Objective: To systematically assessed the diagnostic performance of contrast-enhanced computed tomography (CT) compared to other imaging modalities for diagnosing and staging renal-cell carcinoma in adults.
Methods: A comprehensive literature search was conducted through various electronic databases. Data from the selected studies were extracted and pooled, and median sensitivity and specificity were calculated wherever possible. Forty studies analyzing data of 4354 patients were included. They examined CT, magnetic resonance imaging (MRI), positron emission tomography-CT, and ultrasound (US).
Results: For CT, median sensitivity and specificity were 88% (interquartile range [IQR] 81%-94%) and 75% (IQR 51%-90%), and for MRI they were 87.5% (IQR 75.25%-100%) and 89% (IQR 75%-96%). Staging sensitivity and specificity for CT were 87% and 74.5%, while MRI showed a median sensitivity of 90% and specificity of 75%. For US, the results varied greatly depending on the corresponding technique. Contrast-enhanced US had a median diagnostic sensitivity of 93% (IQR 88.75%-98.25%) combined with mediocre specificity. The diagnostic performance of unenhanced US was poor. For positron emission tomography-CT, diagnostic accuracy values were good but were based on only a small amount of data. Limitations include the strong heterogeneity of data due to the large variety in imaging techniques and tumor histotypes. Contrast-enhanced CT and MRI remain the diagnostic mainstay for renal-cell carcinoma, with almost equally high diagnostic and staging accuracy.
Conclusion: For specific questions, a combination of different imaging techniques such as CT or MRI and contrast-enhanced US may be useful. There is a need for future large prospective studies to further increase the quality of evidence.
Keywords: Computed tomography; Kidney cancer; MRI; PET; Ultrasound.
Copyright © 2018 Elsevier Inc. All rights reserved.