Model to Predict Prolonged Delayed Graft Function (DGF) Following Kidney Transplantation from Deceased Donor with Acute Kidney Injury (AKI Tx).
1Mayo Clinic, Phoenix, AZ
2Mayo Clinic, Rochester, MN
Meeting: 2017 American Transplant Congress
Abstract number: B145
Keywords: Area-under-curve (AUC), Cadaveric organs, Donors, Kidney, marginal
Session Information
Session Name: Poster Session B: Kidney Complications II
Session Type: Poster Session
Date: Sunday, April 30, 2017
Session Time: 6:00pm-7:00pm
Presentation Time: 6:00pm-7:00pm
Location: Hall D1
Background
Our aim is to develop a model to predict prolonged DGF following AKI Tx.
Methods
AKI Tx (donor terminal Cr>/=2) from 2003 to 2016 were included. Logistic regression analysis starting with baseline characteristics was used to develop a model for predicting DGF >/=5 days. 10-fold cross-validation was used.
Results
Cohort n=358. Median days DGF 8 (IQR 0-14). 63.4% had DGF>/=5.
| DGF>/=5(n=227) | Not(n=131) | P | |
| Age | 55.6 (12.2) | 57.8 (14.0) | 0.12 |
| Female | 34.8 | 51.2 | 0.003 |
| On dialysis | 94.3 | 72.5 | <.001 |
| KDPI | 52 (23) | 50 (24) | 0.63 |
| CIT | 22.4 (6.6) | 20.3 (7.4) | 0.005 |
| Donor age | 37.7 (13.9) | 38.0 (14.1) | 0.86 |
| DCD | 5.7 | 4.6 | 0.64 |
| donor oliguria | 55.1 | 34.1 | 0.0003 |
| Terminal Cr | 5.08 (2.70) | 4.00 (2.14) | <.001 |
| admit Cr | 1.73 (0.99) | 1.49 (0.63) | 0.01 |
| peak Cr | 5.66 (2.78) | 4.38 (2.35) | <.001 |
Table shows coefficients (SE) for the model
| Estimate | Pr(>|z|) | |
| (Intercept) | -4.85 (0.94) | 0.000 |
| Recipient male | 0.74 (0.26) | 0.005 |
| Recipient Black | -0.64 (0.42) | 0.123 |
| dialysis pre Tx | 1.69 (0.45) | 0.0002 |
| years of dialysis | 0.16 (0.06) | 0.01 |
| Donor pumped | -0.53 (0.27) | 0.05 |
| CIT | 0.06 (0.02) | 0.003 |
| Donor age | 0.03 (0.01) | 0.001 |
| Donor Weight | -0.01 (0.00) | 0.07 |
| DCD donor | 1.23 (0.62) | 0.04 |
| donor diabetes | -0.64 (0.44) | 0.15 |
| Donor oliguria | 0.60 (0.28) | 0.03 |
| Admit Cr | 0.27 (0.17) | 0.12 |
| Peak Cr | 0.20 (0.06) | 0.002 |
The AUC for ROC was 0.797. The mean cross-validation AUC was 0.708. 
Discussion
Using baseline characteristics we have developed a model for estimating the risk for prolonged DGF with good discriminatory value. This model will help centers in estimating the risk of prolonged DGF after AKI transplantation.
CITATION INFORMATION: Heilman R, Smith B, Johnson C, Khamash H, Reddy K, Kremers W, Kaplan B. Model to Predict Prolonged Delayed Graft Function (DGF) Following Kidney Transplantation from Deceased Donor with Acute Kidney Injury (AKI Tx). Am J Transplant. 2017;17 (suppl 3).
To cite this abstract in AMA style:
Heilman R, Smith B, Johnson C, Khamash H, Reddy K, Kremers W, Kaplan B. Model to Predict Prolonged Delayed Graft Function (DGF) Following Kidney Transplantation from Deceased Donor with Acute Kidney Injury (AKI Tx). [abstract]. Am J Transplant. 2017; 17 (suppl 3). https://atcmeetingabstracts.com/abstract/model-to-predict-prolonged-delayed-graft-function-dgf-following-kidney-transplantation-from-deceased-donor-with-acute-kidney-injury-aki-tx/. Accessed May 19, 2025.« Back to 2017 American Transplant Congress