Study Population
The protocol used for this cross-sectional, retrospective review of medical records was approved by the institutional review board of our institute. We enrolled women who experienced a DRF between September 2016 and April 2019 and met the following inclusion criteria: (1) acute DRF caused by minor trauma, such as a fall from standing height, and (2) underwent a DXA scan (Lunar Prodigy; GE Lunar, WI) within two weeks after the fracture. Although our institute does not have a fracture liaison service, an osteoporosis examination is routinely recommended for all patients with DRF on their first follow-up visit to the outpatient clinic or after admission for operation. Finally, 227 women who met the criteria were enrolled. The mean age of the cohort was 65.1 ±10.1 years, mean BMI was 23.4 ± 3.1 kg/m2, and 163 patients (71.8% of 227 patients) were treated surgically.
Among the 227 patients with DRF, 203 experienced a first time DRF (initial DRF group) and 24 had a previous history of DRF (subsequent DRF group). The demographic characteristics, osteoporosis treatment history, and bone fragility parameters, including BMD, TBS, hip geometry parameters, BCT of the distal radius, and FRAX scores, were compared between the two groups. To reduce bias, patients in the subsequent DRF group were propensity score matched at a 1:2 ratio with patients in the initial DRF group, and additional comparison was performed between these groups. The propensity score was calculated for each patient based on logistic regression analysis, using subject age, BMI, and sex for matching.
BMDs and TBS
At our institute, BMD (g/cm2) was measured in the lumbar spine, femoral neck, trochanter, Ward’s triangle, and the total hip using Lunar Prodigy DXA scans (GE Healthcare, Madison, WI) and was analyzed using Encore Software ver.11.0. The lowest BMD T-score was derived from the BMDs of the lumbar spine, total hip, and femoral neck only, in accordance to the classification proposed by the World Health Organization [15]. Osteoporosis was defined as a lowest BMD T-score < -2.5. The BMD precision errors (percentage of the coefficient of variation)—measured by assessing 30 individuals with two scans at our institution—were 1.9% for the lumbar spine, 2.5% for the femoral neck, and 1.8% for the total hip. The least significant changes in BMD, calculated as 2.77× precision error and at a 95% confidence level, were 0.053 g/cm2 for the lumbar spine, 0.069 g/cm2 for the femoral neck, and 0.050 g/cm2 for the total hip. For the lumbar spine BMD, the L1–4 value was used for analysis. All TBS measurements were performed retrospectively using TBS iNsight Software, ver. 3.02 (Med-Imaps, Needham, MA, USA) based on spine DXA files from the database to ensure that the investigators are blinded to all clinical parameters. The software uses the raw DXA images of the anteroposterior spine for the same region of interest as the lumbar spine BMD measurements.
Hip geometry parameters
Geometric bone structure properties in all scans were further analyzed using the advanced hip assessment (AHA) program included with the GE Lunar Prodigy software, as described previously [16, 17]. The AHA program automatically set the region of interest, defined as the narrow neck (NN), transversing the narrowest width of the femoral neck. The AHA program yielded data for HAL, NSA, mean cortical thickness (mm), femur neck width (FNW, mm), CSA, cross-sectional moment of inertia (CSMI, cm4), section modulus (SM, cm3), and buckling ratio (BR) at the NN. The short-term coefficients of variance of AHA indices calculated from the images used for the precision assessment of BMD appeared to be slightly greater than those of conventional BMD, but were approximately 2%, similar to the previously reported precision data [18].
Cortical thickness of the distal radius
Cortical bone thickness was measured and analyzed based on a previously described method of analyzing the relationship between the BMD and cortical thickness of the distal radius [14]. In patients with initial DRF or recurrent DRF on the same side, an image of the contralateral side was selected. In patients with subsequent DRF on the side contralateral to that of the previous DRF, an old DRF side image was selected. In patients with bilateral subsequent DRF, an image of the dominant hand was selected. All images were randomly sorted after removing the personal information of patients and were reviewed by one orthopedic surgeon and one orthopedic resident. All radiographic measurements were performed using the picture archiving and communication systems program of our institute (Petavision®, OOO, OOO, OOO). Varying image magnification was normalized by standardizing longitudinal capitate lengths on all radiographs to 21.65 mm [19]. BCT was measured 50 and 70 mm proximal to the distal radio-ulnar joint, with the mean of the two measurements defined as average BCT. The mean value of each measurement was used for the analysis.
FRAX
FRAX® is a simple Fracture Risk Assessment Tool developed by the World Health Organization. FRAX algorithms calculate the ten-year probability of major osteoporotic fractures and hip fractures. We calculated the ten-year probability of fracture by including clinical risk factors such as previous fracture, hip fracture in parents, smoking habits, use of steroid medicine, rheumatoid arthritis, secondary osteoporosis, and alcohol habits. FRAX scores were acquired using the web-based calculation tool for OOO. We adjusted the FRAX score with TBS in each patient.
Statistical analysis
All statistical analyses, including propensity score matching analysis, were performed using the R statistical software (ver. 3.1.0; Foundation for Statistical Computing, Vienna, Austria; http://cran.r-project.org/), with p < 0.05 considered significant. Descriptive statistics, including means and 95% confidence intervals, were estimated for both groups. After assessing the normality of the distribution of the tested parameters, between-group differences in continuous variables including demographic data, BMD, hip geometry parameters, cortical thickness of the distal radius, and ten year probability of osteoporotic fracture were assessed using the Student’s t-test or Mann‒Whitney U-test, as appropriate. Categorical variables, including the proportions of female patients and those with underlying diseases, were compared in the two groups using the chi-square test or Fisher’s exact test. The reliability of measurements of distal radius BCT was calculated using the single measures intra-class correlation coefficient (ICC) from a two-way random effect ANOVA. Correlations between all parameters were evaluated using the Pearson correlation test. The correlation coefficient was interpreted using the scale proposed by Evans: 0.00–0.19, very weak; 0.20–0.39, weak; 0.40–0.59, moderate; 0.60–0.79, strong; and 0.80–1.00, very strong, respectively [20].