Efficiently removing blood from the brain vasculature is critical to evaluate accurately the brain penetration and biodistribution of drug candidates, especially for biologics as their blood concentrations are substantially higher than the brain concentrations. Transcardial perfusion has been used widely to remove residual blood in the brain; however, the perfusion conditions (such as the perfusion rate and time) reported in the literature are quite varied, and the performance of these methods on blood removal has not been investigated thoroughly. In this study, the effectiveness of the perfusion conditions was assessed by measuring brain hemoglobin levels. Sodium nitrite (NaNO2) as an additive in the perfusate was evaluated at different concentrations. Blood removal was significantly improved with 2% NaNO2 over a 20 min perfusion in mouse without disrupting the integrity of the blood‐brain barrier (BBB). In mice, the optimized perfusion method significantly lowered the measured brain‐to‐plasma ratio (Kp,brain) for monoclonal antibodies due to the removal of blood contamination and small molecules with a moderate‐to‐high BBB permeability and with a high brain‐unbound‐fraction (fu,brain) presumably due to flux out of the brain during perfusion. Perfusion with or without NaNO2 clearly removed the residual blood in rat brain but with no difference observed in Kp,brain between the perfusion groups with or without 2% NaNO2. In conclusion, a perfusion method was successfully developed to evaluate the brain penetration of small molecules and biologics in rodents for the first time. The transcardial perfusion with 2% NaNO2 effectively removed the residual blood in the brain and significantly improved the assessment of brain penetration of biologics. For small molecules, however, transcardial perfusion may not be performed, as small molecule compounds could be washed away from the brain by the perfusion procedure.