HAS-BLED Score and Bleeding Risk in RC: Dataset

Objective To evaluate the predictive value of the preoperative HAS-BLED score in estimating clinically significant blood loss in the course of radical cystectomy (RC). Materials and Methods We retrospectively reviewed 176 patients who underwent RC. Major hemorrhage was defined as a change in hemoglobin level (delta hemoglobin = preoperative hemoglobin level – postoperative hemoglobin level) of ≥4 g/dL, the need for blood transfusion to manage bleeding, or the necessity of early reintervention due to postoperative hemorrhage or related complications within 14 days. HAS-BLED scores were calculated preoperatively. Univariable and multivariable logistic regression analyses were performed to identify independent predictors. ROC analysis was used to assess the discriminative ability of the HAS-BLED score. Results Major postoperative bleeding occurred in 22 patients (12.5%). Bleeding patients had significantly higher HAS-BLED scores compared to those without bleeding (p < 0.001). Each 1-point increase in the HAS-BLED score tripled the odds of bleeding (OR: 3.01, 95% CI: 1.83–4.93). Multivariable analysis identified hypertension, renal or hepatic dysfunction, and history of stroke as independent risk factors. ROC analysis showed an AUC of 0.796, and the Youden-optimal cut-off was 2.5, corresponding to a clinical threshold of ≥3. In a subgroup of patients on anticoagulants (n = 35), 40% experienced bleeding, and HAS-BLED remained predictive (OR: 2.02, p = 0.045). Conclusion The HAS-BLED score effectively predicts clinically meaningful bleeding after RC and may aid in preoperative risk stratification. A threshold of ≥3 appears suitable for identifying high-risk patients, particularly in those on anticoagulant therapy.

研究目的：本研究旨在评估术前HAS-BLED评分对根治性膀胱切除术（RC）术中及术后临床显著性失血的预测价值。材料与方法：本研究回顾性分析了176例行根治性膀胱切除术的患者资料。将符合以下任一情况定义为严重出血：血红蛋白差值（ΔHb=术前血红蛋白水平-术后血红蛋白水平）≥4 g/dL、需输血控制出血，或术后14天内因出血或相关并发症需早期再次干预。术前计算所有患者的HAS-BLED评分。采用单变量及多变量logistic回归分析筛选独立预测因素，通过受试者工作特征（Receiver Operating Characteristic, ROC）分析评估HAS-BLED评分的判别能力。结果：术后共22例患者（12.5%）发生严重出血。出血患者的HAS-BLED评分显著高于未出血患者（P<0.001）。HAS-BLED评分每升高1分，出血发生的比值比升至3倍（OR=3.01，95%CI：1.83~4.93）。多变量分析显示，高血压、肝肾功能不全及卒中病史为独立危险因素。受试者工作特征分析显示曲线下面积（Area Under the Curve, AUC）为0.796，Youden指数最优截断值为2.5，对应临床界值为≥3分。在35例接受抗凝治疗的亚组患者中，40%发生了出血，HAS-BLED评分仍具有预测价值（OR=2.02，P=0.045）。结论：HAS-BLED评分可有效预测根治性膀胱切除术后的临床显著性出血，有助于术前风险分层。以≥3分为界值可准确识别高危患者，尤其适用于接受抗凝治疗的人群。