When should vitamin K be given to correct INR?
Vitamin K rapidly reduces raised INR values in patients treated with warfarin but does not affect the risk of bleeding.
INR values greater than 4 are associated with an increase in bleeding complications and the risk of bleeding, in particular intracranial, markedly increases with an INR>4.5.
A total of 724 patients treated with warfarin, with INR values of 4.5 to 10 but no bleeding, were studied in a multicentre trial in anticoagulant clinics in Canada, the US and Italy.
The patients were randomised to oral vitamin K, 1.25 mg, or placebo.
The primary endpoint was all forms of bleeding events occurring during the first 90 days.
Results were analysed for 347 patients in the vitamin K group and 365 in the placebo group.The results showed that vitamin K was associated with a rapid reduction in INR. The day after treatment there was a mean reduction of 2.8 in the vitamin K group compared with 1.4 in the placebo group (P<0.001). However there was no significant difference in the primary endpoint between the two groups, bleeding occurred in 56 patients (15.8%) in the vitamin K group compared with 60 (16.3%) in the placebo group (P=0.86).
We are all aware of the unpredictable dose-response characteristics of warfarin and the current lack of a proven alternative. There are also increasing numbers of patients likely to require warfarin such as those with atrial fibrillation.
Therefore the late afternoon calls from the pathology laboratory to alert us to a high INR are likely to become increasingly commonplace. The traditional response to this has been to advise the patient to stop warfarin and if the patient is either bleeding or felt to be at high risk of a bleed then vitamin K should be considered (orally or intravenously).
This study would support the notion that stopping warfarin should be sufficient in patients with a high INR and that vitamin K although improving the time to INR recovery makes no difference to the risk of bleeding.
However, it is possible that because of the lower than expected bleeding rate in this study there were not enough events to show a benefit or it may be that the INR does not correlate well with bleeding risk when rapid changes are occurring such as during administration of vitamin K.
I think I will stick with the BNF advice and only consider vitamin K when there is a major bleed or the patient has an INR>8 with other risk factors for bleeding.
Crowther M, Ageno W, Garcia D et al. Oral vitamin K versus placebo to correct excessive anticoagulation in patients receiving warfarin. A randomised trial. Ann Intern Med 2009; 150:293-300Reviewer
Dr Peter Savill
GPwSI Cardiology, Southampton