Introduction Optimal follow-up after primary colorectal resection for adenocarcinoma is yet to be determined.1 2 Our centre uses surveillance CT scanning of the thorax, abdomen and pelvis at 1, 2 and 5 years postresection.

The aim of this study was to examine the role of a fifth-year surveillance CT scan in detecting resectable recurrence in our population.

Methods All patients at our centre who underwent potentially-curative resections of colorectal adenocarcinomas between 2003 and 2004 were identified using hospital coding. Retrospective data collection from hospital case-notes allowed details of radiological and clinical follow-up to be recorded in an excel database. Analyses of outcomes from fifth-year surveillance CT scans were performed using Microsoft Office Excel 2007 and GnuPSPP statistical software. The timing of presentation and clinical outcomes (ie, survival) between symptomatic and asymptomatic metastasis were charted and compared.

Results Of 213 patients studied, 121 (57%) were alive and disease free at 5 years, of whom 81 (67%) received a fifth-year surveillance CT scan. Fifth-year scanning did not demonstrate any new colorectal metastases however 6 (7%) scans revealed new, undiagnosed, non-colorectal malignancies.

All 40 of the patients who developed metastatic disease had presented by 3½ years postinitial resection. Nearly one third of these patients presented after their second-year surveillance CT scan. Asymptomatic metastasis at discovery was associated with improved 5-year survival.

Conclusion Our study showed no role for a fifth-year surveillance CT scan in the detection of resectable metastases, however there was a 7% pick-up rate for detecting new, unrelated malignancies. CT scanning beyond 2 years was needed to identify about one-third of the recurrences reported in this study.