Colorectal cancer: is the surgeon a prognostic factor? A systematic review.

OBJECTIVE: To assess whether there is evidence that the surgeon is a prognostic factor in the treatment of colorectal cancer. DATA SOURCE: MEDLINE 1985-February 1999, and bibliographies of retrieved articles. STUDY SELECTION: Publications which analysed the outcome of patients with colorectal cancer and in which one of the variables analysed was the surgeon. RESULTS: Thirteen studies were identified which addressed the outcome measures: post-operative mortality, anastomotic leak rate, local recurrence rate, and long-term survival. For these outcomes, different surgeons achieve significantly different results, with experienced and specialist surgeons achieving significantly better results than other surgeons. CONCLUSION: The current data strongly suggest that the surgeon is an important prognostic factor in the treatment of colorectal cancer.

Five-layer repair of rectovaginal fistula using a vaginal approach. A case report.

We present a case of a rectovaginal fistula which was revealed as an incidental finding at the time of posterior colporrhaphy. We describe a previously unreported 5-layer repair through a vaginal approach in preference to the more frequently reported approaches of endoanal flap or conversion to a fourth degree tear. The diagnosis and management of rectovaginal fistulas is discussed.

J ileal pouch-anal anastomosis for chronic ulcerative colitis: complications and long-term outcome in 1310 patients.

AIM: The purpose of the study was to determine the risk of postoperative complications and the functional outcome after a hand-sewn ileal pouch-anal anastomosis (IPAA) for ulcerative colitis using a single J-shaped pouch design. METHODS: Preoperative function, operative morbidity and long-term functional outcome were assessed prospectively in 1310 patients who underwent IPAA between 1981 and 1994 for ulcerative colitis. RESULTS: Three patients died after operation. Postoperative pelvic sepsis rates decreased from 7 per cent in 1981-1985 to 3 per cent in 1991-1994 (P = 0.02). After mean follow-up of 6.5 (range 2-15) years, the mean number of stools was 5 per day and 1 per night. Frequent daytime and nighttime incontinence occurred in 7 and 12 per cent of patients respectively, and did not change over a 10-year period. The cumulative probability of suffering at least one episode of 'clinical' pouchitis was 18 and 48 per cent at 1 and 10 years and the cumulative probability of pouch failure at 1 and 10 years was 2 and 9 per cent respectively. CONCLUSION: These results indicate that increased experience decreases the risk of pouch-related complications and that with time the functional results remain stable, but the failure rate increases.

Carcinoid tumor of the rectum. DNA ploidy is not a prognostic factor.

PURPOSE: This study was designed to evaluate the clinical characteristics, surgical treatment, and outcome of carcinoid tumors of the rectum and to assess flow cytometry deoxyribonucleic acid (DNA) analysis as a potential prognostic factor for management of these tumors. METHODS: Medical records, tumor registry database, and pathology slides were retrospectively reviewed. Flow cytometry DNA analysis was performed on archived specimens. RESULTS: One hundred nine patients with rectal carcinoid tumors underwent surgery between 1962 and 1987. Follow-up was available in 86 patients for a mean period of 12 years. Of 100 patients with tumors less than 2 cm, only one with a 1.5 cm ulcerated tumor developed liver metastases. Of nine patients with a tumor more than or equal to 2 cm, three with known liver metastases underwent rectal biopsy only, and three had rectal biopsy and laparotomy with biopsy of liver metastases. Three patients underwent radical resection. Following abdominoperineal resection, one patient died with local recurrence after 5 years, and one developed hepatic recurrence after 5.5 years and died at 9 years. One patient with coloanal anastomosis developed local and hepatic metastases seven years after surgery and died at ten years. No patients developed carcinoid syndrome. DNA ploidy did not correlate with metastases at presentation or recurrence of carcinoid tumor. CONCLUSION: Radical resection of rectal carcinoids with ulceration or size greater than or equal to 2 cm is associated with a poor prognosis; however, survival may be long term, even in the presence of metastatic disease. DNA ploidy does not appear to be a useful prognostic factor for rectal carcinoid tumors.

Rectal mucosal electrosensitivity - what is being tested?

The results of rectal mucosal electrosensitivity (RME) testing have been used to support theories regarding the aetiology of both idiopathic constipation and bowel dysfunction following rectopexy. The aim of this study was to assess the validity of tests of RME. Sixty-eight patients, comprising three groups (group 1: 50 patients undergoing assessment in the Anorectal Physiology Unit, group 2: 10 patients with coloanal or ileoanal anastomosis, group 3: 8 patients with a stoma) underwent mucosal electrosensitivity testing, with the threshold stimulus required to elicit sensation being recorded. In addition the RME was measured in groups 1 and 2 when placing the electrode, mounted on a catheter with a central wire, against the anterior, posterior, right and left rectal or neorectal walls. To asses the influence on this test of loss of mucosal contact due to faeces, a further 8 cases with a normal rectum had RME performed with and without a layer of water soaked gauze around the electrode to stimulate faeces and prevent the electrode from making contact with the rectal mucosa. There was marked variance in the sensitivity of the different regions of rectal wall tested (P < 0.001). In group 1 patients the mean sensitivities were: central 36.6 mA, anterior 27.4 mA, posterior 37.9 mA, right 22.3 mA and left 25.6 mA. This circumferential variation suggests that the pelvic floor rather than rectal mucosa was being stimulated. All patients in group 2 had recordable sensitivities, and the mean sensitivity threshold was significantly higher than group 1 patients in the central (P = 0.03), right (P = 0.03) and left (P = 0.007) positions. In group 3 the sensitivity was greater within the stoma at the level of the abdominal wall muscle than intra-abdominally or subcutaneously, again suggesting an extra-colonic origin of the sensation. The sensitivity threshold was significantly greater with the electrode wrapped in gauze (P < 0.01), and loss of mucosal contact was not detected by the EMG machine. Therefore RME testing would seem not to measure mucosal sensitivity, and is probably influenced by the presence of faeces.

Twelve-year experience with the long intestinal tube.

A number of surgical methods, including use of the long intestinal tube, have been designed to decrease the incidence of recurrent small bowel obstruction. The aim of the present study was to review the indications, morbidity, and long-term results of the long intestinal tube at the Mayo Clinic. During the 12-year period 1981-1992, 47 patients had such tubes placed. The patients formed a complex surgical group: 46 patients had previously undergone at least one laparotomy (median 4, range 1-10); 41 patients had been hospitalized at least once for small bowel obstruction (median 3, range 1-15); and all 41 of these patients had undergone at least one previous laparotomy for obstruction (median 2, range 1-7). Eleven patients had a history of inflammatory bowel disease, and eight had a history of irradiation. In addition to dense adhesions in 46 patients, operative findings included large bowel tumors in six patients, intraperitoneal carcinomatosis in four, intraabdominal abscess in four, and small bowel stricture in three. Twenty patients required either a small bowel or large bowel resection, and three had a stoma fashioned. Only one case of morbidity (tube retraction) was related to tube placement. Among the 36 patients with complete follow-up, nine patients developed episodes of recurrent adhesional small bowel obstruction after a mean follow-up of 48 months, although only one required laparotomy. Of the remaining 11 patients it is known that two developed small bowel obstruction, one of whom required laparotomy.(ABSTRACT TRUNCATED AT 250 WORDS)

Autonomic nerves cross the posterior plane of rectal dissection.

This investigation was an anatomical study to determine whether branches of the pre-sacral autonomic plexus cross the posterior plane of surgical dissection to supply the rectum. Initially four cadaver hemi-pelves were dissected. Twelve patients undergoing full rectal mobilization were then studied at operation. In all subjects the pre-sacral nerves were arranged as a plexus below the sacral promontory, rather than as individual left and right nerve trunks. Structures thought to be nerves were identified crossing the plane of posterior mobilization of the rectum. They were traced towards their origin and destination, photographed and representative fibres biopsied. A total of 42 such structures were biopsied (16 in cadavers, 26 in operative cases) and 40 were confirmed to be nerves. These nerves connect the pre-sacral autonomic plexus with the posterior aspect of the rectum and were found at all sacral levels. In the operative cases the level of the positive biopsies were S1-six, S2-six, S3-five, S4-four, S5-three. The posterior plane of rectal dissection is therefore crossed by autonomic nerves that innervate the rectum. The pre-sacral nerves have been found to form a plexus in all subjects.

Scintigraphic assessment of colonic function during defaecation.

The aim of this study was to develop a method by which rectal and colonic activity could be examined during defaecation under physiological conditions, in order to evaluate whether the colon plays a role in defaecation. Subjects presented to the Nuclear Medicine department on the day following ingestion of oral In-111 labelled DTPA, when they developed the normal urge to defaecate. Defaecation took place in a private room while dynamic scintigraphy of the rectum and colon was recorded. Fourteen subjects were studied (8 normal subjects, 4 with constipation, 2 with irritable bowel syndrome). In 13 subjects the left colon was visualized during defaecation and emptying was clearly observed in 12. The right colon was visualised in 11 subjects and emptying was seen in 7. Mean percentage segmental evacuation was right colon 20%, left colon 32% and rectum 66%. Colonic emptying occurs during defaecation, which is not a process of rectal evacuation only. This has implications for the understanding of the pathophysiology of obstructed defaecation.

The long-term effect of adjuvant postoperative chemoradiotherapy for rectal carcinoma on bowel function.

OBJECTIVE: The authors assessed the long-term effect of postoperative chemoradiotherapy on bowel function. SUMMARY BACKGROUND DATA: Adjuvant postoperative radiation therapy, often combined with chemotherapy, is being used increasingly often for rectal carcinoma. However, the long-term effect of this treatment on bowel function has not been investigated. METHODS: The records were reviewed of all patients undergoing anterior resection for rectal carcinoma 2 to 5 years previously. During this period, patients with Astler-Coller stage B2 or C tumors generally were given postoperative radiation therapy with chemotherapy, whereas those with earlier stage tumors were not. To minimize possible confounding factors that may have been more common in the group receiving chemoradiotherapy and that may affect bowel function, extensive exclusion criteria were used, such as invasion of contiguous organs, local or distant metastases, use of a dysfunctioning stoma, and anastomotic or pelvic complications. One hundred remaining patients were suitable for inclusion in the study and participated in a telephone questionnaire; 41 patients had postoperative chemoradiotherapy, and 59 did not. RESULTS: The two groups were well matched for sex, level of anastomosis, and length of follow-up, although the group receiving chemoradiotherapy was slightly younger. The group that had chemoradiotherapy had more bowel movements per day than the group that did not have radiation therapy (median 7 vs. median 2, p < 0.001); the former group had "clustering" of bowel movements more often (42% vs. 3%, p < 0.001), had nighttime movements more often (46% vs. 14%, p < 0.001), had occasional or frequent incontinence more often (39% and 17% vs. 7% and 0%, p < 0.001), wore a pad more often (41% vs. 10%, p < 0.001), and were unable to defer defecation for more than 15 minutes more often (78% vs. 19%, p < 0.001). The group that had chemoradiotherapy also had stool of liquid consistency, used antidiarrheal medications, had perianal skin irritation, were unable to differentiate stool from gas, and needed to defecate again within 30 minutes of a movement significantly more often than the group that did not receive chemoradiotherapy. CONCLUSION: Adjuvant postoperative chemoradiotherapy for rectal carcinoma has a major long-term detrimental effect on bowel function.

Right hemicolectomy with a linear cutting stapler.

PURPOSE: The aims of this study were to describe and evaluate a new method of performing a stapled, functional end-to-end anastomosis following right hemicolectomy. METHODS: The widely patent anastomosis is constructed by two "firings" of a nondisposable linear cutting stapler. The medical records and follow-up details of all patients undergoing this procedure were reviewed. RESULTS: This technique, which can be performed quickly and safely, has been used in 42 patients. There have been no instances of short-term or long-term anastomotic complications. CONCLUSION: The initial experience with this simplified technique of stapled anastomosis following right hemicolectomy has been favorable.

Does colonoscopic polypectomy reduce the incidence of colorectal carcinoma?

The study's objective was to examine whether there is evidence that colonoscopic polypectomy reduces the incidence of colorectal cancer. The records of all patients who underwent colonoscopic polypectomy by a single surgeon between 1974 and 1991 were reviewed. Patients with colorectal cancer diagnosed at the initial colonoscopy, with a history of colorectal cancer, inflammatory bowel disease or familial adenomatous polyposis or with only hyperplastic polyps were excluded. There were 1008 remaining patients, of whom 645 have attended at least one follow-up colonoscopic examination, and these 645 patients from the basis of the study, because the incidence of cancer is known exactly in this group. The mean period of follow up was 4.4 years and the mean number of follow-up colonoscopic examinations was 2.2. There was a total of 2847 person-years of colonoscopic follow up. The expected incidence of cancer, age and sex adjusted, is calculated using Australian epidemiological figures. The observed incidence of cancer was 3 cases (all asymptomatic) per 2847 person-years, which is indistinguishable from the general population's risk of 3.75 cases per 2847 person-years. Analysis of previous publications suggests that patients with adenomas are at an increased risk of developing colorectal cancer of about 2.5 times the general population's risk. If correct, then the observed incidence of 3 cases per 2847 person-years is less than the expected incidence of 9.4 cases per 2847 person-years. This analysis suggests colonoscopic polypectomy does reduce the incidence of colorectal cancer.

Incidental cholecystectomy during colorectal surgery.

OBJECTIVE: To assess the risks and benefits of incidental cholecystectomy in patients having colorectal surgery. SUMMARY BACKGROUND DATA: Cholelithiasis is found commonly during abdominal surgery. Previous studies used disparate methods to assess the risks and benefits of incidental cholecystectomy and have reached contradictory conclusions. METHODS: All patients in whom asymptomatic cholelithiasis was noted during colorectal surgery between January 1982 and December 1986 were studied. Operative morbidity and long-term outcome were assessed by chart review and questionnaire. RESULTS: Three hundred five patients were identified, of whom 195 (63.9%) had an incidental cholecystectomy and 110 (36.1%) did not. The two groups were similar in terms of age, sex, primary disease, and associated medical conditions, although fewer emergency procedures, abdominoperineal resections, and Hartmann's procedures were needed in the cholecystectomy group. The overall operative morbidity rate was the same in both groups. The long-term risk for developing small bowel obstruction was also similar. After a median follow-up of 6 years after hospital discharge, biliary pain or cholecystitis developed in 16 patients (14.6%) in the "no cholecystectomy" group, 12 of whom have had cholecystectomy. Two additional patients had cholecystectomy for acute postoperative cholecystitis while still in the hospital. Six more patients have had incidental cholecystectomy at subsequent laparotomies. The cumulative probability of needing cholecystectomy at 2 and 5 years after the initial colorectal operation was 12.1% and 21.6%, respectively. CONCLUSIONS: Incidental cholecystectomy was not associated with increased postoperative morbidity, whereas the long-term risk that previously asymptomatic gallstones would become symptomatic was substantial. Unless there are clear contraindications, patients with asymptomatic gallstones who have colorectal surgery should have concomitant cholecystectomy.

Anal lymphoma presenting as a perianal abscess.

A case of perianal abscess secondary to anal lymphoma affecting a human immunodeficiency virus negative patient is presented. Causes of perianal abscess other than anal glandular infection are important, although the proportion of abscesses reported to be secondary to these other causes is found to vary widely in the literature. A biopsy of the wall of an anorectal abscess should be taken when suspicion of an unusual cause arises.

Bisacodyl reduces the volume of polyethylene glycol solution required for bowel preparation.

PURPOSE: This study was designed to compare the efficacy and patient tolerance of standard orthograde bowel preparation using 4 liters of polyethylene glycol solution with only 2 liters of polyethylene glycol preceded by the stimulant laxative bisacodyl. METHODS: Three hundred eighty-two consecutive patients undergoing outpatient colonoscopy under two surgeons over a six-month period were randomized to receive one of the two preparations. Patients were asked to record the effects of the preparation and to give it a "discomfort rating" on a scale from 1 to 5. Surgeons were blinded to the preparation used and rated the quality of bowel cleansing on a scale of 1 to 5. One hundred ninety-one patients were randomized to the 4-liter preparation and 191 to the 2-liter preparation. RESULTS: Ninety-three percent of patients in the 2-liter group drank all of the solution as opposed to only 66 percent of patients in the 4-liter group. Patients in the 4-liter group gave the preparation significantly lower comfort scores than those in the 2-liter group (Fisher's exact test; P = 0.0001). The quality of bowel cleansing was not significantly different between the groups (P = 0.88). A total of 6.8 percent of the 2-liter group and 9.4 percent of the 4-liter group were considered by the surgeon to have had a poor preparation (rating 4 or 5). The quality of the preparation was found to correlate with the patients' age and compliance with the preparation regimen, and was found not to correlate with presenting symptoms, pathology, or a previous colonic resection. CONCLUSION: Bowel preparation with bisacodyl and 2 liters of polyethylene glycol is more acceptable to patients than a 4-liter regimen and is equally effective in cleansing the colon.

Alternative surgical procedures for primary rectal carcinoma.

This review summarizes the recent literature on alternative surgical procedures for treating rectal carcinoma. After a brief examination of publications on preoperative staging, current reports on alternative surgical procedures, from local excision to anterior resection and extended resection, are discussed. Although radiotherapy and chemotherapy are assuming increasingly prominent roles in the management of rectal carcinoma, detailed examination of these therapies is beyond the scope of this review.

The cough response of the anal sphincter.

Rises in intra-abdominal pressure are common and are accompanied by reflex contraction of the external sphincter. Voluntary contraction of the sphincter is an uncommon event but is routinely used as a measure of sphincter strength. Cough pressure and squeeze pressure were compared in 75 patients using a 4 channel perfused catheter (3 anal and 1 rectal side-hole). Maximum anal cough pressure was higher than squeeze pressure (mean 158 vs 133 cm H2O, P = 0.0015). Intra-individual variance was less using cough pressure (mean 20% vs 29%, P = 0.005). There was significant overall correlation between cough pressure and squeeze pressure (P < 0.001) although in some cases there were wide differences, suggesting that use of both cough and squeeze pressure in manometry will assess sphincter strength more reliably. Measurement of the recto-anal pressure gradient during coughing correlated significantly better with degree of incontinence than cough or squeeze pressure (P = 0.005). The presence of a positive gradient was 100% specific for incontinence but the sensitivity was only 43% suggesting that factors other than simple mechanical sphincter weakness are involved in incontinence. Measurement of cough pressure has both clinical and research importance and should be added to standard manometric protocols.

Non-operative treatment of small bowel obstruction following appendicectomy or operation on the ovary or tube.

This study examined the hypothesis that small bowel obstruction in patients whose only previous laparotomy was for appendicectomy or operation on the ovary or tube is unlikely to resolve with non-operative management. Results of 330 admissions for small bowel obstruction were examined. In 40 cases the only previous laparotomy was at appendicectomy or tubo-ovarian operation; in 38 (95 per cent) of these division of adhesions was undertaken compared with 154 (53 per cent) of the remaining 290 (P < 0.00001). In the former group band adhesions were commoner (86 versus 45 per cent, P < 0.00001), bowel resection was required more frequently (22 versus 10 per cent, P = 0.02) and it was considered safe to give a trial of non-operative management less often (60 versus 85 per cent, P = 0.0004). A trial of conservative management may be unsafe or not worth while in patients with obstruction following earlier appendicectomy or operation on the ovary or tube.