Evaluation of CHA₂DS₂-VASc risk stratification tool in pregnant women with atrial fibrillation or atrial flutter

BACKGROUND: The CHA₂DS₂-VASc score has not been validated as a risk stratification tool to determine need for anticoagulation therapy in pregnant women with atrial fibrillation (AF) and atrial flutter (AFL).

HYPOTHESIS: The CHA₂DS₂-VASc score will underestimate thromboembolic event (TE) risk in the hypercoagulable setting of pregnancy. existing or newly diagnosed AF or AFL delivering at a tertiary care institution from 2004-2017. AF or AFL was confirmed by electrocardiogram, holter monitor or electrophysiology study. The primary outcome was systemic TE up to six months post-partum (stroke, transient ischemic event, left atrial appendage [LAA] thrombus, other systemic TE); secondary outcome was bleeding events (post-partum hemorrhage or transfusion requirement).

METHODS: This is a restrospective cohort study of women with pre-existing or newly diagnosed AF or AFL delivering at a tertiary care institution from 2004-2017. AF or AFL was confirmed by electrocardiogram, holter monitor or electrophysiology study. The primary outcome was systemic TE up to six months post-partum (stroke, transient ischemic event, left atrial appendage [LAA] thrombus, other systemic TE); secondary outcome was bleeding events (post-partum hemorrhage or transfusion requirement).

RESULTS: AF was diagnosed in 42/53 (79%) women; 7/53 (13%) had AFL and 4/53 (8%) had both. Therapeutic anticoagulation was prescribed to 11 (21%) women during pregnancy and 13 (25%) post-partum; aspirin was prescribed to 13 (25%) during pregnancy and 11 (21%) post-partum. TE occurred in 3/53 (5.7%) women; TE occurred in 2/35 (5.7%) of women with CHA₂DS₂-VASc scores of 1 (predicted risk 0.5%), and 1/13 (7.7%) with CHA₂DS₂-VASc scores of 2 (predicted risk 1.0%). There was no increased bleeding risk for women on therapeutic anticoagulation during pregnancy (18% vs. 14%, p=0.7), but there was a trend towards more bleeding in women on therapeutic anticoagulation post-partum (31% vs. 10%, p=.09).

CONCLUSIONS: The CHA₂DS₂-VASc score underestimates the risk of systemic TE events in pregnant women with AF/AFL. Pregnancy is a hypercoagulable state which may increase the risk of TE in women with AF/AFL, even postpartum.