Minimal-preparation CT colon in detection of colonic cancer, the Oxford experience

Background: the main colonic imaging modalities, including barium enema, colonoscopy and computed tomography colonography, require bowel preparation. Performing these imaging procedures in the elderly can difficult due to immobility, incontinence and poor tolerance of bowel cleansing. Minimal preparation CT (MPCT) colon was introduced in the early 1990s in the UK. Much of the published literature on MPCT colon is limited by small patient numbers and short duration of follow-up. Objective: the aim of this study is to review our experience with the MPCT technique involving a large consecutive cohort of patients with long follow-up. Methods: all studies of MPCT performed in a 1-year period between July 2000 and July 2001 at our institution were reviewed retrospectively. MPCT reports were cross-referenced with the cancer registry to allow for an average period of 30 months follow-up. A definite diagnosis of cancer was only given following the appearance on the cancer registry. Those patients who had negative MPCT colon were assumed to be true negatives if no corresponding name was identified on the cancer registry. In the event of data mismatch, patient notes were reviewed to ascertain a diagnosis. Results: 391 MPCT examinations were performed during the period of the study (209 males, median age 82: age range 56-91 years). Thirty-four patients who had MPCT colon during the study period appeared on the cancer registry. A further three patients with disseminated colorectal malignancy identified on MPCT colon died without histological confirmation (tumour prevalence = 9.5%). Thirty-two of the registry confirmed 34 cases were detected on MPCT colon, giving a sensitivity of 0.94 (95% confidence interval 0.86-1.00). Including the three cases without histological confirmation gives a slightly higher sensitivity of 0.95. There were seven patients with definitely abnormal MPCT colons, who did not appear on the registry, resulting in specificity for definite abnormality of 0.98 (confidence interval 0.97-1.0). However, three of these seven are those who died of disseminated colorectal malignancy as above, raising the specificity to 0.99. Fourteen cases (3.5%) of extra-colonic malignancies were observed in this study. Conclusion: even with the longer follow-up of this large cohort of patients the sensitivity and specificity in our study for the diagnosis of colorectal cancer with MPCT remains comparable with that of other studies and this technique competes well with other common colonic imaging modalities.