engleski

Analysis of Errors in Ultrasonographic Detection and Characterisation of Renal Tumors

Ultrasound (US) is performed by various specialists and general practitioners. However, it is very operator-dependent method and errors in detection and characterisation of renal tumors are common. We analyse the number and types of errors in US findings encountered in patients who had renal tumors diagnosed in our institution. Materials and Methods: Medical history was analyzed retrospecitvely of 68 patients (44 m, 24 f, age range 5-82 years) in whom renal tumors were diagnosed in our institution by US and CT in a four-years period. Only those patients were included in the study who had one or more renal US examinations performed in other institutions, private practices, or in our institution by non-radiologists, 1 day to 12 months prior to examination in our department. 66 solid renal lesions and 6 solid-cystic renal lesions were observed, with the size range 1.8 – 9 cm. 55 patients had renal cell carcinoma, 4 transitional cell carcinoma, two Willms tumors, 1 multilocular cystic nephroma, five angiomyolipoma, one oncocytoma. All were operated and diagnosis confirmed histopathologically, except four patients with typical US and CT findings of AML. Logiq 9 scanner (GE Healthcare) was used in our department with a variety of multifrequency transducers equiped with compound and native-harmonic imaging capabilities, color and power Doppler, and our US findings were compared to previous ones. Previous US findings were cathegorized as: (1) correct recognition of solid renal resion ; (2) complete non-visualization of solid lesion ; (3) erroneous characterization of visualized lesions. Results: In 42 patients (61.8%) solid lesions were visualized, and patients were referred for further imaging or urologic examination with a high suspicion or definite diagnosis of renal tumor. In 12 patients (17.6%) solid lesions were not visualized at all. Those lesions were 1.8-6 cm in size. Nine patients had renal carcinoma, two patients had TCC, and one had AML. In 14 patients (20.6%) lesions were observed, but were erroneously characterized. Two AMLs were interpreted as renal cancers, four renal carcinomas were interpreted as AMLs, 1 MLCN was misinterpreted as cyst, and seven renal cancers were interpreted as &laquo ; most probably&raquo ; or definitely renal cyst, including the largest observed renal cancer that had 9 cms in diameter and was cleary hypervascular on color and power Doppler. Some of these patients were seen several times by two or more different US examiners. In 26 patients with wrong initial diagnosis, examination was performed most commonly by general practitioners (20), but also by nephrologists (4), urologists (4), gastroenterologists (3), abdominal surgeons (2), and in two patients by radiologists. The diagnosis was delayed up to one year, and one patient was even referred for the laparoscopic removal of a large renal cyst, that was actually a large clear-cell renal carcinoma. Conclusion: Non-visualization or erroneous characterization of solid renal lesions on US is very common in our clinical practice, particulary when examinations are performed by insuffitiently trained physicians and non-radiologists. Good education in US and experience are essential for proper visualisation and characterisation of renal tumors.

renal tumors; ultrasound; errors

nije evidentirano