MULTIVARIATE ANALYSES OF ACUTE REJECTION IN RENAL TRANSPLANT RECIPIENTS RECEIVING EARLY CORTICOSTEROID WITHDRAWAL

THERAPEUTIC STRATEGIES FOR KIDNEY TRANSPLANTATION

A.R. Shields¹, R.R. Alloway², G. Mogilishetty², M. Cardi³, S. Huang³, L. Arend⁴, P. Brailey⁵, R. Munda⁶, J. Everly⁷, E.S. Woodle^6
1Transplant Surgery, University of Cincinnati, Cincinnati/Ohio/UNITED STATES OF AMERICA, ²Nephrology, University of Cincinnati, cincinnati/UNITED STATES OF AMERICA, ³Nephrology, The Christ Hospital, cincinnati/UNITED STATES OF AMERICA, ⁴Pathology, University of Cincinnati, cincinnati/UNITED STATES OF AMERICA, ⁵Hoxworth Immunology, University of Cincinnati, cincinnati/UNITED STATES OF AMERICA, ⁶Transplant Surgery, University of Cincinnati, cincinnati/UNITED STATES OF AMERICA, ⁷, Oncology Hematology, Inc, cincinnati/UNITED STATES OF AMERICA

Body:
Introduction: Multivariate analyses of risk factors for acute rejection (AR) under early corticosteroid withdrawal (≤ 7 days of steroids, ECSWD) have been previously conducted with short-term follow-up. Our experience with ECSWD now has long-term follow-up in which this patient population can be rigorously analyzed. The purpose of this study is to determine risk factors for AR under ECSWD regimens and whether there are differences in death-censored graft survival (DCGS) based on timing of AR in renal transplant recipients. Methods: Data was collected prospectively in 630 patients who received ECSWD regimens from January 2000-August 2009 at our center. AR was defined by Banff 97 criteria and included Banff 1A or greater. Separate analyses were done for AR < 6 months (early AR), AR ≥ 6 months (late AR), and overall AR. Risk factors evaluated in univariate analyses (UVA) include: age, gender, race, repeat transplant, current CDC PRA, peak CDC PRA, DGF, DR mismatches (MM), donor type (living versus deceased), donor characteristics, and pretransplant diabetes types 1 and 2. UVA were performed by chi-square analysis, with Cox Proportional Hazards regression performed for factors that entered the multivariate model (MVA). Variables with a p < 0.25 entered the MVA. Kaplan Meier survival analyses were done to determine AR timing and DCGS. Results: There were 116 patients (18.4%) who experienced an AR post-transplant (69 patients with an early AR and 47 patients with a late AR). Mean follow-up in AR patients was 1573 ± 960 days. There were 148 AR episodes for a mean of 1.3 episodes per patient. Mean time to first AR was 328 ±493 days post-transplant. Results of MVA’s are in table below. Deceased and living donor patients with early AR had significantly greater DCGS compared to patients with late AR (p<0.001) (Figures 1 & 2).

MVA for AR Overall	HAZARD RATIO		P	95% CI
HLA DR MM = 2	2.12		<0.001	1.293 – 2.807
African American Race	1.90		0.001	1.465 – 3.082
Recipient Age	0.97		<0.001	0.960 – 0.988
MVA for AR < 6 Months
Repeat Transplant		4.02	0.001	1.709 – 9.479
HLA DR MM = 2		3.43	0.003	1.534 – 7.695
Donor SCr		0.07	0.042	0.006 – 0.909
MVA for AR ≥ 6 Months
African American Race	2.48		0.022	1.137 – 5.432
Recipient Age	0.95		0.004	0.930 – 0.986

Conclusion: Early and late AR both reduce GS, with late AR having a greater effect. Interestingly, risk factors for early and late AR were different. Importantly, graft half lives in this large experience with long-term followup were numerically greater than those previously derived from SRTR data.

Disclosure: All authors have declared no conflicts of interest.