Parotid gland lesions: Diagnosis of malignancy with MRI and flow cytometric DNA analysis and cytology in fine-needle aspiration biopsy

Background. The purpose of this study was to assess the capability of magne tic resonance imaging (MRI) and cytology and flow cytometric (FCM) deoxyrib onucleic acid (DNA) analysis in fine-needle aspiration biopsy (FNAB)-derive d materials for diagnosing malignancy of the parotid lesions and the effica cy of FCM analysis in FNAB. Methods. Magnetic resonance imaging findings and FCM results (ploidy and S + G2 + M phases [S + G2M] fraction) and cytology in FNAB-derived materials in 26 patients with 26 parotid lesions (12 benign lesions, 14 malignancies) were assessed for predicting malignancy. Flow cytometric results in aspira tes were compared with those in surgically resected tissues. Results. When a single predictor was used, cytology (92% accuracy) was most accurate for malignancy, followed by ill-defined margin (88% accuracy) and aneuploidy (88% accuracy). The combination of FCM and cytology raised the rate of sufficient materials from 92% to 100% and accuracy from 92% to 96% compared with cytology alone. The same highest accuracy (96%) was obtained with the combination of the ill-defined margin or other findings such as cy tology, aneuploidy, or a high (S + G2M) fraction (6% <). Deoxyribonucleic a cid ploidy in the FNAB showed full agreement with that in the surgical spec imens. Receiver operating characteristic curves showed that the diagnosis o f malignancy with (S + G2M) fraction in FNAB was superior to that in surgic al specimens, but no significant difference was noted. Conclusions. A combination of MRI findings, cytology, and FCM results is op timal for diagnosing malignancies of the parotid lesions, and FNAB may repl ace the surgical specimens in FCM analysis. (C) 1999 John Wiley & Sons, Inc .