Occult Colocutaneous Fistula Following Pelvic Fracture
S Thomas, M Walsh, D Zimmer, B Speicher, M Sarb, B Boyer, A Newbold, C Arnold, M Donnino
colocutaneous fistula, ct scan, diagnostic delay, enterocutaneous fistula, pelvic fracture, trauma
S Thomas, M Walsh, D Zimmer, B Speicher, M Sarb, B Boyer, A Newbold, C Arnold, M Donnino. Occult Colocutaneous Fistula Following Pelvic Fracture. The Internet Journal of Emergency and Intensive Care Medicine. 2009 Volume 12 Number 1.
An enterocutaneous fistula is a rare complication following a pelvic fracture, and a colocutaneous fistula is even rarer. Whereas most cases of bowel entrapment due to a pelvic fracture involve the mobile small bowel, we are reporting a case in which the sigmoid colon became entrapped in the pelvic fragments. Bowel entrapment led to the delayed development of an adynamic ileus, followed by an open communicating fistula between the colon and the skin. Twenty-one cases of bowel entrapment within a pelvic fracture following blunt trauma have been reported in the literature since 1907.1,2 Only five of the twenty-one cases involved the large bowel, excluding the rectum. 2,3,5,6 However, none of the subjects in these cases presented with spontaneous rupture of a colocutaneous fistula, as did the patient in the present study.
Work was done in the Department of Emergency Medicine at Memorial Hospital, South Bend, IN.
A fifty-three year old man presented to the emergency department after being struck by a five hundred pound clamp used to align natural gas pipeline. The clamp struck the patient in the left flank/hip region and pinned him to the seat of a tractor. The findings during his primary emergency department evaluation included normal vital signs and bruising of the abdominal wall over his left iliac crest. X-rays revealed a three centimeter closed fracture of his iliac wing, which was non-displaced (Figure 1). A minor contusion of his spleen was identified on the abdominal CT scan. He was discharged home on bed rest and given oral narcotics for pain control.
Forty-eight hours post-injury, the patient visited a different emergency department in a neighboring state complaining of abdominal discomfort and constipation. At this second visit, his vital signs were normal. Physical exam revealed that the patient was in a moderate amount of pain. He complained of a tender left flank with a large, red, and warm hematoma covering the flank and iliac crest. His abdomen was soft, without guarding or rebound tenderness, and auscultation revealed normal bowel sounds. A repeat CT scan of the abdomen was unchanged from his initial presentation. The CBC showed a WBC of 10,700 with a normal hemoglobin level. He was discharged home with a stool softener, additional narcotics for pain control, and crutches for ambulation.
Nine days after his initial injury, while sitting on the toilet to defecate, the patient expelled projectile stool through a fistula on the left side of his abdominal wall. Food particles and stool were splattered on the bathroom wall. He was transported by paramedics to yet another hospital. On exam, his vital signs included an oral temperature of 100.3 F, a pulse of 125, and a blood pressure of 114/52. He had a large defect in the skin over his left flank and a tender abdomen (Figure 2). Surgery consultation confirmed an obvious colocutaneous fistula which required immediate exploratory laparotomy and colostomy.
Exploration through laparotomy revealed a small, non-bleeding splenic laceration and a large abscess cavity between the sigmoid colon, the iliac crest, and the skin. There was a three centimeter defect in the sigmoid colon that communicated with the fistula. It was clear from surgical exploration that the mechanism behind the etiology of this fistula involved the fractured segment of the pelvis. Initial speculation was raised as to whether the etiology of this fistula may simply have been due to the crushing effect of the trauma, however surgical examination put this issue to rest. The fractured iliac crest protruded into the abscess cavity confirming it as the source of colonic perforation leading to the fistula formation. The necrotic tissue was debrided, the abdomen was irrigated, and sigmoid colectomy with colostomy and Hartmann rectal stump closure were performed.
During his hospital course the patient required intravenous antibiotics, parenteral hyperalimentation, and local wound management. On post-operative day five the patient began a liquid diet, and later advanced to a soft diet that he tolerated well. On post-operative day seven he was transferred, in good condition, by fixed wing aircraft to his hometown. No subsequent follow-up was available.
Twenty-one cases of bowel entrapment within a pelvic fracture following blunt trauma have been reported in the literature since 1907.1,2 Five of the twenty-one cases involved the large bowel, excluding the rectum.1,2,3,5,6 None of these cases presented with spontaneous rupture of a colocutaneous fistula, as did the present subject.
When pelvic fracture occurs, the fracture segment most often injures the small bowel, either directly or through entrapment, and does not usually injure the adherent colon. One author has suggested that the colon is less frequently involved because the colon, as compared to the small bowel, is relatively fixed via attachments to the retroperitoneum.1 When the colon is involved, fistulation can be caused by the fixed descending colon being crushed on the anterior superior iliac crest2, with disruption of the iliacus muscle allowing communication between the peritoneum and the fracture fragment.3 Alternatively, a fistula can form when a bone fragment perforates the bowel.4
The clinical diagnosis of bowel injury following pelvic trauma can be made difficult by a delay in presentation and can masquerade as an ileus.1 Adynamic ileus occurs in up to 18% of all patients with pelvic fractures and lasts an average of 2.6 days.7,8 Of the five reported cases of colonic injury associated with pelvic fracture, the longest delay in diagnosis was 90 days.5
CT scan with the addition of enteric contrast can increase radiologic sensitivity, and is recommended to evaluate prolonged ileus in patients with pelvic fractures.1 However, even with radiologic testing, bowel injuries are often missed and lead to life-threatening consequences. Generally, missed abdominal injuries have been reported in 2% of blunt and penetrating abdominal trauma.9 Although CT scanning is usually effective in detecting impending colocutaneous fistula following abdominal and pelvic trauma, CT scan with oral contrast may not detect the herniated or entrapped bowel due to the normal transit time of contrast through the bowel.10 In fact, in the present case, the two initial CT scans of the abdomen done with oral and IV contrast were normal. Another reason that the injury eluded early detection was that the patient was not examined radiologically with a pelvic Judet view, an oblique view that allows better delineation of the iliac crests. This may have revealed the offending pelvic fragment.
In a review of the literature, we were unable to find other cases wherein a patient presented with defecation through a colocutaneous fistula due to a pelvic fracture, particularly one that initially seemed to be relatively minor and stable. It is also uncommon to find examples of fistulas that formed in such a brief period of time and presented so explosively. In cases involving pelvic fractures with concern for an enterocutaneous fistula, a CT examination with oral contrast should be performed to determine any abnormalities in the bowel or other organs near the fracture. Moreover, when pelvic fractures are noted, an oblique Judet view should be obtained in an attempt to better delineate the fracture segment. If the patient's clinical findings suggest the development of an enterocutaneous fistula, despite a normal abdominal CT scan, an exploratory laparotomy should be seriously considered as the next course of action.