RSS Feeds
Recurrent Pancytopenia, Coagulopathy, and Renal Failure Associated with Multiple Quinine-dependent Antibodies
- Robert B. Maguire, MD;
- David F. Stroncek, MD; and
- Allan C. Campbell, MD
- From the University of Illinois College of Medicine at Peoria, St. Francis Medical Center, and Proctor Hospital, Peoria, Illinois; the University of Minnesota Medical School, Minneapolis, Minnesota; the American Red Cross, St. Paul, Minnesota. Requests for Reprints: David Stroncek, MD, University of Minnesota Hospital and Clinic, Department of Laboratory Medicine and Pathology, Blood Bank Laboratory, Box 198 UMHC/D211 Mayo, 420 Delaware Street SE, Minneapolis, MN 55455. Acknowledgments: The authors thank Mary Egging, Gail Eiber, Greg Herr, Jane Swanson, and Shelley Pulkrabek for their technical assistance and Bobbie Gibson for help in manuscript preparation. Grant Support: In part by funds from the Residents' Research Fund, St. Francis Medical Center, and The American Red Cross St. Paul Regional Blood Services.
Immune thrombocytopenia mediated by drug-dependent antibodies to platelets is a well-known adverse reaction to quinine [1, 2]. Recent reports of disseminated intravascular coagulation [3] and the hemolytic uremic syndrome [4, 5] in association with multiple quinine-dependent antibodies suggest the existence of immune-mediated effects on cell lines other than platelets. We describe two patients who were not known to be taking quinine and who had a recurrent febrile illness associated with pancytopenia and renal failure along with serologic evidence of drug-dependent platelet, neutrophil, and erythrocyte antibodies.
Case Reports
Patient 1
A 41-year-old white woman came to the emergency department in August 1992 with a 24-hour history of nausea, vague abdominal pain, and fever. She had been hospitalized on at least four other occasions with a similar illness in the 5 years before hospitalization. The patient was vague about her past medical history, but the first episode occurred after the ingestion of a quinine compound taken for leg cramps. Because she had a hepatotoxic reaction, she was instructed never to take quinine again. The second and third episodes occurred in 1991, when the patient was hospitalized twice for an acute febrile illness characterized by hypotension, pancytopenia, coagulopathy, and acute renal failure. Disseminated intravascular coagulation was diagnosed, but no cause was identified. Records from both hospitalizations indicated that she had a quinine allergy, but no records indicated that she admitted to or was questioned specifically about quinine ingestion. Her hematologic abnormalities and renal insufficiency resolved with supportive care and empiric antibiotics for presumed sepsis.
During her work-up at our hospital, she listed an allergy to quinine but repeatedly denied quinine ingestion. Physical examination at the time of hospitalization showed a pulse of 114 beats per minute, a blood pressure of 84/40 mm Hg, and a temperature of 38.2 C (100.7 F). Extremities were cool with acrocyanosis. Lung fields were clear and heart sounds were normal. Results of her physical examination were otherwise unremarkable. Laboratory data are summarized in Table 1. Oliguric renal failure developed. Serum and urine obtained at the time of hospitalization tested positive for quinine although the patient repeatedly denied quinine ingestion. Acute-phase serum was obtained to test for quinine-dependent antibodies.
The patient required hemodialysis and was treated empirically with broad-spectrum antibiotics. Her hematologic abnormalities improved gradually. After 4 months of hemodialysis, her renal function normalized.
Six weeks after discharge from the hospital, the patient came to the outpatient hemodialysis center and complained of fever and chills. She was noted to be thrombocytopenic, with quinine present in her urine and serum. She admitted drinking tonic water the previous evening, despite having been given strong warnings about the use of quinine-containing beverages. Patient 1's recurrent illness appeared to be precipitated by intentional quinine ingestion and to represent a variant of the Munchausen syndrome [6, 7].
Patient 2
A 65-year-old white woman with a history of a recurrent febrile illness came to another hospital in July 1992. She reported the acute onset of back pain, nausea, vomiting, and fever and was transferred to our facility for treatment. In November 1989, she had identical symptoms and was found to be hypotensive with pancytopenia, coagulopathy, and acute renal failure. She was treated for thrombotic thrombocytopenic purpura with plasma exchange and hemodialysis, and her hematologic abnormalities resolved, although her renal failure did not. She remained dependent on chronic ambulatory peritoneal dialysis. She was hospitalized again in August 1991 with fever, hypotension, pancytopenia, and coagulopathy and responded to supportive therapy without requiring plasma exchange.
Physical examination showed mild distress. Her blood pressure was 86/58 mm Hg; her temperature was 39.7 C (103.5 F); her pulse was 176 beats per minute; and her respirations were 28 per minute. Cardiorespiratory and neurologic examinations showed normal results. Laboratory data are summarized in Table 1. The patient was empirically treated with plasma exchange, and her hematologic abnormalities gradually improved.
When questioned 3 months later, she reported taking quinine sulfate (260 mg) occasionally for leg cramps, although she could not remember taking the tablets before her episodes. Her pharmacist reported that the original prescription for 20 tablets of quinine sulfate was filled in 1988 and was refilled in January 1989. Three tablets were missing from the latter pill bottle, corresponding to the number of acute febrile episodes. Serum was obtained to test for quinine-dependent antibodies.
Methods
Sera were mixed with cells from normal donors in the presence and absence of quinine and quinidine. Granulocyte agglutination and immunofluorescence, neutrophil immunoprecipitation, platelet immunofluorescence, monoclonal antibody-specific immobilization of platelet antigens, and erythrocyte agglutination assays were done as previously described [5].
Results
Drug-dependent antibodies to neutrophils, platelets, and erythrocytes were detected in the sera of both patients in the presence of quinine as well as quinidine. Characterization of the antibodies' titer and class are shown in Table 2.
The sera was tested against platelets in a monoclonal antibody-antigen capture assay. Quinine plus serum 1 reacted with platelet membrane glycoprotein complex IIb/IIIa, but in the presence of quinine, serum 2 did not react with any of the platelet membrane complexes tested (IIb/IIIa, Ib/IX, and Ib/IIa). In the presence of quinine, both sera immunoprecipitated neutrophil membrane molecules. Sera from patient 1 reacted with a 60-kd molecule and serum 2 reacted with 85-kd, 60-kd, and 32-kd molecules.
Discussion
Both patients had a recurrent febrile illness characterized by hypotension, pancytopenia, coagulopathy, and renal failure. Recorded drug histories in each patient showed no history of recent quinine use, and, on retrospective questioning, one patient denied quinine ingestion and the other could not recall quinine ingestion. Both patients had high titers of quinine-dependent antibodies and showed cross-reactivity with quinidine, a quinine stereoisomer. Quinine- and quinidine-dependent antibody cross-reactivity has been noted previously with antierythrocyte and antiplatelet antibodies [8] but not with antineutrophil antibodies.
The most commonly reported adverse reaction to quinine is thrombocytopenia, secondary to quinine-dependent platelet antibodies [1, 2]. Recently, disseminated intravascular coagulation and recurrent episodes of the hemolytic-uremic syndrome after quinine ingestion have been reported in association with quinine-dependent platelet antibodies [3-5]. These reports are summarized in Table 3. Drug-dependent erythrocyte antibodies have also been reported to be associated with a high incidence of acute renal failure and disseminated intravascular coagulation [9]. Our experience and that of others [5] suggest that erythrocyte and neutrophil antibodies might also be involved. In patients with blood dyscrasias, unexplained coagulopathy, and renal failure, drug-dependent antibodies should be considered. Fulminant adverse reactions to quinine may be clinically underappreciated and a high index of suspicion must be maintained even when carefully done histories do not show quinine ingestion.
Quinine is available in many over-the-counter preparations and is recommended by many physicians for the treatment of leg cramps. Because of the serious adverse reactions noted here and in previous reports [3-5], consideration should be given to making quinine available by prescription only. At minimum, warnings of quinine's potential harmful effects should be printed on all over-the-counter preparations and on bottles of tonic water.
- Copyright 2004 by the American College of Physicians
