Introduction: Liver disease is often marked by changes in hemostasis. Vitamin K is frequently administered to cirrhotic patients with an elevated INR to improve their coagulopathy, though strong evidence justifying this approach is lacking. Questions regarding the efficacy of vitamin K have been gathering based on an increased understanding of the rebalanced hemostasis of liver disease. This study evaluated the effect of vitamin K on the INR 24-72 hours after administration.

Methods: This retrospective chart review used the VA Informatics and Computing Infrastructure (VINCI) database to identify 886 admissions for patients with liver disease who received vitamin K between January 1, 2001 and March 31, 2014. Patients were included if they had a coded diagnosis of cirrhosis, acute hepatitis, non-alcoholic steatohepatitis, hepatocellular carcinoma, or end stage liver disease. Charts for patients with one of those diagnoses who received vitamin K at the LSCDVAMC were included.

All data was collected from the Computerized Patient Record System (CPRS). Medication route and dosing was determined from the pharmacy administration record. Patients were excluded if they received heparin, LMWH, FFP, or if they did not have an INR value before the administration of vitamin K or 24-72 hours after the dose was given.

Results: A total 886 individual admissions were identified, 333 admissions met inclusion criteria for analysis. The mean INR on admission for the included encounters was 1.88 (95% CI 1.798 - 1.955) the lowest INR was 0.86 and the highest was 5.99. In the 333 admissions analyzed the mean decrease in the INR was 0.08 (95% CI 0.028 - 0.132). 180 encounters had a repeat INR during the hospitalization. The mean decrease in INR from admission to the second post-vitamin K INR was 0.123 (95% CI 0.058 - 0.187). Of the 333 included patient encounters 37 had a change in INR (increase or decrease) >0.4, of which in 11 the INR increased and in 24 the INR decreased (mean change in INR 0.313; 95% CI -0.139 to 0.765). The average INR of those 37 encounters was 2.83 (95% CI 2.497 - 3.171). There was no significant difference in albumin in encounters when the INR increased vs. decreased in response to vitamin K. There was a trend towards higher total bilirubin (TBILI) when the INR did not decrease in response to vitamin K. Mean TBILI was 5.9 in INR responders (95% CI 5.011 - 6.789) and was 6.66 (95% CI 5.219 - 8.101) in the encounters where the INR increased despite vitamin K.

Conclusion: Vitamin K administration to improve the coagulopathy of patients with liver disease is common and often administered in response to an elevated INR. This is the largest retrospective review to date evaluating the effect of vitamin K on the INR of patients with liver disease. While a statistically significant decrease in INR of 0.08 was found, it is unclear if such a difference from vitamin K would be clinically significant. Moreover, only a very small portion of the patients included in this study (24/333) had an INR decrease of greater than 0.4. A limitation of our study is that patient centered outcomes, such as bleeding events, were not assessed. Future research should evaluate if there is a role for vitamin K in liver disease patients with a significantly elevated INR and low total bilirubin.


No relevant conflicts of interest to declare.

Author notes


Asterisk with author names denotes non-ASH members.

This icon denotes a clinically relevant abstract