Peripheral Leukocyte Count and Risk of Bleeding in Patients with Non-Valvular Atrial Fibrillation Taking Dabigatran: A Real-World Study
Atrial fibrillation (AF) is a prevalent cardiac arrhythmia associated with increased risks of stroke, thromboembolism, and mortality. The introduction of oral anticoagulants has significantly reduced these risks, with dabigatran, a direct thrombin inhibitor, emerging as a preferred alternative to warfarin due to its favorable risk-benefit profile and convenience. However, anticoagulation therapy, including dabigatran, is associated with bleeding risks. This study explores the relationship between peripheral leukocyte count and bleeding events in patients with non-valvular atrial fibrillation (NVAF) treated with dabigatran, aiming to identify potential predictors of bleeding and improve patient safety.
Background
Atrial fibrillation is a major contributor to stroke and systemic embolism, particularly in patients with NVAF. Dabigatran, approved for stroke prevention in NVAF, has demonstrated superior efficacy and safety compared to warfarin. Despite its advantages, dabigatran is not without risks, as bleeding events, both minor and major, have been documented. Leukocytosis, or elevated leukocyte count, has been linked to various inflammatory and pathological conditions, including bleeding. Previous studies have suggested a correlation between high leukocyte counts and bleeding in conditions such as subarachnoid hemorrhage and acute coronary syndromes. This study investigates whether a similar association exists in NVAF patients treated with dabigatran.
Methods
The study enrolled 851 NVAF patients from 12 centers in China between February 2015 and December 2017. Patients were prescribed dabigatran 110 mg twice daily and followed up for six months. Baseline leukocyte counts were measured, and bleeding events were recorded during follow-up. Bleeding events were classified as major or minor based on established criteria. Major bleeding was defined as a significant drop in hemoglobin, transfusion of at least two units of blood, or symptomatic bleeding in critical areas requiring hospitalization. All other bleeding events were considered minor.
Covariates included demographic data, medical history, and laboratory parameters such as age, body mass index (BMI), blood pressure, hemoglobin, platelet count, and estimated glomerular filtration rate (eGFR). Statistical analyses involved univariate and multivariate Cox proportional hazards models to assess the relationship between leukocyte count and bleeding. Non-linearity was addressed using cubic spline functions and smooth curve fitting, with an inflection point calculated using a recursive algorithm.
Results
During the six-month follow-up, 87 patients experienced bleeding events, including 83 minor and four major cases. The most common bleeding events were hematuria (51 cases), gingival bleeding (11 cases), gastrointestinal bleeding (10 cases), and skin ecchymosis (10 cases). The baseline characteristics of participants were analyzed according to leukocyte count tertiles: tertile 1 (6.88×10⁹/L). Patients in tertile 3 were younger, had higher platelet and hemoglobin levels, and a higher prevalence of hypertension compared to those in tertiles 1 and 2.
Multivariate Cox proportional hazards models revealed that for every 1×10⁹/L increase in leukocyte count, the risk of bleeding increased by 11% (hazard ratio [HR]: 1.11, 95% confidence interval [CI]: 0.99–1.25). Smooth curve fitting demonstrated a non-linear relationship between leukocyte count and bleeding events, with an inflection point at 6.75×10⁹/L. Below this threshold, the HR for bleeding was 0.88 (95% CI: 0.69–1.13), indicating no significant association. Above the threshold, the HR increased to 1.28 (95% CI: 1.09–1.51), suggesting a significant rise in bleeding risk.
Discussion
This study identified a J-shaped association between baseline leukocyte count and bleeding risk in NVAF patients treated with dabigatran. Patients with leukocyte counts above 6.75×10⁹/L were at a higher risk of bleeding, while those below this threshold showed no significant association. These findings align with previous studies linking leukocytosis to increased bleeding in various clinical settings, including acute coronary syndromes and cardiac surgery.
The potential mechanism underlying this association may involve the interaction between leukocytes and platelets, reducing platelet-mediated procoagulation and increasing bleeding risk. Additionally, leukocytosis may reflect an underlying inflammatory state, which could exacerbate bleeding tendencies. However, further research is needed to elucidate the exact mechanisms.
The study has several clinical implications. First, it highlights the importance of monitoring leukocyte counts in NVAF patients on dabigatran, as elevated levels may indicate a higher risk of bleeding. Second, it suggests that patients with leukocyte counts above 6.75×10⁹/L may benefit from closer monitoring or dose adjustments to mitigate bleeding risks. Finally, the findings contribute to the development of predictive models for bleeding in NVAF patients, potentially improving treatment outcomes.
Limitations
The study has some limitations. First, the participants were exclusively Chinese, limiting the generalizability of the findings to other populations. Second, the study used a 110 mg dose of dabigatran, and the results may not apply to patients on the 150 mg dose. Third, leukocyte counts were measured only once at baseline, which may not account for fluctuations over time. Fourth, as an observational study, there may be unmeasured confounders influencing the results.
Conclusion
This study found a J-shaped association between baseline leukocyte count and bleeding risk in NVAF patients treated with dabigatran. Patients with leukocyte counts above 6.75×10⁹/L were at a significantly higher risk of bleeding, while those below this threshold showed no significant association. These findings underscore the importance of monitoring leukocyte counts in NVAF patients on dabigatran and may inform the development of predictive models for bleeding risk. Further research is needed to validate these findings and explore the underlying mechanisms.
doi.org/10.1097/CM9.0000000000000423
Was this helpful?
0 / 0