Red blood cell transfusion is associated with infection and extracerebral complications after subarachnoid hemorrhage

Red blood cell transfusion (RBCT) is associated with medical complications in general medical and surgical patients. We examined the hypothesis that RBCT during intensive care unit (ICU) care is associated with medical complications after subarachnoid hemorrhage (SAH).

We retrospectively analyzed a prospective observational database containing 421 patients with SAH (mean age, 51.5 years; standard deviation, 14.6 years). Logistic regression models were used to adjust for age, admission hemoglobin (Hgb), clinical grade, average ICU Hgb, and symptomatic vasospasm.

Two hundred fourteen patients received an RBCT during their ICU stay. Medical complications were identified in 156 patients and were more common in those who received blood (46%) than in those who did not (29.8%) (P < .001). Major medical complications (cardiac, pulmonary, renal, or hepatic) occurred in 111 patients, and minor complications (eg, skin rash, deep vein thrombosis) occurred in 45 patients. Any non-central nervous system infection (n = 183; P < .001), including pneumonia (n = 103; P < .001) or septicemia (n = 36; P = .02), was more common with RBCT. Central nervous system infections (meningitis, cranial wound, n = 15) also were associated with RBCT (P = .03). Mechanically ventilated patients (n = 259) were more likely to have received an RBCT than those who did not (P < .001). When logistic regression was used to control for age, admission clinical grade and Hgb, average ICU Hgb, symptomatic vasospasm, and other admission variables associated with outcome, the following factors (odds ratio; 95% confidence interval) were associated with RBCT: any medical complication (1.8; 1.1-3.0), major medical complications (2.1; 1.2-3.7), any infection (2.8; 1.7-4.5), pneumonia (2.6; 1.5-4.7), septicemia (2.9; 1.2-6.8), and need for mechanical ventilation (2.8; 1.5-5.1).

These data suggest that RBCTs are associated with medical complications after SAH. However, the data do not infer causation, and further study is necessary to better define the indications for transfusion after SAH.