PEDIATRICS Vol. 105 No. 2 February 2000, pp. 402-404

Unintentional Pediatric Superwarfarin Exposures: Do We Really Need a Prothrombin Time?

Michael E. Mullins, MD, Christina L. Brands, BSN, CSPI, and Mohamud R. Daya, MD

From the Oregon Poison Center, Department of Emergency Medicine, Oregon Health Sciences Center, Portland, Oregon.

	ABSTRACT

Top Abstract Methods Results Discussion Conclusion References

Objective.  To determine whether routine follow-up coagulation studies are useful in children with accidental exposures to rodenticides containing superwarfarin compounds.

Design.  Retrospective review of poison center charts involving pediatric superwarfarin exposures occurring in two 2-year periods.

Setting.  An American Association of Poison Control Centers-certified regional poison control center with an annual call volume of 55 000 calls per year from a 2-state area with a combined population of 4 million people.

Outcome Measures.  Prothrombin times and/or international normalized ratios and reported clinical signs of excessive anticoagulation after exposure.

Results.  Of 542 children in 4 years of data collection, follow-up prothrombin times and/or international normalized ratios measurements did not detect any significant coagulation abnormalities. No child developed bleeding complications. No child required or received antidotal treatment with vitamin K.

Conclusion.  Normal preschool-aged children with unintentional acute exposures to superwarfarin rodenticides do not require any routine follow-up laboratory studies and do not require any medical intervention.  Key words:  poisoning, children, anticoagulant, rodenticide.

In 1997, the American Association of Poison Control Centers (AAPCC) recorded 13 405 human exposures to long-acting anticoagulant rodenticides, commonly known as superwarfarins. Of these, 12 005 (89.6%) involved children under 6 years of age.¹ This figure underestimates the actual number of exposures. Some children may stay at home with no call by the care provider to the nearest poison center; others may present to a pediatrician's office or to the emergency department (ED) with no call to the poison center.

In 1989, Smolinske et al² reviewed 110 pediatric exposures reported to a poison center. They concluded that all superwarfarin rodenticide (SWR) exposures required follow-up prothrombin times (PT) at 24 and 48 hours after exposure. POISINDEX (MICROMEDEX, Englewood, CO) subsequently incorporated this recommendation despite the fact that neither this study nor any other studies demonstrated that this approach detected any clinically significant coagulopathy or reduced morbidity and mortality in this setting.

	METHODS

Top Abstract Methods Results Discussion Conclusion References

We reviewed poison center charts for 398 consecutive pediatric SWR exposures in a 2-year period (January 1993 to December 1994) during which our poison center recommended follow-up PT measurements at 24 and 48 hours. We then reviewed poison center charts for 144 consecutive cases for a 2-year period (January 1996 to December 1997) commencing 1 year after our local protocol changed to recommend only a single PT measurement at 48 hours. Registered nurses trained and certified as specialists in poison information made follow-up calls to parents, their pediatricians, or their medical laboratories to obtain information on subsequent measurement of PT and/or international normalized ratios (INR) and to ask about the presence of unusual bleeding or bruising.

	RESULTS

Top Abstract Methods Results Discussion Conclusion References

In the first 2-year review including 1993 and 1994, the mean 24-hour PT was 12.5 seconds (range: 10.7-14.9) and the mean 48-hour PT was 12.1 seconds (range: 9.9-18.8). Only 1 child had a PT >15 seconds at any time. The PT of this child was 18.8 seconds at 48 hours and declined to 14.0 seconds at 72 hours without therapy. No child required or received vitamin K. No child had any apparent bleeding complications.

In the second 2-year review including 143 cases from 1996 and 1997, the poison center recommended follow-up PT in 82 cases, of which 80 had documented follow-up. The actual PT value was available in 58 charts, described only as normal in 8 charts, performed but not documented in 3 charts, and not performed in 8 charts. Of the 58 cases with known PT values, the mean follow-up PT was 12.1 seconds (range: 10.4-13.9). No child had a PT >14 seconds. No child required or received vitamin K. No child had any apparent bleeding complications.

	DISCUSSION

Top Abstract Methods Results Discussion Conclusion References

Several case reports of serious pediatric SWR poisonings exist. All these reports describe hospitalized children with abnormal PT values and serious bleeding complications. In each case, the history of SWR exposure was unknown at the time of admission and was not the presenting chief complaint. All had unusual circumstances surrounding the exposure. Greeff et al³ described 2 SWR-poisoned children, 1 with a neck hematoma requiring intubation for airway control and 1 with profuse epistaxis and hemarthrosis. Both stayed with the same babysitter who had 2 tenants on her property; 1 tenant used pink pellets of bromadiolone around the dwelling, and the other tenant sold a confection closely resembling the rat poison.

Watts et al⁴ described a chronic brodifacoum intoxication in a child with several admissions for intracranial bleed after minor head trauma, a calf hematoma with compartment syndrome, epistaxis, hematemesis, and extremity bleeding. Even after removal of brodifacoum pellets from the home, her bleeding complications continued. A subsequent social service visit found an unkempt home with rat feces on the floor. The analysis of the rat feces revealed the presence of brodifacoum.

Travis et al⁵ described a 36-month-old girl with excessive bruising and bleeding from the nose and mouth. Subsequent laboratory analysis detected brodifacoum in the blood. Although the mother thought that child did not have access to the rat poison in her home, the child did have a known history of pica. Her mother was noncompliant with outpatient vitamin K therapy, and the child required foster care placement to assure completion of her therapy.

Babcock et al⁶ presented a 24-month-old child who presented with unexplained bruising and a PT of >125 seconds. The child was the victim of Munchausen syndrome by proxy. Once confronted with detection of brodifacoum in the blood and a container of brodifacoum found in his hospital room, the mother admitted to sprinkling brodifacoum on his cereal for the 4 days before admission and the first 10 days in the hospital.

None of these few bizarre incidents involved primary contact with a poison control center. The situation of an unintentional pediatric SWR exposure reported to a poison center or to the primary care provider is much different. In 1989, Sullivan et al⁷ reported on 88 SWR ingestions followed by the Pittsburgh Poison Center. All follow-up PT values at 24 and 72 hours were normal, and no patient developed any untoward event. The authors attributed these results to early gastric emptying, usually with syrup of ipecac.

The review by Smolinske² of 110 cases at the Rocky Mountain Poison Center actually provides little credible evidence supporting the perceived need for serial PT testing in this population. Of 110 children with SWR ingestions followed with PT ratios at 24 and 48 hours, the authors detected only 8 children with any PT ratio 1.20 times control. The highest ratio in this series occurred in a 12-month-old child whose PT ratios were .97 at 24 hours, 1.44 at 48 hours, and .95 at 72 hours. None of the children had any clinical evidence of abnormal bleeding, and none received vitamin K.

Smolinske et al did note in an addendum that 6 months after the end of their study, they consulted on the case of one 2-year-old child who ingested brodifacoum. Serial PT values were 10.5 seconds (ratio: 1.00) at 24 hours and 21.5 seconds (ratio: 2.05) at 48 hours. The child had no clinical evidence of bleeding but received 3 mg of vitamin K1 intramuscularly. Follow-up PTs were normal up to 12 days later. They did not give information on the total number of pediatric superwarfarin exposures that occurred in that same 6-month period.

In the same year, Katona and Wason⁸ reviewed the issue of SWR poisoning and argued that although repeated or intentional ingestions required medical attention, unintentional pediatric exposures rarely, if ever, had any bleeding complications. Unfortunately, despite this well-reasoned conclusion, the Micromedex incorporated the recommendation of the Denver group for serial PT determinations at 24 and 48 hours in POISINDEX. That recommendation remains in place a decade later.

The Washington State Poison Center reported on 1789 warfarin and superwarfarin (82%) unintentional exposures.⁹ Approximately half received syrup of ipecac, yet none had any bleeding complications regardless of whether they had gastric emptying. A subset of 138 patients with health department home visits revealed only 1 occurrence of a minimal elevation of PT in a child who remained asymptomatic. The authors highlighted their conclusion by pointing out that the AAPCC had recorded over 72 000 warfarin and SWR exposures in children <6 years of age from 1987 through 1993, yet no death occurred in this population. Additional review of the AAPCC reports for 1985-1986 and 1994-1997 bring the 13-year total to nearly 136 101 recorded anticoagulant rodenticide exposures in this age group with no pediatric deaths.^110-12

Shepard et al¹³ more closely reviewed the AAPCC Toxic Exposure Surveillance System data for brodifacoum from 1993-1996. Of 23 481 brodifacoum exposures in children under 6 years of age, less than half (10 733) received poison center follow-up to document the outcome. Few of these had either minor (2.88%) or moderate (.57%) effects. Coagulation abnormalities were rare (abnormal PT: .36%; bleeding: .15%). Ninety children received prophylactic vitamin K despite the fact that 71 of these children had no effects from the brodifacoum. Half of the children underwent gastrointestinal decontamination with no difference in outcomes.

Even in the rare instance of mild, transient changes in coagulation laboratory studies, the risk of a serious or life-threatening bleeding event is most likely exceedingly low. The only available epidemiologic evidence comes from several recent studies of adults chronically taking warfarin.^14-20 Various long-term studies have detected rates of major hemorrhage or intracranial bleeding between 1.1 and 7 per 100 patient-years. Rates of fatal bleeding events were between .9 and 2.1 per 100 patient-years. Complication rates are highest in the first 3 months of warfarin therapy.^19,20 Gitter et al²⁰ observed a 1.6% incidence of major hemorrhage in the first month of warfarin therapy in a retrospective cohort study of 241 adults. In a randomized, double-blinded, placebo control of 3404 patients taking warfarin, 14 intracranial bleeds occurred. All 14 of these patients had an INR >3.0.¹⁵ Generally, complications tended to occur in older patients with significant co-morbidities.

In a young healthy child who ingests a SWR, the .36% incidence of abnormal PT multiplied by up to 1.6% incidence of major hemorrhage in the first month of anticoagulation²⁰ yields a crude estimate of only a .006% risk of serious hemorrhage or a 6 in 100 000 chance. Given that the child with the highest PT described by Smolinske only had a PT that was twice the control value and that the PT remained normal at least 12 days after a single dose of vitamin K,² the risk of a serious bleeding event occurring in the rare child with a transient, minimal increase in PT/INR approaches zero.

In our institution, the charge for a level 2 pediatric office visit is $64.80 for an established patient. The local charge for a level 2 ED visit is $172.70, which includes a facility charge of $91.30 and a professional fee of $81.40. Charges for outpatient laboratory testing for PT/INR include $10.60 for the test and $15.10 for phlebotomy for a total charge of $25.70. In 1997, 4515 (34%) SWR-exposed patients received treatment in a health care facility.¹ Among patients evaluated and managed in health care facilities after poisoning exposures that were reported to AAPCC in 1997, 85.9% went to a hospital ED, and 11.8% went to an outpatient clinic.¹

We can crudely estimate the annual charges for US patients using charge data from our hospital and assuming compliance with the Smolinske guidelines in medically evaluated patients (Table 1). Because poison center data often fail to capture a large proportion of exposed or poisoned patients,^21,22 this estimate of nearly 1 million dollars vastly understates the actual charges. This rough estimate also does not include costs of unnecessary gastric decontamination, additional laboratory tests, prophylactic administration of vitamin K, hospital admission for observation, etc.

	CONCLUSION

Top Abstract Methods Results Discussion Conclusion References

Preschool-aged children with single acute unintentional superwarfarin exposures do not develop any significant laboratory or clinical evidence of excessive anticoagulation. Current guidelines for medical evaluation and repeated laboratory testing in these patients result in unnecessary expenditure probably well in excess of 1 million dollars annually in the United States. Poison centers and pediatricians should abandon routine evaluation and testing. Parents should receive advice to seek medical attention only in the unlikely occurrence of unusual bleeding or bruising. Only intentional ingestions, ingestions involving suspected child abuse or neglect, and ingestions by children with abnormal neurological or psychological development warrant close follow-up and laboratory monitoring.

	FOOTNOTES

This abstract was presented at Society for Academic Emergency Medicine Western Regional Research Forum; March 6, 1999; Redondo Beach, CA.

Received for publication Feb 2, 1999; accepted Jun 18, 1999.

Reprint requests to (M.E.M.) Department of Emergency Medicine, Washington University School of Medicine, Campus Box 8072, 660 S Euclid Ave, St Louis, MO 63110-8072. E-mail: mullinsm{at}msnotes.wustl.edu

	ABBREVIATIONS

AAPCC, American Association of Poison Control Centers; ED, emergency department; SWR, superwarfarin rodenticide; PT, prothrombin times; INR, international normalized ratios.

	REFERENCES

Top Abstract Methods Results Discussion Conclusion References

1. Litovitz TL, Klein-Schwartz W, Dyer KS, 1997 annual report of the American Association of Poison Control Centers Toxic Exposure Surveillance System. Am J Emerg Med 1998; 16:443-497 [CrossRef][Medline]
2. Smolinske SC, Scherger DL, Kearns PS, Superwarfarin poisoning in children: a prospective study. Pediatrics 1989; 84:490-494 [Abstract/Free Full Text]
3. Greeff MC, Mashile O, MacDougall LG "Superwarfarin" (bromodialone) poisoning in two children resulting in prolonged anticoagulation. Lancet 1987; 2:1269 [Medline]
4. Watts RG, Castleberry RP, Sadowski JA Accidental poisoning with a superwarfarin compound (brodifacoum) in a child. Pediatrics 1990; 86:883-887 [Abstract/Free Full Text]
5. Travis SF, Warfield W, Greenbaum BH, Spontaneous hemorrhage associated with accidental brodifacoum poisoning in a child. J Pediatr 1993; 122:982-984 [Medline]
6. Babcock J, Hartman K, Pedersen, et al Rodenticide-induced coagulopathy in a young child: a case of Munchausen syndrome by proxy. Am J Pediatr Hematol Oncol 1993; 15:126-130 [Medline]
7. Sullivan MP, Dean BS, Krenzelok EP Long-acting anticoagulant rodenticides: an evaluation of 88 cases. Vet Hum Toxicol 1989; 31:361
8. Katona B, Wason S Superwarfarin poisoning. J Emerg Med 1989; 7:627-631 [CrossRef][Medline]
9. Morrisey B, Burgess JL, Robertson WO Washington's experience and recommendations re: anticoagulant rodenticides. Vet Hum Toxicol 1995; 37:362-363 [Medline]
10. Litovitz TL, Smilkstein MJ, Felberg M, 1996 annual report of the American Association of Poison Control Centers Toxic Exposure Surveillance System. Am J Emerg Med 1997; 15:447-500 [CrossRef][Medline]
11. Litovitz TL, Felberg L, White S, 1995 annual report of the American Association of Poison Control Centers Toxic Exposure Surveillance System. Am J Emerg Med 1996; 14:487-537 [CrossRef][Medline]
12. Litovitz TL, Felberg L, Soloway RA, 1994 annual report of the American Association of Poison Control Centers Toxic Exposure Surveillance System. Am J Emerg Med 1995; 13:551-597 [CrossRef][Medline]
13. Shepard G, Klein-Schwartz W, Anderson B Acute pediatric brodifacoum ingestions. J Toxicol Clin Toxicol 1998; 36:464
14. Wandell P A five-year follow-up of 115 patients treated with anticoagulants: bleeding complications may be underestimated. Lakartidningen 1998; 95:3673-3674 [Medline]
15. Azar AJ, Koudstaal PJ, Wintzen AR, Risk of stroke during long-term anticoagulant therapy in patients after myocardial infarction. Ann Neurol 1996; 39:301-307 [CrossRef][Medline]
16. McMahan DA, Smith DM, Carey MA, Zhou XH Risk of major hemorrhage for outpatients treated with warfarin. J Gen Intern Med 1998; 13:311-316 [CrossRef][Medline]
17. Podesser BK, Khuenl-Brady G, Eigenbauer E, Long-term results of heart valve replacement with the Edwards Duromedics bileaflet prosthesis: a prospective ten-year clinical follow-up. J Thorac Cardiovasc Surg 1998; 115:1121-1129 [Abstract/Free Full Text]
18. Steffensen FH, Kristensen K, Ejlersen E, Major haemorrhagic complications during oral anticoagulant therapy in a Danish population-based cohort. J Intern Med 1997; 242:497-503 [Medline]
19. Arom KV, Emery RW, Nicoloff DM, Petersen RJ Anticoagulant related complications in elderly patients with St. Jude mechanical valve prostheses. J Heart Valve Dis 1996; 5:505-510 [Medline]
20. Gitter MJ, Jaeger TM, Petterson TM, Bleeding and thromboembolism during anticoagulant therapy: a population-based study in Rochester, Minnesota. Mayo Clin Proc 1995; 70:806-808 [Medline]
21. Crouch BI, Cook L, Caravati EM, Probabilistic linking of poison center data to other health care databases. J Toxicol Clin Toxicol 1998; 36:500
22. Hoppe J, Lloyd L, Chyka P Poison center-reported mortality data inconsistent with national trends. J Toxicol Clin Toxicol 1998; 36:500