Predictive Value of HAS-BLED Score Regarding Bleeding Events and Graft Survival following Renal Transplantation

Objective: Due to the high prevalence and incidence of cardio- and cerebrovascular diseases among dialysis-dependent patients with end-stage renal disease (ERSD) scheduled for kidney transplantation (KT), the use of antiplatelet therapy (APT) and/or anticoagulant drugs in this patient population is common. However, these patients share a high risk of complications, either due to thromboembolic or bleeding events, which makes adequate peri- and post-transplant anticoagulation management challenging. Predictive clinical models, such as the HAS-BLED score developed for predicting major bleeding events in patients under anticoagulation therapy, could be helpful tools for the optimization of antithrombotic management and could reduce peri- and postoperative morbidity and mortality. Methods: Data from 204 patients undergoing kidney transplantation (KT) between 2011 and 2018 at the University Hospital Leipzig were retrospectively analyzed. Patients were stratified and categorized postoperatively into the prophylaxis group (group A)—patients without pretransplant anticoagulation/antiplatelet therapy and receiving postoperative heparin in prophylactic doses—and into the (sub)therapeutic group (group B)—patients with postoperative continued use of pretransplant antithrombotic medication used (sub)therapeutically. The primary outcome was the incidence of postoperative bleeding events, which was evaluated for a possible association with the use of antithrombotic therapy. Secondary analyses were conducted for the associations of other potential risk factors, specifically the HAS-BLED score, with allograft outcome. Univariate and multivariate logistic regression as well as a Cox proportional hazard model were used to identify risk factors for long-term allograft function, outcome and survival. The calibration and prognostic accuracy of the risk models were evaluated using the Hosmer−Lemshow test (HLT) and the area under the receiver operating characteristic curve (AUC) model. Results: In total, 94 of 204 (47%) patients received (sub)therapeutic antithrombotic therapy after transplantation and 108 (53%) patients received prophylactic antithrombotic therapy. A total of 61 (29%) patients showed signs of postoperative bleeding. The incidence (p < 0.01) and timepoint of bleeding (p < 0.01) varied significantly between the different antithrombotic treatment groups. After applying multivariate analyses, pre-existing cardiovascular disease (CVD) (OR 2.89 (95% CI: 1.02−8.21); p