Current status of hemodilution in acute cerebral ischemia.

T 4 C r | ^ hinning the blood" is such an intuitively reasonable and understandable treatment for ischemic stroke that most clinicians have invoked this phrase to explain to families one of the ways of possibly treating their recently hemiplegic relative. The Scandinavian Stroke Study Group reports in this issue of Stroke, however, that hemodilution (or probably any treatment) does not work when applied by general physicians seeing patients several days after a stroke. This is a clinically important finding but it should not be a surprising one considering the conservative design of the study. In their thoughtful discussion, the authors address many of the questions raised by their results, but in addition, it should be emphasized that while it is necessary to validate any therapy in the "real world" and not in just the specialized confines of tertiary referral centers, the delay in treatment and relatively indiscriminate selection of patients characteristic of this study contradicts the emerging concept, based on modern studies of cerebral blood flow (CBF) and metabolism, that therapy of cerebral ischemia should be instituted early and tailored to subgroups of patients. The concept of hemodilution has been extensively reviewed in this journal and it is based on the close correlation of hematocrit and whole blood viscosity. Lowering hematocrit results in reduced viscosity and less resistance to blood flow, especially in regions of relative stasis. Arguments in favor of hemodilution therapy for cerebral ischemia are based on observations that hemodilution increases CBF and has improved clinical outcome in animal and preliminary human studies."* Hypervolemic hemodilution is so popular for treating ischemic complications of vasospasm after subarachnoid hemorrhage that most clinicians dealing with this condition are reluctant to perform a controlled trial including untreated patients. Arguments against hemodilution are based on the poor correlation of CBF and outcome after stroke and the generally negative results obtained with other therapies aimed at raising CBF. Furthermore, improved