Thank you for the very interesting comments on the statistical methodology related to the article “Fine needle aspiration cytology and frozen section in the diagnosis of malignant parotid tumours” by Fakhry et al. It is indeed a frequently encountered bias in studies evaluating the performance of FNAC. However, as the authors of the comments point out, this bias is valid only for studies where all tumours are not operated on. This is the case for example in thyroid tumours, which sometimes cannot be operated on when the preoperative evaluation indicates a benign tumour. This is not the case for parotid tumours. FNAC alone cannot be used to guide surgical management, and imaging – magnetic resonance imaging (MRI) in particular, with the use of new sequences (perfusion and diffusion-weighted sequences) – now has a very important place in the management of parotid masses in combination with FNAC. FNAC and MRI are important examinations that provide valuable information for the preoperative diagnostic work-up and together alert the surgeon to the possible presence of malignancy; however, in view of the considerable false-negative rate, FNAC and MRI cannot formally exclude malignancy and therefore cannot reliably reassure the surgeon and the patient. Moreover the majority of parotid tumours, both benign and malignant, are not homogeneous. Therefore careful analysis of the entire tumour specimen after surgery is required before a final diagnosis can be established. Finally, considering the high rate of malignancy in parotid tumours (26–32%), apart from a very few selected cases (patients in poor general health, or very old patients, for example), histological examination after parotidectomy is always necessary.