Background/Purpose: Bone mineral density (BMD) measurement by dual X-ray absorptiometry (DXA) is the basis of World Health Organization (WHO) classification criteria for osteoporosis. Traditionally unilateral hip bone density has been measured and the utility of bilateral reporting remains unclear. Recent data have shown that a significant difference may be noted in bone density measurements between left and right hips with resultant failure to identify and treat individuals at high risk of fracture. In the 2023 Position Statement, the International Society of Clinical Densitometry (ISCD) noted that the impact of discordant hip bone density on diagnosis and monitoring of osteoporosis is a significant knowledge gap. This study aims to examine concordance vs discordance between left and right hip bone density measurement at a large hospital system. Results of this study may be helpful to further evaluate the impact and need for bilateral hip BMD testing as standard of care for osteoporosis screening.

Methods: A retrospective cohort study was performed using data from electronic medical record at the University of Pennsylvania. After IRB approval, patients with bilateral DXA were identified. A subset of charts with the following criteria were randomly selected for review: age 50 years or older, DXA report within the past 4 years within the health system, and diagnosis of osteopenia or osteoporosis. Manual chart review was conducted for capture of T scores for lumbar spine, left femoral neck, left total hip, right femoral neck and right total hip. We descriptively report baseline characteristics. Scatter plots were used to assess correlation between the right and left sides. Paired T tests were employed to test whether the left was significantly different from the right side for both the total hip and femoral neck.

Results: Within the health system, 29,258 patients > age 50 years had received a DXA scan at 3 primary sites between 2021 and 2024. The initial search was then narrowed to limit patients to those seen by Rheumatology providers (N&#3f1171). Among these, 85.7% patients were female and 81% were White. 200 charts were randomly selected among these and reviewed manually. 143 (71.5%) patients had DXA results reporting bilateral hip BMD. 27 (19%) patients had a T score difference of 0.5 between left and right femoral neck or total hip. 13 patients were classified into a different category: 2 patients with normal T score on one side vs low bone mass on the other side, and 11 patients with low bone mass on one side vs osteoporosis on the other side. This resulted in a reclassification rate of 7.8%, with 7% of patients with missed diagnosis of osteoporosis. Correlation between the two sides was high (Figure 1A and B) and there were no statistically significant differences between the left and right sides (Table 2).

Conclusion: In this retrospective study of patients >50 years, measurement of bilateral hip bone density demonstrated a high overall correlation between the two sides but resulted in a reclassification rate of 7% from low bone mass into osteoporosis when compared to measuring only one side. There may be a significant missed opportunity for primary fracture prevention which needs further examination.

Table 1: Demographics

Table 2: Difference between left and right femoral neck and total hip T scores

Figure 1A and 1B: Scatter plots comparing right and left sides for femoral neck (A) and total hip (B)

« Back to ACR Convergence 2025

ACR Meeting Abstracts - https://acrabstracts.org/abstract/evaluating-discordance-between-bilateral-hip-bone-density-measurements-in-individuals-with-low-bone-mass/