Bee envenomation: a rare cause of thrombotic thrombocytopenic purpura. (Case Report).

Abstract

Thrombotic thrombocytopenic purpura (TTP) is characterized by the pentad of microangiopathic hemolytic anemia, thrombocytopenia, neurologic symptoms, renal insufficiency, and fever. Although many underlying diseases and conditions are associated with TTP, there has been only one previously documented case of bee-sting-associated TTP. In this article, we describe the case of a woman who developed life-threatening TTP temporally related to a honeybee sting and who required prolonged plasma exchange before her condition improved. We also briefly review evolving concepts in the pathogenesis of TTP and speculate on the etiopathogenicity of TTP and bee envenomation.

Key Words: bee sting, microangiopathic hemolytic anemia, plasma exchange, thrombotic thrombocytopenic purpura, von Willebrand factor 

**********

Thrombotic thrombocytopenic purpura (TTP) is characterlized by the pentad of microangiopathic hemolytic anemia (MAHA), thrombocytopenia, central nervous system symptoms, renal dysfunction, and fever. Most patients present with the dyad of thrombocytopenia and MAHA alone. (1) Various insults can lead to the development of TTP, including drugs, toxins, infectious agents, malignancy, immune dysregulation, and other factors affecting platelet aggregation or endothelial function. (2-7) Frequently, however, the specific cause of TTP is never firmly established. The understanding of the pathophysiology underlying TTP is evolving. Although vascular endothelial injury is still implicated as a possible inciting event, there is accumulating evidence that abnormal von Willebrand factor (vWF) multimers are central to the pathogenesis of TTP. (7,8) In this article, we detail the clinical course of a woman who developed TTP after a bee sting. This represents only the second reported instance of such an association. (9) We a lso briefly review evolving concepts in the pathogenesis of TTP and the possible link between insect envenomation and microangiopathic processes.

Discussion

On the basis of postmortem examinations, patients with TTP have characteristic hyaline arteriole and capillary lesions characterized by a damaged endothelium, platelet remnants, little to no fibrinogen or fibrin, the presence of vWF, and no associated perivascular inflammation. (5,10) These platelet microthrombi occur in the arterioles and capillaries of the brain, kidney, pancreas, heart, and adrenal glands. This leads to end-organ ischemia, intravascular hemolysis as red cells traverse occluded vessels, and the classic clinical symptomatology of TTP. (5)

Endothelial cells normally synthesize and polymerize vWF into large multimers. (11,12) These large multimers do not normally travel through the circulation, but rather are cleaved by a protease into smaller multimers. (7,8,13,14) This proteolytic activity has been recently shown to be deficient in cases of TTP. The absence of proteolytic activity can be because of a constitutional deficiency of the enzyme seen in familial inherited forms of TTP, or because of immunoglobulin G antibody interference with enzymatic function seen in cases of idiopathic TTP. (13-16) By interfering with vWF protease activity, these immunoglobulin G molecules prevent the breakdown of unusually large vWF into smaller multimers. This leads to elevated serum levels of unusually large vWF. (14) Because unusually large vWF is more effective than normal vWF at binding platelets, enhanced platelet aggregation can occur in areas subject to high shear stress, such as the cerebral and renal vasculature. (8) Such a cascade of events could expl ain the clinical manifestations and postmortem pathologic findings that are seen in patients with TTP. It could also explain why patients with nonfamilial TTP benefit from plasma exchange therapy, either because of the removal of vWF protease autoantibodies or because of infusion of vWF protease with FFP.

Our patient developed TTP shortly after a bee stung her. Honeybees (Apis mellifera) are the cause of most bee stings. (17) Honeybee venom is composed of amines, peptides, and enzymatically active proteins. These compounds have varied activities, including cell lysis, mast cell degranulation, and the induction of inflammation. Moreover, most of the higher molecular weight compounds have allergenic activity. (17,18) Although most envenomations are characterized by a transient local reaction of erythema, edema, and pain, anaphylaxis and other unusual manifestations can occur. Vasculitis, neuritis, and encephalitis have also been reported several days to weeks after a bee sting. (19) How a bee sting could cause TTP remains speculative, although we offer two hypotheses. Bee envenomation can cause diffuse endothelial damage, thus promoting hyperaggregation of platelets and alterations in vWF synthesis or function. Second, a bee sting could induce allergen sensitization to venom proteins, resulting in the productio n of autoantibodies that could cross-react with vWF protease epitopes, thereby interfering with its function. Such an event could result in elevated levels of unusually large vWF multimers and the enhanced platelet aggregation that is characteristic of TTP. With the recent commercialization of a laboratory assay that can provide semiquantitative analysis of vWF cleaving protease activity and vWF inhibitor activity, this becomes a readily testable hypothesis. Unfortunately, such a test had not yet been licensed when we were asked to consult on her care.

To date, only one other case of bee sting-associated TTP has been reported in the medical literature. Jones et al (9) described a 44-year-old woman who developed TTP shortly after she was stung by a honeybee. She had been stung by a bee and developed anaphylaxis three previous times, but in each instance she recovered without obvious long-term sequelae. Subsequently, she underwent Hymenoptera whole-body extract desensitization injections but had stopped receiving the injections 2 years earlier. After being stung by a bee for the fourth time, she briefly lost consciousness and was immediately taken to an emergency room. Over a 2-week period, she was monitored carefully but, because her clinical condition and MAI{A remained mild, active intervention was not pursued. She recovered spontaneously and remained in good health when, a year later, her case was published (Table 1).

Given the temporal relationship of our patient's symptom onset with a bee sting and the lack of other readily identifiable causes, we believe that bee envenomation was the most likely cause of TTP. We suggest that the envenomation led to an autoimmunologic cascade of events, resulting in enhanced platelet aggregation and the clinical manifestations of TTP. In conclusion, clinicians are reminded to question patients with TTP regarding potential immunologic, infectious, or pathologic triggers. A history of recent bee envenomation should also be sought before one concludes that TTP is idiopathic.

[FIGURE 1 OMITTED]

Table 1

Comparison of two bee sting-associated cases of TTP

Data                          Jones et al,        Ashley et al,
                              1979 (9)            2003 (present study)

Age (yr)                      44                  40
Sex                           Female              Female
Initial symptoms              Headaches followed  Crampy abdominal
                               by loss of          pain, diarrhea,
                               consciousness       head and chest pain

Time from bee sting to        72                  96
 medical evaluation (h)
Temperature on                37.6                37
 presentation ([degrees]C)
Initial laboratory studies
 Platelets (X [10.sup.9]/L)   13                  5
 Hematocrit (%)               28                  15.1
 Lactate dehydrogenase (U/L)  808                 1058
 Serum creatinine (mg/dl)     1.8                 0.9
Treatment                     Observation         Plasma exchange,
                                                   intravenous
                                                   immunoglobulin,
                                                   corticosteroids,
                                                   and aspirin
Outcome                       Well at 12-mo       Well at 18-mo
                               follow-up           follow-up

Acknowledgment

We thank Arleen Sierra for manuscript preparation.

Accepted October 28, 2002.

References

(1.) Rock GA. Management of thrombotic thrombocytopcnic purpura. Br J Haematol 2000; 109:496-507.

(2.) Elliot MA, Nichols WL. Thrombotic thrombocytopenic purpura and hemolytic uremic syndrome. Mayo Clin Proc 2001;76:1154-1 162.

(3.) George JN, El-Harake M. Thrombocytopenia due to enhanced platelet destruction by nonimmunologic mechanisms, in Beutler E, Lichtman MA, Coller BS, Kipps TJ, Seligsohn U (eds): Williams Hematology. New York, McGraw-Hill, 1995, ed 5, pp 1290-1315.

(4.) Kaplan BS, Towards a new classification of HUS and TTP. J Nephrol 1995;8:3 (editorial).

(5.) McCrac KR, Cines D. Thrombotic thrombocytopenic purpura and hemolytic uremic syndrome, in Hoffman R, Banz EJ Jr, Shattil SJ, Furie B, Cohen HJ, Silberstein LE, et al (eds): Hematology: Basic Principles and Practice. Philadelphia, Churchill Livingstone, 2000, ed 3, pp 2126-2138.

(6.) Ellingson TL, Wilske K, Aboulafia DM. Thrombotic thrombocytopenic purpura in a patient with polymyositis: Therapeutic importance of early recognition and discussion of pathogenic mechanisms. Am J Med Sci 1992;303:407-410.

(7.) Ruggenenti P, Noris M, Remuzzi G. Thrombotic microangioparhy, hemolytic uremic syndrome, and thrombotic thrombocytopenic purpura. Kidney Int 2001;60:831-846.

(8.) Allford SL, Machin SJ. Current understanding of the pathophysiology of thrombotic thrombocytopenic purpura. J Clin Pathol 2000;53:497-501.

(9.) Jones MB, Armitage JO, Stone DB. Self-limited TTP-like syndrome after bee sting. JAMA 1979;242:2212-2213.

(10.) Moake JL. Studies on the pathophysiology of thrombocytopenic purpura. Semin Hematol 1997;34:83-89.

(11.) Moake JL, Rudy CK, Troll JH, Weinstein MJ, Colannino NM, Azocar J, et al. Unusually large plasma factor VIII: Von Willebrand factor multimers in chronic relapsing thrombotic thrombocytopenic purpura. N Engl J Med 1982;307:1432-1435.

(12.) Moore JC, Murphy WG, Kelton JG. Caplain proteolysis of von Wille-brand factor enhances its binding to platelet membrane glycoprotein Ilb/IIIa: An explanation for platelet aggregation in thrombotic thrombocytopenic purpura. Br J Haematol 1990;74:457-464.

(13.) Cines DB, Konkle BA, Furlan M. Thrombotic thrombocytopenic purpura: A paradigm shift? Thromb Hoemost 2000;84:528-535.

(14.) Furlan M, Lammle B. Von Willebrand factor in thrombotic thrombocytopenic purpura. Thromb Hoemost 1999;82:592-600.

(15.) Furlan M, Robles R, Galbusera M, Remuzzi G, Kyrle PA, Brenner B, et al. Von Willebrand factor-cleaving protease in thrombotic thrombocytopenic purpura and the hemolytic-uremic syndrome. N Engl J Med 1998;339: 1578-1584.

(16.) Tsai HM, Lian EC. Antibodies to von Willebrand factor-cleaving protease in acute thrombotic thrombocytopenic purpura. N Engi J Med 1998;339:1585-1594.

(17.) Lintner TJ, Guralnick WMW, Lowenstein H. Allergens from stinging insects, in Kay AB (ed): Allergy and Allergic Diseases. Maiden, MA, Blackwell Science, 1997, vol 2, pp 927-941.

(18.) Yunginger JW. Insect allergy, in Middleton E Jr, Reed CE, Elliot EF, et al (eds): Allergy: Principles and Practice. St. Louis, Mosby-Year Book, 1998, ed 5, pp 1063-1072.

(19.) Reisman RE. Allergy to stinging insects, in Grammer LC, Greenberger PA (eds): Patterson Allergic Diseases. Philadelphia, Lippincott Williams & Wilkins, 2002, ed 6, pp 317-329.

RELATED ARTICLE: Key Points

* Thrombotic thrombocytopenic purpura (TTP) is a clinical syndrome characterized by the pentad of microangiopathic hemolytic anemia, thrombocytopenia, neurologic symptoms, renal insufficiency, and fever.

* Current evidence implicates deficient von Willebrand factor metalloproteinase activity in the pathogenesis of TTP.

* Diagnosis of TTP is made on the basis of clinical presentation and a new von Willebrand factor metalloproteinase assay.

* Bee envenomation is a rare cause of TTP.

Case Report

A 40-year-old woman was stung by a honeybee in August 2001 while gardening in her yard in western Washington state. She experienced immediate right flank pain at the site of the eavenomation. By the next day, she had low-grade fever and was bothered by abdominal cramps and pain as well as watery diarrhea. For an additional 4 days, she noted intermittent chest discomfort in conjunction with headache and easy bruising. When her symptoms failed to improve, she sought a medical evaluation in a local emergency room.

At the first evaluation, the patient appeared slightly anxious and irritable, had a temperature of 37[degrees]C, and her blood pressure was 105/60 mm Hg. A skin examination revealed scattered ecchymoses involving her extremities and, on her right flank, a 2-cm erythematous target lesion corresponding to her recent bee sting. Despite diffuse abdominal tenderness and hyperactive bowel sounds, she was without peritoneal signs or hepatosplenomegaly. A cardiac examination revealed only a resting tachycardia, and her lungs were without wheezes, rhonchi, or rales. The remainder of the physical examination was unremarkable.

The following initial laboratory study results were notable: hematocrit, 15.1%; reticulocytes, 6.2%; platelet count, 5 X [10.sup.9]/L and white blood cell count, 11.4 X [10.sup.9]/L. The peripheral smear showed 3+ schistocytes, occasional nucleated red cells, an absence of platelets, and normal-appearing leukocytes. Other relevant laboratory test results included the following: serum creatinine, 0.9 mg/dl; lactate dehydrogenase (LDH), 1,058 U/L (normal, 122-220 U/L); aspartate aminotransferase, 51 IU/L (normal, [less than or equal to]40); alanine aminotransferase, 28 IU/L (normal, <35 IU/L); and bilirubin, 1.9 g/dl (normal, <1.1 g/dl). Haptoglobin was undetectable (normal, 35-260 mg/dl), direct and indirect Coombs' tests were negative, serum immunoglobulin levels were normal, and the levels of the third and fourth components of complement were also within normal limits. The prothrombin and partial thromboplastin times also were normal, and fibrinogen was 577 mg/dl (normal, 200-450 mg/dl). Serum tests for preg nancy, antinuclear antibodies, lupus anticoagulant, and anticardiolipin antibody titers were negative, as were stool cultures for Shigella and Escherichia coli 0157:117. An electrocardiogram showed flattening of T waves in an inferolateral pattern, and this finding, in conjunction with the initial elevation of cardiac enzymes, was suggestive of myocardial ischemia.

The patient was diagnosed with TTP on the basis of the history of fever and the presence of headache, MAHA, and thrombocytopenia. She received 5 U packed red cells and 3 U fresh frozen plasma (FFP) before the initiation of plasma exchange. She was also placed in a telemetry unit, where she received nitrates and supplemental oxygen. Her chest pain resolved after the first infusion of FFP, and myocardial infarction was subsequently ruled out. For 4 consecutive days, I V plasma exchanges with FFP yielded no improvement. On Day 5 of her hospitalization, the patient's LDH was 679 U/L, platelets were 11 x [10.sup.9]/L, and stools were intermittently guaiac-positive for occult blood. On Day 5 of plasmapheresis, the patient became disoriented and demonstrated right lower-extremity weakness. An emergent computed tomographic scan of the head showed no abnormalities, and the patient's neurologic deficits resolved spontaneously 2 hours later. Because of concerns that her condition was worsening, she was administered 60 mg/d prednisone and 325 mg/d aspirin. In subsequent exchange transfusions, cryosupematant was used instead of FFP.

After 3 additional days of 1 V cryosupematant plasma exchange, she remained intermittently disoriented and her laboratory study results included the following: LDH, 1,177 U/L; platelets, S )< 109/L; and hematocrit, 26%. Her peripheral blood smear showed persistent schistocytes. Consequently, the cryosupernatant exchange was increased to 1.5 V/d plasma. She also received intravenously single doses of immune globulin (1 g/kg) and vincristine (2 mg). Within 48 hours, she appeared more lucid and was without focal neurologic deficits. Laboratory parameters also improved, and by Day 12 of her hospitalization, the platelet count was 17 X [10.sup.9]/L, LDH had decreased to 300 U/L, and hematocrit was stable without transfusion support at 27%. Over the next 5 days, her condition improved sufficiently so that the last two plasma exchanges were carried out on an outpatient basis. By then, her laboratory studies included a normal hematocrit, reticulocyte count, LDH, and hepatic panel (Fig. 1). She received a total of 17 plasma exchanges and 8 units of leuko-poor and irradiated packed red blood cells. More than 18 months after her hospitalization, she remains well and is no longer receiving both steroids and aspirin. Her most recent laboratory studies are within normal limits, and no schistocytes are present on peripheral blood smear.

From the Departments of General Internal Medicine and Hematology/Oncology, Virginia Mason Medical Center, and Division of Hematology, University of Washington, Seattle, WA.

Reprint requests to David Aboulafia, MD, Section of Hematology/Oncology, Virginia Mason Medical Center, 1100 Ninth Avenue, P.O. Box 900 (H14-HEM), Seattle WA 98111-0900. Email: [e-mail]

Copyright [c] 2003 by The Southern Medical Association 0038-4348/03/9606-0588