Abstract
TO THE EDITOR: The desire to overcome the many complexities of reaching adequate anticoagulation status with warfarin therapy continues to propel investigators to explore the perceived benefit of using a protocol for initiation of warfarin therapy. The recent article published by Roberts et al. 1 provides interesting details and concluded that elderly patients “rapidly achieved a stable INR [international normalized ratio] with minimal overanticoagulation”; however, a bit of clarification may be needed to substantiate such findings. The definition of overanticoagulation and the incidence of bleeding observed, the warfarin response beyond the 1-week study period, and the association of low albumin levels with an exaggerated warfarin response should be considered.
There is question regarding the number of patients outside of the defined stable INR range of 2–3 observed during the study period. Patients with an INR of 3.0 on day 3 who experienced a further increase in their INR to 3.5 on day 4 were included in the 63% of patients who attained a stable INR. In practice, however, an INR of 3.5 may be considered overanticoagulation and warrant some intervention. Also, it is unclear as to what group patients with an INR of 3.5–3.9 were included. Were they also considered as having a stable INR since overanticoagulation was defined as an INR ≥4?
There is no mention of the incidence of minor and major bleeding observed. This is of particular interest due to the possible association of age as a risk factor for bleeding in the elderly population. The study noted 4 of the 5 patients experiencing an INR ≥4 at completion of the loading dose protocol as being >65 years of age, further suggesting that the elderly appear to have a more exaggerated response to therapy.
Many practitioners in the US do not routinely follow a protocol in initiating warfarin therapy due to potential bleeding complications that have developed shortly beyond the designated initiation period. An understanding of the pharmacokinetic profile of warfarin enables clinicians to realize that the full effect of initiation doses may take 10–12 days to be seen due to the prolonged half-life of certain clotting factors being inhibited. To have a patient quickly reach a target INR within a 1-week period may seem desirable, but it is imperative that the INR is reevaluated 3–5 days later to ensure that overanticoagulation has not occurred. The concept of administering a loading dose of warfarin, as promoted by the authors, may be a potential danger in terms of overanticoagulation and is not commonly practiced by many clinicians.
Lastly, the association of low serum albumin levels with an exaggerated warfarin response may influence practitioners in their dosing decisions. 2 However, it should be noted that low serum albumin levels resulting in an increased free fraction of warfarin is accompanied by an increase in plasma clearance of the drug, with no effect on the INR, and is clinically irrelevant. 3