Distal femur BMD declines ∼20% following total knee arthroplasty (TKA) potentially leading to adverse outcomes. BMD knowledge before and following TKA might allow interventions to optimize outcomes. We hypothesized that distal femur and proximal tibial BMD could be reproducibly measured with existing DXA technology. Elective TKA candidates were enrolled and standard clinical DXA plus bilateral PA and lateral knee scans acquired. Manual regions of interest (ROIs) were placed at distal femur and proximal tibia sites based on required TKA machining and periprosthetic fracture location. Intra- and inter-rater BMD reliability was assessed by intra-class correlation (ICC). Custom and standard proximal femur BMD were correlated by linear regression and paired t test evaluated BMD differences between planned surgical and contralateral side. One hundred subjects (68F/32M), mean (SD) age and BMI of 67.2 (7.7) yr and 30.8 (4.8) kg/m² were enrolled. Lowest clinical BMD T-score was < -1.0 in 65% and ≤ -2.5 in 16%; 34 had prior fracture. BMD reproducibility at all custom ROIs was excellent; ICC > 0.96. Mean BMD at custom ROIs ranged from 0.903 to 1.346 g/cm² in the PA projection and 0.891 to 1.429 g/cm² in the lateral. Lower BMD values were observed at the proximal tibia, while the higher measurements were at the femur condyle. Custom knee ROI BMD was highly correlated (p < 0.0001) with total and femur neck with better correlation at ROIs adjacent to the joint (R² = 0.62–0.67, 0.49–0.55 respectively). In those without prior TKA (n = 76), mean BMD was lower (2.8%–6.6%; p < 0.05) in the planned surgical leg at all custom ROIs except the PA tibial regions. Individual variability was present with 82% having a custom ROI with lower BMD (up to 53%) in the planned operative leg. Distal femur and proximal tibial BMD can be measured using custom ROIs with good reproducibility. Suboptimal bone status is common in TKA candidates and distal femur/proximal tibia BMD is often lower on the planned operative side. Routine distal femur/proximal tibial BMD measurement might assist pre-operative interventions, surgical decision-making, subsequent care and outcomes. Studies to evaluate these possibilities are indicated.

全膝关节置换术 (TKA) 后股骨远端BMD 下降约 20%，可能导致不良后果。TKA 之前和之后的 BMD 知识可能允许干预以优化结果。我们假设可以使用现有的 DXA 技术重复测量股骨远端和胫骨近端 BMD。选修 TKA 候选人被招募并获得标准临床 DXA 加双侧 PA 和侧膝扫描。根据所需的 TKA 加工和假体周围骨折位置，将手动感兴趣区域 (ROI) 放置在股骨远端和胫骨近端部位。通过类内相关性 (ICC) 评估评估者内和评估者间 BMD 的可靠性。自定义和标准股骨近端 BMD 的相关性为线性回归和配对t检验评估了计划手术侧和对侧之间的 BMD 差异。一百名受试者 (68F/32M)，平均 (SD) 年龄和 67.2 (7.7) 岁和 30.8 (4.8) kg/m ²的 BMI被招募。65% 的最低临床 BMD T 评分 < -1.0 和 16% 的≤ -2.5；34 人曾有过骨折。所有自定义 ROI 的 BMD 重现性非常好；ICC > 0.96。自定义 ROI 处的平均 BMD 范围为 PA 投影中的 0.903 至 1.346 g/cm ²和横向中的 0.891 至 1.429 g/cm ²。在胫骨近端观察到较低的 BMD 值，而在股骨髁处观察到较高的测量值。自定义膝关节 ROI BMD 高度相关（p< 0.0001)，总颈和股骨颈在与关节相邻的 ROI 处具有更好的相关性（R ²  = 0.62–0.67，分别为 0.49–0.55）。在没有既往 TKA 的患者（n = 76）中，平均 BMD 较低（2.8%–6.6%；p< 0.05) 在除 PA 胫骨区域之外的所有自定义 ROI 的计划手术腿中。存在个体差异，82% 的患者在计划的手术腿中具有较低 BMD（高达 53%）的自定义 ROI。可以使用具有良好重现性的自定义 ROI 测量股骨远端和胫骨近端 BMD。次优骨状态在 TKA 候选者中很常见，股骨远端/胫骨近端 BMD 在计划的手术侧通常较低。常规股骨远端/胫骨近端 BMD 测量可能有助于术前干预、手术决策、后续护理和结果。指出了评估这些可能性的研究。