CYP2C9*3 polymorphism presenting as lethal subdural hematoma with low-dose warfarin.

Warfarin is the most common and cheap oral anticoagulant currently used in clinical practice. A high inter-individual variation is seen in the response to warfarin. Recently, pharmacogenetics has gained importance in managing patients on warfarin, both in predicting the optimum required dose as well as in decreasing the risk of bleeding. This case report is a description of a 49-year-old patient who had a lethal subdural hematoma with low-dose warfarin. He was subsequently found to have CYP2C9 gene polymorphism (*1/*3). This case report stresses the importance of pre-prescription assessment of genetic analysis for those initiated on warfarin.

Introduction

Warfarin is the most commonly used oral anticoagulant both for prophylaxis as well as treatment for arterial and venous thromboembolic conditions. It has a narrow therapeutic index and is metabolized by hepatic cytochrome P450 2C9 enzyme (CYP2C9). Various polymorphisms exist in the gene that code for CYP2C9, of which *1 is the wild form expressing normal activity. Amongst the variants, *2 and *3 are more common than the rest and exhibit lesser enzymatic activity.[1] The anticoagulant effect of warfarin is measured by prothrombin time-international normalized ratio (PT-INR). Warfarin exerts its anticoagulant effect by reducing the regeneration of active vitamin K through inhibition of vitamin K epoxide reductase, encoded by vitamin K epoxide reductase complex subunit 1 gene (VKORC1). Increased risk of deranged INR or bleeding has been reported in patients harboring mutant form of either CYP2C9 or VKORC1. This case report describes a patient on low-dose warfarin, who succumbed to a subdural hematoma and was found to have polymorphic CYP2C9 (*1/*3).

Case Report

A 49-year-old male patient, who was a known case of superior sagittal sinus thrombosis for the past 8 years on tablet warfarin 2.5 mg every night, was brought to the hospital unconscious. There was no history of any head injury, headache, fever or vomiting. The patient was non-compliant in not following up with his physician nor did he check his PT-INR levels regularly for the past 1 year. The patient's relative, however, maintained that the patient was regular in taking the prescribed dose of warfarin (2.5 mg) every night. On physical examination, he was found to be unconscious (Glasgow Coma Scale score of 3), pulse was feeble and the blood pressure was not recordable. Pupils were 6 mm dilated with no reaction to the light. Immediate resuscitative measures including invasive ventilation and inotropic support were given in the ICU. A plain computed tomography (CT) brain revealed 24.2 mm hyperdense cresentic subdural hematoma in left fronto-temporo-parieto-occipital region with subfalcine extension and midline shift [Figure 1]. His PT-INR was found to be 9 [normal (in the absence of anticoagulation therapy) is 1 and it has to be maintained between 2 and 3 for an appropriate therapeutic effect for warfarin]. Fresh frozen plasma was initiated at a dose of 15 ml/kg/day. Blood (5 ml) was collected from the patient and genotyping was done for both CYP2C9 and VKORC1. Genomic DNA was extracted by phenol-chloroform method. CYP2C9 was found to be polymorphic for *3 (*1/*3) and VKORC1 was found to be wild-type (normal). The patient succumbed on 5 day of admission.

Figure 1

Plain CT brain revealing 24.2 mm hyperdense cresentic subdural hematoma (red arrow) in left fronto-temporo-parieto-occipital region

Discussion

Pharmacogenetics is the study of genetic influence on inter-individual variation in the response of a drug.[2] Pharmacogenetics has been shown to be of value in detecting an optimal response as well as in preventing adverse reactions following drugs such as warfarin, clopidogrel, and phenytoin. Warfarin is metabolized by CYP2C9 and over 30 polymorphic alleles have been identified in the gene coding for CYP2C9. The wild form of the gene is *1 and the most common polymorphic forms include *2 and *3. Allelic frequency of CYP2C9*2 and *3 in various populations including Indians were found to be in the range of 0-20%.[3] A recent systematic review concluded that in comparison to the patients with CYP2C9*1/*1 genotype, the CYP2C9*1/*2, CYP2C9*1/*3, CYP2C9*2/*2, CYP2C9*2/*3, and CYP2C9*3/*3 patients required warfarin doses that were 19.6, 33.7, 36.0, 56.7, and 78.1% lower, respectively.[4] Similarly, in the VKORC1 gene, two alleles have been identified in VKORC1 namely G and A. Patients with GG (wild type) are resistant requiring a higher dose of the drug whereas AA are sensitive requiring a lesser dose and GA requires intermediate. A randomized clinical trial has also shown that pharmacogenetic-based dosing of warfarin was associated in maintaining the PT-INR levels within the acceptable range for a longer time than by the standard dosing.[5] United States Food and Drug Administration (FDA) has recently modified the warfarin labeling by suggesting CYP2C9 and VKORC1 genotyping before initiating the drug to ensure warfarin safety.[6]

Our patient had a fatal subdural hematoma despite low-dose (2.5 mg HS) of warfarin. He had massively deranged INR of 9, was brought in with Glasgow Coma Scale of 3. The *1/*3 variant CYP2C9 explained this serious adverse event with low-dose warfarin. The *3 variant is less than 5% as active as the wild type whereas *2 retains around 12% of the activity. A recent Indian review of oral anticoagulant use revealed lack of proper laboratory facilities, with irregular PT-INR monitoring in 25% patients and deranged INR values in a high proportion of patients.[7] Pharmacogenetic testing needs to be done only once and hence can be considered cost-effective. Studies have shown that along with age, gender, body weight, CYP2C9 and VKORC1 polymorphisms account for almost 64% of the variability in the dose of warfarin required for an individual.[8] Hence, by doing these genetic testing a priori to the treatment initiation, there is a high possibility that the patients may receive an appropriate dose than just by routine practice. Even, several international algorithms for predicting dose of warfarin have been developed with these predicting factors of which, Gage's and Wadelius’ algorithms have been shown to be the most accurate.[9] Recently, an algorithm for Indian patients has also been developed, which was found to be more accurate, sensitive and significantly reduced the risk of overestimation of dose of warfarin.[10] This algorithm can be used for predicting the appropriate dose of warfarin in our population with the genotype details of CYP2C9 and VKORC1. To conclude, this case highlights the necessity of adopting the practice of routine genetic testing for CYP2C9 and VKORC1 while initiating warfarin therapy.