PURPOSE: The aims of this study were to measure the accuracy of computerised tomography (CT) and magnetic resonance imaging (MRI) scans in detecting colorectal liver metastases (CRLM) and to determine if patients who are staged with MRI in addition to CT have longer liver recurrence-free survival compared to those having CT alone in a unit performing routine intra-operative ultrasound. METHODS: A retrospective analysis of patients undergoing liver resection for CRLM was performed. Patients staged pre-operatively with CT or with additional MRI were included and those with additional PET imaging were excluded from survival analysis. Timing and site of tumour recurrence were recorded. RESULTS: During a 7-year period, 303 patients underwent resection for CRLM of whom 47 (15.5 %) were staged with CT alone and 36 (11.9 %) with additional MRI. The overall accuracy of CT (63 %) and MRI (61.9 %) was similar in the detection of tumour nodules (P = 0.905). There was no difference in the rate of intra-hepatic recurrence between groups with 13/47 and 8/36 cases, respectively (P = 0.737). There was no difference in the disease-free survival curves between the groups (P = 0.487). CONCLUSIONS: Our recommendation is that MRI should not be a mandatory imaging modality in referral guidelines for patients with hepatic CRLM, as the cost and delay associated with the scan outweigh any potential benefit in terms of improved sensitivity compared to CT.
PURPOSE: The aims of this study were to measure the accuracy of computerised tomography (CT) and magnetic resonance imaging (MRI) scans in detecting colorectal liver metastases (CRLM) and to determine if patients who are staged with MRI in addition to CT have longer liver recurrence-free survival compared to those having CT alone in a unit performing routine intra-operative ultrasound. METHODS: A retrospective analysis of patients undergoing liver resection for CRLM was performed. Patients staged pre-operatively with CT or with additional MRI were included and those with additional PET imaging were excluded from survival analysis. Timing and site of tumour recurrence were recorded. RESULTS: During a 7-year period, 303 patients underwent resection for CRLM of whom 47 (15.5 %) were staged with CT alone and 36 (11.9 %) with additional MRI. The overall accuracy of CT (63 %) and MRI (61.9 %) was similar in the detection of tumour nodules (P = 0.905). There was no difference in the rate of intra-hepatic recurrence between groups with 13/47 and 8/36 cases, respectively (P = 0.737). There was no difference in the disease-free survival curves between the groups (P = 0.487). CONCLUSIONS: Our recommendation is that MRI should not be a mandatory imaging modality in referral guidelines for patients with hepatic CRLM, as the cost and delay associated with the scan outweigh any potential benefit in terms of improved sensitivity compared to CT.
Authors: Shandra Bipat; Maarten S van Leeuwen; Emile F I Comans; Milan E J Pijl; Patrick M M Bossuyt; Aeilko H Zwinderman; Jaap Stoker Journal: Radiology Date: 2005-08-11 Impact factor: 11.105