Diagnostic Pelvic Computed Tomography in the Rectal-Injured Combat Casualty

Transcription

1 MILITARY MEDICINE, 173, 3:293, 2008 Diagnostic Pelvic Computed Tomography in the Rectal-Injured Combat Casualty MAJ Eric K. Johnson, MC USA*; MAJ Timothy Judge, MC USA*; CPT Jonathan Lundy, MC USA*; MAJ Mark Meyermann, MC USA ABSTRACT Background: The current standard for evaluating trauma patients for penetrating rectal injury is to perform a rigid proctoscopy. This can be laborious and inaccurate. Injuries are often not visualized and a small number of unnecessary colostomies may be created. Computed tomography (CT) scanning of the pelvis may be useful in identifying penetrating rectal injuries. Study Design: A retrospective analysis was performed on data regarding all casualties admitted to the 10th Combat Support Hospital during the period of November 2005 through March Nineteen patients were identified. Patients that were hemodynamically stable underwent preoperative CT scanning. All rectal injuries diagnosed preoperatively were confirmed through a different diagnostic modality in the OR. Results: Nineteen patients with rectal injury or suspected rectal injury were identified. Eight of the 19 were hemodynamically unstable in the emergency medical treatment area and were taken emergently to surgery. For discussion, only stable patients with gunshot wound or blast/fragmentation injury mechanisms were included. No injuries were missed by CT scanning, but there were two false-positive scans. Conclusions: In our brief experience, CT scanning was a useful screening tool to assist in identifying patients with penetrating traumatic rectal injuries. It allowed us to improve triage and make effective use of limited operative resources. INTRODUCTION Penetrating injury to the rectum is an uncommon but serious injury seen in combat casualties. These injuries are typically diagnosed based on initial physical examination findings. Palpation of a rectal injury on digital rectal examination is quite accurate but is not always possible or necessary to drive additional diagnostic investigation. The finding of gross blood on a digital rectal examination should prompt further diagnostic workup. The current gold standard is to perform a rigid proctoscopy in search of the injury. In the past, the presence of gross blood alone was an indication for laparotomy and presumptive treatment of a rectal injury unless a proximal source of bleeding could be identified during exploration. 1 Identification of a proximal source of bleeding still required proctoscopic evaluation unless a rectal injury could be identified at exploration. If the pattern of wounding made rectal injury likely, then one would be required to proceed with appropriate treatment even in the absence of an identified injury. It is likely that unnecessary colostomies were performed in some circumstances. The presence of gross blood on digital rectal examination is commonly seen in battle casualties. It can sometimes be difficult to determine whether this blood is arising from the *Department of Surgery, Dwight David Eisenhower Army Medical Center, 300 Hospital Road, Fort Gordon, GA Department of Radiology, William Beaumont Army Medical Center, 5005 North Piedras Street, Fort Bliss, TX The opinions and assertions contained herein are the private views of the authors and are not to be construed as official or as reflecting the views of the Department of Defense. Use of commercial products in this project does not imply endorsement by the U.S. government. This manuscript was received for review in April The revised manuscript was accepted for publication in November rectal lumen or whether it is coming from soft tissue or other associated injuries. Adequate rigid proctoscopy is difficult to perform in the trauma bay of a combat support hospital. Often these patients must be taken to the operating room (OR) with sedation or anesthetic to perform a suitable examination and, even in this setting, injuries may be difficult to identify. Limited OR space often complicates this situation when a surgeon is confronted with multiple casualties at once. This scenario is commonly encountered in current military operations. The computed tomography (CT) scanner is being used more frequently to evaluate hemodynamically stable trauma patients with penetrating wounds to the abdomen to direct operative and nonoperative management. 2 6 A CT scan of the pelvis may in fact assist in the preoperative diagnosis and triage of patients with suspected penetrating rectal injury. METHODS Study Design This is a retrospective analysis of a case series made up of combat casualties that had either intraoperative identification of a rectal injury or had a preoperative CT scan that suggested rectal injury. Nineteen patients were identified. Radiographic findings that were considered suggestive of rectal injury were: air in the pararectal space, thickening of the rectal wall, perirectal fat stranding, perirectal and pelvic free fluid with no other obvious source, and metallic fragments located in close proximity to the rectum. All rectal injuries were either confirmed or refuted with a different diagnostic modality in the OR despite CT findings. This was done because of the lack of experience in the use of this modality to diagnose rectal injuries. This study was reviewed by the hospital re- 293

2 search committee and did not require institutional review board approval due to the small number of patients in the series. Setting This study was performed at the 10th Combat Support Hospital in Baghdad, Iraq during the period of November 2005 through March The majority of the patients evaluated and treated at this facility were treated for blast, burn, and penetrating trauma. Patient Population The study group was made up of U.S. military personnel, coalition military personnel, civilian contractors, Iraqi military, and Iraqi civilians who were injured in the combat environment. Data Collection The first casualty encountered which was included in the study underwent CT scanning of the abdomen and pelvis before surgical consultation. The CT revealed findings consistent with rectal injury. Casualties subsequently encountered with wounding patterns suspicious for rectal injury underwent preoperative CT scanning of the pelvis if they were hemodynamically stable. Hemodynamically unstable patients with potential intra-abdominal, intrathoracic, or peripheral vascular sources of hemorrhage were taken immediately to the OR without preoperative CT. All CT scans were done with intravenous contrast after the administration of oral gastrografin. No triple contrast scans were performed. A Siemens Somatom Plus 4 scanner (four slice, 8-mm beam collimation with a pitch of 1A/T140 kv, 155 mas; Siemens, Malvern, Pennsylvania) was used in all cases. One hundred to 150 ml of Isovue 300 (Bracco Diagnostics, Princeton, New Jersey) were administered at 2.5 to 3 ml/sec intravenously. Data regarding admission vital signs, admission laboratory values, blood products transfused, time elapsed between admission and start time in the OR, CT findings, and intraoperative findings were collected prospectively (Table I). Data Analysis The data were analyzed retrospectively to determine the accuracy of CT scanning in the identification of rectal injuries. Comparison between patients who did and did not receive preoperative CT scans was done to demonstrate the time delay that occurs with preoperative CT scanning. RESULTS Nineteen patients with suspected rectal injury were identified. The average patient age was 30 years (range, 17 63) and 100% were male. Eight of the 19 were hemodynamically unstable in the emergency medical treatment area and were taken emergently to surgery. Two of the remaining 11 patients were injured in motor vehicle accidents and sustained impalement-type injuries. One of these patients was unstable TABLE I. Comparison of Admission, Time to Treatment, and Transfusion Variables between the CT and Non-CT Groups Variable CT Scan Group (n 11) No CT Scan (n 8) Admit temperature (Fahrenheit) Mean arterial pressure (mm Hg) Base deficit Admit hemoglobin (g/dl) Injury severity score Time to OR (minutes) Time in OR (minutes) Packed red blood cells transfused (units) Fresh-frozen plasma transfused (units) This data includes all patients identified prior to exclusion criteria. TABLE II. Findings That Were Noted on CT Scans in All Patients Who Underwent Study, Excluding the One Impalement Number Demonstrating CT Finding Finding Air in pararectal space 10/10 Thickening of rectal wall 4/10 Fat stranding 1/10 Free fluid 3/10 Adjacent fragment 7/10 and did not undergo CT scanning. Because of the difference in mechanism of injury, these patients will be excluded from the discussion. Two of the 10 patients to be discussed had false-positive CT scans. Stable patients with rectal injuries encountered during the study period underwent preoperative CT scanning. Signs suggesting a rectal injury were obvious on all CT scans performed. Seven of the eight patients with rectal injuries had gross blood on digital rectal examination, although the true source of this blood was difficult to determine in some patients because of bleeding from associated injuries. Two additional patients were diagnosed as having rectal injuries preoperatively based on CT findings, but injuries were not found in the OR. Neither of these patients had rectal bleeding when examined in the emergency medical treatment area. The most common CT finding was that of air in the pararectal space. This was present in all patients studied (Table II). Pararectal air and adjacent fragments were seen in both patients who turned out to be false-positive diagnoses by CT scan. None of the other mentioned findings were seen in these two patients. A total of 2,110 CT scans including the pelvis were performed on trauma patients during the study period. Findings suggesting penetrating rectal injury were noted in 10 patients (0.4%). There were no missed rectal injuries during the study period. Mechanism of injury was split evenly between high-velocity gunshot wounding and blast/fragmentation injury, each representing 50% (Table III). High and low extraperitoneal injuries 294

3 TABLE III. Breakdown of Injury by Mechanism of Injury and Location/Zone of Injury in Each Group Injury CT Group (n 10) Non-CT Group (n 7) High-velocity gunshot wound 5 4 Blast/fragmentation injury 5 3 Intraperitoneal 2 2 Extraperitoneal, upper, (7 10 cm) 3 1 Extraperitoneal, lower, ( 6 cm) 3 2 Combined intraextraperitoneal 0 2 Note that only eight patients are represented in the CT group since there were two false positives diagnosed by CT scan. TABLE IV. Associated Injuries in the 10 Patients Who Underwent CT Scan Associated Injuries Number Soft tissue injuries 7 Bladder injuries 4 Extremity fractures 3 Small bowel 2 Pelvic fractures 2 Urethra 1 Hemo/pneumothoraces 1 Ureter 1 Testicle/scrotum 1 Sciatic nerve 1 FIGURE 1. CT image through the pelvis showing an adjacent fragment, rectal wall thickening, and mesorectal stranding in a patient who sustained an extraperitoneal rectal injury secondary to an improvised explosive device (IED) blast with fragmentation injuries through the left thigh. This includes the two patients without luminal rectal injury. occurred with equal frequency, each in 38% of the patients studied. Intraperitoneal injuries made up the remainder of our cohort. Combined intra- and extraperitoneal injuries were only encountered in the unstable group of patients. The most frequently seen associated injury was destructive soft-tissue trauma. Bladder injuries were the next most common, followed by extremity fractures (Table IV). These each occurred in 70%, 40%, and 30% of the study group, respectively. The diagnosis of probable rectal injury was made based on CT findings in 10 patients but was only correct in 8. Of these 10 patients, 8 had the CT scan ordered specifically to evaluate the pelvis due to a pattern of wounding. The two-false positive results were from this group. Three patients in the group of 10 had CT scans ordered to evaluate the abdomen for injury in addition to the pelvis. Two of these patients would very likely have been taken to the OR without a CT scan if a surgeon had evaluated the patient first. These patients proved to have intraperitoneal rectal injuries as well as small bowel injuries. Representative CT images can be viewed in Figures 1 7. All diagnoses made by CT scan were confirmed by another diagnostic modality in the OR. Six diagnoses were confirmed by proctoscopy, three were confirmed by physical examination, and two were confirmed by laparotomy (Table V). One patient had the diagnosis confirmed by two modalities. No injuries in the CT group were identified preoperatively by physical examination. Of the two patients with false positives, one had the diagnosis refuted at laparotomy which FIGURE 2. CT image through the upper pelvis showing several fragments, sacral fracture, rectal wall thickening, pararectal air, and free fluid in a patient with an intraperitoneal rectal injury secondary to high-velocity gunshot wound through the upper left buttock. was done to repair a bladder injury, and one was refuted by rigid proctoscopy done by two independent surgeons. DISCUSSION Military medical experience and practice during armed conflict has driven the surgical management of penetrating rectal injury since World War I. A decrease in the mortality rate related to these injuries from as high as 50% to as low as 15% has been noted This reduction in mortality is largely related to proximal fecal diversion, distal rectal washout, presacral drainage, debridement and closure of injuries, if possible, and broad spectrum antibiotics. Current surgical methods dictate that a trauma patient who presents with gross 295

4 FIGURE 4. Pelvic CT image showing a fragment in the rectal wall with a small amount of adjacent air and rectal wall thickening in a patient with an intraperitoneal rectal injury secondary to an IED blast. FIGURE 5. Low pelvic CT image showing a large metallic fragment adjacent to the rectum with pararectal air and fluid. This patient sustained an extraperitoneal rectal injury secondary to an IED blast. FIGURE 3. (a) Pelvic CT image showing pararectal air, intravesical air, and rectal wall thickening in a patient who sustained an extraperitoneal rectal injury secondary to high-velocity gunshot wound to the medial thigh with no exit wound. (b) Lower pelvic image from the same patient in (a) showing a pubic ramus fracture and a urethral injury with adjacent fragment. (c) Scout film from the patient in (a) and (b) showing several metallic fragments in the thigh and lower pelvic region, not necessarily suggestive of rectal injury. This is comparable to the information that would be obtained from a standard plain radiograph of the abdomen. blood on digital rectal examination should undergo rigid proctoscopy to confirm a rectal injury. Surgeons should have a high index of suspicion for rectal injury in the absence of gross rectal blood when the wounding pattern suggests the possibility. Low rectal injuries that are within the reach of the examiner s finger may be easily diagnosed in the trauma bay provided the patient is cooperative with the examination. Four patients in the CT scan group had their injuries identified by physical examination, but this was only possible in the OR with the patient under anesthesia. Patients with injuries located higher in the rectum must undergo rigid proctoscopic examination or laparotomy to make the diagnosis. If an injury is not visualized, but blood is present in the rectal lumen in the absence of a proximal source, the patient is typically treated as if they have a rectal injury. In theory, physical 296

5 TABLE V. Method of Rectal Injury Diagnosis in Both Groups Method of Diagnosis CT Group Non-CT Group Physical examination (palpable defect) 0 a 2 Rigid proctoscopy 1 1 CT scan 8 b,c 0 Laparotomy 1 4 a Three injuries were confirmed on a more aggressive physical exam in the OR that the patient would not have tolerated without anesthesia. b An additional two diagnoses were made that turned out to be false positives. One diagnosis was refuted with rigid proctoscopy and one with laparotomy. c Because of inexperience with this diagnostic modality, all CT diagnoses were either confirmed or refuted with proctoscopy, laparotomy, or physical exam; 6-proctoscopy, 3-physical exam, 2-laparotomy (the sum here equals 11 because injuries could be confirmed by more than one modality). FIGURE 6. Low pelvic CT image showing pararectal air. Other images showed fragments adjacent to the rectal wall. This patient was a false positive and had no rectal or intra-abdominal injury. The mechanism of wounding was an IED blast from beneath a vehicle where all fragments penetrated in an upward direction and did not take a trans-pelvic trajectory. FIGURE 7. Upper pelvic CT image showing mesorectal stranding and thickening. Additional images showed pararectal air with a large fragment embedded in the right psoas muscle. This patient was explored and had an injury to the right branch of the superior rectal artery that was ligated and also had an injury to the bladder that was repaired. There was no luminal rectal injury. This patient sustained a high-velocity gunshot wound to the upper right thigh. examination, proctoscopy, and laparotomy have all bases covered. When one considers the rarity of an adequately prepped rectum in the casualties encountered in the combat theater, as well as the fact that most soldiers are moderately dehydrated and on constipating diets, it becomes obvious that proctoscopy may be difficult and inaccurate. The lack of examination setup in the trauma bay must also be considered. In fact, any casualty that required an accurate proctoscopic examination had it done in the OR. Previous studies have shown accuracy rates of rigid proctoscopy to be in the range of 72% to 89%. 11,12 The majority of these patients did not have the injury definitively identified; however, the diagnosis was made based on the presence of intraluminal blood alone. Additionally, the origin of blood seen on rectal examination in this patient population is not necessarily from a rectal injury. If every patient who had rectal blood underwent proctoscopy, considerable resources would be consumed. In the combat setting, the resources referred to are more than just a proctoscope and examining bed. As previously stated, these patients must often be taken to an OR which is already stressed by the number of patients requiring surgery. One patient in the CT group had a rectal injury with no gross blood and no other injury requiring laparotomy. The CT scan was ordered based on a pattern of wounding, clearly demonstrating the injury. Medical facilities operating in a combat theater are often presented with multiple casualties at once. The importance of triage becomes quite obvious when a surgeon is presented with several patients and limited OR space. The ability to better determine the patients that require operative resources is quite important. CT scans of the pelvis in patients with suspected rectal injuries may be part of the solution. There has been a push toward evaluation of penetrating abdominal injuries with CT scanning in our civilian trauma centers in the United States. 2 6 CT evaluation has been reliable in assisting with the decision to operate or not. It has also provided additional information that may change patient management. 2 Many casualties present with multiple small fragmentation injuries and equivocal physical examination findings. Local wound exploration can be quite time consuming, but CT scans of the abdomen and pelvis help to identify those that need to go to the OR for exploration. One CT-related drawback seen in this study population was a delay in operative treatment. However, if time to arrival in the OR of 97.9 minutes is compared to the time achieved by a prominent U.S. trauma center of 130 minutes, 1 it seems to be a minor issue. The patients in this study had penetrating rectal injuries with injury severity scores similar to our CT group. A recent 297

6 FIGURE 8. Algorithm showing possible management of combat casualties with suspected penetrating rectal injuries. Suspicious wounding patterns and perineal bleeding without an obvious source will typically be present in these patients. study conducted in a pediatric population with mainly blunt rectal trauma showed that CT was at least as accurate as rigid proctoscopy in detecting rectal injury. 13 Two patients in this population had injuries that were initially missed on proctoscopy, but later diagnosed on CT scan after deterioration in their condition. A false-negative rate of 31% using proctoscopy to diagnose rectal injury has previously been reported. 14 One author (E. K. Johnson) performed several of the proctoscopic examinations and only definitively identified injuries after marked persistence because of positive CT scan findings. While it is very difficult to draw any definitive conclusions based on a retrospective study with such a small patient population, these findings are compelling. We do not propose that pelvic CT is mandatory in the evaluation of these casualties. Placing a stable patient in the CT scanner has little associated risk. Based on data presented, it is unlikely that a penetrating rectal injury will be missed by a CT examination. False positives will most likely be encountered. The main advantage of a preoperative CT scan will be avoidance of rigid proctoscopy in the nonbleeding casualty who has a pattern of wounding suspicious for rectal injury. This may assist in triage and more effective utilization of limited resources. The addition of rectal contrast would likely increase the specificity of CT examination. The modern combat support hospital has the ability to perform highquality CT examination which may be quite helpful in managing this group of casualties. Pelvic CT is merely an additional tool to assist the surgeon in managing these casualties (Fig. 8). These methods may also be useful to our civilian trauma colleagues. However, the differences in mechanism of injury may limit its usefulness in the civilian setting. CONCLUSION Pelvic CT is useful to assist the physician in making a preoperative diagnosis of penetrating rectal injury in the military setting. Further prospective evaluation of this diagnostic modality using rectal contrast may show improved results. REFERENCES 1. Thomas DD, Levison MA, Dykstra BJ, Bender JS: Management of rectal injuries, dogma versus practice. Am Surg 1990; 56: Velmahos GC, Constantinou C, Tillou A, Brown CV, Salim A, Demetriades D: Abdominal computed tomographic scan for patients with gunshot wounds to the abdomen selected for nonoperative management. J Trauma 2005; 59: Shanmuganathan K, Mirvis SE, Chiu WC, Killeen KL, Hogan GJ, Scalea TM: Penetrating torso trauma: triple contrast helical CT in peritoneal violation and organ injury: a prospective study in 200 patients. Radiology 2004; 231:

7 4. Grossman MD, May AK, Schwab CW, et al: Determining anatomic injury with computed tomography in selected torso gunshot wounds. J Trauma 1998; 45: Ginzburg E, Carillo EH, Kopelman T, et al: The role of computed tomography in selective management of gunshot wounds to the abdomen and flank. J Trauma 1998; 45: Chiu WC, Shanmuganathan K, Mirvis SE, Scalea TM: Determining the need for laparotomy in penetrating torso trauma: a prospective study using triple contrast enhanced abdominopelvic computed tomography. J Trauma 2001; 51: Wallace C: A study of 1200 cases of gunshot wounds to the abdomen. Br J Surg 1917; 4: Ogilvie WH: Abdominal wounds in the western desert. Surg Gynecol Obstet 1944; 78: Lavenson GS, Cohen A: Management of rectal injuries. Am J Surg 1971; 122: Armstrong RG, Schmitt HJ, Patterson LT: Combat wounds of the extraperitoneal rectum. Surgery 1973; 74: Levine JH, Longo WE, Priutt C, Mazoski JE, Shapiro MJ, Durham RM: Management of selected rectal injuries by primary repair. Am J Surg 1996; 172: Ivatury RR, Licata J, Gunduz Y, Rao P, Stahl WM: Management options in penetrating rectal injuries. Am Surg 1991; 57: Leaphart CL, Danko M, Cassidy L, Gaines B, Hackam DJ: An analysis of proctoscopy vs. computed tomography scanning in the diagnosis of rectal injuries in children: which is better? J Pediatr Surg 2006; 41: Grasberger RC, Hirsch EF: Rectal trauma: a retrospective analysis and guidelines for therapy. Am J Surg 1983; 145: