Related Experiment Video
Updated: Jan 10, 2026

Software-Assisted Quantitative Measurement of Osteoarthritic Subchondral Bone Thickness
Published on: March 18, 2022
Novel Visual Grade and Hounsfield Unit Predict Adequate Bone Strength for Cementless Total Knee Arthroplasty
Dong Hwan Lee1,2, Dai-Soon Kwak3, Sheen-Woo Lee4
1Department of Orthopaedic Surgery, Yeouido St. Mary's Hospital, Seoul 07345, Republic of Korea.
None:
Background and Objectives: The use of cementless total knee arthroplasty (TKA) is increasing, but established methods for assessing bone quality to prevent early failure remain undefined. Current preoperative assessments using central bone mineral density (BMD) do not accurately reflect peripheral bone quality, and intraoperative evaluation is subjective. This study aimed to establish objective assessment methods by analyzing the correlations between a novel visual grading system, CT Hounsfield units (HU), and actual bone strength. Materials and Methods: This prospective study included 131 patients undergoing posterior-stabilized TKA. We developed a novel visual grading system (Excellent, Good, Fair, Poor) based on femoral cutting surface characteristics. CT HUs were measured preoperatively by an assisting surgeon in the box bone area. Femoral box specimens underwent indentation testing to determine their actual bone strength. Minimum Required Strength (MRS) was defined at 2.5-fold the patient's body weight, and Estimated Withstanding Strength (EWS) was determined by scaling first failure load using area ratios. Patients were classified as "cementless suitable" (EWS > MRS) or "cemented mandatory" (EWS < MRS). Correlations were assessed using Spearman's rank correlation for visual grade and Pearson correlation for Hounsfield units. ROC curve analysis determined diagnostic accuracy. Results: Visual grade exhibited an exceptionally robust relationship to bone strength (Spearman ρ = 0.903, p < 0.01), whereas HU showed substantial correlation (Pearson r = 0.660, p < 0.01, R2 = 0.435). Visual grading achieved excellent diagnostic accuracy (AUC = 0.974, sensitivity 95.1%, specificity 95.9%) using "Good" grade as cutoff. HU demonstrated AUC of 0.938 with 92.7% sensitivity and 81.6% specificity at a cutoff value of 65.2. Conclusions: Our novel visual grading system and CT HU demonstrated excellent correlations with actual distal femoral bone strength and outstanding diagnostic performance for identifying cementless TKA candidates. Unlike traditional subjective intraoperative assessments such as the "thumb test", this system provides objective visual criteria directly correlated with actual bone strength. Preoperative HU screening with intraoperative visual grading can help prevent early failure.

