Warfarin dosing nomogram for initiating therapy

A 60 year old female patient presents to the emergency department with complaints of swelling and pain in her right leg.  She recently had her knee replaced and has healed well since the procedure.  She has a PMH of HTN and CKD (Stage IV).  She has no other complaints at this time.  She is diagnosed with a proximal DVT on lower extremity ultrasound and the decision is made to anticoagulate her with warfarin plus a parenteral anticoagulant.  Her baseline INR is 1.28.  What strategy can we use to initiate her on warfarin to reach a therapeutic INR in a reasonable amount of time without overanticoagulating her?

The problem

Though we have several decades of clinical experience using warfarin for a variety of conditions, prescribers today still struggle with dosing strategies to maintain a therapeutic INR.  A recent study found that of patients that were "stable" for 6 months (80% INRs within therapeutic range), only 34% remained "stable" over the next year.1  

In the setting of acute VTE, it is desirable to achieve a goal INR in a reasonable amount of time without overshooting the therapeutic range and increasing the patients' risk of bleeding.  Each patient has individual needs for warfarin dosing depending partly on the following variables:

• Genetic factors - eg. CYP450 2C9 & VKORC1 mutations
• Dietary factors - eg. fluctuating levels of dietary vitamin K
• Drug interactions - affecting absorption, metabolism, distribution, or elimination
• Herbal supplements - often because they affect the above two factors
• Comorbid conditions - eg. hepatic dysfunction, hyperthyroidism, heart failure

Many models have been developed that incorporate the various above factors in an effort to better predict warfarin requirements (initial and maintenance dosing) and improve the time in therapeutic range.  Some difficulties are the limited availability and cost of testing for genetic mutations and the fact that even comprehensive prediction models (including both genetic and clinical factors) only account for about 50% of the interindividual variability with respect to warfarin dose.2  

Example dosing nomogram

One warfarin dosing nomogram that is simple to use and does not require testing for genetic variations was developed and used by UCLA Medical Center.  Ordering physicians there are even able to write, "Warfarin dosing per Pharmacy", which allows the pharmacist to manage warfarin according to their guidelines.

The UCLA nomogram is simple to use and essentially involves two steps:

     1.  Determine if your patient likely has high, moderate, or low warfarin sensitivity as determined by the following table:

High warfarin sensitivity	Moderate warfarin sensitivity	Low warfarin sensitivity
Baseline INR >1.5 Age >65 Significant hepatic disease Decompensated HF Malnourishment Malabsorption syndrome/ chronic diarrhea Cancer Hypoalbuminemia (<2g/dL) Thyrotoxicosis Genetic polymorphism of CYP450 2C9	Baseline INR 1.2-1.5 Age 50-65 Concurrent CYP450 enzyme inhibitor	Baseline INR <1.2 Age <50

     2.  Look at the INR every day of therapy and order the corresponding dose according to this table:

Day	INR	High warfarin sensitivity	Moderate warfarin sensitivity	Low warfarin sensitivity
Day 1	Baseline INR	2-5 mg	5 mg	7.5 mg
Day 2	<1.5 1.5-1.9 2-2.5 >2.5	2-5 mg 2 mg 1-2 mg none	5 mg 2.5 mg 1-2.5 mg none	5-7.5 mg 2.5-5 mg 1-2.5 mg non
Continue below for all patients
Day 3	<1.5 1.5-1.9 2-2.5 2.6-3 >3	5-10 mg 2.5-5 mg 0-2.5 mg 0-2.5 mg none
Day 4	<1.5 1.5-1.9 2-3 >3	10 mg 5-7.5 mg 2.5-5 mg 0-2.5 mg
Day 5	<1.5 1.5-1.9 2-3 >3	10 mg 5-7.5 mg 2.5-5 mg 0-2.5 mg
Day 6	<1.5 1.5-1.9 2-3 >3	7.5-12.5 mg 5-10 mg 2.5-5 mg 0-2.5 mg
Day 7	Make adjustment based on total weekly dose (increase or decrease) by 5-20% depending on current INR and target INR

Notes about the UCLA nomogram:

• The nomogram should be initiated on day number one of treatment, not after the patient has already taken various doses as there is a significant lag time on each dose affecting the INR
• While it provides guidance on dosing, it is not completely rigid in many cases and still allows for/requires clinical judgement.  For example, on Day 3 a person with an INR of <1.5 could take between 5-10 mg.  This is where clinical judgement comes into play and requires a keen look at comorobidities, diet, other medications, etc..
• The nomogram has not been validated in prospective studies but validation studies are underway
• The dosing strategy in the nomogram was based on the ACCP Chest Guidelines

Back to the patient case

For treatment of a new VTE, full anticoagulation with either a non-vitamin K oral anticoagulant (NOAC) or a warfarin/heparin bridge is necessary.  This patient's stage IV CKD makes the NOACs and LMWH undesirable or these would be our primary choices - leaving us with initiation of warfarin plus unfractionated heparin.3,4  

Because our patient has not received any warfarin yet, the UCLA warfarin dosing nomogram is a viable tool to help determine the daily warfarin dose.  The first step would be to identify her warfarin sensitivity, which in this case would be "moderate sensitivity" based on her age, baseline INR, and lack of the listed comorbid conditions above.  The next step would be to select the dose in the corresponding "moderate sensitivity" column of the table above - 5mg on day 1.  On day 2 we can proceed down the table to select the dose based on day 2's INR.  We should continue in this fashion, checking the INR daily and dosing as per the table, until the INR is therapeutic (2-3) and at least 5 days have elapsed, at which point the parenteral anticoagulant should be discontinued but warfarin continued.

Take home points:

• Initial warfarin dosing is challenging because of many factors affecting patient response
• Nomograms and websites have been developed to assist in initial dosing, none of which completely account for variability in response
• The above nomogram is a straight-forward example that does not require the use of genetic testing 

References:

1.  Pokorney SD, Simon DN, Thomas L, et al.  Stability of International Normalized Ratios in patients taking long-term warfarin therapy.  JAMA  2016;316(6):661-3.
2.  Gage B, Eby C, Johnson J, et al.  Use of pharmacogenetic and clinical factors to predict the therapeutic dose of warfarin.  Clin Pharmacol Ther  2008;84(3):326-31.
3.  Kearon C, Akl EA, Comerota AJ, et al.  Antithrombotic therapy for VTE disease.  Antithrombotic therapy and prevention of thrombosis, 9th ed:  American College of Chest Physicians evidence-based clinical practice guidelines.  CHEST 2012;141(2):e419S-94S.
4.  Kearon C, Akl EA, Ornelas J, et al.  Antithrombotic therapy for VTE disease.  CHEST  2016;149(2):315-52.

photo by Kevin Gilmour