In this study, we present an application of EVA descriptors for a QSAR model of inhibition of carbonic anhydrase isozyme CA II by an heterogeneous set of 66 sulfonamide compounds. For each of the compounds, geometry optimization and frequency calculations have been performed using the DFT/B3LYP level of the theory in conjugated with the 6-31G* basis set. Different numbers of EVA descriptors for each structure were produced by applying various values of Gaussian kernel of a fixed standard deviation, sigma (cm(-1)) and sampled at fixed increments of L (cm(-1)) during the evaluation of the descriptors based on their vibrational frequencies. The set of compounds was divided into two subsets. The first subset contained the 22 compounds that were used as the test compounds. The remaining 44 compounds were used as the training set. Several QSAR models have been developed using these calculated EVA descriptors and the carbonic anhydrase isozyme CA II inhibitory data (K(i)) of the compounds. Among the QSAR models evaluated, the one that produced the best statistical results had the parameters sigma and L both equal to 5 cm(-1). This model produced correlation values (R(2)) of 0.777 and 0.616 for the training and test sets, respectively. The results of this study showed that EVA descriptors perform well as explanatory and predictive tools for modeling the inhibition activity of carbonic anhydrase by a set of sulfonamide compounds.