Purpose <p>Knowledge of direct vertebral rotation (DVR) in idiopathic scoliosis is largely based on small studies. By analyzing pooled data from relevant studies, this review aimed to explore the impact of DVR on the clinical and radiographic outcomes of idiopathic scoliosis patients.</p> Methods <p>Studies that met the following inclusion criteria were included: idiopathic scoliosis, use of DVR in one group plus a comparative group (non-DVR), and available pre- and postoperative data. A study was excluded if the two groups were not similar before the intervention. The PubMed and Web of Science databases were used to identify studies and were last searched in January 2025. The risk of bias was assessed via the ROBIN-I tool. The outcomes were based on imaging as well as scoliometer readings and Scoliosis Research Society (SRS) scores. Screw-related adverse events were noted. Random-effects model meta-analyses were used.</p> Results <p>Sixteen studies with 931 participants were included. Vertebral rotation and trunk rotation were reversed more in the DVR group than the control group, as were the mean differences in degrees and confidence intervals (MD 4.71, CI 1.95 to 7.46, <i>p</i> = 0.001, I² = 84.1%; studies = 9) and (MD 1.19, CI 0.18 to 2.19, <i>p</i> = 0.02, I² = 53.8%; studies = 5), respectively. The coronal Cobb angle and thoracic kyphosis angle were similar between the groups (MD 1.70, CI -0.54 to 3.93, <i>p</i> = 0.14, I² = 74.8%; studies = 16), (MD 0.40, CI -0.94 to 1.75, <i>p</i> = 0.56, I² = 52% studies = 14), respectively. The SRS score was not improved by DVR (MD -0.05, CI -0.25 to 0.14, <i>P</i> = 0.59, I² = 75.1%; studies = 3). The studies included did not report screw-related complications in the DVR group.</p> Conclusion <p>In addition to a marginal correction in the axial plane, DVR did not affect other dimensions of the deformity, nor did it affect self-reported outcomes.</p>