In this work the effects of citric acid and of two common fillers, lactose (soluble) and tricalcium phosphate (insoluble) are examined on the release profiles from pellets, using ibuprofen as a model drug with pH-dependent solubility. Also studied is the dependence of these profiles on the specific surface area, bulk density, apparent density, porosity and porosity parameters (pore size distribution, total pore surface area, mean pore diameter and pore shape), as determined by mercury intrusion porosimetry. Pellets with high porosity and total pore surface area but small median pore diameter (tricalcium phosphate pellets-IPM) are found to produce similar dissolution results to those of low porosity and low total pore surface area, but having a high median pore diameter (lactose pellets-ILM), irrespective of the solubility of excipients. Addition of citric acid causes a delay in the initial dissolution for both formulations. During dissolution, however, citric acid reduces the median pore diameter of lactose-based pellets. In contrast, in tricalcium phosphate/citric acid pellets (CIPM), this parameter increases considerably during dissolution, when compared to the IPM formulation. These findings may justify the contrasting dissolution behaviors of CIPM and CILM (lactose/citric acid) pellets, after their common behavior in the initial stages, and show that porosity and its related parameters, along with physical properties of excipients such as solubility, density and specific surface area, are helpful to predict pellet behavior in drug release profiles.