Abstract: TH-PO929
Cardiovascular Risk Prediction in Renal Transplant: Post-Hoc FAVORIT Trial Analyses
Session Information
- Transplantation: AKI, Cardiovascular, and Metabolic Complications
November 02, 2017 | Location: Hall H, Morial Convention Center
Abstract Time: 10:00 AM - 10:00 AM
Category: Transplantation
- 1702 Transplantation: Clinical and Translational
Authors
- Shireman, Theresa I., Brown University, Providence, Rhode Island, United States
- Merhi, Basma Omar, Rhode Island Hospital, East Greenwich, Rhode Island, United States
- Ogarek, Jessica A, Brown University, Providence, Rhode Island, United States
- Bostom, Andrew, Memorial Hospital of Rhode Island, Chepachet, Rhode Island, United States
Background
Cardiovascular disease (CVD) is the leading cause of mortality and kidney graft failure in renal transplant recipients (KTR), but predictive risk algorithms consistently underestimate the incidence of arteriosclerotic outcomes.
Methods
We conducted a secondary analysis of data from the Folic Acid for Vascular Outcome Reduction in Transplantation (FAVORIT) randomized clinical trial. Measures included traditional CVD risk factors along with an expanded list of variables clinically important for KTRs, such as baseline phosphorus, cholesterol (HDL, LDL, remnant), triglycerides, graft vintage, donor type, glomerular filtration rate (eGFR), albumin-to-creatinine ratio (ACR), and diastolic (DBP) and systolic blood pressure (SBP). The cohort was split into training (2/3) and validation (1/3) samples after excluding individuals with pre-existing CVD. After deriving the most parsimonious risk calculator for the clinically adjudicated CV endpoint (CVE) in the training sample, we then applied the risk score to the validation sample and assessed model fit with area under the curve (c-statistic). Secondary outcomes, all-cause mortality (ACM) and graft failure, were also modeled using the CV risk score.
Results
From the training sample (n=1,892), key variables associated with CVD endpoint were age, smoking status, diabetes, living donor, eGFR ≥ 45, ln(ACR), DBP (<70) and SBP (>140). Combined with race and gender, the c-statistics for the training model was 0.754 for CVD endpoint, 0.762 for all-cause mortality, and 0.83 for graft failure. The model results from the validation sample generated c-statistics of 0.666 (CVE), 0.682 (ACM), and 0.643 (graft failure).
Conclusion
Important kidney transplant related risk factors (donor type, eGFR, and ACR) add significantly to cardiovascular risk prediction that include more typical measures. Further testing in other cohorts is needed to validate our findings and strengthen the model.
Funding
- NIDDK Support