Original article 341
Isolated Bowel Injury in Blunt Abdominal Trauma in Childhood G. Schimpl, B. Schmidt, H. Sauer UniversilätskJinik für Kinderchirurgie Graz, HeinrichsIr. 31, Graz, Auslria
From a total of 734 children with a blunl abdominallrauma admilled lo lhe hospital in lhe pasl15 years, 21 patients (3 %) sustained an isolated injury of the bowel (8 duodenal, 9 jejunal and 4 colon ruptures). All patients were laparotomized ",cilhoul a postoperative mortalily. Accompanying abdominal injuries were seen only in duodenal ruplures (pancreatitis and one choledochal and pancreatic ruptures). In 85 % lhe blunt \iolence was caused by bicycle accidents due lo lhe handle bar, in one case by a car accident and in 3 children by falls. Accurale diagnosis was only possible regarding lhe hislory, lhe mechanism of lhe accident and an exact repeated clinical examinalion. Despite further investigations of blood chemistry laboralory findings, ultrasound and x-ray, no further confirmalion of lhe diagnosis could be achieved. Complications, occurring in 14 % of our patients, were not related lo the lrauma itself, bul caused by a delayed diagnosis and therapy. Key words Blunt abdominal trauma - Isolated traumatic rupture of the bowel Resume Sur un total de 734 enfants atteints de traumatisme abdominal ferme (SBT) dans les 15 dernieres annees, 21 patients (3 %) presentaient une plaie isolee de l'intestin (8 ruptures duodenales, 9 jejunales et 4 coliques). Tous les patients ont ete operes sans mortalite post-operatoire. Les lesions abdominales d'accompagnement ont ete seulement observees dans les ruptUl'es duodenales (pancreatite, rupture du choledoque ou du pancreas). Dans 85 % des cas le traumatisme etait cause par le guidon de bicyclette, dans 1 cas par un accident de voiture, et par une chute chez 3 enfants. Le diagnostic precis a seulement ete etabli en tenant compte de I'histoire, du mecanisme d'accident et d'un examen clinique exact. Le diagnostic precis a ete etabli essentiellement par les antecedents, le
mecanisme de I'accidenl, el un examen clinique repele. Les recherches de laboratoire, echographie et aux rayons X, n'ont pas pu amener de detail parliculier dans le diagnostic d'une rupture inleslinale isolee. Les complications (14 %) n'ont pas resulle de la plaie inleslinale en elle meme, mais onl ele consecuti\'es a un diagnostic el un lraitement differe. Mols-eIes ----Traumalisme abdominal ferme - Huplure intestinale traumatique isolee Zusammenfassung Insgesamt wurden in den letzten 15 Jahren 734 Kinder nach einem stumpfen Bauchlrauma (SBT) behandelt. 21 von ihnen (3 %) hallen eine lraumatische Darmruptur (8 Duodenalrupturen, 9 Jejunalrupturen und 4 Kolonrupturen). Intraabdominelle Begleitverletzungen lagen bei allen Duodenalrupturen vor (Pankreatitis, Pankreasund Choledochusabriß), jedoch bei keinem Kind mit einer Jejunum- oder Kolonruptur. Alle Patienten wurden laparolomiert, ohne posloperative Mortalität. In 85 % führten Fahrradstürze mit in 60 % sichtbaren abdominellen Prellmarken und in einem Patienten ein Autounfall und 3mal Stürze aus mittlerer Höhe zu Darmverletzungen. Zur Diagnose haben sich ausschließlich die Erhebung einer genauen Anamnese, unter Berücksichtigung des Unfallmechanismuses und die wiederholte klinische Untersuchung als sicheres diagnostisches Vorgehen bewährt. Mit laborchemischen Untersuchungen und durch den Einsatz bildgebender Verfahren konnte keine wesentliche Erweiterung in der Diagnostik isolierter Darmrupturen erzielt werden. Komplikationen (14 %) resultieren nicht aus der Verletzung des Darmes selbst, sondern waren Folge einer verzögerten Diagnose und Therapie. SchI üsselwörter Stumpfes Bauchtrauma - Isolierte traumatische Darmruptur
Received Januar.\· -1, 1991 Eur J Pediatr Surg 2 (1992) 3-11-3-1-1 © HiPPOkrates \-erlag Slullgali .\ \asson Editeur Paris
Downloaded by: NYU. Copyrighted material.
Eur J Pediatr Surg 2 (1992)
G. Schimpl et al
Blunt abdominal trauma still presents diagnostic difficulties in cases of isolated injury to the bowel in spite of diagnostic imaging and chemi try laboratory findings. These injuries, unlike ruptures of the liver or spleen, do not lead to great blood loss. In addition, a further factor leading to postponement of therapy and diagnosis i the recent trend not to operate children with blunt abdominal trauma \\'ho e clinical condition is stable. Signs of peritonitis or ileus should be considered as delayed symptoms of traumatic intestinal rupture. Patients In the past 15 years, /31 children were treated for blunt abdominal trauma althe l'niycrsilätsklinik für Kinderchirurgie in Graz. T\I'enly-onc patients (3 %).20 boys and one girl aged 5-1 J years old. suffered an isolated injury of the bo\\el (8 duodenal. 9 jejunal and I colon ruplures). Causes of injury' \lere bicycle accidenLs (I/ cases), car accidenls (I case). and falls (3 cases). Age distribution \las 5-9 years for lhe bicycle accidcnts. 10.5 years for the car accidenl and / years for olher injuries. Laboralor.\ findings, ultrasound, x-ray and in one case a laparoscopy \I ere performed. Includcd in the laboratory findings \\'ere: the complete blood count, serum electrolytc, li\'er funclion tests, serum and urine amyla e \'alues, blood gas analyse and chemical and microscopic urine analyses. All palients were analyzed retrospectiyely with regard to case hislory, c1inical examination, laboratOl'y findings and imaging diagnostic .
a) Duodenal injuries (Table 1) Eight boys, ages 6-10 years old, sustained injuries of the duodenum. Cause of injury was in 6 ca es a bicycle accident (75 %), in one case a car accident as a passenger, and in one case a fall from a jungle gym. Accompanying injuries were in one case craniocerebral trauma \\~th jaw fracture and in a second case lower leg fracture. Clini-
cal examinations showed contusions and tenderness in the upper abdomen. The general condition of alt chiJdren was stahle. The abdomen natiye x-ray in upright position and, to exclude perforation, in lateral posi. tion \\~th horizontal radiograph was positive in only 2 cases (25 %). Twice we used ultrasound. The tentatiye diagnosis wa duodenal injUI}'. In addition, injuries of parenchymatous organs, a weil as fTee intraperitoneal nuid, could be excluded. In one ca e, gastro-inteslinal passage witb contrast showed duodenal stenosis in a duodenal mural hematoma. In 6 children (75 %) blood chemistries revealed an unspecific leukocytosis, and in 7 patients (87 %) an eleyated amylase. All 8 children were laparotomized between land ..J8 hours (mean 3.8 hours) after repeated clinical examination because of persistent or worsening symptoms. Six times an oblique semieireular rupture in the pars inferior of the duodenum was found intraoperatiyely. Onee a pronounced duodenal mural hematoma and onee a sublotal rupture of the pars deseendens with simultaneous ayulsion of the ductus eholedoehus and head of the panereas \\'a found. L'ncompliealcd ruplures were re· anaslomosed. The mural hematoma \\'a drained. Solely a eomplex injury was lreated by a eholedoehopanereatieojejunostomy after V-Roux as weil as a gastrie reseetion aeeording to B 11. Compliealions resulled in lwo ca e . In one case the complicalion lIas a pancreas pseudoeysl thaI was trealed operatively in another hospital in spile of spontaneous regression. In anolher case the abdominal \I'all could inilially nol be c10sed bceause of a se\'ere peritoni· tis and panereatitis. A temporary' zipper lIas se\\cd in for repeated la\'alion of the abdominal ea\ity and c10sure followed 12 days later.
b) Jejunal ntptures (Table 2) )line boys, ages 5-9 years, sustained perforation of the jejunum. In eight eases (88 %) the trauma was eaused bya bieycle aeeiden!. Six of these patients (66 %) exhibited the typieal bruise marks on the abdominal waJl eaused by the handle bar. Concomitant injuries were in one ease eranioeerebral trauma and in another case fraeture of the forearm. AJI of the ehildren were stable with regard to cardiopulmonary funclion; one ehild exhibited neurologie symptoms due to eranioeerebral trauma. Palpatory findings were abdominal tenderness as weil as localized tenderness on pressure in the upper and midabdominal region. The native abdomen x·ray showed subphrenie air in one patient and in two eases signs of bowel obstruetion (66 %). In 4 ca es ultrasound findings revealed minimal free intraabdominal fluid. All of the chemistry laboratory findings were normal. Because of repeated clinieal examinations and persisting symptoms, all 9 patients were
Duodenal rup ures in 8 patlents (+ pathologie flndlllgs, - normal flndlngs).
M. A. male 8a S. R. male 10a
N A. male 7a
R P male8a
K. 5 male 6a
M M male 7a
G 5 male9a
methods used x·ray· -
fall (clim blllg) automobile
K. P male 8a
x-ray: + ultrasound: +
con uSlon mark contuslon mark con uSlon mark contuslon mark
x-ray x-ray' x·ray ~ ultrasound: x-ray ul rasound +
x-ray ultrasound x·ray ultrasound -
Laboratory findings amylase + leukocytosls + amylase + leukocytosls +
amylase + leu kocytosls + amylase amylase .,. leukocytosls amylase + leukocytosls + contrast medium + amylase +
rupture pars Inferior rupture, tearoff of he choledochus and pancrea IC head rupture pars Inferior rupture pars Inferior rupture pars IIlferior hematoma
rupture pars IIlfenor rup ure pars Infenor
Y·en-Roux, B 11
SHT Jaw fracture
secondary closure of abdominal wall
oversuturlng pancreatic pseudoeys
oversuturlng oversuturlng removal oversutunng
frac ure tlbla
Downloaded by: NYU. Copyrighted material.
Eur J Pediatr Surg 2 (1992)
/so/ated Bowe/ /njury in Blunt Abdominal Trauma in Childhood Table 2
JeJunal ruptures In 9 patlents (+ pathologlc flndlngs. - normal flndlngs).
A T male 5a
R. B. male9a
DA male8a FM. male 7a
contusion mark contuslon mark
A. male 6a
con uSlon mark
H. S. male 9a
G. L. male 7a
R. M. male 8a
K. C. male 8a
x-ray: + ultrasound. + x-ray: + laparoscopy' + x-ray: + x-ray ultrasound + x-ray ul rasound + x-ray ultrasound + x-ray ultrasound x-ray CT +x-ray -
rupture of JeJunum
rupture of JeJunum hematoma rupture of JeJunum rupture of JeJunum
lower arm fracture
rupture of jeJunum
rupture of JeJunum
rup ure of JeJunum
rupture of JeJunum
rupture of JeJunum
laparotomizcd \\'ithin 2-6 hours of admission (mean 3.6 hours). One patient \\ ith craniocerebral trauma undcn\'cnt a preopcrati\'e laparoscopy due to his ;,omnolent state. Intraoperati\'ely, antimesenteric rupturc;, of the jejunum \\'crc found that \\,erc;) mm to 2 cm in diameter and situated J 5-80 cm di tal of thc duodenojejunal f1exure. In 8 cases, ,uture of the site of rupture \\'a sufficienl. In another case_ a portion of thc intestines \\'as resected because of a concomitant mesenteric hematoma and hemorrhagic infarction. _ one of the patients had po toperati\c complications.
c) Colon ruptures (Table 3) Three boys and' one girl of 6- J I years of age ustained colon ruptures. Causes of injury \\'ere in 3 cases bicycle accidents (75 %) and in one case a fall out of a \\'indo\\'. Concomitant injuries "'ere limited to craniocerebral trauma in one case. In all cases, no bo\\'el injuries were suspected in the prima!)' c1inical examinations. Ho\\'ever in three ca es (75 %) bruises \\'ere found on the abdominal \\'all. The nati\'e abdomen x-ray as \\'ell as the uJtrasound \\'ere unremarkable. Blood chemistry revealed solely a sJight leukocytosis. 8-20 hours later (rnean 11.6 hours) c1inical and radiological signs of an ileus \\'ere found in 2 children. Two other children showed a clinical deterioration of the abdominal ymptoms with increasing abdominal disten ion. During surgery, in 2 children perforations in the ascending colon were found, in one a stenosing colon mural hernatoma in the right colon flexure and in one patient a lengthwise tear in the colosigmoidal junction. ln three cases the rupture was sutured and in one ca e the portion of the colon with the hematoma was resected and reanastomosed due to localized circuJatory complications. 3 1/2 months later, this child developed an adhesion ileu requiring relaparotomy.
OS male 6a
J M male 11a
fall (willdow) cycle
K. B. female lla
R. R male 10a
Discussion Injury of the bO\\'el in lhe form of blunl abdominal trauma is fairi)' rare and is cited in lhe lilerature as occurring bel\\'een 1 % and 15 %. Con\'ersely, more frequenl forms of bO\\'el injuries are perforating abdominal trauma (89 %) and injuries through impalement (30 %) (5, 13). The mortality rate of isolated bowel injuries is 3 % (4). These types of bowel injuries initially do nol appeal' as dramatic as ruptures of the li"er 01' spleen since there is no hypovolemia and no uniform clinical findings can be made. If signs of peritonitis should occur then these are delayed symptoms (21). The rare occurrence of bowel injury in blunt abdominal trauma is reJated to the anatomy and topography of the bowel as weil as its suspensory structures. Even though the bowel is onJy protected by the abdominal wall in front and rests against the vertebral column in the back, it is very resistant against outside influences, as long as these are not limited 01' localized. In cases of limited 01' localized influences, however, the possibility of injury of a fluid-filled hollow organ is much higher (20). Bowel injuries often occur unexpectedly and often are not diagnosed and subsequenUy treated too late (7). Bowel segments may rupture by the following pathomechanisms: a) direct, localized forces, b) deceleration with occurrence of shearing forces, c) through bursting due to the volume of fluid and d) through abnormal bowel fixations, as occur with postoperative adhesions (24). According to the literature, geograpruc differences are seen according to the mechanism of injury (1, 4,16). The most frequent cause of injury (82 %) in our patients wa a localized abdominal trauma due to the handle
x-ray ultrasound' x-ray ultrasound x-ray -
rupture colon ascendens rupture colon ascendens hematoma of wall of colon rupture of slgmold colon
oversu unng avers
resectlon oversu urlng
Concomltan Compltcatlon Leslons
Downloaded by: NYU. Copyrighted material.
Eur J Pediatr Surg 2 ( J 992)
bar of a bicycle, as in lhe cases of bicycle accidenls, as welJ as car accidenls and fall from medium heighl. Accompanying injuries which can be presenl in up lo 40 % of lhe cases (5) and are usually easier to diagnose, generally make lhe diagnosis of bowel injury more difficull. Se\'en of our eight patienls v.ith duodenal injuries had a pancreatitis (87 %). Howe\'er, in cases of ruptures of the jejunum and lhe colon, no accompanying intradeslomial injuries were presenl. _ one of lhe diagnostic procedure , whether c1inical examination, blood chemistry analysis, x-ray, ullrasound, or more an im'asi\'e procedures, can in ilself be more decisi\'ely adyantageous in cases of bowel injuries. The abdomen nati\'e x-ray, \\'hether in upright position or lateral \\'ith horizontal radiograph, presenls a good picture of the distribution of gas. In 15 % of lhe cases, accumulation of gas was found in lhe intra- or retroperitoneaJ areas (25). Abdominal ultrasound, a procedure that is quick and can ah\ays be repeated, is adntntageous in the clinical assessment of solid organs and free inlraperiloneal fluid, as weil as alteralions of the duodenum. This diagnostic procedure, howe\'er, cannot be lhe sole determinant of injury since the layered anatomy of the remaining intestines as \\'ell as the air \\'ilhin them makes determination of status in lhese areas more difficuJt (6, 8, 19). A similar problem occurs \\'ith computed tomography wh ich causes a high incidence of false negatiye findings in cases of bO\\'el injury (1-1, 15, 22). ,\1ore ad\'antageous are diagnoslic imagines \\ith conlrasl, OIthograde as weil as retrograde, since stenosing intestinal wall hemalomas as \\'ell as extra\'asations can be diagnosed. Ilowe\'er, in case of distal bo\\'el injuries, due to intestinal paralysis, diagnosis may be delayed. HighJy conlrO\'ersially discussed is lhe diagnostic perinealla\'age (DPL) as weil as laparo copy, because up lo 24 % false positi\'e resulls leaded lo unnece sary laparotomies (2, 11, 17, 18, 22). In addilion, chemieal as weH as microseopie analyses for leukoeytes or enzymes laken from lhe fluid from lhe periloneal Ja\'age did nol inerease the diagnostie preei ion of DPL (9, 12). Seintigraphie and angiographie depietions are valuable in diagnosing parenchyrnous organ injuries, especially for the diagnostie exclusion of rarely oeeurring vascular 1esions in childhood (3). Summing up, a possibiJity of isolated bowel injury in eases of blunt abdominallrauma may be present (after injury of parenchyrnous organs has been excluded) if - as a resull of patient history and meehanism of injury, the influence of localized oulside forees (i.e. lhe handlebar of a bieycle) can be delermined; - clinieal signs of a eireumseribed abdominal wall eontusion (for example bruises) are present, in spite of primary negative sonographie, radiographie, or laboratory findings, lf there i a traumatie ruplure of the bowe1, sulure of lhe site of rupture is usually sufficienl. In isolated inlances; in cases of duodenal wall hemalomas, a more eonseryalive procedure can be effective. The high rate of postoperalive eompJicalions (up to 50 %) is usually not a result of the operati\'e procedure ilself, but rather lhe result of delayed diagnosis and therapy (3, 10).
G. Schimp/ et al
Billing A, Zülke C, HufR, el al: Slumpfe \'erlelzungen des :\1agen,Darm, Trakles. Unfallchirurg 93 (1990) 62-65 2 Bl/mey RE, ,\lueller GL' Diagnosis of isolaled small bo\\el injury follo\\', ing blunl abdominal lrauma. Ann Emerg .\\ed 12 (1983) 7 I-7-1 3 Cywes S, Bass DH, ,\lillar AjW: Blunl abdominal lrauma in children, Pedialr Surg Inl5 (1990) 350-35-1 1 Dal/terive All, Flancbal/m L' Blunl abdominal lrauma. A modern re, I·ie\\. ,\nn Surg (I (85) I U8-2(J:j 5 GrosfeldjL. Reswria]. Wes/I\I\', el al: Gaslroinleslinal injuries in child, hood: .\nal,lsis of 53 palienls, J Pedialr Surg 2-1 (I U8U) 580-583 6 Gn'issller R..lIelllges B. Dübel' C. el al: Sonography I'ersus periloneal lal'age in blunt abdominallrauma, J Trauma 2U (I \189) 212-211 7 Iltlrm eR ßlunl abdominallrauma causingjejunal ruplure. Ann Emerg .\\ed I I (1985) 916-U18 H 1101)/11(/1111 I?, Pohlemallll T \\'ippermal1l1 H, et al: \\anagemenl der Sonographie bei slumpfen Bauchtrauma. L'nfallchirurg \12 (1989) -1711'16 \I jellob, I)(J, IllgI" L. RoJrigl(e~ I, et al: Peritoneall,l\'age \\ h,le counl: a reas,e"menl J Trauma 30 (IU\lO) 60'1-612 111 I\a/llillttllll CR Rir(/ra Fl', .lIttier R I' Pediatric t"auma: 'eed o[ surgical management J Trauma 2U (1 U8U) I 120-1 126 11 Lombardo C, .\loretli R Perforazione deli inle,llno lenue da lrauma chiu,o addome (rel'isione di I (j casi ,ollopo,ti a Irallemento chirurgico). ( hir Ilal :J 1 (I U82) ,I I '1-,125 12 .lIanj.1. 1I0rre EE, Bar /) I'eriloneallal'age in pl'netraling injuries of the ,mall bo\\el and colon I'alue of enzl'me delel'l11lnalion Ann Emerg \\ed 12 (HI8:.l) 68-'10 I~ ,lIa,oll LC. l"illocur CD, II'aglla CII". el al: Inteslinal perforation due lo blunllrmllna in children in an era of increasednonopel'ati"e trealment. J Trauma 26 (I U89) 161-163 11 .\Ja/subara TI\. FOllg lUfT Bums C.II Computed lomographyofabdomen (CT,\) in managemenl of blunl abdominal Irauma. J Trauma 30 (1990) 110 11 l 15 .\Jeyer. /).11, Thai ER, I\'eigelt j,l, el al: Elalualion of compuled lomography and diagnosic peritoneallal'age in blunl abdominallrauma. J Trauma 29 (1989) 1168-11'12 16 Oliveira Fj, Concalves 0, SanlosjD, el al: Les perforations du grele au cours des lraumatismes fermes de I'abdomen. Apropos de 63 observations. J Chir Paris (198-1) 97- I00 11 Powell RW, GreenjB, Ochsner ,\JG, el aI: Periloneallal'age in pedialric patienls suslaining blunl abdominaJ lrauma: a reappraisel. J Trauma 27 (1987) 6-10 18 Rothenbers: S, Moore EE, lllarxjA, el aI: Selecli"e managemenl of blunl abdominal lrauma in children - lhe triage role of periloneal lavage, J Trauma 27 (1987) 1101-1106 19 RufW, Friedl 11', Weber G, el al: SlellL der sonographische Nachweis von Blul im Abdomen nach slumpfen Bauchlrauma in jedem Fall eine Operationsindikalion dar? L'nfallchirurg 93 (1990) 132-136 20 Sauer, H: Bauchlraurna, in H. Sauer. Das I'erlelzle Kind. Georg Thieme \'erlag, Slullgarl" eil" York (I 98~) pp 339-373 21 Schenk \FG, Lonchyna I', l\1oylanjA: Perforalion of lhe jejunum from blunl abdominallrauma. J Trauma 23 (1983) 54-56 22 herck jP, Oakes 00' Inleslinal injuries missed by compuled lomography. J Trauma 30 (1990) 1-7 23 Sorkey Aj, Farnell MB, \\'illiams Hj, el al: The complemenlary roles of diagnoslic periloneallal·age and compuled lomography in lhe el'alualion of blunl abdominallrauma. Surgery 106 (1989) 791-80 I 2~ S/e7.'ens L..lIal/lI KI- Small bo\\el injuries. Surgical Clinics of "lorlh America 70 (I (90) 511-560 25 lI'elch I