Bevacizumab is a novel treatment for the recurrent high-grade gliomas (rHGG). However, only a subset of the patients shows response to the bevacizumab treatment and the response evaluation using conventional criteria is difficult. The purpose of our study was to evaluate the early response for rHGG treated with bevacizumab using volumetric analysis of diffusion-weighted imaging (DWI). Twenty-nine patients who received bevacizumab therapy for rHGG were included in our study. All patients received a conventional MRI scan with DWI before and after the initial bevacizumab dose. For each MRI, we measured the total volume of the T2 hyperintense lesion (H(T2)) of the rHGG, the volume of foci with a lower ADC value than that of the normal cortex (L(ADC)), and the proportion of L(ADC) to H(T2) (L(ADC)/H(T2)). The Changes in the H(T2), L(ADC) and L(ADC)/H(T2) after bevacizumab treatment were also determined. Thereafter, those volumetric data were compared to the progression free survival (PFS). After the analyses, we found a significant negative correlation between the PFS and the L(ADC) for the post-bevacizumab ADC maps (r = -0.413, P = 0.026). The patients with an L(ADC) of <2.5 cm(3) showed a longer PFS than those with an L(ADC) of ≥2.5 cm(3) (median = 135 vs. 91 days, P = 0.002) on the post-bevacizumab ADC maps. A multiple linear regression analysis revealed that only the post-bevacizumab L(ADC) was a significant predictor of the PFS (P = 0.026). In conclusion, the post-bevacizumab L(ADC) can be used for an early response evaluation and can predict the PFS for rHGG patients treated with bevacizumab.