Multimodal image fusion-based reconstruction combined with 3D printing for preoperative planning in intracranial tumor resection: a retrospective comparative study

Abstract Objective To evaluate the clinical utility of multimodal image fusion-based three-dimensional reconstruction combined with 3D printing for preoperative planning in intracranial tumor surgery, with particular attention to its role in supratentorial tumors and cerebellopontine angle (CPA) tumors. Methods We retrospectively reviewed 41 consecutive patients with intracranial tumors who underwent microsurgical resection at our institution between January 2022 and December 2024. According to the preoperative planning strategy, patients were assigned to a conventional group (n = 18) or a multimodal-planning group (n = 23). In the conventional group, surgical planning was based on routine imaging assessment. In the multimodal-planning group, routine imaging was supplemented by multimodal image fusion, three-dimensional reconstruction, and 3D printing-assisted planning. Among 20 patients with supratentorial tumors, 3D-printed positioning guides were mainly used for surface localization, craniotomy design, and cortical venous reconstruction. Among 21 patients with CPA tumors, multimodal reconstruction focused on the tumor, bony structures, vessels, and relevant cranial nerves, followed by fabrication of patient-specific 3D-printed physical models. All procedures were performed under microscopy with intraoperative electrophysiological monitoring and ultrasonic aspiration; fluorescence guidance was additionally used in selected glioma and metastatic tumor cases. Perioperative variables, extent of resection, and postoperative complications were compared between groups. Results All 41 patients underwent surgery successfully. Compared with the conventional group, the multimodal-planning group had a shorter operative time (4.86 ± 1.19 h vs. 5.92 ± 1.38 h), lower intraoperative blood loss (298.7 ± 136.4 mL vs. 428.6 ± 171.9 mL), and a shorter postoperative hospital stay (8.8 ± 2.5 days vs. 10.9 ± 3.0 days) (all P < 0.05). The rate of intraoperative secondary adjustment of the incision/craniotomy was lower in the multimodal-planning group (4.3% vs. 27.8%), although the difference did not reach statistical significance. Gross total resection was achieved more frequently in the multimodal-planning group (82.6% vs. 61.1%), while the overall complication rate was lower (13.0% vs. 33.3%) (both P < 0.05). In supratentorial tumors, preservation of important cortical veins was more frequent in the multimodal-planning subgroup (90.9% vs. 55.6%, P < 0.05). In CPA tumors, the multimodal-planning subgroup showed a higher gross total resection rate and lower rates of new postoperative facial nerve dysfunction and dysphagia (all P < 0.05). Conclusions Multimodal image fusion-based reconstruction combined with 3D printing may improve the visualization and personalization of preoperative planning for intracranial tumor resection. This strategy appears to facilitate more efficient surgery, reduce blood loss, and improve resection and functional preservation outcomes. It may be particularly useful for accurate localization and cortical venous protection in supratentorial tumors, and for delineating complex neurovascular anatomy and enabling preoperative simulation in CPA tumors.