Sanchez EQ, Gonwa TA, Levy MF, Goldstein RM, Mai ML, Hays SR, Melton LB, Saracino G, Klintmalm GB. Preoperative and perioperative predictors of the need for renal replacement therapy after orthotopic liver transplantation.
Transplantation 2004;
78:1048-54. [PMID:
15480173 DOI:
10.1097/01.tp.0000137176.95730.5b]
[Citation(s) in RCA: 97] [Impact Index Per Article: 4.9] [Reference Citation Analysis] [Abstract] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 01/31/2023]
Abstract
BACKGROUND
Acute renal failure developing after orthotopic liver transplantation (OLTx) requiring renal replacement heralds a poor prognosis. Our center has previously reported a 1-year survival of only 41.8%. We undertook this study to determine whether we could identify preoperative and perioperative factors that would predict which patients are at risk.
METHODS
OLTxs performed between January 1, 1996, and December 31, 2001, were included in our retrospective database review. Combined kidney-liver transplants or patients with preoperative renal replacement therapy (RRT) were excluded. A total of 724 OLTxs were studied, which were divided into group I: no RRT, n=637; group II: hemodialysis only post-OLTx, n=17; and group III: continuous RRT post-OLTx, n=70. Univariate and stepwise logistic multivariate analyses were performed.
RESULTS
Preoperative serum creatinine greater than 1.9 mg/dL (odds ratio [OR] 3.57), preoperative blood urea nitrogen greater than 27 mg/dL (OR 2.68), intensive care unit stay more than 3 days (OR 10.23), and Model for End-Stage Liver Disease score greater than 21 (OR 2.5) were significant. A clinical prediction model was constructed: probability of requiring dialysis posttransplant=(-2.4586+1.2726 [creatinine >1.9] + 0.9858 [blood urea nitrogen >27] + 0.4574 [Model for End-Stage Liver Disease score >21] + 1.1625 [intensive care unit days >3]). A clinical prediction rule for patients with a score greater than 0.12 was applied to OLTx recipients who underwent transplantation in 2002. A total of 15 of 20 patients who received RRT and 111 of 121 who did not were correctly classified with the model.
CONCLUSIONS
This model allowed us to identify patients at high risk for developing the need for RRT postoperatively. Strategies for these patients to prevent or ameliorate acute renal failure and reduce the need for RRT postoperatively are needed.
Collapse