CHAPTER XXIII
Wounds of the Spleen (341 Casualties)
H. Leon Poole, M. D.
The experience of the 2d Auxiliary Surgical Group with woundsof the spleen was at some variance with previous experiences with theseinjuries. The frequency was greater, the proportion of multivisceral injurieswas larger, and the damage to the organ was more severe than previous recordsshowed, but the case fatality rate was lower, probably because policies oftreatment, including operative approaches, were different.
Records of 3,546 abdominal wounds sustained by the AmericanExpeditionary Forces in World War I include only 49 instances1 ofinjury to the spleen, a frequency of 1.4 percent. Bailey,2 in 1942, reportedonly 54 splenic injuries among British troops and estimated their frequency at5.6 percent. Jolly3 reported a percentage of 4.6 injuries of thespleen in a series of 238 cases of abdominal injuries in the Spanish Civil War.Giblin4 reported 3 splenic injuries among 90 abdominal woundsobserved during the El Alamein campaign. Splenic injuries occurred in 32 cases,5.1 percent of the total number, in the two periods covered by Ogilvie's5report in 1942 on abdominal wounds in the Western Desert.
In contrast to these figures, the surgeons of the 2d Auxiliary Surgical Groupobserved 341 instances of splenic injury in the 3,154 patients with abdominalinjuries treated during 1944 and 1945 (10.8 percent) (table 79). This wasgreater than the frequency of 6.4 percent (22 of 346 abdominal injuries)recorded by the same surgeons in 1943.
Two hundred and fifty-three of the three hundred and forty-one wounds (74.2percent) were thoracoabdominal injuries. In their entire experience from 1943through 1945, the group surgeons treated a total of 903 thoracoabdominalinjuries, in which the left diaphragm was involved 468 times. There were 277wounds of the spleen (59.2 percent) in these 468 cases.
NATURE OF INJURY
The agents causing injury to the spleen were in general ofthe same type and operated with essentially the same frequency as thoseresponsible for other abdominal injuries. Blast was recorded as the cause ofthree wounds. Of
1(1) The Medical Departmentof the United States Army in the World War. Washington: U. S. Government Printing Office, 1927, vol. XI, pt. 1. (2) This number does not include two cases ofinjury to the spleen among American troops in Russia.
2Bailey, Hamilton: Surgery of Modern Warfare. Edinburgh: E. & S.Livingstone, 1941-42.
3Jolly, Douglas W.: Field Surgery in Total War. New York: Paul B. Hoeber, Inc., 1941.
4Giblin, T.: AbdominalSurgery in the Alamein Campaign. Australian & New Zealand J. Surg. 13:37-64,July 1943.
5Ogilvie, W.H.: Abdominal Wounds in the Western Desert. Surg., Gynec. & Obst. 78:225-238, March 1944.
292
TABLE 79.-Essential data in 341 injuries of spleen
Type of wound | Cases |
| Deaths | Case fatality rate | |
| In 341 splenic injuries | ||||
|
|
| Percent |
|
|
Univisceral | 100 | 3.2 | 29.3 | 12 | 12.0 |
Multivisceral | 241 | 7.6 | 70.7 | 73 | 30.3 |
| 341 | 10.8 | 100.0 | 85 | 24.9 |
Thoracoabdominal | 253 | 8.o | 74.2 | 67 | 26.5 |
Abdominal | 88 | 2.8 | 25.8 | 18 | 20.5 |
| 341 | 10.8 | 100.0 | 85 | 24.9 |
the 3 non-battle-incurredinjuries in the series, 2 werethe result of accidental falls and 1 the result of a vehicular accident.
All degrees of damage were present after wounding, ranging from smallfissures to complete fragmentation of the body of the spleen. In a fewinstances, the organ was penetrated. Severe lacerations, penetrations, andperforations produced essentially the same gross lesions as a splenic fracture;that is, irregular rents in the capsule radiating from the track of thecausative agent. Injury to the splenic pedicle, without injury to the body ofthe organ, occurred 8 times and subcapsular hematoma 3 times.From the standpoint of degree, the injuries were classified as severe in 61.3percent of the cases, asmoderate in 29.3 percent,and as slight in only 9.4 percent.
Active hemorrhage from the injured spleen was only occasionally encounteredwhen the abdomen was opened. In these cases, either the pedicle was injured orthe body of the organ was seriously damaged. Unless, however, the injury wasslight, there was always evidence of previous hemorrhage, and active bleedingusually recurred during the maneuvers incidental to splenectomy. In someinstances of thoracosplenic injury in which very little blood was encountered inthe peritoneal cavity, an extensive left-sided hemothorax was present. Thefinding was so commonly associated with injuries of the spleen, in fact, thattheir presence was suspected whenever a large hemothorax was present inassociation with missile wounds in the lower portion of the left chest. Thisphenomenon was probably the result of negative intrapleural pressure, whichcaused the blood in the upper abdomen to be sucked up through the diaphragmaticrent.
Multivisceral and associated injuries-One hundred of the three hundred and forty-one wounds of the spleen (29.3percent) were univisceral (table 79). Eighty-twoof these occurred in thoracoabdominal injuries. The proportion
293
of univisceral injuries is lower than that reported in other wars. One-thirdof the 49 splenic injuries recorded among American Expeditionary Forces in WorldWar I were univisceral,6 as were 59.2 percent of the 54 cases reported amongBritish troops by Bailey7 in 1942. Jolly,8 while citing no exact figures, statedthat univisceral injuries were "rare" in the Spanish Civil War. Ten ofthe twenty-two injuries of the spleen treated by the 2d Auxiliary Surgical Groupin 1943 were univisceral.
The 241 multivisceral injuries of the spleen in this series were chieflyassociated with wounds of the stomach (100 cases), the colon (92 cases), theleft kidney (84 cases), the small intestine (53 cases), the liver (50 cases),and the pancreas (22 cases). The jejunum was most frequently involved in woundsof the small intestine, the duodenum being injured only three times. The mostcommonly involved portions of the colon were the splenic flexure, the lefttransverse colon, and the upper portion of the descending colon. In a number ofinstances, two separate portions of the colon were injured.
Other structures were injured with much less frequency. The adrenal gland wasinjured in three cases and the gallbladder once. Major vascular channels,including the gastroepiploic artery, the left renal vein, the celiac axis, andthe thoracic aorta, were involved in four instances. A second wounding agent hadmade a separate entry in each of the three cases in which the urinary bladder,the sigmoid colon, and the rectum were also injured.
Severe associated injuries, which were themselves multiple in20 instances, occurred in 128 cases (34.6 percent). The most important were asfollows: 53 soft-tissue injuries with 14 deaths, 36 compound fractures with 10deaths, 13 spinal cord injuries with 6 deaths, and 6 amputations with 1 death.These injuries were often so severe that they contributed materially to thedevelopment of shock prior to surgery and to the development of complicationsafter operation.
CLINICAL CONSIDERATIONS
Shock was an extremely common clinical finding in injuries of the spleen(table 80). Its absence, however, could not be taken to indicate that the spleenwas not involved in the injury. It was mentioned as not present when the patientwas first seen in 78 of 319 recorded cases (24.4 percent), in 48 of which theinjury was multivisceral. Mild shock was present in 13.5 percent of the injuriesand moderate or severe shock in 62.1 percent.
Tenderness and muscle defense in the left upper quadrant of the abdomen werealways present in splenic injury. Pain referred to the left shoulder and to thebase of the neck was observed in a few cases. The frequency of extensiveleft-sided hemothorax has already been commented on.
6See footnote 1, p. 291.
7See footnote 2, p.291.
8Seefootnote 3, p. 291.
294
Degree of shock |
| Multivisceral wounds | Total | ||||||
Cases | Deaths |
| Cases | Deaths | Case fatality rate | Cases | Deaths | Case fatality rate | |
None | 30 | 1 | 3.3 | 48 | 2 | 4.2 | 78 | 3 | 3.8 |
Mild | 18 | 1 | 5.6 | 25 | 5 | 20.0 | 43 | 6 | 14.o |
Moderate | 29 | 3 | 10.3 | 66 | 15 | 22.7 | 95 | 18 | 18.9 |
Severe | 16 | 7 | 43.8 | 87 | 49 | 56.3 | 103 | 56 | 54.4 |
| 93 | 12 | 12.9 | 226 | 71 | 31.4 | 319 | 83 | 26.0 |
1These data are missing in 22cases.
TREATMENT
The "surgical abstention" practiced in World War Iwas never seriously considered as a method of treatment for wounds of the spleenmanaged in World War II by the 2d Auxiliary Surgical Group. It was the generalopinion that splenectomy was the procedure of choice in all cases and that itwas imperative in every injury of more than minimal severity. It was performedin 299 of the 341 wounds in this series (87.7 percent) (table 81), in 9instances with supplemental drainage.
Three of the forty-one patients in whom splenectomy was not performed died onthe operating table, before the operation could be completed. In one severeinjury in which the splenic vein was severed, surgery was limited to ligationof the vein. In 12 cases (table 81) in which the injury was not severe,drainage, pack, or suture was considered sufficient. In the majority of thesecases, though not in all, the lesion consisted merely of a small fissure, withno active hemorrhage and little or no evidence of previous hemorrhage. In theremaining cases, the other visceral injuries or the associated injuries were ofsuch severity that they took precedence over the splenic injury, or it wasthought that the addition of splenectomy to the procedures already accomplishedwould be beyond the limits of the patient's tolerance.
Suture of the spleen, which was carried out in seven cases,proved an inadequate and unsatisfactory measure for the control of activehemorrhage, the prevention of subsequent hemorrhage, and the repair of damage tothe spleen; the risk of secondary hemorrhage, in fact, was substantial.Moreover, this measure was as time consuming as splenectomy. The supplementaryuse of a muscle strip, which was employed in one case, was no more effectivethan simple suture. Packing of the spleen also proved an ineffective method ofmanagement.
Modern hemostatic agents such as absorbable gelatin sponges wereavailable in overseas theaters during World War II only in limited quantitiesand
295
for designated purposes. If the situation had been different, the necessityfor the removal of only slightly damaged spleens might have been prevented.
TABLE 81.-Methods of treatment in 340 injuries of spleen1
Methods of treatment |
| Total | |
Thoracoabdominal |
| ||
Splenectomy: |
|
|
|
| Cases | 224 | 75 | 299 |
| Deaths | 60 | 13 | 73 |
| Case fatality rate | 26.8 | 17.3 | 24.4 |
Other measures: |
|
|
|
| Cases | 29 | 12 | 41 |
| Drainage only | 1 | 2 | 3 |
| Pack | 1 | 1 | 2 |
| Suture | 6 | 1 | 7 |
| No operation on spleen2 | 21 | 8 | 29 |
| Deaths | 7 | 5 | 12 |
| Case fatality rate | 24.1 | 41.7 | 29.3 |
1The record of 1 patient whodied on the operating table does not make clear what had been accomplished whendeath occurred.
2In the 29 cases in thisgroup, the wound of the spleen, for various reasons, was not attacked. Twenty-one of theoperations were performed through a thoracoabdominal incision.
Surgical approach.-The surgical approach toinjuries of the spleen in this series (table 82) differed materially from thetechniques employed in previously reported series. The most striking departurefrom earlier methods was the frequent use of a transdiaphragmatic incision inthoracoabdominal wounds. This approach was used in 171 cases, 67.6 percent ofthe 253 thoracoabdominal injuries and 50.1 percent of the 341 injuries whichmake up the total series of wounds of the spleen. In the other 82thoracoabdominal injuries, the abdominal approach was employed in 75 cases (29.6percent), and both thoracic and abdominal incisions were used in the other 7cases. Since thoracoabdominal wounds furnished almost three-quarters of thewounds of the spleen, it is not surprising to find that the transthoracicapproach was used in so large a proportion of the cases. Anesthesia is animportant consideration in this technique, but the ready availability ofcompetent anesthetists, who were provided with adequate equipment, prevented anydifficulties in that respect.
The case fatality rates for the transdiaphragmatic and abdominal approachesto thoracoabdominal injuries of the spleen, 19.9 percent against 41.3 percent,are so greatly in favor of the transdiaphragmatic approach as to suggest thatall the advantages lay with it. Actually, the marked disparity in the figuresshould not be interpreted as furnishing a true appraisal of the respective ratesby the two techniques. Rather, the lower rate for the transdiaphragmatic
296
TABLE 82.-Surgicalapproaches in 337 wounds of spleen1
Surgical approach |
| Total | |||||||
| Abdominal only | ||||||||
Cases | Deaths |
| Cases | Deaths | Case fatality rate | Cases | Deaths | Case fatality rate | |
Transdiaphragmatic | 171 | 34 | 19.9 | 0 | 0 | 0 | 171 | 34 | 19.9 |
Abdominal | 75 | 31 | 41.3 | 84 | 18 | 21.4 | 159 | 49 | 30.8 |
Transdiaphragmatic and abdominal | 7 | 2 | 28.6 | 0 | 0 | 0 | 7 | 2 | 28.6 |
| 253 | 67 | 26.5 | 84 | 18 | 21.4 | 337 | 85 | 25.2 |
1This information is lackingin 4 cases.
approach is a reflection of the feasibility and ease ofintimate surgical exploration of the left upper abdominal quadrant by thisincision in appropriate cases. In many of the thoracoabdominal injuriesinvolving the spleen handled by the abdominal route, the multiplicity andlocation of the visceral injuries made approach from below mandatory. A numberof cases showed that it was questionable wisdom to repair the diaphragm frombelow under these circumstances, although some surgeons followed this plan, onthe ground that in a very sick patient it might be better to proceed withceliotomy, in the hope of avoiding two major incisions.
In spite of the advantages of transdiaphragmatic incision in the managementof thoracoabdominal injuries involving the spleen, this approach was not usedwhen the thorax was not involved in the wound, for two reasons:
1. It was seldom possible before operation to exclude the presence of otherintraperitoneal injuries which could not be properly explored and repairedthrough the diaphragm.
2. When wounds of the stomach, small intestine, or colon complicated thesplenic injury, there was always a risk of contaminating an undamaged pleuralcavity during the transdiaphragmatic repair of the gastrointestinal wounds.
Although surgeons were in general agreement, as already mentioned, that thetransdiaphragmatic approach was not the best technique for injuries of thespleen confined to the abdomen, there was no agreement as to how this type ofinjury should be approached. The left rectus and left paramedian incisions weremost frequently used. Occasionally the one or the other was extended to theleft, but as a rule splenectomy, as well as other operations in the left upperquadrant and other parts of the abdomen, could be accomplished through theoriginal incision. The left subcostal incision and certain transverse incisionsalso proved satisfactory for splenectomy, though their usefulness
297
was limited by the possibility of other visceral injuriesand the fact that full exploration of the peritoneal cavity was difficult oractually impossible through any of them. An additional serious objection, inview of the many instances in which injury to the large bowel complicated thesplenic injury, was the fact that all of these incisions utilize the space bestadapted for exteriorization of the colon.
POSTOPERATIVE COMPLICATIONS
Although the records of 69 patients, 20.2 percent of the total number, wereincomplete, it was possible to determine indirectly, from various sources, thata large proportion of this group was evacuated without the development ofpostoperative complications. Aside from peritonitis and shock, which werepresent alone or in combination in many cases at operation, major complicationswere recorded in 21.0 percent of the 341 splenic injuries. Anuria, woundinfection, and atelectasis were the most frequent. Portal thrombosis andsecondary hemorrhage from the spleen or splenic pedicle apparently did notoccur. There was also no mention of accidental injury to the pancreas or stomachincurred during the operative procedure. Among miscellaneous complications weremalaria (3 instances); intestinal obstruction, pneumothorax, anaerobicinfection, and psychosis (2 instances each); massive gastric hemorrhage,paroxysmal tachycardia, femoral phlebitis, pelvic abscess, jaundice, decubitusulcer, cardiac failure, spinal meningitis, pulmonary edema, transfusionreaction, and breakdown of the diaphragmatic repair (1 instance each).
Anuria occurred only in patients who had been in severe shock before orduring operation or at both times, or who had received large amounts of blood.Eight of the nine instances of atelectasis occurred in patients with thoracoabdominal injuries. Two of the three patients with infections of the leftsubphrenic space had complicating injuries of the colon. All three instances ofpulmonary embolism were fatal. The first occurred in a patient with aunivisceral injury of the spleen. The second patient also had a compoundfracture of the femur on one side and a traumatic amputation of the leg on theother. The third had severe complicating injuries of the stomach and liver.
CASE FATALITY RATES
The case fatality rate of 24.9 percent (85 cases) in these 341 injuries ofthe spleen is considerably lower than the 66.7 percent reported by Giblin9in 1943 from El Alamein, the 50 percentreported by Ogilvie10 inhis combined series from the Western Desert in 1944, and the 67 percent reportedin American troops in World War I.11
It is also lower than the 33.3-percent rate reported by the surgeons of the2d Auxiliary Surgical Group in 1943. Except for the latter series and Ogilvie's
9See footnote 4, p.291.
10See footnote 5, p. 291.
11See footnote 1 (1), p. 291.
298
series, it is not clear whether the figures cited do or do not includepostevacuation studies.
Bailey12 estimated that the case fatality rate inthe univisceral wounds of the spleen in his series was 40 percent. In World WarI, the British rate for univisceral injuries was 50 percent and formultivisceral injuries 60 percent.13 In the 2d Auxiliary Surgical Groupseries, these respective rates were 12.0 and 30.3 percent. The rate roseprogressively and rapidly as the number of complicating visceral injuriesincreased, reaching 61.5 percent when three or more organs were involved inaddition to the spleen. The case fatality rate of 24.4 percent followingsplenectomy in this series is to be compared with the "practically 100percent" reported for this procedure in World War I.14
Death occurred on the operating table in 10 of 67 patientssubmitted to splenectomy and occurred in the first 24 hours after operation in23 others (table 83). Three other patients died on the operating table beforesplenectomy could be performed.
TABLE 83.-Primary cause of death and time ofdeath in 67 fatalities following splenectomy
Cause |
| To 12 hours | 12 to 24 hours | 1st day | 2d through 3d days | 4th through 7th days | After 7th day | Total |
Shock | 7 | 13 | 8 | 6 | 4 | 0 | 0 | 38 |
Anuria | 0 | 0 | 0 | 0 | 6 | 5 | 1 | 12 |
Peritonitis | 0 | 0 | 1 | 0 | 1 | 2 | 1 | 5 |
Pulmonary embolism | 0 | 0 | 0 | 0 | 2 | 1 | 0 | 3 |
Pneumonia | 0 | 0 | 0 | 0 | 0 | 1 | 0 | 1 |
Atelectasis | 0 | 0 | 0 | 1 | 0 | 0 | 0 | 1 |
Other | 3 | 1 | 0 | 0 | 1 | 2 | 0 | 7 |
| 10 | 14 | 9 | 7 | 14 | 11 | 2 | 67 |
In comparing the case fatality rate (table 81) forsplenectomized patients (24.4 percent) with the rate for nonsplenectomizedpatients (29.3 percent), it must be borne in mind that in most of the lattergroup operation was omitted because the splenic wound was not consideredserious. In nine cases, however, either severe multivisceral or severeassociated injuries, or both, contraindicated additional surgery, or thecomplications which caused death could not be attributed solely to wounds of thespleen.
Shock was the primary cause of death in 38 splenectomized patients (table 83)and in 6 others in whom splenectomy was not performed. Peritonitis,
12See footnote 2, p. 291.
13History of the Great WarBased on Official Documents. Medical Services Surgery of the War. London:His Majesty's Stationery Office, 1922, vol. I.
14See footnote 1 (1), p. 291.
299
pneumonia, and atelectasis were responsible for one death each in thenonsplenectomized patients. In the remaining cases in this group, the cause ofdeath was either undetermined or unrecorded.
Additional complications in the fatal cases included the vagovagal reflexafter bronchoscopy, transfusion reaction, spinal meningitis, massive empyema,bilateral hemothrax, gastric hemorrhage, and disruption of the diaphragmaticrepair complicated by atelectasis and by herniation of the stomach and coloninto the chest.
