Objective: To assess the use of contrast-enhanced magnetic resonance imaging (CEMRI) in addition to computed tomography in the pre-operative assessment of patients for radical surgery in malignant pleural mesothelioma.
Methods: Over a 45-month period, 51 of 76 patients assessed (69 men and seven women), underwent extra-pleural pneumonectomy or radical pleurectomy/decortication. Post-operative pathological stage was correlated with radiological staging, with particular emphasis on tumour resectability.
Results: Seventeen (22%) patients were found on CEMRI to have unresectable, but histologically unconfirmed disease, not previously seen on CT. Fifty-one (67%) patients proceeded to radical surgery, but pathological nodal data were incomplete in three, so excluding these patients from further analyses. The median pre-operative interval after CEMRI was 17 days. Two patients were found to have unexpectedly extensive disease at thoracotomy, thus the sensitivity of CEMRI for prediction of resectability was 97%. Using the International Mesothelioma Interest Group system, tumour stage was correctly predicted by CEMRI in 48% of patients, but understaged in 50% of cases, largely due to the underestimation of pericardial involvement, but this did not affect resectability and had no significant effect on prognosis. Nodal stage was correctly identified in 60% of patients. CEMRI was successful in predicting pathological tumour stage T3 or less (sensitivity of 85%; specificity of 100%), but less so in identifying tumour stage T2 or less (sensitivity of 23%; specificity of 96%) or N2 nodal disease (sensitivity 66%; specificity 73%).
Conclusions: CEMRI is most useful in the differentiation of T3 and T4 disease and may be unnecessary at earlier stages. Its multiplanar tumour localisation abilities are of value in the assessment of resectability. It is unlikely to contribute significantly to nodal staging, but it remains a valuable adjunct in the selection of patients for radical surgery.