Bidirectional endoscopy in patients with fecal occult blood.

BACKGROUND: Current screening protocols for colorectal cancer depend primarily on fecal occult blood testing (FOBT). However, positive test results do not always indicate the presence of a colonic neoplasm. METHODS: We reviewed the results of 100 consecutive bidirectional (upper and lower) endoscopic procedures performed to evaluate positive FOBT results. Patients were excluded if they presented with gross bleeding, a history of bowel lesions, or previous intestinal operations. There were 31 women and 69 men whose mean age was 51 years. RESULTS: Major abnormalities were found on esophagogastroduodenoscopy (n = 24), colonoscopy (n = 13), or both studies (n = 2). Active bleeding was manifested in two patients, (Barrett's ulcer, duodenal arteriovenous malformation). Two other patients had malignancy: One had a cecal adenocarcinoma and the other a gastric adenocarcinoma. Various benign lesions also were identified in the stomach including esophagitis (n = 8), ulcers/erosions (n = 8) varices (n = 5), and arteriovenous malformations (n = 2). Colonic pathology included polyps (n = 8), arteriovenous malformations (n = 3), and rectal varices (n = 1). Diverticulosis and hemorrhoidal disease were present in 29 and 16 patients, respectively, but were not considered to be likely sources of a positive FOBT. CONCLUSION: Positive FOBT results may indicate the presence of either upper or lower intestinal pathology, and bidirectional endoscopy is an efficient and accurate technique for the comprehensive evaluation of occult bleeding.

Laparoscopic surgery for stage III colon cancer: long-term follow-up.

BACKGROUND: The role of laparoscopic surgery in the management of colorectal cancer is controversial. This study was undertaken to determine the oncological adequacy, in terms of margins of resection, lymph node harvest, and anastomotic and locoregional recurrence of laparoscopic colectomy in patients with stage III (node-positive) colorectal cancer. METHODS: The results of laparoscopic colectomy in 50 consecutive patients with stage III colorectal cancer operated on at a single hospital between 1991 and 1998 were analyzed with respect to postoperative morbidity, mortality, and long-term survival by the Kaplan-Meier method. Methodical patient follow-up was the mainstay of the study. RESULTS: There were 31 men (52%) and 19 women (38%) with a mean age of 67.7 years (range, 40-88). Low anterior resection was performed in 17 cases, abdominal perineal resection in five cases sigmoid colectomy in 10 cases, left hemicolectomy in six cases, right hemicolectomy in seven cases, transverse colectomy in one case, and subtotal colectomy in four cases. Conversion was necessary in three cases (6%). Major complications included one leak, one pelvic abscess, one perineal wound infection, and three anastomotic strictures early in the experience, with none in the past 4 years. One early death occurred due to massive stroke. Median length of stay was 6 days (range, 3-37). Forty-six patients were staged as CII and four as CI colon cancer. The average number of positive nodes was 5.1 (range, 3-58). The margins of resection were adequate in all patients. Follow-up ranged from 3 to 75 months (average, 29.3; median, 24). Overall cancer-related mortality was 34% (17 patients); three patients died of unrelated causes with no detectable cancer. All who died of cancer had distant disease; three of them also had pelvic recurrence. Mean time of death was 21.7 months. There were no anastomotic recurrences or trocar site implants. Overall 3- and 5-year survival was 54.5% and 38.5%, respectively; cancer-adjusted survival was 60.8% and 49.1%. CONCLUSIONS: Based on this study, laparoscopic colectomy in patients with stage III colorectal cancer is oncologically adequate. It results in a long-term outcome comparable to that of traditional open surgery and is associated with low perioperative mortality and morbidity (lower wound infection rate, lower wound recurrences at trocar sites) and a shortened length of stay.

Laparoscopic management of enterocutaneous fistula.

Enterocutaneous fistulas develop in settings of prior abdominal surgery, inflammatory bowel disease, diverticulitis, radiation or malignancy. Traditional surgical management requires laparotomy with bowel resection and anastomosis and is associated with a high incidence of wound infection. Recent advances in instrumentation and accumulation of experience has allowed minimally invasive surgery to become an alternative and often preferred approach to handling complex surgical problems. We present a case of successful laparoscopic management of an enterocutaneous fistula that developed in the setting of prior colectomy and laparoscopic inguinal hernia repair with prosthetic mesh. Laparotomy and its attending complications were avoided facilitating recovery and return to work.

Effects of granulocyte colony-stimulating factor in severe pancreatitis.

BACKGROUND: The effect of granulocyte colony-stimulating factor (G-CSF) on the rate of secondary infections in acute pancreatitis was evaluated in a canine model. Infectious complications are the major determinant of morbidity and mortality in severe pancreatitis. Bacterial translocation has been shown to be a cause of these secondary infections. The relative immunosuppression found with pancreatitis may promote translocation and the spread of bacteria to the pancreas. METHODS: Thirty-four mongrel dogs were studied. Pancreatitis was induced in 18 dogs; 9 were treated with 100 micrograms G-CSF/day and 9 were given only saline solution. Laparotomy alone was done in 16 dogs of which one half were given 100 micrograms G-CSF/day and one half were given saline solution. Daily blood counts and cultures were obtained. All dogs were killed on day 7, and the mesenteric lymph nodes, pancreas, liver, spleen, and peritoneal fluid were cultured and studied histologically. RESULTS: G-CSF caused a significant and sustained increase in mature granulocytes in dogs given pancreatitis. No difference was found in the rate of translocation to mesenteric lymph nodes in dogs given G-CSF (n = 4) versus dogs given saline solution (n = 6). However, a significant decrease occurred in the spread of bacteria to distant sites in dogs given G-CSF (1 versus 15, p < 0.05). CONCLUSIONS: Although G-CSF does not decrease the rate of translocation, it does decrease the rate of distant infection in severe acute pancreatitis.

Pancreatic pseudocyst resolution after parathyroidectomy for hyperparathyroidism.

Hyperparathyroidism is a rare cause of pancreatic inflammatory disease. Appropriate treatment of coexistent hyperparathyroidism and pancreatitis, especially when complicated by pseudocyst formation, is unsettled. We describe two patients with primary hyperparathyroidism who developed pancreatitis associated with multiple pseudocysts. The largest cyst in each patient was 9 and 5 cm, respectively. After correction of hyperparathyroidism and normalization of serum calcium levels by removal of a parathyroid adenoma, the pseudocysts resolved in both patients, as documented with computed tomography. We conclude that uncomplicated pancreatic pseudocysts in patients with primary hyperparathyroidism can be treated expectantly. Surgical correction of hyperparathyroidism and normalization of serum calcium levels should precede pancreatic intervention when possible, since pseudocyst resolution is likely and the risks of postoperative hypercalcemia are avoided.

Plasmid labeling confirms bacterial translocation in pancreatitis.

To examine whether the gut is a source of infection in acute pancreatitis, bacterial translocation and alterations of intestinal microecology and morphology were studied in 16 dogs. Dogs were colonized with a strain of Escherichia coli (E. coli 6938K) bearing the plasmid pUC4K, which confers kanamycin resistance. In eight dogs (group I), pancreatitis was induced by sodium taurocholate/trypsin injection. Eight other dogs (group II) underwent laparotomy only. The pancreas, mesenteric lymph nodes, peritoneal fluid, liver, and spleen were harvested 7 days later for culturing and histologic analysis. Identification of E. coli 6938K was accomplished by plasmid DNA analysis. Group I dogs had severe pancreatitis and ischemic changes in small bowel mucosa. Group II dogs had no changes. Translocation to the pancreas occurred in five dogs and to mesenteric lymph nodes in six dogs with pancreatitis. No translocation occurred in group II dogs (p < 0.05). In addition to E. coli 6938K, other gram-negative kanamycin-resistant species were isolated, including E. coli (other than 6938K) and Enterobacter cloacae. Enteric origin of these strains was confirmed by antibiography and plasmid DNA analysis. No overgrowth of cecal gram-negative bacteria was found. This study suggests that the gut is a primary source of infection in pancreatitis and that ischemic damage of intestinal mucosa may promote bacterial translocation.