MR analysis of nasopharyngeal carcinoma: correlation of the pattern of tumor extent at the primary site with the distribution of metastasized cervical lymph nodes. Preliminary results.

2000 
The purpose of this study was to correlate the pattern of tumor extent of nasopharyngeal carcinoma at the primary site on magnetic resonance (MR) imaging with the distribution of cervical lymph node metastasis. MR images of 32 patients with biopsy-proven nasopharyngeal carcinoma were reviewed and classified into five patterns of tumor extent in correlation with the distribution of cervical lymphadenopathy. The assessment of cervical lymph node metastasis was done on the basis of the computed tomography (CT) findings. The tumor volume was also correlated with the occurrence of contralateral lymphadenopathy. Of the 32 patients, five (16%) presented as type 1, tumor limited to the nasopharyngeal mucosa; 12 (38%) as type 2 a, tumor which had invaded either lateral side but did not extend over the roof of nasopharynx; three (9%) as type 2 b, tumor which had invaded bilaterally across the midline but did not extend over the roof of nasopharynx; three (9%) as type 2 c, tumor which had invaded mainly the skull base but did not cross the midline; and nine (28%) as type 3, tumor which had extended anteriorly to the nasal cavity without invasion. Twenty-five patients (78%) demonstrated cervical lymphadenopathy. Patients with type 1, type 2 b and type 3 spread had frequent bilateral cervical lymphadenopathy; those with type 2 a had only ipsilateral lymphadenopathy. There was statistical significance (P<0.005) regarding the existence of contralateral lymphadenopathy with midline tumors as well as the absence of contralateral cervical lymphadenopathy with non-midline tumors. This study therefore suggests that the distribution of metastasized lymph nodes depends on the pattern of tumor extent at the primary site.
    • Correction
    • Source
    • Cite
    • Save
    • Machine Reading By IdeaReader
    28
    References
    19
    Citations
    NaN
    KQI
    []