Obaro, A;
Plumb, AAO;
Fanshawe, TR;
Torres, US;
Baldwin-Cleland, R;
Taylor, SA;
Halligan, MS;
(2018)
Post-imaging colorectal cancer or interval cancer rates after computed tomographic colonography: A systematic review and meta-analysis.
The Lancet Gastroenterology and Hepatology
, 3
(5)
pp. 326-336.
10.1016/S2468-1253(18)30032-3.
Preview |
Text
Lancet_Gastro_Hep_Accepted.pdf - Accepted Version Download (1MB) | Preview |
Abstract
Background: CT colonography (CTC) is highly sensitive for colorectal cancer, but “interval” or postimaging colorectal cancer (PICRC) rates (diagnosis of cancer after initial negative CTC) are unknown, as are their underlying causes. Methods: We conducted a systematic review and meta-analysis of post-CTC PICRC rates and causes by searching MEDLINE, EMBASE and the Cochrane Register. We included randomised, cohort, cross-sectional or case-control studies published Jan 1994-Feb 2017, using CTC performed according to international consensus standards with aim of detecting cancer or polyps, and reporting PICRC rates or sufficient data to allow their calculation. Two independent reviewers extracted data from the study reports. We used random-effects meta-analysis to estimate pooled PICRC rates, expressed using (a) total number of cancers and (b) total number of CTC scans as denominators, and (c) per 1000 person-years. Primary study authors provided details of retrospective CTC image review and causes for each PICRC. The study is registered (PROSPERO:CRD42016046838). Findings: 2977 articles were screened and 12 analysed. These reported 19,867 patients (18-96 years; of 11,590 with sex data available, 6532 (56·4%) female) from March 2002-May 2015. At mean 34 months’ follow-up (range: 3 to 128·4 months), CTC detected 643 cancers and 29 PICRCs were diagnosed. The pooled PICRC rate was 4·42 PICRCs/100 cancers detected; 95%CI 3·03-6·42, corresponding to 1·61 PICRCs/1000 CTCs (95%CI 1·11-2·33) or 0·64 PICRCs/1000 person-years (95%CI 0·44-0·92). Heterogeneity was low (I2 =0%). Over half (17/28, 61%) of PICRCs were due to perceptual error and visible in retrospect. Interpretation: The 3-year PICRC rate post-CTC is 4·4%, or 0·64 per 1000 person-years, towards the lower end of range reported for colonoscopy. Most arise from perceptual errors. Radiologist training and quality assurance may help reduce PICRC rates. Funding: St Mark’s Hospital Foundation and the UCL/UCLH Biomedical Research Centre.
Archive Staff Only
 |
View Item |
