Clinical factors influencing the sensitivity to warfarin when restarted after surgery

Effect of Early Amiodarone Use on Warfarin Sensitivity, Blood Product Use, and Bleeding in Patients With a Left Ventricular Assist Device

Data are scarce on the effect of amiodarone on warfarin sensitivity and related outcomes after placement of a left ventricular assist device (VAD). This retrospective study compared 30-day outcomes between patients on amiodarone vs no amiodarone after VAD implant. After exclusions, 220 patients received amiodarone and 136 patients did not. Compared to the no amiodarone group, the amiodarone group had a higher warfarin dosing index (0.53 [0.39, 0.79] vs 0.46 [0.34, 0.63]; P = 0.003), incidence of INR ≥ 4 (40.5 vs 23.5%; P = 0.001), incidence of bleeding (24.1 vs 14%; P = 0.021), and use of INR reversal agents (14.5 vs 2.9%, P ≤ 0.001). Amiodarone was associated with bleeding (OR, 1.95; 95% CI, 1.10-3.47; P = 0.022), but not after adjusting for age, estimated glomerular filtration rate, and platelet count (OR, 1.67; 95% CI, 0.92-3.03; P = 0.089). After VAD implant, amiodarone was associated with increased warfarin sensitivity and administration of INR reversal agents.

Identification of clinical factors predicting warfarin sensitivity after cardiac surgery

Background

Warfarin is commonly initiated post-cardiac surgery to reduce the risk of intracardiac thrombus formation. Studies have found that sensitivity is increased after cardiac surgery and anticoagulation is subsequently difficult to manage. This study set out to identify clinical markers of increased warfarin sensitivity in patients after cardiac surgery, and build a model that can predict warfarin sensitivity, and improve safety in this setting.

Methods

The study was an observational, retrospective cohort design. Clinical parameters including left ventricular ejection fraction (LVEF), cross-clamp time, age, serum albumin and C-reactive protein concentrations were collected from consenting patients who had undergone cardiac surgery and were prescribed postoperative warfarin. The warfarin dose index (WDI) was calculated for each patient from their international normalized ratio (INR) and warfarin dose, as a measure of sensitivity.

Results

A total of 41 patients were recruited to the study. Logarithmically transformed WDI (log WDI) significantly correlated with LVEF, cardiopulmonary bypass (CPB) time, cross-clamp time, baseline INR and co-administration of amiodarone (p < 0.05). When added to a linear regression model, LVEF and cross-clamp time produced a model that accounted for 41% of the variance in log WDI (R² = 0.41), p = 0.0002). Applying a log WDI cutoff value of -0.349 discriminated between patients who develop an INR > 4 and those who do not, with a sensitivity of 75% and a specificity of 70%.

Conclusions

This single-centre study has highlighted two risk factors for increased warfarin sensitivity post-cardiac surgery. Further research is needed to confirm these findings in a wider, more diverse population, and to validate this model.

Comparing Usual Care With a Warfarin Initiation Protocol After Mechanical Heart Valve Replacement

BACKGROUND

The immediate postoperative warfarin sensitivity for patients receiving heart valve prostheses is increased. Established warfarin initiation protocols may lack clinical applicability, resulting in dosing based on clinical judgment.

OBJECTIVE

To compare current practice for warfarin initiation with a known warfarin initiation protocol, with doses proportionally reduced to account for the increased postoperative sensitivity.

METHODS

We compared the Mechanical Heart Valve Warfarin Initiation Protocol (Protocol group) with current practice (clinical judgment-Empirical group) for patients receiving mechanical heart valves in an observational before-and-after format. End points were the time to achieve a stable therapeutic international normalized ratio (INR), doses held in the first 6 days, and overanticoagulation in the first 6 days.

RESULTS

The Protocol group (n = 37) achieved a stable INR more rapidly than the Empirical group (n = 77; median times 5.1 and 8.7 days, respectively; P = 0.002). Multivariable analysis indicated that the Protocol group (hazard ratio [HR] = 2.22; P = 0.005) and men (HR = 1.76; P = 0.043) more rapidly achieved a stable therapeutic INR. Age, serum albumin, amiodarone, presence of severe heart failure, and surgery type had no impact. Protocol patients had fewer doses held (1.1% vs 10.1%, P < 0.001) and no difference in overanticoagulation (2.7% vs 9.1%, P = 0.27).

CONCLUSION

The Mechanical Heart Valve Warfarin Initiation Protocol provided a reliable approach to initiating warfarin in patients receiving mechanical aortic or mitral valves.

Assessment of warfarin dosing requirements after bariatric surgery in patients requiring long-term warfarin therapy

STUDY OBJECTIVE

To quantify the change in weekly warfarin dose after bariatric surgery in patients requiring long-term warfarin therapy.

DESIGN

Retrospective matched-cohort study.

SETTING

Anticoagulation management service in an integrated health care delivery system.

PATIENTS

Patients receiving long-term warfarin anticoagulation who underwent bariatric surgery between January 1, 1996, and December 31, 2010 (27 patients), were matched by date of surgery (± 2 years), age (± 5 years), and target international normalized ratio (INR) range to patients receiving long-term anticoagulation therapy who underwent other abdominal surgical procedures: cholecystectomy or endoscopic retrograde cholangiopancreatography (59 patients [control group]).

MEASUREMENT AND MAIN RESULTS

The main end point was change in postoperative warfarin dose from baseline (preoperative dose), measured at weekly postoperative intervals from weeks 1 to 8 and again at months 3 and 6. After surgery, patients in the bariatric surgery group had statistically significant decreases in weekly warfarin doses compared with preoperative dose at all postoperative time points (week 1 dose vs preoperative dose, p<0.01; doses at all other time points vs preoperative dose, p<0.001), except at 6 months (p>0.05). No statistically significant decreases in warfarin dose were detected at any postoperative time points in the control group. Twenty patients (74.1%) in the bariatric surgery group experienced a 20% or more decrease in weekly warfarin dose compared with 19 patients (32.2%) in the control group (p=0.004). No significant differences in warfarin-related adverse events were noted between groups.

CONCLUSION

Weekly warfarin doses decreased in the immediate postoperative period in anticoagulated patients after bariatric surgery but returned to their preoperative doses after approximately 6 months. A similar pattern was not observed in patients in the control group who underwent other types of abdominal surgery. Compared with preoperative anticoagulation control, this resulted in reduced anticoagulation control despite close INR monitoring. If a causal relationship between bariatric surgery and warfarin sensitivity is established in future research, developing and validating a postbariatric surgery warfarin-dosing algorithm would be valuable.

Clinical impact of temporary therapy interruptions on anticoagulation control in patients treated with warfarin

This retrospective cohort study was completed to describe the impact of short-term therapy interruptions on anticoagulation control in patients receiving warfarin. Patients seen in a pharmacist-managed anticoagulation clinic were included if they were on a stable warfarin dose and then underwent a planned interruption in therapy. Patients were excluded if phytonadione was administered before the interruption or if medications known to interact with warfarin were altered during the interruption. Data were analyzed for 2 groups: (1) patients with a single interruption in therapy (group 1) and (2) patients with a single interruption in therapy plus patients with an extended interruption in therapy (group 2). The primary endpoint was the change in weekly maintenance warfarin dose from preinterruption to postinterruption. Evaluation of 199 patients resulted in 31 interruptions in group 1 and 34 interruptions in group 2. A change in dose was required in 58% of patients in group 1 and 56% of patients in group 2. The mean absolute change in dose was 2.03 ± 2.79 mg (P < 0.003) in group 1 and 1.96 ± 2.72 mg (P < 0.002) in group 2. For the majority of patients, the dose change represented <10% of their preinterruption weekly dose. Of patients requiring a dose change, 50% required an increase in dose. In conclusion, close follow-up is warranted after a warfarin therapy interruption as dose adjustments will likely be needed to regain anticoagulation control and the direction of this dose change cannot be predicted.

Patient Factors That Influence Warfarin Dose Response

Warfarin has long been the mainstay of oral anticoagulation therapy for the treatment and prevention of venous and arterial thrombosis. The narrow therapeutic index of warfarin, and the complex number of factors that influence international normalized ratio (INR) response, makes optimization of warfarin therapy challenging. Determination of the appropriate warfarin dose during initiation and maintenance therapy requires an understanding of patient factors that influence dose response: age, body weight, nutritional status, acute and chronic disease states, and changes in concomitant drug therapy and diet. This review will examine specific clinical factors that can affect the pharmacokinetics and pharmacodynamics of warfarin, as well as the role of pharmacogenetics in optimizing warfarin therapy.