TY - JOUR T1 - Solitary pulmonary nodule: How accurate are our predictions? JF - European Respiratory Journal JO - Eur Respir J VL - 44 IS - Suppl 58 SP - P2731 AU - Juliana Lopes AU - Ana Machado AU - Pedro Leuschner AU - Miguel Ricardo Y1 - 2014/09/01 UR - http://erj.ersjournals.com/content/44/Suppl_58/P2731.abstract N2 - Background: To manage patients with solitary pulmonary nodules (SPN) one depends critically on estimating its probability of malignancy. This can be accomplished through one of several proposed mathematical models.Objectives: To analyse clinical and radiological findings that predict the histology of SPN; to validate a previously developed model to estimate the probability of malignancy; and to develop a standard approach to SPN in our centre.Methods: A retrospective study involving 75 patients with SPN diagnosed at a single centre, between 2010-13. Demographical, clinical and radiological data were assessed, and the patients' final diagnosis was compared with the probability of malignancy, as predicted by the Mayo Clinic model (MCM). The accuracy of the model was assessed by calculating areas under the receiver operating characteristic (ROC) curve and calibrated by comparing predicted and observed rates of malignancy.Results: We found a significantly relation between SPN malignancy and maximum transverse diameter (p= 0,004 for diameter >20 mm) and a positive non-significant association between SPN malignancy and age >65 years, previous history of malignancy, spiculated margins and upper lobe localization. The area under the ROC curve for this model was 0,662 ± 0,068 (IC95%: 0,529 – 0,794).Conclusions: SPN size was significantly associated with the probability of malignancy. MCM proved sufficiently accurate in predicting SPN malignancy probability. The lack of a standard approach is probably accountable for the observed variability in the chosen strategy. Differences in geographical and institutional prevalence may warrant a specific approach for each setting. ER -