From the Division of Neurosurgery, Department of Surgery, University of British Columbia, Vancouver, Canada (O.G.S.A., P.A.G.).
Division of Neurosurgery, Hospital for Sick Children (G.M.I.).
Stroke. 2018 Aug;49(8):1859-1865. doi: 10.1161/STROKEAHA.117.020260.
Background and Purpose- Anemia after aneurysmal subarachnoid hemorrhage is common and potentially modifiable. Here, we first evaluate the effect of anemia on neurological outcome and death and second, study the effects of packed red blood cell transfusion on outcome. Methods- A secondary analysis on 413 subjects in the CONSCIOUS-1 study (Clazosentan to Overcome Neurological Ischemia and Infarction Occurring After Subarachnoid Hemorrhage). Multivariable logistic regression identified independent risk factors for anemia and determined the effect of anemia on neurological outcome and death, while adjusting for selected covariates. Optimal predictive thresholds for hemoglobin levels were determined using receiver operating characteristic curve analysis. Finally, patients were pseudorandomized to transfusion using propensity score matching to study the effect of transfusions on outcome. Results- Anemia, defined as hemoglobin <10 g/dL, was present in 5% of patients at presentation, in 29% of patients after aneurysm securing (days 1-3), and in 32% of patients during the peak delayed cerebral ischemia risk period (days 5-9). Anemia after aneurysm securing (odds ratio, 1.96; 95% confidence interval, 1.07-3.59; P=0.03) and during the delayed cerebral ischemia window (odds ratio, 2.63; 95% confidence interval, 1.46-4.76; P=0.0014) was independently associated with poor neurological outcome. Anemia postaneurysm securing (odds ratio, 3.50; 95% confidence interval, 1.15-10.62; P=0.027) but not during the delayed cerebral ischemia window was associated with death. Using propensity score-matched cohorts, we found that transfusion of anemic patients did not improve long-term outcome (P=0.8) or mortality rates (P=0.9). Transfusion of patients with a hemoglobin concentration >10 g/dL was associated with improved neurological outcomes (odds ratio, 0.09; 95% confidence interval, 0.002-0.72; P=0.015), with no differences in mortality. Conclusions- Anemia after aneurysmal subarachnoid hemorrhage is associated with poor long-term neurological outcome and death. Transfusion of packed red blood cells is beneficial for patients who are not considerably anemic beforehand, suggesting further work needs to define the threshold but also the time period of anemia that is sufficient and necessary to contribute to poor outcomes. Clinical Trial Registration- URL: https://www.clinicaltrials.gov . Unique identifier: NCT00111085.
背景与目的- 蛛网膜下腔出血后贫血很常见，且具有潜在可干预性。在此，我们首先评估贫血对神经功能预后和死亡的影响，其次，研究红细胞输注对预后的影响。方法- 对 CONSCIOUS-1 研究（克拉生坦治疗蛛网膜下腔出血后神经缺血和梗死）413 例患者进行二次分析。多变量逻辑回归确定贫血的独立危险因素，并在调整选定协变量的情况下，确定贫血对神经功能预后和死亡的影响。采用受试者工作特征曲线分析确定血红蛋白水平的最佳预测阈值。最后，使用倾向评分匹配对患者进行伪随机分组以研究输血对预后的影响。结果- 入院时存在贫血（定义为血红蛋白＜10 g/dL）的患者占 5%，在血管内夹闭动脉瘤后 1-3 天的患者占 29%，在迟发性脑缺血风险期 5-9 天的患者占 32%。血管内夹闭动脉瘤后（比值比，1.96；95%置信区间，1.07-3.59；P=0.03）和在迟发性脑缺血期间（比值比，2.63；95%置信区间，1.46-4.76；P=0.0014）发生的贫血与不良神经功能预后独立相关。血管内夹闭动脉瘤后贫血（比值比，3.50；95%置信区间，1.15-10.62；P=0.027）但不是在迟发性脑缺血期间发生的贫血与死亡相关。使用倾向评分匹配队列，我们发现输注贫血患者并不能改善长期预后（P=0.8）或死亡率（P=0.9）。输注血红蛋白浓度＞10 g/dL 的患者与改善的神经功能预后相关（比值比，0.09；95%置信区间，0.002-0.72；P=0.015），死亡率无差异。结论- 蛛网膜下腔出血后贫血与长期神经功能预后不良和死亡相关。输注红细胞对之前没有明显贫血的患者有益，这表明需要进一步研究以确定贫血的阈值以及足以导致不良预后的贫血时间周期。临床试验注册- URL：https://www.clinicaltrials.gov。独特标识符：NCT00111085。
