Secondary Logo

Journal Logo

Institutional members access full text with Ovid®

Complications associated with anemia and blood transfusion in patients with aneurysmal subarachnoid hemorrhage

Kramer, Andreas H. MD, MSc, FRCPC; Gurka, Matthew J. PhD; Nathan, Bart MD; Dumont, Aaron S. MD; Kassell, Neal F. MD; Bleck, Thomas P. MD, FCCM

doi: 10.1097/CCM.0b013e31817c1095
Clinical Investigations

Objective: Patients with subarachnoid hemorrhage (SAH) frequently develop delayed cerebral ischemia and may be especially vulnerable to the effects of anemia. However, the potentially harmful effects of allogeneic red blood cells are increasingly being recognized. The optimal transfusion threshold is unknown, but current practice most often uses a liberal approach. We assessed the association between anemia or transfusion and subsequent adverse outcomes.

Design: Retrospective cohort study.

Setting: Neuroscience intensive care unit of a university hospital.

Patients: A total of 245 consecutive patients with aneurysmal SAH.

Interventions: None.

Measurements: Logistic regression models were used to adjust for baseline differences in age, severity of neurologic impairment, and amount of blood on computed tomography. Patients were dichotomized based on whether symptomatic vasospasm was diagnosed.

Main Results: Individually, anemia (nadir hemoglobin <10 g/dL) and the use of transfusions were both associated with the combined outcome of death, severe disability, or delayed infarction (odds ratio [OR] for anemia, 2.7; 95% confidence interval [CI] 1.5–5; p < .01; OR for transfusion, 4.8; 95% CI, 2.5–9.1; p < .01). When both variables were together introduced into a logistic regression model, only transfusion remained significantly predictive (OR, 4.3; 95% CI, 1.5–9.3; p < .01). The relationship between anemia and adverse outcomes was stronger among patients diagnosed with vasospasm, whereas for transfusion, it was stronger among patients without vasospasm. Transfusion also was associated with the development of nosocomial infections (OR, 3.2; 95% CI, 1.7–5.5; p < .01). There was no statistically significant difference in complications based on the duration of blood storage before transfusion.

Conclusions: Although anemia is predictive of adverse outcomes in patients with SAH, this observation cannot be considered justification for a liberal transfusion strategy. Appropriate transfusion thresholds may vary depending on the presence or absence of clinical vasospasm. Randomized trials that compare liberal and restrictive transfusion strategies in patients with SAH are needed.

From the Departments of Critical Care Medicine and Clinical Neurosciences (AHK), University of Calgary, Calgary, AB, Canada; Division of Biostatistics and Epidemiology (MJG), Department of Public Health Sciences, University of Virginia, Charlottesville, VA; Departments of Neurology and Internal Medicine (BN), University of Virginia, Charlottesville, VA; Department of Neurological Surgery (ASD, NFK), University of Virginia, Charlottesville, VA; The Ruth Cain Ruggles Chairman of Neurology (TPB), Evanston Northwestern Healthcare, Department of Neurology, Neurological Surgery, and Medicine, Northwestern University Feinberg School of Medicine, Evanston, IL.

The authors have not disclosed any potential conflicts of interest.

Supported, in part, by the Louise Nerancy endowment of The University of Virginia.

For information regarding this article, E-mail:

This study was performed at the University of Virginia.

© 2008 by the Society of Critical Care Medicine and Lippincott Williams & Wilkins