Postpartum Thrombotic Thrombocytopenic Purpura (TTP) Complicating Pregnancy-Associated Immune Thrombocytopenic Purpura (ITP)
- Martee Olenich, MD; and
- Elaine Schattner, MD
- From The New York Hospital-Cornell Medical Center, New York, New York. Requests for Reprints: Elaine Schattner, MD, Division of Hematology-Oncology, Room C-606, The New York Hospital-Cornell Medical Center, 525 East 68th Street, New York, NY 10021.
Although immune thrombocytopenic purpura in pregnancy is common, thrombotic thrombocytopenic purpura is rare and requires aggressive therapy. We report the case of a young pregnant woman with refractory immune thrombocytopenic purpura who developed postpartum thrombotic thrombocytopenic purpura and was successfully treated with plasmapheresis. The case emphasizes the importance of reviewing peripheral blood smears for evaluation of thrombocytopenia even when the diagnosis is felt to be clear. The coexistence of these two disorders has not been previously reported in the English-language literature.
Case Report
A 15-year-old girl, gravida 1, para 0, presented at 11 weeks of pregnancy with hyperemesis gravidarum. A manual platelet count was 56 × 109/L. The patient had had no previous blood counts and no history of bleeding or bruising. She was normotensive and had no petechiae, bruises, or organomegaly. The hemoglobin was 107 g/L; hematocrit, 0.33; mean corpuscular volume, 83 fL; and leukocyte count, 11.5 × 109/L. The peripheral blood smear showed normal red cell structure, with some platelet clumping and an overall decreased platelet count. The prothrombin time was 12.1 s (normal range, 9 to 13 s), and the activated partial thromboplastin time was 31.8 s (normal range, 23 to 38 s). Circulating antiplatelet antibody assays were negative, and renal and liver functions were normal. Idiopathic thrombocytopenic purpura was diagnosed, based on persistent thrombocytopenia during an otherwise normal pregnancy. The patient was followed until week 36 without pharmacologic intervention. Her platelet count ranged from 40 × 109/L to 68 × 109/L, and she had no clinical evidence of bleeding. At 36 weeks the patient was hospitalized with early labor, cervical dilatation, and a platelet count of 20 × 109/L. Peripheral blood smears still showed normal red cell structure and thrombocytopenia (Figure 1, top). The patient was treated with intravenous methylprednisolone, 20 mg every 8 hours, and gammaglobulin, 0.5 g/kg body weight daily for 4 days. Because she did not respond to therapy, bone marrow aspiration and biopsy were done to confirm the diagnosis of immune thrombocytopenic purpura. The aspirate showed fully maturing granulocytic and erythroid elements, with abundant megakaryocytes and decreased iron stores. She refused administration of vincristine, and a second course of methylprednisolone, 1 g/d for 2 days, and gammaglobulin, 2 g/kg for 3 days was given.
On day 13, the fetus developed late cardiac decelerations. The patient went into active labor, and an uncomplicated cesarean section was done. She received 12 units of platelets intraoperatively. The patient delivered a normal infant boy whose initial platelet count was 100 × 109/L. After the cesarean section the patient appeared well despite a decrease of 2 g/L in hemoglobin that was probably caused by intraoperative blood loss. The platelet count increased to 41 × 109/L. Pneumovax 0.23 [Merck and Company; West Point, Pennsylvania] was given in preparation for a possible splenectomy. On postoperative day 3 she was afebrile and asymptomatic except for some ecchymoses and petechiae, but her hemoglobin decreased to 56 g/L and her platelet count, to 7 × 109/L. Review of the peripheral blood smear showed a new, marked schistocytosis with nucleated red blood cells and polychromasia (Figure 1, bottom). The lactate dehydrogenase level was 19.32 µkat/L (Table 1).
On the basis of the postpartum development of microangiopathic hemolysis and profound thrombocytopenia, combined with the absence of hypertension, uremia, or disseminated intravascular coagulation, we diagnosed thrombotic thrombocytopenic purpura. Although overt neurologic symptoms were absent, trace proteinuria (0.3 g/L) and microscopic hematuria were present. Fever may have been masked by the high dose of steroids she received. We treated her with plasmapheresis, antibiotics, and continued steroids. The hemolysis cleared, and her platelet count increased steadily. After 4 days the platelet count was more than 100 × 109/L, and the steroids were tapered. During the next month, while not receiving steroids, her platelet count decreased to 64 × 109/L, but hemoglobin and lactate dehydrogenase levels remained normal. The peripheral blood smear showed no schistocytes. Subsequent outpatient evaluation has included a negative antinuclear antibody test, normal C3 and C4 levels, and a negative test for human immunodeficiency virus (HIV-1). The patient has chronic immune thrombocytopenic purpura, is asymptomatic, and currently requires no therapy.
Discussion
This case shows the importance of reviewing peripheral blood smears for pregnant patients with thrombocytopenia, even when the diagnosis seems to be clear. Burrows and Kelton [1] have shown that immune thrombocytopenic purpura is relatively common in pregnancy, occurring in 11 of 6715 consecutive deliveries at McMaster University, or approximately 1 in 500 deliveries. The condition is usually detected incidentally and usually does not require medical intervention. In contrast, thrombotic thrombocytopenic purpura is a rare complication of pregnancy. The McMaster series had no cases of thrombotic thrombocytopenic purpura. Thrombotic thrombocytopenic purpura must be treated aggressively, as illustrated by Weiner's review of 45 cases of pregnancy-associated thrombotic thrombocytopenic purpura reported in the literature. He found a mortality rate of 44% for the mothers and of 80% for the infants [2]. Most cases of pregnancy-associated thrombotic thrombocytopenic purpura occur antepartum; in Weiner's series, only 5 (11%) presented after delivery.
We speculate that the occurrence of these two disorders in one person was not accidental. The fact that both immune and thrombotic thrombocytopenic purpura occur with increased frequency among persons with systemic lupus erythematosus, HIV disease, or pregnancy supports the hypothesis that some pathophysiologic factors are shared. Circulating antibodies or antigen-antibody complexes, caused by a primary autoimmune aberration, might induce endothelial cell damage and, consequently, thrombotic thrombocytopenic purpura. Alternatively, both immune and thrombotic purpura may represent separate hematologic manifestations of one underlying process.
The distinctions among various microangiopathic hemolytic anemias associated with pregnancy are subtle [3, 4]. The diagnosis of thrombotic thrombocytopenic purpura can be obscured by signs of eclampsia or the hemolytic-uremic syndrome. In one case that may have been similar to ours, a woman had severe preeclampsia with persistent postpartum hemolysis and thrombocytopenia, which responded to plasmapheresis [5]. Similarly, our patient developed thrombotic thrombocytopenic purpura superimposed on a relatively common disorder, idiopathic thrombocytopenic purpura. The prognosis and management of these conditions differ completely; accurate diagnosis with review of the peripheral blood smear is crucial.
- Copyright ©2004 by the American College of Physicians
RSS Feeds










