Grandi Alessandro, Bertoglio Luca, Lepidi Sandro, Kölbel Tilo, Mani Kevin, Budtz-Lilly Jacob, DeMartino Randall, Scali Salvatore, Hanna Lydia, Troisi Nicola, Calvagna Cristiano, D'Oria Mario
Department of Vascular Medicine, University Heart and Vascular Center, 20251 Hamburg, Germany.
Division of Vascular Surgery, Department of Clinical and Experimental Sciences, ASST Spedali Civili of Brescia, 25123 Brescia, Italy.
J Clin Med. 2023 Aug 24;12(17):5505. doi: 10.3390/jcm12175505.
The present scoping review aims to describe and analyze available clinical data on the most commonly reported risk prediction indices in vascular surgery for perioperative mortality, with a particular focus on ruptured abdominal aortic aneurysm (rAAA).
A scoping review following the PRISMA Protocols Extension for Scoping Reviews was performed. Available full-text studies published in English in PubMed, Cochrane and EMBASE databases (last queried, 30 March 2023) were systematically reviewed and analyzed. The Population, Intervention, Comparison, Outcome (PICO) framework used to construct the search strings was the following: in patients with aortic pathologies, in particular rAAA (population), undergoing open or endovascular surgery (intervention), what different risk prediction models exist (comparison), and how well do they predict post-operative mortality (outcomes)?
The literature search and screening of all relevant abstracts revealed a total of 56 studies in the final qualitative synthesis. The main findings of the scoping review, grouped by the risk score that was investigated in the original studies, were synthetized without performing any formal meta-analysis. A total of nine risk scores for major vascular surgery or elective AAA, and 10 scores focusing on rAAA, were identified. Whilst there were several validation studies suggesting that most risk scores performed adequately in the setting of rAAA, none reached 100% accuracy. The Glasgow aneurysm score, ERAS and Vancouver score risk scores were more frequently included in validation studies and were more often used in secondary studies. Unfortunately, the published literature presents a heterogenicity of results in the validation studies comparing the different risk scores. To date, no risk score has been endorsed by any of the vascular surgery societies.
The use of risk scores in any complex surgery can have multiple advantages, especially when dealing with emergent cases, since they can inform perioperative decision making, patient and family discussions, and post hoc case-mix adjustments. Although a variety of different rAAA risk prediction tools have been published to date, none are superior to others based on this review. The heterogeneity of the variables used in the different scores impairs comparative analysis which represents a major limitation to understanding which risk score may be the "best" in contemporary practice. Future developments in artificial intelligence may further assist surgical decision making in predicting post-operative adverse events.
本综述旨在描述和分析血管外科手术中最常报告的围手术期死亡率风险预测指标的现有临床数据，尤其关注破裂性腹主动脉瘤（rAAA）。
按照PRISMA循证医学系统评价扩展版方案进行综述。系统检索并分析了在PubMed、Cochrane和EMBASE数据库（最后检索日期为2023年3月30日）中以英文发表的全文研究。用于构建检索词的人群、干预措施、对照、结局（PICO）框架如下：在患有主动脉病变，特别是rAAA的患者（人群）中，接受开放或血管内手术（干预措施），存在哪些不同的风险预测模型（对照），以及它们对术后死亡率的预测效果如何（结局）？
对所有相关摘要进行文献检索和筛选后，最终定性综合分析纳入了56项研究。本综述的主要结果按原始研究中调查的风险评分进行分组，未进行任何形式的荟萃分析。共确定了9个主要血管手术或择期腹主动脉瘤的风险评分，以及10个针对rAAA的评分。虽然有几项验证研究表明，大多数风险评分在rAAA情况下表现良好，但没有一个达到100%的准确性。格拉斯哥动脉瘤评分、加速康复外科（ERAS）和温哥华评分在验证研究中被更频繁地纳入，并且在二次研究中使用得更多。不幸的是，在比较不同风险评分的验证研究中，已发表的文献呈现出结果的异质性。迄今为止，没有任何一个风险评分得到血管外科学会的认可。
在任何复杂手术中使用风险评分都有多种优势，尤其是在处理急诊病例时，因为它们可以为围手术期决策、患者及家属讨论以及事后病例组合调整提供依据。尽管迄今为止已经发表了多种不同的rAAA风险预测工具，但基于本综述，没有一种优于其他工具。不同评分中使用的变量的异质性损害了比较分析，这是理解哪种风险评分在当代实践中可能是“最佳”评分的一个主要限制。人工智能的未来发展可能会进一步辅助手术决策，以预测术后不良事件。
