Thrombotic thrombocytopenic purpura masquerading as Evans syndrome

1 CASE PRESENTATION

A 26-year-old female with a history of chronic urticaria (treated with omalizumab in the past) and recently treated chlamydia trachomatis infection presented to an outside emergency department after a syncopal episode. She endorsed a three-day history of nausea, vomiting, fatigue, and ecchymoses. Initial laboratory analysis revealed macrocytic anemia (hemoglobin 4.6 g/dL) with a mean corpuscular volume (MCV) of 100.7 fL, thrombocytopenia (2000 plts/μL), indirect hyperbilirubinemia (indirect bilirubin 3.7 mg/dL), lactate dehydrogenase (LDH) 1245 units/L, absolute reticulocyte count of 0.237 million/μL, haptoglobin 1 mg/dL, international normalized ratio (INR) 1.2 and creatinine 0.75 mg/dL. Peripheral blood smear identified numerous spherocytes and microspherocytes, true thrombocytopenia, and increased reticulocytes without evidence of erythrocyte fragmentation (Figure 1). Direct antiglobulin testing (DAT) was positive for IgG (2+) and C3 (2+) at 37°C, although quantitative analysis was not performed. She received one unit of packed red blood cells and one unit of platelets and was transferred to our institution for further evaluation.

FIGURE 1

Open in figure viewerPowerPoint

Peripheral smear on admission (two sections) which showed spherocytes and microspherocytes, but no increased schistocytes.

This patient presented with profound thrombocytopenia and hemolytic anemia as evidenced by elevated LDH, elevated reticulocyte count, decreased haptoglobin, and indirect hyperbilirubinemia. The lack of schistocytes or evidence of erythrocyte fragmentation on peripheral smear as well as positive DAT was consistent with Evans syndrome—warm autoimmune hemolytic anemia (AIHA) with concomitant immune thrombocytopenia (ITP). Evans syndrome can arise spontaneously (primary) or secondary to diseases that generate autoantibodies. Etiologies may include infections, autoimmune disorders, lymphoproliferative disorders, or pregnancy.

On physical exam, the patient was well-appearing. She was afebrile with the remainder of her vital signs within normal limits. Physical examination was only notable for mild gingival bleeding. No rashes, petechiae, purpura, or ecchymoses were identified. She conversed appropriately and was oriented to person, location, and time. She had not received a blood product transfusion prior to this admission. She did not have a history of pregnancy. Family history was pertinent for Hashimoto's thyroiditis in her mother.

Overall, the patient appeared clinically well. She reported mild and nonspecific infectious symptoms including malaise, nausea, and emesis after recent domestic travel with her family. No other family members reported similar symptoms. Additionally, she lacked red flag signs and symptoms for malignancy or autoimmune disease including weight loss, night sweats, lymphadenopathy, myalgias, arthralgias, and rash. Nevertheless, broad workup for underlying infectious etiology, autoimmune disease and malignancy was pursued.

Infectious workup included negative bacterial blood cultures, EBV, CMV, HIV, HSV-1, HSV-2, HHV-6, hepatitis A, B, and C serologies and gastrointestinal pathogen panel. Autoimmune labs were negative for antinuclear antibodies (ANA) as well as antibodies against double-stranded DNA, Sjögren's-syndrome-related antigen A (SSA), Sjögren's-syndrome-related antigen B (SSB), and Smith. Thyroid function tests were within normal limits. Peripheral blood flow cytometry was sent given concern for atypical lymphocytes on further review of peripheral smear as well as hepatomegaly (22 cm in craniocaudal dimension) as noted on computed tomography (CT) abdominal imaging. Flow cytometry and positron emission tomography were both negative for a malignant process.

The above workup for secondary precipitants of Evans syndrome was unrevealing. Thus, idiopathic AIHA with concomitant ITP was most likely. As such, the patient was started on intravenous immunoglobulin (IVIG, 1 mg/kg) and high dose prednisone (1 mg/kg).

Her hemoglobin responded appropriately to transfusions, but her platelet count remained low despite high-dose steroids and IVIG. On day three of hospital admission, the patient endorsed acute onset headaches and new right arm weakness. On hospital day four, she was found to be obtunded. Urgent neurologic imaging was obtained with head CT negative for acute hemorrhage. Brain magnetic resonance imaging (MRI) demonstrated cortical hypoxia raising concern for encephalitis or seizures. She was transferred to the medical intensive care unit (MICU) for neurologic monitoring.

This acute change in clinical status now raised high suspicion for a life-threatening, alternate diagnosis. With new onset neurologic manifestations, thrombotic microangiopathies (TMAs)—most notably, thrombotic thrombocytopenic purpura (TTP)—ascended on the differential as a diagnosis that must be considered. Thus, the medical team reconsidered critical data that argued against MAHA—the positive DAT results and lack of erythrocyte fragmentation on peripheral smear.

On presentation, the patient lacked hallmark manifestations of TTP including fever and encephalopathy. However, the classic pentad of fever, anemia, thrombocytopenia, renal, and neurologic symptoms is rarely observed.¹ Patients often present with few of these symptoms or nonspecific symptoms that mimic other etiologies. Furthermore, neurologic manifestations of TTP range from mild confusion to acute ischemic stroke in greater than 60% of patients.^{2, 3} Prompt diagnosis and treatment are tantamount, as without therapy, mortality rates approach 90%.⁴ Even with timely plasma exchange (PLEX) and corticosteroids, mortality remains 10%–15%.⁴

The PLASMIC score is a validated, clinical diagnostic tool that aids clinicians in assessing the likelihood of severe a disintegrin and metalloproteinase with a thrombospondin type 1 motif, member 13 (ADAMTS13) deficiency.⁵ Notably, the PLASMIC score is only validated in those with TMAs. Thus, although the patient had a PLASMIC score of 6 on admission—placing her in the high-risk category for TTP (with a 72% risk of severe ADAMTS13 deficiency)—her peripheral smear lacked clear evidence of intravascular hemolysis indicative of a MAHA.⁶ Thus, empiric PLEX was deferred. Nevertheless, ADAMTS13 activity level was sent stat on admission; results were pending.

Additional examination of her peripheral smear on day four of hospitalization demonstrated many nucleated erythrocytes, acanthocytes, and atypical lymphocytes (Figure 2). Still, no schistocytes were noted. ADAMTS13 activity level returned undetectable and identified an ADAMTS13 inhibitor (anti-ADAMTS13 antibodies) at a titer of 2.5. Urgent PLEX was initiated, and high-dose corticosteroids continued. A peripheral blood smear from the subsequent day revealed increased numbers of schistocytes (Figure 3). After two sessions of PLEX, her hemoglobin, platelets, LDH, and total bilirubin all improved. With further PLEX sessions, her hemoglobin reached 10.3 g/dL, platelets reached 248 000 plts/μL, LDH 337 units/L, indirect bilirubin 0.2 mg/dL. Her mental status gradually returned to baseline. Her ADAMTS13 level improved from undetectable to 63% after four sessions of PLEX. Her DAT continued to remain positive for IgG and C3, though titers decreased from 2+ to only weakly positive.

FIGURE 2

Open in figure viewerPowerPoint

Peripheral smear on day four of admission (two sections) did not reveal increased number of schistocytes, despite clinical concern for TTP.

FIGURE 3

Open in figure viewerPowerPoint

Peripheral smear on day five of admission (two sections) which showed increased number of schistocytes. In conjunction with the undetectable ADAMTS13 activity, this confirmed the diagnosis of TTP.

This case underscores the need for clinicians to consider alternative diagnoses in the face of new data and a rapidly changing clinical course. Persistent anemia and thrombocytopenia despite IVIG and prednisone with the development of new acute encephalopathy demanded an alternate diagnosis. Despite initial data supporting Evans syndrome, TMA—most particularly TTP—became much more likely. When ADAMTS13 activity returned undetectable, TTP was confirmed. PLEX treatment was swiftly initiated.

Despite continued PLEX, platelets precipitously dropped from 218 000 to 25 000 plts/μL within 3 days. LDH also increased from 365 units/L to 659 units/L. Rituximab 375 mg/m² weekly was added. Daily PLEX sessions continued. Interval ADAMTS13 testing again returned undetectable and identified an increase in ADAMTS13 inhibitor with a titer of 11.1. Thrombocytopenia continued to worsen, reaching a nadir of 14 000 plts/μL at which time caplacizumab was initiated for recalcitrant TTP. After 1 week of caplacizumab, her ADAMTS13 activity increased to 25% with undetectable ADAMTS13 inhibitor and normal hemoglobin and platelet count. A peripheral blood smear showed decreased number of schistocytes (Figure 4). Following 1 month of caplacizumab, ADAMTS13 activity fluctuated to 17%, but counts remained stable. Caplacizumab therapy was extended to 6 weeks and after also completing four weekly doses of rituximab, her ADAMTS13 improved to 21%.

FIGURE 4

Open in figure viewerPowerPoint

Peripheral smear after initiation of PLEX, rituximab and caplacizumab (two sections) showed improvement in number of schistocytes.