Comparison of warfarin dosage needed to achieve target INR before and after cardiac surgery

Br J Cardiol 2016;23:(1)doi:10.5837/bjc.2016.006 Leave a comment
Click any image to enlarge
First published online February 2, 2016

Delays in achieving target international normalised ratio (INR) with warfarin after cardiac surgery can lead to suboptimal outcomes. The aims of the study are to determine the difference in warfarin dosage requirements, before and after cardiac surgery, needed to achieve therapeutic anticoagulation.

A single-centre, retrospective review was conducted from 2012 to 2014 in cardiac surgery patients who were on warfarin pre-operatively and who had warfarin therapy resumed postoperatively in hospital. The primary outcome was the difference in warfarin dosage needed to achieve target INR before and after cardiac surgery. 

Ninety-five patients were included in the study. The mean daily postoperative warfarin dose needed for achieving a therapeutic INR was 0.18 mg lower than the mean pre-operative dose (5.03 ± 2.10 vs. 4.85 ± 2.25 mg; p=0.358). 

In conclusion, there was no statistically significant difference in the warfarin dosage before and after cardiac surgery needed to achieve target INRs.


Warfarin is an anticoagulant commonly used in atrial fibrillation, venous thromboembolism, prosthetic cardiac valve replacement and postoperative atrial fibrillation.1 Warfarin is usually discontinued prior to cardiac surgery and subsequently re-initiated postoperatively to achieve the target therapeutic international normalised ratio (INR).2 At the cardiac surgery unit of St. Paul’s Hospital, it has been observed that the warfarin dosage needed to achieve therapeutic anticoagulation is often lower post-cardiac surgery, compared with the patient’s warfarin dose prior to cardiac surgery. Serious complications, such as postoperative thromboembolic and haemorrhagic events, are consequences of the cardiac surgery itself and under- and over-anticoagulation.3,4 It is important to optimise anticoagulation in post-cardiac surgery patients, and delays can lead to prolonged hospitalisations due to the above complications.

No studies have been published that quantify the difference, if any, in warfarin dosage needed to achieve therapeutic INR, specifically, before and after cardiac surgery. The purpose of this study is to determine if there is a difference in warfarin dosage before and after cardiac surgery needed to achieve therapeutic anticoagulation.



This single-centre, retrospective chart review was approved by the Providence Health Care Research Ethics Board and by the Fraser Health Research Ethics Board (Providence Health Care Reference number UBC PHC H13-02182). The need for informed consent was waived.


Eligible patients were identified through the pharmacy database and patient chart reviews. Patients were included if they were admitted to St. Paul’s Hospital for cardiac surgery between 1 February 2012 and 4 February 2014, were on warfarin prior to cardiac surgery and achieved therapeutic anticoagulation, had warfarin therapy resumed postoperatively in hospital, and achieved INRs within the therapeutic range (as appropriate for the indication for warfarin) during hospitalisation. Exclusion criteria were: warfarin use for less than two months prior to the cardiac surgery date; undocumented or unclear pre-operative warfarin dose; postoperative warfarin therapy lasting less than five days prior to discharge; postoperative admission to the intensive care unit (ICU).


The primary outcome was the difference in warfarin dose, before and after cardiac surgery, required to achieve and maintain the patient’s target INR. The target INR was defined as the therapeutic target INR based on the patient’s indication for anticoagulation postoperatively. The secondary outcomes were the rates of major and minor bleeding and thromboembolic events during the postoperative hospitalisation period. Major bleeding events were defined as those that were intracranial or retroperitoneal, led to death, necessitated transfusion, warranted interruption of antithrombotic treatment, or required operation. All other haemorrhagic events documented in the patient charts were considered to be minor bleeding events. Thromboembolic events were defined as ischaemic stroke, transient ischaemic attack, systemic embolism, deep vein thrombosis, pulmonary embolism, or cardiac valve thrombosis.

For the primary outcome analysis, the postoperative warfarin dose required to achieve and maintain a therapeutic INR was determined by averaging the doses from the day warfarin was initiated post-operatively until the day that a therapeutic INR was achieved. The pre-operative warfarin dose was obtained from chart records of the patient’s home warfarin dose prior to cardiac surgery.

Statistical analysis

The pre-operative and postoperative maintenance warfarin doses were compared using the paired samples t-test; p values less than 0.05 were considered to be statistically significant. Descriptive statistics were used to analyse the secondary outcomes of bleeding and thromboembolic event rates.


There were 1,402 patients screened, of which 95 were eligible for the study. The majority of patients received coronary artery bypass graft surgery. Warfarin therapy was restarted after a median of one day postoperatively, and the median postoperative length of stay was nine days.

The 95 patients achieved a therapeutic INR after a median of five postoperative days. The overall mean postoperative maintenance warfarin dose was 0.18 mg lower than the pre-operative dose (5.03 ± 2.10 vs. 4.85 ± 2.25 mg) and this difference was not statistically significant (p=0.358). Similarly, no statistically significant differences in warfarin dose were found in any of the subgroups, which were defined by cardiac surgery type and by whether warfarin-interacting medications or products were initiated pre-operatively (see table 1).

Table 1. Primary outcome (n=95)
Table 1. Primary outcome (n=95)

Four patients (3.6%) had one major bleeding event postoperatively during their hospital stay. At the time of the major bleeding events, none of the four patients had supratherapeutic INRs. Half (50%) of major bleeds also occurred in patients taking three or more antithrombotic agents in addition to warfarin, whereas all patients with no bleeding events or minor bleeding events were taking two or fewer concomitant antithrombotic agents. Ten minor bleeding events occurred overall and the INRs were supratherapeutic at the time of only two of these events.

Three patients (2.7%) had one thromboembolic event and all events were ischaemic strokes. In two of these patients, the INR was subtherapeutic at the time of the stroke, and the INRs had been consistently subtherapeutic for the previous three and six days prior to the event.


This is the first study undertaken to determine the changes in warfarin dosage requirements, specifically, before and after cardiac surgery. In the postoperative setting, practitioners rely heavily on frequent INR monitoring, patients’ past warfarin dosage requirements and on the patient’s history of bleeding and thromboembolic events to guide warfarin dosage adjustments. No statistically or clinically significant difference in warfarin dosage requirements was found overall, or in any of the subgroups.

Although no previous studies have quantified changes in warfarin dosage requirements after cardiac surgery, alterations in warfarin response in the postoperative period have been reported. A study by Schulman et al. examined postoperative changes in warfarin response among general surgical patients, 156 of which had received cardiac surgery.6 Warfarin responses after cardiac surgery were classified as altered in 34% of patients: resistant in 19%, and sensitive in 15%.6 However, Schulman et al.’s criteria for warfarin responsiveness classifications were not clearly defined. Of the clinical factors studied, postoperative commencement of amiodarone was the only factor identified to influence warfarin responses to a statistically significant degree. 6 This effect of concomitant amiodarone use was reproduced in our study as well.

Studies by Rahman et al. and Lee et al. showed that patients receiving heart valve replacements had decreased warfarin requirements immediately following the surgery.7,8 The authors concluded that the differences were attributed to postoperative physiological changes, such as improvements in cardiac function, with consequent decreases in fluid retention, relief of hepatic congestion, increases in warfarin metabolism, the deleterious effect of cardiopulmonary bypass (CPB) on clotting factors that worsens as the CPB time is prolonged, and hypoalbuminaemia following CPB, which affects warfarin’s half-life and volume of distribution.8,9 Such physiological changes post-cardiac surgery did not translate into any clinical difference in warfarin dosage needed to achieve therapeutic anticoagulation in our study.

Peri-operative bleeding events are of particular concern in cardiac surgery patients.10 Our results suggest that major bleeding events in this study may have been associated with the number of concomitant antithrombotics administered. The INRs were subtherapeutic at the time of the majority of bleeding events, which suggests that the intensity of anticoagulation with warfarin may not be a primary factor associated with the bleeding events, but perhaps the number of antiplatelet and anticoagulants are.

Our study has a number of limitations. The largest limitation was the retrospective nature of the study. Important factors may not have been detected by our retrospective review, such as undocumented bleeding or thromboembolic events, as well as factors that affect warfarin response, such as dietary vitamin K intake, gut malabsorption and liver dysfunction. Our study was also limited by its small sample size, which may have contributed to the lack of statistical significance detected among the analyses. As well, while our study demonstrates that therapeutic anticoagulation was achieved using pre-operative warfarin dosages, it is unclear if this would translate to long-term therapeutic anticoagulation. This would be more difficult to determine as the times in therapeutic range with warfarin vary depending on the anticoagulation care available, ranging from 29% to 75% in outpatient atrial fibrillation patients.11

As well, our findings are only applicable to our institution and, possibly, post-cardiac surgery patients in general, as hospital diet composition varies greatly between hospitals. Other factors, such as cultural preferences for certain types of food and the ability to tolerate an oral diet, would affect dietary intake of vitamin K. This variability in dietary intake of vitamin K post-operatively would influence the warfarin dosage needed to achieve therapeutic anticoagulation. Thus, we would not be able to extrapolate our findings to other hospitalised patients or other institutions given the great variability in dietary vitamin K intake. Further studies should be undertaken to validate our findings in other surgical and hospitalised patient populations in different institutions.

The fact that our study demonstrated no difference in warfarin dosage needed for therapeutic anticoagulation pre-operatively and postoperatively raises some interesting mechanistic questions. Other factors postoperatively, such as an increased inflammatory state, effect of an acutely stressed physiological state on clotting factors and intra-operative anticoagulation may influence dose-responsiveness to warfarin postoperatively. Future mechanistic studies are warranted to elucidate the physiologic factors post-cardiac surgery that would impact warfarin dose-responsiveness.


In conclusion, there were no statistically significant differences in warfarin dosage needed to achieve therapeutic INR, before and after cardiac surgery. The majority of recorded bleeding and thromboembolic events occurred while the INR was subtherapeutic. As the study was limited by its retrospective nature and small sample size, larger prospective studies are warranted to validate these findings.

Key messages

  • Delays in achieving therapeutic anticoagulation with warfarin after cardiac surgery can lead to prolonged hospitalisation and potential for thromboembolism
  • Cardiac surgery can cause physiological changes that may alter the body’s response to anticoagulation
  • This study illustrates that the warfarin dose needed to achieve therapeutic anticoagulation after cardiac surgery is similar to pre-operative warfarin dosage requirements
  • Patients will likely require similar warfarin dosages postoperatively compared with pre-operatively to achieve therapeutic anticoagulation


We acknowledge Jessica Chang, BSc(Pharm), and Flora Yu, BSc(Pharm), for assistance in the data collection for this study.

Conflict of interest

None declared.


1. Tadros R, Shakib S. Warfarin – indications, risks and drug interactions. Aust Fam Physician 2010;39:476–9. Available from:

2. Guyatt GH, Akl EA, Crowther M, Gutterman DD, Schuünemann HJ. Executive summary: antithrombotic therapy and prevention of thrombosis, 9th edition. American College of Chest Physicians evidence-based clinical practice guidelines. Chest 2012;141(2 suppl):7S–47S.

3. Irwin AN, McCool KH, Delate T, Witt DM. Assessment of warfarin dosing requirements after bariatric surgery in patients requiring long-term warfarin therapy. Pharmacotherapy 2013;33:1175–83.

4. Zareh M, Davis A, Henderson S. Reversal of warfarin-induced hemorrhage in the emergency department. West J Emerg Med 2011;12:386–92.

5. Gulseth M. Managing Anticoagulation Patients in the Hospital: The Inpatient Anticoagulation Service. Bethesda (MD): American Society of Health-System Pharmacists (ASHP), 2007.

6. Schulman S, El Bouazzaoui B, Eikelboom JW, Zondag M. Clinical factors influencing the sensitivity to warfarin when restarted after surgery. J Intern Med 2008;263:412–17.

7. Rahman M, BinEsmael TM, Payne N, Butchart EG. Increased sensitivity to warfarin after heart valve replacement. Ann Pharmacother 2006;40:397–401.

8. Lee J, Lee B, Kim K et al. Factors affecting warfarin therapy following cardiac valve surgery. Ann Pharmacother 2002;36:1845–50.

9. Harker LA, Malpass TW, Branson HE, Hessel EA, Slichter SJ. Mechanism of abnormal bleeding in a patient undergoing cardiopulmonary bypass: acquired transient platelet dysfunction associated with selective alpha-granule release. Blood 1980;56:824–34.

10. Tettey M, Aniteye E, Sereboe L et al. Predictors of post operative bleeding and blood transfusion in cardiac surgery. Ghana Med J 2009;43:71–6.

11. Wan Y, Heneghan C, Perera R et al. Anticoagulation control and prediction of adverse events in patients with atrial fibrillation. Circ Outcomes 2008;1:84–91.