Author + information
- Published online September 13, 2017.
- Pablo Martinez-Legazpi, MEng, PhD∗ (, )
- Lorenzo Rossini, AEng,
- Candelas Pérez del Villar, MD, PhD,
- Yolanda Benito, DCS, DVM,
- Carolina Devesa-Cordero, MD,
- Raquel Yotti, MD, PhD,
- Antonia Delgado-Montero, MD,
- Ana Gonzalez-Mansilla, MD, PhD,
- Andrew M. Kahn, MD,
- Francisco Fernandez-Avilés, MD, PhD,
- Juan C. del Álamo, AEng, PhD and
- Javier Bermejo, MD, PhD
- ↵∗Department of Cardiology, Hospital General Universitario Gregorio Marañón, Dr. Esquerdo 46, 28007 Madrid, Spain
During the subacute phase of acute myocardial infarction (AMI), the incidence of left ventricular thrombosis (LVT) can be as high as 15% to 20%. A method for assessing the risk of LVT would be of particular value in the setting of AMI, because prophylactic anticoagulation must be balanced against the bleeding risk of triple antithrombotic regimens. Recently, it has become possible to obtain patient-specific maps of left ventricular (LV) stasis using conventional echocardiography (1,2). To test the potential of stasis mapping in AMI, we prospectively studied 73 patients admitted to our institution for a first anterior ST-segment elevation AMI from July 8, 2013 to January 2, 2016. Additional inclusion criteria were sinus rhythm, absence of greater than mild aortic regurgitation, LV ejection fraction ≤45% within the first 72 h of AMI onset, stable clinical status, and Killip class less than IV. All patients underwent a full echocardiographic examination both in the early phase (within 72 h of admission) and after 4 to 5 months of follow-up. Contrast ultrasound was used to rule out LVT. For comparison, we studied 37 control subjects of similar age based on the absence of cardiovascular disease, no history of diabetes mellitus or hypertension, and a normal electrocardiogram and echocardiogram. The institutional review board approved the study, and all participants provided written informed consent.
We used color Doppler velocimetry to obtain the unsteady 2-dimensional (2D+t) blood flow field in the apical long-axis view (3). From the 2D+t velocity field, we mapped the residence time in the LV, a magnitude that accounts for the time spent by blood particles in transit through the chamber (Figure 1A) (4). To characterize global stasis, we measured the average residence time of the entire blood volume inside the LV after 8 beats (1). Additionally, because local stasis can be particularly meaningful for mural thrombosis, we identified and characterized stagnant regions, defined as regions with all their blood particles having residence time ≥2 cycles.
The AMI patient population consisted of 73 patients, and follow-up data were available in 62 (85%; median follow-up of 4.6 months). Global residence time was >50% higher in the early phase of AMI than in control subjects (2.6 ± 0.9 cycles vs. 1.7 ± 0.9 cycles; p < 0.001) (Figure 1B). Stagnant regions were larger (44 ± 15% vs. 27 ± 20% of total LV area; p < 0.001) and had longer regional residence times (4.4 ± 1.1 cycles vs. 3.8 ± 1.1 cycles; p = 0.01) in the early phase of AMI than in control subjects. All global and regional metrics of stagnant regions improved toward control values in the follow-up studies.
LVT was found in 15 patients (20%; blind analysis), 3 of them in the follow-up study. LVT-positive patients showed significantly different stasis metrics than LVT-negative patients in early-phase studies, as demonstrated by a longer global residence time (3.2 ± 0.7 cycles vs. 2.4 ± 0.8 cycles; p = 0.001) (Figure 1B), larger stagnant regions (52 ± 10% vs. 42 ± 20% of total LV area; p = 0.004), and longer regional residence times (5.1 ± 0.9 cycles vs. 4.3 ± 1.0 cycles; p = 0.02). Although apical wall motion score (AWMS) was higher in LVT-positive patients (18.4 ± 3.3 vs. 16.0 ± 5.8; p = 0.04), the performance of the global residence time for LVT detection (receiver-operating characteristic area under the curve [AUC] = 0.79) outweighed that of the AWMS (receiver-operating characteristic AUC = 0.64; p = 0.05 for comparison of both AUCs). According to bivariate logistic regression analysis that included average residence time and AWMS, only the former was significantly associated with LVT (p = 0.001 and p = 0.2, respectively). LV volumes and ejection fraction were of no use to predict LVT detection in this cohort.
To the best of our knowledge, this is the first prospective study demonstrating that objective indices of LV stasis can be derived from bedside echocardiography. Our results suggest that stasis imaging is a more powerful predictor of thrombosis than AWMS (5) in the setting of AMI. Thus, stasis imaging is a promising tool to address the risk of cardioembolic stroke in the clinical setting.
Please note: Presented in part at the American College of Cardiology 2015 Scientific Sessions, March 14 to 16, 2015, in San Diego, California. Dr. Martinez-Legazpi, Mr. Rossini, Dr. Yotti, Dr. Kahn, Dr. del Álamo, and Dr. Bermejo are inventors of a method for quantifying intracardiac stasis from imaging data under submitted Patent Cooperation Treaty and U.S. patent applications (application number 15/360,783). All other authors have reported that they have no relationships relevant to the contents of this paper to disclose.
- 2017 American College of Cardiology Foundation
- Rossini L.,
- Martinez-Legazpi P.,
- Vu V.,
- et al.
- Bermejo J.,
- Martínez-Legazpi P.,
- del Álamo J.C.
- Weinsaft J.W.,
- Kim J.,
- Medicherla C.B.,
- et al.