Archived: This report is greater than 3 years old. Findings may be used for research purposes, but should not be considered current.
Colorectal cancer is both the third most common type of cancer and the third most common cause of cancer-related death for both men and women. Treatment options for colorectal cancer differ depending on the clinical stage of disease at diagnosis. Our objective was to synthesize the available evidence on the comparative effectiveness of imaging tests for staging colorectal cancer.
We searched Embase®, MEDLINE®, PubMed, and the Cochrane Library from 1980 through November 2013 for published, English-language, full-length articles on using endoscopic rectal ultrasound (ERUS), computed tomography (CT), magnetic resonance imaging (MRI), and positron emission tomography (PET)/CT for staging colorectal cancer. The searches identified 4,683 citations; after dual screening against the inclusion criteria, 8 systematic reviews and 65 primary comparative studies were included.
We performed dual data abstraction from the included studies and constructed evidence tables. Where possible, we pooled the data using binomial-bivariate random-effects regression models (for diagnostic accuracy outcomes) or using random-effects meta-analysis (for under- and overstaging and under- and overtreatment outcomes). We rated the risk of bias of individual studies using internal validity instruments and graded the overall strength of evidence of conclusions using four domains (study limitations, consistency, precision, directness).
For preoperative rectal cancer T (tumor) staging, ERUS is more accurate than CT; ERUS is less likely than CT to incorrectly stage (odds ratio [OR] = 0.36; 95% confidence interval [CI], 0.24 to 0.54), less likely to understage (OR = 0.63; 95% CI, 0.44 to 0.89), and less likely to overstage (OR = 0.47; 95% CI, 0.28 to 0.80), supported by evidence of low strength. For preoperative rectal cancer T staging, MRI and ERUS were similar in accuracy, supported by evidence of low strength. There was no statistical difference in accuracy between MRI and CT, but there were few patients in the available studies. For preoperative rectal cancer N (lymph node) staging, CT, MRI, and ERUS were similar in overall accuracy, but all three modalities had limited sensitivity. MRI was less likely to overstage preoperative rectal cancer N stage than CT (OR = 0.498; 95% CI, 0.308 to 0.806), supported by evidence of low strength. We identified only one study of preoperative T and N staging of colorectal cancer using CT versus PET/CT. Nine studies reported on preoperative colorectal M (metastasis) staging. MRI is more sensitive than CT for detecting colorectal liver metastases (OR = 1.3, 95% CI, 1.0 to 1.8), supported by evidence of moderate strength.
There is no statistically significant difference in accuracy across MRI, CT, or ERUS for interim rectal T restaging, supported by evidence of low strength. The evidence was insufficient for drawing conclusions regarding other interim restaging outcomes for rectal or colorectal cancer following initial treatment.
While perforation, bleeding, and pain are potential complications of ERUS, most studies reporting complications mentioned only mild discomfort or minor bleeding. Rare but potential harms of MRI are associated with use of gadolinium-based contrast agents. Harms of CT include radiation exposure and adverse events from intravenous contrast agents.
ERUS is preferable to CT for preoperative rectal cancer T staging (based on evidence of low strength). Moderate-strength evidence suggests MRI is preferable to CT for detecting colorectal liver metastases. Low-strength evidence suggests that CT, MRI, and ERUS are comparable for rectal cancer N staging, where all have limited sensitivity, and for interim rectal cancer T restaging. Evidence was insufficient to allow any evidence-based conclusions about the use of PET/CT for colorectal cancer staging.