Medullary thyroid cancer (MTC) is an uncommon and challenging malignancy. In spite of aggressive primary treatment with total thyroidectomy and modified neck lymph node dissection, approximately 50 % of the patients have persistent or recurrent disease [1,2]. The recent European Thyroid Association guidelines [3] recommend surgery and other local treatment modalities (such as external beam radiation therapy, radiofrequency ablation, and chemoembolization) for the treatment of distant metastases in patients with low tumor burden (distant metastases limited to single organ) and stable disease, while patients with symptoms, significant tumor burden and progressive disease should receive novel targeted systemic therapies. However, imaging of recurrent disease in MTC is difficult and still poses a major problem in the management of the disease.The serum tumor markers calcitonin (Ct) and carcinoembryonic antigen (CEA) and their doubling times (Dt) are used to assess progression rate and also reflect tumor burden in MTC. However, conventional imaging (CI) is often negative in patients with Ct concentrations \150 pg/ mL [3]. It is estimated that CI identifies metastases in about 40 % of patients with biochemical evidence of recurrence. The probability of detecting metastases increases in patients with increasing Ct and CEA concentrations and shortening Dts. Of note, some patients with poorly differentiated and aggressive metastatic MTC have low Ct or discrepantly high CEA concentrations. The recommended CI protocol is extensive and comprises neck ultrasonography for detection of cervical lymph node metastases, computed tomography (CT) of thorax for lung and mediastinal lesions, magnetic resonance imaging (MRI) for liver and bone metastases of the spine and pelvis and bone scintigraphy. Modern whole body imaging with positron emission tomography (PET) and integrated CT could thus represent an alternative to CI, at least for patients with negative CI imaging.In this issue of the Endocrine, Treglia et al.[4] report a meta-analysis consisting of 24 studies comprising altogether 538 patients on the performance of fluorine-18 fluorodeoxyglucose ( 18 F-FDG) PET or PET/CT in detection of recurrent MTC. The pooled detection rate on a per patient-based analysis was 59 % (range 24-95 %). The authors conclude that 18 F-FDG-PET/(CT) is non-optimal in surveillance of MTC since about 40 % of suspected recurrences remain unidentified. Of note, the definition of false negative and true negative imaging results differed. Some considered patients with increased Ct concentrations and negative 18 F-FDG-PET/other imaging as false negative results, while other considered them as true negative. However, a detection rate of 59 % should be related to the corresponding overall detection rates of CI of about 40 %. The current meta-analysis did not present data on whether the 18 F-FDG-PET studies primarily included patients for whom CI remained negative or not. In clinical practice, 18 F-FDG-PET is performed in patients with biochemical proof of re...