Management of anastomotic leak: lessons learned from a large colon and rectal surgery training program.

BACKGROUND: Anastomotic leak is a dreaded surgical complication that can lead to significant morbidity and mortality. Despite its prevalence, there is no consensus on the management of anastomotic leak. This study aimed to review the management of anastomotic leak in the Division of Colon and Rectal Surgery at two institutions. METHODS: This is a retrospective review of all anastomotic leaks occurring after surgery in the Division of Colon and Rectal Surgery at two teaching institutions during 1997-2008. RESULTS: Altogether, 103 leaks occurred in 1,707 anastomoses (6 %), with a median time to diagnosis of 20 days (2-1,400 days). The 90-day mortality rate was 3 %. The majority of cases were managed nonoperatively (73 %), and the majority of leaks were from an extraperitoneal anastomosis (67 %). Success (i.e., radiographic demonstration of a healed leak, restored gastrointestinal continuity) occurred in 54 % of operatively managed leaks and 57 % of nonoperatively managed leaks (56 % overall). Operative management differed by leak location. In 91 % of patients with intraperitoneal leaks, the anastomosis was resected. In 76 % of patients with extraperitoneal leaks, diversion and drainage alone was performed without manipulating the anastomosis. Nonoperative management was successful for 57 % of extraperitoneal leaks and 58 % of intraperitoneal leaks. There was no significant difference in the success rates based on type of management (operative/nonoperative) for either extraperitoneal or intraperitoneal leaks. CONCLUSIONS: Anastomotic leak continues to result in patient morbidity and mortality. Its diverse presentation requires tailoring management to the patient. Nonoperative and operative treatments are viable options for intraperitoneal and extraperitoneal leaks based on patient presentation.

Hand-sewn versus stapled intestinal anastomoses in a chronically steroid-treated porcine model.

Chronic steroid treatment is known to delay wound healing; however, there are no trials examining stapled intestinal anastomoses in subjects chronically treated with steroids. The current study compares mechanically stapled to manually sutured intestinal anastomoses in a steroid-treated porcine model. Twenty adult male pigs were treated with daily prednisolone (0.5 mg/kg IM) for 14 days. The pigs were divided between hand-sewn and stapled groups. All pigs underwent a laparotomy with construction of a jejunojejunostomy, an ileocecostomy, and a colocolostomy. Pigs were sacrificed on postoperative day 4. Bursting pressure, collagen content, and inflammatory scores were used to evaluate each anastomosis. Bursting pressure was greater for the hand-sewn group at the jejunojejunostomies only (1.52 vs 0.79 psi; P < 0.001). The collagen content and microscopic inflammatory scores were not significantly different at any location. The mean gross inflammatory score was greater for the hand-sewn colocolostomies only (5.20 vs 4.00; P < 0.01). Stapled ileocecostomies and colocolostomies appear to heal as well as hand-sewn anastomoses. This may not be true of the jejunojejunostomies as evidenced by the decreased bursting pressures. These data may provide evidence in support of the use of stapled anastomoses in steroid-treated patients.

Retrorectal cyst: a rare tumor frequently misdiagnosed.

BACKGROUND: The rarity of retrorectal cysts and their nonspecific clinical presentations often lead to misdiagnoses and inappropriate operations. In recent years, several such patients have been referred to our institutions for evaluation and treatment of misdiagnosed retrorectal cysts. A review of these patients is presented. STUDY DESIGN: Medical records of the colorectal surgery divisions at two institutions were reviewed. Patients found to have previously misdiagnosed retrorectal cysts were identified. Preliminary diagnoses, radiologic examinations, operative procedures, and final diagnoses were obtained. RESULTS: Seven patients with retrorectal cysts who had been misdiagnosed before referral were identified. These patients had been treated for fistulae in ano, pilonidal cysts, perianal abscesses; psychogenic, lower back, posttraumatic, or postpartum pain, and proctalgia fugax before the correct diagnosis was made. Patients underwent an average of 4.1 operative procedures. Physical examination in combination with CT scanning made the correct diagnosis in all patients. All patients underwent successful resection through a parasacrococcygeal approach, and six of seven did not require coccygectomy. The resected tumors included four hamartomas, two epidermoid cysts, and one enteric duplication cyst. CONCLUSIONS: Retrorectal cysts are a rare entity that can be difficult to diagnose without a high index of clinical suspicion. A history of multiple unsuccessful procedures should alert the clinician to the diagnosis of retrorectal cyst. Once suspected, the correct diagnosis can be made with physical examination and a CT scan before a definitive surgical procedure.

Early experience with stapled hemorrhoidectomy in the United States.

INTRODUCTION: We report the early results of patients treated with stapled hemorrhoidectomy, which has recently been introduced into the United States. METHODS: Sixty-eight patients with symptomatic hemorrhoids were treated at two institutions with the Proximate HCS Hemorrhoidal Circular Stapler supplied by Ethicon Endo-Surgery. Patients were prospectively evaluated for functional recovery and postoperative pain on a 1 to 10 scale. RESULTS: There were 45 (66 percent) males and 23 (34 percent) females with a mean age of 56 years and median duration of symptoms of 5 years. The mean operative time was 22.2 minutes. The operation was performed with spinal (50 percent), local (40 percent), or general (10 percent) anesthesia and as an outpatient (56 percent) or overnight admission (44 percent). Ninety-three percent of patients remained asymptomatic with a mean follow-up of 34 weeks, whereas the remaining 7 percent required either surgical excision or rubber band ligation for persistent symptoms. There was no mortality, new incontinence, fecal impaction, or persistent pain. The total morbidity was 19 percent, with urinary retention as the most common complication (12 percent). The mean pain score decreased from 3.6 on postoperative Day 1 to 1.4 at postoperative Day 7. Ninety-nine percent of patients made a complete functional recovery by postoperative Day 7. CONCLUSIONS: Stapled hemorrhoidectomy is safe, effective, and can be performed as an outpatient procedure with local or regional anesthesia. There seems to be minimal postoperative pain and early recovery, although a benefit over traditional hemorrhoidectomy needs to be proven in a randomized trial.

Superselective mesenteric embolization with microcoils in a porcine model.

PURPOSE: There is a lack of data regarding the degree of ischemic change that may occur in small and large bowel after superselective arterial embolotherapy with platinum microcoils. The purpose of this study was to gain a clearer understanding of the ischemic complications of superselective embolization of peripheral mesenteric arteries with microcoils by gross and histologic examination of postembolectomy bowel specimens at several time intervals in an adult porcine model. METHODS: Two-millimeter platinum microcoils with fibrils were superselectively deployed into six isolated mesenteric branch vessels in each of nine adult pigs. The animals were observed for two days (n = 3), one month (n = 3), or three months (n = 3) and then killed. At necropsy, intestines were examined for gross abnormalities. Sections of bowel containing microcoils were identified under fluoroscopy, resected, and evaluated histologically. RESULTS: A total of 54 microcoils were deployed into the distal arterial mesentery supplying the jejunum (n = 14), ileum (n = 26), and colon (n = 14) of nine adult pigs. Each animal received six microcoils. There were no clinical complications, and all pigs gained weight during their observation periods. Gross examination of the intestines did not reveal any evidence of acute or chronic ischemia. The coils were found in the distal arterial vasculature of the intestine. Histologic examination revealed mild superficial necrosis of villous tips in several samples; however, this finding is likely related to a fixation artifact rather than ischemic injury. There was no other histologic evidence suggestive of ischemic injury. CONCLUSIONS: Gross and histologic findings after superselective arterial embolotherapy demonstrated minimal changes in the bowel, and there were no significant clinical consequences to the animals. We conclude that transcatheter arterial embolotherapy in pigs is safe and may be applicable in the control of massive lower gastrointestinal hemorrhage in humans.