Abstract: Objective  To explore the effect of stroke severity on the efficacy of head-down position therapy in acute ischemic stroke.
Methods  This study was a post-hoc analysis of the head-down position for acute ischemic stroke with large artery atherosclerosis (HOPES 2) trial. Enrolled patients with acute ischemic stroke were stratified by admission NIHSS score into two subgroups: 6-10 points (moderately mild) and 11-16 points (moderately severe). Within each subgroup, patients received either head-down position therapy or routine position therapy. Linear regression analysis was performed to assess the effects of different treatments on the changes in NIHSS scores at 7 days (ΔNIHSS7 d) and 12 days (ΔNIHSS12 d) within each subgroup. Multivariable logistic regression analysis was used to investigate the effects of different treatments on the 90 d favorable (mRS score of 0-2) and excellent (mRS score of 0-1) outcomes outcome within each subgroup. Meanwhile, the interaction term between treatment and admission NIHSS score subgroup was included in the regression models to test for interaction.
Results  A total of 89 patients with acute ischemic stroke were enrolled, including 64 males, with a mean age of (62.6±11.5) years. Among them, 64 patients were in the admission NIHSS score 6-10 subgroup (31 received head-down position therapy, 33 received routine position therapy), and 25 patients were in the admission NIHSS score 11-16 subgroup (11 received head-down position therapy, 14 received routine position therapy). For the primary outcome, there was no significant difference in the proportion of patients with a 90-day favorable outcome between head-down position therapy and routine position therapy in either subgroup. For secondary outcomes, in the admission NIHSS score 6-10 subgroup, patients receiving head-down position therapy showed greater improvement in ΔNIHSS7 d (adjusted β=−2.23, 95%CI −4.81-−0.35, Padjusted=0.004) and ΔNIHSS12 d (adjusted β=−2.67, 95%CI −5.32-−0.02, Padjusted<0.001) than those receiving routine position therapy. In the admission NIHSS score 11-16 subgroup, there was no statistically significant difference in neurological improvement between patients receiving different treatments. Early neurological deterioration at 48 hours occurred in only 2 patients (6.1%) receiving routine position therapy within the admission NIHSS score 6-10 subgroup. For 90-day all-cause mortality, in the admission NIHSS score 6-10 subgroup, there was 1 death with head-down position therapy and 3 with routine position therapy, while in the 11-16 subgroup, there were 2 deaths with routine position therapy. Interaction analysis revealed that the interaction between treatment and admission NIHSS score was not statistically significant for either the primary or secondary outcomes.
Conclusions  Head-down position therapy can improve early neurological impairment in patients with moderately mild acute ischemic stroke, but has no significant effect on 90-day favorable outcome. No clear benefit of head-down position therapy was observed in patients with moderately severe acute ischemic stroke. Future studies may
optimize the head-down position therapy protocol and further explore its long-term benefits in patients with moderately mild acute ischemic stroke.

Key words: Head-down position; Acute ischemic stroke; National Institutes of Health stroke scale score

摘要： 目的 探讨卒中严重程度对头低位治疗急性缺血性卒中疗效的影响。
方法 本研究为大动脉粥样硬化型急性缺血性卒中的头低位治疗（head-down position for acute ischemic stroke with large artery atherosclerosis，HOPES 2）试验的事后分析。根据入院NIHSS评分，将入组的急性缺血性卒中患者分为入院NIHSS评分6～10分（中度较轻）亚组和入院NIHSS评分11～16分（中度较重）亚组。各亚组内分别接受头低位治疗与常规体位治疗。采用线性回归分析评估两个亚组内不同治疗方法对患者7 d（ΔNIHSS7 d）和12 d（ΔNIHSS12 d）NIHSS评分变化值的影响；采用多因素logistic回归分析探讨两个亚组内不同治疗方法对患者90 d预后［良好预后（mRS评分0～2分）和极佳预后（mRS评分0～1分）］的影响。同时，在回归模型中纳入治疗与入院NIHSS评分分组的交互项，以判断是否存在交互作用。
结果 本研究最终纳入89例急性缺血性卒中患者，男性64例，平均年龄（62.6±11.5）岁。其中入院NIHSS评分6～10分亚组64例（头低位治疗31例、常规体位治疗33例），入院NIHSS评分11～16分亚组25例（头低位治疗11例、常规体位治疗14例）。主要结局方面，两个亚组内接受头低位治疗与常规体位治疗患者在90 d良好预后率方面的差异均无统计学意义。次要结局方面，在入院NIHSS评分6～10分亚组中，接受头低位治疗患者ΔNIHSS7 d（校正β=-2.23，95%CI -4.81～-0.35，P校正=0.004）和ΔNIHSS12 d（校正β=-2.67，95%CI -5.32～-0.02，P校正<0.001）的改善幅度均优于接受常规体位治疗的患者。在入院NIHSS评分11～16分亚组中，接受不同治疗患者在神经功能改善方面的差异均无统计学意义。48 h早期神经功能恶化仅在入院NIHSS评分6～10分亚组内接受常规体位治疗的患者中出现2例（6.1%）。在发病90 d全因死亡方面，入院NIHSS评分6～10分亚组中，接受头低位治疗患者死亡1例，常规体位治疗患者死亡3例；入院NIHSS评分11～16分亚组中，接受常规体位治疗患者死亡2例。交互作用分析显示，治疗方法与入院NIHSS评分的交互作用在主要及次要结局指标中均无统计学意义。
结论 头低位治疗可改善中度较轻急性缺血性卒中患者的早期神经功能缺损，但对90 d良好预后无显著影响。在中度较重急性缺血性卒中患者中未观察到头低位治疗的明确疗效。未来研究可进一步优化头低位治疗方案，并探索其在中度较轻急性缺血性卒中患者中的长期获益。

关键词: 头低位; 急性缺血性卒中; 美国国立卫生研究院卒中量表评分