Objective To investigate short-term effectiveness of spinal navigation with the intra-operative three-dimensional (3D)-imaging modality in pedicle screw fixation for congenital scoliosis (CS). Methods Between July 2010 and December 2011, 26 patients with CS were treated. Of 26 patients, 13 patients underwent pedicle screw fixation using the spinal navigation with the intra-operative 3D-imaging modality (navigation group), while 13 patients underwent the conventional technique with C-arm X-ray machine (control group). There was no significant difference in gender, age, hemivertebra number and location, major curve Cobb angle, and Risser grade between 2 groups (P gt; 0.05). Operation time, operative blood loss, frequency of the screw re-insertion, and postoperative complication were observed. The pedicle screw position was assessed by CT postoperatively with the Richter’s standard and the correction of Cobb angle was assessed by X-ray films. Results All patients underwent the surgery successfully without major neurovascular complication. There was no significant difference in operation time, operative blood loss, and pedicle screw location between 2 groups (P gt; 0.05). A total of 58 screws were inserted in navigation group, and 3 screws (5.2%) were re-inserted. A total of 60 screws were inserted in control group, and 10 screws (16.7%) were re-inserted. There was significant difference in the rate of pedicle screw re-insertion between 2 groups (χ2=3.975, P=0.046). Patients of navigation group were followed up 6-24 months, and 6-23 months in control group. According to Richter’s standard, the results were excellent in 52 screws and good in 6 screws in navigation group; the results were excellent in 51 screws, good in 5 screws, and poor in 4 screws in control group. Significant difference was found in the pedicle screw position between 2 groups (Z= — 1.992, P=0.046). The major curve Cobb angle of 2 groups at 1 week and last follow-up were significantly improved when compared with preoperative value (P lt; 0.05), but there was no significant difference between 1 week and last follow-up (P gt; 0.05). No significant difference in correction rate of the major curve Cobb angle was found between 2 groups at last follow-up (t=0.055, P=0.957). Conclusion Spinal navigation with the intra-operative 3D-imaging modality can improve the accuracy of pedicle screw implantation in patients with CS, and effectually reduce the rate of screw re-insertion, and the short-term effectiveness is satisfactory.