Plenary Poster 33rd ASM of the Australian & New Zealand Bone & Mineral Society 2023

Fracture following kidney and simultaneous pancreas-kidney transplantation is predicted by DXA-derived bone mineral density and advanced hip analysis (#106)

Grahame J Elder 1 2 3 , Tahira Scott 4 5 , Jasna Aleksova 6 7 , Carmel Hawley 5 8 9 , Mina Khair 9 , Chris Schultz 10 , Ryan Gately 9 , Mirna Vucat-Dzumhur 1 , Harpreet Kaur 1
  1. Department of Renal Medicine, Westmead Hospital, Sydney, NSW, Australia
  2. Medical faculty, University of Notre Dame, Sydney, NSW , Australia
  3. Skeletal Diseases Program, Garvan Institute of Medical Research, Sydney, NSW, Australia
  4. Department of Nephrology, Cairns Hospital, Cairns, Qld, Australia
  5. School of Medicine, University of Queensland, Brisbane, Qld , Australia
  6. School of Clinical Sciences, Monash University, Melbourne, Vic, Australia
  7. Department of Endocrinology, Monash Health, Melbourne, Vic, Australia
  8. Australian Kidney Trials Network, University of Queensland, Brisbane, Qld, Australia
  9. Department of Nephrology, Princess Alexandra Hospital, Brisbane, Qld, Australia
  10. Department of Nuclear Medicine, PET and Bone Densitometry, Royal Adelaide Hospital, Adelaide, SA, Australia

Patients with kidney failure have accelerated trabecular and cortical deterioration and elevated fracture risk that remains high following transplantation. However, BMD is less predictive of fracture in these patients. This study aimed to determine if the DXA-derived trabecular bone score (TBS) and advanced hip analysis (AHA) improved post-transplant fracture prediction.

Patients receiving kidney or simultaneous pancreas-kidney (SPK) transplants within two Australian centres were included. Baseline information included demographics, medications and laboratory data. A DXA provided BMD, TBS and AHA parameters; femoral neck, calcar and shaft cortical thickness (CTh), and the femoral neck buckling ratio (BR), an indicator of structural instability defined as radius/CTh. Patients received treatment to reduce post-transplant BMD loss using a risk algorithm.(1) Parameters with skewed distributions were log transformed, and hazard ratios were determined using Kaplan Meier and Cox proportional hazard models with multivariable adjustment.

Of 357 kidney and SPK transplant recipients, 289 (83%) received a kidney-only transplant. Mean age was 48±13 years, 62% were male, 20% had type 1 diabetes mellitus (T1DM) and median dialysis vintage was 29 months (IQR: 12, 60). There were 81 incident fractures, with median time to fracture or censoring 4.4 years (2.5, 5.5). Incident fracture was predicted by T1DM (p<0.001), former smoking (p=0.017), lower serum 25OHD (p=0.045), axial BMD (p<0.01), CTh (all sites p≤0.01) and the BR (p=0.004; HR 1.83 (1.34, 2.49) for each log-unit increase), but not by the TBS.  After multivariate adjustment, T1DM, 25OHD, smoking, prevalent fracture and hip BMD remained significant predictors.  Using the BMD-based risk algorithm, inclusion of an interaction parameter for BR above or below the median improved the model fit (HR 2.36 (0.96, 5.80); p=0.06).

DXA-derived BMD, AHA cortical parameters and the BR predict incident fracture in kidney and SPK transplant recipients, but TBS does not. AHA by DXA may improve fracture risk assessment.

  1. Elder G. Transplantation Feb. 2023