• Colorectal neoplasms;
  • referral and consultation;
  • survival analysis;
  • colonoscopy;
  • barium enema;
  • sigmoidoscopy


Objective  To review all preceding ‘negative’ large bowel investigations in patients with a final diagnosis of colorectal cancer, and to examine whether delayed diagnosis was associated with worse outcome.

Method  Details were gathered on all patients with a new diagnosis of colorectal adenocarcinoma presenting over 4.5 years. For each patient the hospital’s clinical workstation and radiology and endoscopy databases were interrogated for all flexible sigmoidoscopies, colonoscopies and barium enemas during the 5 years prior to diagnosis.

Results  Among the 570 patients, 28 (5%) had undergone colonoscopy and/or flexible sigmoidoscopy that had not shown colorectal cancer during the 5 years preceding final diagnosis, and a further 28 (5%) had undergone ‘negative’ barium enemas. Polyp surveillance might have missed four lesions destined to become malignant. Correspondingly there were three patients undergoing IBD surveillance found to have CRC, having had a negative complete colonoscopy within the preceding 5 years. Among patients undergoing de novo colonoscopy for diagnosis the true miss rate was only one patient per year. At August 2007, 29 (58%) of those with delayed diagnosis were still alive, compared with 216 (42%) of those diagnosed during initial investigation (χ2 = 5.04, < 0.05).

Conclusions  Colonoscopic miss rates are in line with previous studies. The application of simple clinical ground rules will avoid most pitfalls. The methodology described herein may assist in auditing the quality assurance of lower gastrointestinal diagnostic services. Despite the delay, late diagnosis was found to be associated with improved survival and a lower likelihood of metastatic disease.