编者按：2024年9月22~24日，第20届世界食管疾病大会（ISDE）在苏格兰爱丁堡盛大召开。渥太华大学医学院Andrew Seely教授一项关于在食管癌围术期预测生命体征监测的研究入选大会摘要展示（摘要号：S19.05）。《肿瘤瞭望消化时讯》在大会现场针对该研究对Andrew Seely教授进行了专访。

肿瘤瞭望消化时讯：目前用于食管癌围术期预测性生命体征监测的技术有哪些？您能否举一些已经显示出显著获益的例子？

Oncology Frontier: What are the current technologies used for predictive vital sign monitoring peri-esophagectomy, and how effective are they in improving patient outcomes? Can you provide some examples that have demonstrated significant benefits? 

Andrew Seely: There is currently no proven technologies that use predictive monitoring to improve care. It's an area of burgeoning interest and exciting research, development and commercialization. We are at the early stages of applying predictive monitoring tools based on machine learning, where we integrate the information of multiple vital signs measured continuously over time. So, for example, there is a tool called the Visensia Safety Index, which integrates all of the vital signs to provide a prediction of subsequent deterioration, which can then be tracked in real time over time. And we are evaluating that at the University of Ottawa in surgical patients at this time. But there have been no proven clinical trials demonstrating benefit to predictive monitoring. In addition, novel technology for the future includes where we analyze patterns of variation of heart rate and respiratory rate, and then do machine learning on that information to create predictive clinical decision support tools to improve care in the emergency room, the ICU, or even on ward patients. That is an innovative use of heart rate variability and respiratory rate variability analysis in order to derive a better understanding of the patient status, their trajectory and their risk of deterioration. A challenge in this technology is getting access to high quality continuous waveform or continuous vital sign data with which to be to perform predictive monitoring. However, there's been new improvements in wearable monitors, new improvements in being able to access waveform data, data from patients in hospitals, and that is allowing us to develop and apply this technology further. So it's an exciting domain of research, development and commercialization that is going to evolve further in the coming years.

肿瘤瞭望消化时讯：在围手术期，预测性生命体征监测面临的主要挑战是什么？您是如何应对这些挑战？

Oncology Frontier: What are the main challenges associated with predictive vital sign monitoring in the perioperative setting? How do you address these challenges?

Andrew Seely: Well, a principal challenge is access to continuously monitored vital sign data and or waveform data, such as electrocardiogram or capnography data. So you have to have that data in order to perform predictive analyses on continuously monitored data. So that is one challenge, but as I mentioned, things are improving in that respect. A second challenge is that when everyone talks about predictive information, you are providing that predictive clinical decision support to clinicians such that they can improve a decision that they are making. For example, when to take a patient off a breathing machine in an ICU, we have a tool called extubation advisor that can optimize a clinician's ability to determine the timing and safety of taking a patient off a breathing machine, off a ventilator. But a challenge is that clinical decision support requires a physician to utilize that information and factor it into their decision making. And we are just starting to understand how doctors, in fact, utilize a predictive clinical decision support, how they trust it when they do, when they don't, and so that is a challenge as well.

肿瘤瞭望消化时讯：展望未来，您认为食管癌围术期预测性生命体征监测方面会有哪些进展或创新？这些进展或创新对患者护理和手术效果有哪些影响？

Oncology Frontier: Looking ahead, what advancements or innovations do you foresee in predictive vital sign monitoring for peri-esophagectomy? How might emerging technologies or approaches further enhance patient care and surgical outcomes in the future?

Andrew Seely教授：在食管切除术或任何复杂手术患者的管理中，预测性监测能发挥关键作用，我想重点阐述两个主要方面：

首先，我们可以通过改进术前评估来识别术后不良事件的风险，进而为每位患者制定个性化的靶向治疗方案，以预防特定术后不良事件的发生。例如，相较于为所有患者采用统一的加速康复路径，我们更倾向于采用模块化方法。这意味着，对于术后肺部并发症风险较高的患者，我们会在术前进行如高流量加温加湿给氧、呼吸肌训练等针对性干预，以减少这些并发症的发生。这是一种灵活的、个性化的、风险导向的术后管理方法。

其次，在术后阶段，我们可以利用连续预测性监测来提供不良事件的早期预警，从而改善管理和预防工作。理想情况下，这种监测能够及时发现如吻合口瘘、肺部感染或房颤等风险，并预防这些不良事件或其后果的恶化。这样我们就能更早地干预，更有效地保护患者的健康。

Andrew Seely: So I would highlight two principal areas where predictive monitoring can be useful in the management of esophageal resections or any patient undergoing complex surgery. First, I think that we can utilize better preoperative assessments to identify risks of postoperative adverse events to create targeted pathways that are individualized to help prevent certain post operative adverse events. For example, if instead of having one enhanced recovery pathway for all patients, the concept is that to have a modular approach where patients who are at increased risk for pulmonary complications postoperatively get targeted interventions like high flow, heated humidity, oxygen, or inspiratory muscle training preoperatively, and those things would help prevent postoperative pulmonary complications. So it's a modular, individualized, targeted approach to risk mitigation postoperatively. So that is one. A second, I think, is to use continuous predictive monitoring in the post operative period to provide early warning for adverse events and improved management and prevention, ideally of deterioration secondary to those adverse events. So earlier detection of anastomotic leak ,or pulmonary infection, or atrial risk for atrial arrhythmias, all those things, I think, could help prevent the development of those adverse events or the consequences of those adverse events.