The purpose of this study is to determine if giving blood transfusions to anemic patients with subarachnoid hemorrhage will reduce their chances of having a stroke from vasospasm.

Each year, approximately 30,000 people suffer aneurysmal subarachnoid hemorrhage (SAH) in the United States. The most common and potentially treatable cause of secondary neurological injury in this population is delayed ischemic deficit (DID). As the name implies, this phenomenon is fundamentally a reduction of cerebral blood flow (CBF) and oxygen delivery below critical ischemic thresholds, occurring days after the onset of hemorrhage. Three inter-related physiological processes appear to be involved in the reduced oxygen delivery: severe narrowing of intracranial arteries (arterial vasospasm), intravascular volume depletion and a loss of normal autoregulatory function in the distal circulation. DID occurs in up to 40% of patients surviving SAH. One third of these patients will die from this phenomenon and another third will be left with permanent and severe disability.

The optimal treatment of vasospasm is not known. Medical management involves a number of hemodynamic manipulations and is usually referred to as hypervolemic, hypertensive, hemodilution (or Triple-H) therapy. Our knowledge of the physiological impact of the individual components or a combination of them is limited and clinical efficacy has not been established. The information gained in this study has great potential to advance our knowledge regarding the role of hematocrit in the optimal treatment of this often-devastating condition.

Changes in hematocrit can potentially impact brain oxygen delivery in two ways. First, there is a linear relationship between hemoglobin and arterial oxygen content, lower hematocrit less oxygen. Thus at a given CBF lowering hematocrit reduces brain oxygen delivery. Fortunately, the brain responds to this by increasing blood flow to restore oxygen delivery to baseline levels. Additionally, lowering hematocrit has another effect, it reduces viscosity which in and of itself can raise CBF, but in a non-linear way. It is the relative contribution of these two effects that will determine if oxygen delivery improves.

It has been proposed by largely on theoretical consideration that the "optimal" hematocrit that achieves this balance is 30-35%. Yet no study to date has assessed the relationship between hematocrit and oxygen delivery in SAH patients. Other observations, however, suggest that higher hemoglobin levels in SAH patients was associated with better outcomes. Finally another retrospective review suggested that receiving transfusions increased risk for vasospasm and poor outcome after subarachnoid hemorrhage.

We are proposing to begin a series of studies to determine the appropriate management of hematocrit in SAH patients. The first is to define the appropriate physiologic response (cerebral oxygen delivery and metabolism) to a change in hematocrit. Then the "optimal" hematocrit can be defined. Only then will we be able to properly design clinical outcome trials.

Inclusion Criteria:

1. Aneurysmal SAH confirmed by angiography

2. Hemoglobin < 12.5 gm/dl

3. One of the following:

   • Considered at increased risk for vasospasm by care team
   • Angiographic vasospasm
   • Delayed ischemic deficit

4. Able to be studied within 2 weeks after subarachnoid hemorrhage

Exclusion Criteria:

1. Active Coronary Artery Disease
2. Severe congestive heart failure
3. Jehovah's witness
4. Unable to obtain appropriately matched blood
5. Other contraindications for transfusion
6. Pregnancy