Atypical thrombotic thrombocytopenic purpura in a middle-aged woman who presented with a recurrent stroke


  • Conflict of interest: Nothing to report

  • A physician or group of physicians considers presentation and evolution of a real clinical case, reacting to clinical information and data (boldface type). This is followed by a discussion/commentary

Correspondence to: Modupe Idowu, Division of Hematology, Department of Medicine, University of Texas, Houston, TX 77030. E-mail:

A 48-year-old African American woman was brought to the emergency department by ambulance for evaluation of acute onset of altered mental status and neurological deficits. She was found leaning over the side of her bathtub with aphasia, left-sided facial droop, and left hemiparesis, suggestive of a stroke. Prior to this event, she was able to walk and perform daily activities. At baseline, she had mild dysarthria, minimal expressive aphasia, and right upper extremity weakness and spasticity. These were residual neurological deficits from the cerebrovascular accident (CVA) she had 6 years ago. Her past medical history included ischemic strokes 8 years ago associated with post-CVA seizures, another ischemic stroke 6 years ago, and thrombotic thrombocytopenic purpura (TTP). On physical examination, she appeared lethargic, disoriented to person, place, and time; she was unable to name objects or repeat words; and she was unable to follow commands. She was found to have a left facial droop, partial right-sided gaze preference, left hemianopia, and spastic left hemiparesis. A CT scan of the head without contrast showed two previous middle cerebral artery (MCA) territory cerebral infarctions with an acute right MCA occlusion at the origin of the M2 segment. An additional thrombus was present in a branch of the right MCA along the distribution of the posterior insular cortex. An MRI of her brain revealed a large area of acute ischemia within the distribution of the anterior division of right MCA, small foci of petechial hemorrhage in the right insula around the area of ischemia, and small foci of acute ischemia in the right cerebellar hemisphere. A CT angiogram confirmed these findings.

This patient presented with acute stroke involving the distribution of the right MCA. Her vital signs on admission were within normal limits. Given that her clinical presentation was indicative of a stroke and having arrived within 3 hr of the onset of her symptoms, she was given pharmaceutical tissue plasminogen activator (tPA). Intra-arterial thrombolysis and thrombectomy were performed; however, her neurological symptoms did not improve.

Her family history was negative for premature strokes or thrombotic disorders. The patient did not smoke, drink alcohol, or use illicit drugs. Her home medications included Keppra and Dilantin for seizure prophylaxis. A transesophageal echocardiogram performed on admission was unremarkable. Her lipid profile and HgbA1c were within normal limits. Admission laboratory data included hemoglobin (Hgb) 13.2 g/dl, hematocrit (Hct) 38.6%, and platelet (PLT) count of 113,000 per cubic mm of whole blood. Other laboratory values included lactate dehydrogenase (LDH) 238 U/l (normal range, 98–192 U/l), reticulocyte count 1.6%, international normalized ratio (INR) 1.22, prothrombin time (PT) 15.6 sec, activated partial thromboplastin time (aPTT) 26.7 sec, blood urea nitrogen (BUN) 6 mg/dl, and serum creatinine 0.8 mg/dl. Peripheral blood smear (PBS) did not show schistocytes. Patient's neurological deficits and altered mental status persisted. She was then intubated to avoid respiratory compromise secondary to drowsiness and was transferred to the intensive care unit. Systolic blood pressures were maintained at less than 140. On further questioning, the patient's mother reported that both of patient's prior strokes were determined to be secondary to TTP because in each instance, she had associated hemolytic anemia, thrombocytopenia, and her symptoms subsided following therapeutic plasma exchange (TPE).

In the absence of cardiovascular risk factors or cardiogenic emboli, CVA in a middle-aged patient increases the suspicion of unusual thrombophilia. An atypical TTP episode could not be ruled out in this patient, even though when she presented to the hospital, she had normal Hgb, slightly decreased PLT count, an absence of schistocytes on PBS, mildly elevated LDH, and normal renal function. This patient's medical records did not indicate that a disintegrin and metalloprotease with thrombospondin type 1 motif 13 (ADAMTS13) level was checked during her prior TTP episodes or during periods of remission. Her medical records indicated that her clinical presentation during prior episodes of TTP was typical. At those times, she had microangiopathic blood smear with Hgb under 8 g/dl, high LDH of above 600 U/L, and PLT counts as low as 20,000 per cubic mm of blood. It would have been quite beneficial to obtain a detailed history in this patient who lacked typical atherosclerotic risk factors prior to subjecting her to major interventions such as thrombolytic therapy. Failure to obtain thorough history caused this patient to be exposed to a great magnitude of unnecessary risks such as bleeding complications that might occur as a result of thrombolytic therapy.

The lack of response to thrombolysis and her history of recurrent strokes in the setting of TTP prompted hematology consultation on hospital day 3 and she was started on empiric TPE at that time. On Day 3, it was noted that her PLT count had decreased from 113,000 to 97,000 per cubic mm and her Hgb had decreased from 13.2 to 11.1 g/dl, but without microangiopathic blood smear. Hematology also considered other differential diagnoses of stroke and thrombocytopenia in a young patient; however, further work-up and review of her medical records failed to show evidence of vasculitis, antiphospholipid syndrome, infections, or exposure to toxic drugs. Prior to initiation of TPE, a blood sample was drawn from her and sent for ADAMTS13 activity level using the fluorescence resonance energy transfer (FRET) assay. It took few days for her test results to come back, but her TPE was continued during this period. A repeat brain MRI performed on hospital day 5 showed new areas of infarction in the right parietal lobe and in the right frontal operculum. A new area of hemorrhagic change was also found in the right MCA infarct.

These new brain MRI findings may be due to the intra-arterial thrombolysis done on admission because thrombolysis can be detrimental in patients with TTP. Hematology should have been consulted immediately because this patient had a history of previous strokes associated with her past TTP episodes. Considering her history of recurrent cerebral infarctions secondary to TTP episodes, the most appropriate initial step in her management should have been empiric TPE as opposed to thrombolysis. A repeat brain imaging should have also been done 24 hr after thrombolysis per this institution's acute ischemic stroke post-thrombolytic protocol to evaluate response to therapy or complication from therapy. In this patient's case, repeat brain imaging was done 4 days later.

Quantitative analysis of this patient's plasma revealed an ADAMTS13 activity of 16% of normal control. Moreover, mixing studies and an ELISA assay were negative for both an ADAMTS13 inhibitor and IgG antibody, respectively, in our patient. After the first of five procedures, she recovered some of her neurological deficits. After all five treatments, she no longer had evidence of left facial droop, left hemianopia, right gaze preference, and spastic left hemiparesis. She was able to speak and exhibited only the mild dysarthria, minimal expressive aphasia, and right upper extremity weakness and spasticity that she had at baseline. TPE was stopped after five treatments because she already had significant neurologic recovery and her laboratory findings did not suggest further microangiopathic hemolytic anemia (MAHA). Her laboratory data showed PLT of 225,000 per cubic mm, stable Hgb/Hct, and normal LDH. ADAMTS13 genetic mutation analysis was not done at that time because most patients with congenital TTP have severely reduced, or absent, ADAMTS13 activity level, and failure to detect ADAMTS13 antibody may reflect limited sensitivity of the test system that was used [[1, 2]]. Several months after discharge, she remained stable without any concern for TTP relapse.


Differential diagnosis of stroke and thrombocytopenia in a young patient without atherosclerotic risk factors include vasculitis, antiphospholipid syndrome, certain infections (e.g., syphilis, Lyme disease, tuberculosis, and fungal), toxic drugs, or unusual thrombophilia. Large artery occlusive changes can also result from various causes such as cardiogenic embolism, thromboembolism, fibromuscular dysplasia, vasospasm, and dissection of cerebral vessel. These diagnoses were also considered in our patient; nevertheless, atypical presentation of TTP was included and favored in the differential diagnosis of stroke in our patient because of her history of recurrent strokes in the setting TTP.

TTP is categorized as being idiopathic or nonidiopathic and congenital or acquired. Many TTP cases are idiopathic. However, some cases may occur in association with autoimmune diseases [3], infection [4], pregnancy [5], bone marrow transplantation [6], and certain drugs such as cyclosporine A, tacrolimus, mitomycin, or ticlopidine [7], which were not suggested in this patient's history. TTP is commonly associated with a deficiency of ADAMTS13, the von Willebrand factor-cleaving plasma metalloprotease that cleaves the von Willebrand factor at the Tyr1605-Met1606 peptide bond in the central A2 subunit [8, 9]. Many cases of idiopathic TTP are caused by an underlying ADAMTS13 deficiency [2]. Patients with acquired TTP may have circulating anti-ADAMTS13 inhibitory autoantibodies or IgG antibodies that neutralize its enzymatic activity and increases ADAMTS13 removal from the circulation, leaving the patient with low plasma levels of the enzyme [10, 11]. Some patients with idiopathic TTP may have no demonstrable ADAMTS13 inhibitor, just as in our patient, and this may be reflective of the imperfect sensitivity of the test system that is used [2]. The prevalence of ADAMTS13 inhibitors in patients with idiopathic TTP who had severe ADAMTS13 deficiency was 44% in a prospective study [2]. It is also interesting to note that the severity of ADAMTS13 deficiency in patients with idiopathic TTP does not correspond to the presenting features or response to TPE according to publications from Oklahoma TTP-HUS Registry [12, 13].

TTP is described by a pentad of thrombocytopenia, anemia, fever, renal, and neurologic abnormalities [14]. Conversely, patients with TTP rarely present with the classic pentad. Patients presenting with MAHA and thrombocytopenia without an identifiable cause are typically suspected to have TTP, especially if they have associated neurological symptoms [15]. Key features of TTP are attributed to the presence of microthrombi in small blood vessels that shear red blood cells and consume PLTs. Widespread microvascular occlusions in multiple organs of patients with TTP result in the typical presentation of nonfocal organ dysfunction. Large artery occlusive changes are rare in patients with TTP, but it is possible that a large thrombus may develop and block medium to large size arteries without resulting in significant MAHA and thrombocytopenia.

In recent literature, atypical TTP has been described in patients presenting with strokes or myocardial infarction with normal or near normal, laboratory values, and a low ADAMTS13 activity [16-20]. Downes et al. [16] reported two interesting cases of atypical TTP. The first case was of a 40-year-old African-American woman with history of chronic relapsing TTP who presented with facial numbness, left upper extremity weakness, dysarthria, and diplopia [16]. On admission, she had Hgb of 9.3 g/dl, PLT of 239 per cubic mm, LDH of 296 U/L (normal range, 68–239 U/L), and rare schistocytes on PBS. Her brain MRI study showed an acute right hemispheric infarction. Her history of TTP led to consideration of relapsed TTP as the etiology for her stroke, and therefore, she was started on TPE. Twelve hours after the first TPE, she had increased strength in her left hand and less dysarthria. She showed significant improvement in her neurological status after receiving 7 TPEs. Her ADAMTS13 activity level was 11% at presentation and an ADAMTS13 inhibitor was detected. The second case described by the same authors was of a 42-year-old Caucasian woman with chronic relapsing TTP and history of strokes in the setting of TTP [16]. Her fourth stroke was associated with motor weakness, dysphasia, and confusion without laboratory findings typical for TTP. An MRA showed an acute infarct in the left basal ganglia and old cerebral infarcts. Because of her history of stroke in the setting of TTP, she was immediately started on TPE. Her ADAMTS13 activity level was 12% on admission and she had ADAMTS13 IgG antibody. Her neurological condition showed significant improvement prior to discontinuation of TPE, but she had multiple relapses in the following 3 years.

There were also other cases of atypical presentations of TTP described in literature. One case report was regarding a 36-year-old African-American woman who had her first episode of TTP at 27 years [18]. She did well until 8 years later, when she experienced a stroke with motor deficits and dysarthria; however, her PLT count, bilirubin, and LDH were normal and she had no schistocytes on PBS. Three weeks after the stroke, she developed an overt episode of TTP with PLT count of 6 per cubic mm and schistocytes on PBS. Her neurological status and blood counts improved after TPE and prednisone. Likewise, O'Brien and Crum [19] reported two cases of TTP where there were delays in diagnosis because neurologic deficits were initially the dominant findings, followed weeks later by pronounced hematologic manifestations [19]. A recent retrospective study indicated that abnormal brain neuroimaging is common in TTP, but large infarctions and hemorrhage are infrequent [21]. This study also showed that abnormal brain neuroimaging does not seem to impact patient outcome, and full neurologic recovery is possible even in patients with extensive brain abnormalities on MRI.

Approximately two-thirds of the few cases of patients described in literature who presented with atypical TTP episodes in association with isolated focal neurologic symptoms had a prior history of typical TTP episodes. Although it can be argued that recurrent TTP is rare in patients who do not have severe ADAMTS13 deficiency (≤10%), our patient's history of this disorder and her clinical response to TPE strongly suggest TTP as the etiology of her stroke. The cases described by Downes et al. were similar to our patient's case in the sense that these patients had recurrent TTP and their ADAMTS13 levels were higher than 10% [16]. Further studies of ADAMTS13 levels in patients who present with atypical TTP may help delineate the value of such measurements in the management of this disorder.

Our patient's case and similar cases presented in the literature illustrated the fact that TTP patients may present soon after the onset of symptoms such that laboratory abnormalities like profound thrombocytopenia or microangiopathic changes had little time to become manifest. Because PLT aggregation and thrombosis are not widespread in the early stages of TTP, thrombocytopenia and MAHA may not be clinically evident. Nonetheless, if PLT thrombi affect the vasculature of a vital area, patients may become symptomatic. A timely diagnosis of TTP in a patient with acute stroke is essential because TPE may lead to a dramatic improvement in neurologic status; however, a delay in the diagnosis and treatment may lead to devastating outcomes. Serious consequences may result if thrombolytic or anticoagulation therapy is initiated as part of acute ischemic stroke protocol prior to recognition of TTP [22].


An atypical clinical presentation makes TTP diagnosis difficult, which prevents prompt management of TTP. This case highlights the importance of early recognition of TTP in patients with atypical presentations who may not have the expected clinical or laboratory findings. The level of suspicion should be especially high in young and middle-aged patients with strokes or acute coronary syndrome who do not have other risk factors for cardiovascular events. This is particularly so for those patients who have a previous episode of TTP or patients with subtle laboratory abnormalities, which may suggest the potential existence of a thrombotic microangiopathic process.