The management of suspected tumours of the colon: the role of colonoscopy in general surgery.

The role of colonoscopy in the management of 100 cases of suspected large bowel tumour is described. Forty out of 47 patients with an equivocal barium enema examination were satisfactorily examined and the possibility of a tumour being present was either confirmed or refuted. Examination of 30 patients with otherwise unexplained blood loss from the bowel revealed the cause in 50 per cent of them. Twenty-nine patients had pedunculated polyps which were snared and removed, thus avoiding laparotomy and colotomy. The problem of distinguishing tumours from other lesions or retained faeces by radiology arises most often in the sigmoid colon. This segment is easily examined using a short flexible endoscope. General surgeons taking up colonoscopy will find a 1-m single-channel instrument convenient to use and capable of reaching approximately 90 per cent of colonic tumours that require further investigation or endoscopic resection. Facilities for the examination of the whole colon should be available in established endoscopy units.     

Increasing incidence of right-sided lesions in colorectal cancer.

One hundred ninety-eight patients with 211 cancers of the colon and rectum underwent elective resection at the University of Vermont College of Medicine during the five year period 1971 through 1975. Analysis of this series demonstrated that 35 per cent of all cancers were located in the cecum and ascending colon, that a similar percentage were classified as Dukes' A cancers, that a synchronous cancer was present in 5.5 per cent of the patients and that diagnosis by rectal examination and sigmoidscopy was possible in only 32 per cent of the patients. Comparison of these results with published data during the past thirty years indicate that there is an increasing incidence of carcinoma of the right colon with an associated decrease in the incidence of carcinoma of the sigmoid colon and rectum. It is recommended that patients be screened by examination of the stool for occult blood rather than by rectal examination and sigmoidoscopy so that these proximal lesions can be diagnosed at an earlier stage. Preoperative evaluation of patients with distal colorectal cancer should include double contrast barium enema examinations and colonscopy to rule out synchronous right-sided lesions.     

Effect of preoperative colonoscopy on the incidence of synchronous and metachronous neoplasms

Colonoscopy and air-contrast barium enema performed preoperatively in 389 patients with colorectal cancer revealed synchronous cancer in 4% and polyp in 14%. Nine of the 16 synchronous cancers were located in other surgical segments than the index cancer, and six of the nine were in stage A or B1. Of the 54 synchronous polyps, 28 were located in such other segments. Half of the synchronous cancers and almost half of the synchronous polyps were missed at double-contrast barium enema. All synchronous cancers and three-fourths of the synchronous polyps were detected at colonoscopy. No patient with preoperative colonoscopy presented with metachronous cancer within 3 years from surgery, and only two were subsequently found to have adenocarcinoma arising from an adenomatous polyp. Endoscopic polypectomy was performed in 21 cases during follow-up. Extensive use of preoperative colonoscopy is recommended in the evaluation of colorectal cancer, in order to promote detection of synchronous tumors, reduce the incidence of 'early metachronous' cancer and avoid malignant degeneration of adenomatous polyp.     

Synchronous and 'early' metachronous carcinomas of the colon and rectum

In a review of cases of colorectal cancer presenting to St. Mark's Hospital over the 16-year period 1970-85, 59 patients were found to have a synchronous carcinoma (3.4 per cent). Although 82 per cent of these synchronous tumours were distal to the splenic flexure (and hence within reach of a 60 cm flexible sigmoidoscope) only 42 per cent were detected pre-operatively: the remaining tumours were noted at surgery (24 per cent) or found incidentally on pathological examination of the resected specimen (34 per cent). Histological examination of these synchronous lesions revealed a high proportion with favourable stage (Dukes' A-75 per cent) and grade (well or moderate differentiation-90 per cent). Over the same 16-year period, 10 patients presented with an 'early' metachronous lesion (less than 3 years from initial surgery). Review of these cases noted negative findings on the initial barium studies in four patients and a failure to conduct full examination of the colon at initial presentation in the remaining six. It is concluded that full examination of the colon in all patients presenting with primary colorectal cancer is mandatory and that, in the light of this experience and recent reports in the literature, this should be by pre- or peroperative colonoscopy.     

Intraoperative colonoscopy in patients with colorectal cancer.

Sixty-seven patients underwent intraoperative colonoscopy during elective surgery for colorectal cancer. Complete examination of the colon was achieved in 65 patients (97 per cent), albeit with insertion through a colotomy in three (4 per cent). A synchronous carcinoma was found in six patients (9 per cent), which necessitated a change of planned surgical procedure. Synchronous polyps were detected and removed in 24 patients (36 per cent); two had polyps with carcinoma in situ. The mean age of patients with synchronous carcinoma was significantly higher than that of those without (74.1 versus 61.2 years, P = 0.02). Intraoperative colonoscopy took a mean of 15 min surgical time and only two minor complications (serosal lacerations) were encountered. In patients with colorectal cancer, intraoperative colonoscopy allows complete assessment of the colon and identifies synchronous lesions.     

The effect of preoperative endoscopy on recurrence and survival following surgery for colorectal carcinoma

Numerous studies have elucidated the benefits of endoscopy before surgery for carcinoma of the colon and rectum. In patients with known colon cancer, the incidence of synchronous colon cancers is 1.5 to 7.6 per cent and synchronous colon polyps is 25 to 40 per cent. Standard barium contrast studies are inferior to endoscopic examination in detecting these synchronous lesions. Endoscopy has been shown to alter the planned surgical procedure in 11 to 13 per cent of patients with colorectal cancer. Nevertheless, some authors avoid preoperative endoscopy because of concern that neoplastic cells may be seeded throughout the colon during the examination. They fear that manipulation of the tumor may promote hematogenous or lymphatic spread. Our study seeks to demonstrate whether this concern is valid by comparing rates of local recurrence, distant metastases, and survival between patients who have undergone preoperative endoscopy with those who have not.     

Is total colonic evaluation for anal cancer necessary?

The purpose of this study was to evaluate the necessity of total colonic evaluation in patients diagnosed with anal cancer. We reviewed the cases of 69 patients treated for neoplasm of the anus from January 1973 through November 1998. Patients underwent a barium enema, flexible sigmoidoscopy or colonoscopy; findings from these procedures were analyzed. Anal pathology included: squamous cell carcinoma, 48 patients (72%); adenocarcinoma, 6 patients (9%); and squamous cell carcinoma in situ, 10 patients (15%). Chemoradiation was completed in 41 patients (61%), and 6 patients (9%) underwent abdominal perineal resection. Wide excision was the primary therapy in 15 patients (22%). Procedures included: colonoscopy, 31 patients (46%); flexible sigmoidoscopy only, 15 patients (22%); barium enema only, 3 patients (5%); and a combination of flexible sigmoidoscopy and barium enema, 18 patients (27%). Eighty-five per cent of patients had a normal colonic evaluation. Ten patients (15%) had single or multiple polyps identified. Six adenomatous polyps and two hyperplastic polyps were found. No synchronous colorectal malignancy was identified. We conclude that colonoscopy is an integral part of colonic surveillance in patients diagnosed with anal carcinoma, but our study fails to find evidence to support such intervention. The standard screening recommendation for colorectal cancer based on age and risk factors is appropriate intervention for patients diagnosed with anal cancer; anal cancer itself does not appear to be one of these risk factors.     

Is preoperative colonoscopy necessary in carcinoma of the colon and rectum?

The incidence of synchronous polyps of the colon has been shown to be 25 to 40 per cent and the incidence of synchronous carcinomas to be two to eight per cent. Because of this, many surgical groups now advocate routine preoperative colonoscopy on all patients with colon carcinoma. The possibility of spreading and implanting tumor cells with the colonoscope has prompted the authors to purposely avoid preoperative colonoscopy and then clear the colon of any missed lesions with an early postoperative colonoscopy. This study is a retrospective review of 104 patients who have undergone a partial colectomy for colon and rectal carcinoma followed by a postoperative colonoscopy between June 1982 and June 1986. The purpose is to determine the adequacy of intraoperative palpation to detect synchronous neoplasms, and to further define the role of perioperative colonoscopy. The results of 34 per cent synchronous polyps and 5.8 per cent synchronous carcinomas parallels previous studies. The six patients with synchronous carcinomas were discovered by intraoperative palpation and the operation was modified in four of the six patients. Postoperative colonoscopy revealed polyps in 20 per cent of the patients, but all of these were amenable to snare polypectomy or electrocoagulation. No carcinomas were overlooked by palpation. It is our conclusion that intraoperative palpation is adequate for detection of synchronous carcinomas and therefore the risk and expense of preoperative colonoscopy can be avoided. Early postoperative colonoscopy, however, is imperative to clear the colon of small polyps which have the potential to progress to carcinoma.     

An assessment of occult blood testing to determine which patients with large bowel symptoms require urgent investigation

One hundred and fifty-two consecutive patients with symptoms suggestive of colorectal disease were offered occult blood testing before undergoing barium enema examination or colonoscopy; one hundred and thirty-nine successfully completed the test. Thirty-four had positive results of whom thirteen had a cancer and eight an adenomatous polyp (diagnostic yield for neoplasia of 59 per cent). No false negative results occurred, a sensitivity of 100 per cent, and only 21 false positives occurred, a specificity for malignancy of 84 per cent. Subjects attending outpatients should be offered occult blood testing; those with a positive test should undergo colonoscopy. The cost-benefit of such a scheme is emphasized.     

Colonoscopy as a routine preoperative procedure for carcinoma of the colon

Ninety patients with cancer of the colon who had total colonoscopy in the perioperative period have been reviewed. Almost half of the examinations revealed positive findings, three being unsuspected synchronous carcinomas in an area that would not have been resected with the proposed primary cancer. In addition, 79 polyps in 36 patients, the majority of which were also undetected by barium enema, were found and removed at colonoscopy. Thus, patients with carcinoma of the colon, in view of its tendency to be associated with synchronous polypoid disease, should have colonoscopy in the perioperative period. Whenever possible, this should be carried out preoperatively to confirm the diagnosis, to remove suspected or unsuspected polyps, and to detect unsuspected synchronous carcinoma.     

Colonoscopy after curative resection of colorectal cancer

Colonoscopy is generally considered to be an important part of the follow-up program for patients who have undergone curative resection of colorectal cancer. However, there are few data available concerning the frequency with which colonoscopy should be performed and for what length of time after operation. Since 1978, our policy has been to examine the colon annually in these patients using colonoscopy alternating with barium enema. We have evaluated the results in 100 patients over a four-year period. Based on size and histology, the significant colonoscopic findings were new colon cancers in three patients and 11 polyps demonstrating increased risk for malignancy in nine patients. This represents an interval yield of 3% per year. From these results and other reports, we recommend that these patients undergo total colonoscopy in the perioperative period to identify and remove synchronous lesions of the colon, and that examination of the remaining colon should be performed annually, preferably with colonoscopy, for at least the first four years after curative resection.     

Routine double contrast barium enema and fiberoptic colonoscopy in the diagnosis of colorectal carcinoma

The diagnostic efficacy of routine double-contrast barium enema and fiberoptic coloscopy for detection of cancer was retrospectively studied in 303 patients consecutively operated for colorectal adenocarcinoma in a 52-month period from January 1980. Double-contrast barium enema was performed in 180 patients with 184 carcinomas, 157 (85%) of which were revealed by this examination. The detection rate of carcinoma according to site was 89% between cecum and descending colon, 92% in sigmoid colon and 71% in rectum. Fiberoptic colonoscopy was done in 176 patients with 181 carcinomas, 163 (90%) of which were detected by the endoscopical examination. The detection rate of carcinoma according to site was 86% between cecum and descending colon, 90% in sigmoid colon and 95% in rectum. The two methods were equally effective in detecting carcinoma of the colon. Colonoscopy was superior to barium enema in detection of rectal carcinoma.     

Duplication of the proximal colon mimicking volvulus: a case report.

The case of a 56 year old female with intermittent pain, weight loss, anaemia, and a palpable tympanic abdominal mass is reported. Barium enema showed a very redundant loop of proximal colon, which was thought to have undergone recurrent volvulus. At colonoscopy, the findings seemed normal, but a much shorter length of colon was intubated to the caecum than expected from the barium findings. The duplicated colon was discovered only intra-operatively. This unusual diagnosis should be considered when a barium enema shows a long redundant colon which is not confirmed on colonoscopy.     

DUPLICATION OF THE PROXIMAL COLON MIMICKING VOLVULUS: A CASE REPORT

The case of a 56 year old female with intermittent pain, weight loss, anaemia, and a palpable tympanic abdominal mass is reported. Barium enema showed a very redundant loop of proximal colon, which was thought to have undergone recurrent volvulus. At colonoscopy, the findings seemed normal, but a much shorter length of colon was intubated to the caecum than expected from the barium findings. The duplicated colon was discovered only intra-operatively. This unusual diagnosis should be considered when a barium enema shows a long redundant colon which is not confirmed on colonoscopy.     

Ureterosigmoidostomy and carcinoma of the colon.

The etiology for development of colon carcinoma associated with ureterosigmoidostomy seems to be related to the urine. The incidence of colon carcinoma associated with ureterosigmoidostomy is 500 times greater than in the normal population, indicating about a 5% life time risk. The development time of these lesions varies from 6 to 50 years postoperatively but it is significantly less in patients more than 40 years old. The possibility exists that colon carcinoma may develop in primary sigmoid urinary diversion conduits or sigmoid internal conduits to either bladder or bowel. No reported bowel carcinoma has developed in an ileal urinary diversion. Followup examination should include stools for blood every 3 months after 2 years, an excretory urogram yearly after 5 years, sigmoid or colonoscopy every 5 years and barium enema every 5 years. If the patient has hematochezia or the excretory urogram demonstrates ureteral obstruction sigmoid or colonoscopy and a barium enema should be done.     

Etiologic and therapeutic aspects of intussusception in childhood

Fifty-two instances of intussusception in fortynine patients treated at the Hadassah University Hospital during a fifteen year period were analyzed. Intussusception was reduced with no subsequent complications in 47 per cent of the children who underwent barium enema reductions. Recurrence of intussusception occurred in 6 per cent, all of them successfully reduced again by barium enema. Sixty-five per cent of the patients underwent operation, of whom seven had pathologic lead points (lymphosarcoma, Meckel's diverticulum, leiomyoma, and intestinal polyp). Our experience and the collected evidence in the literature indicate the importance and usefulness of the barium enema examination as an excellent diagnostic and therapeutic modality in intussusception. Only when incomplete or uncertain reduction occurs should operative intervention be undertaken. The only contraindication to barium enema should be very marked intestinal obstruction or peritonitis.     

Colonoscopic diagnosis and treatment of lower gastrointestinal bleeding

Methods of diagnosis and treatment of lower gastrointestinal bleeding depend on the rate of bleeding and the amount of blood lost. If bleeding is occult, colonoscopy is the single best way to determine the source, if bleeding is gross but mild, causing melena or small amounts of hematochezia, colonoscopy or a combination of flexible sigmoidoscopy and double-contrast barium enema should be used to evaluate the colon. In most patients with melena, the upper tract must be examined endoscopically. Acute lower gastrointestinal bleeding stops spontaneously in 75 to 90 per cent of patients, permitting preparation of the colon before colonoscopy. If bleeding is continuing, diagnostic options include colonoscopy with no preparation of the colon, relying on the cathartic effect of blood, or a red cell radionuclide scan followed by angiography if the scan is positive. A bleeding lesion seen on angiography is usually treated by infusion of vasopressin. Colonoscopic treatment of a bleeding site uses the BICAP probe, heater probe, or argon laser. Patients who bleed severely and those who do not respond to treatment or rebleed after treatment are candidates for operation. Segmental resection is preferred if the bleeding site is known. If not, total colectomy with ileorectal anastomosis may be necessary. A mortality rate of 10 to 15 per cent in patients with severe bleeding reflects the advanced age of many of these patients and the difficulty of managing gastrointestinal bleeding in the presence of associated medical conditions.     

Is preoperative colonoscopy in carcinoma a realistic and valuable proposition?

Colonoscopy was performed preoperatively in 100 consecutive patients as a prospective study to establish the feasibility of the procedure and its value, which was considered "adequate" when the colon remaining after surgical resection had also been examined preoperatively. Colonoscopy was adequate in 35 of 46 patients (76%) with malignant tumour located in the cecum and ascending and transverse colon, but in only 15 of 54 patients (28%) who had a tumour in the left colon or rectum. Synchronous malignant tumours were present in 2 patients, and 54 additional adenomas were discovered in 29 patients. These adenomas could be removed endoscopically in 13 patients and were included in the standard resection in 12. Extension of the planned operation was necessary in only four patients with synchronous adenomas. Routine preoperative colonoscopy to assess the presence of synchronous colonic tumours is more likely to be adequate in proximal or right-sided large-bowel tumours than in left-sided tumours because of the annular configuration of the latter. Because of the high rate (46%) of synchronous adenomas in adequate examinations, inadequate preoperative colonoscopy should be complemented by a repeat examination 3 to 6 months postoperatively in every patient with a malignant tumour of the large bowel.     

The impact of an incomplete preoperative colonoscopy in patients with colorectal cancer

Total large bowel evaluation remains an essential step in the treatment of patients with colorectal cancer (CRC). Colonoscopy is the gold standard in the evaluation of the colon for colorectal tumors, but may be incomplete due to tumor obstruction, which is a frequent event in distal cancers. Double-contrast barium enema has a lower accuracy and is not ideal in the presence of signs of obstruction. In theory, intraoperative colonoscopy is a valid alternative, but its routine use is impeded by various practical limitations. Preoperative survey of the colon in 521 consecutive patients treated for CRC in our department was based on colonoscopy (92.5%). Our series was characterised by a high percentage of distal lesions (76.4%) and therefore by a high percentage of incomplete preoperative colonoscopies (50.4%) due to tumor obstruction. In the presence of an incomplete preoperative colonoscopy, we evaluated the entire colon with a double-contrast barium enema in selected cases (36.7%) and with a postoperative colonoscopy within 3 months of surgery in almost all patients (93.4%). The overall rate of complete endoscopic evaluation, either pre- or postoperatively, was 96.7%. The incidence of adenomas was significantly higher in the preoperative examinations as compared to early postoperative colonoscopy. This means that in distal cancer the vast majority of polyps will be located in the distal colon and therefore included in a standard resection. Four patients (0.8%), required a second operation for treatment of a missed lesion (2 benign and 2 malignant). The need for a repeat surgery apparently did not affect the therapeutic results in these four patients. On the basis of our experience, intraoperative colonoscopy would not appear to be a mandatory procedure in all cases of incomplete preoperative evaluation of the colon. However, in the absence of prospective, randomised trials comparing intraoperative vs early postoperative colonoscopy, the dilemma as to the strategy of choice remains.     

DIAGNOSTIC FAILURE IN COLONOSCOPIES FOR MALIGNANT DISEASE

Background: While colonoscopy has become established as more accurate than double contrast barium enema for detecting colonic polyps and cancers, as well as offering the opportunity for therapy, there are occasional instances where colonoscopy is misleading. The present study is to determine what problems occur, with a view to finding a solution. Methods: The records of the Colorectal Project at the Princess Alexandra Hospital indicate retrospectively that 346 patients have been correctly diagnosed with cancer of the colon and rectum by colonoscopy in the 5 years up to October 1996. During the same time eight patients (2.3%) were recorded at the same hospital as being misdiagnosed by colonoscopy, the lesion being either missed completely or misplaced. Results: In five of these patients there was failure to recognize that the whole colon had not been examined endoscopically, thereby missing a more proximal lesion. In two patients the lesion was missed although the entire colon was examined. In one patient the lesion was discovered but inaccurately sited. Six of these mistakes would have been obviated by the routine use of fluoroscopy to confirm the totality of the colonoscopy and to site any lesions found. The other two cases occurred because of failure to remember that colonic examination during withdrawal should be performed meticulously back as far as the anal canal. Failure to diagnose a colon cancer on the initial colonoscopy led to an average delay of 6 months for definitive care. Conclusions: It is recommended that fluoroscopy be used routinely during colonoscopy to site accurately any lesions found, and to confirm the completeness of insertion if reliable landmarks, including terminal ileum, are not clearly identified.