Abstract: Purpose: To use clinical and laboratory tests instead of kidney biopsy for the diagnosis of type IV WHO renal involvement in SLE. Methods: Patients with SLE were divided in two groups according to their kidney biopsy: type IV WHO (110 patients) and other forms (98 patients). Each group was further divided in two groups: the learning sample and the validation ample. The classification and regression tree method was used for the analysis of the data (38 clinical and laboratory parameters). Results: The simplest tree having the best accuracy was chosen as bellow. The tree had a sensitivity of 75%, a specificity of 94%, and an accuracy of 84.5%.