Purpose To determine the diagnostic performance of three-dimensional (3D) MRI for the depiction and characterization of cartilage defects within the knee joint by using arthroscopy and/or open surgery as the standard of reference. Materials and Methods A systematic literature search was performed to extract diagnostic studies published between January 1985 and October 2017. Two independent investigators assessed the methodologic quality of each study by using Quality Assessment of Diagnostic Accuracy Studies 2. Bivariate random-effects model was used to compare the diagnostic odds ratio (DOR) of 3D and two-dimensional (2D) MRI for helping to detect knee cartilage defects and to assess the effect of relevant covariates on diagnostic performance of 3D MRI. Meta-regression analysis was performed to assess DOR of 3D MRI during the last 3 decades. Results Twenty-seven studies (composed of 1710 MRI examinations) were included. Of those, 16 (59%) studies compared the diagnostic performance of 3D and 2D MRI. The diagnostic performance of 3D MRI statistically significantly improved over the last 3 decades (P = .003). Three-dimensional MRI obtained by using 3.0-T field strength had higher DOR relative to 1.5-T or lower field strength (relative DOR, 4.05; P = .01). Three-dimensional multiplanar reformation was associated with higher specificity (P = .001) compared with conventional axial, sagittal, and coronal 2D MRI planes. Three-dimensional fast-spin-echo sequences provided higher sensitivity and specificity (P < .05) than did 2D MRI. Conclusion Three-dimensional MRI currently provides comparable diagnostic performance to two-dimensional MRI, with improvement in diagnostic performance achieved by using 3.0-T field strength, three-dimensional fast-spin-echo sequences, and multiplanar reformation. © RSNA, 2018 Online supplemental material is available for this article.