Warfarin therapy is usually monitored using the international normalized ratio (INR), and prolonged INR means coagulation impairment that can leads elevating transfusion requirement after operation. This study was designed to assess the relationships with warfarin and TEG values related to transfusion amount, to set up blood transfusion strategy in patients on atrial fibrillation undergoing cardiac surgery.

This study was a retrospective evaluation of patients who have received cardiac surgery with Maze-operation from March 2008 to November 2013.

Patients who have atrial fibrillation undergoing cardiac surgery with Maze-operation were included.

The following exclusion criteria are applied:

1. Patients who have taken medicine related to coagulation (such as antiplatelets like aspirin and clopidogrel) except warfarin or bridged heparin
2. Patients whose laboratory values could not be confirmed by medical records review.

Patients were categorized by two groups.

1. W group: patients on atrial fibrillation who have taken warfarin for p.o medication.
2. C group: patients on atrial fibrillation who have not taken warfarin for p.o medication.

Using blood sample, authors examined as follows:

1. Preoperative (maximum 5days before operation) and postoperative (30min after ICU admission) coagulation values with haematologic values: PT/aPTT, INR, Hb, Hct, platelet levels.
2. ROTEM values at 30min after anaesthetic induction and 30min after ICU admission: Clotting time (CT), clot formation time (CFT), maximum clot firmness (MCF), maximum lysis (ML)
3. Intraoperative (during operation) and postoperative (during admission, an expected average of 5 weeks) total transfusion amount: RBC, FFP, PC, cryoprecipitate.
4. Post operative 24 hr bleeding amount in the chest tube drain bottle.
5. The duration (maximum 2 weeks) of warfarin discontinuation. (in W group)

Statistical analysis was performed using unpaired t-test and chi-square test to compare patients demographic data and other parameters. To determine the relationship of INR and ROTEM parameters with bleeding amount and transfusion amount, multiple linear regression model that accounted for all confounder variables was used. For each factor in the multiple regression model, variables with a p value of less than 0.10 were kept in the final model. The data was analyzed using the program Statistical Package for the Social Sciences ver. 18.0 (SPSS Inc, Chicago, USA). A value of p < 0.05 was considered statistically significant.

Inclusion Criteria:

1. Patients who have atrial fibrillation undergoing cardiac surgery with Maze-operation.

• W group : Patients who have had warfarin therapy before operation due to treat Atrial fibrillation
• C group : Patients who have not had warfarin therapy before operation

Exclusion Criteria:

• 1. Patients who have taken medicine related to coagulation (such as antiplatelets like aspirin and clopidogrel) except warfarin or bridged heparin.
• 2. Patients whose laboratory values could not be confirmed by medical records review.

Patients undergoing cardiac surgery with Maze-operation