Overcoming anatomical distortion in complex and severe spinal deformity via a novel self-stabilizing mortise-tenon design 3D-printed guide template: a retrospective before-and-after cohort study

Complex and severe spinal deformity (CSSD) causes marked anatomical distortion, complicating pedicle screw placement and three-column osteotomy (3-CO). Evidence supporting patient-specific 3D-printed guides in this setting remains limited. We compared a digital-planning and mortise–tenon 3D-printed osteotomy guide workflow with conventional freehand surgery. We retrospectively reviewed 40 patients with CSSD who underwent posterior 3-CO at a tertiary center between 2020 and 2024. Patients treated from January 2020 to December 2021 underwent conventional freehand surgery (CG), whereas those treated from January 2022 to January 2024 received the combined workflow (TG). Co-primary outcomes were the major deformity angle (MA) correction rate, adjusted for preoperative MA using analysis of covariance, and acceptable adjacent-segment pedicle screw placement, defined as Gertzbein–Robbins Grade A or B and analyzed using mixed-effects logistic regression. The MA correction rate was higher in TG than CG (58.5% ± 13.1% vs. 46.5% ± 12.7%, P = 0.006), with an adjusted between-group difference of 12.8% points (95% CI 4.5–21.0, P = 0.003). Patient-level adjacent-segment screw accuracy was also higher in TG (98.8% ± 5.6% vs. 85.0% ± 12.6%, P < 0.001; mixed-effects OR 10.42, 95% CI 1.93–56.32, P = 0.006). Preoperative sagittal vertical axis (SVA) was greater in TG (30.8 ± 26.6 vs. 7.4 ± 18.5 mm, P = 0.003), but SVA change did not differ significantly between groups ( P = 0.197). Neurological complications occurred in 3 TG and 7 CG patients ( P = 0.273). In TG, planned and executed osteotomy angles showed agreement, with matching rates above 91%, mean signed biases of − 2.35° coronally and − 4.19° sagittally, and mean absolute errors of 3.99° and 5.51°, respectively. In this selected retrospective historical-control cohort, the combined workflow was associated with greater early MA correction and higher adjacent-segment screw accuracy than freehand surgery. Plan-to-execution findings support internal agreement but do not establish superior execution over controls. Prospective studies are needed to determine the workflow’s independent contribution, long-term benefits, and neurological safety. Not applicable. This is a retrospective observational study and was not registered as a clinical trial.