Professor MRB Keighley
The optimum prophylactic operation for cancer in familial adenomatous polyposis
University Department of Surgery, Queen Elizabeth Hospital, Birmingham, England
Familial adenomatous polyposis is a Mendelian dominant with an inevitable risk of colon cancer if left untreated. However, the risk of colon cancer and the time at which cancer develops varies with the genetic mutation on chromosome 5, which also influences the incidence of extra-intestinal disease and other factors which could determine the most appropriate surgical operation.
There is thus enormus heterogeneity in familial adenomatous polyposis in terms of colorectal polyps, other of potential malignancy and desmoids. The number of colorectal polyps veries with the sepcific site of mutation as do the malignant potential of the polyps and the distribution of the polyps. Furthermore, the risk of bile induced cancer in the bile duct, duodenum and small bowel also is influenced by the mutation on the APC gene. In addition, the risk of gastric polyps of thyroid cancer, central ner-vous system tumours, bone tumours and subcutaneous tumours is also influenced by the genetic predisposition and site of mutation. Finally, the risk of desmoids either in the wound, in the retroperitoneal tissues or in the pelvis is genetically determined. Since all of these issues play a role in deciding upon the most appropriate surgical operation in FAP, more genetic information and its impact on prognosis is needed.
Quite apart from genetic information, there are other issues that need to be taken into consideration when deciding upon the most appropriate operation for familial adenomatous polyposis. These include patient intelligence, patient expectations, reliability of individuals in attending follow up, body image and the fact the their operation is a prophylactic procedure when patients have no symptoms must also be considered.
The principal surgical options in the ablation of the large bowel for prevention of colorectal cancer include: a) proctocolectomy and permanent ileostomy, b) subtotal colectomy and ileorectal anastomosis, which leaves the rectal stump still at risk and c) restorative proctocolectomy and ileopouch anal anastomosis. However for ileopouch anal anastomosis to completely protect against malignancy a mucosectomy of the anal canal is needed. Having said that, the risk of cancer occurring in the anal canal is extremely low and it could, therefore, be argued that the risk is so small that a mucosectomy is not necessary and that it is perfectly justified to perform an operation with better functional results by stapled ileoanal anastomosis without anal mucosectomy.
As conventional proctocolectomy and permanent ileostomy has such an enormus stigmata in terms of persuading the next generation to come forward for prophylactic surgery, and because very few patients who are asymptomatic really are prepared to consider an ileostomy as their first choice, conventional proctocolectomy will not be futher considered. Despite this there are a few patients, who are worried about the consequences of complications following ileopouch anal anastomosis or rectal cancer after ileorectal anastomosis and decide on totally ablative surgery on the grounds that a proctocolectomy and ileostomy as a single operation and that is the end of it.
When comparing ileorectal anastomosis versus ileopouch anal anastomosis, it is important for patients to realize that cancer prophylaxis is incomplete with ileorectal anastomosis, whereas with a mucosectomy follow up is not necessary in patients treated by ileoanal pouch surgery.
In terms of morbidity, ileorectal anastomosis scores over ileopuch anal anastomosis. Hospital stay is shorter (8% v 10%), leak rates are lower (2% v 6%), pelvic sepsis is lower (4% v 8%) and the need for a covering loop ileostomy to protect the operation is also less (2% v 16%). Also in terms of complications, ileorectal anastomosis has a lower rate of complications than ileopouch anal anastomosis, in particular small bowel obstruction occurs in approximately 5% of all patients treated by ileorectal anastomosis compared with a cumulative figure of 20% after ileopouch anal anastomosis. Furthermore, if desmoid tumours are present the risk of small bowel obstruction is significantly higher in patients treated by ileopouch anal anastomosis as compared with ileorectal anastomosis.
It used to be thought that the functional results of ileopouch anal anastomosis and ileorectal anastomosis were comparable. This is not entirely true. Bowel frequency is more predictable after ileorectal anastomosis, occurring from between one and three times a day, whereas after ileopouch anal anastomosis pouch frequency varies tremendously from four to 12 time a day; rising at night time to have one´s bowels open rarely is necessarily after ileorectal anastomosis, whereas 30% of patients after ileopouch anal anastomosis need to get up at night time. If a mucosectomy has been performed there is at least a 15% risk of soiling with ileopouch anal anastomosis and 5% of patients are troubled with urgency. However if a stapled ileopouch anal anastomosis is used then the incidence of urgency is 4% and soiling is less than 7%.
The biggest downside to ileorectal anastomosis is that at least 70% of patients are going to require a further major operation to remove the rectum during life-time. The risk of rectal cancer is low after ileorectal anastomosis provided patients attend for regular follow up is low. The risk of rectal cancer after ileopouch anal anastomosis is purely anecdotal. The proportion of patients eventually requiring an ileostomy treated initially by ileorectal anastomosis is about 5% compared with 10% after ileopouch anal anastomosis.
More genetic information is required if we are going to select the most appropriate operation as prophylaxis in asymptomatic patients for familial adenomatous polyposis. However, in this author´s experience, follow up is variable in FAP patients, hence the majority of this author´s patients are offered an ileopouch anal anastomosis, provided that there is no evidence of desmoids.