Comparative study in predictive value of ORBIT score and HAS-BLED score for warfarin anticoagulation bleeding risk after heart valve replacement

Objective To compare the predictive value of ORBIT score and HAS-BLED score for bleeding risk after warfarin anticoagulation following heart valve replacement.

Methods The clinical data of 304 patients who underwent heart valve replacement and required warfarin anticoagulation therapy were retrospectively analyzed. The patients were stratified and grouped based on the ORBIT score and HAS-BLED score. The correlation between risk stratification based on different scores and postoperative bleeding events during warfarin anticoagulation was analyzed. The consistency of risk stratification between the two scoring methods and their predictive value for bleeding events were compared.

Results Among 304 patients, there were 32 bleeding patients and 272 non-bleeding patients. The postoperative ORBIT and HAS-BLED scores of the bleeding patients were higher than those of non-bleeding patients(P < 0.05). The bleeding risk of medium and high-risk patients based on the ORBIT score was 6.092 (95%CI, 2.694 to 13.775) times and 9.373 (95%CI, 1.465 to 59.943) times higher than that of low-risk patients, respectively. The bleeding risk of medium and high-risk patients based on the HAS-BLED score was 3.750 (95%CI, 1.383 to 10.166) times and 14.250 (95%CI, 5.489 to 36.995) times higher than that of low-risk patients, respectively. The analysis of the receiver operating characteristic curve showed that the area under the AUC (95%CI, 0.589 0.698 9) of ORBIT score for predicting the risk of postoperative warfarin anticoagulation bleeding was 0.646 (95%CI, 0.589 to 0.699), with a sensitivity of 46.88% and a specificity of 87.87%, and an optimal cutoff value of 3 points. The AUC of HAS-BLED score predicting the risk of postoperative anticoagulation bleeding with warfarin was 0.768 (95%CI, 0.717 to 0.814), with a sensitivity of 75.00% and a specificity of 69.85%, and an optimal cutoff value of 2 points. The Delong test results showed that there was a statistically significant difference in AUC between the two scores (D=0.122; 95%CI, 0.001 to 0.245; Z=1.962; P < 0.05).

Conclusion Both ORBIT score and HAS-BLED score have certain predictive value for bleeding risk during warfarin anticoagulation after heart valve replacement, but the predictive efficiency of HAS-BLED score is better.