- Email: [email protected]
Predictive indicators for bowel injury in pediatric patients who present with a positive seat belt sign after motor vehicle collision
Journal of Pediatric Surgery (2010) 45, 921–924
www.elsevier.com/locate/jpedsurg
Predictive indicators for bowel injury in pediatric patients who present with a positive seat belt sign after motor vehicle collision Catherine Paris, Mahli Brindamour, Alain Ouimet, Dickens St-Vil ⁎ Division of Pediatric Surgery, Centre Hospitalier Universitaire Sainte-Justine, Montréal, Québec, Canada, H3T 1C5 Received 24 January 2010; accepted 2 February 2010
Key words: Intraabdominal injury; Abdominal wall bruising; Vertebral fracture; Seat belt syndrome
Abstract Purpose: Abdominal wall bruising (AWB) is a frequent finding in children wearing seat belts involved in motor vehicle collision (MVC) and is highly suspicious but not indicative of intestinal injury. The aim of this study was to find objective clinical and radiologic predictors for the need of an abdominal exploration in these children. Materials and Methods: A retrospective chart review of children admitted from 1998 and 2008 with AWB after MVC was conducted. Demographics, vital signs, physical examinations, radiologic investigations, associated injuries, management, and outcome were extracted. Univariate and multivariate statistical analyses were done. Results: Fifty-three children with a median age of 9 years (range, 3-16 years) were included. Forty-four patients (83%) had abdominal pain on arrival, and 25 (47%) had free intraabdominal fluid on ultrasound/scan. Intraabdominal injuries were noted in 29 patients (55%), and the most common were mesenteric or bowel injuries (25%), splenic injuries (13%), and hepatic injuries (8%). Ten patients (19%) needed therapeutic laparotomy, and all were victims from collision involving 2 moving vehicles, had abdominal pain, free intraabdominal fluid, and tachycardia. Five patients (50%) operated on had lumbar fracture compared to only 4 patients (9%) in the nonoperative group. Pulse rate higher than 120 (P = .048), lumbar fracture (P = .008), and free intraabdominal fluid (P ≤ .001) were significant predictors for intestinal perforation. Overall survival was 98% with 1 death because of head trauma. Conclusion: Intraabdominal injuries in children with AWB after MVC are frequent. Associated lumbar fracture, the presence of free intraabdominal fluid, and pulse rate higher than 120 are significant predictors of intestinal injuries. An abdominal exploration should be considered in these patients. © 2010 Elsevier Inc. All rights reserved.
Presented at the 41st Annual Meeting of the Canadian Association of Paediatric Surgeons, Halifax, Nova Scotia, Canada, October 1-3, 2009. ⁎ Corresponding author. Tel.: +1 514 345 4879; fax: +1 514 345 4964. E-mail address: [email protected] (D. St-Vil). 0022-3468/$ – see front matter © 2010 Elsevier Inc. All rights reserved. doi:10.1016/j.jpedsurg.2010.02.023
After the introduction of lap seat belts in cars in the 1960s and its generalized use, the incidence of deaths related to motor vehicle collision (MVC) decreased dramatically [1]. This change in mortality rate is explained in numerous studies by a significant decrease in severe head traumas that remain the major cause of death in MVC [2]. However, with
922
C. Paris et al.
the introduction of seat belts, patients involved in MVC sustained different types of injuries from those previously described. In 1962, Garrett and Braunstein [3] described a complex of lesions called the seat belt syndrome, composed of a combination of lumbar spine fractures, abdominal wall bruising (AWB), and intraabdominal injuries. Early diagnosis of intraabdominal lesions of the mesentery or the bowel is difficult to establish and leads to therapeutic delays that increase morbidity and mortality. Abdominal wall bruising is a frequent finding in children wearing seat belts involved in MVC and is highly suggestive, but not indicative, of intestinal injury. This study was done to define clinical and/or radiologic objective findings that can predict the need for therapeutic laparotomy in children with AWB.
1. Materials and methods A retrospective chart review of all patients with AWB admitted at Centre Hospitalier Universitaire Ste-Justine, a tertiary pediatric trauma center in Montreal, Quebec, Canada, between June 1998 and October 2008 was conducted. All patients with AWB after MVC were included in the study. Patients presenting with AWB not related to seat belt were excluded. Collected data included patient's age, sex and weight, accident mechanism, location of the patient inside the vehicle and type of seat belt used as described in the prehospital medical report, vital signs on arrival, symptoms and physical findings at initial evaluation, investigations done and their results, associated injuries, type and timing of surgical interventions, complications, length of hospital stay, and outcome. The relationship of intestinal perforations with early available findings was evaluated. Univariate and multivariate statistical analyses were done using the Fisher's Exact test and logistic regression model.
2. Results Fifty-three children were admitted during the study period with AWB secondary to MVC. There were 25 girls (47%)
Table 1
Vitals signs of 53 patients with AWB on arrival
Vital sign (median)
Operative group (n = 10)
Nonoperative group (n = 43)
Total
Cardiac pulse rate Blood pressure Respiratory rate Pulse oximetry Glasgow
126 119/66 22 98 14
104 112/64 24 98 14
108 118/64 24 98 14
Fig. 1
Distribution of injuries in patients with AWB.
and 28 boys (53%), with median age and weight of 9 years (range, 3-16 years) and 36.8 kg (range, 15-91 kg). Ten patients required laparotomy (operative group), and 43 were treated without abdominal surgery (nonoperative group). From the prehospital reports, 9 children were front seat passengers, 42 were rear seat passengers, whereas the location inside the vehicle was unspecified for 2 children. The seat belt was apposed in 2 points (abdominal seat belt) in 10 patients, in 3 points (abdominal and thoracic) in 13 patients, and undefined in 30 patients. The accident mechanism involved 2 moving vehicles in 39 cases and collisions with a fixed object in 14 cases. Forty-one patients were transferred to our center after stabilization in a peripheral hospital. The most frequent symptom at initial evaluation was abdominal pain in 44 children (83%). Four children sustained severe head trauma and were intubated. Five patients (9%) without abdominal pain did not sustain any intraabdominal injuries. Physical examination revealed abdominal guarding in 4 children, only one eventually requiring surgery for perforated bowel. No difference in arterial blood pressure, respiratory rate, blood oximetry, and Glasgow score were seen between patients needing laparotomy and the nonoperative group (Table 1). However, when cardiac pulse was categorized in 2 groups: more than 120/min and 120/min or less; a cardiac pulse rate higher than 120/min was encountered significantly more often in children with perforated hollow viscus needing surgery than in children without bowel perforation (P = .048; odds ratio [OR], 9.75). Screening for intraabdominal injuries was done with abdominal ultrasonography, computed tomographic (CT) scan, and x-ray in 34 (64%), 25 (47%), and 10 (19%) patients, respectively. Twenty-five children (47%) had free intraabdominal fluid on abdominal ultrasound and/or CT scan. Other radiologic studies more frequently done were cervical spine x-ray in 35 children (66%), chest x-ray in 33 patients (62%), pelvic x-ray in 29 patients (55%), and lumbar spine x-ray in 23 children (43%). Intraabdominal injuries were noted in 29 patients (55%). Intraabdominal organs most frequently injured were the mesentery or bowel in 13 (25%), the spleen in 7 (13%), and the liver in 4 (8%) children. Other associated injuries included 32 musculoskeletal injuries (60%), of which 9
Predictive indicators for bowel injury in pediatric patients Table 2
923
Parameters and injuries in children with and without surgery
Median age (y) Median weight (kg) 2 moving vehicles Abdominal pain Free liquid Lumbar fracture Intra-abdominal lesion Perforated intestines Solid organs injuries (%) Complications (%) Mean length of stay (d)
Laparotomy (10)
Observation (43)
Total (53)
9.8 39.1 10 (100%) 9 (90%) 10 (100%) 5 (50%) 10 (100%) 10 (100%) 4 (40%) 5 (50%) 19.3
9.3 36.4 29 (67%) 35 (81%) 15 (35%) 4 (9%) 10 (19%) 0 12 (29%) 0 4.6
9.4 36.8 39 (74%) 44 (83%) 25 (47%) 9 (17%) 20 (38%) 10 (19%) 16 (30%) 5 (9%) 7.4
were lumbar spine fractures (17%) and 16 head traumas (30%) (Fig. 1). Ten patients (19%) developed hollow viscus perforation and required therapeutic laparotomy with bowel resection and primary anastomosis. Indications for surgery were clinical deterioration with signs of peritonitis in 9 children and hemodynamic instability in 1. Six laparotomies were done within 24 hours of admission, 3 within 48 hours, and 1 on the third day. All were victims from collision involving 2 moving vehicles and had abdominal pain and free intraabdominal fluid on the initial imaging studies (Table 2). Five patients (50%) had associated lumbar fracture compared to only 4 patients (9%) in the nonoperative group. Univariate analysis with Fisher's Exact test showed that pulse rate higher than 120/min (P = .048; OR, 9.75), presence of free intraabdominal fluid (P ≤ .001; no OR), and associated lumbar fracture (P = .008; OR, 4.77) were significant predictors for intestinal perforation. Logistic regression was used to see the effect of all the above predictors together. They remained significant at 0.05 level except free intraabdominal fluid. Because all 10 patients with bowel perforation also had free intraabdominal fluid, the logistic regression model was not stable when adding it as a predictor. Postoperative complications occurred in 5 patients and included wound infections (2), anastomotic leak (1), intraabdominal abscess (1), and hemodynamic instability (1). Only one child operated on after 24 hours of admission had complications. Operated patients had a mean hospital stay of 19.3 days compared to 4.6 days in children who did not require surgery. Overall survival was 98% with 1 death because of associated severe head trauma.
3. Discussion Children are particularly at risk for seat belt syndrome. Because of their specific constitution, lap belts tend not to be apposed correctly on children, leading to a tendency to migrate from their immature iliac crests toward the abdomen
[4]. Classically, injuries reported as being part of the seat belt syndrome are abdominal wall contusions, intraabdominal injuries to both solid and hollow organs, as well as fractures of the lumbar spine, most often from L2 to L4 [5]. The mechanism of this constellation of lesions is rapid deceleration resulting in compression of the lower abdomen and sudden hyperflexion of the upper torso around the seat belt, leading to crushing of the abdominal content against the spine. In a recent study, the incidence of seat belt syndrome has been estimated at one child per 1000 children injured in a MVC [6]. Few studies have recognized the relationship between the presence of AWB and concurrent intraabdominal injuries, with AWB being a strong predictor of more serious internal injuries [7,8]. Wotherspoon et al [9] found that 9 patients of 60 adults with a seat belt sign had intestinal lesions and that no intestinal lesions were seen without it. In the present study, 29 patients (55%) presenting with AWB had an intraabdominal injury and 10 (19%) had an intestinal injury. Therefore, there is a 1 in 5 chance for children with AWB to have intestinal perforation and to require surgical intervention. Although most patients who required laparotomy underwent surgery within the first 24 hours, 40% had delayed laparotomy (between 24 and 72 hours) and the rate of complications appeared to be independent from the timing of surgery; this can be explained by the small numbers of complications and children operated on. The relationship between the presence of lumbar spine fractures and intraabdominal injuries in children wearing seat belt is well described in the literature. It has been reported that between 48% and 62% of patients with lumbar fracture will have simultaneous intraabdominal injuries [10-12]. Beaunoyer et al described a higher rate of laparotomy in children wearing seat belts as opposed to unbelted patients, reaching 68% in those with spinal fractures and AWB [13]. The present study shows similar and even greater rates of intraabdominal injuries among patients with lumbar spine fractures. Bowel injury was encountered in 55% of these children, and the rate of intraabdominal injuries reached
924 78%. Although it is well described that radiologic investigation should be done in children with AWB to search for and exclude spine fracture, 15 patients in our study were considered free of lumbar fracture based only on physical examination, without any lumbar spine imaging. In our institution, lumbar x-rays are now standard of care in children with AWB. Hollow viscus injuries in the setting of blunt traumas are difficult to identify because of the limits of radiologic tools and require a high degree of suspicion from the clinician [14-16]. In light of these results, we strongly suggest that intraabdominal injuries be suspected in the presence of AWB after MVC. Moreover, the presence of associated lumbar spine fracture, pulse rate higher than 120/min, and intraabdominal free fluid on ultrasound or CT scan should raise suspicion to an even higher degree. Lap belts protect children involved in MVC from severe head trauma but can cause AWB as well as intraabdominal and spinal injuries leading to major surgery, higher rates of complications, and prolonged hospital stays. Associated lumbar fracture, the presence of free intraabdominal fluid, and tachycardia are highly predictive of intestinal injuries in children with AWB after MVC. Abdominal exploration should be considered in these patients.
References [1] Sokolove PE, Kuppermann N, Holmes JF. Association between the “seat belt sign” and intra-abdominal injury in children with blunt abdominal trauma. Acad Emerg Med 2005;12:808-13. [2] Galasko CSB, Edwards DH. The use of seat belts by motor car occupants involved in road traffic accidents. Injury 1975;6:320-4.
C. Paris et al. [3] Garrett JW, Braunstein PW. The seatbelt syndrome. J Trauma 1962;2: 220-8. [4] Santschi M, Echavé V, Laflamme S, et al. Seat-belt injuries in children involved in motor vehicle crashes. Can J Surg 2005;5:373-6. [5] Durbin DR, Arbogast KB, Moll EK. Seat-belt syndrome in children: a case report and review of the literature. Pediatr Emerg Care 2001;17: 474-7. [6] Santschi M, Lemoine C, Cyr C. The spectrum of seat belt syndrome among Canadian children: results of a two-year population surveillance study. Pediatr Child Health 2008;13(4):279-83. [7] Sivit CJ, Taylor GA, Newman KD, et al. Safety-belt injuries in children with lap-belt ecchymosis: CT findings in 61 patients. AJR Am J Roentgenol 1991;157:111-4. [8] Lutz N, Nance ML, Kallan MJ, et al. Incidence and clinical significance of abdominal wall bruising in restrained children involved in motor vehicles crashes. J Pediatr Surg 2004;39:972-5. [9] Wotherspoon S, Chu K, Brown AFT. Abdominal injury and the seat-belt sign. Emerg Med 2001;13:61-5. [10] Mulpuri K, Reilly CW, Perdios A, et al. The spectrum of abdominal injuries associated with chance fractures in pediatric patients. Eur J Pediatr Surg 2007;17:322-7. [11] Anderson PA, Rivara FP, Maier RV, et al. The epidemiology of seat belt-associated injuries. J Trauma 1990;31:1133-8. [12] Anderson P, Henley M, Rivara F, et al. Flexion distraction and chance injuries to the thoraco-lumbar spine. J Orthop Trauma 1991;5(2): 153-60. [13] Beaunoyer M, St-Vil D, Lallier M, et al. Abdominal injuries associated with thoraco-lumbar fractures after motor vehicle collision. J Pediatr Surg 2001;36(5):760-2. [14] Sherck J, Shatney C, Sensaki D, et al. The accuracy of computed tomography in the diagnosis of blunt small bowel perforation. Am J Surg 1994;168(6):670-5. [15] Albanese CT, Meza MP, Gardner MJ, et al. Is computed tomography a useful adjunct to the clinical examination for the diagnosis of pediatric gastrointestinal perforation from blunt abdominal trauma in children? J Trauma 1996;40(3):417-21. [16] Peters E, LoSasso B, Foley J, et al. Blunt bowel and mesenteric injuries in children: do nonspecific computed tomography findings reliably identify these injuries? Pediatr Crit Care Med 2006;7(6):551-6.
www.elsevier.com/locate/jpedsurg
Predictive indicators for bowel injury in pediatric patients who present with a positive seat belt sign after motor vehicle collision Catherine Paris, Mahli Brindamour, Alain Ouimet, Dickens St-Vil ⁎ Division of Pediatric Surgery, Centre Hospitalier Universitaire Sainte-Justine, Montréal, Québec, Canada, H3T 1C5 Received 24 January 2010; accepted 2 February 2010
Key words: Intraabdominal injury; Abdominal wall bruising; Vertebral fracture; Seat belt syndrome
Abstract Purpose: Abdominal wall bruising (AWB) is a frequent finding in children wearing seat belts involved in motor vehicle collision (MVC) and is highly suspicious but not indicative of intestinal injury. The aim of this study was to find objective clinical and radiologic predictors for the need of an abdominal exploration in these children. Materials and Methods: A retrospective chart review of children admitted from 1998 and 2008 with AWB after MVC was conducted. Demographics, vital signs, physical examinations, radiologic investigations, associated injuries, management, and outcome were extracted. Univariate and multivariate statistical analyses were done. Results: Fifty-three children with a median age of 9 years (range, 3-16 years) were included. Forty-four patients (83%) had abdominal pain on arrival, and 25 (47%) had free intraabdominal fluid on ultrasound/scan. Intraabdominal injuries were noted in 29 patients (55%), and the most common were mesenteric or bowel injuries (25%), splenic injuries (13%), and hepatic injuries (8%). Ten patients (19%) needed therapeutic laparotomy, and all were victims from collision involving 2 moving vehicles, had abdominal pain, free intraabdominal fluid, and tachycardia. Five patients (50%) operated on had lumbar fracture compared to only 4 patients (9%) in the nonoperative group. Pulse rate higher than 120 (P = .048), lumbar fracture (P = .008), and free intraabdominal fluid (P ≤ .001) were significant predictors for intestinal perforation. Overall survival was 98% with 1 death because of head trauma. Conclusion: Intraabdominal injuries in children with AWB after MVC are frequent. Associated lumbar fracture, the presence of free intraabdominal fluid, and pulse rate higher than 120 are significant predictors of intestinal injuries. An abdominal exploration should be considered in these patients. © 2010 Elsevier Inc. All rights reserved.
Presented at the 41st Annual Meeting of the Canadian Association of Paediatric Surgeons, Halifax, Nova Scotia, Canada, October 1-3, 2009. ⁎ Corresponding author. Tel.: +1 514 345 4879; fax: +1 514 345 4964. E-mail address: [email protected] (D. St-Vil). 0022-3468/$ – see front matter © 2010 Elsevier Inc. All rights reserved. doi:10.1016/j.jpedsurg.2010.02.023
After the introduction of lap seat belts in cars in the 1960s and its generalized use, the incidence of deaths related to motor vehicle collision (MVC) decreased dramatically [1]. This change in mortality rate is explained in numerous studies by a significant decrease in severe head traumas that remain the major cause of death in MVC [2]. However, with
922
C. Paris et al.
the introduction of seat belts, patients involved in MVC sustained different types of injuries from those previously described. In 1962, Garrett and Braunstein [3] described a complex of lesions called the seat belt syndrome, composed of a combination of lumbar spine fractures, abdominal wall bruising (AWB), and intraabdominal injuries. Early diagnosis of intraabdominal lesions of the mesentery or the bowel is difficult to establish and leads to therapeutic delays that increase morbidity and mortality. Abdominal wall bruising is a frequent finding in children wearing seat belts involved in MVC and is highly suggestive, but not indicative, of intestinal injury. This study was done to define clinical and/or radiologic objective findings that can predict the need for therapeutic laparotomy in children with AWB.
1. Materials and methods A retrospective chart review of all patients with AWB admitted at Centre Hospitalier Universitaire Ste-Justine, a tertiary pediatric trauma center in Montreal, Quebec, Canada, between June 1998 and October 2008 was conducted. All patients with AWB after MVC were included in the study. Patients presenting with AWB not related to seat belt were excluded. Collected data included patient's age, sex and weight, accident mechanism, location of the patient inside the vehicle and type of seat belt used as described in the prehospital medical report, vital signs on arrival, symptoms and physical findings at initial evaluation, investigations done and their results, associated injuries, type and timing of surgical interventions, complications, length of hospital stay, and outcome. The relationship of intestinal perforations with early available findings was evaluated. Univariate and multivariate statistical analyses were done using the Fisher's Exact test and logistic regression model.
2. Results Fifty-three children were admitted during the study period with AWB secondary to MVC. There were 25 girls (47%)
Table 1
Vitals signs of 53 patients with AWB on arrival
Vital sign (median)
Operative group (n = 10)
Nonoperative group (n = 43)
Total
Cardiac pulse rate Blood pressure Respiratory rate Pulse oximetry Glasgow
126 119/66 22 98 14
104 112/64 24 98 14
108 118/64 24 98 14
Fig. 1
Distribution of injuries in patients with AWB.
and 28 boys (53%), with median age and weight of 9 years (range, 3-16 years) and 36.8 kg (range, 15-91 kg). Ten patients required laparotomy (operative group), and 43 were treated without abdominal surgery (nonoperative group). From the prehospital reports, 9 children were front seat passengers, 42 were rear seat passengers, whereas the location inside the vehicle was unspecified for 2 children. The seat belt was apposed in 2 points (abdominal seat belt) in 10 patients, in 3 points (abdominal and thoracic) in 13 patients, and undefined in 30 patients. The accident mechanism involved 2 moving vehicles in 39 cases and collisions with a fixed object in 14 cases. Forty-one patients were transferred to our center after stabilization in a peripheral hospital. The most frequent symptom at initial evaluation was abdominal pain in 44 children (83%). Four children sustained severe head trauma and were intubated. Five patients (9%) without abdominal pain did not sustain any intraabdominal injuries. Physical examination revealed abdominal guarding in 4 children, only one eventually requiring surgery for perforated bowel. No difference in arterial blood pressure, respiratory rate, blood oximetry, and Glasgow score were seen between patients needing laparotomy and the nonoperative group (Table 1). However, when cardiac pulse was categorized in 2 groups: more than 120/min and 120/min or less; a cardiac pulse rate higher than 120/min was encountered significantly more often in children with perforated hollow viscus needing surgery than in children without bowel perforation (P = .048; odds ratio [OR], 9.75). Screening for intraabdominal injuries was done with abdominal ultrasonography, computed tomographic (CT) scan, and x-ray in 34 (64%), 25 (47%), and 10 (19%) patients, respectively. Twenty-five children (47%) had free intraabdominal fluid on abdominal ultrasound and/or CT scan. Other radiologic studies more frequently done were cervical spine x-ray in 35 children (66%), chest x-ray in 33 patients (62%), pelvic x-ray in 29 patients (55%), and lumbar spine x-ray in 23 children (43%). Intraabdominal injuries were noted in 29 patients (55%). Intraabdominal organs most frequently injured were the mesentery or bowel in 13 (25%), the spleen in 7 (13%), and the liver in 4 (8%) children. Other associated injuries included 32 musculoskeletal injuries (60%), of which 9
Predictive indicators for bowel injury in pediatric patients Table 2
923
Parameters and injuries in children with and without surgery
Median age (y) Median weight (kg) 2 moving vehicles Abdominal pain Free liquid Lumbar fracture Intra-abdominal lesion Perforated intestines Solid organs injuries (%) Complications (%) Mean length of stay (d)
Laparotomy (10)
Observation (43)
Total (53)
9.8 39.1 10 (100%) 9 (90%) 10 (100%) 5 (50%) 10 (100%) 10 (100%) 4 (40%) 5 (50%) 19.3
9.3 36.4 29 (67%) 35 (81%) 15 (35%) 4 (9%) 10 (19%) 0 12 (29%) 0 4.6
9.4 36.8 39 (74%) 44 (83%) 25 (47%) 9 (17%) 20 (38%) 10 (19%) 16 (30%) 5 (9%) 7.4
were lumbar spine fractures (17%) and 16 head traumas (30%) (Fig. 1). Ten patients (19%) developed hollow viscus perforation and required therapeutic laparotomy with bowel resection and primary anastomosis. Indications for surgery were clinical deterioration with signs of peritonitis in 9 children and hemodynamic instability in 1. Six laparotomies were done within 24 hours of admission, 3 within 48 hours, and 1 on the third day. All were victims from collision involving 2 moving vehicles and had abdominal pain and free intraabdominal fluid on the initial imaging studies (Table 2). Five patients (50%) had associated lumbar fracture compared to only 4 patients (9%) in the nonoperative group. Univariate analysis with Fisher's Exact test showed that pulse rate higher than 120/min (P = .048; OR, 9.75), presence of free intraabdominal fluid (P ≤ .001; no OR), and associated lumbar fracture (P = .008; OR, 4.77) were significant predictors for intestinal perforation. Logistic regression was used to see the effect of all the above predictors together. They remained significant at 0.05 level except free intraabdominal fluid. Because all 10 patients with bowel perforation also had free intraabdominal fluid, the logistic regression model was not stable when adding it as a predictor. Postoperative complications occurred in 5 patients and included wound infections (2), anastomotic leak (1), intraabdominal abscess (1), and hemodynamic instability (1). Only one child operated on after 24 hours of admission had complications. Operated patients had a mean hospital stay of 19.3 days compared to 4.6 days in children who did not require surgery. Overall survival was 98% with 1 death because of associated severe head trauma.
3. Discussion Children are particularly at risk for seat belt syndrome. Because of their specific constitution, lap belts tend not to be apposed correctly on children, leading to a tendency to migrate from their immature iliac crests toward the abdomen
[4]. Classically, injuries reported as being part of the seat belt syndrome are abdominal wall contusions, intraabdominal injuries to both solid and hollow organs, as well as fractures of the lumbar spine, most often from L2 to L4 [5]. The mechanism of this constellation of lesions is rapid deceleration resulting in compression of the lower abdomen and sudden hyperflexion of the upper torso around the seat belt, leading to crushing of the abdominal content against the spine. In a recent study, the incidence of seat belt syndrome has been estimated at one child per 1000 children injured in a MVC [6]. Few studies have recognized the relationship between the presence of AWB and concurrent intraabdominal injuries, with AWB being a strong predictor of more serious internal injuries [7,8]. Wotherspoon et al [9] found that 9 patients of 60 adults with a seat belt sign had intestinal lesions and that no intestinal lesions were seen without it. In the present study, 29 patients (55%) presenting with AWB had an intraabdominal injury and 10 (19%) had an intestinal injury. Therefore, there is a 1 in 5 chance for children with AWB to have intestinal perforation and to require surgical intervention. Although most patients who required laparotomy underwent surgery within the first 24 hours, 40% had delayed laparotomy (between 24 and 72 hours) and the rate of complications appeared to be independent from the timing of surgery; this can be explained by the small numbers of complications and children operated on. The relationship between the presence of lumbar spine fractures and intraabdominal injuries in children wearing seat belt is well described in the literature. It has been reported that between 48% and 62% of patients with lumbar fracture will have simultaneous intraabdominal injuries [10-12]. Beaunoyer et al described a higher rate of laparotomy in children wearing seat belts as opposed to unbelted patients, reaching 68% in those with spinal fractures and AWB [13]. The present study shows similar and even greater rates of intraabdominal injuries among patients with lumbar spine fractures. Bowel injury was encountered in 55% of these children, and the rate of intraabdominal injuries reached
924 78%. Although it is well described that radiologic investigation should be done in children with AWB to search for and exclude spine fracture, 15 patients in our study were considered free of lumbar fracture based only on physical examination, without any lumbar spine imaging. In our institution, lumbar x-rays are now standard of care in children with AWB. Hollow viscus injuries in the setting of blunt traumas are difficult to identify because of the limits of radiologic tools and require a high degree of suspicion from the clinician [14-16]. In light of these results, we strongly suggest that intraabdominal injuries be suspected in the presence of AWB after MVC. Moreover, the presence of associated lumbar spine fracture, pulse rate higher than 120/min, and intraabdominal free fluid on ultrasound or CT scan should raise suspicion to an even higher degree. Lap belts protect children involved in MVC from severe head trauma but can cause AWB as well as intraabdominal and spinal injuries leading to major surgery, higher rates of complications, and prolonged hospital stays. Associated lumbar fracture, the presence of free intraabdominal fluid, and tachycardia are highly predictive of intestinal injuries in children with AWB after MVC. Abdominal exploration should be considered in these patients.
References [1] Sokolove PE, Kuppermann N, Holmes JF. Association between the “seat belt sign” and intra-abdominal injury in children with blunt abdominal trauma. Acad Emerg Med 2005;12:808-13. [2] Galasko CSB, Edwards DH. The use of seat belts by motor car occupants involved in road traffic accidents. Injury 1975;6:320-4.
C. Paris et al. [3] Garrett JW, Braunstein PW. The seatbelt syndrome. J Trauma 1962;2: 220-8. [4] Santschi M, Echavé V, Laflamme S, et al. Seat-belt injuries in children involved in motor vehicle crashes. Can J Surg 2005;5:373-6. [5] Durbin DR, Arbogast KB, Moll EK. Seat-belt syndrome in children: a case report and review of the literature. Pediatr Emerg Care 2001;17: 474-7. [6] Santschi M, Lemoine C, Cyr C. The spectrum of seat belt syndrome among Canadian children: results of a two-year population surveillance study. Pediatr Child Health 2008;13(4):279-83. [7] Sivit CJ, Taylor GA, Newman KD, et al. Safety-belt injuries in children with lap-belt ecchymosis: CT findings in 61 patients. AJR Am J Roentgenol 1991;157:111-4. [8] Lutz N, Nance ML, Kallan MJ, et al. Incidence and clinical significance of abdominal wall bruising in restrained children involved in motor vehicles crashes. J Pediatr Surg 2004;39:972-5. [9] Wotherspoon S, Chu K, Brown AFT. Abdominal injury and the seat-belt sign. Emerg Med 2001;13:61-5. [10] Mulpuri K, Reilly CW, Perdios A, et al. The spectrum of abdominal injuries associated with chance fractures in pediatric patients. Eur J Pediatr Surg 2007;17:322-7. [11] Anderson PA, Rivara FP, Maier RV, et al. The epidemiology of seat belt-associated injuries. J Trauma 1990;31:1133-8. [12] Anderson P, Henley M, Rivara F, et al. Flexion distraction and chance injuries to the thoraco-lumbar spine. J Orthop Trauma 1991;5(2): 153-60. [13] Beaunoyer M, St-Vil D, Lallier M, et al. Abdominal injuries associated with thoraco-lumbar fractures after motor vehicle collision. J Pediatr Surg 2001;36(5):760-2. [14] Sherck J, Shatney C, Sensaki D, et al. The accuracy of computed tomography in the diagnosis of blunt small bowel perforation. Am J Surg 1994;168(6):670-5. [15] Albanese CT, Meza MP, Gardner MJ, et al. Is computed tomography a useful adjunct to the clinical examination for the diagnosis of pediatric gastrointestinal perforation from blunt abdominal trauma in children? J Trauma 1996;40(3):417-21. [16] Peters E, LoSasso B, Foley J, et al. Blunt bowel and mesenteric injuries in children: do nonspecific computed tomography findings reliably identify these injuries? Pediatr Crit Care Med 2006;7(6):551-6.