OBJECTIVES: To provide a graded diagnosis of benign and malignant lesions in the parotid gland by conventional ultrasound, and thus to predict the probability of malignancy of the lesions. METHODS: Retrospective analysis of conventional ultrasound images of 150 patients with parotid lesions by two observers. Parotid lesions were classified into seven patterns and then categorised into eight grades: Grade 0, unsatisfied illustration on ultrasound; Grade 1, normal parotid gland; Grade 2, definitively benign; Grade 3, probably benign; Grade 4, indeterminate; Grade 5, probably malignant; Grade 6, highly suggestive malignant and Grade 7, already had malignant diagnosis. Combined with the pathological results, the conventional ultrasound diagnostic grade of parotid lesions was evaluated for predicting the probability of malignancy. RESULTS: There was excellent interobserver agreement of both readers for patterns and grades (K = 0.89 and 0.90, p < 0.01). The proportions of the malignancies in conventional ultrasound Grade 2, 3, 4, 5 and 6 according to the two readers 0 and 0, 0 and 0, 8.7% and 8.8%, 54.2% and 50%, 100% and 100%, respectively. The sensitivity, specificity and area under ROC curve were 64.0%, 91.2%, 0.809 and 64.0%, 89.6%, 0.802, respectively, using Grade 5 of the two readers as the best grade for diagnosing benign and malignant parotid lesions. CONCLUSION: The conventional ultrasound diagnostic grade of parotid lesions can be used to evaluate the risk of malignancy and will be helpful to improve the imaging diagnosis and clinical treatment.
OBJECTIVES: To provide a graded diagnosis of benign and malignant lesions in the parotid gland by conventional ultrasound, and thus to predict the probability of malignancy of the lesions. METHODS: Retrospective analysis of conventional ultrasound images of 150 patients with parotid lesions by two observers. Parotid lesions were classified into seven patterns and then categorised into eight grades: Grade 0, unsatisfied illustration on ultrasound; Grade 1, normal parotid gland; Grade 2, definitively benign; Grade 3, probably benign; Grade 4, indeterminate; Grade 5, probably malignant; Grade 6, highly suggestive malignant and Grade 7, already had malignant diagnosis. Combined with the pathological results, the conventional ultrasound diagnostic grade of parotid lesions was evaluated for predicting the probability of malignancy. RESULTS: There was excellent interobserver agreement of both readers for patterns and grades (K = 0.89 and 0.90, p < 0.01). The proportions of the malignancies in conventional ultrasound Grade 2, 3, 4, 5 and 6 according to the two readers 0 and 0, 0 and 0, 8.7% and 8.8%, 54.2% and 50%, 100% and 100%, respectively. The sensitivity, specificity and area under ROC curve were 64.0%, 91.2%, 0.809 and 64.0%, 89.6%, 0.802, respectively, using Grade 5 of the two readers as the best grade for diagnosing benign and malignant parotid lesions. CONCLUSION: The conventional ultrasound diagnostic grade of parotid lesions can be used to evaluate the risk of malignancy and will be helpful to improve the imaging diagnosis and clinical treatment.
Authors: V Cantisani; E David; P S Sidhu; B Sacconi; A Greco; F Pandolfi; M Tombolini; L Lo Mele; F Calliada; L Brunese; C Catalano; M De Vincentiis; N Di Leo; G Ascenti; F D'Ambrosio Journal: Ultraschall Med Date: 2016-06-14 Impact factor: 6.548