Superwarfarin Poisoning

ORIGINAL INVESTIGATION Superwarfarin Poisoning

Jimmy Dy Chua, MD; William R. Friedenberg, MD

Background: Superwarfarin sodium exposure or poi- accidentally ingested superwarfarin or attempted sui- soning is a growing public health problem. There were 5133 cide using it were easily diagnosed, while diagnosis was reported cases of superwarfarin exposure and poisoning markedly delayed for the 2 patients with Munchausen in 1988 and 13 423 cases in 1995. Cases may be associ- syndrome. Full reversal of anticoagulation was quickly ated with accidental exposure, suicide attempts, or Mun- achieved in the cases of accidental ingestion and at- chausen syndrome, and may be difficult to diagnose. tempted suicide. We examined and treated the patients with Munchausen syndrome for months before estab- Patients and Methods: Patients from northern Wis- lishing a diagnosis and fully reversing the anticoagula- consin with superwarfarin exposure or poisoning were tion. None of the patients in our study died of superwar- examined at a tertiary referral center in rural Wisconsin farin poisoning. to determine what led to their exposure and to review the clinical manifestations, diagnosis, treatment, and pre- Conclusions: Superwarfarin exposure or poisoning is vention of superwarfarin poisoning. a growing public health problem that should be part of the differential diagnosis of patients who present with a Results: Eleven cases satisfied the criteria for superwar- coagulopathy consistent with vitamin K deficiency in the farin exposure or poisoning. All 7 children included in absence of coumadin therapy, liver disease, or the use of the study had accidentally ingested superwarfarin, 2 adults an inhibitor, and whose conditions do not resolve with had Munchausen syndrome, and 1 teenager and 1 adult large doses of parenteral vitamin K1 therapy. had attempted suicide using superwarfarin. Nine of the 11 cases had taken brodifacoum. The patients who had Arch Intern Med. 1998;158:1929-1932

UPERWARFARINS constitute a tamin K epoxide,3 the inactive form of class of rodenticides devel- vitamin K (Figure 1). Although the events oped in the 1970s to over- that lead to superwarfarin exposure or poi- come resistance to warfarin in soning are diverse, the cause is usually rats.1,2 They are popular and straightforward on admission. Some- readilyS available in stores and homes. Su- times the diagnosis of superwarfarin poi- perwarfarins are long-acting, fat-soluble soning can be challenging. Multiple ad- anticoagulants with a terminal half-life of missions and evaluations6-9 are reported for at least 24.2 days and are 100 times more this disorder when it is a manifestation of potent than warfarin.3 Superwarfarins and Munchausen syndrome. warfarins concentrate in the liver on in- There have been no epidemiological gestion.4 There are 3 major groups of su- studies of patients with superwarfarin ex- perwarfarins: hydroxy coumarin deriva- posure or poisoning. We reviewed the re- tives with a 4-bromo(1-1 biphenyl) side cords of 11 patients found to have super- chain; coumatetryls; and indanediones. warfarin poisoning over 15 years. We also The most commonly used superwarfarin reviewed the literature on the causes, clini- is brodifacoum, a 4-hydroxy coumarin cal manifestations, diagnosis, treatment, and with a 4-bromo side chain sold as D-Con prevention of brodifacoum poisoning. (Reckitt & Colman, Montvale, NJ.)5 Su- perwarfarins inhibit the carboxylation of RESULTS From the Marshfield Clinic, vitamin K–dependent factors (II, VII, IX, Marshfield, Wis. Dr Chua is and X) in the liver. They specifically in- Eleven cases (7 children, 1 teenager, and now with the Cleveland Clinic hibit the vitamin K 2,3-epoxide reduc- 3 adults) satisfied the criteria of super- Foundation, Cleveland, Ohio. tase enzyme, leading to an increase in vi- warfarin exposure or poisoning. Seven pa-

ARCH INTERN MED/ VOL 158, SEP 28, 1998 1929

Downloaded From: https://jamanetwork.com/ on 09/25/2021 Vitamin K Metabolism SUBJECTS AND METHODS Epoxidase

We reviewed all cases (128 charts) at Marshfield Clinic, Marshfield, Wis, with International Statistical Classi- fication of Diseases, 10th Revision (ICD-10)10 codes 989.4 (toxic effects of other pesticides), including E codes E863.7 (accidental poisoning–rodenticide), E950.6 (suicide-agricultural agent), E962.1 (assault Vitamin K Liver Microsomes poisoning with solid/liquid), and E980.7 (undeter- Vitamin K mined poisoning with agricultural agent), and code Epoxide 286.7 (acquired coagulation factor deficiency) from January 1, 1979, to March 31, 1996. Exposure was defined as any history of oral intake or inhalation of superwarfarin, or superficial con- tact with a superwarfarin near the oral mucosa with- out any biochemical or clinical manifestations of coagulopathy. Reductase (Superwarfarin Blocks) Poisoning was defined by exposure to superwarfarin plus clinical and biochemical changes consis- Figure 1. Vitamin K is converted to vitamin K epoxide (inactive) in the liver. tent with coagulopathy, including an elevated pro- Brodifacoum (a superwarfarin) interferes with vitamin K metabolism by thrombin time (PT), an elevated activated partial inhibiting the 2,3-epoxide reductase enzyme, which increases levels of vitamin K epoxide (inactive vitamin K) and inhibits the synthesis of active thromboplastin time (APTT), an increase in the vi- factors II, VII, IX, and X. tamin K epoxide–vitamin K1 ratio, or depressed levels of vitamin K–dependent coagulation factors (II, VII, IX, and X) with the presence of superwarfarin in tient’s plasma and normal plasma at a ratio of 1:1 cor- the blood. The definitions of exposure and poison- rected the PT to normal. The patient’s plasma was negative ing were established before the epidemiological data for warfarin. His vitamin K epoxide level (following high- were collected. dose vitamin K1 therapy) was 50.2 ng/mL (reference range, Ͻ0.05 ng/mL) and his phylloquinone level (reduced vitamin K) was 9.5 ng/mL (reference range, 0.1-1.0 ng/ mL). The results of his initial screening for superwarfa- tients were male and 4 were female. There were 3 major rins were negative. Subsequently, a specific assay for brodi- categories of exposure or poisoning: accidental inges- facoum revealed a concentration of 78 ng/mL. After the tion (in children), attempted suicide, and deliberate self- patient was confronted with this finding and a possible poisoning with denial (Munchausen syndrome). police investigation, he admitted to taking a handful of All the children in our study were individuals brodifacoum (Talon). The patient received large doses of younger than 6 years who had accidentally ingested parenteral vitamin K1 in addition to fresh frozen plasma superwarfarin. None of the children had documented to reverse his coagulopathy as soon as the initial coagu- elevated PTs, clinical signs of coagulopathy, or compli- lation levels were reported. He was discharged after 11 days cations. One child received 5 mg of vitamin K subcuta- in the hospital, requiring 5 months of outpatient fol- neously, and 5 children received syrup of ipecac. Test re- low-up by a home health nurse and 20 weekly doses of sults showed no PT abnormalities in any of the children subcutaneous vitamin K1 (410 mg) before the coagulopa- on follow-up more than 48 hours after treatment. There thy resolved. Although he had been instructed to take oral was no morbidity or mortality reported in this group over vitamin K1, it was discovered that he never had the pre- many years of follow-up. scription filled. The patient underwent psychiatric evalu- The second group, patients who attempted suicide, ation but refused follow-up visits. included 1 teenager and 1 adult. The teenager presented The second patient, a 50-year-old woman, was ad- with some bruising and a slightly elevated PT with an in- mitted because of easy bruising and a subungual hema- ternational normalized ratio of 1.3. She was sent home and toma. Laboratory studies yielded the following values: PT, found to be normal on follow-up 3 days later with no clini- 39.6 seconds; APTT, 108.2 seconds; factor VII, 7%; and cal signs of coagulopathy. The man who had attempted factor IX, less than 2%. Mixing studies performed as de- suicide had an initial PT of 11.7 seconds and no coagu- scribed earlier corrected the PT to normal. The patient’s lopathy during his 4-day stay. The final outcome of this plasma was negative for warfarin; however, she was found patient was unknown since he was transferred to another to have a brodifacoum level of 310 ng/mL. As a result, she institution on his fourth day at the hospital. was given large doses of parenteral vitamin K1 as well as The third group of patients included those with Mun- fresh-frozen plasma to reverse her coagulopathy, which chausen syndrome. The first patient was a 44-year-old di- resolved after 11 months of outpatient treatment. The treat- vorced man who was admitted with multiple ecchymosis ment included prolonged intake of oral vitamin K and nu- and gross hematuria. Laboratory studies yielded the fol- merous emergency department and clinic visits for fresh- lowing values: PT, greater than 90 seconds; APTT, greater frozen plasma transfusions and parenteral vitamin K1. The than 160 seconds; international normalized ratio, 60.8; fac- patient was suspected of continuing to ingest brodi- tor VII, 1%; and factor X, 4%. Mixing studies with the pa- facoum, but she refused psychiatric referral.

ARCH INTERN MED/ VOL 158, SEP 28, 1998 1930

Downloaded From: https://jamanetwork.com/ on 09/25/2021 18,30,34 14 000 orrhage. Sixteen of the 24 reported cases presented with hematuria. Ecchymosis was present in 6 of

12 000 the 24 cases. Workup of brodifacoum poisoning is not a problem in cases of acute poisoning but may be challenging in patients with suspected Munchausen syn- 10 000 drome. We advise physicians confronted with cases of brodifacoum poisoning to determine the PT of the pa- 8000 tient on initial exposure and at 48 hours.35 Any prolon- gation in PT indicates that poisoning has occurred. We 6000 No. of Cases also suggest that for patients with possible Munchausen syndrome, a mixing study of the patient’s plasma with 4000 control plasma should be performed. Normalization of the PT and APTT after mixing studies implies the pres- 2000 ence of factor deficiency and the absence of an inhibitor. Furthermore, we advise physicians to obtain samples 0 of a few of the vitamin K–dependent factors (II, VII, 1988 1989 1990 1991 19921993 1994 Year IX, and X) and factor V, which is produced in the liver but is not vitamin K dependent. Although not usually Figure 2. The incidence of brodifacoum (superwarfarin) exposures and necessary, one can determine the vitamin K epoxide– poisoning from 1988 to 1994.11-17 1 reduced vitamin K ratio. A deficiency of vitamin K– related factors plus a very high vitamin K1 epoxide– COMMENT reduced vitamin K1 ratio is consistent with warfarin or 36,37 superwarfarin poisoning. Vitamin K1 epoxide is a prod- Superwarfarin exposure or poisoning is a growing prob- uct of vitamin K metabolism and is the inactive form of lem. Data on exposure or poisoning with long-acting ro- vitamin K (Figure 1). The presence of warfarin or su- denticides from the American Association of Poison Con- perwarfarin in the blood confirms warfarin or superwar- trol Centers from 1988 to 199411-17 point to a gradual farin exposure or poisoning. Usually none is detected. If increase (Figure 2) in cases of superwarfarin exposure Munchausen syndrome is suspected and the patient’s and poisoning. In 1995, 13 423 individuals were ex- blood is negative for warfarin, a specific assay for brodi- posed and treated in 5051 facilities that serve a total popu- facoum (Talon) and other superwarfarins should be per- lation of 218.5 million people.18 formed. In our first case, levels of all the vitamin K– Accidental exposure or poisoning usually occurs in dependent factors were low and the factor V level was children. About 90% of the exposures reported by the normal, which is consistent with a lack of vitamin K1 or American Association of Poison Control Centers in 1995 inhibition of its activity by warfarin or superwarfarin. Af- occurred in individuals younger than 6 years.18 In our ter transient resolution, repeated and continuous treat- study, 7 of 11 cases of exposure were in children younger ment with large doses of vitamin K1 was necessary to cor- than 6 years. Fortunately, there was no notable morbid- rect the coagulopathy. The vitamin K1 epoxide–reduced ity or mortality in these patients. When confronted with vitamin K ratio confirmed the presence of a warfarin- patients with accidental exposure or ingestion, physi- type compound while assays for warfarin were being per- cians should follow guidelines recommended by a poi- formed. Results of the original screening test for super- son center.5 warfarin were reportedly negative, but brodifacoum was Brodifacoum poisoning in adults is less frequent. Poi- later detected at a concentration of 78 ng/mL. soning may be due to suicide attempts,19,20 industrial ex- The treatment of brodifacoum poisoning in adults 21 22 posure, psychiatric conditions, accidental inges- is supportive. We believe vitamin K1 should be intrave- tion,15 or Munchausen syndrome.6 Brodifacoum poisoning nously administered to patients with severe coagulopa- is the most common type of superwarfarin poisoning, yet thy. The total amount of vitamin K1 should be diluted only 24 cases of brodifacoum poisoning have been re- with saline or glucose and infused at a rate not exceed- ported in adults. There have been 11 cases of Munchau- ing 5% of the total dose per minute. This treatment can sen syndrome,6-9,23-27 6 attempted suicides,19,20,28-31 2 in- be administered every 6 to 8 hours.38 Intramuscular ad- dustrial exposures,21 3 cases of unknown cause,18,32 1 case ministration should be reserved for adults with mild cases of accidental ingestion,15 and 1 case involving a psychi- of poisoning by ingestion, while subcutaneous admin- atric disorder.22 Poisoning in the adults and the teen- istration may be used for maintenance therapy. We be- ager in our study resulted from 2 suicide attempts and lieve treatment for severe bleeding should include fresh- 2 cases of Munchausen syndrome. Thus far, there have frozen plasma and red blood cell transfusion as necessary been no reports that superwarfarin has been used to to quickly correct coagulopathy. The role of phenobar- commit murder or that the drug has been dispensed bital in the treatment of brodifacoum poisoning has yet erroneously.33 to be defined. There are some case reports19 of its en- Clinical manifestations of brodifacoum poisoning in- hancement of hepatic microsomal enzymes. The dose of clude vaginal bleeding,9,22 epistaxis,7,27,30,34 hematu- phenobarbital is 100 to 180 mg/d.5 ria,7,8,19,26-34 bleeding from the gums,20,25,31 gastrointesti- Treatment for superwarfarin poisoning requires long- nal bleeding,8,30-32 ecchymosis,6,19,34 spontaneous abortion,22 term outpatient care. Coagulopathy may persist for hemoptysis,23,25 abdominal pain,20 and intracranial hem- months, and some patients may die if discharged too

ARCH INTERN MED/ VOL 158, SEP 28, 1998 1931

Downloaded From: https://jamanetwork.com/ on 09/25/2021 early.29,34 Adequate compliance is crucial to achieving com- 9. Routh CR, Triplett DA, Murphy MJ, Felice LJ, Sadowski JA, Bovill EGT. Super- plete reversal of anticoagulation. One of our patients with warfarin ingestion and detection. Am J Hematol. 1991;36:50-54. 10. World Health Organization. International Statistical Classification of Diseases, 10th Munchausen syndrome required continued parenteral Revision (ICD-10). Geneva, Switzerland: World Health Organization; 1992. therapy because he never filled his prescription for oral 11. Litovitz TL, Schmitz BF, Holm KC. 1988 Annual report of the American Associa- vitamin K1. Although psychiatric evaluation may be in- tion of Poison Control Centers National Data Collection System. Am J Emerg Med. dicated, patients with Munchausen syndrome rarely com- 1989;7:495-545. ply with therapy. 12. Litovitz TL, Schmitz BF, Bailey KM. 1989 Annual report of the American Asso- ciation of Poison Control Centers National Data Collection System. Am J Emerg Most patients with superwarfarin exposure or poi- Med. 1990;8:394-442. soning do not have major complications. Of the 79 025 13. Litovitz TL, Bailey KM, Schmitz BF, Holm KC, Klein-Schwartz W. 1990 Annual cases of exposure or poisoning reported over a period of report of the American Association of Poison Control Centers National Data Col- 8 years,11-18 there were 8 reported deaths and only 67 pa- lection System. Am J Emerg Med. 1991;9:461-509. tients (0.08%) with major residual effects (residual dis- 14. Litovitz TL, Holm KC, Bailey KM, Schmitz BF. 1991 Annual report of the Ameri- can Association of Poison Control Centers National Data Collection System. Am ability or disfigurement). However, among 24 patients J Emerg Med. 1992;10:452-505. with brodifacoum poisoning reported in the literature, 15. Litovitz TL, Holm KC, Clancy C, Schmitz BF, Clark LR, Oderda GM. 1992 Annual 6 died (1 committed suicide after being discharged).29 Re- report of the American Association of Poison Control Centers Toxic Exposure ported causes of death include sepsis in 1 case,9 intrace- Surveillance System. Am J Emerg Med. 1993;11:494-555. rebral bleeding in 2 cases,30,34 and events related to the 16. Litovitz TL, Clark LR, Soloway RA. 1993 Annual report of the American Asso- 15,18 ciation of Poison Control Centers Toxic Exposure Surveillance System. Am J Emerg central nervous system in 2 cases. Med. 1994;12:546-584. 17. Litovitz TL, Felberg L, Soloway RA, Ford M, Geller R. 1994 Annual report of the CONCLUSIONS American Association of Poison Control Centers Toxic Exposure Surveillance Sys- tem. Am J Emerg Med. 1995;13:551-597. Superwarfarin exposure or poisoning is a growing pub- 18. Litovitz TL, Felberg L, White S, Klein-Schwartz W. 1995 Annual report of the Ameri- can Association of Poison Control Centers Toxic Exposure Surveillance System. lic health problem in the United States affecting those of Am J Emerg Med. 1996;14:487-537. all ages, but primarily children. The diagnosis of super- 19. Jones EC, Growe GH, Naiman SC. Prolonged anticoagulation in rat poisoning. warfarin poisoning is usually straightforward in cases of JAMA. 1984;252:3005-3007. acute poisoning or exposure but may be difficult in pa- 20. Chong L-L, Chau W-K, Ho C-H. A case of “superwarfarin” poisoning. Scand J tients with Munchausen syndrome. Patients with a Haematol. 1986;36:314-315. 21. Park BK, Choonara IA, Haynes BP, Breckenridge AM, Malia RG, Preston FE. Ab- coagulopathy consistent with vitamin K1 deficiency that normal vitamin K metabolism in the presence of normal clotting factor activity persists or recurs despite a history of adequate replace- in factory workers exposed to 4-hydroxycoumarins. Br J Clin Pharmacol. 1986; ment should be evaluated for warfarin or superwarfarin 21:289-294. poisoning. 22. Lipton RA, Klass EM. Human ingestion of a “superwarfarin” rodenticide result- ing in a prolonged anticoagulant effect. JAMA. 1984;252:3004-3005. 23. Barnett VT, Bergmann F, Humphrey H, Chediak J. Diffuse alveolar hemorrhage Accepted for publication February 5, 1998. secondary to superwarfarin ingestion. Chest. 1992;102:1301-1302. We wish to thank John W. Suttie, PhD, from the Depart- 24. Ross GS, Zacharski LR, Robert D, Rabin DL. An acquired hemorrhagic disorder ment of Biochemistry, University of Wisconsin, Madison, from long-acting rodenticide ingestion. Arch Intern Med. 1992;152:410-412. for performing the vitamin K ratio tests, and Alice Stargardt 25. Exner DV, Brien WF, Murphy MJ. Superwarfarin ingestion. CMAJ. 1992;146: 34-35. for her expertise in the preparation of the manuscript. We 26. Hollinger BR, Pastoor TP. Case management and plasma half-life in a case of also appreciate the advice and care given to the patients with brodifacoum poisoning. Arch Intern Med. 1993;153:1925-1928. Munchausen syndrome by Dale Sternberg, MS, and Shelly 27. Stanton T, Sowray P, McWaters D, Mount M. Prolonged anticoagulation with Stanton, MD. long-acting coumadin derivatives: case report of a brodifacoum poisoning with Corresponding author: William R. Friedenberg, MD, pharmacokinetic data. Blood. 1988;72(suppl):310A. 28. Hoffman RS, Smilkstein MJ, Goldfrank LR. Evaluation of coagulation factor Department of Hematology/Oncology, Guthrie Clinic, Guth- abnormalities in long-acting anticoagulant overdose. Clin Toxicol. 1988;26: rie Square, Sayre, PA 18840. 233-248. 29. Helmuth RA, McCloskey DW, Doedens DJ, Hawley DA. Fatal ingestion of a brodifacoum-containing rodenticide. Lab Med. 1989:25-27. REFERENCES 30. Kruse JA, Carlson RW. Fatal rodenticide poisoning with brodifacoum. Ann Emerg Med. 1992;21:331-336. 1. Hadler MR, Shadbolt RS. Novel 4-hydroxycoumarin anticoagulants active against 31. Sheen SR, Spiller HA, Grossman D. Symptomatic brodifacoum ingestion requir- resistant rats. Nature. 1975;253:275-277. ing high-dose phytonadione therapy. Vet Hum Toxicol. 1994;36:216-217. 2. Lund M. Comparative effect of the three rodenticides warfarin, difenacoum and 32. Chen TW, Deng JF. A brodifacoum intoxication case of mouthful amount. Vet brodifacoum on eight rodent species in short feeding periods. J Hyg Camb. 1981; Hum Toxicol. 1986;28:488. 87:101-107. 33. O’Reilly RA, Aggeler PM. Covert anticoagulant ingestion: study of 25 patients 3. Park BK, Leck JB. A comparison of vitamin K antagonism by warfarin, difenacoum and review of world literature. Medicine (Baltimore). 1976;55:389-399. and brodifacoum in the rabbit. Biochem Pharmacol. 1982;31:3635-3639. 34. Basehore LM, Mowry JM. Death following ingestion of superwarfarin rodenti- 4. O’Bryan SM, Constable DJC. Quantification of brodifacoum in plasma and liver cide: a case report. Vet Hum Toxicol. 1987;29:459. tissue by HPLC. J Anal Toxicol. 1991;15:144-147. 35. Smolinske SC, Scherger DL, Kearns PS, Wruk KM, Kulig KW, Rumack BH. 5. Rumack BH, Hess AJ, Gelman CR, eds. POISINDEX System. Englewood, Colo: Superwarfarin poisoning in children: a prospective study. Pediatrics. 1989;84: Micromedex Inc; 1996:89. 490-494. 6. Swigar ME, Clemow LP, Saidi P, Kim HC. “Superwarfarin” ingestion: a new prob- 36. Sadowski JA, Suttie JW. Mechanism of action of coumarins: significance of vi- lem in covert anticoagulant overdose. Gen Hosp Psychiatry. 1990;12:309-312. tamin K epoxide. Biochemistry. 1974;13:3696-3699. 7. Weitzel JN, Sadowski JA, Furie BC, et al. Surreptitious ingestion of a long-acting 37. Whitlon DS, Sadowski JA, Suttie JW. Mechanism of coumarin action: signifi- vitamin K antagonist/rodenticide, brodifacoum: clinical and metabolic studies of cance of vitamin K epoxide reductase inhibition. Biochemistry. 1978;17:1371- three cases. Blood. 1990;76:2555-2559. 1377. 8. Wallace S, Worsnop C, Paull P, Mashford ML. Covert self poisoning with brodi- 38. de la Rubia J, Grau E, Montserrat I, Zuazu I, Paya´ A. Anaphylactic shock and vi-

facoum, a “superwarfarin.” AustNZJMed. 1990;20:713-715. tamin K1. Ann Intern Med. 1989;110:943.

ARCH INTERN MED/ VOL 158, SEP 28, 1998 1932

Downloaded From: https://jamanetwork.com/ on 09/25/2021