ObjectiveIn this multicenter study, we investigated the management algorithm for thyroid nodules with benign cytology using US features from data collected from 7 institutions.Materials and MethodsThe institutional review board approved this retrospective study. Data on 700 focal thyroid nodules in 673 consecutive patients were collected from 7 university-affiliated hospitals. This study included nodules that were diagnosed as benign at initial cytologic evaluation and that underwent pathologic or follow-up study. The risk of malignancy was compared according to the US assessments of each institution as well as looking at all the nodules together as a whole.ResultsOf the 700 nodules, 688 (98.3%) were benign and 12 (1.7%) were malignant. If initial cytologic results were benign, the likelihood of the nodule actually being malignant was from 1 to 3%, varying by institution. The likelihood of a cytologically benign nodule with positive US being malignant (4.7%, 8/169) was higher than that of one without positive US (0.8%, 4/531) (P = .002).ConclusionBased on our multicenter study, repeat FNA should be performed in thyroid nodules with initial benign cytology showing suspicious US features in order to decrease the number of false negative cases.