Background and Objectives

Prior research on volume-based patient outcomes related to acute ischemic stroke (AIS) have demonstrated contradictory results and fail to reflect recent advances in stroke care. We sought to examine contemporary relationships between hospital AIS volumes and outcomes.

Methods

We used complete Medicare datasets in a retrospective cohort study using validated International Classification of Diseases Tenth Revision codes to identify patients admitted with AIS from January 1, 2016 through December 31, 2019. AIS volume was calculated as the total number of AIS admissions per hospital during the study period. We examined several hospital characteristics by AIS volume quartile. We performed adjusted logistic regressions testing associations of AIS volume quartiles with: inpatient mortality, receipt of tissue plasminogen activator (tPA) and endovascular therapy (ET), discharge home, and 30-day outpatient visit. We adjusted for sex, age, Charlson comorbidity score, teaching hospital status, MDI, hospital urban-rural designation, stroke certification status and ICU and neurologist availability at the hospital.

Results

There were 952400 AIS admissions among 5084 US hospitals; AIS 4-year volume quartiles were: 1^st: 1-8 AIS admissions; 2^nd: 9-44; 3^rd: 45-237; 4^th: 238+. Highest quartile hospitals more often were stroke-certified (49.1% vs 8.7% in lowest quartile, p<0.0001), with ICU bed availability (19.8% vs 4.1%, p<0.0001) and with neurologist expertise (91.1% vs 3%, p<0.0001). In the highest AIS quartile (compared to the lowest quartile), there was lower inpatient mortality (odds ratio [OR] 0.71 [95%CI 0.57-0.87, p<0.0001]), lower 30-day mortality (0.55 [0.49-0.62], p<0.0001), greater receipt of tPA (6.60 [3.19-13.65], p<0.0001) and ET (16.43 [10.64-25.37], p<0.0001, and greater likelihood of discharge home (1.38 [1.22-1.56], p<0.0001). However, when the highest quartile hospitals were examined separately, higher volumes were associated with higher mortality despite higher rates of tPA and ET receipt.

Conclusions

High AIS-volume hospitals have greater utilization of acute stroke interventions, stroke certification and availability of neurologist and ICU care. These features likely play a role in the better outcomes observed at such centers, including inpatient and 30-day mortality and discharge home. However, the highest volume centers had higher mortality despite greater receipt of interventions. Further research is needed to better understand volume-outcome relationships in AIS to improve care at lower volume centers.

中文翻译：

---

美国医院急性缺血性卒中量、医院特征和结果之间的关系

背景和目标

先前关于与急性缺血性卒中 (AIS) 相关的基于容量的患者结果的研究已经证明了相互矛盾的结果，并且未能反映卒中护理的最新进展。我们试图研究医院 AIS 数量和结果之间的当代关系。

方法

我们在一项回顾性队列研究中使用了完整的医疗保险数据集，使用经过验证的国际疾病分类第十版修订代码来识别 2016 年 1 月 1 日至 2019 年 12 月 31 日期间因 AIS 入院的患者。AIS 量计算为期间每家医院的 AIS 入院总数学习期间。我们通过 AIS 体积四分位数检查了几个医院特征。我们对 AIS 体积四分位数与以下关联进行了调整逻辑回归测试：住院死亡率、接受组织纤溶酶原激活剂 (tPA) 和血管内治疗 (ET)、出院回家和 30 天门诊就诊。我们调整了性别、年龄、查尔森合并症评分、教学医院地位、MDI、医院城乡指定、卒中认证地位以及医院的 ICU 和神经科医生可用性。

结果

美国 5084 家医院中有 952400 人次 AIS 入院；AIS 4 年数量四分位数为： 第 1^名：1-8 名 AIS 入院；第二名： ^9-44；第三名：45-237 ^；4^号: 238+。排名最高的四分位数医院更经常获得卒中认证（49.1% 对最低四分位数的 8.7%，p<0.0001），ICU 病床可用性（19.8% 对 4.1%，p<0.0001）和具有神经科医生专业知识（91.1% 对 3%， p<0.0001）。在最高 AIS 四分位数（与最低四分位数相比）中，住院死亡率较低（比值比 [OR] 0.71 [95%CI 0.57-0.87，p<0.0001]），30 天死亡率较低（0.55 [0.49-0.62] ], p<0.0001)，更多接受 tPA (6.60 [3.19-13.65]，p<0.0001) 和 ET (16.43 [10.64-25.37]，p<0.0001，出院回家的可能性更大 (1.38 [1.22-1.56] , p<0.0001). 然而，当分别检查最高四分位数医院时，尽管 tPA 和 ET 接收率较高，但较高的容量与较高的死亡率相关。

结论

AIS 处理量大的医院可以更多地利用急性卒中干预、卒中认证以及神经科医生和 ICU 护理的可用性。这些特征可能在这些中心观察到的更好结果中发挥作用，包括住院和 30 天死亡率和出院回家。然而，尽管接受了更多的干预措施，但容量最大的中心死亡率更高。需要进一步研究以更好地了解 AIS 中的容量-结果关系，以改善低容量中心的护理。