A systematic review was undertaken to select studies that compared the accuracy of 2-[18F]-fluoro-2-deoxy-D-glucose positron emission tomography with computed tomographic imaging in detecting mediastinal lymph node metastases in patients with nonsmall cell lung cancer. Two authors selected relevant articles according to predefined criteria. With a meta-analytic method, summary receiver operating characteristic curves were constructed. The point on the receiver operating characteristic curve with equal sensitivity and specificity for 2-[18F]-fluoro-2-deoxy-D-glucose positron emission tomography was Q* = 0.90 (95% confidence interval [CI], 0.86 to 0.95). For computed tomography it was 0.70 (95% CI, 0.65 to 0.75). The difference was highly significant (p < 0.0001). We conclude that 2-[18F]-fluoro-2-deoxy-D-glucose positron emission tomography is more accurate than computed tomography in detecting mediastinal lymph node metastases.