Results
Among the 510 patient enrolled in this study, 164 patients were excluded due to the usage of inappropriate dose of edoxaban. Accordingly, data on 346 patients, who were receiving recommended dosages of edoxaban, were used for the analysis. The median age of the included patients was 73 years (range, 34-93 years), and there were 168 (48.6%) females. Thirty two patients (9.2%) experienced bleeding complications after taking edoxaban. Among them, 16 patients experienced major bleeding complications. Two and eight patients experienced intracranial hemorrhage and gastrointestinal bleeding complications, respectively. Six patients showed urological bleeding and 11 had unclassified bleeding outcomes. Eleven patients experienced stroke or recurrent DVT/PE.
As shown in Table 1, patients less than or equal to 60 kg had more bleeding complications than those who were heavier than 60 kg (p=0.018). Also, patients with cancer had more bleeding outcomes than those without cancer (p=0.030).
Multivariable analysis (Table 2) included sex, age, CrCl, and factors with p<0.05 in univariate analysis (weight and cancer). After adjusting for related covariates, patients with cancer or weights less than or equal to 60 kg showed approximately 3.4-fold higher bleeding complications than patients without cancer or with weights higher than 60 kg. The Hosmer–Lemeshow test showed that the fitness of the multivariable analysis model was satisfactory (χ2=11.335, 6 degrees of freedom, p=0.079).
Since more than 10% of patients had bleeding complications even after dose reduction, we performed subgroup analysis to investigate factors on bleeding complications in high-risk patients receiving a low dose of edoxaban. We found that weight was a significant factor for bleeding outcomes in both univariate and multivariable analysis (Table 3 and Table 4). Patients with weight ≤60 kg showed 3.3-times higher bleeding complications even with low dose of edoxaban compared to those with weight >60 kg. Hosmer–Lemeshow test showed that the fitness of the multivariable analysis model was satisfactory (χ2=3.636, 6 degrees of freedom, p=0.726).