Prognostic ability of simplified nuclear grading of renal cell carcinoma.

BACKGROUND: The Fuhrman grading system is an established predictor of survival in patients with renal cell carcinoma (RCC). The predictive accuracy of various Fuhrman grading schemes was tested with the intent of improving the prediction of RCC-specific survival (RCC-SS). METHODS: The analyses targeted 5453 patients from 14 institutions. Univariable, multivariable, and predictive accuracy analyses addressed RCC-SS. The statistical significance of the gain in predictive accuracy was quantified with the Mantel-Haenszel test. RESULTS: The median follow-up time was 4.5 years. In both univariable and multivariable analyses, Fuhrman grade achieved independent predictor status regardless of the coding scheme. When Fuhrman grade was not considered in multivariable analyses, the predictive accuracy was 83.8%. Addition of Fuhrman grade to the multivariable model resulted in predictive accuracy gains of 0.8% for all 3 grading schemes tested. CONCLUSION: Fuhrman grade must to be considered when RCC-SS is assessed. However, modified or conventional Fuhrman grading schemes perform equally well as the conventional grading system.