Relationship of Hematocrit to Ischemic Stroke Outcome

University Of Virginia, Charlottesville VA

Investigators

Abstract

[unreadable] DESCRIPTION (provided by applicant): The purpose of this study is to examine the association between baseline hematocrit (HCT) level, stroke severity, and outcome following ischemic stroke. HCT is the main determinant of blood oxygen carrying capacity and blood viscosity. Central Nervous System (CNS) oxygen transport diminishes at high HCT levels due to increased blood viscosity and reduced cerebral blood flow. Oxygen transport is also reduced at low HCT levels due to low blood oxygen content. Several studies have found ischemic stroke more prevalent among persons with high HCT levels. Studies employing animal models of focal ischemia have shown a reduction in infarct size and increase in cerebral oxygen flow with experimental reductions in HCT to levels of 30 to 35%. Yet, multicenter clinical trials using hemodilution techniques designed to reduce HCT to similar target levels based on these animal studies have failed to demonstrate a clinical benefit and in some instances has resulted in worse outcome. Preliminary research suggests that HCT levels around 45% may be associated with optimal stroke outcome in humans, a value considerably higher than suggested by animal models. Clinical variables including stroke subtype and comorbidities may significantly influence this relationship. This study involves reanalysis of an existing clinical database, the Randomized Trial of Tirilazad Mesylate in Acute Stroke (RANTTAS). Specific aims are (1) to examine the association between initial HCT level and clinical/functional outcome as measured by NIHSS, Bl and COS scores following stroke; (2) to assess whether the association between HCT level and outcome is influenced by stroke subtype; (3) to examine whether certain co-variables influence this relationship; (4) to further study the association between HCT level and outcome using infarct volume as a secondary outcome measure. Rigorous statistical evaluation of the RANTTAS dataset will be carried out using statistical modeling techniques. These models allow for the evaluation of the effect of HCT in acute ischemic stroke adjusting for other possible effects, and for comparing the effect of HCT in different stroke subtypes. A number of models are proposed, each of which is intended to assess different aspects of the effect of HCT in this setting. Fitting these models will result in a more complete understanding of the association between HCT and stroke outcome. Findings could lead to the development of new interventions to improve stroke outcome. [unreadable] [unreadable]