Hematologic Causes of Stroke

Hematologic disorders account for a small but clinically important subset of ischemic strokes, particularly in younger patients without traditional vascular risk factors. These conditions include acquired and inherited hypercoagulable states, hemoglobinopathies, myeloproliferative neoplasms, thrombotic microangiopathies, and cancer-associated thrombosis. Recognition is critical because management often diverges significantly from standard antiplatelet-based secondary prevention.

When to Suspect a Hematologic Cause

Clinical red flags that should prompt evaluation for an underlying hematologic disorder include:

• Stroke in a young patient (<50 years) without traditional vascular risk factors
• Recurrent arterial or venous thrombosis, particularly in unusual sites (cerebral venous thrombosis, splanchnic vein thrombosis)
• Concurrent or prior deep vein thrombosis or pulmonary embolism
• Recurrent pregnancy losses or obstetric complications
• Known malignancy or unexplained weight loss
• Abnormal CBC findings: erythrocytosis, thrombocytosis, or cytopenias with schistocytes
• Hemolytic anemia (elevated LDH, low haptoglobin) with negative Coombs test — consider PNH
• Platelet drop 5–10 days after heparin exposure — consider HIT
• Palpable purpura with low C4 and/or hepatitis C — consider cryoglobulinemia
• Livedo reticularis or other skin findings suggesting systemic vasculopathy
• Family history of thrombophilia or early-onset thrombosis

Classification of Hematologic Stroke Etiologies

Antiphospholipid Syndrome

Pathophysiology and Diagnosis

Antiphospholipid syndrome (APS) is an acquired autoimmune disorder characterized by arterial and/or venous thrombosis in the presence of persistent antiphospholipid antibodies (aPL). Stroke is among the most common arterial manifestations, accounting for approximately 20% of strokes in patients under 45 years.

Diagnosis requires both clinical criteria (vascular thrombosis or pregnancy morbidity) and laboratory confirmation of aPL persistence on two occasions at least 12 weeks apart (revised Sydney criteria). The three clinically relevant antibodies are:

• Lupus anticoagulant (LA): Detected by clotting assays (dRVVT, aPTT mixing studies)
• Anticardiolipin antibodies (aCL): IgG or IgM at medium-to-high titers (>40 GPL/MPL or >99th percentile)
• Anti-β2-glycoprotein I antibodies: IgG or IgM >99th percentile

Trial Evidence: DOACs vs Warfarin in APS

Multiple randomized trials have now established that DOACs are inferior to warfarin for stroke prevention in APS, particularly in high-risk patients:

The TRAPS trial (2018) randomized 120 patients with high-risk triple-positive APS to rivaroxaban 20mg daily vs warfarin (INR 2–3). The trial was stopped early due to excess events in the rivaroxaban arm: 19% vs 3% experienced the composite outcome (HR 6.7, P=0.01), driven primarily by arterial thrombotic events including stroke. The TRAPS 2-Year Outcomes follow-up reinforced these findings, with a composite outcome of 33.3% in patients remaining on DOACs vs 5.7% on warfarin (HR 6.9, P=0.018).

The ASTRO-APS trial (2022) compared apixaban to warfarin in thrombotic APS and was also stopped early for excess strokes in the apixaban group (6/23 vs 0/25 patients). Major bleeding was similar between groups.

The RAPS trial (2016) studied rivaroxaban vs warfarin in a lower-risk APS population (VTE without arterial events, 28% triple-positive). While the primary endpoint was a laboratory surrogate (thrombin generation), no thrombotic events occurred in either arm during 6 months of follow-up. This suggests DOACs may be acceptable in carefully selected lower-risk APS patients with isolated VTE.

Paroxysmal Nocturnal Hemoglobinuria

Pathophysiology

Paroxysmal nocturnal hemoglobinuria (PNH) is an acquired clonal disorder caused by somatic mutations in the PIGA gene, leading to deficiency of glycosylphosphatidylinositol (GPI)-anchored proteins on blood cell surfaces. The absence of complement regulatory proteins CD55 and CD59 renders red blood cells susceptible to complement-mediated lysis and creates a profoundly prothrombotic state.

Thrombosis is the leading cause of death in PNH, occurring in 30–50% of untreated patients. Unlike most hematologic conditions, PNH causes both arterial and venous thrombosis with a particular predilection for unusual sites:

• Cerebral venous sinus thrombosis: Most common neurological manifestation
• Arterial ischemic stroke: Less common but well-documented
• Hepatic vein thrombosis (Budd-Chiari syndrome): Classic PNH association
• Mesenteric and portal vein thrombosis: Often presenting with severe abdominal pain
• Dermal vein thrombosis: Painful skin lesions

Clinical Recognition

PNH should be suspected in patients with stroke (particularly cerebral venous thrombosis) who have:

• Hemolytic anemia with negative direct Coombs test
• Pancytopenia or history of aplastic anemia/MDS
• Dark or red-colored urine, classically worse in the morning
• Recurrent abdominal pain or thrombosis in unusual sites
• Elevated LDH with low haptoglobin (intravascular hemolysis)
• Iron deficiency despite hemolysis (urinary iron loss)

Diagnosis: Flow cytometry demonstrating deficiency of GPI-anchored proteins (CD55, CD59) on red blood cells and granulocytes. FLAER (fluorescent aerolysin) assay on granulocytes and monocytes is most sensitive.

Treatment and Thrombosis Prevention

Complement inhibitors have transformed PNH management and dramatically reduce thrombotic risk:

• Eculizumab: Terminal complement inhibitor (anti-C5 monoclonal antibody); reduces thrombosis risk by 85–90% and is the standard of care
• Ravulizumab: Long-acting anti-C5 antibody with 8-week dosing interval; similar efficacy to eculizumab
• Pegcetacoplan: Proximal complement inhibitor (C3) for patients with inadequate response to C5 inhibition

Anticoagulation is indicated for acute thrombosis and often continued long-term given high recurrence risk. In patients on complement inhibitors, the role of primary anticoagulation prophylaxis is less clear, though many experts recommend it for patients with large PNH clones (>50%).

Heparin-Induced Thrombocytopenia

Pathophysiology and Clinical Presentation

Heparin-induced thrombocytopenia (HIT) is an immune-mediated prothrombotic disorder caused by antibodies against complexes of platelet factor 4 (PF4) and heparin. Despite causing thrombocytopenia, HIT is paradoxically associated with a high risk of thrombosis — earning it the designation "HIT with thrombosis" (HITT) when clots occur.

HIT typically develops 5–10 days after heparin exposure (or earlier with prior heparin exposure within 100 days). The thrombotic risk is substantial:

• Venous thrombosis: DVT, PE (more common)
• Arterial thrombosis: Stroke, limb ischemia, MI (less common but devastating)
• Overall thrombosis rate: 30–50% if untreated
• Stroke specifically: ~3–5% of HIT patients with thrombosis

Diagnosis

The 4Ts Score is used for pretest probability assessment:

Interpretation: 0–3 points = low probability (HIT rare); 4–5 = intermediate; 6–8 = high probability

Laboratory confirmation:

• PF4/heparin ELISA: High sensitivity (~99%), moderate specificity; negative test essentially rules out HIT
• Serotonin release assay (SRA): Gold standard functional assay; highly specific but not widely available

Management

Alternative anticoagulants for HIT:

• Argatroban: Direct thrombin inhibitor; hepatically cleared; preferred in renal impairment; requires aPTT monitoring
• Bivalirudin: Direct thrombin inhibitor; shorter half-life; often used in PCI settings
• Fondaparinux: Factor Xa inhibitor; does not cross-react with HIT antibodies; increasingly used off-label; no monitoring required
• DOACs: Emerging data support use after initial parenteral therapy and platelet recovery; rivaroxaban and apixaban most studied

Continue anticoagulation for at least 4 weeks (HIT without thrombosis) or 3–6 months (HITT). Transition to warfarin only after platelet recovery, with overlap of parenteral agent until INR therapeutic.

Hereditary Thrombophilias

Overview

Inherited thrombophilias are genetic disorders that predispose to venous thromboembolism. Their association with arterial stroke is considerably weaker than with VTE, and routine testing in unselected stroke patients is not recommended. However, testing becomes relevant in specific clinical scenarios.

The major inherited thrombophilias include:

• Factor V Leiden (activated protein C resistance): Most common inherited thrombophilia (~5% of Caucasians); 3–7× increased VTE risk; minimal arterial stroke association
• Prothrombin G20210A mutation: Present in ~2% of Caucasians; 2–3× increased VTE risk
• Protein C deficiency: 0.2–0.4% prevalence; 7–10× increased VTE risk
• Protein S deficiency: Similar prevalence and risk to protein C deficiency
• Antithrombin III deficiency: Rarest but highest risk (15–20× VTE risk)
• Elevated Factor VIII: Associated with both VTE and possibly arterial thrombosis

Management Considerations

For patients with cryptogenic stroke and confirmed hereditary thrombophilia without concurrent VTE, antiplatelet therapy is reasonable (AHA Class 2a, LOE C-LD). Whether anticoagulation provides additional benefit over antiplatelet therapy in this population is unknown.

For patients with thrombophilia and documented VTE or paradoxical embolism through PFO, anticoagulation is typically indicated based on VTE guidelines. PFO closure may also be considered in appropriate candidates, though the interaction between PFO closure and thrombophilia status requires individualized decision-making.

Testing considerations: Testing should be deferred 4–6 weeks after acute stroke or VTE, as acute illness, anticoagulation, and inflammation can affect protein C, protein S, and antithrombin levels. Warfarin decreases protein C and S; DOACs do not affect these assays but can interfere with lupus anticoagulant testing.

Sickle Cell Disease

Epidemiology and Pathophysiology

Sickle cell disease (SCD) is the most common cause of stroke in children. Without primary prevention, approximately 11% of patients with HbSS will have an overt stroke by age 20, with silent cerebral infarcts affecting an additional 20–35%. The risk extends into adulthood, with stroke incidence remaining elevated throughout life.

Stroke mechanisms in SCD include:

• Large artery arteriopathy: Intimal hyperplasia and progressive stenosis of internal carotid and middle cerebral arteries
• Moyamoya syndrome: Develops in approximately 30% of patients with SCD-related arteriopathy
• Small vessel disease: Border zone infarcts from impaired perfusion
• Hemorrhagic stroke: Accounts for approximately one-third of strokes in adults with SCD, often from moyamoya-related aneurysms

Primary Prevention: TCD Screening

Transcranial Doppler (TCD) screening revolutionized stroke prevention in children with SCD. The STOP trial (1998) demonstrated that chronic transfusion therapy in children with elevated TCD velocities (≥200 cm/s in ICA or MCA) reduced stroke risk by 92% compared to standard care.

The STOP II trial (2005) showed that discontinuing transfusions after TCD normalization led to reversion to high-risk velocities and stroke occurrence, establishing that transfusion must be continued indefinitely once initiated for primary prevention. However, the TWiTCH trial (2016) demonstrated that hydroxyurea can replace transfusions for primary stroke prevention in children with SCD and abnormal TCD who have been on transfusions with stable, low TCD velocities — providing an alternative for those who achieve disease control.

For silent cerebral infarcts detected on MRI, the SIT trial (2014) showed that chronic transfusion reduced recurrence of silent infarcts compared to observation (6% vs 14%), though the clinical significance of preventing silent infarcts remains debated.

Myeloproliferative Neoplasms

Overview

Myeloproliferative neoplasms (MPNs) are clonal hematopoietic stem cell disorders characterized by overproduction of mature blood cells. The classic BCR-ABL-negative MPNs — polycythemia vera (PV), essential thrombocythemia (ET), and primary myelofibrosis (PMF) — are associated with increased thrombotic risk, including stroke.

Driver mutations include JAK2 V617F (present in ~96% of PV and ~55% of ET), CALR (~25% of ET), and MPL (~3% of ET). These mutations contribute to constitutive activation of JAK-STAT signaling and clonal expansion.

Polycythemia Vera

Polycythemia vera is characterized by clonal erythrocytosis with variable leukocytosis and thrombocytosis. Arterial thrombotic events, including stroke and myocardial infarction, are the leading cause of morbidity and mortality. Risk factors for thrombosis include age >60 years and prior thrombotic events.

Two landmark trials established the evidence base for PV management:

CYTO-PV: This Italian RCT randomized 365 patients with JAK2-positive PV to intensive hematocrit control (<45%) vs less intensive control (45–50%). At median follow-up of 31 months, the intensive control group had significantly fewer cardiovascular deaths and major thrombotic events (2.8% vs 9.8%; HR 3.91, P=0.007). This trial established the 45% hematocrit threshold as evidence-based rather than arbitrary.

ECLAP: This European RCT randomized 518 patients with PV to low-dose aspirin (100mg) vs placebo. Aspirin reduced the composite of nonfatal MI, nonfatal stroke, PE, major venous thrombosis, or cardiovascular death (RR 0.40, 95% CI 0.18–0.91, P=0.03) without significantly increasing major bleeding.

Essential Thrombocythemia

Essential thrombocythemia is characterized by sustained thrombocytosis (platelets ≥450 × 10⁹/L) without features of other MPNs. Both arterial and venous thrombotic events occur, along with paradoxical bleeding risk at very high platelet counts.

The PT-1 trial (UK MRC, 2005) compared hydroxyurea + aspirin vs anagrelide + aspirin in 809 high-risk ET patients. Hydroxyurea was superior, with the anagrelide group having increased arterial thrombosis (P=0.004), serious hemorrhage (P=0.008), and transformation to myelofibrosis (P=0.01). However, anagrelide was associated with fewer venous thrombotic events (P=0.006). This established hydroxyurea as first-line cytoreductive therapy for high-risk ET.

The ANAHYDRET study (2013) using WHO diagnostic criteria found anagrelide non-inferior to hydroxyurea, suggesting diagnostic criteria may affect treatment response. Current practice favors hydroxyurea as first-line, with anagrelide reserved for hydroxyurea-intolerant patients.

Cancer-Associated Stroke

Mechanisms and Recognition

Cancer increases stroke risk through multiple mechanisms:

• Hypercoagulability (Trousseau syndrome): Mucin-secreting adenocarcinomas (lung, GI, pancreatic) are most thrombogenic; often presents with markedly elevated D-dimer and multiterritory infarcts
• Nonbacterial thrombotic endocarditis (NBTE/marantic endocarditis): Sterile vegetations on heart valves; requires echocardiography (often TEE) for detection
• Direct tumor effects: Compression or invasion of vessels; intracardiac tumors
• Treatment-related: Radiation arteriopathy; chemotherapy effects (L-asparaginase, cisplatin, bevacizumab)
• Paradoxical embolism: Cancer-associated VTE crossing through PFO

Cancer-associated stroke may be the presenting manifestation of occult malignancy in 5–10% of ESUS patients. Clinical clues include: markedly elevated D-dimer (>2–3× ULN), multiterritory infarcts, unexplained weight loss, and anemia or thrombocytosis.

Management Approach

Anticoagulation is the mainstay for cancer-associated thrombosis. Several RCTs have compared DOACs to LMWH for cancer-associated VTE:

• Hokusai VTE Cancer (2018): Edoxaban vs dalteparin; edoxaban non-inferior for recurrent VTE but with higher GI bleeding in GI malignancies
• SELECT-D(2018): Rivaroxaban vs dalteparin; lower VTE recurrence with rivaroxaban but higher clinically relevant non-major bleeding
• CARAVAGGIO: Apixaban vs dalteparin; apixaban non-inferior without significantly increased GI bleeding, suggesting it may be preferred in non-GI cancers

For patients with atrial fibrillation and active cancer, DOACs are reasonable over warfarin based on the AF population data and cancer-VTE trials (AHA Class 2a).

Thrombotic Microangiopathies

Thrombotic Thrombocytopenic Purpura (TTP)

TTP is characterized by microangiopathic hemolytic anemia (MAHA), thrombocytopenia, and microvascular thrombosis due to severe ADAMTS13 deficiency (<10% activity). The deficiency leads to accumulation of ultra-large von Willebrand factor multimers that cause platelet aggregation in the microvasculature.

Neurological manifestations are present in 40–80% of cases and include confusion, seizures, focal deficits, and stroke. The classic pentad (MAHA, thrombocytopenia, neurological symptoms, renal dysfunction, fever) is present in only 7–10% of patients at diagnosis.

Treatment: Plasma exchange (TPE) is the cornerstone of therapy. The HERCULES trial (2019) demonstrated that caplacizumab — an anti-vWF nanobody — added to TPE and immunosuppression accelerated platelet normalization and reduced the composite of TTP-related death, recurrence, or thromboembolic events by 74% compared to placebo. The earlier TITAN trial (2016) provided phase 2 evidence supporting these findings.

Hemolytic Uremic Syndrome (HUS)

HUS shares features with TTP but typically has more prominent renal involvement and normal ADAMTS13 activity. Typical HUS is caused by Shiga toxin-producing E. coli (STEC-HUS, usually diarrhea-associated), while atypical HUS (aHUS) results from complement dysregulation.

Neurological involvement is less common than in TTP but can occur, particularly in STEC-HUS. Treatment is supportive for typical HUS; complement inhibitors (eculizumab, ravulizumab) are effective for aHUS.

Disseminated Intravascular Coagulation (DIC)

DIC is a consumptive coagulopathy with simultaneous thrombosis and hemorrhage, typically triggered by sepsis, malignancy, or obstetric complications. Cerebral manifestations include microinfarcts, hemorrhage, or both. Treatment focuses on the underlying cause; factor replacement for bleeding and anticoagulation for thrombosis may both be needed.

Hyperviscosity Syndromes

Hyperviscosity can cause stroke through impaired microcirculatory flow. Causes include:

• Waldenström macroglobulinemia: IgM paraprotein increases serum viscosity; symptoms typically occur when viscosity exceeds 4 centipoise
• Multiple myeloma: Less commonly causes hyperviscosity than Waldenström due to lower molecular weight of IgG/IgA
• Cryoglobulinemia: Type I (monoclonal) can cause hyperviscosity; mixed types (II/III) cause vasculitis (see dedicated section below)
• Polycythemia: Discussed above under MPNs

Symptoms include visual disturbances, headache, confusion, and stroke. Treatment is plasmapheresis for acute symptoms, with disease-directed therapy for the underlying condition.

POEMS Syndrome

POEMS syndrome (Polyneuropathy, Organomegaly, Endocrinopathy, M-protein, Skin changes) is a rare paraneoplastic syndrome associated with plasma cell dyscrasias. While primarily recognized for its neurological manifestations (progressive demyelinating polyneuropathy), POEMS carries significant thrombotic risk.

Stroke mechanisms in POEMS:

• Hypercoagulability from elevated VEGF and cytokines
• Endothelial dysfunction
• Hyperviscosity (less common than in Waldenström)
• Arterial and venous thrombosis both described

Clinical clues suggesting POEMS:

• Progressive sensorimotor polyneuropathy (often the dominant feature)
• Organomegaly (hepatomegaly, splenomegaly, lymphadenopathy)
• Endocrine abnormalities (hypogonadism, hypothyroidism, diabetes)
• Monoclonal protein (usually IgA or IgG lambda)
• Skin changes (hyperpigmentation, hypertrichosis, hemangiomas)
• Papilledema, extravascular volume overload
• Elevated VEGF levels (often markedly elevated)

Diagnosis requires the presence of both mandatory criteria (polyneuropathy + monoclonal plasma cell disorder) plus at least one major criterion (Castleman disease, sclerotic bone lesions, elevated VEGF) and one minor criterion.

Treatment: Directed at the underlying plasma cell clone — radiation for localized disease, systemic chemotherapy (often with autologous stem cell transplant) for disseminated disease. Thromboprophylaxis should be considered given elevated thrombotic risk.

Cryoglobulinemia

Overview

Cryoglobulinemia is a disorder characterized by circulating immunoglobulins (cryoglobulins) that precipitate at low temperatures and dissolve on rewarming, leading to vascular occlusion, immune-complex vasculitis, or both. Although stroke is uncommon, cryoglobulinemia should be considered in younger patients with stroke plus systemic inflammatory features, abnormal complement levels, or evidence of hyperviscosity or small-vessel vasculitis.

Cryoglobulinemia is classified into three types (Brouet classification) based on immunoglobulin composition. The underlying causes differ by type and are clinically important because management is cause-directed.

Classification, Mechanism, and Important Causes

Stroke-Relevant Mechanisms

• Type I: Hyperviscosity and vascular occlusion can cause ischemia (including stroke), especially with high cryocrit or concomitant paraproteinemia.
• Type II/III: Immune-complex vasculitis can produce multifocal ischemic injury, small/medium vessel stenoses, and systemic inflammatory manifestations that help distinguish it from atherosclerosis.

Clinical Clues

• Palpable purpura (often lower extremities), arthralgias, fatigue
• Peripheral neuropathy (painful, distal, sensory > motor)
• Renal involvement (hematuria/proteinuria; membranoproliferative GN pattern)
• Low complement (especially low C4) and often positive rheumatoid factor (mixed cryoglobulinemia)
• History of hepatitis C (key association), autoimmune disease, or B-cell lymphoproliferative disorder

Diagnosis

• Serum cryoglobulins (pre-analytic handling matters: keep sample warm until serum separated to avoid false negatives)
• Complement levels (C3/C4), rheumatoid factor, hepatitis C antibody + RNA, hepatitis B testing, HIV when indicated
• Urinalysis and kidney function tests when systemic disease suspected
• Consider tissue biopsy (skin/kidney) if vasculitis/organ-threatening disease suspected

Management Principles

Diagnostic Approach

The extent of hematologic workup depends on clinical suspicion based on patient age, presentation, and initial laboratory findings.

Summary: Guideline Recommendations

Trial Comparison Table

References

1. Kleindorfer DO, et al. 2021 Guideline for the Prevention of Stroke in Patients With Stroke and Transient Ischemic Attack. Stroke. 2021;52:e364–e467.
2. Prabhakaran S, et al. 2024 Guideline for the Early Management of Patients With Acute Ischemic Stroke. Stroke. 2024.
3. Bushnell CD, et al. 2024 Guideline for the Primary Prevention of Stroke. Stroke. 2024.
4. Pengo V, et al. Rivaroxaban vs warfarin in high-risk patients with antiphospholipid syndrome (TRAPS). Blood. 2018;132:1365–1371.
5. Pengo V, et al. Trial of Rivaroxaban in AntiPhospholipid Syndrome: Two-year outcomes. Thromb Haemost. 2021;121:111–116.
6. Ordi-Ros J, et al. Rivaroxaban Versus Vitamin K Antagonist in Antiphospholipid Syndrome (ASTRO-APS). Ann Intern Med. 2022.
7. Cohen H, et al. Rivaroxaban versus warfarin in antiphospholipid syndrome (RAPS). Lancet Haematol. 2016;3:e426–436.
8. Hillmen P, et al. The complement inhibitor eculizumab in paroxysmal nocturnal hemoglobinuria. N Engl J Med. 2006;355:1233–1243.
9. Hillmen P, et al. Effect of eculizumab on hemolysis and transfusion requirements in patients with paroxysmal nocturnal hemoglobinuria. N Engl J Med. 2004;350:552–559.
10. Lee JW, et al. Ravulizumab (ALXN1210) vs eculizumab in adult patients with PNH naive to complement inhibitors: the 301 study. Blood. 2019;133:530–539.
11. Cuker A, et al. American Society of Hematology 2018 guidelines for management of venous thromboembolism: heparin-induced thrombocytopenia. Blood Adv. 2018;2:3360–3392.
12. Greinacher A. Heparin-induced thrombocytopenia. N Engl J Med. 2015;373:252–261.
13. Adams RJ, et al. Prevention of a first stroke by transfusions in children with sickle cell anemia (STOP). N Engl J Med. 1998;339:5–11.
14. Adams RJ, Brambilla D. Discontinuing prophylactic transfusions used to prevent stroke in sickle cell disease (STOP II). N Engl J Med. 2005;353:2769–2778.
15. Ware RE, et al. Hydroxycarbamide versus chronic transfusion for maintenance of transcranial doppler flow velocities in children with sickle cell anaemia (TWiTCH). Lancet. 2016;387:661–670.
16. Marchioli R, et al. Cardiovascular Events and Intensity of Treatment in Polycythemia Vera (CYTO-PV). N Engl J Med. 2013;368:22–33.
17. Landolfi R, et al. Efficacy and safety of low-dose aspirin in polycythemia vera (ECLAP). N Engl J Med. 2004;350:114–124.
18. Harrison CN, et al. Hydroxyurea compared with anagrelide in high-risk essential thrombocythemia (PT-1). N Engl J Med. 2005;353:33–45.
19. Scully M, et al. Caplacizumab Treatment for Acquired Thrombotic Thrombocytopenic Purpura (HERCULES). N Engl J Med. 2019;380:335–346.
20. Raskob GE, et al. Edoxaban for the Treatment of Cancer-Associated Venous Thromboembolism (Hokusai VTE Cancer). N Engl J Med. 2018;378:615–624.
21. Agnelli G, et al. Apixaban for the Treatment of Venous Thromboembolism Associated with Cancer (CARAVAGGIO). N Engl J Med. 2020;382:1599–1607.
22. Dispenzieri A. POEMS syndrome: 2019 Update on diagnosis, risk-stratification, and management. Am J Hematol. 2019;94:812–827.
23. Cacoub P, et al. Cryoglobulinemia Vasculitis. Am J Med. 2015;128:950–955.