K. W. Eu

A defunctioning ileostomy does not prevent clinical anastomotic leak after a low anterior resection: a prospective, comparative study.

PURPOSEDefunctioning ileostomy or colostomy is still routinely performed after low anterior resection in the belief that diverting the fecal stream will prevent anastomotic dehiscence. However, an ileostomy is not without morbidity for the patient. This study aims to determine if a diverting stoma is really necessary after a low anastomosis.METHODSAll low or ultralow anterior resections done in this department were performed by consultant-grade surgeons in a standardized manner. The patients were all monitored closely after surgery for clinical signs of an anastomotic leak. There were 1078 patients who underwent elective low or ultralow anterior resections in a ten-year period between 1994 and 2004. Twelve of them were irradiated before surgery; they were excluded from the study. During a seven-month period from February 2004 through August 2004, 324 patients who underwent such procedures were not defunctioned. These were compared with 742 patients who were previously defunctioned with a proximal stoma. The results were analyzed using the Pearson chi-squared test.RESULTSThirteen (4 percent) patients who were not defunctioned developed a clinical anastomotic leak, whereas the leak rate for those who were defunctioned was 3.8 percent. There was no statistical difference demonstrated. Ninety-five percent of patients who developed a leak required surgical intervention; the remaining 5 percent could be dealt with by radiologic drainage. The overall mortality rate for anastomotic leak in this department is 7.3 percent.CONCLUSIONA diverting stoma does not reduce postoperative anastomotic leak rate. Rather, it reduces the otherwise catastrophic effects of an anastomotic leak such as fecal peritonitis and septicemia. An ileostomy carries certain morbidity and also adds to the cost of the entire operation. Therefore, it should not be performed routinely. Instead, it should be performed selectively in patients with poorly prepared bowels, coupled with a distal limb washout, and in patients with significant comorbidities who can ill afford the complications of a leak.

Comparison of J-Pouch and Coloplasty Pouch for Low Rectal Cancers: A Randomized, Controlled Trial Investigating Functional Results and Comparative Anastomotic Leak Rates

ObjectiveTo assess the efficacy of a novel coloplasty colonic pouch design in optimizing bowel function after ultralow anterior resection. Summary Background DataA colonic J-pouch may reduce excessive stool frequency and incontinence after anterior resection, but at the risk of evacuation problems. Experimental surgery on pigs has suggested that a coloplasty pouch (CP) may be a useful alternative. Although CP has recently been shown to be feasible in patients, there is no randomized controlled trial comparing bowel function with the J-pouch. MethodsAfter anterior resection for cancer, patients were allocated to either J-pouch or CP-anal anastomoses. Continence scoring, anorectal manometry, and endoanal ultrasound assessments were made before surgery. All complications were recorded, and these preoperative assessments were repeated at 4 months. The assessments were repeated again at 1 year, and a quality of life questionnaire was added. ResultsEighty-eight patients were recruited from October 1998 to April 2000. Both groups were well matched for age, gender, staging, adjuvant therapy, and mean follow-up. There were no differences in the intraoperative time and hospital stay. CP resulted in more anastomotic leaks. At 4 months, J-pouch patients had 10.3% less stool fragmentation but poorer stool deferment and more nocturnal leakage. However, there were no differences in the bowel function, continence score, and quality of life at 1 year. There were no differences in the anorectal manometry and endoanal ultrasound findings. ConclusionsColoplasty pouches resulted in more anastomotic leaks and minimal differences in bowel function. At present, the J-pouch remains the benchmark for routine clinical practice, and due care (including defunctioning stoma) should be exercised in situations requiring CP.

A simple and effective treatment for hemorrhagic radiation proctitis using formalin

Radiation proctitis is a common complication of radiotherapy for pelvic malignancy. In the more severe form, it leads to intractable or massive hemorrhage, which may require repeated hospital admissions and blood transfusions. Medical therapy in patients with radiation proctitis is usually ineffective, whereas surgery is associated with a high morbidity and mortality. Eight patients (seven females and one male) with hemorrhagic radiation proctitis were Treated over a six-month period with endoluminal formalin. The technique used ensured minimal contact with formalin. The median age of the patients was 68 years (range, 42–73 years). Seven patients had had cancer of the uterine cervix, and one patient had had cancer of the prostate treated with radiotherapy at a median time of 30 months (range, 9–46 months) previously. The median duration of time of symptomatic rectal hemorrhage before formalin therapy was eight months (range, 1–12 months). The median number of units of blood transfused previously per patient was four (range, 2–32). The time taken for formalin therapy was 20 minutes (range, 10–70 minutes). One patient required repeat formalin application at two weeks. Bleeding ceased immediately in seven patients after formalin treatment. No further bleeding was noted, nor was any blood transfusion needed, at follow-up at four months (range, 1–6 months). Formalin therapy is a simple, inexpensive, and effective treatment for hemorrhagic radiation proctitis.

Primary colorectal lymphomas

Objective  The incidence of primary colorectal lymphomas is rare, comprising 10–20% of gastrointestinal lymphomas and only 0.2–0.6% of large bowel malignancies. There is a male predominance, with a maximal reported incidence in the 50‐ to 70‐year age group. Patients often present delayed with nonspecific symptoms and consequently have advanced disease at the time of diagnosis. Inflammatory bowel disease and immunosuppression have been reported as risk factors, although a direct causal link has yet to be established. Treatment often involves a multimodality approach, combining surgery and chemotherapy, with the use of radiotherapy in selected cases. We present our experience in the management of primary colorectal lymphomas over a 10‐year period (1989–1999).

Early postoperative results of a prospective series of laparoscopic vs. open anterior resections for rectosigmoid cancers

PURPOSE: This study was undertaken to compare postoperatively laparoscopic (LAR) with open (OAR) anterior resection in patients with rectosigmoid cancers. METHODS: Forty consecutive patients were divided into two groups: 20 patients (9 males) were allocated to LAR and 20 patients (6 males) to OAR. RESULTS: Median age in the LAR group was 62 (range, 39–77) years, and in the OAR group, it was 61 (range, 43–84) years (P=0.9). Median lengths of the distal margin of clearance beyond the tumor were 4 (range, 2–8) cm and 4.5 (range, 3–7.5) cm in the LAR and OAR groups, respectively (P=0.35). Median numbers of lymph nodes harvested were 20 (range, 7–49) and 19 (range, 7–97) for the LAR and OAR groups, respectively (P=0.44). Median operating times were 90 (range, 55–185) minutes and 73 (range, 40–140) minutes in the LAR and OAR groups, respectively (P=0.08). Blood losses were 50 (range, 50–800) ml and 50 (range, 50–1,500) ml in the LAR and OAR groups, respectively. There was no intraoperative complication in either group, and no laparoscopic patient was converted to an open procedure. Median length of extraction site incision in the LAR group was 5.5 (4–13) cm, and length of incision in the OAR group was 18 (8–25) cm (P<0.002). CONCLUSION: There were no significant differences between the two groups with regard to duration of parenteral analgesia, starting of fluid and solid diet after surgery, or time to first bowel movement and time to discharge from the hospital.

Experience of 3711 stapled haemorrhoidectomy operations

Stapled haemorrhoidectomy has been routinely performed in the Department of Colorectal Surgery, Singapore General Hospital since 1999.

Anal sphincter injuries from stapling instruments introduced transanally

PURPOSE: Injury sustained from the transanally introduced stapling technique was assessed by comparison with biofragmentable anastomotic ring anastomosis, which excluded anal manipulation. METHODS: A randomized, controlled trial was conducted on consecutive patients undergoing sigmoid colectomy (where pelvic nerve injury was avoided). A bowel function questionnaire was administered six months after surgery. Anorectal manometry and endoanal ultrasonography were performed preoperatively and at six months postoperatively. The observers were blinded to the randomization. RESULTS: There were 18 patients in the transanally introduced stapling technique group and 17 patients in the biofragmentable anastomotic ring group, with no differences in age, gender, Dukes staging, and follow-up. Three of the transanally introduced stapling technique patients had occasional liquid soiling, which was absent in biofragmentable anastomotic ring patients. Mean change in resting anal pressures was also significantly impaired when compared with patients with biofragmentable anastomotic ring (P=0.007). Endosonographic internal sphincter fragmentation was found in five transanally introduced stapling technique patients but none after biofragmentable anastomotic ring anastomosis (P=0.046). Internal sphincter fragmentation was associated with the impaired resting pressures (P=0.007). External sphincter deficiencies were found after transanally introduced stapling technique in two patients (biofragmentable anastomotic ring = 0), and these were associated with the soiling (P=0.005). CONCLUSIONS: The transanally introduced stapling technique may result in anal sphincter defects and impaired anal pressures when assessed at six months of follow-up.

Meta-analysis of randomized clinical trials comparing drainage alone vs primary sphincter-cutting procedures for anorectal abscess–fistula

Background and aimConcurrent definitive treatment of underlying fistulas from infected anal glands at the time when the anorectal abscesses are drained is controversial as this is associated with a higher incidence of faecal incontinence, failure and recurrence. This meta-analysis was conducted to determine the merits of drainage alone vs primary sphincter-cutting procedures (which includes fistulotomy and fistulectomy) for anorectal abscess–fistula.MethodsMedline, Embase and Cochrane Central Register of Controlled Trials database searches identified all randomized controlled trials using the keywords: anorectal abscess, anal sepsis, drainage, fistulotomy, fistulectomy or surgery from 1966 to 2004. The outcome variables analysed were recurrence, faecal continence and wound-healing times.ResultsFive trials were considered suitable for the meta-analysis, with a total of 405 patients. Sphincter-cutting procedures for anorectal abscesses resulted in 83% reduction in recurrence rate [relative risk (RR) 0.17, 95% confidence interval (CI) 0.09–0.32, p<0.001]. However, there was a tendency to a higher risk of faecal incontinence to flatus and soiling when primary sphincter-cutting procedure was performed (RR 2.46, 95% CI 0.75–8.06, p=0.140).ConclusionThere is no conclusive evidence if simple drainage or sphincter-cutting procedure is better in the treatment of anorectal abscess–fistula.

High preoperative serum carcinoembryonic antigen predicts metastatic recurrence in potentially curative colonic cancer: Results of a five-year study

INTRODUCTION: Serum carcinoembryonic antigen is used mainly for tumor follow-up to detect recurrence of colonic cancer. However, raised preoperative carcinoembryonic antigen levels may be helpful for the identification of understaged cases and of patients meriting more intensive preoperative and postoperative diagnostic workup. METHODS: From a prospectively collected database, the data on 261 patients who had curative colonic carcinoma with a minimal follow-up of five years and who had preoperative carcinoembryonic antigen levels assessed were retrieved and analyzed. Outcome parameters were local and/or distant recurrence and time to recurrence. These parameters were correlated with Dukes staging and preoperative carcinoembryonic antigen levels. RESULTS: The cumulative diseasefree survival of patients with a preoperative carcinoembryonic antigen level within the normal range was significantly better than that of those whose carcinoembryonic antigen was 5 ng/ml or more (P=0.001). No patient with carcinoembryonic antigen levels less than 1 ng/ml developed metastatic recurrence. Twenty-three percent of all patients with a raised carcinoembryonic antigen above 5 ng/ml compared with 2.1 percent of patients with carcinoembryonic antigen below 5 ng/ml developed a metastasis at two years. At five years, these figures were 37.2 percent and 7.5 percent, respectively. Dukes staging and carcinoembryonic antigen levels were found to be directly correlated (P<0.001) when all patients were included. Carcinoembryonic antigen of more of 15 ng/ml was found to be a significant adverse prognostic indicator for disease-free survival irrespective of Dukes staging (P<0.02). Raised carcinoembryonic antigen levels predicted distant metastatic recurrence (P<0.001) but did not predict local recurrence (P=0.72). CONCLUSIONS: High preoperative carcinoembryonic antigen levels above 15 ng/ml predicted an increased risk of metastatic recurrence in potentially curative colonic cancer and may indicate undetectable disseminated disease. Preoperative carcinoembryonic antigen levels predict understaging and the possibility of distant recurrence. Such patients may therefore be selected for adjuvant therapy where indicated. Therefore, carcinoembryonic antigen is complementary to conventional Dukes staging for the prediction of recurrence and survival.

Randomized controlled trial of primary fistulotomy with drainage alone for perianal abscesses

PURPOSE: Primary fistulotomy may be advantageous for perianal abscesses because unlike ischiorectal abscesses, fistulas are more commonly found and can be laid open with full preservation of the external anal sphincters. Therefore, a randomized, controlled trial was conducted to compare primary fistulotomy with incision and drainage alone, specifically for perianal abscesses. METHODS: Fifty-two consecutive patients (43 males; mean age, 40 (standard error of mean, 2) years) with perianal abscesses were randomized to treatment by either incision and drainage (controls; N=28) or fistulotomy (N=24). Patients were followed up clinically for a mean of 15.5 (standard error of the mean, 0.7) months. Anorectal manometry was also performed before, six weeks, and three months after surgery. RESULTS: Persistent fistulas developing after surgery were significantly more common after incision and drainage (N=7; 25 percent) than after fistulotomy (N=0;P=0.009). One patient in each group was also found to have a residual abscess, which required repeat drainage. All patients remained fully continent. The anal pressures after incision and drainage and fistulotomy were not significantly different. Operative time, hospital stay, and time for the wound to heal completely were the same in both groups. CONCLUSIONS: Primary fistulotomy at the time of drainage for perianal abscesses results in fewer persistent fistulas and no added risk of fecal incontinence.