An Exploratory Study of the Texture Research Investigational Platform (TRIP) to Evaluate Bone Texture Score of Distal Femur DXA Scans – A TBS-Based Approach

Abstract

Poor bone status is associated with increased complications following orthopedic surgery. Therefore, assessing site-specific skeletal status prior to or after orthopedic surgery to optimize outcomes is appealing. The trabecular bone score (TBS) approach, a surrogate for microarchitecture, was adapted to the Texture Research Investigational Platform (TRIP), which allows assessment of many skeletal sites imaged by various modalities. TRIP generates a bone texture score (TBS ORTHO), which could potentially guide surgical decision-making and offer insight into postsurgical fracture risk. As distal femur bone loss occurs following total knee arthroplasty (TKA), we hypothesized that TBS ORTHO after TKA would identify poorer texture in the operated femur compared to the nonoperated. We evaluated 30 subjects (15 M/15 F) with unilateral TKA 2–5 yr previously, mean age 67.9 yr and body mass index 30 kg/m². Using a Lunar iDXA, lumbar spine and entire femur scans were obtained, the latter using the atypical femur fracture feature. Distal femur bone mineral density (BMD) and TBS ORTHO were obtained using manual regions of interest (ROI) at 15% and 25% of leg length from the intercondylar notch. TBS ORTHO was determined using distal femur DICOM images and TRIP v1.0 (Medimaps, France). Differences in operated vs nonoperated femur were evaluated by paired t test. As previously reported, operated leg BMD was approx 10% lower at 15% and 25% ROIs. Similarly, TBS ORTHO values in the operated leg were approx 5% lower (p < 0.05) at these same ROIs. Distal femur TBS ORTHO and BMD were largely unrelated. TBS ORTHO reproducibility at these ROIs was approx 3.5%. In conclusion, this pilot study documents the feasibility of reproducibly obtaining distal femur TBS ORTHO values. Lower values were observed in the surgical leg, consistent with the bone loss that follows TKA. Further work is indicated to refine TRIP use and evaluate whether such data provides guidance for surgical decision-making and improves periprosthetic fracture prediction.

Introduction

Total knee arthroplasty (TKA) is commonly performed; approx 15% of Americans age 70+ have had a TKA (1) and this number will increase for the foreseeable future (2, 3). Low bone density is common in these individuals with approx 1/3 having osteoporosis preoperatively (4, 5). Furthermore, TKA causes distal femoral bone loss of approx 15% (6) within 12 mo, further (7, 8, 9, 10, 11) predisposing to periprosthetic fracture. As low bone mineral density (BMD) is common and these fractures are generally caused by falls (12), they have many similarities with classical osteoporosis-related fractures. These fractures are generally in the supracondylar region of the distal femur (13), and occur in 0.3%–5.5% of patients after a primary TKA, and in 1.6%–38.0% after a TKA revision (14, 15, 16). Femoral fractures after TKA are technically challenging to repair (17) and are associated with longer hospital stay, frequent readmission, and high mortality rates (18, 19, 20, 21); indeed, the morbidity and mortality consequences are strikingly similar to classic osteoporotic hip fractures (14, 15, 16, 18, 19, 20, 21, 22). As the number of TKA procedures grows and implant long-term survival increases, periprosthetic fracture risk will become a greater healthcare issue (14, 15, 16). Thus, increased recognition of the scope of this problem and efforts to reduce periprosthetic fracture risk are essential. Additionally, poor bone quality may increase the risk for revision surgery following total joint arthroplasty, thus bone-active medications may enhance implant survival. Consistent with this, a meta-analysis (23) found long-term bisphosphonate use reduced revision risk after total knee or total hip arthroplasty by 52%. Finally, orthopedic surgeons report that bone quality knowledge affect surgical decision making and choice of arthroplasty implant (24).

As DXA is a widely utilized tool to assess skeletal status, it is logical that this methodology be utilized to assist with orthopedic preoperative planning and subsequent periprosthetic fracture mitigation. As could be expected, those with low DXA-measured BMD are at increased risk of periprosthetic fracture and other postoperative complications (25) and distal femur BMD is lower in the operated leg post-TKA (26). In addition to measuring BMD, DXA can assess bone texture; a surrogate for microarchitecture. Currently, trabecular bone score (TBS) is used as a surrogate of lumbar spine microarchitecture using gray-level variation thereby improving fracture prediction and risk assessment independent of BMD (27). Until recently, TBS utility was limited to the lumbar spine. New software, Texture Research Imaging Platform (TRIP) applies the principles of gray-level texture scores to any skeletal site using various image modalities to generate a bone texture score (TBS ORTHO). This value is analogous to the TBS of the lumbar spine. It is logical that femoral microarchitecture is adversely affected by TKA and that this predisposes to periprosthetic fracture. However, distal femur bone microarchitecture after TKA has not been previously studied using DXA. Thus, the purpose of this study was to evaluate the utility and reproducibility of TRIP in the distal femur of patients with unilateral knee replacements. It is plausible that this measurement could ultimately serve as a supplement to BMD to facilitate surgical decision-making and improve prediction of postoperative periprosthetic fracture risk.