Mechanisms, Predictors and Clinical Markers of Dysphagia in Cardiac Surgical Patients

项目摘要 吞咽困难(吞咽障碍)是心脏外科手术的常见并发症，导致 营养不良，脱水，吸入性肺炎，重新插管，卫生保健利用率增加，持续时间 住院和经济负担。虽然可以预防，但与吞咽困难相关的吸入性肺炎是一种主要的 死亡原因。因此，早期发现和准确监测吞咽困难对于促进 及时干预，减轻发展后遗症。目前，吞咽困难的临床护理受到以下因素的阻碍 知识上的根本差距，包括1)吞咽困难的危险因素未知，禁止 使用分类的个性化护理路径；2)没有经过验证的工具来准确检测和监测 存在心脏重症监护病房；以及3)吞咽损伤和恢复的调控机制 功能未知，阻碍了机械引导疗法的发展和优化 显著的术后评估时间点。我们的三个具体目标是针对这些广泛的知识差距 目标是将护理转变为主动、多方面和数据驱动的综合吞咽围手术期模式 关怀(马赛克)。为此，我们将在四年内招募360名心脏手术患者，并确定1) 吞咽困难的独立危险因素，2)吞咽障碍的敏感临床标志物，以及3)管理 急性、亚急性和长期不安全和低效吞咽的生理机制 术后一段时间。登记的参与者将接受手术前的光纤内窥镜评估 吞下(费用)以筛选出先前存在吞咽困难的患者。候选预测变量将为 在围手术期的整个过程中系统收集。手术后及术后48小时内 拔管时，将同时进行视频透视和费用(VF-费用)，以及 简单的床边临床测试。将使用经过验证的吞咽指标执行详细的盲法分析 安全性、有效性、计时和运动学检查急性期吞咽功能和相关 不安全或低效吞咽的生理机制。急性术后相性吞咽困难患者 将在1个月和6个月时重新检查，以确定亚急性和长期吞咽困难的轨迹和 管理损害和恢复的机制。吞咽困难危险因素的多变量建模将 制作实用的吞咽困难风险分层工具，以实现准确的预测和个性化分类 术后护理路径。将开发一个附带的开放获取电子应用程序，以提供 无缝的临床采用和数据驱动的临床决策工具。临床数据的判别能力 将确定标志物，并将验证实用的床边吞咽困难筛查工具，以实现准确的 心脏重症监护病房吞咽困难的检测和监测。成果将推动未来的目标 治疗和预防策略，并增强个性化医疗保健模式，以最终改善 病人的结果。

Project Summary Dysphagia (swallowing impairment) is a common complication of cardiac surgical procedures, leading to malnutrition, dehydration, aspiration pneumonia, reintubation, increased health care utilization, length of hospitalization, and economic burden. Although preventable, dysphagia-related aspiration pneumonia is a major cause of mortality. Early detection and accurate monitoring of dysphagia are therefore important to facilitate timely interventions to mitigate developing sequelae. Currently, clinical care of dysphagia is hindered by fundamental gaps in knowledge, including 1) contributing risk factors of dysphagia are unknown, prohibiting the use of triaged personalized care pathways; 2) no validated tools to accurately detect and monitor dysphagia in the cardiac intensive care unit exist; and 3) governing mechanisms of swallowing impairment and recovery of function are unknown, impeding the development of mechanistically guided therapeutics and optimization of salient postoperative evaluation time points. Our three specific aims target these knowledge gaps with the broad goal to shift care toward a proactive, multifaceted, and data-driven perioperative Model of Swallowing Integrated Care (MOSAIC). To this end, we will enroll 360 cardiac surgical patients over a four-year period and identify 1) independent risk factors for dysphagia, 2) sensitive clinical markers of swallowing impairment, and 3) governing physiologic mechanisms of unsafe and inefficient swallowing throughout the acute, sub-acute, and long-term postoperative period. Enrolled participants will undergo a preoperative Fiberoptic Endoscopic Evaluation of Swallowing (FEES) to screen out patients with pre-existing dysphagia. Candidate predictor variables will be systematically collected throughout the perioperative time course. Following surgery and within 48 hours of extubation, a simultaneous videofluoroscopy and FEES (VF-FEES) will be performed as well as a battery of simple bedside clinical tests. Detailed blinded analyses will be performed using validated metrics of swallowing safety, efficiency, timing and kinematics to examine acute-phase swallowing function and associated physiologic mechanisms of unsafe or inefficient deglutition. Patients with acute postoperative phase dysphagia will be re-examined at one- and six-months to determine sub-acute and long-term dysphagia trajectories and governing mechanisms of impairment and recovery. Multivariable modeling of dysphagia risk factors will produce a practical dysphagia risk stratification tool to enable accurate forecasting and personalized triaged postoperative care pathways. An accompanying open-access electronic application will be developed to provide seamless clinical adoption and a data-driven clinical decision making tool. The discriminant ability of clinical markers will be determined, and a practical bedside dysphagia screening tool will be validated to enable accurate detection and monitoring of dysphagia in the cardiac intensive care unit. Outcomes will drive future targeted therapeutic and preventative strategies and enhance personalized health care models to ultimately improve patient outcomes.