Park YH, Kim BJ, Kim JS, Yang MH, Jang MS, Kim N, et al. Impact of Both Ends of the Hemoglobin Range on Clinical Outcomes inAcute Ischemic Stroke. Stroke. 2013
From a mechanistic and pathophysiological standpoint, stroke could be viewed as a sudden blockage of oxygen delivery to the brain. Hence, the oxygen carrying capacity of blood could affect outcomes, and anemia could render the brain more vulnerable to ischemia. Park et al studied the effect of hemoglobin concentrations on clinical outcome, examining both ends of the concentration range and different time points of measurement.
A prospective registry identified 2681 patients with acute ischemic stroke. Hemoglobin initially, at nadir, time-averaged, discharge and hemoglobin drop was collected, and hemoglobin concentration was grouped in quintiles. Poor outcomes (higher mRS) and mortality at 3-months were related to the first quintiles (lower hemoglobin) of the initial, nadir, time-averaged, and discharge hemoglobin. With hemoglobin drop, mortality and higher mRS significantly increased only in the fifth quintile (largest hemoglobin drop). No significance was found with the remaining quintiles. They also accounted for blood pressure drop from baseline, and this didn’t change results. The authors concluded that poor outcome was related to the lower but not higher end of the hemoglobin range, regardless of when and how hemoglobin concentrations were measured.
So then, should we transfuse red blood cells in acute ischemic stroke? Is blood transfusion an option in our patients with lacunes or intracranial stenosis for example, having fluctuations despite hemodynamic optimization? If so, should we be restrictive or liberal in our transfusion thresholds? There are no randomized trials that have examined the outcome of transfusion in patients with acute ischemic stroke. Current practices extrapolate from studies in other populations such as myocardial infarction and subarachnoid hemorrhage. Even then, the data is not as straightforward and some evidence suggests transfusion itself being an independent predictor of worse outcomes.