Bladder cancer: evaluation of staging accuracy using dynamic MRI.

ABSTRACT

AIM: To assess the accuracy of magnetic resonance imaging (MRI) in staging bladder cancer and to assess whether dynamic gadolinium-enhanced sequences have any added benefit in staging. MATERIALS AND METHODS: Over a 22 month period, the MRI findings of 100 consecutive patients with histologically proven transitional cell carcinoma (TCC) of the bladder were reviewed. The T stage was assessed independently on T2-weighted imaging alone and in combination with gadolinium-enhanced MRI. The final histological diagnosis was considered the reference standard. Statistical analysis was performed to ascertain stage-by-stage accuracy. Accuracy of MRI in differentiating superficial (≤ T1) from invasive (≥ T2) and in differentiating organ-confined (≤ T2) from non-organ-confined (≥ T3) disease was assessed. RESULTS: On a stage-by-stage basis, tumours were correctly staged using MRI in 63% of patients (observed agreement=0.63, weighted kappa=0.57). The sensitivity and specificity of MRI to differentiate between superficial (≤ T1) from invasive (≥ T2) disease was 78.2 and 93.3%. The observed agreement for this group was 85% (kappa=70%; p<0.0001). The sensitivity and specificity of MRI to differentiate between organ-confined (≤ T2) from non-organ confined (≥ T3) disease was 90.5 and 60%. The observed agreement for this group was 89% (kappa=30%; p<0.01). Gadolinium-enhanced images improved staging in only three patients. CONCLUSION: In the present study MRI was found to be a moderately accurate tool in assessing the T stage. Agreement on a stage-by-stage basis was good. Agreement for differentiating between non-invasive versus muscle-invasive disease was good and that for organ-confined versus non-organ-confined disease was fair. Routine use of gadolinium-enhanced images is not routinely required.