Sickle Cell Disease and Stroke
A 24-year-old man with sickle cell disease (SCD) was noncompliant with medication and presented in pain crisis with severe abdominal and back pain. He had no neurological complaints or abnormalities on neurological examination. Later, he was transferred to the Medical Intensive Care Unit with lethargy, reduced hemoglobin, and impaired kidney, liver, and heart function. The hemoglobin S percentage (HbS%) was 84, and he was transfused to reduce the HbS% to 55; however, his multiorgan failure worsened.
The next day he developed left arm weakness, but the time of onset was unclear. His examination showed a left hemiparesis and hemisensory loss. His National Institutes of Health stroke score was 10, and computed tomography of the head was unremarkable. Thrombolytics were not given because of the unknown onset. MRI of the brain showed scattered diffusion-positive lesions throughout both cerebral hemispheres. Magnetic resonance angiography of the head and neck showed no large vessel occlusion. Transesophageal echocardiogram was negative for a cardioembolic source of emboli, and telemetry records were unremarkable. He started aspirin 81 mg daily and underwent exchange transfusion. His HbS% stabilized at 23 to 25, and his neurological examination improved.
A few days later, he developed severe headache and had a generalized tonic clonic seizure. Computed tomography now showed a left frontal subarachnoid hemorrhage. He was treated with levetiracetam and underwent magnetic resonance angiography, magnetic resonance venography, and cerebral angiogram, which showed no aneurysms or other vascular malformations. His coagulation panel was unremarkable, but the HbS% was 16. Of note, imaging requiring iodinated dye was delayed until after his crisis resolved because of potential vasospastic complications from the contrast material during crisis.
The incidence of ischemic and hemorrhagic strokes is increased in adults with SCD relative to the general population. The prevalence of stroke is 3.75% in patients with SCD.1 Eleven percent of patients have a clinically evident stroke by 20 years of age and 24% by 45 years of age.1 Recurrent cerebral infarction occurs in two thirds of patients, usually within 2 to 3 years.2 The risk of stroke is highest during the first decade of life, drops in the second decade, and rises again after 29 years of age when hemorrhagic strokes become more common.1 Because more children with SCD survive, adult neurologists must become familiar with these complications. There is little evidence to guide management in adults, and most recommendations are extrapolated from pediatric studies.
The pathophysiology of SCD involves not only red blood cells (RBCs) but also the vascular endothelium, inflammation, and coagulopathy. The gene encoding the β-globulin chain of hemoglobin is mutated in SCD, causing substitution of valine for glutamic acid in the sixth position of the β-globulin chain. This substitution makes deoxygenated hemoglobin poorly soluble and prone to polymerization and vessel occlusion.3 Furthermore, sickled hemoglobin distorts the cell membrane, causing ion imbalances, dehydrating RBCs and making them stiff, and facilitating occlusion. Patients with SCD have chronic inflammation with high levels of interleukins, chemokines, and cytokines. Increased expression of adhesion molecules enhances attachment of sickled RBCs to the endothelium.3 Nitric oxide is also depleted, furthering ischemia via vasoconstriction. Finally, increased thrombin and reduced antithrombotic protein S and C induce a hypercoagulable state.3 All of these processes increase vascular injury and stroke.
Stroke risk factors in SCD differ for each stroke type. Clinical ischemic strokes in patients with SCD are associated with prior transient ischemic attack, increased systolic blood pressure, acute coronary syndrome, prior silent infarcts, and nocturnal hypoxemia.4 However, hemorrhagic strokes are associated with older age, low steady state hemoglobin, low steady state leukocytes count,3 transfusion within 2 weeks, and treatment with corticosteroids or nonsteroidal anti-inflammatory drugs within 2 weeks.4 Silent infarcts are associated with leukocytosis, seizures, lower baseline hemoglobin, male sex, and hypertension.4
Clinically apparent ischemic stroke, hemorrhagic stroke, and clinically silent infarcts are all increased in SCD. Ischemic strokes are secondary to vasculopathy and intracranial large artery stenosis.3 Hemorrhages account for one third of SCD-related stroke1 and may be subarachnoid, intraparenchymal, intraventricular, or a combination. Most intracranial hemorrhages in patients with SCD are associated with aneurysms, which tend to be multiple, and rupture despite smaller size.4 Clinically silent infarction is defined as increased T2 signal abnormalities on MRI without corresponding deficits. Twenty-two percentage of children with SCD between 6 and 19 years of age have silent infarcts, usually involving small vessels in watershed distributions. Diminished reserve and impaired cerebral perfusion pressure likely contribute to infarction.4
Moyamoya syndrome is present in 20% to 30% of patients with SCD undergoing cerebral angiogram5 and predisposes to both ischemic stroke and intracranial hemorrhage. Moyamoya disease is idiopathic, progressive stenosis of the terminal internal carotid arteries and their main branches. A compensatory collateral network develops at the base of the brain.5 On catheter angiography, this network appears as a puff of smoke or moyamoya in Japanese. Moyamoya syndrome describes radiographic features similar to idiopathic moyamoya disease that may coexist with other conditions such as SCD.
Stroke symptoms in SCD are similar to those in other patients. Patients should be evaluated for traditional risk factors that may coexist with SCD. Vascular imaging should be obtained to evaluate for aneurysms or moyamoya. In the case of hemorrhage, if computed tomographic angiography/magnetic resonance angiography is unrevealing, conventional angiography should be performed. Importantly, hyperosmolar intravenous contrast may increase the risk of ischemic stroke. Patients should be pretreated with intravenous hydration and transfusion to maintain the HbS% <20 to 50 before iodinated contrast.6
In addition to standard stroke care, patients with SCD and acute stroke should receive transfusion with a goal of decreasing the HbS% to <30.2 With simple transfusion, patients do not have their own blood removed but receive transfused blood. In exchange transfusion, the patient’s blood is removed and replaced with an equal volume of packed RBCs. Exchange transfusion can be either manual, in which whole blood is replaced with donor blood, or automated. In erythrocytapheresis, sickled RBCs are identified and removed by automated centrifugation and replaced with donor packed RBCs. A retrospective study suggests that exchange transfusion at the time of presentation is associated with lower stroke recurrence compared with simple transfusion.7 Blood products should be thoroughly matched for minor antigens, and leukocyte-filtered blood should be used. Importantly, packed RBCs should be screened for sickle hemoglobin. Unfortunately, recurrent transfusions increase the risk of iron overload, allergic reactions, infections, alloimmunization, and transfusion reactions.
The role of thrombolysis in patients with SCD is unclear. To our knowledge, there are no published data on thrombolytic therapy in patients with SCD. Most experts recommend that thrombolysis be considered in patients meeting criteria. The risk of hemorrhage is, at least theoretically, increased because of the propensity toward hemorrhage in SCD, and this additional risk should be discussed with patients and families.2
There are no data specifically guiding management of intracerebral hemorrhage in adults with SCD. Treatment does not differ from intracranial hemorrhage in other patients, and neurosurgical expertise should be sought for aneurysms and surgical decompression if warranted.8
The Stroke Prevention Trial in Sickle Cell Anemia (STOP) trial demonstrated that elevated transcranial doppler velocity can identify SCD children with a high risk of stroke and that exchange transfusion reduces that risk.9,10 Unfortunately, studies fail to show similar protection in adults.2 Increased transcranial doppler velocity identifies arterial stenosis but does not predict stroke.2 Because of the high risk of stroke in adults with SCD, aggressive identification and treatment of traditional risk factors are warranted. Alternative SCD therapies including hydroxyurea and hematopoietic stem cell transplantation can be considered. Secondary prevention of hemorrhage depends on identifying and treating underlying causes of hemorrhage, such as aneurysms, hypertension, and coagulopathies.
Cerebral revascularization may decrease the risk of recurrent stroke in patients with SCD with moyamoya disease. Two approaches are used for revascularization: direct and indirect revascularization. Direct vascularization involves anastomosing a branch of the superficial temporal artery to the brain (superficial temporal artery to middle cerebral artery bypass), and the indirect approach, for example, encephalomyosynangiosis, involves placing vascularized tissue in contact with the brain, leading to growth of new collaterals. Both approaches are effective in reducing the risk of stroke, but there are no randomized trials comparing the 2 directly; the choice between different procedures is often dependent on the preference of the surgeon.5
Sickle cell disease increases risk of ischemic stroke and intracranial hemorrhages. Stroke management is largely extrapolated from pediatric studies.
It is important to evaluate for and treat other potential causes of stroke.
The role of exchange transfusion in patients with stroke with sickle cell disease is based on pediatric studies. Data supporting this approach in adults are lacking.
Although exchange transfusion to maintain the hemoglobin S percentage <30 is common, alternative therapies such as hydroxyuria or bone marrow transplant can be considered.
- Received March 24, 2014.
- Revision received March 24, 2014.
- Accepted April 1, 2014.
- © 2014 American Heart Association, Inc.
- Ohene-Frempong K,
- Weiner SJ,
- Sleeper LA,
- Miller ST,
- Embury S,
- Moohr JW,
- et al
- Hulbert ML,
- Scothorn DJ,
- Panepinto JA,
- Scott JP,
- Buchanan GR,
- Sarnaik S,
- et al