To the Editor: Warfarin therapy is prescribed for the prevention of stroke and embolism, not only for patients who have a prosthetic valve but also for a growing number of elderly patients with atrial fibrillation. Control of the international normalized ratio (INR), by which warfarin therapy is monitored, is a daunting challenge. As stated by Gage, “Variation in the INR is unavoidable. …”
1
The annual risk of major hemorrhage can approach 3% to 4% and is proportional to the percentage of patient-days of INR at 5.0 or higher.1
In my prescribing experience, a key factor is the dosing schedule, about which little has been published. Warfarin is available in several strengths (1, 2, 2.5, 3, 4, and 5 mg) of scored tablets. Patients often require a daily dose that is neither a precise multiple of a single tablet nor a multiple of its half-strength. For example, a weekly requirement of 30 mg (4.28 mg/d) is customarily prescribed in some alternating-day fashion (that is, one 5-mg tablet on Sunday, Tuesday, Wednesday, Friday, and Saturday and one-half tablet on Monday and Thursday).
1
, 2
The term “customary” is used advisedly. This dosing pattern is not found in the manufacturer's package insert, standard pharmacological texts, or clinical trials. Rather, it is commonly used on the basis of tradition1
or personal experience.2
Because of day-to-day variation, confusion may arise on the part of the patient or prescriber.1
Wong et al3
compared “customary” dosing with “consistent” daily dosing (in multiples of 1 mg). Patients marginally preferred consistent daily dosing, but INR control was unaffected.3
As part of a continuous-improvement effort within my private practice in Pittsburgh, Pa, beginning January 1991 I switched all 22 patients (15 with prosthetic valves) receiving warfarin therapy to a consistent daily dose. This regimen differed from Wong's in that the warfarin dosing “menu” was constructed not in increments of 1 mg, but rather in increments of 0.25 mg. This “precise daily dose” strategy necessitated a 2-tablet system (Table 1). Patients were given 2 different-strength tablets (generally 2.5-mg and 1-mg tablets for patients age 65 and older and 5-mg and 1-mg tablets for younger patients). In the aforementioned example, a patient requiring 30 mg weekly would have been prescribed 4.25 mg daily. Periodic dose adjustments, in response to INR changes, were made in the customary way. Dose changes became transparent. Dosages were easier for the physician to calculate and communicate and easier for the patient (or surrogate) to understand. Only a brief educational effort for the patient was required. In 1992, I retrospectively reviewed the (then standard) prothrombin time test results of all my patients receiving customary warfarin therapy and compared these results with the prothrombin time test results of my patients who were managed with warfarin therapy in the 3 years before the warfarin dosing system conversion and then were managed with precise daily dosing for the next 2 years (Table 2).
Table 1Warfarin Doses in a 2-Tablet System
| No. of tablets | |||
|---|---|---|---|
| Daily amount (mg) | 1-mg (pink) | 2.5-mg (green) | 5-mg (peach) |
| 0.5 | ½ | na | na |
| 1 | 1 | na | na |
| 1.25 | na | ½ | na |
| 1.5 | 1 ½ | na | na |
| 1.75 | ½ | ½ | na |
| 2 | 2 | na | na |
| 2.25 | 1 | ½ | na |
| 2.5 | na | 1 | na |
| 2.75 | 1 ½ | ½ | na |
| 3 | ½ | 1 | na |
| 3.25 | 2 | ½ | na |
| 3.5 | 1 | 1 | na |
| 3.75 | na | 1 ½ | na |
| 4 | 1 ½ | 1 | na |
| 4.25 | ½ | 1 ½ | na |
| 4.5 | 2 | 1 | na |
| 4.75 | 1 | 1 ½ | na |
| 5 | na | 2 | na |
| na | na | 1 | |
| 5.25 | 1 ½ | 172 | na |
| 5.5 | ½ | na | 1 |
| 5.75 | 2 | 172 | na |
| 6 | 1 | na | 1 |
| 6.5 | 1 ½ | na | 1 |
| 7 | 2 | na | 1 |
| 7.5 | na | na | 172 |
| 8 | 3 | na | 1 |
* NA = not applicable.
Table 2Comparison of Prothrombin Time Test Results for Customary and Precise Daily Dosing
| Manner of warfarin dosing (calendar years) | No. of patients | No. of PT tests | No. (%) of PT tests within target range | No. (%) of PT tests ±1 second of target range |
|---|---|---|---|---|
| Customary (1988-1990) | 22 | 511 | 222 (43) | 311 (61) |
| Precise daily dosing (1991-1992) | 37 | 581 | 328 (57) | 421 (72) |
* PT = prothrombin time.
† Includes 15 new patients after January 1991.
I continued to use the refined and consistent daily dose in all patients in my practice throughout the next decade, even as additional clarity and control were gained through laboratories reporting results in INR.
In 2000, Samsa et al
4
published the first normative data of INR control of patients with chronic atrial fibrillation. Twenty-five systematically surveyed physician practices did not have anticoagulation services available. They serve as the benchmark against which most physicians’ practices can be compared. Discouragingly, patients’ INRs were within the target range on less than half of the days (Table 3, rows 1 and 2).Table 3Comparison of INR Control in Normative Data to Author's Clinical Observation in Patients With Chronic Atrial Fibrillation
| Practice location | No. of practice sites | No. of patients | Patient-days INR within target range (%) | Patient-days (per 1000) of INR ≥5 |
|---|---|---|---|---|
| Research Triangle Park, NC 4 | 8 | 92 | 36 | 28 |
| Rochester, NY 4 | 17 | 58 | 47 | 14 |
| Pittsburgh, Pa | 1 | 25 | 69 | 2 |
* INR = international normalized ratio.
† 2.0-3.0.
‡ Estimated.
§ Without anticoagulation service available.
‖ Author's private practice audit.
In an accompanying editorial, Ansell
5
commented that this work brought much-needed insight into the everyday real-world management of oral anticoagulation therapy. Ansell also called for better understanding of the manner in which other smaller practices fare in this regard.Using the method described by Samsa et al,
4
I reviewed the INR control of my 25 patients with atrial fibrillation (2000-2001) (Table 3, row 3). In comparison with the patients in Samsa's study, my patients were older (21 patients [84%] were age =70 years; 15 patients [60%] were age =75 years) and had greater comorbidity. Moreover, this degree of INR control was accomplished with a longer (mean) interval between blood tests, 24.9 days vs 22.2 days in Samsa et al4
(25 practices). The only disadvantage of the precise daily dosing system was the expense of 50% more tablet consumption.REFERENCES
- Warfarin therapy for an octogenarian who has atrial fibrillation.Ann Intern Med. 2001; 134: 465-474
- Oral anticoagulant therapy3/450 years later.Arch Intern Med. 1993; 153: 586-596
- Influence of warfarin regimen type on clinical and monitoring outcomes in stable patients in an anticoagulation management services.Pharmacotherapy. 1999; 19: 1385-1391
- Quality of anticoagulation management among patients with atrial fibrillation: results of a review of medical records from 2 communities.Arch Intern Med. 2000; 160: 967-973
- The quality of anticoagulation management [editorial].Arch Intern Med. 2000; 160: 895-896
Article info
Identification
Copyright
© 2002 Mayo Foundation for Medical Education and Research. Published by Elsevier Inc. All rights reserved.
