Blunt Trauma to the Abdomen * Jom H. MORTON, M.D., J. RAYMOND HINSHAW, M.D., D.PHIL. (OXON.) JOHN J. MORTON, M.D. Rochester, New York FromtheDepartment ofSurgery, University of Rochester School of Medicine and Dentistry, Rochester, New York IT IS GENERALLY AGREED that penetrating Forty-nine of our patients, listed in Table wounds of the abdomen necessitate sur- III, underwent laparotomy. Seven of the gical intervention as soon as the patient's cases operated upon ended fatally, the general condition permits. When, however, cause of death being outlined in Table IV. the patient suffers blunt trauma to the ab- The accidents producing these injuries are domen, the surgeon must decide whether summarized in Table V. Accidents reported or not laparotomy is indicated, and an in- in the literature parallel these but also in- correct decision may be catastrophic. In an clude certain oddities such as injury from attempt to determine what clinical and a pneumatic drill 7 and from the high heel laboratory findings are most useful in diag- of a shoe in the hand of an irate wife.55 nosis, which signs may be misleading, and Any abdominal structure may be dam- whatfactors may cause an unjustified delay aged by non-penetrating trauma, and many in treatment, the cases of blunt abdominal bizarre injuries have been reported. How- trauma at the University ofRochester Med- ever, the organs most commonly injured in ical Center have been reviewed and per- this way are the kidney, spleen and liver.1, tinent literature has been surveyed. 19,20,81 Since injuries to the kidney can usu- Case histories of 120 patients from ally be managed conservatively, rupture of Strong Memorial and Rochester Municipal the spleen is the most common single entity Hospitals have been studied. This review requiring operation. Our series, in which includes all non-penetrating abdominal splenectomy was done in 32 of 49 cases trauma treated surgically or discovered at undergoing operation, bears this out. The autopsy plus representative injuries man- frequency of hepatic damage varies con- aged conservatively. The sites of injury are siderably from one series to another. Ex- summarized in Table I. All patients who cluding those patients with multiple injury died after sustaining multiple severe in- in whom operation was out of the question, juries and in whom abdominal trauma was only three patients with surgically signif- of little or no importance have been ex- icant trauma to the liver or extrahepatic cluded from further evaluation. In asmaller biliary system were encountered at this group of cases, death was directly due to center. The two patients who sustained an undiagnosed abdominal injury discov- major wounds of the liver both died of ered at autopsy. These patients, whose hemorrhage. The one patient with injury deaths might have been prevented by sur- of the extrahepatic biliary system recovered gical treatment, are included in the report satisfactorily and has been reported else- and the sites of trauma listed in Table II. where.21 Welch and Giddings81 reported *Presented before the Southern Surgical As- that 9.5 per cent of their 200 patients sus- sociation, Boca Raton, Florida, December 4-6, tained hepatic injury, and other figures 11,28 1956. agree closely with theirs. However, at Har- 699 700 MORTON, HINSHAW AND MORTON Annals of Surgery May 1957 TABLE I. Cases ofBlunt Abdominal Trauma Reviewed, TABLE II. Cases DiagnosedatAutopsy UniversityofRochesterMedicalCenter (Surgical TreatmentPotentially Curative) SiteofInjury No. SiteofInjury No. Spleen 37 Spleen 5 Liver 5 Liver 2 Pancreas 2 Gastro-intestinal tract 1 Gastro-intestinal tract 13 Urinary bladder 1 Kidney 26 Urinary bladder 4 Total 9 Abdominalwall 3 Retroperitoneal space 2 Multiple severe injuries 32 with a two year history of crampy abdom- Total 120* inal pain and loss of weight. He dis- was *Multipleinvolvement in4cases: (1) spleenandleft covered to have marked anemia and a left kidney, (2) spleen and stomach, (3) pancreas and upper quadrant mass. Pathologic examina- stomach, (4) ileum, bladderand urethra. tion of the specimen removed at laparot. showed old hematoma surrounding omy an lem Hospital,85 over 33 per cent of the pa- an atrophic spleen. Olander and Reimann53 tients withintra-abdominal injury sustained reported a patient with symptoms begin- significant hepatic damage. ning two years after the only known ab- Rupture of the spleen is a well recog- dominal trauma. Operation six months after nized entity which has been extensively the onset of symptoms disclosed a hema- reported.38 5 7 The usual case requires toma surrounding a spleen which had early operation to prevent exsanguination, apparently been bleeding intermittently. but cases are encountered in which rupture Schwartz67 reported a large symptomatic and hemorrhage are late events. Mac- cyst of the spleen which he believed to be Auley43 reported a patient in whom the the late result of a subcapsular rupture. spleen was entirely avulsed fromitspedicle, One of the unusual late events following but massive hemorrhage was prevented by rupture of the normal spleen is the im- early thrombosis of the severed vessels. The plantation of splenic tissue throughout the spleen was removed more than 24 hours peritoneal cavity,40' 72 and secondary in- after injury primarily because of abdominal testinal obstruction may result from this. pain. Delayed rupture following subcap- All authors agree that mortality from sular hematoma is reported in about 14 rupture of the liver is very high.23' 24,85 The per cent of cases.88 Although this com- relatively low incidence of liver trauma in monly occurs during the first two weeks various hospital series is probably due to following trauma48 occasional patients have death from hemorrhage at the scene of the minor undiagnosed difficulties over long accident in many severe hepatic injuries.65 periods of time. One of our cases was hos- When central subcapsular hepatic or rup- pitalized elsewhere for three weeks after ture occurs, later biliary tract hemorrhage injury during which time he complained of may result.27' 64, 73 These patients present a intermittent abdominal pain. One week puzzling picture of intermittent major gas- after his discharge he was admitted to tro-intestinal hemorrhage, biliary colic or Strong Memorial Hospital with abdominal incomplete extrahepatic biliary obstruc- tenderness, leukocytosis and hypotension. tion. Hemobilia also follow surgical may At operation both fresh and organized repair of rupture of the liver.18 Sympto- hematomas were found around a lacerated matic intrahepatic bile cysts following sub- spleen. Another patient entered the hospital capsular rupture have been reported.54 The Volume 145 BLUNT TRAUMA TO THE ABDOMEN 701 Number 5 extrahepatic bile ducts may be injured in TABixJII. Non-PenetratingAbdominal Trauma, the absence of damage to the liver or other Surgical Cases organs.46 These patients may have mild No.of intermittent symptoms with consequent de- SiteofInjury Cases Fatalities lay in diagnosis and treatment. Both im- mediate and delayed rupture of the gall LSipvleerenandbiliary tract 332 42 bladder have occurred,3,15,50,52,71 although Gastro-intestinal tract 12 2 injury to the liver or the bile ducts is more Pancreas 2 0 Cholecystectomy has been Kidney 2 0 common. accom- Urinarybladder 1 0 plished with almost surgical precision by Retroperitoneal space 1 0 blunt trauma to the right upper quadrant.6 44' 55 In two of these cases the gall bladder Total 49* 7* was recovered from the pelvis at laparot- *Multipleinvolvementin4cases (1 fatal): (1) spleen omy. and left kidney, (2) spleen and stomach (fatal case), ShallowandWagner69reportedthat (3) pancreas and stomach, (4) ileum, bladder and non- urethra. penetrating abdominal trauma accounts for two to four per cent of cases of acute pan- TABLE IV. DeathsFollowingOperation creatitis, but our own figure is not nearly this high. Pancreatitis is more common fol- SiteofInjury No. lowingbluntthanpenetratingtrauma 49 and Spleen4 may be produced by a trivial accident.36,80 Multipleinjuries* 1 Symptoms may be immediate but are often BTernosnicohnoppnneeuummootnhioarax, untreated 11 delayed for 72 hours or even longer.86 Latebowelinfarction, unrecognized 1 Most commonly the organ is contused but Liver2 laceration or complete transection may oc- GasHtermo-oirnrthesatgienal tract 1 2 cur.36 Pancreatic calcification and pain have Latediagnosis 1 been reported late sequelae following as blunt abdominal injury37 although most *Rupture of stomach and open chest wound also present. cases of this syndrome are not associated with known trauma.33 TABLE V. TypeofAccident Any part of the gastro-intestinal tract may be damaged by blunt trauma, al- AFaultlomobile 3137 though gastric injury is rare.82 The sites of Struck bycar 14 injury in our 13 cases with gastro-intestinal Blow toabdomen 13 trauma were as follows: stomach two, retro- USlnekdnown 36 peritoneal duodenum one, jejunum four, jejunum and ascending colon ileum one, two, transverse colon one, descending colon occurred in the jejunum and ileum, ten per one, sigmoid colon one. The only ruptures cent in the duodenum and four per cent at fixed points in the bowel were those in in the large bowel. These figures have not the fourth portion of the duodenum and been significantly changed by later re- in the sigmoid colon at the peritoneal re- views.4 56 Injuries of the retroperitoneal flection. Most of the small intestinal rup- duodenum have attracted special interest tures occurred along the anti-mesenteric because of the difficulty in diagnosis even border although in one instance the mid- at laparotomy.12,20,81 Siler 70 reported one jejunum was completely transected. Coun- patient with duodenal injury in whom op- seller and McCormack16 found that 80 per eration was notperformed untilfive months cent of non-penetrating intestinal injuries after injury. At this time a large walled off 702 MORTON, HINSHAW AND MORTON Annals of Surgery May 1957 retroperitoneal abscess was present and the tended down into the renal vein. The sec- initial diagnosis perforatedcarcinoma ond patient developed hematuria several was a of the ascending colon. Unusual injuries of days after being gored by bull. Because a the bowel include two cases of traumatic hematuria recurred intermittently for six appendicitis,7'22 the expulsion per rectum weeks and because retrograde pyelograms of a segment of gangrenous ileum,2 and an showed a mass in the superior pole of the example of jejunal obstruction from sub- right kidney, he was explored. A bifid serosal hematoma.39 Two additional re- ureterwas presentwith the upper one lead- ports32 42 described intestinal obstruction ing from a huge hydronephrotic sac which dueto the trapping of a loop of small bowel contained an organized hematoma and between injured bony structures. bloody fluid. The lower ureter communi- Rather than affecting the bowel directly, cated with the rest of the kidney and a injury may involve primarily the mesentery fracture of its upper calyx with hematoma and blood supply.29' 32,78 In one of ourcases formation was present. The kidney was re- of injury to the ileum the mesentery had moved. A most unusual occurrence follow- been torn away from a nine centimeter ing trauma was the complete transection length of bowel which was otherwise un- of a horseshoe kidney reported by Corco- damaged. The second case of ileal injury ran.13 Non-penetrating damage to the ure- was misdiagnosed until very late in its fatal ters is extremely rare and almost always course, and the damage was probably of associated with other major injury. Rusche this same nature. Mesenteric damage may and Hager63 reported two cases of this result in late perforation of the bowel sev- type. Either intraperitoneal or extraperi- eral days after injury.26 One of our patients toneal rupture of the urinary bladder may showed at laparotomy shortly after injury be produced by trauma.58 In the majority a large retroperitoneal hematoma involving of cases an associated pelvic fracture is the descending mesocolon, but the bowel present,14 and bony fragments projecting wall itself appeared viable. Six days later through the bladderwall maypreventheal- perforation of this segment necessitated an ing. One of our fatal cases might possibly emergency left colonic resection. Chylous have survived if a small bladder perfora- peritonitis may also occur secondary to tion with pelvic peritonitis had been recog- mesenteric injury.41 nized and treated. Only the most severe lacerations of the Direct trauma to the adrenal glands, in kidney require nephrectomy for the control the absence of other serious injury, must of hemorrhage. Conservative management be extremely rare. Sevitt68 recorded post in milder injuries gives uniformly good re- traumatic adrenal hemorrhage occurring, sults.74 Postmortem examination of two of he felt, as a primary effect of trauma. He our patients who died of head injuries reported that crushing injury to adrenal shows why conservative management is so vessels led to multiple focal hemorrhagic successfuL Both patients had gross hema- areas in the adrenal substance. Although turia within 48 hours of death, but the only adrenal hemorrhage was present in several lesions of the urinary tract demonstrable at of our fatal cases, it is impossible to deter- autopsy were a minute calyceal fracture in mine what part this played in the patients' one and several small, shallow lacerations deaths. of the kidney parenchyma plus a perirenal Rupture of the uterus from abdominal hematoma in the other. One of our two pa- trauma has occurred only during preg- tients treated surgically sustained a lacera- nancy, and even then it is rare. McClure47 tion which divided the superior pole from recorded one case and mentioned less than the remainder of the left Iddney and ex- 50 others from the literature. Woodhull's Volume 145 BLUNT TRAUMA TO THE ABDOMEN 703 Number 5 patient83 underwent operation one month pain and priapism. However, one of our after injury. A macerated fetus was found patients with shoulder pain as a primary in the pelvis outside a badly torn uterus. complaint proved to have only an abdom- Vaginal bleeding had not occurred in this inal wall contusion. A latent period be- patient and was minimal in McClure's case. tween injury and onset of symptoms may Injury to the great vessels in the abdo- be important in the diagnosis of such ab- men is rare. Strassmann,75 in reviewing 72 dominal injuries as subcapsular ruptures ruptures of the aorta from blunt force, and retroperitoneal injuries. This has been found only three cases involving its abdom- emphasized in retroperitoneal ruptures of inal portion. Traumatic aortic aneurysms the duodenum by several authors,12, "I but have occurred in the thoracic aorta,25 but it is not an invariable rule since our one no case has been reported involving the patient with this injury had steady, severe abdominal aorta. Only penetrating wounds pain from the time of accident. A history of the inferior vena cava are on record al- of vomiting blood or of passing bloody though caval thrombosis following abdom- stools or urine may aid in localizing dam- inal trauma may at times have been due to age. The constant desire to void with in- an unsuspected injury of the vessel wall. ability to do so is almost pathognomonic of At least one case of portal vein thrombosis bladder or urethral injury.14 Dyspnea is following trauma is on record.SO commonly a prominent symptom with splenic rupture, but this is rarely useful in DIAGNOSIS, AIDS AND PITFALLS diagnosis since other for dyspnea, causes Whenever a patient complains of abdom- particularly fractured ribs, aresofrequently inal pain following non-penetrating abdom- present concomitantly. inal trauma, the physician is confronted The prominent physical findings to be with a diagnostic problem. If a serious anticipated are those of shock and of peri- intra-abdominal condition is overlooked, toneal irritation. Although retroperitoneal delay may lead to long morbidity or to a hemorrhage may produce this combination, fatal outcome. If a minor injury is misin- it is usually indicative of a lesion necessitat- terpreted, an unnecessary laparotomy may ing operation. Shifting dullness, if present, be performed. If the patient has major in- is a valuable sign. The harder problem is juries elsewhere, this unnecessary proce- presented by the patient in whom signs are dure may be extremely ill advised. To aid less pronounced or are masked by trauma the physician in his decision, a variety of elsewhere. The full blown picture of shock diagnostic procedures has been suggested. is not an early event in most abdominal An accurate history and repeated thor- trauma, and a plan of therapy should be ough physical examinations are of primary established before it occurs. With certain importance. The type of trauma sustained injuries of surgical importance, particularly may give a clue to the potential severity of the retroperitoneal ones, the usual signs of the injury but it can also be misleading. parietal peritoneal irritation may be en- One of our patients suffered a ruptured tirely lacking. Profound shock may also spleen by falling off a couch; another com- obscure the picture. In one of our patients pletely transected his jejunum during the with a ruptured liver, abdominal examina- simple act of bowling. The type and loca- tion was unremarkable at the time of hos- tion of pain are important although local- pital admission. Several hours later, after ization of pain does not always indicate the blood transfusion, there was obvious evi- site of injury. Injuries in the diaphragmatic dence of intraperitoneal fluid, and at lap- area may be suggested by shoulder pain, arotomy major hemorrhage from a massive in the retroperitoneal area by testicular rupture ofthe liver was discovered.Trauma 704 MORTON, HINSHAW AND MORTON Annals of Surgery May 1957 to either the abdominal wall or viscera usu- dicative of pancreatic damage but it does ally produces a decrease in peristaltic ac- not, of course, rule out other abdominal tivity, but continued peristalsis has been re- pathology. The value of serum amylase de- ported even after rupture of the intestine.43 terminations in non-penetrating trauma is The findings from routine laboratory in- thus somewhat limited. vestigation are equally inconclusive. Some In an effort to establish a more accurate degree of anemia is the rule after these in- diagnosis, abdominal paracentesis has been juries, but exceptions are numerous. Ane- advocated. Byrne,9 who advises tapping mia, when present, does not determine the each of the four abdominal quadrants with type of injury or its proper management. a small needle, reported 83 per cent ac- Leukocytosis is similarly non-diagnostic. curacy with this technic. Even in his hands, Elevation of the white blood count is a however, the paracentesis was negative in finding with intra-abdominal hem- eight where ruptured spleen common cases a was orrhage but is not rare with hematomas of removed at laparotomy and in two with in- the abdominal wall or with kidney damage, juries to the small intestine. The recovered neither of which requires operation in most material is usually bloody fluid, but air, in- cases. The white blood count may be nor- testinal contents or bile-stained fluid may mal, although rarely so, with any of the also be found. If traumatic pancreatitis is more serious intra-abdominal injuries. Ex- suspected, the amylase activity of the fluid amination of the urine for red blood cells can be measured. Amylase activity usually is important in establishing the presence reaches higher levels in abdominal fluid of injury to the urinary tract. Hematuria is than in serum, but the elevation may not ordinarily noticed early and frequently occur for two or more days after injury.9 leads the patient to seek medical advice, Paracentesis may be misleading even when but its appearance is occasionally delayed positive. In one of our cases bloody fluid for several days. Once present, it usually was recovered, but at operation it was dis- disappears slowly over nine or ten days, but covered that this fluid had been aspirated infrequently stops as abruptly as it began. from an unusually large retroperitoneal When a patient is unable to void, cautious hemorrhage rather than from theperitoneal catheterization of the urinary bladder for cavity. Needle paracentesis was performed diagnosis is desirable. The immediate re- in 25 of our patients who underwent subse- turn of abnormally large volumes of fluid quent laparotomy, and positive taps were is considered indicative of bladder rup- reported in 16. However, non-diagnostic ture.10 It has been suggested 28 that sterile taps were recorded in six ofthe 15ruptured saline be injected through the catheter and spleens and in three of the intestinal seven then aspirated, the recovery of an equiv- injuriesinwhichparacentesiswasdone.One alent amount of fluid being viewed as evi- patient with a negative tap had more than dence of integrity of the bladder wall. This one liter ofbloodfree in the peritoneal cav- practice is generally condemned,34 58 both ity when a ruptured spleen was removed because of the danger of further extravasa- a short time later. tion and because the recovery of equivalent Several types of x-ray examination have or greater amounts of fluid can occur with been suggested as aids in diagnosing ab- bladder rupture. Failure to appreciate this dominal injuries. The most commonly em- fact led to a fatal misdiagnosis in one of ployed are the plain film of the abdomen our patients with a bladder rupture. The and the upright or lateral decubitus film attempt to pass a cystoscope as a diagnostic forfreeair.35 Haziness ofthe plain film may aid is not advised.58 Elevation of the level suggestfree abdominalfluid, buteven large of serum amylase following trauma is in- amounts of fluid may be impossible to de- Volume 145 BLUNT TRAUMA TO THE ABDOMEN 705 Number 5 tect.A7 Obliteration of renal outlines or of amined with the usual roentgenologic tech- psoas shadows suggests retroperitoneal niques, and positive or suspicious findings edema or hemorrhage.79 The gas pattern in werepresent in ten. Findings were negative the gastro-intestinal tract can be evaluated, in 14 of the 19 cases of splenic rupture and and a search can be made for injuries fre- in four of the six intestinal injuries in which quently associated withthe intra-abdominal x-ray studies were done. We have been less pathologic conditions. These include frac- successful in establishing the diagnosis of tures of ribs, lumbar transverse processes ruptured spleen by plain abdominal films and the bony pelvis. The finding of collec- than have some authors, but our negative tions of retroperitoneal gas is indicative of findings in intestinal injury parallel the ex- extraperitoneal injury to the duodenum, perience of others. but the films in our case were not diagnos- Several other x-ray technics have been tic. Wyman 87 felt that enlargement of the reported occasionally. The use of a lipiodol splenic outline following trauma was pa- swallow to demonstrate tears of the retro- thognomic of splenic injury and that it was peritoneal duodenum has been recom- a reasonably constant early sign. He con- mended.20 The ingestion of barium to out- sidered serration of the greater curvature line the greater curvature of the stomach of the stomach, a sign mentioned by many and to demonstrate gastric displacement in authors, to be an unusual finding. Wang suspected splenic injuries has been sug- and Robbins 79 concluded that the presence gested.1' 7 Rudolph62 reported an unsuc- of a small, well outlined spleen on a good cessful attempt to evaluate possible splenic plain film practically rules out splenic in- injury by percutaneous splenoportal venog- jury even with a history of abdominal raphy, and Burke and Madigan8 attempted trauma. The free air films have been dis- to diagnose hepatic and splenic injuries by appointing in the diagnosis of traumatic in- the intravenous injection of ThorotrastO testinal rupture. For example, Jacobson and followed by abdominal films in several Carter31 found subdiaphragmatic free air hours. Estes found the latter technic of 19 in only two of 16 small bowel perforations, some value in cases of ruptured liver. None and because of the paucity of positive find- ofthese technics has been widely employed ings they considered x-ray examination of and, with the possible exception of the little valueinexcluding intestinal injury fol- lipiodol their further swallow, use seems lowing blunt trauma. Intravenous pyelog- unwarranted in the acutely injured patient. raphy and retrograde cystography may be Using the combination of paracentesis of considerable value in diagnosing injuries and appropriate x-ray examination, the sur- of the urinary tract. Many patients with geon can make a proper decision in most renal trauma have normal intravenous py- difficultcases whereclinicaljudgmentalone elograms even in the period immediately is not conclusive. In our series both meth- after trauma, but the damage in these cases ods of study were used in 17 of the pa- is usually slight and they can be managed tients undergoing laparotomy, and both conservatively. The demonstration of a con- were negative in only four. The four fail- tralateral kidney with normal function is of ures occurred in two of the ten splenic rup- paramount importance in the unusual case tures and in two of the four intestinal in- where nephrectomy must be considered. juries in which both tests were performed. Gravity cystograms, if carefully done, can Thus, in a certain small number of patients demonstrate bladder injuries with minimal an exploratory laparotomy must be under- risk. They provide the best method of taken diagnostic procedure.57 Careful es- as a tablishing this diagnosis.'4 Twenty-eight of examination at the operating table should our patients requiring operation were ex- demonstrate readily any intraperitoneal 706 MORTON, HINSHAW AND MORTON Annals of Surgery May 1957 damage, but retroperitoneal injury may has been long abandoned because of the produce no intraperitoneal pathologic frequent occurrence of subsequent hemor- change. Duodenal trauma has been fre- rhage. quently overlooked at operation with disas- Although most surgeons now feel that trous results.51 This injury should be sus- patients with lacerations of the liver should pected whenever there is emphysema, undergo early operation, conservative man- hematoma or bile-staining of the transverse agement is still advocated by some.19 In mesocolon or of the posterior parietal peri- operative management the surgeon's prime toneum. In the presence of any of these consideration is the control of hemorrhage, signs the duodenum should be mobilized but how this can be most safely accom- so that its entire length can be visualized. plished with the least morbidity will vary Even when bile-stained fluid is found in according to the location and severity of the peritoneal cavity, the site of rupture in the injury. Small rents may be closed,28 but the biliary system may not be apparent. obviously necrotic tissue should be re- The use of operative cholangiography moved before placing the sutures. Depend- should be helpful in this situation. ence on Oxycel® gauze or Gelfoam® as a There is no easy road to establishing definitive treatment can not always be the proper diagnosis in these interesting avoided when dealing with large lacera- but difficult cases. Attempts such as that tions. If a foreign substance or necrotic of Knopp and Harkins 38 to describe the tissue must be left in the liver, subsequent "typical ruptured spleen patient" represent escape of bile can be expected and the an ill-advised use of statistics and may give abdominal cavity should, therefore, be a false sense of security to the unwary. The drained. Fragments of liver tissue may ap- surgeon must realize that diagnostic aids pear in the drainage.65 Sparkman73 sug- are of importance only when they are posi- gests that hemobilia occurs as a complica- tive. Careful, repeated physical examina- tion of the treatment of lacerations of the tions and thoughtful analysis of each case liver because bile under pressure in the still afford the best opportunity for correct resulting cavity may produce continued diagnosis.76 autolysis of the lining hepatic tissue. Since surgical decompression of the biliary tract OPERATIVE TREATMENT is therecommended treatmentfor this com- Once the decision for laparotomy has plication, perhaps it might also prevent its been reached, the surgical management occurrence. The only alternatives seem to often presents no major problem. However, be close apposition of viable liver tissue, the difficulty in reaching this decision led which is desirable and drainage anyway, Breidenbach 5 to feel that penetrating ab- of the potential cavity below the sutured dominal wounds, all of which require op- surface. The latter has the disadvantage of eration, are usually better managed than other packing methods, the possibility of non-penetrating ones. Many authors 11,56,57 secondary hemorrhage after removal of the agree with this point of view and urge packs. The inadvisability of suturing lacera- that exploration be done whenever a well tions which extend almost through the sub- founded suspicion of intra-abdominal in- stance of the liver is demonstrated by one jury is present. of our cases in which a major portion of The treatment for ruptured spleen is, of the left lobe was all but completely tran- course, splenectomy. It has been esti- sected. Although suturing provided tem- mated " that the mortality without opera- porary hemostasis, at autopsy the disrupted tion is at least 90 per cent. The practice fragment was completely necrotic. Removal of packing or suturing a ruptured spleen of the injured tissue would have been bet- Volume 145 BLUNT TRAUMA TO THE ABDOMEN 707 Number 5 ter treatment. Large lacerations in the because one site of damage is readily ap- dome of the liver may best be approached parent. The chief complications which may through a thoraco-abdominal incision.'7 Al- be avoided by the choice of the correct though plain gauze packing is frequently procedure are stenosis at the site of injury condemned, if the surgeon is unable to and delayed perforation subsequent to im- control hemorrhage by more desirable pairment of the blood supply. Stenosis is means he should be prepared to do this. less likely if the more extensive injuries are Before resorting to this least satisfactory treated by bowel resection rather than by form of treatment, however, he should simple closure. Our case of damage to the think of the invariable complication of sec- mesentery of the descending colon demon- ondary infection and the frequent of strated the difficulty of determining the one subsequent hemorrhage. adequacy of the remaining blood supply. Injury to the extrahepatic bile ducts usu- Here the surgeon convinced himself that ally requires little more than an escape the segment of colon was viable but, never- route for the leaking bile, and decompres- theless, on the sixth day after injury had sion of the ducts with a catheter in the to reoperate because of delayed perfora- common duct or gall bladder. Subsequent tion. In retrospect, this combination of con- cholangiography should be used to check tused bowel and decreased blood supply the integrity of the duct. The usual lesion would have been better handled by ex- will not require operative repair either pri- teriorization or resection. marily or secondarily, and, if the rent is Only the most severe lacerations of the small, primary suturing maybe inadvisable. kidney demand immediate operation to Traumatic ruptures of the gall bladder prevent exsanguination. Although bleeding may require cholecystectomy, but cholecys- may at times be controlled by pressure or tostomy or suture of the rent may also be suturing, injury involving the renal pedicle, satisfactory.71 The treatmentwilldependon similar to that in one of our cases, requires the injury but is, in any case, no problem. nephrectomy. That later nephrectomy may Isolated pancreatic injury, if the diagno- be necessary following renal trauma is il- sis can be established, is not an indication lustrated by our case described in a previ- for immediate operation but it may lead to ous section. late complications despite the initial treat- Rupture of the urinary bladder requires ment.80 An obvious tear, discovered at op- immediate operation. The perforation eration, should be sutured to prevent fur- should be closed and a suprapubic cystos- ther bleeding or leakage of pancreatic tomy done, but if the injury can not be secretions,34 and drainage of the lesser sac readily located or is found in an inacces- might.render subsequent development of a sible area, cystostomy alone will suffice.34' pseudocyst less likely. Both of our cases of 58 Bony spicules which may extend from a traumatic pancreatitis required drainage of pelvic fracture into the bladder must be a collection of fluid in the lesser sac three removed to permit satisfactory healing. weeks after the initial trauma. Uneventful Rupture of the pregnant uterus can usu- recovery followed in each instance. ally be handled by extraction of the fetus Injuries to the gastro-intestinal tract and and placenta followed by repair of the mesentery should be managed by suture lacerated uterus. In the severely ill patient, of lacerations, resection of bowel, side- however, hysterectomy be faster and may tracking anastomosis, exteriorization or con- safer than attempted suture of a large tear trol of hemorrhage as the individual case in the vascular uterine wall.83 demands. The surgeon should not overlook Injury to vital blood vessels would or- the possibility of multiple injuries merely dinarily be treated by repair or replace- 708 MORTON, HINSHAW AND MORTON Annals of Surgery ment of the involved segment. An interest- and thorough physical examination still ing example is Ulvestad's patient78 in provide the best information for minimiz- whom suture of the superior mesenteric ing error. artery successfully controlled hemorrhage and prevented massive bowel gangrene. BIBLIOGRAPHY In 1931 Robertson 'I reviewed the ltera- 1. Bancroft, F. W.: Cited by Zabinski and ture and his own experience in an attempt Harkins.88 to determine the best management of non- 2. Benson, C. D., W. S. Carpenter and R. D. penetrating abdominal trauma. Today there Swedenberg: Spontaneous Expulsion of Se- questrated Ileum. Ann. Surg., 137: 261, is still no argument with certain of his con- 1953. clusions. For example, differential diagno- 3. Benson, C. D. and F. W. Prust: Traumatic sis remains difficult and often impossible; Injuries ofthe Liver, Gallbladder and Biliary each case presents a different problem and Tract inthe Infant and Child. S. Clin. North should be dealt with accordingly; and any America, 33: 1187, 1953. 4. Bosworth, B. M.: Perforation of the Small In- trauma involving the abdomen calls for testine from Non-penetrating Abdominal critical study and carefulobservation. How- Trauma. Am. J. Surg., 76: 472, 1948. ever, surgeons no longer agree that con- 5. Breidenbach, L.: Discussion of Bosworth.4 servatism is usually the best policy. The 6. Brown, H. P., Jr.: Traumatic Cholecystectomy. availability of blood for transfusion and Ann. Surg., 95: 952, 1932. 7. Burgess, C. M.: Traumatic Appendicitis. the advances in anesthetic technic today J. A. M. A., 111: 699, 1938. permit earlier operation with consequent 8. Burke, W. F. and J. P. Madigan: The Roent- greater hope for cure in some of the seri- genologic Diagnosis of Rupture of the Liver ously injured patients. and Spleen as Visualized by Thorotrast. The rumble seat concerned Robertson in Radiology, 21: 580, 1933. 9. Byrne, R. V.: Diagnostic Abdominal Tap. the same manner that the steadily increas- West. J. Surg., 64: 369, 1956. ing horse power of current automobiles 10. Campbell, M. F.: Rupture of the Bladder. worries the surgeon of today. Unless safer Surg., Gynec. and Obst., 49: 540, 1929. cars or saner drivers are designed, the sur- 11. Clarke, R.: Closed Abdominal Injuries. Lancet, geon can anticipate an increasing number 267: 877, 1954. 12. Cohn, I., Jr., H. R. Hawthorne and A. S. of these difficult cases as the years go by. Frobese: Retroperitoneal Rupture of the Duodenum in Non-Penetrating Abdominal SUMMARY AND CONCLUSIONS Trauma. Am. J. Surg., 84: 293, 1952. The records of 120 patients seen at the 13. Corcoran, J.: Discussion of MacAuley.43 University of Rochester Medical Center 14. Cottrell, J. C.: Nonperforative Trauma to after blunt abdominal trauma have been Abdomen. Arch. Surg., 68: 241, 1954. 15. Coulter, D. F.: Traumatic Delayed Rupture reviewed. Our experience in the diagnosis of the Gall-bladder in a Child Aged 9. and care of these patients has been com- Brit. M. J., 1: 198, 1948. pared with that of others. 16. Counseller, V. S. and C. J. McCormack: Sub- The injuries, common and bizarre, which cutaneous Perforation of the Jejunum. Ann. may result have been described and their Surg., 102: 365, 1935. 17. Devine, J. W., Jr. and S. Burwell: Thoracico- treatment discussed. Various diagnostic Abdominal Approach to Rupture of the aids may be of help when positive, but our Liver. Am. J. Surg., 78: 695, 1949. study shows that negative findings with 18. Epstein, H. J. and B. Lipshutz: Hemobilia, major injury are too common to permit Cholecystitis, and Gastrointestinal Bleeding greatreliance on anyof them. Correctdiag- with Rupture of Liver. J. A. M. A., 149: 1132, 1952. nosis and proper management are depend- 19. Estes, W. L., Jr.: Present-Day Problems in ent upon individual evaluation of each pa- Non-penetrating Abdominal Trauma. Bull. tient. In this evaluation a careful history Am. Coll. Surgeons, 39: 11, 1954.