Over 14 years 276 patients with rectal cancer underwent surgery; 219 who underwent low anterior resection of the rectum with total mesorectal excision were studied. There were 24 (11.0 per cent) major anastomotic leaks associated with peritonitis or a pelvic collection and 14 (6.4 per cent) minor leaks that were asymptomatic and detected by contrast enema. All major leaks occurred at an anastomotic height of less than 6 cm (P = 0.08). The abdominoperineal excision rate was 9.1 per cent. Major leaks were associated with failure to defunction in 11 of 62 patients and with a defunctioning colostomy in 13 of 157 (P = 0.03). Of the 24 patients with major leaks seven developed peritonitis, one with a defunctioned anastomosis (P = 0.002), and three died (P = 0.02). Use of the sigmoid colon led to major leakage in seven of 32 patients compared with 17 of 187 when the splenic flexure was employed (P = 0.05). There was no increase in the local recurrence rate but only nine patients with major leakage and a temporary stoma have had these closed. Key technical factors include: a clean dry pelvic cavity, pulsatile colonic blood supply, suction drainage started during closure and mobilization of ample tissue to fill the pelvic space.

Risk of peritonitis and fatal septicaemia and the need t o defunction the low anastomosisBr.

The effects of two methods of colonic vascular ligation were studied in 143 consecutive patients who underwent low anterior resection with total mesorectal excision and full mobilization of the splenic flexure. Either the ascending left colic artery (ALCA) was selectively preserved (n = 52) or a flush aortic ligation was performed (n = 91). In those with a protective colostomy, the radiological leak rate was 12 per cent when the ALCA was preserved (n = 41) and 10 per cent when a flush aortic tie was performed (n = 60) (P greater than 0.95; 95 per cent confidence interval (c.i.) for difference -10 to +15 per cent). In those without a colostomy, the clinical leak rates of 9 per cent when the ALCA was preserved (n = 11) and 19 per cent when a flush aortic tie was performed (n = 31) were not significantly different (P greater than 0.10; 95 per cent c.i. for difference -12 to +32 per cent). Proportional hazards analysis showed no association between the method of vascular ligation and the risk of tumour recurrence and death. Anastomotic leak rates, tumour recurrence and survival were not related to the method of vascular ligation.

Summary:The value of follow-up after potentially curative treatment of breast cancer remains controversial. Recurrence-free women (n = 402) attending a breast clinic over a 3 month period were studied. The women attended 423 appointments, 412 of which were routine. Eleven were requested by the patient or general practitioner (interval appointments). All 11 interval and 19 of the routine appointments resulted in investigations for possible recurrence. Three (27%) interval appointments and four (1%) routine appointments resulted in the diagnosis of a recurrence (P < 10-5).The attitudes of 285 patients to follow-up were investigated by questionnaire. Two hundred and twenty-three (78%) questionnaires were completed. Regular follow-up in the breast clinic was preferred to attendence only when symptomatic by 190 (85%) women and 169 (76%) preferred regular breast clinic visits to general practitioner follow-up. Most women (n = 174) (81%) said they felt reassured and less anxious having attended the breast clinic. Routine follow-up after potentially curative treatment of breast cancer is inefficient in the detection of recurrence. It is, however, highly rated for providing reassurance and reducing anxiety. Reassurance rather than detection of recurrence may be the most important function of the breast cancer follow-up clinic.

No abstract