Hypoprothrombinemia

The following report illustrates a case of trimethoprim/sulfamethoxazole-induced hypoprothrombinemia in a patient receiving ongoing warfarin therapy for atrial fibrillation and aortic valve replacement. He was treated with trimethoprim/sulfamethoxazole (TMP/SMX) for sinusitis. During this time, the patient's prothrombin time international normalized ratio (INR) increased 3.5 times higher than the baseline value. The INR values decreased when the antibiotic was discontinued. If a patient is on warfarin and TMP/SMX is added, INR values should be monitored closely.

Key words. Hypoprothrombinemias; sulfonamides; warfarin; trimethoprim/sulfamethoxazole combination.

(J Fam Pract 1994; 39:589-591) Warfarin and trimethoprim/sulfamethoxazole (TMP/SMX) are two commonly prescribed medications. Warfarin remains the most frequently used oral anticoagulant for the treatment of thromboembolic disorders. Likewise, the antimicrobial agent TMP/SMX has significant clinical utility in the treatment of a variety of disorders, including urinary tract infections, otitis media, chronic bronchitis, shigellosis, travelers' diarrhea, and Pneumocystis carinii pneumonia. It is also well known that warfarin interacts with a large number of medications, including sulfonamides. This interaction could result in an exaggerated hypoprothrombinemic response and subsequent bleeding.

The following report describes a case of a patient who was stabilized on warfarin for atrial fibrillation and aortic valve replacement. He subsequently experienced significant hypoprothrombinemia with associated bruising following the concomitant use of TMP/SMX therapy for sinusitis.

Case Report

A 70-year-old man presented to the hospital on February 28, 1994, with a chief complaint of inability to breathe, which had progressively worsened over the 2 days before admission. He was admitted with a diagnosis of chronic obstructive pulmonary disease exacerbation and possible viral syndrome. His medications before admission included potassium chloride tablets, 30 mEq twice daily; furosemide tablets, 80 mg every morning, 40 mg every night; albuterol inhaler, 2 puffs as needed; ipratropium bromide inhaler, 2 puffs four times daily; beclomethasone diproprionate inhaler, 2 puffs every day; ferrous sulfate tablets, 300 mg daily; and warfarin sodium 6 mg daily. Warfarin had been prescribed for an aortic valve replacement and a history of atrial fibrillation. Before admission, the patient had been stabilized on the 6-mg dose (international normalized ratio(1) [INR] at admission, 2.53).

On the 3rd hospital day, the patient complained of hoarseness and pharyngitis. A nasopharyngoscopy showed evidence of purulent drainage of the right sinus and sinusitis. Oral TMP/SMX, one double-strength tablet twice daily, was begun at noon on the same day. On the 5th hospital day, the patient's INR(1) was elevated to 8.76, following 3 full days of TMP/SMX therapy. On the same day, a large bruise appeared on his abdomen. TMP/SMX and warfarin were both eliminated from the patient's therapy on this date, and cefuroxime sodium 1.5 g given intravenously every 8 hours was begun for the sinusitis. Two days later (March 7), oral doxycycline hyclate 100 mg every 12 hours was added to his antimicrobial regimen.

Upon discontinuation of TMP/SMX and warfarin, the patient's prothrombin times (PTs) and INRs began to fall (Table).(*)

Table. Decreasing Prothrombin Times and International Normalized Ratios i

n a Patient in Whom Trimethoprim/Sulfamethoxazole and Warfarin Were Discontinu ed

On March 8, 3 days after the patient stopped taking warfarin and TMP/SMX, he received an isolated 6-mg dose of warfarin at 6 PM. He did not receive any more warfarin until the evening of March 11, when his INR decreased to 3.30. The patient was discharged the following day with a prescription of 6 mg warfarin daily, which was the dosage he had been taking before admission.

Discussion

The interaction between warfarin and TMP/SMX is well documented, with several different mechanisms suggested for its occurrence. One of the mechanisms thought to be responsible is sulfonamide impairment of the hepatic metabolism of warfarin.(2)(3) Because long-acting sulfonamides and warfarin are both highly bound to plasma proteins (primarily albumin), another mechanism may be competition between these agents for protein-binding sites, potentially leading to warfarin displacement and elevation of serum warfarin levels, which could enhance the hypoprothrombinemic effect.(4) Several case reports and letters have been published documenting hypoprothrombinemia induced by the concomitant use of warfarin and TMP/SMX,(2)(3)(5)(6)(7)(8)(9) although most predate 1980. Since it has been almost 15 years since a case report or letter concerning this interaction has been published in the medical or pharmaceutical literature, we present this case to provide current documentation of the interaction between warfarin and TMP/SMX, which continue to be commonly used agents.

The available literature regarding the onset of the hypoprothrombinemia caused by the warfarin and TMP/SMX interaction varies widely. The published cases cite onsets ranging from 1 day of combined therapy, after which a patient's PT rose from 23 to 32 seconds,(6) to 13 days of combined therapy, after which a patient experienced significantly elevated PT and subsequent nosebleeds.(5) In a study conducted by Hassall et al,(7) approximately 20 patients receiving warfarin and TMP/SMX in combination were identified and observed, with PT being recorded before, during, and after concurrent therapy. Results indicated definite hypoprothrombinemia in 6 of the 20 patients, and in each of the 6, the PT rose between the 2nd and 6th day of combined therapy. The results from the published literature are consistent with those observed in the current case. After 3 days of combined therapy with warfarin and TMP/SMX, our patient's PT and INR elevated markedly from the PT and INR recorded at admission.

Documented adverse effects in patients experiencing this interaction range from gross hematuria and severe bilateral flank pain after 3 days of combined therapy to gum bleeding, hematuria, and hematemesis occurring in another patient after 2 weeks of combined therapy.(5)(8) Our patient's morbidity from the interaction manifested as a large abdominal bruise, which was first noted on March 5 and was present throughout the remainder of his hospital stay. The bruise was noted to have increased in size on March 6, which correlates with the date that his INR was the highest (9)(14). It began resolving on March 8, the same day that his PT showed its first significant decrease since the discontinuation of warfarin and TMP/SMX on March 5.

An intriguing aspect of the interaction between warfarin and TMP/SMX is determining the most appropriate way to expedite resolution of the hypoprothrombinemia. In two instances involving patients who received both warfarin and TMP/SMX, PTs remained significantly elevated for 5 days following discontinuation of the warfarin alone.(6)(9) In one patient, when TMP/SMX was discontinued subsequent to warfarin discontinuation, prothrombin time decreased and the hypoprothrombinemia resolved over the next 4 days.(6) In the other case, PTs fell rapidly after intramuscular administration of 10 mg phytonadione.(9) Based on the observation that discontinuation of warfarin alone did not resolve the hypoprothrombinemia, it appears that the appropriate intervention for this interaction is to discontinue the TMP/SMX.(6) In our patient's case, both agents were discontinued until his PT and INR decreased to 16 seconds and 2.0, respectively, which closely approximated his baseline values at admission.

In the current case, the patient's INR decreased over 3 days following discontinuation of warfarin and TMP/SMX. After an isolated warfarin dose was administered, however, his INR remained elevated for the following 2 days (Figure). It is not known whether the INR value of 4.69 for 3 days is attributable to the isolated warfarin dose. A similar scenario was noted in a patient who, after not receiving warfarin or TMP/SMX for 3 days, experienced a sharp increase in PT upon receipt of two consecutive daily warfarin doses.(9) These occurrences support the concept that warfarin and TMP/SMX compete for protein-binding sites, resulting in sulfonamide displacement, a decrease in total sulfonamide plasma concentration, and an increase in its unbound fraction. The concentration of unbound sulfonamide is also increased in tissue.(10) Sulfonamide potentiation of anticoagulants can be observed for 2 weeks or more after discontinuation, possibly because sulfonamide diffuses from the plasma into tissues without subsequent metabolism or excretion from the body.(11)

[CHART OMITTED]

Because of the potential morbidity associated with this interaction, as well as the availability of other agents that possess a spectrum of activity similar to that of TMP/SMX but lack the potential to interact with warfarin, it is recommended that the combination of warfarin and TMP/SMX be avoided. The INR should be closely monitored for pronounced hypoprothrombinemia in any patient taking these medications concomitantly. If the patient displays exaggerated hypoprothrombinemia or bleeding, both medications should be discontinued. Once the INR normalizes, chronic warfarin therapy may be safely reinstituted.

References

(1.)Vanscoy GJ, Krause JR. Warfarin and the international normalized ratio: reducing interlaboratory effects. DICP Ann Pharmacother 1991; 25:1190--2.

(2.)O'Reilly RA, Motley CH. Racemic warfarin and trimethoprim-sulfamethoxazole interaction in humans. Ann Intern Med 1979; 91:34--6.

(3.)O'Reilly RA. Stereoselective interaction of trimethoprim-sulfamethoxazole with the separated enantiomorphs of racemic warfarin in man. N Engl J Med 1980; 302:33--5.

(4.)Deykin D. Warfarin therapy. N Engl J Med 1970; 283:801--3.

(5.)Kaufman JM, Fauver HE. Potentiation of warfarin by trimethoprim-sulfamethoxazole. Urology 1980; 16:601--3.

(6.)Greenlaw CW. Drug interaction between co-trimoxazole and warfarin. Am J Hosp Pharm 1979; 36:1155--6.

(7.)Hassall C, Feetam CL, Leach RH, et al. Potentiation of warfarin by co-trimoxazole. Lancet 1975; 2:1155--6.

(8.)Tilstone WJ, Nimmo-Smith RH, Gray JMB, et al. Interaction between warfarin and sulphamethoxazole. Postgrad Med J 1977; 53: 388--90.

(9.)Errick JK, Keys PW. Co-trimoxazole and warfarin: case report of an interaction. Am J Hosp Pharm 1978; 35:1399--1400.

(10.)Anton AH. A drug-induced change in the distribution and renal excretion of sulfonamides. J Pharmacol Exp Ther 1961; 134:291--303.

(11.)Martin EW. Hazards of medication. 1 st ed. Philadelphia, Pa: JB Lippincott, 1971.

(*)Internationalized normalized ratio (INR) has succeeded prothrombin time (PT) internationally as an indicator to standardize control of anticoagulant therapy. The author has used both terms in this paper in accordance with the form of the data reported in the sources cited.

From the Pharmacy Practice Residency Program, West Virginia University Hospitals (D.E.C.), and the Departments of Clinical Pharmacy and Family Medicine, The Robert C. Byrd Health Sciences Center, West Virginia University (C.D.P), Morgantown, WV 26506. Requests for reprints should be addressed to Donna Cook, RPh, Department of Pharmacy, Health Sciences Center, SC-69, University of Washington, Seattle, WA 98195.

COPYRIGHT 1994 Dowden Health Media, Inc.
COPYRIGHT 2004 Gale Group