The researchers study looks back at patients who had 3 Tesla multi-parametric MRI for further evaluation of small renal tumors, using a standardized scanning and reporting protocol that took into account parameters like T2 HASTE signal intensity, contrast enhancement ratios, apparent diffusion coefficient, and the presence of microscopic/macroscopic fat. Before comparing MRI results to a histopathologic diagnosis, a uniform reporting system was implemented for all scans. Analysis of 2×2 tables (sensitivity, specificity, PPV, and NPV), Fisher’s exact test, and the likelihood ratio Descriptive statistics and an exact test were employed to analyze the correlation between mpMRI findings suggestive of malignancy and subsequent histology. Over the course of 5 years, 67 patients with 75 renal masses were included. Histopathology confirmed 70 tumors (5 had pathognomonic findings for angiomyolipomas; biopsy was therefore considered unethical, so these were included without histopathology). About 1 patient was discovered to have an organized hematoma rather than a tumor, while 2 others had no diagnostic results. Thus, 72 cases were included in the statistical analysis (64 patients, with 7 patients having multiple tumors). All subsequent statistics will focus on tumors, not patients unless otherwise noted. Around 28 (31.8%) were classified as benign, whereas 52 (72.2%) were classified as suspicious or malignant. Confirmation of renal cell carcinoma was achieved in 51 instances (70.8%). Fisher’s exact test showed a p-value for the correlation between MRI malignancy suspicion and histopathology of less than 0.0001. MRI had a sensitivity of 96.1%, a negative predictive value of 90%, a specificity of 85.7%, and a positive predictive value of 94.2%. The de Silva St. George’s categorization method showed promise in distinguishing benign from malignant solid renal masses, suggesting it could be useful in forecasting the possibility of cancer to evaluate the necessity for biopsy or removal. However, there needs to be more testing done before this reporting method can be recommended for clinical usage.
Source: bmcurol.biomedcentral.com/articles/10.1186/s12894-022-01082-9
