UNCORRECTED PROOF. 2 Incidence, diagnostic methods, and evolution of left. 3 ventricular thrombus in patients with anterior

Transcription

1 1 2 Incidence, diagnostic methods, and evolution of left 3 ventricular thrombus in patients with anterior 4 myocardial infarction and low left ventricular ejection 5 fraction: A prospective multicenter study Q36Q2 Philippe Meurin,MD, a Virginie Brandao Carreira,MD, b Raphaelle Dumaine,MD, a Alain Shqueir c Olivier Milleron,MD, d,e 7 Benjamin Safar, MD, d,e Sergio Perna, MD, f Charles Smadja, MD, g Marc Genest, MD, h Jérome Garot, MD, i 8 Bernard Carette, MD, j Laurent Payot, MD, k and Jean Yves Tabet, MD a, For the College National de Cardiologues 9Q4 Français and the Collège National des Cardiologues des Hôpitaux Français Villeneuve Saint Denis, France; Jossigny, 10 France; Esbly, France; 10 rue du Général Leclerc Montfermeil, France; Meaux, France; Tournan en Brie, France; 11 Provins, France; Massy, France; Reims, France; and Montreuil sous bois, France Background and objectives We aimed to assess the incidence and evolution of left ventricular (LV) thrombi in a 15 high-risk population of patients with LV systolic dysfunction after anterior myocardial infarction (ant-mi). We also compared the 16 accuracy of transthoracic echocardiography (TTE) and cardiac magnetic resonance imaging with contrast-delayed Q5 17 enhancement (CMR-DE) in detecting LV thrombi. 18 Methods We prospectively included 100 consecutive patients with LV ejection fraction (LVEF) b45% at the first TTE 19 performed b7 days after ant-mi. A second evaluation with TTE and CMR-DE (by blinded examiners) was performed at 30 days. 20 A third TTE and assessment of clinical status were performed between 6 and 12 months after ant-mi. 21 Results Patients (males 71%; mean age 59.1 ± 12.1 years; mean LVEF 33.5% ± 6.0%) were included at a median of days (interquartile range 25th-75th percentile days) after ant-mi. Thrombi were detected among 26 (26%) patients at 23 a median of 12.0 days after ant-mi (7 patients at 1-7 days after MI; 15 at 8-30 days; and 4 after day 30). Sensitivity and 24 specificity for LV thrombi detection were 94.7% and 98.5%, respectively, for TTE as compared with CMR-DE. Most thrombi (n = 24; %) disappeared after triple antithrombotic therapy (vitamin K antagonist in addition to dual antiplatelet therapy). 26 Conclusion Left ventricular thrombus is a frequent complication after ant-mi with systolic dysfunction. When a search for 27 thrombus is prespecified, the accuracy of TTE is high as compared with CMR-DE. The best antithrombotic strategy is not known. 28 (Am Heart J 2015;0:1-7.) The generalization of reperfusion techniques used 32 during the acute infarction phase has greatly decreased 33 the frequency of left ventricular (LV) thrombi because of From the a Centre de Réadaptation cardiaque de la Brie Les Grands Prés, 27 rue Sainte Christine, Villeneuve Saint Denis, France, b Marne La Vallée Hospital, 4 cours de Gondoire, Jossigny, France, c College National des Cardiologues Français and Cabinet Médical, Esbly, France, d Le Raincy-Montfermeil Hospital, 10 rue du Général Leclerc Montfermeil, France, e Collège National des Cardiologues des Hôpitaux Français, f Meaux Hospital, 6 rue Saint Fiacre, Meaux, France, g Tournan Clinic, 2 rue Jules Lefebvre, Tournan en Brie, France, h Léon Binet Hospital, route Chalautre, Provins, France, i Private Hospital Jacques Cartier, CMR Department, InstitutCardiovasculaireParisSud ICPS, Générale de Santé, 6Ave du Noyer Lambert, Massy, France, j Courlancy Clinic, 38 rue Courlancy, Reims, France, and k André Grégoire Hospital, 56 Boulevard de la Boissière, Montreuil sous bois, France. No conflict of interest. Submitted March 5, 2015; accepted April 25, Reprint requests: Philippe Meurin, MD, Centre de Réadaptation cardiaque de la Brie Les Grands Prés, 27 rue Sainte Christine, Villeneuve Saint Denis, France. philippemeurin@hotmail.com Published by Elsevier Inc. reduced myocardial damage. The use of drugs ameliorating LV remodeling has probably also been beneficial. Until the late 1990s, LV thrombi complicated approximately 40% of cases of anterior wall myocardial infarction (ant-mi), 1,2 and 10% to 20% of the thrombi progressed to arterial embolism. 3 Left ventricular thrombi are still estimated to complicate of 2% to 5% of all MI and 5% to 10% of unselected ant-mi However, we do not know the incidence of LV thrombi after ant-mi complicated with LV dysfunction despite modern management. Furthermore, the natural history of this condition is rather vague. Indeed, in most published studies, thrombi were assessed at a single time; and their size, mobility, and protruding or mural character were usually not described. Cardiac magnetic resonance imaging (MRI) with contrast-delayed enhancement (CMR-DE) is considered the criterion standard in assessing the presence, size, and location of thrombi. 11 In some studies, comparison of transthoracic echocardiography (TTE) and CMR-DE revealed

2 2 Meurin et al American Heart Journal Month Year 69 low sensitivity for TTE, 20% to 25% This finding is a 70 concern because TTE is the examination regularly per- 71 formed in daily practice to search for LV thrombi. 72 We conducted a prospective multicentric study of 73 patients at high risk of LV thrombus, that is, those with LV 74 ejection fraction (LVEF) b45% after ant-mi. 11 We com- 75 pared TTE and CMR-DE in terms of LV thrombus 76 discovery at day 30 and assessed clinical and echocardio- 77 graphic variables at 6 to 12 months of follow-up. 78 Methods 79 Study population 80 This was a prospective multicentric study (5 public 81 hospitals, 1 private hospital, and 1 cardiac rehabilitation 82 center). Between January 1, 2011, and August , all 83 Q6 consecutive patients with acute anterior wall STEMI 84 defined as follows were screened during hospitalization: 85 characteristic symptoms of myocardial ischemia associ- 86 ated with persistent electrocardiographic ST elevation at 87 the J point in 2 adjacent precordial leads (0.2 mv in leads 88 V2 and V3 or 0.1 mv in the other precordial leads, V1 or 89 V4-V6) and subsequent release of biomarkers of myocar- 90 dial necrosis. 16,17 In addition, these patients had to have 91 LVEF b45% at the first TTE (TTE 1 ), performed during the 92 first 7 days after ant-mi. The only exclusion criterion was 93 CMR contraindication at baseline. 94 All patients gave their informed consent to be in the 95 study. The study protocol conformed to the guidelines of 96 the 1975 Declaration of Helsinki as reflected in a priori 97 approval by the institution's human research committee. 98 Transthoracic echocardiography 99 After the inclusion of TTE 1, patients underwent a 100 mandatory second assessment including TTE (TTE 2 )and 101 CMR-DE at a median of 30 days (authorized range days) after ant-mi. The 2 examinations had to be performed 103 as closely as possible to each other. All TTE 2 s were 104 performed and interpreted jointly by 2 experienced 105 operators of the cardiac rehabilitation center, with blinding 106 to CMR results. A predesignated third echocardiographist 107 was consulted in case of discordance, which occurred only 108 once. The patient's cardiologist performed a third TTE 109 (TTE 3 ) at 6 to 12 months after inclusion. Moreover, to have 110 a better understanding of the evolution of the thrombi, in 111 patients in whom a LV thrombus had been discovered, a 112 TTE was performed at least once a week for 4 weeks. 113 Furthermore, intermediate TTEs were encouraged in all 114 patients (particularly between TTE 1 and TTE 2 ) but not 115 mandatory. All TTE 2 examinations involved use of a VIVID type echograph (2.5-MHz gauge, second harmonic 117 imaging; Vingmed GE). 118 Left ventricular thrombi were considered all masses with 119 echogenicity greater than that of blood, with well-defined 120 edges distinct from the endocardium, contiguous to an 121 akinetic or dyskinetic myocardium segment and present during the whole cardiac cycle on 2 different views. 11, A thrombus was considered protruding when it projected 123 predominantly into the LV cavity and mural when it 124 appeared flat and parallel to the endocardial surface. 11 The 125 thrombus area was measured by planimetry on a 4-cham- 126 ber apical view. Left ventricular ejection fraction was 127 measured by the Simpson method. 19 Importantly, we 128 prespecified on the examination prescription that possible 129 LV thrombus was part of the clinical indication for TTE. 130 Contrast-delayed enhancement As mentioned above, CMR-DE was performed at a median of 30 days (authorized range days) after ant-mi, as closely as possible to the TTE 2. Because LV thrombus is rare in patients with good global systolic function, 11 we prespecified that CMR-DE would be unnecessary (and therefore unethical in a survey) for patients with LVEF 50% on intermediate echocardiography (between TTE 1 and TTE 2 ). Contrast-delayed enhancement involved a 1.5 T scanner (Magnetom Espree; Siemens, Erlangen, Germany). Contiguous short-axis locations encompassing LV from the base to the apex were acquired in the cine steady-state free precession sequence. Each slice was acquired during 1 short breath hold (7-12 seconds each, depending on the heart rate). Typical imaging parameters for cine CMR were field of view, 300 to 360 mm; slice thickness, 6.0 mm; 3.1-ms repetition time; 1.6-ms echo time; flip angle, 60 ; image matrix, ; and temporal resolution, 30 to 40 ms. Delayed enhancement images were obtained 10 to 15 minutes after administration of a gadolinium-based contrast agent (Dotarem, mmol/kg; Guerbet) by a 2-dimensional segmented inversion-recovery gradient-echo pulse sequence, with slice position identical to the cine images (field of view, mm; slice thickness, 6.0 mm; and inversion time, ms). Left ventricular thrombi on CMR were defined as filling defects within the LV cavity, typically adherent to regions of abnormal wall motion (hypokinesis, akinesis, or dyskinesis) on cine sequences and confirmed by late DE. Endocardial and epicardial borders were outlined manually on short-axis end-diastolic and end-systolic cine images, then LV end-systolic volume and LV end-diastolic volume were determined, and LVEF was calculated. Evaluators were blinded to echocardiography results. We prespecified that LV thrombus detection was part of the clinical indication for CMR-DE. Follow-up After hospital discharge, 88 patients received care at the cardiac rehabilitation center, then patients were followed up by their cardiologists, until TTE 3 was performed 6 to 12 months after the ant-mi. The other 12 patients were followed up by their cardiologist at hospital discharge, but, as detailed above, even in these 12 patients, all TTE 2 s were performed at the cardiac rehabilitation center Q

3 American Heart Journal Volume 0, Number 0 Meurin et al 3 Q1 Figure consecutive patients with LVEF < 45% after ant-mi First TTE assessment: TTE 1 performed at a median of 5.5 days after ant-mi (IQR ) New LV Thrombus: n = For patients with an LV thrombus, exercise training 177 sessions were stopped, until the resolution of the thrombus 178 Q8 and VKA therapy was advised, 16 with an international 179 normalized ratio target of 2 to 3 for 6 months, in addition to 180 dual antiplatelet therapy. 181 Stroke was defined as new-onset focal or global neuro- 182 logic deficit caused by ischemia or hemorrhage as assessed 183 by appropriate imaging (CT or MRI) within or around the 184 brain and lasting for N24 hours or leading to death. 185 According to the criteria of thrombolysis in the TIMI trial, 186 major bleeding was defined as a decrease in hemoglobin 187 (Hg) level N5 g/dl, intracranial hemorrhage or N15% 188 absolute decrease in hematocrit level, or cardiac tampo- 189 nade and minor bleeding as a decrease in Hg level of 3 g/dl 190 or N10% decrease in hematocrit level from an identified 191 site; N4 g/dl decrease in Hg level if the bleeding site was not 192 identified; or spontaneous gross hematuria, hematemesis, 193 or hemoptysis. 194 No extramural funding was used to support this work. 195 The authors are solely responsible for the design and 196 conduct of this study, all study analyses, the drafting and 197 editing of the paper, and its final contents. 198 Statistical analysis Second TTE assessment: TTE 2 performed at a median of 30.0 days (IQR ) New LVThrombus: n = 18 TTE 2 and CMR-DE (allowing comparison of the 2 techniques) n = As in the largest prospective study of the topic and 200 comparing CMR-DE and TTE with 17 LV thrombi at TTE 2, no CMR-DE, because CMR-DE was not indicated or possible, n = 22: -No need for CMR (LVEF at day 30 50%), n = 9 -CMR contra-indication, n = 13 Patients not continuing (n = 5): 3 deaths; 2 cardiac transplantation Third assessment: TTE3 performed at a median of 270 days (IQR ) n = 95 New LV Thrombus: n = 1 Flow chart of the study. baseline, 12 we aimed to compare CMR-DE and TTE with 18 LV thrombi detected. We finally ended inclusions after comparing the 2 techniques with 19 LV thrombi (which required 100 patients) (see Figure 1). Data are reported as mean ± SD or median (interquartile range 25th-75th percentile [IQR ]) for continuous variables and n (%) for categorical variables. Unpaired Student t test and Fisher exact test were used to compare differences between continuous and categorical variables, respectively. P b.05 was considered statistically significant. Results Patients Between January 1, 2011, and August 31, 2013, we included 100 consecutive patients (Figure 1). The mean age was 59.1 ± 12.1 years, and 71% were male. The mean baseline LVEF was 33.5% ± 6.0% (Table I). All patients except 1 (99%) underwent coronary angiography; 88 (88%) underwent primary percutaneous intervention (PCI) with stenting and for 84, this involved the left anterior descending artery (LAD); for 15, the circumflex; for 2, the right coronary artery; and for 5, the left main artery. Twelve patients did not undergo PCI: 2 had a pure LAD thrombosis, which was treated with anticoagulants; 9 presented at 18 to 24 hours after onset

4 4 Meurin et al American Heart Journal Month Year Table I. Characteristics of patients with and without thrombus t1:1 t1:2 t1:3 Thrombus (n = 26) No thrombus (n = 74) P Figure 2 t1:4 Age, years, mean ± SD 57.4 ± ± t1:5 Male gender 20 (76.1) 52 (70.3).78 t1:6 Current smoker 12 (46.2) 31 (41.9).71 t1:7 Diabetes 3 (11.5) 17 (23.0).21 t1:8 Known hypertension 7 (26.9) 26 (35.1).44 t1:9 Hypercholesterolemia 5 (16.9) 3 (44.6).12 t1:10 BMI, kg/m 2, mean ± SD 26.7 ± ± t1:11 Preexisting atrial fibrillation 1 (3.8) 2 (2.7).77 t1:12 Coronary angiography 25 (96.2) 74 (100).10 t1:13 Primary PCI 20 (77.0) 68 (92.0).04 t1:14 LAD PCI 18 (69.2) 66 (89.2).02 t1:15 Left main PCI 1 (3.8) 4 (5.5).75 t1:16 PCI on several vessels 4(15.3) 15(20.2).57 t1:17 LVEF, %, mean ± SD t1:18 At TTE ± ± t1:19 At TTE ± ± t1:20 At TTE ± ± t1:21 Drug therapy at TTE 1 t1:22 Aspirin 26 (100) 74 (100) 1 t1:23 Clopidogrel 15 (57.7) 38 (51.4).58 t1:24 Prasugrel 8 (30.8) 29 (39.1).47 t1:25 Ticagrelor 2 (7.7) 3 (4.1).50 t1:26 β-blocker 26 (100) 74 (100) t1:27 Ivabradine 1 (3.8) 1 (1.4).45 t1:28 ACE-I 26 (100) 73 (95.9).29 t1:29 Statin 25 (96.2) 74 (100).10 t1:30 MRA inhibitor 16 (61.5) 43 (58.3).77 t1:31 t1:32 Data are n (%) unless indicated. Abbreviations: BMI, body mass index; ACE-I, angiotensin-converting enzyme inhibitor; MRA, mineralocorticoid receptor antagonist. 225 and at that time had no more evidence of ongoing 226 ischemia; and 1 had a contraindication to PCI (subdural 227 hematoma described below). All patients received aspirin 228 ( mg/day), and 95% received a thienopyridine 229 (clopidogrel, n = 53; prasugrel, n = 37; or ticagrelor, n = 230 5). VKA therapy was introduced in all patients with detected 231 LV thrombus (except for an 80-year-old woman with acute 232 MI who had a subdural hematoma due to a fall, before the 233 ant-mi) and was maintained for 6months, exceptfor1 234 patient with a very small thrombus (area 0.5 cm 2 )inwhom 235 VKA therapy was maintained for only 3 months. When VKA 236 was introduced, prasugrel or ticagrelor was replaced by 237 clopidogrel. One other patient received chronic treatment 238 with a VKA for atrial fibrillation. 239 Incidence of LV thrombus at baseline and at follow-up 240 A total of 26 patients (26%) had an LV thrombus that 241 was discovered a median of 12.0 days (IQR ) 242 after anti-mi: 6 at inclusion, 16 at days 8 to 30 after ant-mi, 243 and 4 others after day 30 (Figure 2). For 24 (92.3%), the 244 thrombus had disappeared at TTE 3. On the day of 245 discovery, the mean thrombus area was 2.1 ± 1.7 cm 2 ; 246 and 65% were protruding, 35% mural, and 15% mobile 247 (Table II). Importantly, no LV thrombus was discovered in 248 the 9 patients in whom CMR-DE had not been performed 249 because of LVEF 50% on intermediate echocardiogra- Days after acute anterior myocardial infarction Detection of LV thrombus by days after ant-mi.days after acute anterior myocardial infarction. phy. For patients with and without LV thrombus, the mean LVEF was 32.3% ± 4.6% and 34.0% ± 6.4% at TTE (P =.22), respectively; and 34.1% ± 6.6% and 39.0% ± % at TTE 2 (P =.01)and37.0%±10.4%and43.6%± % at TTE 3 (P =.006) (Table I). 272 Comparison of TTE and CMR-DE in detecting LV thrombus 273 For all patients, TTE 1 was performed at a median of days (IQR ) and TTE 2 at a median of 30.0 days 275 (IQR ) after ant-mi (Figure 1). During follow-up, patients died, and 2 underwent transplantation, so patients underwent TTE 3 at a median of 270 days (IQR ) after ant-mi. For 78 (78%) patients, CMR-DE was 279 performed at a median of 30.0 days (IQR ) 280 after ant-mi. For the 22 other patients, CMR-DE was 281 unnecessary because of LVEF 50% measured at interme- 282 diate echocardiography (between TTE 1 and TTE 2 )(n=9) 283 or was not possible (patient not stable enough to undergo 284 CMR n = 1; patient interrupting the examination because of 285 anxiety n = 1; or pacemaker or cardioverter defibrillator 286 implanted before that day n = 11) (Figure 1). 287 Of the 19 thrombi seen on CMR-DE, 18 were visible on 288 TTE (Table III). For 1 patient, TTE missed a small mural 289 apical thrombus (area 0.5 cm 2 ), and for one other 290 patient, TTE suggested a thrombus, which could not be 291 confirmed on CMR-DE. Therefore, the sensitivity and 292 specificity of TTE 2 compared with CMR-DE were 94.7% 293 and 98.3%, respectively. 294 Clinical follow-up During the median 270 days of clinical observation, 2 patients underwent cardiac transplantation, and 3 died (due to sudden death at day 60; a recurrent subdural hematoma at day 52 in an 80-year-old woman with an LV thrombus and not receiving a VKA; and at day 44, early after coronary artery bypass grafting). No patient was lost to follow-up

5 American Heart Journal Volume 0, Number 0 Meurin et al 5 t2:1 t2:2 Table II. Characteristics of LV thrombi (n = 26) t2:3 Post-MI first discovery, d, mean ± SD 23.2 ± 34.8 (median 12.0 days; IQR ) t2:4 Thrombus maximal area, cm 2, mean ± SD 2.1 ± 1.7 (median 2.1; IQR ) t2:5 Mobile thrombi, n (%) 4 (15) t2:6 Protruding thrombi, n (%) 17 (65) t2:7 Thrombus resolution, n (%) 24 (92.3) t2:8 Mean duration, d, mean ± SD 53 ± 61 (median 12 days; IQR ) t2:9 Thromboembolic event, n (%) 1 (3.8) 303 One thromboembolic event occurred (arterial inferior 304 limb and splenic embolism at day 15 after ant-mi) in a year-old patient with a large protruding LV thrombus 306 (area 9 cm 2 ) discovered at day 6 after ant-mi, despite the 307 introduction of low-molecular-weight heparin (enoxa- 308 parin 100 IU/kg twice a day as a bridge before reaching an 309 international ratio 2) and VKA as soon as the thrombus 310 was discovered. The evolution was favorable. 311 Five severe hemorragic events occurred: 2 (7.7%) 312 among the 26 patients receiving triple antithrombotic 313 therapy (1 patient 67 years old with posttraumatic in- 314 tracerebral hematoma of favorable evolution and 1 with 315 rectorragias due to unknown colorectal cancer) and (4%) among 74 others not receiving VKA (1 each with 317 severe rectorragia, psoas hematoma, and the recurrent 318 lethal subdural hematoma mentioned above). 319 In total, 7 patients were hospitalized for decompensat- 320 ed heart failure during follow-up. 321 Overall, 15 patients required implantation of an 322 internal cardiac defibrillator and/or a cardiac resyn- 323 chronization therapy; 6 others required cardiac surgery: heart transplantation (mentioned above), 2 coronary artery 325 bypass grafting (1 died), 1 mitral valvular replacement, and interventricular communication correction. Finally, cancers 327 were discovered in 3 patients (1 each of colorectal cancer, 328 mentioned above; lung cancer; and laryngeal cancer). 329 Discussion 330 Our results indicate that the incidence of LV thrombi 331 among patients after major acute anterior-wall MI is high 332 (26%) and that this complication does not seem to be 333 prevented by potent dual antiplatelet therapy. This result 334 should not be surprising. Indeed, our study was precisely 335 designed to evaluate the incidence of LV thrombi in 336 patients with the 2 main risk factors of this complication 337 occurrence: preferential location at the apical scar of an 338 ant-mi and low LVEF. In former studies, LV thrombi 339 occurred in 5% to 10% of unselected cases of ant-mi, and in the ASTAMI trial, among 100 patients who 341 survived ant-mi with moderate LV dysfunction (mean 342 LVEF 45%), 15 LV thrombi occurred. 10 In the present 343 study, patients had a very low LVEF (equal to 33.5% ± Table III. Sensitivity and specificity of thrombus detection by TTE t3:1 and cardiovascular MRI with CMR-DE t3:2 CMR-DE t3:3 TTE No thrombus Thrombus No thrombus 58 1 Thrombus % at baseline) after ant-mi. Thus, this result confirms that 344 the lower the LVEF, the higher the thrombus incidence. 345 The second reason why this high incidence should not 346 be unexpected is that our study involved 3 TTE searches 347 for LV thrombus, so we could precisely assess the natural 348 history of thrombus formation and resolution: most of the 349 LV thrombi (25/26) occurred during the first 6 weeks after 350 the MI, half (13/26) disappeared within 1 month, and nearly 351 all (24/26) had disappeared at day 270. The last reason 352 explaining this high thrombus incidence is that, importantly, 353 a search for LV thrombus was prespecified on the TTE form. 354 The second objective of our study was to assess the 355 accuracy of TTE as compared with CMR-DE in detecting 356 LV thrombi. This is an important question because even if 357 it is considered the criterion standard, CMR-DE has major flaws: high cost and low accessibility. Furthermore, 359 LV thrombus can develop at various times after ant-mi. 360 Thus, if TTE is accurate, performing regular TTEs rather 361 than a single CMR-DE could be useful in patients with 362 normal echogenicity. To our knowledge, only 4 studies 363 have tried to perform such a comparison Two were 364 retrospective, 13,15 which does not allow for a fair compar- 365 ison. Indeed, TTE is more operator dependent than is 366 CMR-DE. This important notion is highlighted by the 367 finding that echocardiographic performance varies by 368 indication: if LV thrombus search is prespecified, sensitivity 369 is multiplied by 2 (60% vs 26%) and positive predictive 370 value by 3 (75% vs 21%) as compared with unfocused 371 routine TTE In 2 prospective studies, 12,14 41 LV thrombi were 373 detected by CMR-DE; TTE had surprisingly low sensitivity, % and 42%, respectively, but good specificity, 95% and %. However, in our study, TTE accuracy was excellent, 376 with sensitivity, 94.7% and specificity, 98.3%. The authors 377 of the smaller study (12 LV thrombi 14 ) emphasized that 378 echocardiographic nonvisualized thrombi were usually 379 mural and small, which agrees with our results. Indeed, the 380 only LV thrombus that we missed at TTE was mural and was cm 2. Unfortunately, in the second prospective study, of LV thrombi (17 at baseline, 12 at 4 months later 12 ), the 383 thrombi characteristics (size, mobility, and protruding 384 character) were not described. The very low sensitivity of 385 TTE in this study could be due to the CMR-DE detection of 386 small mural thrombi of less clinical interest because of low 387 embolic risk 11 or to methodological issues: indeed, because 388 t3:4 t3:5 t3:6 Sensitivity, 94.7%; specificity, 98.5%. t3:7

6 6 Meurin et al American Heart Journal Month Year 389 this study was a substudy of the HEBE trial 20 in which CMR 390 and TTE were primarily performed to detect change in 391 regional myocardial function and not to compare their 392 accuracy in LV thrombus detection, the low TTE sensitivity 393 in LV thrombus detection could be due to the lack of 394 prespecifying a thrombus search on the examination 395 prescription. Moreover, in this study, 2% of the patients 396 were excluded because of poor-quality CMR-DE. 397 In the present study, LV thrombus occurred in 22.7% of 398 patients who underwent primary PCI versus 50% of the 399 others (P =.04). Furthermore, patients with LV thrombus 400 showed worse recovery of systolic function as shown by 401 the LVEF evolution between TTE 1 and TTE 3. These points could suggest worse viability in the LV apex of 403 patients developing an LV thrombus, but this aspect was 404 beyond the scope of our study. 405 Regarding LV thrombus treatment, no prospective 406 randomized study has ever shown the efficacy of VKA; 407 however, a meta-analysis of 7 observational studies found an association of anticoagulation therapy for months and reduced rate of embolization (odds ratio , 95% CI ). Therefore, despite the lack of 411 strong evidence, it is usually recommended that patients 412 with an LV thrombus after MI receive a VKA, without 413 stopping dual antiplatelet therapy. 11,16 However, many 414 questions remain. First, the optimal duration of this triple 415 therapy is not known. Repeated imaging of the left 416 ventricle after 3 months of therapy may allow for VKA 417 discontinuation earlier than 6 months if evidence of 418 thrombus is no longer present, particularly with recovery 419 of apical wall motion. 16 Second, in the WOEST study, the use of clopidogrel without aspirin for patients already 421 under oral anticoagulant therapy and undergoing PCI was 422 associated with reduced risk of bleeding without 423 increased thrombotic risk. Even if these mostly stable 424 patients differed from the patients included in our study, 425 these results could lead to a wider use of this new 426 antithrombotic strategy. 427 Our results suggest that this triple antithrombotic 428 therapy strategy is feasible, but, obviously, we cannot 429 conclude whether other antithrombotic strategies would 430 have had better or worse results. 431 We observed only 1 thromboembolic event (ie, 3.8% of 432 patients with an LV thrombus), which agrees with recent 433 findings with identical treatment. 7,8,10,11 However, the 434 rate of severe hemorragic complications (7.7%, n = 2) 435 remained high. 436 Our study has several limitations; first, even if CMR-DE 437 is considered a criterion standard, validation of CMR-DE 438 findings was not histopathologically confirmed. Second, 439 we did not use contrast echo because our objective was 440 to test TTE accuracy in daily practice. Third, whether 441 management of LV thrombus requires VKA therapy in 442 addition to dual antiplatelet therapy cannot be definitive- 443 ly concluded from this observational study and requires 444 confirmation in a randomized trial of patients with LV thrombus after acute MI. Finally, with the study calendar, even if unlikely, LV thrombi occurring after day 30 and disappearing spontaneously before TTE 3 could have been missed. In conclusion, we show that the frequency of LV thrombus is high after acute ant-mi with LV systolic dysfunction, mostly in the first 6 weeks. As well, when LV thrombus detection is specified in the examination prescription, TTE has excellent performance as compared with CMR-DE. Therefore, in this high-risk population, regular echocardiographic follow-up is required, at least in the first 6 weeks, whereas CMR-DE should be reserved for patients in whom the LV apex is not well observed during echocardiography. References 1. Van Dantzig JM, Delamarre J, Bot H, et al. Left ventricular thrombus in acute myocardial infarction. Eur Heart J 1996;17: Domenicucci S, Chiarella F, Bellotti P, et al. Long term prospective assessment of left ventricular thrombus in anterior wall acute myocardial infarction and implications for a national approach to embolic risk. Am J Cardiol 1999;83: Vaitkus P, Barnathan E. Embolic potential, prevention and management of mural thrombus complicating anterior myocardial infarction: a meta-analysis. J Am Coll Cardiol 1993;22: Motro M, Barbash GI, Hod H, et al. Incidence of left ventricular thrombus formation after thrombolytic therapy with recombinant tissue plasminogen activator, heparin and aspirin in patients with acute myocardial infarction. Am Heart J 1991;122: Kalra A, Jang IK. Prevalence of early left ventricular thrombus after primary coronary intervention for acute myocardial infarction. J Thromb Thrombolysis 2000;10: Greaves SC, Zhi G, Lee RT, et al. Incidence and natural history of left ventricular thrombus following anterior wall acute myocardial infarction. Am J Cardiol 1997;80: Meurin P, Tabet JY, Renaud N, et al. Treatment of left ventricular thrombi with a low molecular weight heparin. Int J Cardiol 2005;98(2): Gianstefani S, Douiri A, Delithanasis I, et al. Incidence and predictors of early left ventricular thrombus after ST-elevation myocardial infarction in the contemporary era of primary percutaneous coronary intervention. Am J Cardiol 2014;113(7): Osherov AB, Borovik-Raz M, Aronson D, et al. Incidence of early left ventricular thrombus after acute anterior wall myocardial infarction in the primary coronary intervention era. Am Heart J 2009;157: Solheim S, Seljeflot I, Lunde K, et al. Frequency of left ventricular thrombus in patients with anterior wall acute myocardial infarction treated with percutaneous coronary intervention and dual antiplatelet therapy. Am J Cardiol 2010;106: Delewi R, Zijlstra F, Piek JJ. Left ventricular thrombus formation after acute myocardial infarction. Heart 2012;98: Delewi R, Nijveldt R, Hirsch A, et al. Left ventricular thrombus formation after acute myocardial infarction as assessed by cardiovascular magnetic resonance imaging. Eur J Radiol 2012;81: Srichai MB, Junor C, Rodriguez LL, et al. Clinical, imaging, and pathological characteristics of left ventricular thrombus: a comparison of contrast-enhanced magnetic resonance imaging, transthoracic

7 American Heart Journal Volume 0, Number 0 Meurin et al echocardiography, and transesophageal echocardiography with 504 surgical or pathological validation. Am Heart J 2006;152: Mollet NR, Dymarkowski S, Volders W, et al. Visualization of 506 ventricular thrombi with contrast-enhanced magnetic resonance 507 imaging in patients with ischemic heart disease. Circulation ;106(23): Weinsaft JW, Kim HW, Crowley AL, et al. LV thrombus detection by 510 routine echocardiography: insights into performance characteristics 511 using delayed enhancement CMR. JACC Cardiovasc Imaging ;4: Steg PG, James SK, Atar D, et al. ESC Guidelines for the management 514 of acute myocardial infarction in patients presenting with ST-segment 515 elevation. Eur Heart J 2012;33: Huang H, Jneid H, Wilson J, et al. Comparison of angiographic 517 findings in patients with acute anteroseptal versus anterior wall 518 ST-elevation myocardial infarction. Am J Cardiol 2011;107: Carpenter K, Adams D. Apical mural thrombus: technical pitfalls. 520 Heart 1988;80(Suppl 1): Lang RM, Bierig M, Devereux RB, et al. Recommendations for chamber quantification: a report from the American Society of Echocardiography's Guidelines and Standards Committee and the Chamber Quantification Writing Group, developed in conjunction with the European Association of Echocardiography, a branch of the European Society of Cardiology. J Am Soc Echocardiogr 2005;18(12): Hirsch A, Nijveldt R, van der Vleuten PA, et al. Intracoronary infusion of mononuclear cells from bone marrow or peripheral blood compared with standard therapy in patients after acute myocardial infarction treated by primary percutaneous coronary intervention: results of the randomized controlled HEBE trial. Eur Heart J 2011;32: Dewilde WJ, Oirbans T, Verheugt FW, et al. Use of clopidogrel with or without aspirin in patients taking oral anticoagulant therapy and undergoing percutaneous coronary intervention: an open-label, randomised, controlled trial. Lancet 2013;381:

8 Our reference: YMHJ 4894 P-authorquery-v11 AUTHOR QUERY FORM Journal: YMHJ Please your responses and any corrections to: Article Number: Dear Author, Please check your proof carefully and mark all corrections at the appropriate place in the proof (e.g., by using on-screen annotation in the PDF file) or compile them in a separate list. Note: if you opt to annotate the file with software other than Adobe Reader then please also highlight the appropriate place in the PDF file. To ensure fast publication of your paper please return your corrections within 48 hours. For correction or revision of any artwork, please consult We were unable to process your file(s) fully electronically and have proceeded by Scanning (parts of) your article Rekeying (parts of) your article Scanning the artwork Any queries or remarks that have arisen during the processing of your manuscript are listed below and highlighted by flags in the proof. Click on the Q link to go to the location in the proof. Location in article Q1 Q2 Q3 Q4 Q5 Q6 Q7 Query / Remark: click on the Q link to go Please insert your reply or correction at the corresponding line in the proof Two sets of captions were provided for Figures 1 and 2 in this article. Please check if the captions used were the correct ones. Please confirm that given names and surnames have been identified correctly. Please provide the professional degree/s (eg, MD, PhD) for author Shqueir. Please supply the location (city, state/country) for Collège National des Cardiologues des Hôpitaux Français. Please check if the spelled out form provided for MRI" is appropriate. Please spell out STEMI" if it is an acronym. Please expand echo" in all occurrences if it is an abbreviation. Q8 Please expand VKA" at first mention if it is an acronym. Please check this box if you have no corrections to make to the PDF file. Thank you for your assistance. Page 1 of 1