Staging and follow-up of nasopharyngeal carcinoma: MR versus CT

Purpose: To compare computerized tomography (CT) and magnetic resonance (MR) in relation to their accuracy in the staging of nasopharyngeal carcinoma (NPC); to compare CT and MR in postirradiation follow-up of NPC. Methods and Materials: Staging: From 1985 to 1993,53 patients affected with NPC were studied with MR and CT. All cases were biopsy-proved epithelial carcinoma. Plain and contrast-enhanced CT scans were performed with third-generation scanners. Magnetic resonance were obtained with 0.5 and 1.5 Tesla units in sagittal, axial, and coronal planes. Computerized tomography was chosen as reference method and findings obtained with MR were compared to those obtained with CT. Follow-up: From 1985 to 1993, 53 patients irradiated with radical intent were followed up with both CT and MR; 71 examinations were performed in all. The baseline follow-up scan was performed, in general, no sooner than 2 months after the end of radiotherapy. All patients were submitted to unlimited clinical follow-up. Results: Staging: Magnetic resonance showed retropharyngeal adenopathies in 6 of 14 cases in which oropharyngeal involvement had been reported after CT; in 3 other patients, adenopathies were recognized on MR, while primary extent to parapharyngeal space had been diagnosed on CT initially. Infiltration of long muscles of the neck was revealed with MR in 14 cases. On the other hand, CT showed bone invasion in 12 patients vs. 8 on MR. Upstaging to T4 occurred in four cases on the basis of CT; no upstaging occurred after MR. Follow-up: Findings on CT were uncertain in 10 out of 53 patients, disease recurrence was excluded by MR in nine cases, whereas progressive disease was confirmed in one patient. Conclusion: Staging: Our series shows that either CT and MR can provide essential information in the staging of NPC. Magnetic resonance, however, seems to provide the most detailed imaging of soft tissue invasion outside the nasopharynx and of retropharyngeal node involvement. Nonetheless, its limitations in evaluating bone details suggest that CT should be always performed when the status of base of skull is uncertain on MR. General reasons and our data indicate that CT can still be considered a valuable tool in routine NPC staging. Follow up: Magnetic resonance may be the modality of choice because it seems to solve, more often than CT, the problems of differentiation between postradiation changes and recurring tumor, apart from those cases showing subtle bone erosions on initial CT scan.