Investigation of Elevated INR

Initial Assessment Framework

The American College of Cardiology and other guideline societies recommend that when investigating an elevated INR, immediately assess for bleeding and determine whether the patient is on warfarin or vitamin K antagonist (VKA) therapy, as the INR is only validated for monitoring VKA therapy and should not be used as a general coagulopathy screen in non-VKA patients 1, 2
The INR was specifically designed and validated only for patients receiving warfarin or other vitamin K antagonists, and in non-VKA patients, an elevated INR lacks biological plausibility as a bleeding predictor and should prompt investigation for underlying liver disease, vitamin K deficiency, or other coagulopathies 1, 2, 3

The American Geriatrics Society recommends checking for overt bleeding (gastrointestinal, intracranial, genitourinary, or other sites) and looking for occult bleeding signs: unexplained anemia, hemodynamic instability, or new neurological deficits 4, 5
The American Geriatrics Society also recommends evaluating for medication non-adherence or accidental overdosing, as well as assessing for new botanicals or herbal supplements, which lack standardization and can unpredictably affect INR 4, 5

The European Heart Journal recommends that advanced age (>65 years) be considered a patient-specific bleeding risk factor, in addition to history of previous gastrointestinal bleeding or stroke, and recent myocardial infarction or severe anemia 5, 6
The American Geriatrics Society recommends that INR-specific bleeding risk be assessed, with INR <5.0 considered a modest elevation with low bleeding risk if no other risk factors are present, INR 5.0-9.0 considered intermediate risk requiring intervention based on patient bleeding risk, and INR >9.0 considered high bleeding risk 4, 5
The European Heart Journal also recommends that INR >4.5 be considered a point at which bleeding risk begins to rise, with INR >6.0 rising steeply and exponentially 7

The Journal of Thrombosis and Haemostasis recommends recognizing that an elevated INR without VKA therapy represents a different clinical scenario requiring investigation for underlying pathology, and checking comprehensive liver function tests and synthetic function (albumin, bilirubin, transaminases) 1, 2, 3
The Journal of Thrombosis and Haemostasis also recommends against using plasma transfusion to "correct" mildly elevated INR in non-bleeding patients, as there is no evidence of benefit and it exposes patients to transfusion risks 2, 3

The Journal of Thrombosis and Haemostasis recommends never assuming INR predicts bleeding risk in non-VKA patients, as it has poor sensitivity for bleeding disorders and provides false reassurance 1, 3
The European Heart Journal recommends avoiding plasma transfusion for INR <2.0, as it does not change INR values below 1.5 and only minimally impacts values below 2.0 3, 7