Department of Cardiac Surgery, University of Michigan, Ann Arbor, Mich.
University of Michigan, Ann Arbor, Mich.
J Thorac Cardiovasc Surg. 2023 Jan;165(1):134-143.e3. doi: 10.1016/j.jtcvs.2021.01.064. Epub 2021 Jan 29.
We evaluated whether interhospital variation in mortality rates for coronary artery bypass grafting was driven by complications and failure to rescue.
An observational study was conducted among 83,747 patients undergoing isolated coronary artery bypass grafting between July 2011 and June 2017 across 90 hospitals. Failure to rescue was defined as operative mortality among patients developing complications. Complications included the Society of Thoracic Surgeons 5 major complications (stroke, surgical reexploration, deep sternal wound infection, renal failure, prolonged intubation) and a broader set of 19 overall complications. After creating terciles of hospital performance (based on observed:expected mortality), each tercile was compared on the basis of crude rates of (1) major and overall complications, (2) operative mortality, and (3) failure to rescue (among major and overall complications). The correlation between hospital observed and expected (to address confounding) failure to rescue rates was assessed.
Median Society of Thoracic Surgeons predicted mortality risk was similar across hospital observed:expected mortality terciles (P = .831). Mortality rates significantly increased across terciles (low tercile: 1.4%, high tercile: 2.8%). Although small in magnitude, rates of major (low tercile: 11.1%, high tercile: 12.2%) and overall (low tercile: 36.6%, high tercile: 35.3%) complications significantly differed across terciles. Nonetheless, failure to rescue rates increased substantially across terciles among patients with major (low tercile: 9.1%, high tercile: 14.3%) and overall (low tercile: 3.3%, high tercile: 6.8%) complications. Hospital observed and expected failure to rescue rates were positively correlated among patients with major (R = 0.14) and overall (R = 0.51) complications.
The reported interhospital variability in successful rescue after coronary artery bypass grafting supports the importance of identifying best practices at high-performing hospitals, including early recognition and management of complications.
我们评估了冠状动脉旁路移植术死亡率的院内差异是否由并发症和未能抢救导致。
对 2011 年 7 月至 2017 年 6 月间在 90 家医院接受单纯冠状动脉旁路移植术的 83747 名患者进行了一项观察性研究。未能抢救定义为发生并发症的患者的手术死亡率。并发症包括胸外科医师学会 5 项主要并发症（中风、再次手术探查、深部胸骨伤口感染、肾衰竭、长时间插管）和更广泛的 19 项总体并发症。在根据观察到的：预期死亡率创建医院绩效三分位数后，根据（1）主要和总体并发症、（2）手术死亡率和（3）主要和总体并发症中的未能抢救（在主要和总体并发症中）的粗发生率，对每个三分位数进行比较。评估了医院观察到的与预期（以解决混杂因素）的未能抢救率之间的相关性。
各医院观察到的：预期死亡率三分位数的胸外科医师学会预测死亡率风险中位数相似（P=0.831）。死亡率随着三分位数的升高而显著升高（低三分位数：1.4%，高三分位数：2.8%）。尽管幅度较小，但主要（低三分位数：11.1%，高三分位数：12.2%）和总体（低三分位数：36.6%，高三分位数：35.3%）并发症的发生率在三分位数之间也有显著差异。尽管如此，在有主要（低三分位数：9.1%，高三分位数：14.3%）和总体（低三分位数：3.3%，高三分位数：6.8%）并发症的患者中，未能抢救的发生率随着三分位数的升高而显著增加。在有主要（R=0.14）和总体（R=0.51）并发症的患者中，医院观察到的和预期的未能抢救率呈正相关。
报告的冠状动脉旁路移植术后成功抢救的院内差异支持确定高绩效医院最佳实践的重要性，包括早期识别和处理并发症。
