Anticoagulation Intensity and Stroke During Warfarin-Use in Atrial Fibrillation Patients
To the Editor:
The interesting study by Poli and colleagues examined risk factors for ischemic stroke during anticoagulation of atrial fibrillation patients.1 For the 21 patients with ischemic events, it would be of interest to know the mean of the last “routine” INR before the ischemic event and the mean of the INR recorded nearest to/at the time of stroke to compare with the means of INRs during total follow-up of these 21 patients and of the mean INR for 343 patients without ischemic events. These data may shed additional light on the potential contribution of anticoagulation intensity to ischemic events in this setting.
Poli D, Antonucci E, Cecchi E, Marcucci R, Liotta AA, Cellai AP, Lenti M, Gensini GF, Abbate R, Prisco D. Culprit factors for the failure of well-conducted warfarin therapy to prevent ischemic events in patients with atrial fibrillation. The role of homocysteine. Stroke. 2005; 36: 2159–2163.
We thank Dr Hart for his interest in our study and for his questions aiming at clarifying our data. The mean INR of the 343 patients without ischemic events was 2.5±0.19, and the mean INR of the 21 patients who experienced an ischemic event during follow-up was 2.5±0.16, a difference not statistically significant. The mean INR related to the ischemic event (obtained at the time of the event or during the preceding 8 days) was 2.1±0.58, statistically lower in comparison to both the mean INR of this group and the mean INR of all the other patients (P=0.002). Fifteen patients had the INR related to the event ≥1.8. The last mean routine INR recorded before the event was 2.34±0.56, not different with respect to both the mean INR value in this group and the mean INR of all the other patients (P=0.1).
To evaluate if the occurrence of ischemic events could be related to a long period of under-anticoagulation, we have calculated the mean INR of the 3 months preceding the event. It was 2.5±0.7, a figure not different from the mean INR of all the patients.
In our opinion, these data confirm that the reduced INR related to the event is only one of the risk factors for the occurrence of ischemic complications during OAT. In fact, 15 out of 21 patients had an INR ≥1.8 at the time of the event. In addition, patients without ischemic events during OAT had been exposed to the risk of a poor anticoagulation for a time similar to patients with ischemic events. Actually, the time spent below the intended therapeutic range was not different between the 2 groups. Oral anticoagulant therapy has an intrinsic variability of response so that even an optimized management cannot avoid significant periods of time with INR <2.
The use of this treatment on a growing population makes this goal more and more difficult to obtain, further underscoring the need for a specifically devoted management. Together with the continuous efforts to ameliorate the quality of anticoagulation, the identification of other risk factors could help to reduce adverse events in nonvalvular atrial fibrillation patients.