Local recurrence of colorectal cancer : clinical and experimental studies
Clinical and experimental studies which investigate factors affecting local recurrence of colorectal cancer are detailed. Three hundred and six patients underwent potentially curative surgery and had their anastomosis randomised to either sutures or staples. There was a 7.5% difference in overall recurrence rates in favour of the stapled patients (Log Rank X2=2.88, 1 df, p=0.09) and there was a similar difference in cancer specific mortality (Log Rank X2=2.41, 1 df, p=0.12). Anastomotic technique was an independent predictor of both tumour recurrence and cancer specific mortality. The effect that anastomotic integrity had on long term outcome was investigated using 177 patients. Tumour recurred in 44.7% of patients with an anastomotic leak and 30.2% of patients with an intact anastomosis (Long Rank X2=8.62, 1 df, p=0.002). Local recurrence occurred in 36.8% and 15.1% respectively. Cancer specific mortality was also significantly better for patients with an intact anastomosis (Log Rank X2=8.19. 1 df, p=0.004). The rates of clinical anastomotic leakage and peri-operative deaths between the surgeons participating in the study varied. Overall recurrence rates varied significantly between the surgeons, however there was a more marked difference in local recurrence rates (Log Rank X2=19.4, 6 df, p=0.004). There was no difference in cancer specific mortality between the surgeons. Experimental studies investigated ways of killing viable exfoliated tumour cells. In vitro, all tumouricidal agents caused 100% tumour cell death. In vivo, only povidone, iodine and sodium hypochlorite reduced the incidence of tumour growth. Tumouricidal agents were inactivated by the presence of whole blood in vitro. Intraperitoneal aqueous mitomycin C (MMC) and MMC adsorbed on activated carbon (MMC-CH) were investigated. The LD10 of MMC-CH was four fold greater and the dose corrected AUC was 17 times greater. Both preparations abolished tumour growth in vivo, but both had variable toxicity and markedly impaired anastomotic healing.