Denti Paolo, Pozzoli Alberto, Geretto Alberto, Vicentini Luca, Di Sanzo Stefania, Monaco Fabrizio, Crivellari Martina, Buzzatti Nicola, De Bonis Michele, La Canna Giovanni, Redaelli Alberto, Alfieri Ottavio
Heart Surgery Unit, Department of Cardiothoracic and Vascular, IRCCS San Raffaele University Hospital, Vita-Salute San Raffaele University, Milan, Italy.
Intensive Care Unit, Department of Cardiothoracic and Vascular, IRCCS San Raffaele University Hospital, Vita-Salute San Raffaele University, Milan, Italy.
Interact Cardiovasc Thorac Surg. 2017 Oct 1;25(4):513-519. doi: 10.1093/icvts/ivx186.
Systolic anterior motion (SAM) can be an insidious complication after mitral repair. Predicting SAM represents a challenge, even for very experienced mitral valve surgeons. The goal of this pilot work was to illustrate for the first time, a computational software able to calculate and prevent SAM during mitral repair.
Using MATLAB graphical user interface, a clinical software to predict SAM, we tested the performances of the software on 136 patients with degenerative mitral valves undergoing repair with standard techniques. A combination of 6 key echocardiographic parameters was used to calculate the SAM risk score. The discriminative performance of the model was assessed by the area under the receiver-operating characteristic curve. The receiver-operating characteristic was used to divide patients into low, medium and high risk for SAM. Simulation of virtual mitral repair (posterior leaflet resection and mitral ring annuloplasty) was also tested to reduce the risk of SAM.
The incidence of SAM was 8.1%; 73% were detected as high risk by the software. The area under the receiver-operating characteristic model discriminant performance was 0.87 (95% confidence interval: 0.78-0.95). Simulating a posterior leaflet resection with the leaflet length fixed at 15 mm, the estimated SAM risk was updated, and all patients were then classified at low risk.
This software is the first computational model designed to predict SAM during mitral repair to show excellent discrimination. This software has the potential to predict SAM risk preoperatively and, after a virtual step-by-step mitral repair simulation, depending on the technique adopted, to always achieve a low-risk SAM profile.
收缩期前向运动（SAM）可能是二尖瓣修复术后的一种隐匿性并发症。即使对于经验丰富的二尖瓣外科医生来说，预测SAM也是一项挑战。这项初步工作的目标是首次展示一种能够在二尖瓣修复过程中计算并预防SAM的计算软件。
我们使用MATLAB图形用户界面创建了一个预测SAM的临床软件，在136例采用标准技术进行修复的退行性二尖瓣患者身上测试了该软件的性能。使用6个关键超声心动图参数的组合来计算SAM风险评分。通过受试者工作特征曲线下的面积评估模型的判别性能。利用受试者工作特征曲线将患者分为SAM低、中、高风险组。还测试了虚拟二尖瓣修复（后叶切除和二尖瓣环成形术）的模拟，以降低SAM风险。
SAM的发生率为8.1%；软件检测出73%为高风险。受试者工作特征模型判别性能曲线下的面积为0.87（95%置信区间：0.78 - 0.95）。模拟后叶切除，将后叶长度固定为15毫米，更新了估计的SAM风险，然后所有患者被归类为低风险。
该软件是首个设计用于预测二尖瓣修复术中SAM的计算模型，具有出色的判别能力。该软件有潜力在术前预测SAM风险，并且在虚拟的逐步二尖瓣修复模拟后，根据所采用的技术，始终实现低风险的SAM情况。
