Long-term survival and patterns of failure after postoperative radiation therapy for subtotally resected rectal adenocarcinoma.

Between 1977 and 1984, 17 patients received external beam irradiation after subtotal resection of rectal carcinoma. Ten patients had microscopic residual disease and 7 had gross residual disease. In the group with microscopic residual disease, 4 had tumor cut through with pathologically involved margins, 5 had adjacent unresected structures that were biopsy positive, and 1 had tumor spillage into the pelvis. The patients with gross residual disease were noted by the surgeon to have visible tumor after maximal debulking. Nine of 17 cases had involved pelvic lymph nodes. Radiation was administered to the pelvis with 4, 6, or 10 MV photons. Doses ranged from 40 to 60 Gy, with a median dose of 50 Gy given at 1.8 to 2.0 Gy per fraction, 5 days per week. Three patients received bacillus Calmette-Guérin (BCG), 2 received 5-fluorouracil (5-FU), and 1 received hycanthone. Thirteen of the 17 patients (76%) experienced local failure and, of these, 10 also developed distant disease. No patients developed distant metastasis in the absence of local failure. Local control was achieved in 3 of 10 patients (30%) with microscopic residual and 1 of 7 (14%) with gross residual. Four of the 17 patients (24%) have remained free of disease for greater than 5 years. External beam irradiation is capable of producing long-term survival and local control in a minority of patients with rectal cancer after subtotal resection. Investigation of more aggressive forms of therapy such as the addition of intraoperative irradiation, brachytherapy, radiation dose modifiers, and chemotherapy is warranted.

Intraoperative irradiation for locally recurrent colorectal cancer in previously irradiated patients.

PURPOSE: Information in the literature regarding salvage treatment for patients with locally recurrent colorectal cancer who have previously been treated with high or moderate dose external beam irradiation (EBRT) is scarce. A retrospective review was therefore performed in our institution to determine disease control, survival, and tolerance in patients treated aggressively with surgical resection and intraoperative electron irradiation (IOERT) +/- additional EBRT and chemotherapy. METHODS AND MATERIALS: From 1981 through 1994, 51 previously irradiated patients with recurrent locally advanced colorectal cancer without evidence of distant metastatic disease were treated at Mayo Clinic Rochester with surgical resection and IOERT +/- additional EBRT. An attempt was made to achieve a gross total resection before IOERT if it could be safely accomplished. The median IOERT dose was 20 Gy (range, 10--30 Gy). Thirty-seven patients received additional EBRT either pre- or postoperatively with doses ranging from 5 to 50.4 Gy (median 25.2 Gy). Twenty patients received 5-fluorouracil +/- leucovorin during EBRT. Three patients received additional cycles of 5-fluorouracil +/- leucovorin as maintenance chemotherapy. RESULTS: Thirty males and 21 females with a median age of 55 years (range 31--73 years) were treated. Thirty-four patients have died; the median follow-up in surviving patients is 21 months. The median, 2-yr, and 5-yr actuarial overall survivals are 23 months, 48% and 12%, respectively. The 2-yr actuarial central control (within IOERT field) is 72%. Local control at 2 years has been maintained in 60% of patients. There is a trend toward improved local control in patients who received > or =30 Gy EBRT in addition to IOERT as compared to those who received no EBRT or <30 Gy with 2-yr local control rates of 81% vs. 54%. Distant metastatic disease has developed in 25 patients, and the actuarial rate of distant progression at 2 and 4 years is 56% and 76%, respectively. Peripheral neuropathy was the main IOERT-related toxicity; 16 (32%) patients developed neuropathies (7 mild, 5 moderate, 4 severe). Ureteral narrowing or obstruction occurred in seven patients. All but one patient with neuropathy or ureter fibrosis received IOERT doses > or =20 Gy. CONCLUSION: Long-term local control can be obtained in a substantial proportion of patients with aggressive combined modality therapy, but long-term survival is poor due to the high rate of distant metastasis. Re-irradiation with EBRT in addition to IOERT appears to improve local control. Strategies to improve survival in these poor-risk patients may include the more routine use of conventional systemic chemotherapy or the addition of novel systemic therapies.

Restorative proctocolectomy and ileal reservoir.

The operation consisting of abdominal colectomy, proximal proctectomy, distal mucosal proctectomy, and ileal pouch-anal anastomosis eliminates colorectal mucosal disease, restores transanal defecation, and avoids the need for a permanent abdominal stoma and the wearing of an external appliance. During the 4-year period from January 1981 through December 1984, 369 such operations were done at the Mayo Clinic for either ulcerative colitis (in 336 patients) or polyposis coli (in 33 patients). None of the patients died in the immediate postoperative period. Follow-up data for the initial 188 patients showed the following complications: anastomotic stricture in 12%, pelvic sepsis in 11%, obstruction of the small intestine in 9%, and reservoir ileitis in 7%. A permanent stoma had to be established in 5% of patients. All patients could defecate spontaneously, and 95% had satisfactory continence for stools and gas. This operation seems safe and effective and provides a quality of life superior to that seen after the conventional Brooke ileostomy.

Surreptitious use of salicylates: a cause of chronic recurrent gastroduodenal ulcers.

We describe five patients who had chronic recurrent gastroduodenal ulcers and pain, intestinal obstruction, bleeding, or a combination of these symptoms. Four patients required surgical intervention because of a poor response to medical therapy. The ulcers recurred in all patients, despite evidence of achlorhydria in two of them. Although the patients denied the use of salicylates, all of them had therapeutic blood levels of salicylates. A salicylate level should be determined in patients with severe ulcer disease that is resistant to medical therapy or that is recurrent after appropriate surgical therapy when the presence of hypergastrinemia or hyperchlorhydria has been definitely excluded. A low serum level of uric acid can also be a clue to the abuse of salicylates.

Strictureplasty for obstructive Crohn's disease: the Mayo experience.

BACKGROUND: Strictureplasty for obstructive Crohn's disease of the small bowel continues to gain favor throughout the world. Although the potential advantages of preserving intestinal length are obvious, the optimal clinical setting for performing strictureplasty remains to be determined. PATIENTS: Of 244 patients who underwent abdominal exploration for complications of Crohn's disease between Jan. 1, 1985, and Jan. 1, 1991, at the Mayo Clinic, 35 had a total of 71 strictureplasties. Concomitant resection of bowel with active disease was performed in 67% of the procedures. RESULTS: In this series, no perioperative deaths occurred, and no anastomotic leaks, enteric fistulas, or intra-abdominal abscesses were noted during a 3-year follow-up. The overall perioperative complication rate was 14%. Postoperatively, 33 of the 35 patients were able to resume enteral nutrition and discontinue medical treatments. The symptomatic recurrence rate at 3 years was 20%; six patients have required reoperation. CONCLUSION: These findings support the use of strictureplasty for isolated, quiescent, stenotic bowel lesions associated with Crohn's disease.

Vascular complications of inflammatory bowel disease.

During an 11-year period from January 1970 to December 1980, 7,199 patients at our institution had chronic ulcerative colitis or Crohn's disease. Thromboembolic complications developed in 92 (1.3%) of these patients. An additional 4 patients had cutaneous vasculitis, and 17 had an arteritis-associated diagnosis. Of the thromboembolic complications, 61 were deep vein thromboses or pulmonary emboli. The mortality among patients with thromboembolic complications was high (25%). Sixty percent of the patients had a thrombocytosis unaffected by sulfasalazine or corticosteroid therapy. In 73% of the patients, the erythrocyte sedimentation rate was increased, and when measured, fibrinogen and factor VIII were commonly elevated. Peripheral arterial thrombosis, coronary thrombosis, and mesenteric and portal vein thrombosis were predominantly postsurgical complications, but 77% of peripheral venous thromboses occurred spontaneously. The role of anticoagulation and surgical intervention in the management of hypercoagulation in patients with inflammatory bowel disease is discussed.

Primary closure and continuous irrigation of the perineal wound after proctectomy.

The aim was to assess the success of primary closure and continuous irrigation of the perineal wound in achieving wound healing after proctectomy. The frequency and degree of primary healing (dry wound within 30 days after removal of irrigation catheters) and the factors affecting the rate of healing were determined in 152 patients undergoing proctectomy for carcinoma (31 patients), Crohn's colitis (32 patients), or ulcerative colitis (89 patients). All patients were followed for at least 2 years. Primary healing occurred in 64.5% of patients, whereas it was delayed (dry wound by 90 days) or failed to occur in 30.2% and 5.3%, respectively. Of those with carcinoma, 87% had healing by 30 days, whereas only 61% and 50% of those with either ulcerative colitis or Crohn's disease, respectively, had healing in that time (P less than 0.02). Females had primary healing more often than males (70% versus 45%, P less than 0.01), and the incidence of persistent perineal sinus was greater in men (28%) than in women (13%) (P less than 0.05). Primary healing was adversely affected in men when drains exited centrally through the wound (31% healed primarily) as opposed to when they exited laterally (65% healed primarily, P less than 0.04). We conclude that primary closure with irrigation of the perineal wound is safe and provides satisfactory healing in most patients.

The continent ileostomy: a viable alternative.

The continent ileostomy is one of the alternative techniques in managing patients who require a proctocolectomy. Although previously plagued with up to 30% revisions, the continent ileostomy appears to contribute significantly to the social and psychologic well-being of these patients. Our experience suggests that improved surgical techniques make this procedure a safe and favorable alternative to the traditional Brooke ileostomy.

Menetrier's disease. Effect of medical and surgical vagotomy.

A 62-year-old woman with incapacitating atherosclerotic vascular occlusive disease was found to have severe hypoproteinemia as a result of Menetrier's disease. That diagnosis was suspected on roentgenographic and gastroscopic examinations and confirmed by examination of full-thickness surgical gastric biopsy specimens. The protein loss from the stomach was significantly decreased by treatment with trimethaphan and atropine and led us to perform proximal gastric vagotomy at the time of endarterectomy. Subsequent protein turnover studies suggest that there has been a significant decrease in protein loss by the stomach. Further evaluation of the possible role of medical and surgical vagotomy in this disease seems warranted.

Radical pancreatoduodenectomy--a procedure to be abandoned?

Radical pancreatoduodenectomy, or Whipple's procedure, has gradually fallen into disrepute during the past decade. This loss of popularity has resulted from excessive postoperative mortality and morbidity due to inherent technical difficulties, the dismal long-term survival when the procedure is done for adenocarcinoma of the pancreas, and the current enthusiasm for total pancreatectomy. Our objective was to assess the technical aspects of the Whipple procedure as they relate to subsequent morbidity and mortality. Conclusions are drawn from the evaluation of 146 consecutive patients who underwent this procedure between 1970 and 1979 for either benign (10 patients) or malignant (136 patients) diseases of the pancreas and periampullary region. The postoperative mortality was 4.1%. Significant morbidity occurred in 34% of patients, and early reoperation was required in 7.5% of these patients. The median postoperative hospital stay was 19 days. The relatively low postoperative mortality indicates that inherent technical difficulties of the Whipple procedure can be overcome and supports its continued use in carefully selected patients. The persistent high morbidity and the rather poor long-term results following treatment of certain pancreatic malignancies place individual surgical judgment and technical expertise at a premium when ever this classic operation is contemplated.

Care of patients and their families with familial adenomatous polyposis.

Familial adenomatous polyposis (FAP) is a dominantly inherited disorder that is typically characterized by the appearance of multiple colorectal adenomas usually by the teenage years, with a risk of early colorectal cancer approaching 100%. Genetic testing can help determine which family members have the disorder and require surveillance endoscopy. Astute physicians may detect unsuspected FAP in patients with extraintestinal manifestations such as hard or soft cutaneous tumors. Colectomy will prevent cancer but is often necessary before the patient is 20 years old. Postoperative lifelong surveillance is indicated to screen for associated duodenal, thyroid, and rectal or ileal neoplasms. Attenuated FAP variants are less typical and may be confused with other types of familial colorectal neoplasia. Chemoprevention, regression, and other treatment strategies are being developed to improve the management of extracolonic neoplasms and desmoid tumors. A better understanding of the natural history of these FAP-associated phenomena will facilitate the rational selection of interventions. Management guidelines that were recently developed at Mayo Clinic Rochester to provide for uniform care and surveillance are discussed.

Proximal gastric vagotomy. Initial experience.

Proximal gastric vagotomy was performed in 223 patients with chronic duodenal ulceration between 1973 and 1977. The follow-up ranged from 6 to 78 months with a mean of 39 months. There was no operative mortality. Adverse postoperative sequelae, including delayed gastric emptying, dumping, diarrhea, and reflux gastritis, were seen in less than 3% of patients. Ulcers recurred in 11 patients (4.9%). We conclude that proximal gastric vagotomy is an effective, safe, and satisfactory operation for chronic duodenal ulcer over the short term.

Changing referral patterns for surgical treatment of ulcerative colitis.

OBJECTIVE: To determine whether referral patterns for surgical treatment among patients with chronic ulcerative colitis have changed in recent years, especially in view of the introduction of a new operation, the ileal pouch-anal anastomosis (IPAA). MATERIAL AND METHODS: Between January 1976 and December 1986, 981 patients underwent proctocolectomy at Mayo Clinic Rochester. Brooke ileostomy (N = 300), colectomy with ileorectal anastomosis (N = 33), proctocolectomy with Kock pouch (N = 180), or ileal pouch-anal anastomosis (IPAA) (N = 468) was performed. The indications for surgical intervention were categorized as emergent or elective, the latter including intractability, cancer, and cancer prophylaxis. For analysis, the duration from diagnosis of disease to operation, indications for surgical treatment, and types of operation were subdivided into pre-IPAA era (before 1981) and post-IPAA era (from 1981 onward) for the entire group and for distant versus local patients. RESULTS: More continence-preserving operations were done in 1981 and thereafter (76%) than before 1981 (46%). In the later segment of the study period in comparison with before 1981, fewer operations were done for emergent reasons (4% versus 8%) and a greater percentage of operations were done for elective indications, especially intractability (74% versus 61%). With the advent of IPAA in 1981, patients underwent operation sooner after the diagnosis was made (7.4 years versus 8.6 years before 1981). A smaller proportion of patients underwent operation for cancer prophylaxis during 1981 through 1986 (19%), however, than before 1981 (28%). CONCLUSION: Referral patterns for surgical treatment of patients with ulcerative colitis have changed in recent years--patients are being referred for operation sooner, before complications develop that necessitate emergent procedures. Although the changed referral pattern may be due to the availability of IPAA, other factors may also have a role.

Comparison of surgical treatment of ulcerative colitis associated with primary sclerosing cholangitis: ileal pouch-anal anastomosis versus Brooke ileostomy.

OBJECTIVE: To compare the operative risks, operative complications, and late outcome of two homogeneous groups of patients with chronic ulcerative colitis (CUC) and primary sclerosing cholangitis (PSC) who underwent either Brooke ileostomy or ileal pouch-anal anastomosis (IPAA). MATERIAL AND METHODS: Between 1970 and 1990, 72 patients with CUC and PSC underwent proctocolectomy with either Brooke ileostomy (group I; N = 32) or IPAA (group II; N = 40). Postoperative data included operative mortality, need for blood transfusion, general postoperative complications, liver-related complications, and proctocolectomy-related complications. RESULTS: Eight group I patients and nine group II patients had a total of 12 and 11 general complications, respectively. Liver-related complications were diagnosed in 16% and 10% of group I and group II patients, respectively. Proctocolectomy-specific complications occurred in 34% of group I and 20% of group II patients. The overall need for blood transfusion was 94% in group I and 47% in group II (P < 0.001). The cumulative probability of proctocolectomy-related complications at 5 years was 23% for group I and 64% for group II patients (P < 0.002). The difference, however, was primarily due to the high frequency of pouchitis after IPAA, estimated at 57% at 4 years. The cumulative 5-year risk of liver-related complications was 37% and 28% for group I and group II, respectively. Peristomal varices and bleeding occurred in eight group I patients but in none of group II. CONCLUSION: Because IPAA avoids bleeding problems, it is the surgical treatment of choice in patients with PSC and CUC.

Endorectal ultrasonographic staging of rectal carcinoma.

Endorectal ultrasonography is a valuable imaging method for examination of the rectum and perirectal tissues. We assessed 50 patients with known rectal carcinoma prospectively by using a 7.0-MHz endorectal transducer to determine the depth of invasion of the rectal wall by tumor and the presence of lymphadenopathy. Tumors were staged by using the Astler-Coller modification of the Dukes staging system, and the results were compared with histologic staging of the surgical specimen. Ultrasonography had an accuracy of 80%, a sensitivity of 92%, and a specificity of 76% for detection of invasion of the perirectal fat. Ultrasonography was sensitive in the detection of perirectal lymphadenopathy but was not specific in distinguishing benign from malignant nodes.

Effect of proximal gastric vagotomy on gastric acid hypersecretion and hypergastrinemia after massive small bowel resection in dogs.

Fasting and postprandial gastric acid secretion and gastrin were determined in Heidenhain pouch dogs before (C) and after (R) massive enterectomy (five dogs) and after additional proximal gastric vagotomy (PGV) in three dogs. Basal acid output was unchanged (C = 7 +/- 2 microneq/hour; R = 11 +/- 4 microneq/hour), but postprandially the hourly (third through eighth hour) and cumulative acid outputs (C = 3.6 +/- 0.3 mEq/8 hour; R = 7.2 +/- 0.4 MEq/8 hour) were significantly increased by resection (P less than 0.001). Similarly, fasting gastrin was unaltered by resection (C = 41 +/- 2 pg/ml; R = 46 +/- 8 pg/ml), whereas hourly gastrin concentrations significantly (P less than 0.05) exceeded control values. Increased gastrin correlated linearly (P less than 0.02) with increased acid output. After PGV, basal acid levels remained unchanged (R = 4 +/- 2 micronEq/hour; PGV = 9 +/- 4 micronEq/hour), but postprandial acid output significantly exceeded prevagotomy values at all time periods (P less than 0.05). Cumulative acid output also was increased (R = 6.8 +/- 0.6 MEq/8 hour; PGV = 11.2 +/- 0.6 mEq/8 hour; P less than 0.001). Serum gastrin, however, remained unchanged. Hypersecretion of acid from Heidenhain pouches after massive enterectomy is increased further by PGV without a concomitant increase in serum gastrin.

Closure of abdominal incisions with subcutaneous catheters.

Five hundred patients who underwent a variety of major abdominal operations had primary closure of their abdominal incisions. The technique of closure included the use of catheters in the subcutaneous space with suction, except when an antibiotic solution was being instilled every 8 hours of the first 9 days after operation. The rate of postoperative wound infection in these 500 patients, a group which included 140 patients who had contaminated or dirty wounds, was 0.8 percent. Infection developed in only one (0.47 percent) of the 211 patients who underwent major colon procedures. No significant side-effects of treatment were observed.

Prognostic factors in pancreatic abscess.

Pancreatic abscess is a highly lethal sequela of pancreatitis. The purpose of this study was to identify clinical and laboratory factors associated with the antecedent episode of pancreatitis or at the time of manifestation of a pancreatic abscess and factors related to its treatment that might have prognostic significance. The records of 66 consecutive patients with pancreatic abscess seen between 1966 and 1980 were reviewed. Factors adversely affecting survival included: severity of precipitating pancreatitis (p less than 0.03); sepsis and pulmonary dysfunction (p less than 0.05); and persistent postoperative sepsis (p less than 0.001). All four patients who were treated nonsurgically died. Of 62 patients treated surgically, 80% experienced serious complications, 31% required reoperations, and 40% died. No single surgical procedure, including marsupialization, proved to be superior. Surgical drainage offers the only possibility of survival for patients with pancreatic abscess. More thorough elimination of all focuses of infection is required, however, to reduce the high morbidity and mortality rates associated with persistent postoperative sepsis.

Hartmann's procedure. Its use with complicated carcinomas of sigmoid colon and rectum.

We assessed the safety of Hartmann's procedure in terms of postoperative morbidity and mortality associated not only with the original operation, but also with reestablishment of colorectal continuity and long-term survival. We studied 107 consecutive patients in whom Hartann's procedure was performed either electively (96) or as an emergency (11) between 1970 and 1975. In 87, the resection extended below the peritoneal reflection. The mean age of the patients was 67 years. The principal indication for Hartmann's procedure was palliation (53%), obstruction (42%) either alone or with perforation, and technical difficulties (5%). Four patients died (3.7%) in the immediate postoperative period, two after elective surgery (2.1%). Complications were observed in 35 patients, including wound infection or dehiscence (22), urinary tract problems (seven), and pelvic abscess (three). Colorectal continuity was reestablished in ten patients without mortality or morbidity. The mean five-year survival for Dukes' B, C, and D lesions was 54%, 23%, and 3%. Hartmann's procedure, when performed for complicated carcinomas of sigmoid colon and rectum, including subsequent restoration of bowel continuity, is safe and can be associated with long-term survival.

Prognosis of symptomatic duodenal adenomas in familial adenomatous polyposis.

The frequent association between familial adenomatous polyposis and duodenal tumors is increasingly recognized, yet many patients do not benefit from adequate diagnosis and follow-up of upper gastrointestinal polyps. A retrospective review of 14 patients with duodenal tumors associated with familial adenomatous polyposis was undertaken to assess the impact of early diagnosis by screening asymptomatic patients. Six of eight patients presenting with symptoms suggesting duodenal disease had invasive cancer. Four of these six patients died after surviving a mean of 13 months after diagnosis. In contrast, none of the six patients diagnosed after screening with upper gastrointestinal endoscopy has had invasive carcinoma. Early diagnosis and long-term surveillance of asymptomatic patients with familial adenomatous polyposis affords the opportunity to diagnose and treat duodenal tumors at an early stage, thereby, avoiding the dismal prognosis once invasive cancer has developed.