Superwarfarin poisoning: challenges still remain

Abstract

Superwarfarin (long-acting anticoagulant rodenticide) poisoning should be suspected in unexplained bleeding with prolonged prothrombin time, especially in the absence of another explanation. Diagnosis and treatment of this intoxication remain a challenge as the direct analysis of superwarfarin in serum is not always possible. Therefore, toxin bioavailability remains unknown and close monitoring and treatment for long periods are required to avoid serious bleeding complications. Here, we discuss a case of suspected superwarfarin poisoning to highlight the challenges in early diagnosis and the challenges we encountered in treatment management and ensuring compliance for long periods.

Background

Around 1929, Dam et al noticed that malnourished chickens were developing haemorrhagic diathesis that could be corrected by feeding the chickens a liposoluble substance named vitamin K, from the first letter of the Danish word ‘koagulation’.^{1 2} Around the same time, on the prairies in Canada and North America, cattle started dying from internal bleeding with no obvious cause, and the ‘sweet clover disease’ was first described.³ Some years later, Link et al identified and isolated the substance that prevented blood from clotting. It was an oxidised form of coumarin named dicoumarol, which acts as a competitive inhibitor of vitamin K epoxide reductase (VKORC1), an enzyme that recycles vitamin K, causing depletion of active vitamin K in the blood.^{2 4} Dicumarol, the first compound to be widely commercialised, was used as a rodenticide.^{5 6} Since then, many forms have been derived from dicumarol with various uses, such as the currently used anticoagulants.

For some years, warfarin and its synthetic analogues were generally known as the first anticoagulant rodenticides, and these drugs dominated the market worldwide. However, heritable warfarin resistance appeared in some species of rats, and better rodenticides were required to overcome this defence mechanism.⁷ This new class of drugs was called superwarfarins, which are long-acting anticoagulant rodenticides (LAARs) that have longer half-lives (from 20 to 209 days) and greater potency to induce coagulopathy than warfarin.^{8 9} In Spain, more than 400 compounds are registered as rodenticides, and most of them are primarily anticoagulants.⁸

It is known that intoxication by ingesting pesticides, including rodenticides, is a major contributor to the global burden of suicide.⁹ Additionally, other types of exposure that induce intoxication, such as transcutaneous poisoning,^{10 11} accidental inhalation¹² and synthetic cannabinoid use,¹³ have been described. According to the Annual Report of the American Association of Poison Control, in 2019, there were more than 3586 cases of poisoning due to LAARs.¹⁴

We present a case of a man who was admitted to the hospital with haemorrhagic diathesis. Initial laboratory data showed the prothrombin time beyond measurable limits in Spain.

Case presentation

A man in his 40s was admitted to the emergency department of a tertiary teaching hospital with mucosal bleeding and macroscopic haematuria 3 years ago. His medical history included Graves’ disease in 2012 that was treated with radioactive iodine and left euthyroid without a need for substitutive hormonal replacement. No other allergies, diseases or surgical interventions were reported.

The patient had normal vital signs; however, haemorrhagic diathesis was found on exploration. Furthermore, no internal organ bleeding was found. Investigations were performed, and treatment was established.

Investigations

The blood test showed a haemoglobin (Hb) level of 128 g/L and a platelet (Plt) count of 182×10⁹ /L. Initially, the prothrombin time (PT) was beyond the measurable limits; the first value was obtained after a correction of 4%, with an international normalised ratio (INR) of 15.8 and a partial thromboplastin time (PTT) of 76.1 s. Urinalysis revealed more than 500 red blood cells per high power field; however, a CT scan of the abdomen showed no inner organ bleeding. Hepatic and kidney parameters and thyroid functions were normal. Considering the evidence suggestive of coagulopathy, we further studied the activity of clotting factors (V, II, VII, IX and X). The activities of all the factors were below normal values except for factor V, which exhibited a normal activity of 92%.

With the evidence of deficiency of all vitamin-K-dependent factors and a normal factor V level, a normal liver function, and no evidence of family or personal history of factor deficiency, an acquired vitamin K antagonist-related bleeding event was assumed. Potential exposure sources were explored thereafter. The patient denied any known exposure to warfarin or rodenticide products or the use of synthetic cannabinoids. Quantitative analysis of superwarfarins in serum could not be performed at our hospital; furthermore, analysis facilities were unavailable in different toxicological centres we contacted around the country. Therefore, the source of the toxin exposure remained unknown.

A psychiatric evaluation was requested to examine evidence of suicidal thoughts or self-injurious behaviour. A major depressive disorder was diagnosed but without clear evidence of autolytic intentions.

Treatment

The patient was treated with 10 mg vitamin K1 intravenously every 8 hours, without prothrombin complex or fresh frozen plasma, since there was no life-threatening bleeding, and the patient was stable.

After the first 24 hours of treatment with vitamin K1, it was possible to reverse the coagulopathy and reach an INR of 1.87 with normal ranges of PT and PTT. However, when vitamin K1 was discontinued, the INR values exceeded 10 again. Therefore, vitamin K1 was reintroduced (figure 1).

Figure 1.

Data from 206 days of laboratory assessments for our patient. INR, international normalised ratio; IV, intravenous posology; PO, oral posology.

A switch to oral treatment was attempted when the INR was stable. Doses of 10 mg every 8 hours were administered. However, the INR increased again, necessitating the reintroduction of the intravenous treatment to re-establish the coagulation values. On day 21 of treatment, the patient was discharged with oral treatment at doses of 60 mg per day, and outpatient follow-up visits were effected. At that point, the activity of all coagulation factors was restored to normal ranges.

Outcome and follow-up

On the first control visit on day 32 of treatment, lab tests showed high INR despite good medication compliance by the patient. Therefore, the dose was progressively increased to 120 mg per day during the follow-up.

When the global pandemic of coronavirus disease 2019 emerged, the patient skipped a few follow-up control visits; therefore, we were unable to properly monitor the coagulation values. Therefore, on day 206 of treatment, we decided to terminate the treatment. However, we managed to contact the patient after 1 year and found that no other bleeding episodes had occurred. We assumed that a reversal of superwarfarin poisoning was achieved.

Discussion

Superwarfarin poisoning in adults has been related to self-harm attempts and accidental exposure.¹⁵ Although rodenticide intoxication is less common in Spain than in China or the USA,^{8 16} there are still some cases being reported, such as the presented one. Therefore, LAAR poisoning should be considered in the differential diagnosis of a patient presenting with signs and symptoms of coagulopathy related to vitamin K deficiency without a more plausible explanation, even if a clear exposure is not identified.⁷

Warfarin acts by inhibiting the activity of the C1 subunit of vitamin K epoxide reductase (VKORC1), thereby preventing the recycling of oxidised vitamin K, an essential cofactor for the four procoagulant vitamin K-dependent proteins (figure 2).^{6 17} The high potency of superwarfarins is reportedly related to the increased inhibition of VKORC1, extremely long biological half-lives and increased liver retention.^{6 18–20} As mammals cannot synthesise vitamin K de novo,¹⁷ VKORC1 inhibition leads to the rapid depletion of this vitamin from the body, and coagulopathy appears when the half-lives of vitamin K-dependent coagulation factors is reached, which range from 0.25 days for factor VII to 2.5 days for prothrombin.^{5 19}

Figure 2.

Mechanism of action of warfarin and superwarfarins. Vitamin K hydroquinone is oxidised to vitamin K epoxide by vitamin K-dependent γ-glutamyl carboxylase, which catalyses the post-translational carboxylation of specific glutamic acid residues to γ-carboxyglutamic acid in a variety of vitamin K-dependent proteins (VKP) including coagulation factors II, VII, IX and X, and protein S and C. Superwarfarins act similar to warfarin exerting their effect by inhibiting vitamin K 2,3-epoxide reductase complex subunit 1 (VKORC1). Inhibition of VKORC1 leads to reduced bioavailability of the metabolically active reduced form of vitamin K, resulting in decreased glutamyl carboxylation of VKP necessary for the final conformational structure of the proteins. The figure was prepared by the author (GM).

Difficulties in diagnosing LAARS intoxication persist as the quantitative analysis of superwarfarins in serum is not widely available during clinical practice.^{5 7} When it is not possible to confirm intoxication using serum detection, the diagnosis should be based on clinical suspicion and laboratory abnormalities such as a prolonged PT and PTT followed by the decreased activity of vitamin K-dependent factors (II, VII, IX and X) compared with the normal activity of factor V, which serves as a control.¹⁷ Immediate diagnosis and prompt treatment are critical as it is a life-treating disorder that is reversible with appropriate therapy.^{7 17}

On establishing a diagnosis of LAAR intoxication, if neither major bleeding nor repercussion on vital signs is found, the intoxication can be reversed using vitamin K1 by restoring inactive clotting factors until LAAR clearance is achieved.5 17 19 21–23

It is known that oral administration has similar efficacy and safety for INR reduction 24 hours after administration compared with intravenous administration.^{5 17 23} Nevertheless, intravenous administration is recommended when bleeding occurs due to its more rapid effect and reaching normal values of INR at 6–12 hours after administration.²⁴ The systemic bioavailability of oral vitamin K1 is approximately 50%,^{25 26} as 5 mg of oral vitamin K1 is equivalent to 1 mg of vitamin K1 when administered intravenously.^{23 27} We suggest that this equivalence could be used when switching to oral therapy, and therefore if 30 mg was needed to control INR intravenously, 150 mg of oral therapy should be used. This conversion matches with that used in our patient, who needed at least 120 mg to maintain INR.

To ensure consistent production of vitamin K-dependent factors, some authors suggest dividing daily doses every 6 hours, as the half-life of factor VII is 3–5 hours, and that of vitamin K1 is around 6 hours.^{19 24 25} However, daily dosing of vitamin K1 has been shown to correct coagulopathy effectively.6 17 18 20 21 23

The treatment regimens to manage LAAR intoxication may vary widely as the doses of vitamin K1 differ between patients, and dosing should be titrated based on coagulation assay results and signs of bleeding.5 17 25 26 It is apparent that, for maintenance treatment, the goal is to achieve the lowest dose possibly administered orally to maintain a PT in the normal range until the toxic effect is cleared.¹⁷

The case review of King et al²⁵ summarises the treatment data for 41 cases of LAAR poisoning. The doses ranged from 10–420 mg of vitamin K1, and the median treatment duration was 140 days. However, there is no clear consensus on the optimal dose, dosing frequency or appropriate time to stop the treatment as coagulation abnormalities persisted in many cases despite undetectable serum levels of LAAR. Additionally, there is no significant dose-effect relationship between the LAAR concentration and vitamin K1 requirements during the maintenance period.15 23 25 28 29

In most cases reported,^{23 25} the treatment during the maintenance phase is provided using the same dose until termination. Some authors suggest that smaller amounts of vitamin K1 could be administered during the maintenance treatment since the sustained stimulation of this compound could increase the expression of VKORC1 analogues that did not bind to the rodenticides.^{17 23} However, to the best of our knowledge, neither the timing of INR control nor adjustment doses related to PT results have been defined yet. In our case, we attempted to reduce the oral dose once; however, there was a subsequent increase in the INR values after stopping. Therefore, the same dose was maintained until treatment termination.

Finally, we intend to raise awareness regarding how challenging it could be to maintain appropriate therapy for longer periods. As LAARs have a long half-life, treatment duration typically spans 3–6 months.²³ Moreover, determining the appropriate time to discontinue vitamin K1 is critical to avoiding the recurrence of bleeding and rehospitalisation.³⁰ Several clinicians have recommended serial quantitative brodifacoum testing to draw an elimination curve to estimate the duration of vitamin K1 therapy.^{19 30 31} However, when these measurements cannot be performed, as in our case, an approach based on discontinuing vitamin K1 therapy followed by measuring coagulation parameters at 48–72 hours has been used to decide if treatment needs to be resumed.^{5 31 32} A safer alternative could be to measure factor VII, which has the shortest half-life (3–5 hours), and it needs to be decreased by more than 70% to alter the coagulation values.¹⁹

Learning points.

• Superwarfarinic poisoning should be considered in the differential diagnosis of a patient presenting with signs and symptoms of coagulopathy related to vitamin K deficiency without a likely explanation, even if clear exposure has not been defined.

• Treatment needs to be empirically determined in each patient, and vitamin K1 dosing should be titrated according to the coagulation values (international normalised ratio).

• Clinicians should switch the patient to oral therapy as soon as possible while taking into account that oral consumption of a 5 mg dose is equivalent to administering 1 mg intravenously.

• Follow-up should be continued for long periods to reassess the requirements of vitamin K1 periodically. Different strategies exist to decide the optimal time to terminate treatment.

Ethics statements

Patient consent for publication

Consent obtained directly from patient(s).