Resumo Objetivo: Comparar protocolos automatizados e manuais de ressonância magnética para estimar a concentração hepática de ferro em 1,5 T. Materiais e Métodos: Foi realizada ressonância magnética hepática em 53 pacientes com suspeita de sobrecarga de ferro hepática e 21 controles, seguida da estimativa cega da concentração hepática de ferro por dois examinadores usando mapas automáticos T2* e T1, assim como o manual T2* e o método signal-intensity-ratio. O desempenho foi medido usando curvas ROC e a correlação interobservador e intraobservador usando o coeficiente de correlação intraclasse bidirecional. Resultados: O desempenho da curva ROC separando pacientes e controles mostrou áreas sob a curva de 0,912 para o mapa automático T2*, 0,934 para o método signal-intensity-ratio, 0,908 para manual T2* e 0,80 para mapa T1 (este difere significativamente dos outros três métodos). Houve boa correlação interobservador e intraobservador (coeficiente de correlação intraclasse entre 0,938 e 0,998; p < 0,05). Correlações envolvendo o mapa T1, embora ainda significativas, foram menores. Conclusão: Em 1,5 T, o mapa T2* representa uma nova ferramenta rápida e promissora para avaliar o diagnóstico de sobrecarga de ferro hepática, enquanto o mapa T1 mostrou menor precisão. O desempenho do mapa T1 foi menor que o dos métodos T2*.

Abstract Objective: To compare automated and manual magnetic resonance imaging protocols for estimating liver iron concentrations at 1.5 T. Materials and Methods: Magnetic resonance imaging examination of the liver was performed in 53 patients with clinically suspected hepatic iron overload and in 21 control subjects. Liver iron concentrations were then estimated by two examiners who were blinded to the groups. The examiners employed automated T2* and T1 mapping, as well as manual T2* and signal-intensity-ratio method. We analyzed accuracy by using ROC curves. Interobserver and intraobserver agreement were analyzed by calculating two-way intraclass correlation coefficients. Results: The area under the ROC curve (to discriminate between patients and controls) was 0.912 for automated T2* mapping, 0.934 for the signal-intensity-ratio method, 0.908 for manual T2*, and 0.80 for T1 mapping, the last method differing significantly from the other three. The level of interobserver and intraobserver agreement was good (intraclass correlation coefficient, 0.938-0.998; p < 0.05). Correlations involving T1 mapping, although still significant, were lower. Conclusion: At 1.5 T, T2* mapping is a rapid tool that shows promise for the diagnosis of liver iron overload, whereas T1 mapping shows less accuracy. The performance of T1 mapping is poorer than is that of T2* methods.

Mucoceles are cystic masses that generally affect the sinuses. It occurs as a result from obstruction of the ostium of a sinus and consequential accumulation of mucus. Frontal and ethmoid sinuses are mostly affected. Usually, the clinical symptoms are insidious, varying with the extent of the affected region. The treatment is surgical and endoscopic surgery is the method of choice in most cases. The present study is aimed at describing the main characteristics of paranasal sinuses mucoceles, demonstrating and illustrating a series of atypical presentations with emphasis on imaging findings.

Trata-se de um relato de caso de tumor de células gigantes (TCG) em falange proximal de terceiro dedo da mão esquerda com metástase pulmonar. A paciente apresentava dor no dedo sem história prévia de trauma. Foram realizados exame clínico, estudo radiográfico e ressonância nuclear magnética. Feito o estudo histológico, a partir de biópsia incisional, com hipótese de TCG. Foi submetida à amputação do dedo, confirmando o diagnóstico pela microscopia da peça. A paciente foi acompanhada devido ao risco de metástase pulmonar, evidenciada em estudo radiográfico e tomografia computadorizada de tórax, sendo submetida à toracotomia. Desde então, houve melhora dos sintomas referidos no pré-operatório e ausência de recidiva local e novas metástases.

This is a case report on a giant cell tumor in the proximal phalanx of the third finger of the left hand, with pulmonary metastasis. The patient presented pain in the finger without any previous history of trauma. Clinical examination, radiographic imaging and magnetic resonance imaging were carried out. A histological evaluation was done from an incisional biopsy, taking the hypothesis of giant cell tumor. The patient underwent amputation of the finger and the diagnosis was confirmed by means of microscopy on the specimen. The patient was followed up because of the risk of lung metastasis, which was shown by radiographic examination and computed tomography on the chest, and thoracotomy was performed. Since then, there has been an improvement of the symptoms that had been reported preoperatively, and no local recurrence or new metastasis has been found.