Preoperative Bone Defect Volume Estimation Using the Hemielipsoid Formula for Allograft Planning in Musculoskeletal Tumor Surgery: A Prospective Validation Study Against Intraoperative Fluid Displacement Measurement
Preoperative Bone Defect Volume Estimation Using the Hemielipsoid Formula for Allograft Planning in Musculoskeletal Tumor Surgery: A Prospective Validation Study Against Intraoperative Fluid Displacement Measurement
Bone defects following primary musculoskeletal tumor resection are frequently reconstructed using structural allografts. Allograft availability is limited, and intraoperative waste due to oversized graft selection remains a practical challenge in orthopedic oncology. Accurate preoperative estimation of the expected defect volume is therefore critical not only for surgical planning but also for minimizing unnecessary allograft use.
This prospective study aims to evaluate the accuracy of the preoperative hemielipsoid formula in estimating bone defect volume after primary musculoskeletal tumor resection, using intraoperative fluid displacement as the reference standard. A secondary aim is to determine whether preoperative CT-based hemielipsoid volume calculation can reliably guide allograft selection and reduce intraoperative graft waste.
Patients scheduled for primary musculoskeletal tumor resection with structural allograft reconstruction at Sakarya University Training and Research Hospital will be enrolled. Preoperative defect volume will be estimated using the hemielipsoid formula (V = 4/3 × π × a × b × c / 2) derived from preoperative CT measurements. Intraoperative defect volume will be measured by fluid displacement as the reference standard. Agreement between methods will be assessed by Bland-Altman analysis and intraclass correlation coefficient (ICC).
**Background:** In orthopedic oncology, structural allograft reconstruction following primary musculoskeletal tumor resection requires precise preoperative planning. Currently, allograft selection at our institution relies on the surgeon's intraoperative visual estimation, with no standardized preoperative volumetric assessment protocol in place. This approach may result in allograft waste due to oversized graft selection, which is particularly problematic given the limited availability of allograft bank resources.
**Rationale:** The hemielipsoid formula (V = 4/3 × π × a × b × c / 2) offers a simple, reproducible method for estimating irregular bone defect volumes from standard cross-sectional imaging. If validated against an intraoperative reference standard, this formula could serve as a practical preoperative planning tool to guide allograft selection and minimize unnecessary graft use.
**Study Design:** This is a single-center prospective observational study conducted at Sakarya University Training and Research Hospital, Department of Orthopedics and Traumatology.
**Preoperative Assessment:** Defect volume will be estimated by the operating surgeon using the hemielipsoid formula applied to routine preoperative CT imaging. The three principal dimensions of the anticipated resection defect (length, width, and depth) will be measured on axial, coronal, and sagittal CT reconstructions, and volume will be calculated accordingly.
**Intraoperative Assessment:** Following tumor resection, the actual bone defect volume will be measured by direct cavity fluid displacement as the intraoperative reference standard. The resection cavity will be filled with a measured volume of saline to determine the true defect volume.
**Outcome Assessment:** The primary outcome is the agreement between preoperative hemielipsoid-estimated volume and intraoperative fluid displacement-measured volume, assessed by Bland-Altman analysis and intraclass correlation coefficient (ICC). The secondary outcome is the correlation between preoperative volume estimation and actual allograft volume used, to evaluate the utility of the hemielipsoid formula in reducing intraoperative allograft waste.
**Sample Size:** A priori power analysis was performed based on intraclass correlation coefficient (ICC) estimation (H0: ICC = 0.70; H1: ICC = 0.90; two-tailed alpha = 0.05; power = 0.80) using the Fisher z-transformation method, indicating a minimum required sample size of 25 patients. Target enrollment was set at 40 patients to account for a 10% dropout rate and to allow for potential subgroup analyses.
Inclusion Criteria:
Exclusion Criteria: