Author + information
- Published online November 6, 2017.
- Barnabas Gellen, MD, PhD∗ (, )
- Loïc Biere, MD,
- Damien Logeart, MD, PhD,
- Olivier Lairez, MD, PhD,
- Eric Vicaut, PhD,
- Alain Furber, MD, PhD,
- Jean-Jacques Mercadier, MD, PhD and
- Marc Sirol, MD, PhD
- ↵∗ELSAN, Polyclinique de Poitiers, 1 Rue de la Providence, 86000 Poitiers, France
Patients with anterior ST-segment elevation myocardial infarction (STEMI) are at high risk for left ventricular thrombi (LVT) (1). Independent predictors of LVT are anterior localization, large infarct size, and reduced left ventricular ejection fraction (2). The largest LVT cardiac magnetic resonance (CMR) study found LVT in 7% of patients with anterior STEMI (3). This is markedly lower than LVT rates found in other CMR cohorts (4). This study sought to assess the detection rate of LVT with regard to the delay between MI and CMR.
Data were obtained from 2 contemporary prospective French STEMI cohorts: the CMR substudy of the PREGICA (Predisposition Génétique à l’Insuffisance Cardiaque [Genetic Predisposition to Heart Failure]) and the REMOVE cohort (Biocollection on Ventricular Remodeling [French governmental hospital-based clinical research program number 2006/0070]). Inclusion periods were 2010 to 2015 and 2005 to 2014, respectively. Exclusion criteria were age <18 years and >80 years, previous STEMI, cardiogenic shock, and contraindication to CMR.
CMR was performed with a 1.5-T CMR system (Siemens, Munich, Germany). The imaging protocol was standard and included cardiac function (cine) and late gadolinium enhanced imaging, 8 to 12 min after 0.2 mmol/kg intravenous gadolinium administration (Dotarem, Guerbet, France). CMR data were analyzed offline by 2 experienced observers blinded to clinical information and angiographic findings. The presence of LVT was evaluated by the mean of late gadolinium enhanced images from 3 different views (short-axis, 2-chamber, and 4-chamber views). LVT was identified as an intracavitary mass with sharp contours and low signal intensity. LVT was classified as mural (borders concave, similar to surrounding endocardial contours) or protruding (projecting into the cavity), and its size measured by planimetry.
Data of 537 STEMI patients were obtained. Patients with low risk for LVT, as defined by a nonanterior infarct localization and by limited infarct size, were excluded from analysis. Out of 265 patients with anterior STEMI, ≥10% of left ventricular necrosis, and valid CMR within 21 days after STEMI, 34 patients had LVT. The baseline characteristics of patients with and without LVT did not differ significantly: age 58 ± 12 years, 85% male, 44% diabetics, 43% with history of smoking, and infarct size 31 ± 12%. Left ventricular ejection fraction and the rate of percutaneous coronary revascularization were lower in patients with LVT than in those without (38 ± 8% vs. 43 ± 10%, p = 0.0036; and 91% vs. 98%, p = 0.0472, respectively).
All LVT were located to the apex, and most (91%) were protruding. No difference in thrombus morphology was detected between patients with respect to the MI-to-CMR delay. All patients with LVT presented at least with apical akinesia (mean apical score, 3.4), whereas patients without LVT had less pronounced wall motion abnormalities (mean apical score, 2.65). Apical dyskinesia was more frequent in patients with LVT when compared with patients with no-LVT (18 of 34, 53% vs. 49 of 231, 21%; p < 0.01). Global wall motion score index was significantly worse in patients with LVT as compared with patients with no LVT (2.12 ± 0.8 vs. 1.86 ± 0.6; p < 0.01). No difference in apical akinesia and in global wall motion score index could be found between patients with late thrombus (>5 days) and patients with early thrombus (≤5 days; 2.23 ± 0.9 vs. 2.20 ± 0.8; p = 0.453).
LVT rates in patients with a CMR ≤5 days as compared with those with CMR >5 days were significantly lower (13 of 160 vs. 21 of 105; p = 0.0047). We further divided the study population in 5 subgroups according to the MI-to-CMR time (Figure 1). The highest LVT detection rate was obtained in patients in whom CMR was performed 9 to 12 days after MI.
The present work demonstrates a significant impact of the delay between STEMI and CMR on the detection of LVT, suggesting that CMR performed within the first 5 days might leave many LVT undetected. The detection rate of LVT at different MI-to-CMR delays was not assessed using repeated CMR in each patient, which constitutes the major limitation of the present work.
The authors thank Mrs. Elise Gand, MSc, for her precious methodological assistance.
Please note: The PREGICA and the REMOVE cohorts were supported by grants from the French governmental hospital-based clinical research program Programme Hospitalier de Recherche Clinique (NCT01113268). The authors have reported that they have no relationships relevant to the contents of this paper to disclose.
- 2017 American College of Cardiology Foundation
- Weinsaft J.W.,
- Kim J.,
- Medicherla C.B.,
- et al.
- Poss J.,
- Desch S.,
- Eitel C.,
- de Waha S.,
- Thiele H.,
- Eitel I.
- Meurin P.,
- Brandao Carreira V.,
- Dumaine R.,
- et al.