Impact of Atrial Fibrillation on Patients with Myeloproliferative Neoplasms

Introduction

Polycythemia Vera (PV) and Essential Thrombocythemia (ET) are two entities characterized by the occurrence of thrombotic events in both arterial and venous territories, as well as the existence of hemorrhagic episodes. The impact of Atrial Fibrillation (AF), a pathology with an intrinsic prothrombotic nature that necessitates anticoagulant treatment in most patients, is not precisely known in this subgroup of patients.

Objective

• To elucidate the impact of AF in patients with Chronic Myeloproliferative Neoplasms (MPNs) concerning the occurrence of thrombotic events, hemorrhagic events, and total/cardiovascular mortality.
• To calculate the net thrombo-hemorrhagic balance in different clinical situations (age stratified >60 vs <60 years, previous event, JAK2 positive, sex...) to understand the inherent risk of anticoagulant treatment use.
• To determine the efficacy and safety profile of antiVitaminK vs. DOACs in this subgroup of patients.
• To understand the predictive power of classical thrombosis and bleeding scales (CHADSVASC-HASBLED).

Material or Patients and Method

This is an observational, ambispective study supported by patients from the GEMFIN Registry, which has involved the active collaboration of more than 60 centers, developed over the last 5 years. It comprises 4169 patients with PV or ET, among whom 313 (7.5%) have concomitant AF. The comparison cohorts have been homogenized using a propensity score matching model 2:1 (2 patients without arrhythmia for each patient with AF). The variables considered in the model are: sex, age, history of stroke/TIA, infarction, Diabetes Mellitus, HTN, venous event, bleeding history, type of myeloproliferative neoplasm (ET or PV), presence of JAK2 mutation, and cytoreductive treatment.

• The balance has been calculated as a difference in the incidence of events (thrombosis minus hemorrhage).
• The comparison between anticoagulants has been performed using a multivariate study.
• The scales have been tested using ROC curves.

Results

The presence of AF was associated with a Hazard Ratio (HR) of total thrombotic events of 1.93 (95% CI 1.34 – 2.77) and TIA/stroke of 2.04 (1.13 – 3.68). Regarding bleeding, the HR of total hemorrhages was 2.33 (1.55 – 3.52), increasing to 4.86 (2.73 – 8.64) in the case of major hemorrhages. When evaluating mortality, patients with AF had an HR of 1.73 (1.36 – 2.20) compared to patients without arrhythmia. Specifically assessing cardiovascular mortality, the result was 3.12 (1.80 – 5.4).

• The net balance shows that the thrombotic risk is higher than the hemorrhagic risk in all the considered scenarios, except in patients with recent previous bleeding.
• DOACs present a similar efficacy profile (total thrombosis: HR 1.94- 95% CI 0.69-5.46) and a superior safety profile (hemorrhage: 3.46 (95% CI 1.06 – 11.22) compared to antiVitK.
• AUC of CHADVASC 0.6990 (95% CI 0.64 – 0.75) and HASBLED 0.6217 (95% CI 0.57 – 0.68).

Conclusions

• The presence of AF concomitant with a Philadelphia-negative MPN is associated with a negative impact concerning the occurrence of thrombotic, hemorrhagic events as well as total and cardiovascular mortality.
• DOACs are suggested as the treatment of choice (lower bleeding rate).
• Classical scales are good discriminators of events in this subgroup.