Subscribe to eTOC


I am writing to follow up on “Ariel Sharon's Stroke: The Treatment He Received – and Why” (May 2, page 8). Neurologists and other physicians at the Hadassah Hospital in Jerusalem have been criticized for the care provided to Israeli Prime Minister Ariel Sharon. I was in frequent contact with Sharon's neurologists during the care for his strokes. I write with the permission of Sharon's neurologist to share the essential facts about his condition and his treatment.

The initial neurological event was the sudden onset of loss of speech. When speech returned, Sharon was aphasic. There were transient slight right face and limb abnormalities. The neurological symptoms and signs cleared rapidly. He had not had past strokes or transient ischemic attacks (TIAs), but was being followed for a monoclonal gammopathy.

Sharon's blood pressure was normal. Diffusion-weighted (DWI) and FLAIR MRI showed a small region of abnormality in the parasylvian cortex of the left cerebral hemisphere. There were no recent or old brain infarcts seen on T2-weighted MRI. T2*-weighted MRI (gradient-echo images) showed several micro-bleeds. Also seen were several small white matter hyperintensities (UBOs). Vascular imaging showed no important neck or intracranial arterial abnormalities in the left internal carotid artery or its tributaries. Electrocardiographic studies revealed normal sinus rhythm and gave no evidence of recent myocardial infarction. Transesophageal echocardiography (TEE) showed spontaneous right-to-left shunting through a large patent foramen ovale (PFO) and an atrial septal aneurysm. No valvular lesions were seen and myocardial function was normal. There were no significant plaques in the ascending portion and arch of the aorta.

The brain lesion was certainly embolic. The sudden onset, rapid clearing, cortical location, and clean arteries leading to the infarcted region make this interpretation quite clear. The only potential source shown was a high-volume shunt at the atrial level and an atrial septal aneurysm.

The presence of a monoclonal gammopathy and the onset of thrombus formation at his age raised the question of a neoplasm with a change in coagulation functions as a component of the thromboembolic event. Acute anticoagulation was indicated to prevent further potentially disabling thromboembolism. The presence of micro-bleeds on gradient-echo MRI raised the question of asymptomatic cerebral amyloid angiopathy (CAA). Although no study has shown an important risk of intracranial bleeding in patients with CAA, long-term anticoagulants (and antiplatelet agents) posed a risk of bleeding into the brain. Considering the benefit versus the risks of various therapeutic strategies, his physicians opted to give a two-week course of low-molecular-weight heparin (enoxyparin), which was to be followed by percutaneous closure of the PFO and subsequent discontinuation of anticoagulation. I feel that this therapeutic decision was eminently reasonable although others might have chosen alternative strategies.

Sharon's physicians followed his anticoagulation by frequent assessments of Factor Xa activity, which always remained within target ranges. A day before the planned PFO closure, Sharon was watching news on television, became upset, and developed a headache. Soon there was an acute decrease in consciousness. He was rushed to Hadassah Hospital. Brain imaging showed a large acute right posterior temporal lobar hemorrhage.

Anticoagulation was reversed with Factor 7 (Novo-7). The hematoma was drained surgically and a ventricular drain was placed. Unfortunately, he remained poorly responsive to stimuli despite the absence of a midline shift or continued pressure effects. Follow-up brain imaging showed bilateral medial thalamic infarcts, a complication of the pressure shifts that accompanied his intracerebral hemorrhage that had diminished the vascular supply to the thalamus provided by penetrating branches at the top of the basilar artery.

The treatment of his lobar hemorrhage was rapid, efficient, and up-to-date. Unfortunately, the sizable brain hemorrhage had accumulated so quickly that little could be done to prevent herniation and displacement of branches of the basilar artery. The location of the hemorrhage, lobar at the surface and right temporal, and the presence of significant mass effect made this hemorrhage the prototypic ICH to decompress. In my experience and that of others, temporal lobe hematomas often produce early herniation. Moreover, good survival is common if pressure effects can be controlled since the deficit in patients with right temporal lobe lesions is often limited to a left hemianopia. I vigorously support the decision to drain this hemorrhage.

I find it difficult using hindsight to criticize Sharon's physicians. The decisions made and the strategies employed were eminently reasonable. Unfortunately, the best laid plans of mice and men oft go astray.

Louis R. Caplan, MD

Harvard Medical School, Boston, MA