In this work, we present a quantitative (statistical) 3D morphological characterization of optical fibers used in electric-field sensing. The characterization technique employs propagation-based x-ray phase-contrast microcomputed tomography (micro-CT). In particular, we investigate specialty optical fibers that contain microstructured holes that are electro-optically modified by thermal poling to induce second-order nonlinear effects (SONE). The efficiency of the SONE is reflected in the characterization parameter, V-pi, which is highly dependent on the dimensions of the fiber. The fiber microstructures must be uniform to support the fabrication of reproducible devices. The results obtained using the micro-CT technique show that uncertainty of +/- 1.7% arises in the determination of the expected value of the voltage that causes a change in the phase of the electromagnetic wave equal to pi rad (V-pi), demonstrating a great advantage, compared with other techniques e.g. SEM, which would need at least 1000 images of the cross-section of an optical fiber, taken at different points, making the process more expensive and time-consuming.