 and 3). One hundred twenty-five patients had a negative CT result. Of these, 116 were managed conservatively, without complication, and because of this were believed not to have experienced any significant intestinal or mesenteric injury (Table 3). Eight patients underwent laparotomy for the management of solid-organ injuries (liver, spleen, and kidney), and none had concurrent significant intestinal or mesenteric injury (Table 3). One patient underwent a delayed laparotomy (at 24 hours) after abdominal examination and CT had yielded negative findings. During the laparotomy, three intussusceptions were found and were successfully reduced. Although the 2. Positive
CT Findings
and Final Diagnosis
Flndmg
Pneumoperitoneum and free fluid +- bowel thrckening and enhancement Unexplained
Retroperitoneal swelling
free fluid
gas and/or
Positwe
Negatwe
10 4
1 8
3
116
3
0
until CT
scan (h)
Table
Outcome
Dlagnosls
Hollow
organ
”
perforation 5
Negative laparotomy findings Intestinal perforation Mesenteric injury requiring
4 1
laparotomy Intestinal or mesenteric hematoma managed conservatively No intraabdomrnal rnjury
2
Duodenal
perforation
Duodenal
hematoma
3 3 1 1
bowel was viable and no mesenteric injury was noted, this case was recorded as a false-negative (Table 3). Based on the data summarized above, the sensitivity, specificity, and positive and negative predictive values were calculated13 (Table 3) (I = intestinal, M = mesenteric). Sensitivity
= (+) I or M injury with (+) CT scan/ All(+)IorMinjury =13of 14 = 0.93
Specificity
= (-) I or M injury iwth (-) CT scan/ All(-)IorMinjury = 124 of 131 = 0.95
PPV
= Positive predictive value = (+) I or M injury with (+) CT scan/ All (+) CT scans = 13of20 = 0.65
NPV
= Negative predictive value = i~i Ic; M injury with (-) Ct scan/All scans = 124 of 125 = 0.99 DISCUSSION
Blunt abdominal trauma is a common injury in children. Solid-organ injury occurs frequently, and the CT scan is a sensitive test for diagnosing this.1-4 Although intestinal and mesenteric injuries occur less frequently, prompt diagnosis is important because the management is surgical and a delay results in an increased risk of morbidity and possibly mortality. Literature to support and also to dispute the use of CT scan in this clinical setting exists, and no consensus has emerged.1,4-11 In our experience, the CT has been very helpful in managing children who have sustained blunt abdominal trauma. Because of the high sensitivity and specificity (0.93 and 0.95, respectively), CT is an excellent screening test for intestinal and mesenteric injuries. A very encouraging finding is the negative
756
predictive value of 0.99. For patients whose physical examination results are unreliable, such as those who have head injury, negative CT findings almost always exclude intestinal and mesenteric injury. In the present study, the only patient with a false-negative CT result was found to have three intussusceptions during the delayed laparotomy. The bowel was viable, and there were no mesenteric hematomas. No cases of perforation or significant mesenteric injuries were missed. The relatively low positive predictive value (0.65) is of concern. One third of the patients with CT findings suggestive of intestinal or mesenteric injury did not have these injuries at the time of surgery or during the period of observation. In this series, two of the seven patients with false-positive CT results had pneumothorax, pnezlmomediastinum, or positive pressure ventilation, all of which can cause pneumoperitoneum.14-l6
GRAHAM
AND
WONG
Presumably, this occurs as air under pressure in the pleura or mediastinum dissects into the abdomen via the retroperitoneum. Three other patients with falsepositive results had small amounts of unexplained free fluid as their only positive CT finding, and this likely resulted from a small solid-organ laceration or an insignificant mesenteric injury. Finally, two patients with false-positive results had only serosal tears at the time of laparotomy. In summary, CT of the abdomen appears to be a sensitive screening test to assess for intestinal and mesenteric injuries in pediatric patients with blunt abdominal trauma. The moderately high rate of false-positive CT scan results must be remembered, and results must be correlated with the patient’s clinical condition. The negative predictive value is very high, indicating that a negative test result virtually excludes intestinal and mesenteric injury.
REFERENCES 9. Fabian TC, Mangiante EC, White TJ, et al: A prospective 1. Taylor GA, Fallat ME, Potter BM, et al: The role of computed tomography in blunt abdominal trauma in children. J study of 91 patients undergoing both computed tomography and Trauma 28:1660-1664,1988 peritoneal lavage following blunt abdominal trauma. J Trauma 2. Kane NM, Cronan JJ, Dorfman GS, et al: Pediatric abdomi26:602-607, 1986 nal trauma: Evaluation by computed tomography. Pediatrics 82:1110. Ceraldi CM, Waxman K: Computerized tomography as an 151988 indicator of isolated mesenteric injury. Am Surg 56:806-810,199O 3. Kaufman RA, Towbin R, Babcock DS, et al: Upper abdomi11. Cobb LM, Vinocur CD, Wagner CW, et al: Intestinal nal trauma in children: Imaging evaluation. AJR 142:449-460, 1984 perforation due to blunt trauma in children in an era of increased 4. Matsubara TK, Fong HMT, Burns CM: Computed tomogranonoperative treatment. J Trauma 26:461-463, 1986 phy of abdomen in management of blunt abdominal trauma. J 12. Nghiem HV, Jeffry J: Mindelzun RE: Blunt trauma to the Trauma 30:410-414,1991 bowel and mesentery. AJR 16053-58, 1993 5. Bulas DI, Taylor GA, Eichlberger MR: The value of CT in 13. Knapp RG, Miller MC: Describing the performance of a detecting bowel perforation in children after blunt abdominal diagnostic test, in Knapp RG, Miller MC (eds): Clinical Epidemiology and Biostastics, chap 3. Baltimore, MD, Williams & Wilkins, trauma. AJR 153:561-564,1989 6. Donohue JH, Federle MP, Griffiths BG, et al: Computed 1992, pp 31-45 tomography in the diagnosis of blunt intestinal and mesenteric 14. Roh JJ, Thompson JS, Harned RK, et al: Value of pneumoinjury. J Trauma 27:11-17,1987 peritoneum in the diagnosis of visceral perforation. Am J Surg 7. Mirris SE, Gens DR, Shanmuganathan K: Rupture of the 146:830-833,1983 bowel after blunt abdominal trauma: Diagnosis with CT. AJR 15. Kane NM, Francis IR, Burney RE, et al: Taumatic pneumo159:1217-1221,1992 peritoneum: Implications of computed tomography diagnosis. 8. Marx JA, Moore EE, Jorden RC, et al: Limitations of Invest Radio1 26:574-578, 1991 computed tomography in the evaluation of acute abdominal 16. Winek TG, Mosley HS, Grout G, et al: Pneumoperitoneum trauma: A prospective comparison with diagnostic peritoneal and its association with ruptured abdominal viscus. Arch Surg lavage. J Trauma 25:933-937,1985 123:709-712,1988
intestinal
injury,
mesen-
P
EDIATRIC BLUNT abdominal trauma most frequently results in injury of the spleen, liver, and kidney, and usually is managed conservatively in hemodynamicaIly stable chiIdren.1-4 Although significant intestinal and mesenteric injury occurs less frequently, the management is surgical.1,2 Computed tomography (CT) is a sensitive test for diagnosing solid-organ injury.1,2 However, controversy surrounds the use of the CT scan to screen for intestinal and mesenteric injury.1,4-11 Often, because of concurrent injuries such as those to the head, CT is the only means to assess for intraabdominal injury. The outcome of management decisions based on CT scan From the Department of Pediattic Surgery, University of Calgary, Calgary, Alberta. Presented at the 27th Annual Meeting of the Canadian Association of Paediatnc Surgeons, Montreal, Quebec, September 2-4, 1995. Address reprint requests to Dr Andrew L. Wong, Depafiment of Pediatric Surgery, AlberTa Children’s Hospital, 1820 Richmond Rd SW Calgary, Alberta, Canada T2T5C7. Copyright o 1996 by WB. Saunders Company
0022-3468/96/3106-0003$03.00l0 754
and
L. Wong
findings was reviewed to determine the suitability of the CT scan to screen for intestinal and mesenteric injuries. MATERIALS
AND METHODS
The records of 214 patients who presented to a tertiary care pediatric hospital between 1987 and 1994 were reviewed retrospectively. One hundred forty-five of them met the inclusion criteria: blunt abdominal trauma, CT scan during initial evaluation, and all necessary clinical documentation. Patient demographics, clinical status, results of investigations (including CT scans), timing between accident and CT scan, mechanism of accident, and the outcome of management (either surgical or conservative) were documented for each patient. All patients had CT scanning from the diaphragm to the symphysis pubis. All were given intravenous and enteral contrast. One patient was hemodynamically unstable during the CT scan and was successfully resuscitated with fluid. The results of each CT scan were reviewed with reference to the specific signs of intestinal and mesenteric injury and also to detect other significant findings such as solid-organ injury.1,6J2 The patients were divided into two cohorts depending on the results of their CT scan. The scans were considered either positive or negative, based on evidence of intestinal or mesenteric injury.2.3j-7 In our study, the findings denoting positive CT scans included pneumoperitoneum and/or unexplained free fluid, bowel wall thickening and/or enhancement, and retroperitoneal swelling and/or gas. Negative CT scans had none of these findings. Evidence of solid-organ injury alone was considered a negative finding. The surgical reports of patients who underwent laparotomy were reviewed, and findings were compared with the CT results. The outcome of patients managed conservatively also was compared with the CT results. Uncomplicated recovery, during a period of observation, was used as clinical evidence of an absence of significant intestinal or mesenteric injury. RESULTS
One hundred forty-five patients were included in this study, 20 of whom had CT scans that were positive for intestinal or mesenteric injury. One hundred twenty-five patients had a negative CT result. The two cohorts were similar with respect to age, trauma score, and the intervals between accident, presentation, and CT scan (Table 1). Most commonly, the abdominal examination findings in the positive-CT group were equivocal because of abdominal wall contusion; in the negative CT group, the examination results usuahy were normal. At least one third of the patients in each group were believed to have equivocal or unreliable abdominal examination findings. Journa/ofPed/atr~
Surgery,
Vol 31, No 6 (June),
1996: pp 754-756
MESENTERIC
Table
INJURY:
1. Patient
755
CT DIAGNOSIS
Characteristics
According
to CT Result
Table
3. Summary
of CT Results
and Patient
CT Result Characteristic
Mean age (yr) Mean trauma score Median time from accident presentatron (h) Median
time from
accident
CT Result
Posltlve
Negatw
Outcome
7.36 (SD, 3.97) 8.95 (SD, 2.69)
8.58 (SD, 3.59) 10.30 (SD, 2.10)
Laparotomyfor intestrnal or mesenteric injury Negative laparotomy or solid-organ injury alone
1.5
2.0
1.0
2.0
3.0
4.5
until
presentatron
CT scan (h) Median time from
Uncomplicated conservative management Intestinal or mesenteric injury managed
until
conservatively
Among the 20 patients with a positive CT result, 14 underwent laparotomy. Ten of these had significant intestinal or mesenteric injury, and four had negative laparotomy results (Table 2). Three patients with a positive CT result (unexplained free fluid) had clinically significant partial bowel obstructions, documented on contrast x-rays, that were secondary to intestinal wall or mesenteric hematomas (Tables 2 and 3). Although these obstructions resolved with conservative therapy, they were recorded as significant injuries. Finally, there were three patients with a positive CT result (small amounts of unexplained free fluid) who had observation only, and they recovered without complication (Tables 2 and 3). One hundred twenty-five patients had a negative CT result. Of these, 116 were managed conservatively, without complication, and because of this were believed not to have experienced any significant intestinal or mesenteric injury (Table 3). Eight patients underwent laparotomy for the management of solid-organ injuries (liver, spleen, and kidney), and none had concurrent significant intestinal or mesenteric injury (Table 3). One patient underwent a delayed laparotomy (at 24 hours) after abdominal examination and CT had yielded negative findings. During the laparotomy, three intussusceptions were found and were successfully reduced. Although the 2. Positive
CT Findings
and Final Diagnosis
Flndmg
Pneumoperitoneum and free fluid +- bowel thrckening and enhancement Unexplained
Retroperitoneal swelling
free fluid
gas and/or
Positwe
Negatwe
10 4
1 8
3
116
3
0
until CT
scan (h)
Table
Outcome
Dlagnosls
Hollow
organ
”
perforation 5
Negative laparotomy findings Intestinal perforation Mesenteric injury requiring
4 1
laparotomy Intestinal or mesenteric hematoma managed conservatively No intraabdomrnal rnjury
2
Duodenal
perforation
Duodenal
hematoma
3 3 1 1
bowel was viable and no mesenteric injury was noted, this case was recorded as a false-negative (Table 3). Based on the data summarized above, the sensitivity, specificity, and positive and negative predictive values were calculated13 (Table 3) (I = intestinal, M = mesenteric). Sensitivity
= (+) I or M injury with (+) CT scan/ All(+)IorMinjury =13of 14 = 0.93
Specificity
= (-) I or M injury iwth (-) CT scan/ All(-)IorMinjury = 124 of 131 = 0.95
PPV
= Positive predictive value = (+) I or M injury with (+) CT scan/ All (+) CT scans = 13of20 = 0.65
NPV
= Negative predictive value = i~i Ic; M injury with (-) Ct scan/All scans = 124 of 125 = 0.99 DISCUSSION
Blunt abdominal trauma is a common injury in children. Solid-organ injury occurs frequently, and the CT scan is a sensitive test for diagnosing this.1-4 Although intestinal and mesenteric injuries occur less frequently, prompt diagnosis is important because the management is surgical and a delay results in an increased risk of morbidity and possibly mortality. Literature to support and also to dispute the use of CT scan in this clinical setting exists, and no consensus has emerged.1,4-11 In our experience, the CT has been very helpful in managing children who have sustained blunt abdominal trauma. Because of the high sensitivity and specificity (0.93 and 0.95, respectively), CT is an excellent screening test for intestinal and mesenteric injuries. A very encouraging finding is the negative
756
predictive value of 0.99. For patients whose physical examination results are unreliable, such as those who have head injury, negative CT findings almost always exclude intestinal and mesenteric injury. In the present study, the only patient with a false-negative CT result was found to have three intussusceptions during the delayed laparotomy. The bowel was viable, and there were no mesenteric hematomas. No cases of perforation or significant mesenteric injuries were missed. The relatively low positive predictive value (0.65) is of concern. One third of the patients with CT findings suggestive of intestinal or mesenteric injury did not have these injuries at the time of surgery or during the period of observation. In this series, two of the seven patients with false-positive CT results had pneumothorax, pnezlmomediastinum, or positive pressure ventilation, all of which can cause pneumoperitoneum.14-l6
GRAHAM
AND
WONG
Presumably, this occurs as air under pressure in the pleura or mediastinum dissects into the abdomen via the retroperitoneum. Three other patients with falsepositive results had small amounts of unexplained free fluid as their only positive CT finding, and this likely resulted from a small solid-organ laceration or an insignificant mesenteric injury. Finally, two patients with false-positive results had only serosal tears at the time of laparotomy. In summary, CT of the abdomen appears to be a sensitive screening test to assess for intestinal and mesenteric injuries in pediatric patients with blunt abdominal trauma. The moderately high rate of false-positive CT scan results must be remembered, and results must be correlated with the patient’s clinical condition. The negative predictive value is very high, indicating that a negative test result virtually excludes intestinal and mesenteric injury.
REFERENCES 9. Fabian TC, Mangiante EC, White TJ, et al: A prospective 1. Taylor GA, Fallat ME, Potter BM, et al: The role of computed tomography in blunt abdominal trauma in children. J study of 91 patients undergoing both computed tomography and Trauma 28:1660-1664,1988 peritoneal lavage following blunt abdominal trauma. J Trauma 2. Kane NM, Cronan JJ, Dorfman GS, et al: Pediatric abdomi26:602-607, 1986 nal trauma: Evaluation by computed tomography. Pediatrics 82:1110. Ceraldi CM, Waxman K: Computerized tomography as an 151988 indicator of isolated mesenteric injury. Am Surg 56:806-810,199O 3. Kaufman RA, Towbin R, Babcock DS, et al: Upper abdomi11. Cobb LM, Vinocur CD, Wagner CW, et al: Intestinal nal trauma in children: Imaging evaluation. AJR 142:449-460, 1984 perforation due to blunt trauma in children in an era of increased 4. Matsubara TK, Fong HMT, Burns CM: Computed tomogranonoperative treatment. J Trauma 26:461-463, 1986 phy of abdomen in management of blunt abdominal trauma. J 12. Nghiem HV, Jeffry J: Mindelzun RE: Blunt trauma to the Trauma 30:410-414,1991 bowel and mesentery. AJR 16053-58, 1993 5. Bulas DI, Taylor GA, Eichlberger MR: The value of CT in 13. Knapp RG, Miller MC: Describing the performance of a detecting bowel perforation in children after blunt abdominal diagnostic test, in Knapp RG, Miller MC (eds): Clinical Epidemiology and Biostastics, chap 3. Baltimore, MD, Williams & Wilkins, trauma. AJR 153:561-564,1989 6. Donohue JH, Federle MP, Griffiths BG, et al: Computed 1992, pp 31-45 tomography in the diagnosis of blunt intestinal and mesenteric 14. Roh JJ, Thompson JS, Harned RK, et al: Value of pneumoinjury. J Trauma 27:11-17,1987 peritoneum in the diagnosis of visceral perforation. Am J Surg 7. Mirris SE, Gens DR, Shanmuganathan K: Rupture of the 146:830-833,1983 bowel after blunt abdominal trauma: Diagnosis with CT. AJR 15. Kane NM, Francis IR, Burney RE, et al: Taumatic pneumo159:1217-1221,1992 peritoneum: Implications of computed tomography diagnosis. 8. Marx JA, Moore EE, Jorden RC, et al: Limitations of Invest Radio1 26:574-578, 1991 computed tomography in the evaluation of acute abdominal 16. Winek TG, Mosley HS, Grout G, et al: Pneumoperitoneum trauma: A prospective comparison with diagnostic peritoneal and its association with ruptured abdominal viscus. Arch Surg lavage. J Trauma 25:933-937,1985 123:709-712,1988