- Email: [email protected]
Risk of splenic salvage after trauma
Risk of Splenic Salvage After Trauma Analysis of 200 Adults
Frederick A. Moore, MD, Denver, Colorado Ernest E. Moore, MD, Denver, Colorado George E. Moore, MD, Denver, Colorado J. Scott Millikan, MD, Denver, Colorado
Management of splenic injury has changed radically over the past decade. Once regarded as mysterii pleni orgunon, the spleen is now considered invaluable as an immunologic factory as well as reticuloendothelial filter. Although risk of postsplenectomy sepsis is greatest in the child less than 2 years of age, the asplenic adult is clearly vulnerable [I ,2]. Danger of overwhelming infection has prompted surgeons to develop alternatives to splenic extirpation after trauma. The following is a critical review of our experience with splenorrhaphy and splenic reimplantation since their inception at our institution in 1978. Material and Methods During the 6 year period ending December 1984, 200 adults with splenic injury underwent laparotomy at the Denver General Hospital. Their ages ranged from 12 to 92 years (median 29.5 years); 166 (83 percent) were men. Of the 138 patients (69 percent) who sustained blunt trauma, the mechanism was motor vehicle accident in 100, assault in 25, fall in 11, and industrial mishaps in 2. Penetrating wounds accounted for the remaining 62 injuries (31 percent). Thirty-two were stab wounds and 30 were gunshot wounds. Iatrogenic splenic injuries were not included. Management of abdominal trauma was standard throughout the study period and has been described in detail in a previous report [3]. The first maneuver for splenic injury was mobilization. The operator retracted the spleen anteromedially to facilitate division of the lienorenal ligament. With blunt dissection posterior to the pancreas, the spleen was gradually rotated into the abdominal wound. Persistent bleeding due to capsular avulsions was controlled with topical microfibrillar collagen applied to the bare parenchyma. Exposed vessels were ligated with From the Deprulment of Sugery. Denver General Hospital and the University of Cokrado Health Sciences Center, Denver, Colorado. Requests for reprlnts should be addressed to Ernest E. Moore, MD, DepMment of bgery. Denver General Hospital, 777 Sanncck Street,, Denver. Colorado 80204-4507. Presented at the 36th Annual Meeting of the Southwestern Surgical Congress, Honolulu, Hawaii. April 21-28, 1984.
800
4-O silk transfixing sutures. For extensive parenchymal defecta, interlocking 2-O polyglycolic mattress sutures were used, supported with TeflonQ or collagen bolsters. Anatomic resection was performed for major splenic fractures. The gastrolienal ligament was divided to expose the splenic hilum. The segmental artery corresponding to the area of injury was ligated. After demarcation, the appropriate segmental veins were divided. The ischemic portion of spleen was then amputated with a knife and the cut edge sutured with pledgeted sutures, Occasionally, the main splenic artery was temporarily occluded with a Silastic@ loop during repair of extensive injuries. When splenectomy was required, the spleen was kept in Ringer’s lactate solution until the remaining intraabdominal procedures were completed. The spleen was then sectioned into five fragments (40 by 40 by 3 mm), which were enclosed in a greater omental pouch [3]. We did not drain the left upper quadrant after splenorrhaphy or splenectomy unless indicated for associated pancreatic injury. Polyvalent pneumococcal vaccine (Pneumovax@) was administered postoperatively to all patients requirii splenectomy or loss of more than half of their spleen. Operative management is summarized in Table I. Eighty-five patients (42 percent) underwent splenorrhaphy, 43 (22 percent) splenectomy with reimplantation, and 50 (25 percent) total splenectomy. The remaining 22 patients (11 percent) died in the operating room. It is noteworthy that the incidence of splenic repair has plateaued at 56 percent over the past 3 years. At the present time, virtually all patients who require splenectomy undergo autotransplantation. Splenorrhaphy: The 85 patients undergoing splenic repair ranged in age from 12 to 61 years (median 29 years). Blunt injuries occurred in 59 of these patients (69 percent); 45 were from motor vehicle accidents, 9 from assaults, and 5 from falls (Table II). The remaining 26 patients (31 percent) sustained penetrating wounds (21 stab injuries and 6 gunshot wounds). Splenic salvage was achieved in 68 percent of the stab injuries, 43 percent of the cases of blunt trauma, and 17 percent of the gunshot wounds. Extraabdominal injuries included chest injury in 32 patients, head injury in 13 patients, pelvic injury in 13 patients, and extremity fractures in 6 patients. Associated intraabdominal
X% Amerkan
Journal ol Surgery
Splenic
TABLE I
Salvage
After Trauma
Operative Management of Splenic Trauma in 200 Adults Splenic Injury (n)
Splenic Repair
Splenic Implant %
n
%
Total Splenectomy n %
n
%
1978 1979 1980 1981 1982 1983
33 31 38 32 39 27
9 7 14 18 22 15
27 23 37 58 56 56
4 4 8 7 11 9
12 13 21 22 28 33
18 17 9 4 1 1
55 54 24 12 3 4
2 3 7 3 5 2
8 10 18 10 13 7
Total
200
85
42
43
22
50
25
22
11
n
Died in OR
DFt = operating room.
TABLE II
Me&a&m
of Injury and Management In 200 Adults Who Sustalned Splenlc Trauma Blunt Automobile n %
Assault n
n
%
Stab wound n 96
Gunshot wound n %
Total n
%
Hemostatic
11
11
2
8
1
8
8
25
2
7
24
12
agents Suture
22
22
4
16
3
23
10
31
3
10
42
21
0 9 3
0 9 3
0 3 0
12 0 0
01 0
0 8 0
1 1
3 3
0 0
0 0
141 4
0.57 2
45
45
9
36
5
41
21
66
5
17
85
42
18
18
8
32
3
23
5
16
9
30
43
22
23
23
7
28
2
15
4
13
14
46
50
25
14
14
1
4
3
23
2
6
2
7
22
11
repair momy Trkegmentectomy Total salvage spienic implant Total spl~omy Died in OFI TOW
100
25
13
32
30
200
OR = operating room.
injuries were present in 48 patient8 (56 percent). The liver, pancreas, and diaphragm predominated (Table III). Of particular interest was that 25 percent of the splenorrhaphy patients had peritoneal contamination from a hollow visceral injury. Splenorrhaphy was accomplished by hemostatic agents in 24 patients (28 percent), debridement and suturing in 42 (50 percent), and partial resection in 19 (22 percent) (Table II). Resections consisted of 14 lobectomies, 4 trisegmentectomies, and 1 segmentectomy. Average intraoperative blood loss was 1,062 ml when repair was performed with hemostatic agents, 1,650 ml when debridement and suturing were employed, and 1,760 ml with partial splenic resection. Splenic reimplantation: The 43 patienta undergoing splenic autotransplantation ranged in age from 16 to 72 years (median 32 years). Injury was due to motor vehicle accident in 18 patients (42 percent), assault in 8 (18 percent), fall in 3 (7 percent), gunshot wound in 9 (21 percent), and stab injury in 5 (12 percent). Extraabdominal injuries included head injury in 6 patients, chest injury in 15 patients, pelvic fracture in 2 patients, spinal cord transection in 1 patient, and extremity fractures in 3 patients. Associated intraabdominal injuries were present in 27 patients
Vol~148,
Dewmbor1984
(63 percent). The pancreas, diaphragm, and kidney were the most frequent sites. The average blood loss during operation was 4,675 ml.
Results In the splenorrhaphy group, six patients (7 percent) died, four from closed head injuries and two from sepsis-related multiple organ failure. Multiple organ failure was due to missed duodenal injury in one patient and recurrent intraabdominal abscesses after multiple gunshot wounds in the other patient. No patient died from delayed overwhelming sepsis. Fifteen complications developed in 14 patients (16 percent). pulmonary morbidity included pneumonia in five patients, major at&ctasis in two patients, and thromboembolism in one patient. One patient required delayed nephrectomy for recurrent bleeding and another had a urinary tract infection. A left subphrenic abscess occurred in a 24 year old man requiring jejunal resection after blunt trauma. Rebleeding developed postoperatively in three patients
801
Moore et al
TABLE III
Associated lntraabdomlnal lnjurles In 200 Adults Who Sustalned Splenlc Trauma Splenic Implant
Splenic Repair n
%
n
%
Liver Pancreas Diaphragm Kidney Colon Stomach Small bowel Major vascular Minor vascular Adrenal Duodenal Biliary Bladder None
20 9 14 10 10 5 5 2 3 0 1 2 2 37
24 11 16 12 12 6 6 2 4 0 1 2 2 44
8 13 10 10 4 7 2 1 3 1 0 0 0 16
18 30 23 23 9 16 5 2 7 2 0 0 0 37
Total patients
85
43
Total Splenectomy n % 18 13 5 9 5 10 5 2 2 5 2 1 0 18 50
32 28 10 18 10 20 10 4 4 10 4 2 0 36
Died in OR
Total
n
%
n
%
10 3 5 3 4 1 1 6 2 3 0 0 0 1
45 14 23 14 18 5 5 27 9 14 0 0 0 4
54 38 33 32 23 23 13 11 10 9 3 3 2 72
27 19 18 16 12 12 6 6 5 4 2 2 1 36
22
200
OR = operating room.
(4 percent). The first patient has been described previously [4]. This 25 year old man had an inferior pole fracture that required running 3-O polyglycolic acid sutures and topical microfibrillar collagen for hemostasis. Five weeks later, shortly after his primary physician instituted aspirin therapy, laparotomy was performed for an acute perisplenic hematoma. The second patient, a 28 year old man assaulted with a lead pipe, had two transverse fractures of the superior spleen and a longitudinal laceration of the inferior pole extending into the hilum. The transverse fractures were sutured and the inferior pole resected. Despite pledget suturing and microfibrillar collagen application, the resected edge continued to bleed. Omentum was wrapped around this area, which finally resulted in hemostasis. Recurrent bleeding ensued on the third postoperative day, necessitating splenectomy. The third patient, a 32 year old woman, had been kicked multiple times. The resulting avulsion injury to the lower pole of the spleen was managed by ligation of the segmental artery, debridement, and suturing. Pledgets, however, were not placed through intact splenic capsule. Rebleeding occurred 18 hours after splenorrhaphy. In the splenic reimplantation group, five patients (12 percent) died, one from closed head injury, three from intraabdominal sepsis and multiple organ failure, and one from delayed overwhelming pneumococcal pneumonia. The latter was a 61 year old cirrhotic patient who recovered uneventfully after blunt abdominal trauma, but returned 5 months later with fulminant pneumococcal sepsis. Eleven complications developed in 10 patients (23 percent). Infections included pneumonia in three, intraabdominal abscess in three, and an infected pancreatic pseudocyst and urinary tract infection in
802
one patient each. Other abdominal problems included a pancreatic fistula, a small bowel obstruction from diffuse adhesions, and a reoperation for hepatic bleeding. No complications could be attributed to the splenic implants per se. Comments Although the first partial splenectomy for trauma was reported in 1590 [5] it took nearly 4 centuries before the concept of operative splenic preservation was accepted. Paralleling elucidation of the spleen’s immunologic significance in the 19509, the anatomic work of Michels [6] and Huu et al [7] delineated the segmental end arterial system of the spleen. Moreover, it became clear that blunt trauma fractures along the avascular intersegmental planes provided an anatomic basis for repair [8]. Development of topical hemostatic agents [9] further aided controlled segmental resection. First reported by Camp0 Christo [IO] in 1962, the principles of anatomic dissection of the human spleen have been promulgated by a number of international groups [7,11]. Favorable results of splenic salvage in children led to enthusiasm for splenorrhaphy in adults [12-141. This early experience, relying predominately on hemostatic agents and including iatrogenic injury, attested to the feasibility of adult splenorrhaphy. As a result, salvage efforts intensified [15,16]. Our aggressive attitude has culminated in splenic repair in 56 percent of our acutely injured patients for each of the past 3 years. Technical complications will no doubt occur as more complex injuries are approached. Three bleeding episodes in this series confirm this fact and emphasize a learning curve, as occurs with other technically demanding operations. To our knowledge, only one other case of rebleed-
The American Journal et &gory
Splenic Salvage
ing after splenorrhaphy has been reported [I 71. Each of our three patients had an isolated but complex splenic injury. The first patient was given an anticoagulant. Although not always an avoidable event, this emphasizes the importance of patient education and careful follow-up. Our experience with the second patient underscores the need for complete hemostasis after splenorrhaphy. When mobilization has not disrupted collateral flow, splenic artery ligation is an alternative [18]. If not feasible, splenectomy should be performed. Inclusion of the disrupted capsule within the pledgetted suture line of a major nonsegmental debridement resulted in our third failure. A formal segmental resection was probably indicated. These three bleeding episodes were well tolerated and easily identified due to the isolated nature of the injury in otherwise healthy persons. The presence of associated hollow visceral injury has not dissuaded us from splenic repair. Although it has been suggested that contamination contraindicates splenorrhaphy, it is these patients who may derive the most benefit from a functioning spleen [11,19]. Of the 22 splenorrhaphy patients with such injuries, infectious complication in the left upper quadrant developed in only one. We minimize foreign material used for splenorrhaphy. Absorbable sutures and a limited amount of hemostatic agents are advocated [20]. Drains are not indicated for splenic repair alone [ 161. Splenectomy is clearly the treatment of choice when the spleen is pulverized or other life-threatening injuries demand immediate attention. In this setting, we currently autotransplant a portion of the spleen and administer polyvalent pneumococcal vaccine [21]. Although numerous animal studies have confirmed immunologic benefits of splenic reimplantation, the extent is variable [22-241. Human data are limited. Our follow-up studies demonstrated uniform implant viability by technetium scanning at 6 months, and showed normalization of the immunoglobulin M levels and platelet numbers [3]. Pate1 et al [.%I have confirmed these findings and, in addition, found disappearance of target cells and Howell-Jolly bodies and return of normal complement levels at 4 weeks. More sophisticated immunologic studies need to be completed. Our experience thus far, however, has demonstrated that splenic autotransplantation is simple and safe. No postoperative complications have occurred with our omental pouch technique. Summary This review was undertaken to analyze critically the complications resulting from operative splenic salvage. Over a 6 year period, 200 adults who sustained splenic trauma underwent laparotomy. The mechanism of injury was blunt in 138 patients (69 percent), a stab wound in 32 patients (16 percent),
Volume 140, December 1984
After Trauma
and a gunshot wound in 30 patients (15 percent). Splenorrhaphy was accomplished in 85 patients (42 percent). Methods of repair included cautery and hemostatic agents in 24 patients (28 percent), debridement and suturing in 42 patients (50 percent), and patial resection in 19 patients (22 percent). Six patients died, four from head trauma and two from multiple organ failure. Postoperative complications occurred in 14 patients. Four were intraabdominal. Three patients required reoperation for splenic hemorrhage; one (2 percent) after suture repair and two (11 percent) after partial resection. A left subphrenic abscess developed in another patient. Splenic reimplantation was performed in 43 patients (22 percent). Five deaths occurred. One was due to head trauma, three to multiple organ failure, and one to overwhelming pneumococcal infection. Eleven postoperative complications occurred, but none was related to splenic autotransplantation. Despite the enthusiasm for splenic salvage, the number of patients suitable for splenorrhaphy plateaued at 56 percent. Complications of splenorrhaphy are infrequent, and the risk increases with more complex salvage attempts. We believe that splenic reimplantation remains a safe procedure. References 1. Dickerman JD. Traumatic asplenia in adults. A defined hazard? Arch Surg 1981;i 18381-3. 2. O’Neal BJ, McDonald X. The risk of sepsis in the asplenic adult. Ann Surg 1981;194:775-8. 3. Mlllikan JS, Moore EE, Moore GE, Stevens RE. Alternatives to splenectomy in adults after trauma: repair, partial resectfon, and reimplantation of splenic tissue. Am J Surg 1982144: 711-8. 4. Van Stiegmann G, Moore EE, Moore GE. Failure of spleen ra pair. J Trauma 1979;19:698-700. 5. Sherman Ft. Perspectives in management of trauma to the spleen: 1979 Presidential address, American Association for the Surgery of Trauma. J Trauma 1981;20:1-13. 6. Michels NA. Blood supply and anatomy of the upper abdominal organs. Philadelphia: JB Lippincott, 1955201. 7. Huu N, Person H, Hong R, Vallee B, Hoan Vu N. Anatomical approach to the vascular segmentation of the spleen (lien) based on controlled experimental partial splenectomles. Anat Clin 1982;4:265-77. 8. Upadhyaya P, Simpson JS. Splenic trauma in children. Surg Gyncecol Obstet 1968;127:781-90. 9. Silverstein ME, Chvapil M. Experimental and clinical experiences with collagen fleece as a hemostatic agent. J Trauma 1981;21:388-93. 10. Campo Chisto M. Segmental resection of the spleen. 0 Hospital (Rio) 1962;62:187-204. 11. Morgenstern L, Shapiro SJ. Techniques for splenic conservation. Arch Surg 1979;114:449-54. 12. Burrington JD. Surgical repair of a ruptured spleen in chiklren. Arch Surg 1977;112:417-9. 13. Weinstein ME, Govin GG. Rice CL, Virgil0 RW. Splenorrhaphy for splenic trauma. J Trauma 1979;19:692-7. 14. Giuliano AE, Lim RC. Is splenic salvage safe in the traumatized patient? Arch Surg 1981;116:651-6. 15. Barrett J, Sheaff C. Abuabara S, Jonasson 0. Splenic preservation In adufts after blunt and penetrating trauma. Am J Sug 1983;145:313-7. 16. Pachter HL, Hofstetter SR, Spencer FC. Evolving concepts in
803
Moore et al
17.
18.
19.
20. 21.
22.
23.
24.
25.
splenic surgery, splenorrhaphy versus splenectomy and post-splenectomy drainage: experience In 105 patients. Ann Surg 1981;194:262-9. Oakes DD, Charters AC. Changing concepts In the management of splenlc trauma. Surg Gynecoi Obstet 1981; 153: 181-5. Horton J, Ogden ME, Williams S, Coin D. The importance of splenlc blood flow in clearing pneumococcal organisms. Ann Surg 1982;195:172-6. Rogers DM, Herrington JL, Morton C. Incidental splenectomy associated with Nissen fundoplication. Ann Surg 1980; 191:153-6. Scher KS, Coil JA. Effects of oxidized cellulose and microfibriiiar collagen on infection. Surgery 1982;91:301-4. Capian ES, Boltansky H, Snyder MJ, et al. Response of traumatized splenectomized patients to immediate vaccination with polyvalent pneumococcal vaccine. J Trauma 1983; 23:801-5. Dickerman JD. Horner SR, Coil JA, Gump DW. The protective effect of intraperitoneal splenic autotransplants in mice exposed to aerosolized suspension of Type iii Streptococcus pneumoniae. Blood 1979;54:354-8. Faschlng MD, Cooney DR. Reimmunlzatlon and spienic autotransplantation: a long-term study of Immunologic response and survival following pneumococcal challenge. J Sug Res 1980;28:449-59. Livingston CD, Levine BA, Sirinek KR. Improved survival rate for lntraperitoneal autotransplantatlon of the spleen foilowlng pneumococcal pneumonia. Surg Gynecoi Obstet 1983; 156:761-6. Pate1 J, Williams JS, Shmigel B, Hinshaw RJ. Preservation of splenic function by autotransplantation of traumatized spleen in man. Surgery 1981;90:683-6.
Discussion David V. Feliciano (Houston, TX): Although all of us recognize the low incidence of fuhninant postsplenectomy sepsis (0.18 cases per hundred person years in the Mayo Clinic series of 1982), most of us try to repair spleens these days rather than perform splenectomy. Splenorrhaphy can be safely performed when good surgical judgment is used. Dr. Moore’s series today documents this, as does our own series from the Ben Taub General Hospital where 135 successful splenorrhaphies, 42 percent of all splenic injuries, have been performed since 1980. It also appears that splenic reimplantation into the leaves of the omentum can be safely performed as well, according to the results just reported, and these fragments will clear red blood cells and platelets, be visible on technetium scans, and produce immtmoglobulin M. However, there are basically two major questions to be considered when discussing the efficacy of extensive splenorrhaphy or reimplantation. First, is the remaining mass of splenic tissue large enough in size to clear an infusion of capsular organisms? Dickerman’s study in 1979 and Van Wyck’s study in 1980 both documented the need for a certain mass of splenic tissue to confer protection against capsular organism infusion in laboratory animals. Second, is the blood flow to the remaining fourth or half of the spleen or a group of splenic fragments sufficient to even get the capsular organisms to the splenic tissue? Coin’s study in 1982, Livington’s study in 1983, and Pabst’s study published in the March 1984 issue of The American Journal of Surgery all emphasized the critical factor of minimal blood flow to the splenic remnants or fragments.
804
Some of the following questions will reflect on these critical points Dr. Moore, why did you not reimplant fragments in 50 patients in this series who had splenectomy? Please comment on the weight of the fragments reimplanted, and the estimated weight of remaining splenic tissue after heroic operations, such as trisegmentectomy, compared with the weight of a normal spleen. How many of the 38 surviving splenic reimplantation patients actually had technetium scans, immunoglobulin M level determinations, platelet counts, and so on and were the results uniformly successful in all 38 patients? Were capsular organisms the cause of death in the splenorrhaphy patient who died from multiple organ failure and the three patients who died from intraabdominal sepsis and multiple organ failure in the reimplantation group? Did you reimplant splenic fragments in the three patients who returned to the operating room after failed splenorrhaphies and, if not, why not? If you are going to use Teflon pledgets to help repair extensive parenchymal defects, how can you say that you must minimize the use of foreign material when splenorrhaphy is performed in the face of gastrointestinal contamination. Finally, I would like to ask about your choice of suture material. I have always used chromic suture or polypropylene suture because either will slide through the spleen and liver a lot easier than some of the newer absorbable sutures. John Moore (Denver, CO): I recently cared for a young man with an isolated splenic injury who underwent splenectomy with splenic implantation. Fever, left quadrant pain, and leukocytosis developed in this patient 9 days after splenic implantation requiring reexploration. At surgery, a fluid collection in the left subphrenic space was found that was consistent with a subphrenic abscess, but more importantly, the entire greater omentum appeared inflamed and was removed. Pathologically, the splenic implants showed liquification necrosis and there was diffuse acute omentitis. Dr. Moore, what has been your experience with this apparent inflammatory response and its clinical significance in the postoperative course? Ernest E. Moore (closing): The incidence of total splenectomy decreased from 55 percent at the inception of this study to 4 percent at the present time. Animal studies suggest that a third of the spleen must be reimplanted for immunologic benefit to occur. We have increased the fragment sizes from 20 by 20 by 3 mm to 40 by 40 by 3 mm to approximate this critical mass. Shaftan and his colleagues have shown in dogs that splenic implants continue to grow over a 2 year period. A standard splenic trisegmentectomy would leave 25 percent of the spleen with its native arterial supply. We are in the process of restudying our patients with splenic implants. In the 16 recalled thus far, all have had technetium uptake by the implants, and normal immunoglobulin M levels as well as platelet counts at 6 months. Pate1 and associates have followed more than 40 patients and have demonstrated normalization of third component of complement levels as well as biopsy-proved regeneration of splenic tissue. The 61 year old cirrhotic patient with multiple organ failure died from overwhelming pneumococcal sepsis 5
The American Journal of Surgery
Splenic Salvage After Trauma
months postoperatively. The three patients with intraperitoneal sepsis had the usual mixed bacterial flora in their abscesses. We reimplanted spleen in both patients with early splenorrhaphy failure, but in the third patient who returned 5 weeks postoperatively this was not feasible. We emphasize that the amount of foreign tissue employed for splenorrhaphy should be minimized. In adults, unlike children, we believe pledgets are warranted for
Voluma 148, Docomhu 1884
major splenic injuries. The issue of polyglycolic acid versus chromic suture is moot. Dr. Moore has raised an important point. Both Tavasoli and Pate1 have shown that splenic autotransplants undergo severe coagulation necrosis during the first 2 postoperative weeks. Early reoperation in these patients may lead to unnecessary removal of the splenic fragments because of this. It may take as long as 6 months for the normal splenic architecture to be reestablished.
Frederick A. Moore, MD, Denver, Colorado Ernest E. Moore, MD, Denver, Colorado George E. Moore, MD, Denver, Colorado J. Scott Millikan, MD, Denver, Colorado
Management of splenic injury has changed radically over the past decade. Once regarded as mysterii pleni orgunon, the spleen is now considered invaluable as an immunologic factory as well as reticuloendothelial filter. Although risk of postsplenectomy sepsis is greatest in the child less than 2 years of age, the asplenic adult is clearly vulnerable [I ,2]. Danger of overwhelming infection has prompted surgeons to develop alternatives to splenic extirpation after trauma. The following is a critical review of our experience with splenorrhaphy and splenic reimplantation since their inception at our institution in 1978. Material and Methods During the 6 year period ending December 1984, 200 adults with splenic injury underwent laparotomy at the Denver General Hospital. Their ages ranged from 12 to 92 years (median 29.5 years); 166 (83 percent) were men. Of the 138 patients (69 percent) who sustained blunt trauma, the mechanism was motor vehicle accident in 100, assault in 25, fall in 11, and industrial mishaps in 2. Penetrating wounds accounted for the remaining 62 injuries (31 percent). Thirty-two were stab wounds and 30 were gunshot wounds. Iatrogenic splenic injuries were not included. Management of abdominal trauma was standard throughout the study period and has been described in detail in a previous report [3]. The first maneuver for splenic injury was mobilization. The operator retracted the spleen anteromedially to facilitate division of the lienorenal ligament. With blunt dissection posterior to the pancreas, the spleen was gradually rotated into the abdominal wound. Persistent bleeding due to capsular avulsions was controlled with topical microfibrillar collagen applied to the bare parenchyma. Exposed vessels were ligated with From the Deprulment of Sugery. Denver General Hospital and the University of Cokrado Health Sciences Center, Denver, Colorado. Requests for reprlnts should be addressed to Ernest E. Moore, MD, DepMment of bgery. Denver General Hospital, 777 Sanncck Street,, Denver. Colorado 80204-4507. Presented at the 36th Annual Meeting of the Southwestern Surgical Congress, Honolulu, Hawaii. April 21-28, 1984.
800
4-O silk transfixing sutures. For extensive parenchymal defecta, interlocking 2-O polyglycolic mattress sutures were used, supported with TeflonQ or collagen bolsters. Anatomic resection was performed for major splenic fractures. The gastrolienal ligament was divided to expose the splenic hilum. The segmental artery corresponding to the area of injury was ligated. After demarcation, the appropriate segmental veins were divided. The ischemic portion of spleen was then amputated with a knife and the cut edge sutured with pledgeted sutures, Occasionally, the main splenic artery was temporarily occluded with a Silastic@ loop during repair of extensive injuries. When splenectomy was required, the spleen was kept in Ringer’s lactate solution until the remaining intraabdominal procedures were completed. The spleen was then sectioned into five fragments (40 by 40 by 3 mm), which were enclosed in a greater omental pouch [3]. We did not drain the left upper quadrant after splenorrhaphy or splenectomy unless indicated for associated pancreatic injury. Polyvalent pneumococcal vaccine (Pneumovax@) was administered postoperatively to all patients requirii splenectomy or loss of more than half of their spleen. Operative management is summarized in Table I. Eighty-five patients (42 percent) underwent splenorrhaphy, 43 (22 percent) splenectomy with reimplantation, and 50 (25 percent) total splenectomy. The remaining 22 patients (11 percent) died in the operating room. It is noteworthy that the incidence of splenic repair has plateaued at 56 percent over the past 3 years. At the present time, virtually all patients who require splenectomy undergo autotransplantation. Splenorrhaphy: The 85 patients undergoing splenic repair ranged in age from 12 to 61 years (median 29 years). Blunt injuries occurred in 59 of these patients (69 percent); 45 were from motor vehicle accidents, 9 from assaults, and 5 from falls (Table II). The remaining 26 patients (31 percent) sustained penetrating wounds (21 stab injuries and 6 gunshot wounds). Splenic salvage was achieved in 68 percent of the stab injuries, 43 percent of the cases of blunt trauma, and 17 percent of the gunshot wounds. Extraabdominal injuries included chest injury in 32 patients, head injury in 13 patients, pelvic injury in 13 patients, and extremity fractures in 6 patients. Associated intraabdominal
X% Amerkan
Journal ol Surgery
Splenic
TABLE I
Salvage
After Trauma
Operative Management of Splenic Trauma in 200 Adults Splenic Injury (n)
Splenic Repair
Splenic Implant %
n
%
Total Splenectomy n %
n
%
1978 1979 1980 1981 1982 1983
33 31 38 32 39 27
9 7 14 18 22 15
27 23 37 58 56 56
4 4 8 7 11 9
12 13 21 22 28 33
18 17 9 4 1 1
55 54 24 12 3 4
2 3 7 3 5 2
8 10 18 10 13 7
Total
200
85
42
43
22
50
25
22
11
n
Died in OR
DFt = operating room.
TABLE II
Me&a&m
of Injury and Management In 200 Adults Who Sustalned Splenlc Trauma Blunt Automobile n %
Assault n
n
%
Stab wound n 96
Gunshot wound n %
Total n
%
Hemostatic
11
11
2
8
1
8
8
25
2
7
24
12
agents Suture
22
22
4
16
3
23
10
31
3
10
42
21
0 9 3
0 9 3
0 3 0
12 0 0
01 0
0 8 0
1 1
3 3
0 0
0 0
141 4
0.57 2
45
45
9
36
5
41
21
66
5
17
85
42
18
18
8
32
3
23
5
16
9
30
43
22
23
23
7
28
2
15
4
13
14
46
50
25
14
14
1
4
3
23
2
6
2
7
22
11
repair momy Trkegmentectomy Total salvage spienic implant Total spl~omy Died in OFI TOW
100
25
13
32
30
200
OR = operating room.
injuries were present in 48 patient8 (56 percent). The liver, pancreas, and diaphragm predominated (Table III). Of particular interest was that 25 percent of the splenorrhaphy patients had peritoneal contamination from a hollow visceral injury. Splenorrhaphy was accomplished by hemostatic agents in 24 patients (28 percent), debridement and suturing in 42 (50 percent), and partial resection in 19 (22 percent) (Table II). Resections consisted of 14 lobectomies, 4 trisegmentectomies, and 1 segmentectomy. Average intraoperative blood loss was 1,062 ml when repair was performed with hemostatic agents, 1,650 ml when debridement and suturing were employed, and 1,760 ml with partial splenic resection. Splenic reimplantation: The 43 patienta undergoing splenic autotransplantation ranged in age from 16 to 72 years (median 32 years). Injury was due to motor vehicle accident in 18 patients (42 percent), assault in 8 (18 percent), fall in 3 (7 percent), gunshot wound in 9 (21 percent), and stab injury in 5 (12 percent). Extraabdominal injuries included head injury in 6 patients, chest injury in 15 patients, pelvic fracture in 2 patients, spinal cord transection in 1 patient, and extremity fractures in 3 patients. Associated intraabdominal injuries were present in 27 patients
Vol~148,
Dewmbor1984
(63 percent). The pancreas, diaphragm, and kidney were the most frequent sites. The average blood loss during operation was 4,675 ml.
Results In the splenorrhaphy group, six patients (7 percent) died, four from closed head injuries and two from sepsis-related multiple organ failure. Multiple organ failure was due to missed duodenal injury in one patient and recurrent intraabdominal abscesses after multiple gunshot wounds in the other patient. No patient died from delayed overwhelming sepsis. Fifteen complications developed in 14 patients (16 percent). pulmonary morbidity included pneumonia in five patients, major at&ctasis in two patients, and thromboembolism in one patient. One patient required delayed nephrectomy for recurrent bleeding and another had a urinary tract infection. A left subphrenic abscess occurred in a 24 year old man requiring jejunal resection after blunt trauma. Rebleeding developed postoperatively in three patients
801
Moore et al
TABLE III
Associated lntraabdomlnal lnjurles In 200 Adults Who Sustalned Splenlc Trauma Splenic Implant
Splenic Repair n
%
n
%
Liver Pancreas Diaphragm Kidney Colon Stomach Small bowel Major vascular Minor vascular Adrenal Duodenal Biliary Bladder None
20 9 14 10 10 5 5 2 3 0 1 2 2 37
24 11 16 12 12 6 6 2 4 0 1 2 2 44
8 13 10 10 4 7 2 1 3 1 0 0 0 16
18 30 23 23 9 16 5 2 7 2 0 0 0 37
Total patients
85
43
Total Splenectomy n % 18 13 5 9 5 10 5 2 2 5 2 1 0 18 50
32 28 10 18 10 20 10 4 4 10 4 2 0 36
Died in OR
Total
n
%
n
%
10 3 5 3 4 1 1 6 2 3 0 0 0 1
45 14 23 14 18 5 5 27 9 14 0 0 0 4
54 38 33 32 23 23 13 11 10 9 3 3 2 72
27 19 18 16 12 12 6 6 5 4 2 2 1 36
22
200
OR = operating room.
(4 percent). The first patient has been described previously [4]. This 25 year old man had an inferior pole fracture that required running 3-O polyglycolic acid sutures and topical microfibrillar collagen for hemostasis. Five weeks later, shortly after his primary physician instituted aspirin therapy, laparotomy was performed for an acute perisplenic hematoma. The second patient, a 28 year old man assaulted with a lead pipe, had two transverse fractures of the superior spleen and a longitudinal laceration of the inferior pole extending into the hilum. The transverse fractures were sutured and the inferior pole resected. Despite pledget suturing and microfibrillar collagen application, the resected edge continued to bleed. Omentum was wrapped around this area, which finally resulted in hemostasis. Recurrent bleeding ensued on the third postoperative day, necessitating splenectomy. The third patient, a 32 year old woman, had been kicked multiple times. The resulting avulsion injury to the lower pole of the spleen was managed by ligation of the segmental artery, debridement, and suturing. Pledgets, however, were not placed through intact splenic capsule. Rebleeding occurred 18 hours after splenorrhaphy. In the splenic reimplantation group, five patients (12 percent) died, one from closed head injury, three from intraabdominal sepsis and multiple organ failure, and one from delayed overwhelming pneumococcal pneumonia. The latter was a 61 year old cirrhotic patient who recovered uneventfully after blunt abdominal trauma, but returned 5 months later with fulminant pneumococcal sepsis. Eleven complications developed in 10 patients (23 percent). Infections included pneumonia in three, intraabdominal abscess in three, and an infected pancreatic pseudocyst and urinary tract infection in
802
one patient each. Other abdominal problems included a pancreatic fistula, a small bowel obstruction from diffuse adhesions, and a reoperation for hepatic bleeding. No complications could be attributed to the splenic implants per se. Comments Although the first partial splenectomy for trauma was reported in 1590 [5] it took nearly 4 centuries before the concept of operative splenic preservation was accepted. Paralleling elucidation of the spleen’s immunologic significance in the 19509, the anatomic work of Michels [6] and Huu et al [7] delineated the segmental end arterial system of the spleen. Moreover, it became clear that blunt trauma fractures along the avascular intersegmental planes provided an anatomic basis for repair [8]. Development of topical hemostatic agents [9] further aided controlled segmental resection. First reported by Camp0 Christo [IO] in 1962, the principles of anatomic dissection of the human spleen have been promulgated by a number of international groups [7,11]. Favorable results of splenic salvage in children led to enthusiasm for splenorrhaphy in adults [12-141. This early experience, relying predominately on hemostatic agents and including iatrogenic injury, attested to the feasibility of adult splenorrhaphy. As a result, salvage efforts intensified [15,16]. Our aggressive attitude has culminated in splenic repair in 56 percent of our acutely injured patients for each of the past 3 years. Technical complications will no doubt occur as more complex injuries are approached. Three bleeding episodes in this series confirm this fact and emphasize a learning curve, as occurs with other technically demanding operations. To our knowledge, only one other case of rebleed-
The American Journal et &gory
Splenic Salvage
ing after splenorrhaphy has been reported [I 71. Each of our three patients had an isolated but complex splenic injury. The first patient was given an anticoagulant. Although not always an avoidable event, this emphasizes the importance of patient education and careful follow-up. Our experience with the second patient underscores the need for complete hemostasis after splenorrhaphy. When mobilization has not disrupted collateral flow, splenic artery ligation is an alternative [18]. If not feasible, splenectomy should be performed. Inclusion of the disrupted capsule within the pledgetted suture line of a major nonsegmental debridement resulted in our third failure. A formal segmental resection was probably indicated. These three bleeding episodes were well tolerated and easily identified due to the isolated nature of the injury in otherwise healthy persons. The presence of associated hollow visceral injury has not dissuaded us from splenic repair. Although it has been suggested that contamination contraindicates splenorrhaphy, it is these patients who may derive the most benefit from a functioning spleen [11,19]. Of the 22 splenorrhaphy patients with such injuries, infectious complication in the left upper quadrant developed in only one. We minimize foreign material used for splenorrhaphy. Absorbable sutures and a limited amount of hemostatic agents are advocated [20]. Drains are not indicated for splenic repair alone [ 161. Splenectomy is clearly the treatment of choice when the spleen is pulverized or other life-threatening injuries demand immediate attention. In this setting, we currently autotransplant a portion of the spleen and administer polyvalent pneumococcal vaccine [21]. Although numerous animal studies have confirmed immunologic benefits of splenic reimplantation, the extent is variable [22-241. Human data are limited. Our follow-up studies demonstrated uniform implant viability by technetium scanning at 6 months, and showed normalization of the immunoglobulin M levels and platelet numbers [3]. Pate1 et al [.%I have confirmed these findings and, in addition, found disappearance of target cells and Howell-Jolly bodies and return of normal complement levels at 4 weeks. More sophisticated immunologic studies need to be completed. Our experience thus far, however, has demonstrated that splenic autotransplantation is simple and safe. No postoperative complications have occurred with our omental pouch technique. Summary This review was undertaken to analyze critically the complications resulting from operative splenic salvage. Over a 6 year period, 200 adults who sustained splenic trauma underwent laparotomy. The mechanism of injury was blunt in 138 patients (69 percent), a stab wound in 32 patients (16 percent),
Volume 140, December 1984
After Trauma
and a gunshot wound in 30 patients (15 percent). Splenorrhaphy was accomplished in 85 patients (42 percent). Methods of repair included cautery and hemostatic agents in 24 patients (28 percent), debridement and suturing in 42 patients (50 percent), and patial resection in 19 patients (22 percent). Six patients died, four from head trauma and two from multiple organ failure. Postoperative complications occurred in 14 patients. Four were intraabdominal. Three patients required reoperation for splenic hemorrhage; one (2 percent) after suture repair and two (11 percent) after partial resection. A left subphrenic abscess developed in another patient. Splenic reimplantation was performed in 43 patients (22 percent). Five deaths occurred. One was due to head trauma, three to multiple organ failure, and one to overwhelming pneumococcal infection. Eleven postoperative complications occurred, but none was related to splenic autotransplantation. Despite the enthusiasm for splenic salvage, the number of patients suitable for splenorrhaphy plateaued at 56 percent. Complications of splenorrhaphy are infrequent, and the risk increases with more complex salvage attempts. We believe that splenic reimplantation remains a safe procedure. References 1. Dickerman JD. Traumatic asplenia in adults. A defined hazard? Arch Surg 1981;i 18381-3. 2. O’Neal BJ, McDonald X. The risk of sepsis in the asplenic adult. Ann Surg 1981;194:775-8. 3. Mlllikan JS, Moore EE, Moore GE, Stevens RE. Alternatives to splenectomy in adults after trauma: repair, partial resectfon, and reimplantation of splenic tissue. Am J Surg 1982144: 711-8. 4. Van Stiegmann G, Moore EE, Moore GE. Failure of spleen ra pair. J Trauma 1979;19:698-700. 5. Sherman Ft. Perspectives in management of trauma to the spleen: 1979 Presidential address, American Association for the Surgery of Trauma. J Trauma 1981;20:1-13. 6. Michels NA. Blood supply and anatomy of the upper abdominal organs. Philadelphia: JB Lippincott, 1955201. 7. Huu N, Person H, Hong R, Vallee B, Hoan Vu N. Anatomical approach to the vascular segmentation of the spleen (lien) based on controlled experimental partial splenectomles. Anat Clin 1982;4:265-77. 8. Upadhyaya P, Simpson JS. Splenic trauma in children. Surg Gyncecol Obstet 1968;127:781-90. 9. Silverstein ME, Chvapil M. Experimental and clinical experiences with collagen fleece as a hemostatic agent. J Trauma 1981;21:388-93. 10. Campo Chisto M. Segmental resection of the spleen. 0 Hospital (Rio) 1962;62:187-204. 11. Morgenstern L, Shapiro SJ. Techniques for splenic conservation. Arch Surg 1979;114:449-54. 12. Burrington JD. Surgical repair of a ruptured spleen in chiklren. Arch Surg 1977;112:417-9. 13. Weinstein ME, Govin GG. Rice CL, Virgil0 RW. Splenorrhaphy for splenic trauma. J Trauma 1979;19:692-7. 14. Giuliano AE, Lim RC. Is splenic salvage safe in the traumatized patient? Arch Surg 1981;116:651-6. 15. Barrett J, Sheaff C. Abuabara S, Jonasson 0. Splenic preservation In adufts after blunt and penetrating trauma. Am J Sug 1983;145:313-7. 16. Pachter HL, Hofstetter SR, Spencer FC. Evolving concepts in
803
Moore et al
17.
18.
19.
20. 21.
22.
23.
24.
25.
splenic surgery, splenorrhaphy versus splenectomy and post-splenectomy drainage: experience In 105 patients. Ann Surg 1981;194:262-9. Oakes DD, Charters AC. Changing concepts In the management of splenlc trauma. Surg Gynecoi Obstet 1981; 153: 181-5. Horton J, Ogden ME, Williams S, Coin D. The importance of splenlc blood flow in clearing pneumococcal organisms. Ann Surg 1982;195:172-6. Rogers DM, Herrington JL, Morton C. Incidental splenectomy associated with Nissen fundoplication. Ann Surg 1980; 191:153-6. Scher KS, Coil JA. Effects of oxidized cellulose and microfibriiiar collagen on infection. Surgery 1982;91:301-4. Capian ES, Boltansky H, Snyder MJ, et al. Response of traumatized splenectomized patients to immediate vaccination with polyvalent pneumococcal vaccine. J Trauma 1983; 23:801-5. Dickerman JD. Horner SR, Coil JA, Gump DW. The protective effect of intraperitoneal splenic autotransplants in mice exposed to aerosolized suspension of Type iii Streptococcus pneumoniae. Blood 1979;54:354-8. Faschlng MD, Cooney DR. Reimmunlzatlon and spienic autotransplantation: a long-term study of Immunologic response and survival following pneumococcal challenge. J Sug Res 1980;28:449-59. Livingston CD, Levine BA, Sirinek KR. Improved survival rate for lntraperitoneal autotransplantatlon of the spleen foilowlng pneumococcal pneumonia. Surg Gynecoi Obstet 1983; 156:761-6. Pate1 J, Williams JS, Shmigel B, Hinshaw RJ. Preservation of splenic function by autotransplantation of traumatized spleen in man. Surgery 1981;90:683-6.
Discussion David V. Feliciano (Houston, TX): Although all of us recognize the low incidence of fuhninant postsplenectomy sepsis (0.18 cases per hundred person years in the Mayo Clinic series of 1982), most of us try to repair spleens these days rather than perform splenectomy. Splenorrhaphy can be safely performed when good surgical judgment is used. Dr. Moore’s series today documents this, as does our own series from the Ben Taub General Hospital where 135 successful splenorrhaphies, 42 percent of all splenic injuries, have been performed since 1980. It also appears that splenic reimplantation into the leaves of the omentum can be safely performed as well, according to the results just reported, and these fragments will clear red blood cells and platelets, be visible on technetium scans, and produce immtmoglobulin M. However, there are basically two major questions to be considered when discussing the efficacy of extensive splenorrhaphy or reimplantation. First, is the remaining mass of splenic tissue large enough in size to clear an infusion of capsular organisms? Dickerman’s study in 1979 and Van Wyck’s study in 1980 both documented the need for a certain mass of splenic tissue to confer protection against capsular organism infusion in laboratory animals. Second, is the blood flow to the remaining fourth or half of the spleen or a group of splenic fragments sufficient to even get the capsular organisms to the splenic tissue? Coin’s study in 1982, Livington’s study in 1983, and Pabst’s study published in the March 1984 issue of The American Journal of Surgery all emphasized the critical factor of minimal blood flow to the splenic remnants or fragments.
804
Some of the following questions will reflect on these critical points Dr. Moore, why did you not reimplant fragments in 50 patients in this series who had splenectomy? Please comment on the weight of the fragments reimplanted, and the estimated weight of remaining splenic tissue after heroic operations, such as trisegmentectomy, compared with the weight of a normal spleen. How many of the 38 surviving splenic reimplantation patients actually had technetium scans, immunoglobulin M level determinations, platelet counts, and so on and were the results uniformly successful in all 38 patients? Were capsular organisms the cause of death in the splenorrhaphy patient who died from multiple organ failure and the three patients who died from intraabdominal sepsis and multiple organ failure in the reimplantation group? Did you reimplant splenic fragments in the three patients who returned to the operating room after failed splenorrhaphies and, if not, why not? If you are going to use Teflon pledgets to help repair extensive parenchymal defects, how can you say that you must minimize the use of foreign material when splenorrhaphy is performed in the face of gastrointestinal contamination. Finally, I would like to ask about your choice of suture material. I have always used chromic suture or polypropylene suture because either will slide through the spleen and liver a lot easier than some of the newer absorbable sutures. John Moore (Denver, CO): I recently cared for a young man with an isolated splenic injury who underwent splenectomy with splenic implantation. Fever, left quadrant pain, and leukocytosis developed in this patient 9 days after splenic implantation requiring reexploration. At surgery, a fluid collection in the left subphrenic space was found that was consistent with a subphrenic abscess, but more importantly, the entire greater omentum appeared inflamed and was removed. Pathologically, the splenic implants showed liquification necrosis and there was diffuse acute omentitis. Dr. Moore, what has been your experience with this apparent inflammatory response and its clinical significance in the postoperative course? Ernest E. Moore (closing): The incidence of total splenectomy decreased from 55 percent at the inception of this study to 4 percent at the present time. Animal studies suggest that a third of the spleen must be reimplanted for immunologic benefit to occur. We have increased the fragment sizes from 20 by 20 by 3 mm to 40 by 40 by 3 mm to approximate this critical mass. Shaftan and his colleagues have shown in dogs that splenic implants continue to grow over a 2 year period. A standard splenic trisegmentectomy would leave 25 percent of the spleen with its native arterial supply. We are in the process of restudying our patients with splenic implants. In the 16 recalled thus far, all have had technetium uptake by the implants, and normal immunoglobulin M levels as well as platelet counts at 6 months. Pate1 and associates have followed more than 40 patients and have demonstrated normalization of third component of complement levels as well as biopsy-proved regeneration of splenic tissue. The 61 year old cirrhotic patient with multiple organ failure died from overwhelming pneumococcal sepsis 5
The American Journal of Surgery
Splenic Salvage After Trauma
months postoperatively. The three patients with intraperitoneal sepsis had the usual mixed bacterial flora in their abscesses. We reimplanted spleen in both patients with early splenorrhaphy failure, but in the third patient who returned 5 weeks postoperatively this was not feasible. We emphasize that the amount of foreign tissue employed for splenorrhaphy should be minimized. In adults, unlike children, we believe pledgets are warranted for
Voluma 148, Docomhu 1884
major splenic injuries. The issue of polyglycolic acid versus chromic suture is moot. Dr. Moore has raised an important point. Both Tavasoli and Pate1 have shown that splenic autotransplants undergo severe coagulation necrosis during the first 2 postoperative weeks. Early reoperation in these patients may lead to unnecessary removal of the splenic fragments because of this. It may take as long as 6 months for the normal splenic architecture to be reestablished.