- Email: [email protected]
Management of Civilian Extraperitoneal Rectal Injuries
Original Article
Management of Civilian Extraperitoneal Rectal Injuries Nawaf J. Shatnawi and Kamal E. Bani-Hani, Department of Surgery, Faculty of Medicine, King Abdullah University Hospital and Jordan University of Science and Technology, Irbid, Jordan.
OBJECTIVE: The mortality and morbidity of rectal injuries are highly unsatisfactory. We retrospectively reviewed our experience with rectal injuries to draw some practical guidelines for management of such injuries. METHODS: The medical records of all patients diagnosed at our hospitals with full-thickness rectal injuries between 1994 and 2003 were retrospectively reviewed. RESULTS: Full-thickness rectal injuries were identified in 23 patients; 19 patients had extraperitoneal injuries and four had both intra- and extraperitoneal injuries. The mean age was 33.5 years (range, 5–73 years). The mechanism of injury was penetrating in 11 patients, blunt in six, impalement in three and iatrogenic in three. Injuries were closed primarily in 17 patients, with variable combinations of adjunct procedures. Eight patients were treated without colostomy. Drainage and rectal washout were performed in 11 and six patients, respectively. Overall, 11 patients developed complications, including eight wound infections and five pelvic septic complications related to the rectal injury. Four of the five pelvic septic complications and all three deaths occurred in patients with shock, at least two associated-organ injuries and more than 6 hours’ delay in treatment. CONCLUSION: Rectal injuries are serious additive mortality and morbidity factors in multi-injured patients. Regardless of treatment modality, wound infection is associated with shock at presentation and more than 6 hours’ delay in treatment. [Asian J Surg 2006;29(1):11–6] Key Words: colostomy, impalement, rectal injury, rectal repair, septic pelvic complication
Introduction Civilian injuries to the extraperitoneal rectum are uncommon.1,2 Their surgical management is based on experience gained during war. Colostomy, rectal wound repair, open presacral drainage and rectal washout are among the modalities used to deal with such injuries.3–5 Despite this, the morbidity of rectal injuries continues to be unsatisfactory, and no consensus regarding the optimal management of civilian rectal injuries has been established.1,2,6 High-velocity missiles are the causative factors in battlefield rectal wounds,2,4,6 while low-velocity missiles, impalement, blunt trauma and perforation secondary to diagnostic and therapeutic instrumentation are recognized causes in civilian rectal injuries.2,4,7 The degree
of rectal and associated-organ damage is dependent on the mechanism of trauma. Rectal injuries are unlikely to be the cause of morbidity or mortality in the first few days following injuries, when polytraumatized patients have other factors that can cause death.2,5 The management of polytraumatized patients should follow an established protocol of primary survey, resuscitation, stabilization, secondary survey and definitive treatment.8 During the secondary survey, digital rectal examination should be performed, especially in patients with suspected rectal injury; gross rectal blood, wounds in close proximity, pelvic fracture, injuries to the genitourinary tract and lower abdominal pain or tenderness point to the possibility of rectal injury.1,5 In most series, negative digital rectal examination did not exclude the diagnosis.9,10 Procto-
Address correspondence and reprint requests to Professor Kamal E. Bani-Hani, Department of Surgery, Faculty of Medicine, Jordan University of Science and Technology, P.O. Box 3030, Irbid 22110, Jordan. E-mail: [email protected] • Date of acceptance: 28 July 2005 © 2006 Elsevier. All rights reserved.
ASIAN JOURNAL OF SURGERY VOL 29 • NO 1 • JANUARY 2006
070/2001
11
■ SHATNAWI & BANI-HANI ■
sigmoidoscopic examination should follow any suspicious findings.9 Unless rectal injuries are diagnosed and treated properly, they can be additive to high-risk injuries. In this paper, we present our experience with management of rectal injuries in order to identify factors that might help in patient selection for proper matching surgical options and, if possible, to develop some guidelines regarding the management of patients with rectal injuries.
Patients and methods The medical records of all patients with full-thickness rectal injury who were managed at our hospitals between 1994 and 2003 were retrospectively reviewed. Clinical data on age, gender, mechanism of injury, presentation, physical findings, diagnostic work-up and time lag before surgical intervention were collected. The anatomical sites of rectal injuries, associated injuries, operative management, type of colostomy, presacral drainage and rectal washout, if done, were also recorded, as were morbidity and mortality. Data were analysed using SPSS version 11 (SPSS Inc, Chicago, IL, USA) for Windows. Only patients with true extraperitoneal injury to the rectal wall, which is uncovered by serosa as defined by McGrath et al,11 were included in this study. None of the patients had only mucosal laceration or complete rectal avulsion. A rectal injury scaling system was used to characterize the degree of rectal wall damage.12
Results During the study period, 23 patients with full-thickness rectal injury were managed at our hospitals. None had complete rectal avulsion, sphincter injuries or mucosal lacerations. The clinical findings are summarized in Table 1. Three patients had iatrogenic injuries, two related to sigmoidoscopic rectal perforation and one due to self-induced
Table 1. Clinical findings among patients with rectal injuries (n = 23) Shock Abdominal pain Wound in close proximity Gross rectal bleeding Microscopic blood Sigmoidoscopic haematoma Sigmoidoscopic laceration
12
13 19 15 14 7 13 12
Table 2. Mechanism of injury
Mechanism
Penetrating Gunshot wound Iron rod Tree branch Blunt Impalement Iatrogenic Total
n
11 8 2 1 6 3 3 23
Associated abdominal injuries
* 2 associated abdominal injuries
9
7
6 2 0 17
3 1 0 11
catheter enema (Table 2). Two impalement injuries occurred as a result of slipping in the bathroom, where the plastic rod of the toilet brush impaled the rectum, while the third was a result of criminal assault by repeated introduction of a heavy wooden stick into the rectum. Only six patients had isolated rectal injury; the remaining 17 patients had associated abdominal injuries (11 with * 2 associated-organ injuries). Extraperitoneal injuries were recorded in 19 patients, while the remaining four patients had both intra- and extraperitoneal injuries. There was rectal bleeding in 91% of patients. The preoperative diagnosis was confirmed by rigid proctosigmoidoscopy in 17 patients. In two patients, the diagnosis was made by on-table intraoperative flexible proctosigmoidoscopy. In the remaining four patients, the diagnosis was made at exploration. Seventeen patients had rectal injury repair using various surgical combinations. Fifteen patients underwent colostomy as part of their treatment (1 had end colostomy with mucous fistula, 4 had transverse loop colostomy and 10 had sigmoid loop colostomy). The remaining eight patients had no colostomy. Rectal washout with 4–6 L of warm saline solution was performed in six patients. Closed vacuum suction presacral drain through a separate lateral abdominal wound was used in 11 patients. All patients received different combinations of antibiotics preoperatively and continued for variable intervals postoperatively. Twelve patients, five with isolated rectal injuries, four with single-organ associated injuries and three with at least two associated-organ injuries, had no complications. Eight abdominal wound infections developed including two suprapubic wound tract infections. Among patients with abdominal wounds, two developed wound dehiscence and later died. Perineal infection occurred in five patients, includ-
ASIAN JOURNAL OF SURGERY VOL 29 • NO 1 • JANUARY 2006
070/20
■ EXTRAPERITONEAL RECTAL INJURIES ■
ing three wound tract local infections and two abscesses in the supralevator space (1 perineal prerectal abscess and 1 pelvic abscess). No patients with penetrating wounds developed wound tract abscesses and no rectal fistula developed in any patient. Comparison between patients who had colostomy and those who did not showed that abdominal wound infection did not differ significantly (p = 0.101; Table 3). Four perineal infections occurred in patients with at least two associated injuries. None of the six patients with isolated rectal injuries had perineal infections. Furthermore, in four of the five pa-
tients with pelvic septic complications, treatment was delayed for more than 6 hours; three had hypotension on admission. Regardless of the cause, complications occurred in 11 patients (48%). Pelvic septic complications occurred in five patients (22%). Three patients died. All deaths and 80% of pelvic infections occurred in patients with shock at presentation, at least two associated-organ injuries and more than 6 hours’ delay in management. In the nonrepair group (n = 6), three pelvic infections and two deaths occurred. The significant risk factors for wound infection were shock at presentation and more than 6 hours’ delay in management (p = 0.029; Table 3).
Table 3. Clinicopathological characteristics of patients with rectal injury (n = 23) and risk factors for patients who developed wound infection (n = 8), pelvic infection (n = 5) or who died (n = 3) Wound infection Yes
No
Age ) 30 yr > 30 yr
2 6
9 6
Mechanism of injury Impalement Penetration Blunt Iatrogenic
1 4 2 1
2 7 4 2
1 1 3 3
5 5 3 2
1 7
9 6
Drain Absent Present
2 6
100 5
Colostomy Absent Present
1 7
7 8
Associated injuries None 1 2 *3 Shock Absent Present
p
No
2 3
9 9
1 3 1 0
2 8 5 3
0 1 3 1
6 5 3 4
1 4
9 9
2 3
100 8
1 4
7 110
2 3
150 3
3 2
3 150
ASIAN JOURNAL OF SURGERY VOL 29 • NO 1 • JANUARY 2006
1 1 1 0
2 100 5 3
0 0 1 2
6 6 5 3
0 3
100 100
0 3
120 8
0 3
8 120
0.624
0.167
0.103
0.052
0.175
0.759 2 1
150 5
2 1
4 160
0.051
0.029 9 6
110 9
0.051
0.056
1 7
0 3
0.433
0.363
0.086
0.231 1 4
9 9
p 0.075
0.538
0.101
2 130
No
0.231
0.057
4 4
Yes
0.205
0.029
Repair Absent Present
p
0.715
0.290
120 3
Death
0.692
0.999
5 3
070/2001
Yes
0.110
Rectal washout Absent Present
Time interval to repair ) 6 hr > 6 hr
Pelvic infection
0.103 0 3
100 100
13
■ SHATNAWI & BANI-HANI ■
Discussion Low-velocity missiles are the most common reported causes of civilian rectal injuries.1 In eight of our patients, the injuries were caused by low-velocity missiles. The high proportion of blunt rectal injuries among our patients probably reflects the high incidence of motor vehicle accidents in Jordan. Sigmoidoscopic examinations were positive in all patients who had the examination. As reported by others, patients presenting with shock needed immediate surgery to deal with life-threatening injuries. The presence of unstable pelvic fracture was the main reason for delay or no proctosigmoidoscopic examination.2 One of our patients with isolated extraperitoneal rectal injury had free intra- and retroperitoneal gas at the same time. Free intraperitoneal gas may be absorbed from the extraperitoneal space.2,7,13 Various combinations of surgical options were used in the management of our patients. This reflects the different surgical backgrounds and individual preferences of the surgeons. Unlike other series, debridement and primary repair was possible in 17 patients (74%); primary repair was possible in only 10–40% of cases reported by others.10,14–17 Rectal fistula is a well-known complication of extraperitoneal rectal injury, but none of our patients had rectal fistula. Nonrepair in the presence of genitourinary injuries increases complications18 and one study found a higher incidence of rectal fistula in the nonrepair group.11 However, other investigators did not find an increase in septic complications and fistula formation in association with nonrepair.16,17,19 Levy et al recommend repair if the wound is easily amenable or if it is found during exploration to repair associated pelvic organs.9 It might be advisable to omit extensive dissection to look for suspected extraperitoneal rectal injuries, especially in patients with shock or haematoma secondary to pelvic fractures.10 However, preoperative sigmoidoscopic findings even on the operating table can be of great help in planning for possible repair, because it can indicate the exact distance of the injury from the anal verge and its location (anterior, posterior or lateral), which might help in minimizing the amount of dissection. In eight of our patients, repair was primarily performed without colostomy as these patients were less severely injured. One patient in this subgroup developed pelvic wound tract infection and wound infection. Seven patients had repair as the only surgical modality, and none of them developed pelvic septic complications. There is growing support for primary
14
repair without diversion in selected patients.7,11,20 Morbidity in this series was 48%, and 22% of patients had pelvic septic complications. Pelvic septic complication can be attributed to rectal injuries.4–6,21 Wound contamination at the time of injury is a potential hazard related to large-bowel injuries.22 In theory, faecal diversion can prevent further contamination,2,23 presacral drainage can prevent development of closed-space abscess,4 and rectal washout can prevent further soiling by removing gross faecal matter24 and decreasing the chance of bacterial translocation related to reduced mucosal defence in stressed patients.25 However, creation of a colostomy, repair, drainage and rectal washout cannot prevent contamination at the time of an accident. Drainage for rectal injuries was used in 11 patients. Perineal presacral drainage is an established part of rectal wound management2–5,15,17 and significant reduction of septic complications is reported with the use of presacral drains.5,15 However, the routine use of presacral drains may be unnecessary for penetrating rectal wounds due to low-velocity missiles.6,9,19,26,27 We found no increase in pelvic septic complications in the non-drain group. Although we used a closedsystem suction drain through a separate lateral abdominal wound, we had comparable septic pelvic complications to others using open presacral perineal drainage.4,15 Four pelvic infections occurred in drained patients or in those with colostomy. We can incriminate neither the colostomy nor the drain for these infections. Seriously injured patients are treated aggressively. We believe that omitting drainage in our patients was not associated with a higher rate of pelvic septic complications. Transperitoneal drains have been used for extraperitoneal rectal injury2 and it is claimed that disruption of the perineum by dissection leads to tissue destruction.28 Rectal washout was used in six patients. Rectal washout is an established part of rectal wound management.5,21,29 In theory, it can prevent gross soiling, as well as bacterial translocation. Shannon et al found significant reduction in pelvic septic complications in patients who had washout.24 On the other hand, rectal washout has been challenged as it adds no benefit.6,15,16,30 Three of the five pelvic infections in this study occurred in the rectal washout group. In a recent randomized prospective trial, elective colorectal surgery without preoperative mechanical preparation did not significantly increase the septic complication rate.31 There is a theoretical disadvantage related to rectal washout because it can lead to spillage of faecal material into prerectal fatty tissue, especially if rectal repair is not performed.17
ASIAN JOURNAL OF SURGERY VOL 29 • NO 1 • JANUARY 2006
070/20
■ EXTRAPERITONEAL RECTAL INJURIES ■
Although diverting colostomy is an accepted standard in dealing with extraperitoneal rectal injuries,2–7 there is increased awareness of complications related to colostomy and colostomy closure.32 Colostomy is intended to prevent further soiling and to protect the integrity of the anastomosis. There are reports of successful management of colonic injuries by primary repair without colostomy31,33 and other scattered reports of successful management of extraperitoneal rectal wounds without colostomy.9,16,30 Safe elective colorectal anastomosis without mechanical preparation further supports that faecal matter cannot disrupt the integrity of a safe, tension-free anastomosis. 31,33 This contradicts the work of Nahai et al in an animal model.34 If rectal repair can be done safely, is there any need for an additional option to prevent further soiling? Three patients died in this group, a mortality rate of 13%. Continuing improvement in mortality resulting from rectal injuries can be attributed to the following factors: reduction in time lag before surgery; improved resuscitative measures; widespread use of effective antibiotics; proper postoperative management; and advances in surgical and anaesthetic skills.1,5,7,9,13,27 Regardless of the treatment modalities used in the management of patients with extraperitoneal rectal injuries, a mortality of 0–15.5% and morbidity of 11–46% has been reported.2,4,9,13,29 Although current mortality is low, complications remain highly unsatisfactory.14,15,17,19,21,29 Risk factors for wound infection in our patients were shock at presentation and more than 6 hours’ delay in treatment. Because of the small sample size and small number of events, our study has insufficient power to detect any association. None of the factors were significantly associated with mortality (Table 3). However, all deaths occurred in patients who presented with shock, had associated injuries (* 2 organs) and more than 6 hours’ delay in treatment and who were older than 30 years. In patients with these risk factors, there was no significant difference between the drained and non-drained groups, or between washout and non-washout, in relation to pelvic infection. Wound infection was more common in the drained group and the colostomy group, but the sample was too small to establish significance. It has been reported that the high morbidity and mortality associated with colorectal injuries are related to associated injuries, shock at presentation and delayed treatment.2,4,5,13,17,35–37 Two-thirds of patients who present more than 12 hours after injury have hypotension and sepsis, and intra-abdominal abscess develops in patients who are neglected or have associated injuries, but there is no relation to the method of drainage.2 Haemorrhagic shock, especially when
ASIAN JOURNAL OF SURGERY VOL 29 • NO 1 • JANUARY 2006
070/2001
coupled with soft-tissue injury, caused marked immune suppression in an animal model.38,39 Even though our patients were young, age above 30 years was associated with death and complications. The immunosuppressive effects of age and haemorrhage, in addition to the significance of associated injury, enhance the development of intra-abdominal abscess.39 Immune suppression in older patients with trauma has been suggested.40 Rectal injuries are serious additive mortality and morbidity factors in multi-injured patients. Regardless of the treatment modalities, wound infection in patients with rectal injury is associated with shock at presentation and more than 6 hours’ delay in treatment.
References 1. Burch JM. Injury to the colon and rectum. In: Mattox KL, Felicianco DV, Moore EE, eds. Trauma. New York: McGraw-Hill, 1999:763–82. 2. Grasberger RC, Hirsch EF. Rectal trauma. A retrospective analysis and guidelines for therapy. Am J Surg 1983;145:795–9. 3. Hanna SS, Jirsch DW. Management of colonic and rectal injuries. Can Med Assoc J 1979;120:1387–91. 4. Ganchrow MI, Lavenson GS Jr, McNamara JJ. Surgical management of traumatic injuries of the colon and rectum. Arch Surg 1970;100: 515–20. 5. Armstrong RG, Schmitt HJ Jr, Patterson LT. Combat wounds of the extraperitoneal rectum. Surgery 1973;74:570–4. 6. Gonzalez RP, Falimirski ME, Holevar MR. The role of presacral drainage in the management of penetrating rectal injuries. J Trauma 1998;45:656–61. 7. Haas PA, Fox TA Jr. Civilian injuries of the rectum and anus. Dis Colon Rectum 1979;22:17–23. 8. Gruen GS, Leit ME, Gruen RJ, et al. The acute management of hemodynamically unstable multiple trauma patients with pelvic ring fractures. J Trauma 1994;36:706–11. 9. Levy RD, Strauss P, Aladgem D, et al. Extraperitoneal rectal gunshot injuries. J Trauma 1995;38:273–7. 10. Levine JH, Longo WE, Pruitt C, et al. Management of selected rectal injuries by primary repair. Am J Surg 1996;172:575–8. 11. McGrath V, Fabian TC, Croce MA, et al. Rectal trauma: management based on anatomic distinctions. Am Surg 1998;64:1136–41. 12. Moore EE, Cogbill TH, Malangoni MA, et al. Organ injury scaling, II: pancreas, duodenum, small bowel, colon, and rectum. J Trauma 1990;30:1427–9. 13. Robertson HD, Ray JE, Ferrari BT, et al. Management of rectal trauma. Surg Gynecol Obstet 1982;154:161–4. 14. Mangiante EC, Graham AD, Fabian TC. Rectal gunshot wounds. Management of civilian injuries. Am Surg 1986;52:37–40. 15. Burch JM, Feliciano DV, Mattox KL. Colostomy and drainage for civilian rectal injuries: is that all? Ann Surg 1989;209:600–10. 16. Ivatury RR, Licata J, Gunduz Y, et al. Management options in penetrating rectal injuries. Am Surg 1991;57:50–5. 17. Tuggle D, Huber PJ Jr. Management of rectal trauma. Am J Surg 1984;
15
■ SHATNAWI & BANI-HANI ■
148:806–8. 18. Franko ER, Ivatury RR, Schwalb DM. Combined penetrating rectal and genitourinary injuries: a challenge in management. J Trauma 1993;34:347–53. 19. Bostick PJ, Johnson DA, Heard JF, et al. Management of extraperitoneal rectal injuries. J Natl Med Assoc 1993;85:460–3. 20. Hughes TM, Elton C, Hitos K, et al. Intra-abdominal gastrointestinal tract injuries following blunt trauma: the experience of an Australian trauma centre. Injury 2002;33:617–26. 21. Brunner RG, Shatney CH. Diagnostic and therapeutic aspects of rectal trauma. Blunt versus penetrating. Am Surg 1987;53:215–9. 22. Fabian TC. Prevention of infections following penetrating abdominal trauma. Am J Surg 1993;165(2A Suppl):14S–9S. 23. Lavenson GS, Cohen A. Management of rectal injuries. Am J Surg 1971;122:226–30. 24. Shannon FL, Moore EE, Moore FA, et al. Value of distal colon washout in civilian rectal trauma—reducing gut bacterial translocation. J Trauma 1988;28:989–94. 25. Moore EE, Jones TN. Benefits of immediate jejunostomy feeding after major abdominal trauma—a prospective, randomized study. J Trauma 1986;26:874–81. 26. Steinig JP, Boyd CR. Presacral drainage in penetrating extraperitoneal rectal injuries: is it necessary? Am Surg 1996;62:765–7. 27. Velmahos GC, Gomez H, Falabella A, et al. Operative management of civilian rectal gunshot wounds: simpler is better. World J Surg 2000;24:114–8. 28. Klein RR, Scarborough RA. Traumatic perforations of the rectum and distal colon. Am J Surg 1953;86:515–22. 29. Vitale GC, Richardson JD, Flint LM. Successful management of injuries to the extraperitoneal rectum. Am Surg 1983;49:159–62.
16
30. Thomas DD, Levison MA, Dykstra BJ, et al. Management of rectal injuries. Dogma versus practice. Am Surg 1990;56:507–10. 31. Zmora O, Mahajna A, Bar-Zakai B, et al. Colon and rectal surgery without mechanical bowel preparation: a randomized prospective trial. Ann Surg 2003;237:363–7. 32. Berne JD, Velmahos GC, Chan LS, et al. The high morbidity of colostomy closure after trauma: further support for the primary repair of colon injuries. Surgery 1998;123:157–64. 33. Sasaki LS, Allaben RD, Golwala R, et al. Primary repair of colon injuries: a prospective randomized study. J Trauma 1995;39:895– 901. 34. Nahai F, Lamb JM, Havican RG, et al. Factors involved in disruption of intestinal anastomoses. Am Surg 1977;43:45–51. 35. Williams MD, Watts D, Fakhry S. Colon injury after blunt abdominal trauma: results of the EAST Multi-Institutional Hollow Viscus Injury Study. J Trauma 2003;55:906–12. 36. Bozorgzadeh A, Pizzi WF, Barie PS, et al. The duration of antibiotic administration in penetrating abdominal trauma. Am J Surg 1999; 177:125–31. 37. Heckbert SR, Vedder NB, Hoffman W, et al. Outcome after hemorrhagic shock in trauma patients. J Trauma 1998;45:545–9. 38. Fabian TC, Croce MA, Fabian MJ, et al. Reduced tumor necrosis factor production in endotoxin-spiked whole blood after trauma: experimental results and clinical correlation. Surgery 1995;118: 63–72. 39. Croce MA, Fabian TC, Patton JH Jr, et al. Impact of stomach and colon injuries on intra-abdominal abscess and the synergistic effect of hemorrhage and associated injury. J Trauma 1998;45:649–55. 40. Weksler ME. Senescence of the immune system. Med Clin North Am 1983;67:263–72.
ASIAN JOURNAL OF SURGERY VOL 29 • NO 1 • JANUARY 2006
070/20
Management of Civilian Extraperitoneal Rectal Injuries Nawaf J. Shatnawi and Kamal E. Bani-Hani, Department of Surgery, Faculty of Medicine, King Abdullah University Hospital and Jordan University of Science and Technology, Irbid, Jordan.
OBJECTIVE: The mortality and morbidity of rectal injuries are highly unsatisfactory. We retrospectively reviewed our experience with rectal injuries to draw some practical guidelines for management of such injuries. METHODS: The medical records of all patients diagnosed at our hospitals with full-thickness rectal injuries between 1994 and 2003 were retrospectively reviewed. RESULTS: Full-thickness rectal injuries were identified in 23 patients; 19 patients had extraperitoneal injuries and four had both intra- and extraperitoneal injuries. The mean age was 33.5 years (range, 5–73 years). The mechanism of injury was penetrating in 11 patients, blunt in six, impalement in three and iatrogenic in three. Injuries were closed primarily in 17 patients, with variable combinations of adjunct procedures. Eight patients were treated without colostomy. Drainage and rectal washout were performed in 11 and six patients, respectively. Overall, 11 patients developed complications, including eight wound infections and five pelvic septic complications related to the rectal injury. Four of the five pelvic septic complications and all three deaths occurred in patients with shock, at least two associated-organ injuries and more than 6 hours’ delay in treatment. CONCLUSION: Rectal injuries are serious additive mortality and morbidity factors in multi-injured patients. Regardless of treatment modality, wound infection is associated with shock at presentation and more than 6 hours’ delay in treatment. [Asian J Surg 2006;29(1):11–6] Key Words: colostomy, impalement, rectal injury, rectal repair, septic pelvic complication
Introduction Civilian injuries to the extraperitoneal rectum are uncommon.1,2 Their surgical management is based on experience gained during war. Colostomy, rectal wound repair, open presacral drainage and rectal washout are among the modalities used to deal with such injuries.3–5 Despite this, the morbidity of rectal injuries continues to be unsatisfactory, and no consensus regarding the optimal management of civilian rectal injuries has been established.1,2,6 High-velocity missiles are the causative factors in battlefield rectal wounds,2,4,6 while low-velocity missiles, impalement, blunt trauma and perforation secondary to diagnostic and therapeutic instrumentation are recognized causes in civilian rectal injuries.2,4,7 The degree
of rectal and associated-organ damage is dependent on the mechanism of trauma. Rectal injuries are unlikely to be the cause of morbidity or mortality in the first few days following injuries, when polytraumatized patients have other factors that can cause death.2,5 The management of polytraumatized patients should follow an established protocol of primary survey, resuscitation, stabilization, secondary survey and definitive treatment.8 During the secondary survey, digital rectal examination should be performed, especially in patients with suspected rectal injury; gross rectal blood, wounds in close proximity, pelvic fracture, injuries to the genitourinary tract and lower abdominal pain or tenderness point to the possibility of rectal injury.1,5 In most series, negative digital rectal examination did not exclude the diagnosis.9,10 Procto-
Address correspondence and reprint requests to Professor Kamal E. Bani-Hani, Department of Surgery, Faculty of Medicine, Jordan University of Science and Technology, P.O. Box 3030, Irbid 22110, Jordan. E-mail: [email protected] • Date of acceptance: 28 July 2005 © 2006 Elsevier. All rights reserved.
ASIAN JOURNAL OF SURGERY VOL 29 • NO 1 • JANUARY 2006
070/2001
11
■ SHATNAWI & BANI-HANI ■
sigmoidoscopic examination should follow any suspicious findings.9 Unless rectal injuries are diagnosed and treated properly, they can be additive to high-risk injuries. In this paper, we present our experience with management of rectal injuries in order to identify factors that might help in patient selection for proper matching surgical options and, if possible, to develop some guidelines regarding the management of patients with rectal injuries.
Patients and methods The medical records of all patients with full-thickness rectal injury who were managed at our hospitals between 1994 and 2003 were retrospectively reviewed. Clinical data on age, gender, mechanism of injury, presentation, physical findings, diagnostic work-up and time lag before surgical intervention were collected. The anatomical sites of rectal injuries, associated injuries, operative management, type of colostomy, presacral drainage and rectal washout, if done, were also recorded, as were morbidity and mortality. Data were analysed using SPSS version 11 (SPSS Inc, Chicago, IL, USA) for Windows. Only patients with true extraperitoneal injury to the rectal wall, which is uncovered by serosa as defined by McGrath et al,11 were included in this study. None of the patients had only mucosal laceration or complete rectal avulsion. A rectal injury scaling system was used to characterize the degree of rectal wall damage.12
Results During the study period, 23 patients with full-thickness rectal injury were managed at our hospitals. None had complete rectal avulsion, sphincter injuries or mucosal lacerations. The clinical findings are summarized in Table 1. Three patients had iatrogenic injuries, two related to sigmoidoscopic rectal perforation and one due to self-induced
Table 1. Clinical findings among patients with rectal injuries (n = 23) Shock Abdominal pain Wound in close proximity Gross rectal bleeding Microscopic blood Sigmoidoscopic haematoma Sigmoidoscopic laceration
12
13 19 15 14 7 13 12
Table 2. Mechanism of injury
Mechanism
Penetrating Gunshot wound Iron rod Tree branch Blunt Impalement Iatrogenic Total
n
11 8 2 1 6 3 3 23
Associated abdominal injuries
* 2 associated abdominal injuries
9
7
6 2 0 17
3 1 0 11
catheter enema (Table 2). Two impalement injuries occurred as a result of slipping in the bathroom, where the plastic rod of the toilet brush impaled the rectum, while the third was a result of criminal assault by repeated introduction of a heavy wooden stick into the rectum. Only six patients had isolated rectal injury; the remaining 17 patients had associated abdominal injuries (11 with * 2 associated-organ injuries). Extraperitoneal injuries were recorded in 19 patients, while the remaining four patients had both intra- and extraperitoneal injuries. There was rectal bleeding in 91% of patients. The preoperative diagnosis was confirmed by rigid proctosigmoidoscopy in 17 patients. In two patients, the diagnosis was made by on-table intraoperative flexible proctosigmoidoscopy. In the remaining four patients, the diagnosis was made at exploration. Seventeen patients had rectal injury repair using various surgical combinations. Fifteen patients underwent colostomy as part of their treatment (1 had end colostomy with mucous fistula, 4 had transverse loop colostomy and 10 had sigmoid loop colostomy). The remaining eight patients had no colostomy. Rectal washout with 4–6 L of warm saline solution was performed in six patients. Closed vacuum suction presacral drain through a separate lateral abdominal wound was used in 11 patients. All patients received different combinations of antibiotics preoperatively and continued for variable intervals postoperatively. Twelve patients, five with isolated rectal injuries, four with single-organ associated injuries and three with at least two associated-organ injuries, had no complications. Eight abdominal wound infections developed including two suprapubic wound tract infections. Among patients with abdominal wounds, two developed wound dehiscence and later died. Perineal infection occurred in five patients, includ-
ASIAN JOURNAL OF SURGERY VOL 29 • NO 1 • JANUARY 2006
070/20
■ EXTRAPERITONEAL RECTAL INJURIES ■
ing three wound tract local infections and two abscesses in the supralevator space (1 perineal prerectal abscess and 1 pelvic abscess). No patients with penetrating wounds developed wound tract abscesses and no rectal fistula developed in any patient. Comparison between patients who had colostomy and those who did not showed that abdominal wound infection did not differ significantly (p = 0.101; Table 3). Four perineal infections occurred in patients with at least two associated injuries. None of the six patients with isolated rectal injuries had perineal infections. Furthermore, in four of the five pa-
tients with pelvic septic complications, treatment was delayed for more than 6 hours; three had hypotension on admission. Regardless of the cause, complications occurred in 11 patients (48%). Pelvic septic complications occurred in five patients (22%). Three patients died. All deaths and 80% of pelvic infections occurred in patients with shock at presentation, at least two associated-organ injuries and more than 6 hours’ delay in management. In the nonrepair group (n = 6), three pelvic infections and two deaths occurred. The significant risk factors for wound infection were shock at presentation and more than 6 hours’ delay in management (p = 0.029; Table 3).
Table 3. Clinicopathological characteristics of patients with rectal injury (n = 23) and risk factors for patients who developed wound infection (n = 8), pelvic infection (n = 5) or who died (n = 3) Wound infection Yes
No
Age ) 30 yr > 30 yr
2 6
9 6
Mechanism of injury Impalement Penetration Blunt Iatrogenic
1 4 2 1
2 7 4 2
1 1 3 3
5 5 3 2
1 7
9 6
Drain Absent Present
2 6
100 5
Colostomy Absent Present
1 7
7 8
Associated injuries None 1 2 *3 Shock Absent Present
p
No
2 3
9 9
1 3 1 0
2 8 5 3
0 1 3 1
6 5 3 4
1 4
9 9
2 3
100 8
1 4
7 110
2 3
150 3
3 2
3 150
ASIAN JOURNAL OF SURGERY VOL 29 • NO 1 • JANUARY 2006
1 1 1 0
2 100 5 3
0 0 1 2
6 6 5 3
0 3
100 100
0 3
120 8
0 3
8 120
0.624
0.167
0.103
0.052
0.175
0.759 2 1
150 5
2 1
4 160
0.051
0.029 9 6
110 9
0.051
0.056
1 7
0 3
0.433
0.363
0.086
0.231 1 4
9 9
p 0.075
0.538
0.101
2 130
No
0.231
0.057
4 4
Yes
0.205
0.029
Repair Absent Present
p
0.715
0.290
120 3
Death
0.692
0.999
5 3
070/2001
Yes
0.110
Rectal washout Absent Present
Time interval to repair ) 6 hr > 6 hr
Pelvic infection
0.103 0 3
100 100
13
■ SHATNAWI & BANI-HANI ■
Discussion Low-velocity missiles are the most common reported causes of civilian rectal injuries.1 In eight of our patients, the injuries were caused by low-velocity missiles. The high proportion of blunt rectal injuries among our patients probably reflects the high incidence of motor vehicle accidents in Jordan. Sigmoidoscopic examinations were positive in all patients who had the examination. As reported by others, patients presenting with shock needed immediate surgery to deal with life-threatening injuries. The presence of unstable pelvic fracture was the main reason for delay or no proctosigmoidoscopic examination.2 One of our patients with isolated extraperitoneal rectal injury had free intra- and retroperitoneal gas at the same time. Free intraperitoneal gas may be absorbed from the extraperitoneal space.2,7,13 Various combinations of surgical options were used in the management of our patients. This reflects the different surgical backgrounds and individual preferences of the surgeons. Unlike other series, debridement and primary repair was possible in 17 patients (74%); primary repair was possible in only 10–40% of cases reported by others.10,14–17 Rectal fistula is a well-known complication of extraperitoneal rectal injury, but none of our patients had rectal fistula. Nonrepair in the presence of genitourinary injuries increases complications18 and one study found a higher incidence of rectal fistula in the nonrepair group.11 However, other investigators did not find an increase in septic complications and fistula formation in association with nonrepair.16,17,19 Levy et al recommend repair if the wound is easily amenable or if it is found during exploration to repair associated pelvic organs.9 It might be advisable to omit extensive dissection to look for suspected extraperitoneal rectal injuries, especially in patients with shock or haematoma secondary to pelvic fractures.10 However, preoperative sigmoidoscopic findings even on the operating table can be of great help in planning for possible repair, because it can indicate the exact distance of the injury from the anal verge and its location (anterior, posterior or lateral), which might help in minimizing the amount of dissection. In eight of our patients, repair was primarily performed without colostomy as these patients were less severely injured. One patient in this subgroup developed pelvic wound tract infection and wound infection. Seven patients had repair as the only surgical modality, and none of them developed pelvic septic complications. There is growing support for primary
14
repair without diversion in selected patients.7,11,20 Morbidity in this series was 48%, and 22% of patients had pelvic septic complications. Pelvic septic complication can be attributed to rectal injuries.4–6,21 Wound contamination at the time of injury is a potential hazard related to large-bowel injuries.22 In theory, faecal diversion can prevent further contamination,2,23 presacral drainage can prevent development of closed-space abscess,4 and rectal washout can prevent further soiling by removing gross faecal matter24 and decreasing the chance of bacterial translocation related to reduced mucosal defence in stressed patients.25 However, creation of a colostomy, repair, drainage and rectal washout cannot prevent contamination at the time of an accident. Drainage for rectal injuries was used in 11 patients. Perineal presacral drainage is an established part of rectal wound management2–5,15,17 and significant reduction of septic complications is reported with the use of presacral drains.5,15 However, the routine use of presacral drains may be unnecessary for penetrating rectal wounds due to low-velocity missiles.6,9,19,26,27 We found no increase in pelvic septic complications in the non-drain group. Although we used a closedsystem suction drain through a separate lateral abdominal wound, we had comparable septic pelvic complications to others using open presacral perineal drainage.4,15 Four pelvic infections occurred in drained patients or in those with colostomy. We can incriminate neither the colostomy nor the drain for these infections. Seriously injured patients are treated aggressively. We believe that omitting drainage in our patients was not associated with a higher rate of pelvic septic complications. Transperitoneal drains have been used for extraperitoneal rectal injury2 and it is claimed that disruption of the perineum by dissection leads to tissue destruction.28 Rectal washout was used in six patients. Rectal washout is an established part of rectal wound management.5,21,29 In theory, it can prevent gross soiling, as well as bacterial translocation. Shannon et al found significant reduction in pelvic septic complications in patients who had washout.24 On the other hand, rectal washout has been challenged as it adds no benefit.6,15,16,30 Three of the five pelvic infections in this study occurred in the rectal washout group. In a recent randomized prospective trial, elective colorectal surgery without preoperative mechanical preparation did not significantly increase the septic complication rate.31 There is a theoretical disadvantage related to rectal washout because it can lead to spillage of faecal material into prerectal fatty tissue, especially if rectal repair is not performed.17
ASIAN JOURNAL OF SURGERY VOL 29 • NO 1 • JANUARY 2006
070/20
■ EXTRAPERITONEAL RECTAL INJURIES ■
Although diverting colostomy is an accepted standard in dealing with extraperitoneal rectal injuries,2–7 there is increased awareness of complications related to colostomy and colostomy closure.32 Colostomy is intended to prevent further soiling and to protect the integrity of the anastomosis. There are reports of successful management of colonic injuries by primary repair without colostomy31,33 and other scattered reports of successful management of extraperitoneal rectal wounds without colostomy.9,16,30 Safe elective colorectal anastomosis without mechanical preparation further supports that faecal matter cannot disrupt the integrity of a safe, tension-free anastomosis. 31,33 This contradicts the work of Nahai et al in an animal model.34 If rectal repair can be done safely, is there any need for an additional option to prevent further soiling? Three patients died in this group, a mortality rate of 13%. Continuing improvement in mortality resulting from rectal injuries can be attributed to the following factors: reduction in time lag before surgery; improved resuscitative measures; widespread use of effective antibiotics; proper postoperative management; and advances in surgical and anaesthetic skills.1,5,7,9,13,27 Regardless of the treatment modalities used in the management of patients with extraperitoneal rectal injuries, a mortality of 0–15.5% and morbidity of 11–46% has been reported.2,4,9,13,29 Although current mortality is low, complications remain highly unsatisfactory.14,15,17,19,21,29 Risk factors for wound infection in our patients were shock at presentation and more than 6 hours’ delay in treatment. Because of the small sample size and small number of events, our study has insufficient power to detect any association. None of the factors were significantly associated with mortality (Table 3). However, all deaths occurred in patients who presented with shock, had associated injuries (* 2 organs) and more than 6 hours’ delay in treatment and who were older than 30 years. In patients with these risk factors, there was no significant difference between the drained and non-drained groups, or between washout and non-washout, in relation to pelvic infection. Wound infection was more common in the drained group and the colostomy group, but the sample was too small to establish significance. It has been reported that the high morbidity and mortality associated with colorectal injuries are related to associated injuries, shock at presentation and delayed treatment.2,4,5,13,17,35–37 Two-thirds of patients who present more than 12 hours after injury have hypotension and sepsis, and intra-abdominal abscess develops in patients who are neglected or have associated injuries, but there is no relation to the method of drainage.2 Haemorrhagic shock, especially when
ASIAN JOURNAL OF SURGERY VOL 29 • NO 1 • JANUARY 2006
070/2001
coupled with soft-tissue injury, caused marked immune suppression in an animal model.38,39 Even though our patients were young, age above 30 years was associated with death and complications. The immunosuppressive effects of age and haemorrhage, in addition to the significance of associated injury, enhance the development of intra-abdominal abscess.39 Immune suppression in older patients with trauma has been suggested.40 Rectal injuries are serious additive mortality and morbidity factors in multi-injured patients. Regardless of the treatment modalities, wound infection in patients with rectal injury is associated with shock at presentation and more than 6 hours’ delay in treatment.
References 1. Burch JM. Injury to the colon and rectum. In: Mattox KL, Felicianco DV, Moore EE, eds. Trauma. New York: McGraw-Hill, 1999:763–82. 2. Grasberger RC, Hirsch EF. Rectal trauma. A retrospective analysis and guidelines for therapy. Am J Surg 1983;145:795–9. 3. Hanna SS, Jirsch DW. Management of colonic and rectal injuries. Can Med Assoc J 1979;120:1387–91. 4. Ganchrow MI, Lavenson GS Jr, McNamara JJ. Surgical management of traumatic injuries of the colon and rectum. Arch Surg 1970;100: 515–20. 5. Armstrong RG, Schmitt HJ Jr, Patterson LT. Combat wounds of the extraperitoneal rectum. Surgery 1973;74:570–4. 6. Gonzalez RP, Falimirski ME, Holevar MR. The role of presacral drainage in the management of penetrating rectal injuries. J Trauma 1998;45:656–61. 7. Haas PA, Fox TA Jr. Civilian injuries of the rectum and anus. Dis Colon Rectum 1979;22:17–23. 8. Gruen GS, Leit ME, Gruen RJ, et al. The acute management of hemodynamically unstable multiple trauma patients with pelvic ring fractures. J Trauma 1994;36:706–11. 9. Levy RD, Strauss P, Aladgem D, et al. Extraperitoneal rectal gunshot injuries. J Trauma 1995;38:273–7. 10. Levine JH, Longo WE, Pruitt C, et al. Management of selected rectal injuries by primary repair. Am J Surg 1996;172:575–8. 11. McGrath V, Fabian TC, Croce MA, et al. Rectal trauma: management based on anatomic distinctions. Am Surg 1998;64:1136–41. 12. Moore EE, Cogbill TH, Malangoni MA, et al. Organ injury scaling, II: pancreas, duodenum, small bowel, colon, and rectum. J Trauma 1990;30:1427–9. 13. Robertson HD, Ray JE, Ferrari BT, et al. Management of rectal trauma. Surg Gynecol Obstet 1982;154:161–4. 14. Mangiante EC, Graham AD, Fabian TC. Rectal gunshot wounds. Management of civilian injuries. Am Surg 1986;52:37–40. 15. Burch JM, Feliciano DV, Mattox KL. Colostomy and drainage for civilian rectal injuries: is that all? Ann Surg 1989;209:600–10. 16. Ivatury RR, Licata J, Gunduz Y, et al. Management options in penetrating rectal injuries. Am Surg 1991;57:50–5. 17. Tuggle D, Huber PJ Jr. Management of rectal trauma. Am J Surg 1984;
15
■ SHATNAWI & BANI-HANI ■
148:806–8. 18. Franko ER, Ivatury RR, Schwalb DM. Combined penetrating rectal and genitourinary injuries: a challenge in management. J Trauma 1993;34:347–53. 19. Bostick PJ, Johnson DA, Heard JF, et al. Management of extraperitoneal rectal injuries. J Natl Med Assoc 1993;85:460–3. 20. Hughes TM, Elton C, Hitos K, et al. Intra-abdominal gastrointestinal tract injuries following blunt trauma: the experience of an Australian trauma centre. Injury 2002;33:617–26. 21. Brunner RG, Shatney CH. Diagnostic and therapeutic aspects of rectal trauma. Blunt versus penetrating. Am Surg 1987;53:215–9. 22. Fabian TC. Prevention of infections following penetrating abdominal trauma. Am J Surg 1993;165(2A Suppl):14S–9S. 23. Lavenson GS, Cohen A. Management of rectal injuries. Am J Surg 1971;122:226–30. 24. Shannon FL, Moore EE, Moore FA, et al. Value of distal colon washout in civilian rectal trauma—reducing gut bacterial translocation. J Trauma 1988;28:989–94. 25. Moore EE, Jones TN. Benefits of immediate jejunostomy feeding after major abdominal trauma—a prospective, randomized study. J Trauma 1986;26:874–81. 26. Steinig JP, Boyd CR. Presacral drainage in penetrating extraperitoneal rectal injuries: is it necessary? Am Surg 1996;62:765–7. 27. Velmahos GC, Gomez H, Falabella A, et al. Operative management of civilian rectal gunshot wounds: simpler is better. World J Surg 2000;24:114–8. 28. Klein RR, Scarborough RA. Traumatic perforations of the rectum and distal colon. Am J Surg 1953;86:515–22. 29. Vitale GC, Richardson JD, Flint LM. Successful management of injuries to the extraperitoneal rectum. Am Surg 1983;49:159–62.
16
30. Thomas DD, Levison MA, Dykstra BJ, et al. Management of rectal injuries. Dogma versus practice. Am Surg 1990;56:507–10. 31. Zmora O, Mahajna A, Bar-Zakai B, et al. Colon and rectal surgery without mechanical bowel preparation: a randomized prospective trial. Ann Surg 2003;237:363–7. 32. Berne JD, Velmahos GC, Chan LS, et al. The high morbidity of colostomy closure after trauma: further support for the primary repair of colon injuries. Surgery 1998;123:157–64. 33. Sasaki LS, Allaben RD, Golwala R, et al. Primary repair of colon injuries: a prospective randomized study. J Trauma 1995;39:895– 901. 34. Nahai F, Lamb JM, Havican RG, et al. Factors involved in disruption of intestinal anastomoses. Am Surg 1977;43:45–51. 35. Williams MD, Watts D, Fakhry S. Colon injury after blunt abdominal trauma: results of the EAST Multi-Institutional Hollow Viscus Injury Study. J Trauma 2003;55:906–12. 36. Bozorgzadeh A, Pizzi WF, Barie PS, et al. The duration of antibiotic administration in penetrating abdominal trauma. Am J Surg 1999; 177:125–31. 37. Heckbert SR, Vedder NB, Hoffman W, et al. Outcome after hemorrhagic shock in trauma patients. J Trauma 1998;45:545–9. 38. Fabian TC, Croce MA, Fabian MJ, et al. Reduced tumor necrosis factor production in endotoxin-spiked whole blood after trauma: experimental results and clinical correlation. Surgery 1995;118: 63–72. 39. Croce MA, Fabian TC, Patton JH Jr, et al. Impact of stomach and colon injuries on intra-abdominal abscess and the synergistic effect of hemorrhage and associated injury. J Trauma 1998;45:649–55. 40. Weksler ME. Senescence of the immune system. Med Clin North Am 1983;67:263–72.
ASIAN JOURNAL OF SURGERY VOL 29 • NO 1 • JANUARY 2006
070/20