Abstract
OBJECTIVES
Left atrial appendage thrombus (LAT) was an exclusion criterion in the seminal transcatheter aortic valve replacement (TAVR) trials; however, such patients do undergo TAVR in the ‘real-world’ setting. This study sought to analyse outcomes after TAVR in patients with LAT or spontaneous echo contrast (SEC).
METHODS
All patients undergoing TAVR at our institution between March 2009 and December 2014 were prospectively analysed. The presence of LAT or SEC was determined via a retrospective chart review. Primary outcomes included 30-day and 1-year neurological events as well as mortality.
RESULTS
Of the 369 patients undergoing TAVR, 3.8% (14) were found to have LAT and 6.8% (25) were found to have SEC, and they were separately compared to patients who did not have LAT or SEC. Significant differences were noted between groups with regard to preoperative renal function, atrial fibrillation and ejection fraction. Preoperative atrial fibrillation was the only independent predictor of LAT. No perioperative complications were associated with the presence of LAT or SEC. Specifically, no patient with LAT or SEC experienced a postoperative neurological event. While neither LAT nor SEC was an independent predictor of 30-day mortality, LAT was an independent predictor of 1-year mortality (odds ratio 3.573, 95% confidence interval 1.040–12.28; P = 0.042).
CONCLUSIONS
The current study suggests that TAVR may be performed in patients with LAT and SEC with a low risk of embolic complications. While neither was an independent predictor of 30-day mortality, LAT was an independent predictor of 1-year mortality. Larger studies are needed to better study this phenomenon.
Keywords: Aortic stenosis , Atrial fibrillation , Left atrial thrombus
INTRODUCTION
Transcatheter aortic valve replacement (TAVR) has emerged as the preferred treatment strategy in patients with severe, symptomatic aortic stenosis (AS) considered at high or extreme risk for surgical aortic valve replacement (SAVR). Its use may soon be extended to patients at intermediate, and eventually, low risk, for SAVR as well [1–4]. As in all randomized clinical trials, specific inclusion and exclusion criteria were applied, leading to a highly selected study cohort. Subsequently, several ‘real-world’ cohort trials described outcomes in more liberally selected patients [5]. Preoperative atrial fibrillation (AF) has been shown to be an independent risk factor for perioperative stroke, heart failure and mortality in patients undergoing SAVR. In addition, AF has also been associated with a higher incidence of cardiac events and long-term mortality after TAVR [6]. Left atrial appendage thrombus (LAT) and spontaneous echo contrast (SEC) are frequently found in association with AF, valvular heart disease and decreased ejection fraction. LAT is believed to result in increased risk for cardioembolic events [7], and it was an exclusion criterion in the PARTNER and CoreValve pivotal trials [1–4]. Despite this, such patients have undergone TAVR in the ‘real-world’ setting. The aim of the present study was to compare baseline characteristics, perioperative characteristics and long-term clinical outcomes in patients with and without LAT or SEC who underwent TAVR.
MATERIALS AND METHODS
Between March 2009 and December 2014, 369 consecutive patients underwent TAVR at our institution, and the data of these patients were analysed. Patients undergoing transfemoral, transapical and transaortic approaches were included, as were patients undergoing ‘valve-in-valve’ procedures. Data regarding preoperative demographic, clinical and echocardiographic parameters, intraoperative echocardiographic and procedural parameters, and postoperative outcomes were collected prospectively. The presence of LAT or SEC was determined via a retrospective chart review. The primary outcomes included 30-day and 1-year neurological events as well as mortality. These were determined by chart review and telephone contact with patients and/or their referring cardiologists. Variables are reported as per definitions of the transcatheter valve therapy registry with the exception of postoperative bleeding and vascular complications, which are reported according to the Valve Academic Research Consortium-2 criteria. The Weill Cornell Medical College institutional review board approved the study.
All patients underwent evaluation by a multidisciplinary Heart Team, including cardiologists and cardiothoracic surgeons. TAVR was performed utilizing the self-expandable Medtronic CoreValve System (Medtronic Inc., Minneapolis, MN, USA) or the balloon-expandable Edwards-SAPIEN System (Edwards Lifesciences Inc., Irvine, CA, USA) at the discretion of the implanting team. All procedures were performed in the cardiac catheterization suite or hybrid operating room, under general anaesthesia, using fluoroscopy and transoesophageal echocardiography (TOE), with which the presence of LAT or SEC was noted.
Statistical analysis was performed with MATLAB R2015a using the Statistics and Machine Learning Toolbox (MathWorks, Inc., MA, USA). Continuous variables are represented as the mean ± standard deviation and compared using the Kruskal–Wallis analysis of variance test. Categorical variables are represented as patient number and percentage and compared with the χ2 test or the Fisher’s exact test, as appropriate (Fisher’s exact test used in the setting of small sample sizes with any cell size < 5). The influence of clinical parameters was first investigated by univariable logistic regression. All parameters with P-value <0.1 were selected for a multivariable logistic regression to identify the most relevant predictors for each outcome. Cox regression analysis was performed to assess independent predictors of overall mortality. Survival curves were estimated by Kaplan–Meier analysis, and the log-rank test was used to compare survival differences. A P-value <0.05 was considered statistically significant.
RESULTS
Between March 2009 and December 2014, 369 patients underwent TAVR at our institution. Of these, 3.8% (14) were found to have LAT and 6.8% (25) were found to have SEC by preoperative or intraoperative TOE. Significant differences were noted between patients with LAT, SEC, or neither with regard to preoperative AF and ejection fraction (Tables 1 and 2). There were no other significant differences between groups with regard to preoperative demographic or clinical characteristics, including echocardiographic parameters of aortic stenosis or other valvular disease (Tables 1 and 2).
Table 1:
Preoperative characteristics
| No LAT/SEC (n = 330) | LAT (n = 14) | SEC (n = 25) | P-value | |
|---|---|---|---|---|
| Age (years) | 85 ± 7.6 | 83.1 ± 5.8 | 83.8 ± 6.9 | 0.258 |
| Female gender | 179 (54) | 6 (43) | 8 (32) | 0.073 |
| PCI | 133 (40) | 6 (43) | 10 (40) | 1.000 |
| CABG | 90 (27) | 3 (21) | 5 (20) | 0.764 |
| Prior MI | 67 (20) | 3 (21) | 4 (16) | 0.896 |
| CVD | 70 (21) | 6 (43) | 6 (24) | 0.142 |
| PVD | 78 (24) | 5 (36) | 9 (36) | 0.216 |
| Hypertension | 290 (88) | 14 (100) | 21 (84) | 0.327 |
| Hypercholesterolaemia | 258 (78) | 12 (86) | 20 (80) | 0.902 |
| Diabetes mellitus | 108 (33) | 5 (36) | 9 (36) | 0.888 |
| Haemodialysis | 11 (3) | 2 (14) | 1 (4) | 0.085 |
| BNP | 740 ± 839 (n = 326) | 775 ± 527 (n = 14) | 709 ± 626 (n = 25) | 0.461 |
| CLD | 88 (27) | 3 (21) | 10 (40) | 0.325 |
| NYHA III–IV | 207 (63) | 12 (86) | 15 (60) | 0.208 |
| STS score | 8.91 ± 4.85 (n = 181) | 7.93 ± 3.97 (n = 9) | 7.98 ± 4.92 (n = 12) | 0.535 |
| Preprocedure AF | 114 (35) | 12 (86) | 15 (60) | <0.001 |
Values are presented as mean ± standard deviation or n (%).
AF: atrial fibrillation; BNP: B-type natriuretic peptide; CABG: coronary artery bypass graft; CLD: chronic lung disease; CVD: cerebral vascular disease; LAT: left atrial appendage thrombus; MI: myocardial infarction; NYHA: New York Heart Association; PCI: percutaneous coronary intervention; PVD: peripheral vascular disease; SEC: spontaneous echo contrast; STS: Society of Thoracic Surgeons.
Table 2:
Haemodynamic and echocardiographic parameters
| No LAT/SEC (n = 330) | LAT (n = 14) | SEC (n = 25) | P-value | |
|---|---|---|---|---|
| Ejection fraction (%) | 51.9 ± 14.0 (n = 317) | 46.3 ± 12.8 (n = 13) | 45.2 ± 14.9 (n = 23) | 0.013 |
| Left atrial diameter (mm) | 4.44 ± 0.74 (n = 314) | 4.74 ± 0.61 (n = 13) | 4.52 ± 0.63 (n = 23) | 0.130 |
| Mean gradient (mmHg) | 49.6 ± 14.7 (n = 325) | 46.8 ± 13.4 (n = 13) | 48.4 ± 17.9 (n = 25) | 0.718 |
| Peak velocity (m/s) | 4.56 ± 0.66 (n = 323) | 4.46 ± 0.66 (n = 13) | 4.51 ± 0.78 (n = 25) | 0.870 |
| Valve area (cm2) | 0.73 ± 0.22 (n = 325) | 0.63 ± 0.19 (n = 13) | 0.71 ± 0.14 (n = 25) | 0.126 |
| AI ≥ moderate | 14 (4.2) | 2 (14) | 3 (12) | 0.058 |
| MR ≥ moderate | 65 (20) | 4 (29) | 8 (32) | 0.227 |
| TR ≥ moderate | 49 (15) | 4 (29) | 7 (28) | 0.059 |
Values are presented as mean ± standard deviation or n (%).
AI: aortic insufficiency; LAT: left atrial appendage thrombus; MR: Mitral regurgitation; SEC: spontaneous echo contrast; TR: tricuspid regurgitation.
In all 14 patients with LAT, the thrombus was in the left atrial appendage. In 4 patients, LAT measurements were available (2.4 ×1.4 cm, 0.9 × 1.2 cm, 1.2 × 1.2 cm and 1.9 × 0.6 cm). LAT was laminar or organized in 3 patients. In 2 patients, LAT was described as ‘small’. In the remaining 5 patients, there was no additional description of LAT.
Of the 14 patients with LAT, 86% (12) had preoperative AF and 71% (10) were on anticoagulation chronically for AF. Of the 4 patients who were not on anticoagulation, 2 had no documented history of AF, 1 was discovered to be in AF during workup for TAVR, and 1 had a contraindication to anticoagulation (gastrointestinal bleeding). Three were started on anticoagulation upon discovery of LAT. All 14 patients underwent TAVR without embolic complications. Ninety-three percent (13) of patients with LAT were discharged from the hospital on anticoagulation, with the 1 patient without documented AF discharged without anticoagulation.
Of the 25 patients with SEC, 60% (15) had preoperative AF. Sixty percent (15) were on anticoagulation chronically, including 12 for AF and 3 for a history of deep vein thrombosis. Of the 40% (10) who were not on anticoagulation, 7 had no documented history of AF and 3 had AF but had contraindications to anticoagulation (fall risk). All 25 patients underwent TAVR without embolic complications. Forty-four percent (11) of patients with SEC were discharged from the hospital on anticoagulation. Eight patients were not discharged on anticoagulation due to no documented history of AF, and 6 patients had contraindications to anticoagulation.
The majority of procedures was performed utilizing the balloon-expandable Edwards-SAPIEN system. There were no significant differences in prosthesis size or type or access route, or intraoperative complications between patients with LAT, SEC or neither (Table 3). In all cases of LAT, postprocedure TOE demonstrated no evidence of thrombus embolization.
Table 3:
Intraoperative variables
| No LAT/SEC (n = 330) | LAT (n = 14) | SEC (n = 25) | P-value | |
|---|---|---|---|---|
| Prosthesis type | ||||
| CoreValve | 8 (2) | 1 (7) | 1 (4) | 0.408 |
| SAPIEN | 234 (71) | 9 (64) | 14 (56) | |
| SAPIEN XT | 54 (16) | 4 (29) | 6 (24) | |
| SAPIEN 3 | 34 (10) | 0 (0) | 4 (16) | |
| Valve size (mm) | ||||
| 20 | 2 (<1) | 0 (0) | 1 (4) | 0.070 |
| 23 | 169 (51) | 8 (57) | 13 (52) | |
| 26 | 135 (41) | 5 (36) | 9 (36) | |
| 29 | 22 (7) | 0 (0) | 2 (8) | |
| 31 | 2 (<1) | 1 (7) | 0 (0) | |
| Transfemoral approach | 203 (62) | 8 (57) | 19 (76) | 0.346 |
| IABP prior to procedure | 5 (1.5) | 0 (0) | 0 (0) | 1.000 |
| IABP placed emergently | 5 (1.5) | 1 (7) | 0 (0) | 0.256 |
| Emergent CPB | 4 (1.2) | 0 (0) | 0 (0) | 1.000 |
| Conversion to open | 2 (0.6) | 0 (0) | 0 (0) | 1.000 |
Values are presented as mean ± standard deviation or n (%).
CPB: cardiopulmonary bypass; IABP: intra-aortic balloon pump; LAT: left atrial appendage thrombus; SEC: spontaneous echo contrast.
No significant differences were noted between patients with LAT, SEC or neither with regard to the incidence of postprocedure complications, including any embolic or haemorrhagic complication (Table 4). Specifically, no patient with LAT or SEC experienced a postoperative neurological event at 30 days, 1 year or at last follow-up. While there were no significant differences in 30-day mortality among groups and no patient with LAT dying at 30 days, there were significant differences in mortality between groups at 1 year [50% (LAT) vs 20% (neither) vs 19% (SEC); P = 0.02]. Postoperative outcomes are listed in Table 4.
Table 4:
Postoperative outcomes
| No LAT/SEC (n = 330) | LAT (n = 14) | SEC (n = 25) | P-value | |
|---|---|---|---|---|
| MI | 1 (<1) | 0 (0) | 0 (0) | 1.000 |
| Renal failure | 6 (1.8) | 0 (0) | 1 (4) | 0.546 |
| Stroke/TIA 30 days | 7 (2) | 0 (0) | 0 (0) | 1.000 |
| Stroke/TIA 1 year | 14 (4.2) | 0 (0) | 0 (0) | 1.000 |
| Major bleed | 67 (20) | 3 (21) | 6 (24) | 0.852 |
| Minor bleed | 62 (19) | 2 (14) | 1 (4) | 0.166 |
| Life-threatening bleed | 13 (4) | 0 (0) | 0 (0) | 1.000 |
| Major vascular complication | 5 (2) | 0 (0) | 0 (0) | 1.000 |
| Minor vascular complication | 10 (3) | 1 (7) | 1 (4) | 0.368 |
| Vascular access complication with treatment | 9 (3) | 0 (0) | 1 (4) | 0.678 |
| AI ≥ moderate (postoperative) | 11 (3) | 2 (14) | 1 (4) | 0.085 |
| 30-Day mortality | 13 (4) | 0 (0) | 3 (12) | 0.196 |
| 1-Year mortality | 62 (19) | 7 (50) | 5 (20) | 0.024 |
| Long-term survival | 221 (67) | 6 (43) | 16 (64) | 0.180 |
Values are displayed as mean ± SD or n (%).
AI: aortic insufficiency; LAT: left atrial appendage thrombus; MI: myocardial infarction; SEC: spontaneous echo contrast; TIA: transient ischemic attack.
Thirty-day mortality was 4.3% (16). In multivariable logistic regression analysis, preprocedure AF, 30-day stroke/transient ischemic attack (TIA), life-threatening bleeding and postprocedure aortic regurgitation of moderate degree or greater were independent predictors of 30-day mortality (Table 5). One-year mortality was 20% (74). In multivariable logistic regression analysis, preprocedure AF, LAT, life-threatening bleeding and preprocedure brain natriuretic peptide were independent predictors of 1-year mortality (Table 6).
Table 5:
Predictors of 30-day mortality
| Univariable |
Multivariable |
|||||
|---|---|---|---|---|---|---|
| Survivors (n = 353) | Non-survivors (n = 16) | Odds ratio (95% CI) | P-value | Odds ratio (95% CI) | P-value | |
| Age (years) | 85 ± 7.4 | 82 ± 9.0 | 0.960 (0.904–1.018) | 0.171 | ||
| Female gender | 185 (52) | 8 (50) | 0.908 (0.332–2.482) | 0.850 | ||
| PCI | 144 (41) | 5 (31) | 0.660 (0.224–1.946) | 0.450 | ||
| CABG | 95 (27) | 3 (19) | 0.627 (0.174–2.257) | 0.473 | ||
| Prior MI | 71 (20) | 3 (19) | 0.917 (0.253–3.318) | 0.894 | ||
| CVD | 76 (22) | 6 (38) | 2.187 (0.768–6.230) | 0.142 | ||
| PVD | 89 (25) | 3 (19) | 0.685 (0.180–2.468) | 0.561 | ||
| Hypertension | 312 (88) | 13 (81) | 0.569 (0.155–2.092) | 0.395 | ||
| Hypercholesterolaemia | 277 (78) | 13 (81) | 1.189 (0.329–4.298) | 0.791 | ||
| Diabetes mellitus | 119 (34) | 3 (19) | 0.454 (0.126–1.630) | 0.224 | ||
| Haemodialysis | 14 (4) | 0 (0) | Only survivors | 1.000 | ||
| BNP | 732 ± 814 | 906 ± 846 | 1.000 (1.000–1.001) | 0.409 | ||
| CLD | 98 (28) | 5 (31) | 1.217 (0.411–3.606) | 0.722 | ||
| NYHA III–IV | 225 (64) | 9 (56) | 0.731 (0.265–2.018) | 0.544 | ||
| STS score | 8.75 ± 4.8 | 10.2 ± 5.2 | 1.053 (0.931–1.192) | 0.409 | ||
| Preprocedure AF | 128 (36) | 14 (88) | 12.457 (2.77–55.96) | 0.001 | 38.71 (3.264–459) | 0.004 |
| SAPIENS (prosthesis type) | 243 (69) | 14 (88) | 3.169 (0.705–14.25) | 0.132 | ||
| Transfemoral approach | 220 (63) | 10 (63) | 1.008 (0.357–2.846) | 0.989 | ||
| Ejection fraction (%) | 51.5 ± 14.0 | 46.8 ± 14.0 | 0.979 (0.946–1.013) | 0.211 | ||
| LAT | 14 (4) | 0 (0) | Only survivors | 1.000 | ||
| SEC | 22 (6) | 3 (19) | 3.472 (0.917–13.15) | 0.066 | 4.141 (0.843–20.36) | 0.079 |
| AI ≥ moderate | 18 (5) | 1 (6) | 1.241 (0.154–9.991) | 0.839 | ||
| MR ≥ moderate | 69 (20) | 8 (50) | 4.115 (1.487–11.39) | 0.006 | 1.990 (0.547–7.243) | 0.295 |
| TR ≥ moderate | 58 (16) | 2 (13) | 0.727 (0.160–3.299) | 0.678 | ||
| Stroke/TIA 30 days | 5 (1) | 2 (13) | 9.940 (1.762–56.10) | 0.009 | 13.72 (1.604–117) | 0.016 |
| Major bleed | 74 (21) | 2 (13) | 0.539 (0.119–2.435) | 0.420 | ||
| Minor bleed | 63 (18) | 2 (13) | 0.658 (0.145–2.981) | 0.586 | ||
| Life-threatening bleed | 8 (2) | 5 (31) | 19.61 (5.491–69.98) | <0.001 | 87.36 (8.665–881) | <0.001 |
| Major vascular complication | 4 (1) | 1 (6) | 5.817 (0.608–55.68) | 0.125 | ||
| Minor vascular complication | 11 (3) | 1 (6) | 2.073 (0.249–17.24) | 0.499 | ||
| Vascular access complication with treatment | 9 (3) | 1 (6) | 2.548 (0.301–21.59) | 0.389 | ||
| AI ≥ moderate (postoperative) | 11 (3) | 3 (19) | 7.174 (1.776–28.98) | 0.006 | 7.846 (1.526–40.34) | 0.013 |
Values are presented as mean ± standard deviation or n (%).
AF: atrial fibrillation; AI: aortic insufficiency; BNP: B-type natriuretic peptide; CABG: coronary artery bypass graft; CI: confidence interval; CLD: chronic lung disease; CVD: cerebral vascular disease; LAT: left atrial appendage thrombus; MI: myocardial infarction; MR: mitral regurgitation; NYHA: New York Heart Association; PCI: percutaneous coronary intervention; PVD: peripheral vascular disease; SEC: spontaneous echo contrast; STS: Society of Thoracic Surgeons; TIA: transient ischemic attack; TR: tricuspid regurgitation.
Table 6:
Predictors of 1-year mortality
| Univariable |
Multivariable |
|||||
|---|---|---|---|---|---|---|
| Survivors (n = 259) | Non-survivors (n = 74) | Odds ratio (95% CI) | P-value | Odds ratio (95% CI) | P-value | |
| Age (years) | 85 ± 6.9 | 84 ± 9.3 | 0.983 (0.952–1.016) | 0.302 | ||
| Female gender | 160 (54) | 33 (45) | 0.679 (0.406–1.136) | 0.139 | ||
| PCI | 120 (41) | 29 (39) | 0.940 (0.557–1.586) | 0.816 | ||
| CABG | 79 (27) | 19 (26) | 0.945 (0.520–1.693) | 0.848 | ||
| Prior MI | 59 (20) | 15 (20) | 1.017 (0.538–1.922) | 0.959 | ||
| CVD | 60 (20) | 22 (30) | 1.657 (0.932–2.946) | 0.084 | 1.449 (0.731–2.871) | 0.286 |
| PVD | 75 (25) | 17 (23) | 0.875 (0.478–1.600) | 0.663 | ||
| Hypertension | 260 (88) | 65 (88) | 0.972 (0.440–2.129) | 0.944 | ||
| Hypercholesterolaemia | 229 (78) | 61 (82) | 1.352 (0.699–2.618) | 0.369 | ||
| Diabetes mellitus | 102 (35) | 20 (27) | 0.701 (0.397–1.237) | 0.219 | ||
| Haemodialysis | 7 (2) | 7 (9) | 4.299 (1.453–12.71) | 0.008 | 2.561 (0.646–10.15) | 0.179 |
| BNP | 662 ± 727 | 1048 ± 1049 | 1.001 (1.000–1.001) | 0.001 | 1.000 (1.000–1.001) | 0.041 |
| CLD | 77 (26) | 24 (32) | 1.359 (0.781–2.364) | 0.276 | ||
| NYHA III–IV | 180 (61) | 54 (73) | 1.725 (0.970–3.037) | 0.058 | 1.529 (0.784–2.970) | 0.211 |
| STS score | 8.0 ± 4.0 | 11.7 ± 6.2 | 1.165 (1.080–1.256) | <0.001 | ||
| Preprocedure AF | 101 (34) | 41 (55) | 2.423 (1.441–4.074) | 0.001 | 2.115 (1.110–4.027) | 0.022 |
| SAPIENS (prosthesis type) | 202 (78) | 55 (74) | 1.3327 (0.747–2.377) | 0.329 | ||
| Transfemoral approach | 187 (63) | 43 (58) | 0.801 (0.470–1.349) | 0.402 | ||
| Ejection fraction (%) | 52 (14) | 48.5 (14.5) | 0.983 (0.966–1.001) | 0.065 | 0.988 (0.966–1.011) | 0.300 |
| LAT | 7 (2) | 7 (9) | 4.299 (1.453–12.71) | 0.008 | 3.573 (1.040–12.28) | 0.042 |
| SEC | 20 (7) | 5 (7) | 0.996 (0.350–2.758) | 0.994 | ||
| AI ≥ moderate | 17 (6) | 2 (3) | 0.454 (0.102–2.021) | 0.299 | ||
| MR ≥ moderate | 55 (19) | 22 (30) | 1.846 (1.034–3.298) | 0.038 | 1.267 (0.625–2.569) | 0.510 |
| TR ≥ moderate | 42 (14) | 18 (24) | 1.936 (1.036–3.619) | 0.038 | 1.182 (0.553–2.526) | 0.665 |
| Stroke/TIA 30 days | 3 (1) | 4 (5) | 5.562 (1.211–25.55) | 0.027 | 5.823 (0.968–35.05) | 0.053 |
| Major bleed | 59 (20) | 17 (23) | 1.193 (0.646–2.205) | 0.572 | ||
| Minor bleed | 50 (17) | 15 (20) | 1.246 (0.653–2.375) | 0.503 | ||
| Life-threatening bleed | 3 (1) | 10 (14) | 15.10 (4.052–57.08) | <0.001 | 19.35 (4.741–79.00) | <0.001 |
| Major vascular complication | 4 (1) | 1 (1) | 0.997 (0.109–9.117) | 0.998 | ||
| Minor vascular complication | 8 (3) | 4 (5) | 2.050 (0.598–7.030) | 0.252 | ||
| Vascular access complication with treatment | 9 (3) | 1 (1) | 0.435 (0.054–3.515) | 0.434 | ||
| AI ≥ moderate (postoperative) | 8 (3) | 6 (8) | 3.166 (1.059–9.459) | 0.038 | 2.521 (0.682–9.318) | 0.164 |
Values are presented as mean ± standard deviation or n (%).
AF: atrial fibrillation; AI: aortic insufficiency; BNP: B-type natriuretic peptide; CABG: coronary artery bypass graft; CI: confidence interval; CLD: chronic lung disease; CVD: cerebral vascular disease; LAT: left atrial appendage thrombus; MI: myocardial infarction; MR: mitral regurgitation; NYHA: New York Heart Association; PCI: percutaneous coronary intervention; PVD: peripheral vascular disease; SEC: spontaneous echo contrast; STS: Society of Thoracic Surgeons; TIA: transient ischemic attack; TR: tricuspid regurgitation.
Overall mortality at an average follow-up of 585 days was 34% (126). Independent predictors of overall mortality as determined by Cox regression analysis included AF [hazard ratio (HR) 1.7, 95% confidence interval (CI) 1.15–2.5; P = 0.008], chronic lung disease (HR 1.72, 95% CI 1.15–2.6; P = 0.008) and life-threatening bleeding (HR 8.03, 95% CI 4.0–16.1; P = 0.00). LAT demonstrated a strong trend towards independently predicting overall mortality (HR 2.3, 95% CI 0.95–5.54; P = 0.065). Kaplan–Meier analysis revealed increased overall mortality in patients with LAT (log-rank P = 0044; Fig. 1). Preprocedure AF was the only independent predictor of LAT (Table 7).
Figure 1:
Overall survival based on the presence of LAT, SEC or neither. LAT: left atrial appendage thrombus; SEC: spontaneous echo contrast.
Table 7:
Predictors of thrombus (LAT)
| Multivariable |
||
|---|---|---|
| Hazard ratio (95% CI) | P-value | |
| CVD | 2.043 (0.623–6.695) | 0.237 |
| Haemodialysis | 2.965 (0.513–17.132) | 0.223 |
| NYHA III–IV | 3.337 (0.688–16.185) | 0.133 |
| Preprocedure AF | 7.620 (1.602–36.26) | 0.010 |
| AI ≥ moderate (postoperative) | 2.725 (0.481–15.41) | 0.255 |
Values are presented as mean ± standard deviation or n (%).
AF: atrial fibrillation; AI: aortic insufficiency; CI: confidence interval; CVD: cerebral vascular disease; LAT: left atrial appendage thrombus; NYHA: New York Heart Association.
DISCUSSION
In our series of patients undergoing TAVR for native aortic valve stenosis or bioprosthetic aortic valve degeneration, the incidence of LAT and SEC was 3.8% and 6.8%, respectively. Both findings were associated with, but not always found in, the presence of AF. While neither was an independent predictor of 30-day mortality on multivariable analysis, LAT was an independent predictor of 1-year mortality and AF was an independent predictor of both. Finally, neither LAT nor SEC increased the risk of procedural complications, specifically embolic events, including stroke or TIA.
In the high-risk population undergoing evaluation for TAVR, TOE may detect LAT in up to 10% of patients, and SEC in another 24% of patients [8]. LAT and SEC are known risk factors for cardioembolic events, including stroke. Subsequent management of this finding in such patients otherwise meeting criteria for TAVR is to date not clearly defined.
In the initial Edwards PARTNER and CoreValve pivotal trials, 30-day major stroke rates of 2.3–5% were noted, and in the PARTNER trials, they were significantly higher in TAVR patients than in surgical patients [1–4]. As TAVR has been rapidly adopted into standard clinical practice, experience and technical modifications have resulted in improved outcomes. Despite this, a significant stroke risk still exists. It was hypothesized that manipulation of large catheters in the ascending aorta and arch, as is necessary during transfemoral approaches, may result in embolization of aortic atheroma. However, in subsequent studies comparing outcomes after transapical versus transfemoral approaches, no difference was found in stroke rates [9, 10]. These findings have been upheld in propensity analyses, given the inherent differences in baseline characteristics between transfemoral and transapical candidates [11]. Other theories focused on prosthesis type, as stroke rates in the balloon-expandable Edwards Partner trials were higher than those of the self-expandable CoreValve pivotal trials [1–4]. This hypothesis has also not been confirmed in larger registry studies and direct comparisons of balloon- versus self-expandable valves [10, 12].
It is likely that the majority of strokes after TAVR are the result of embolization during crossing of the calcified aortic valve or valve deployment. In studies utilizing embolic protection devices, captured debris is typically of arterial and valvular origin [13], thus suggesting that neurological events after TAVR are the result of thromboembolism from a calcified aorta, aortic valve and annulus. It has been suggested that certain technical aspects of the procedure may be modified to reduce this risk, such as the use of preballoon valvuloplasty [14]. Improvements in technology, patient selection and/or operator experience have also resulted in lower stroke rates and mortality [10, 15–17].
In addition to procedural factors, several studies also point to pre-existing baseline characteristics that contribute to postoperative neurological and overall outcomes. In a study by Bosmans et al., of the 996 patients undergoing TAVR with a self-expandable valve, the incidence of periprocedural, early and late neurological events was 1.6%, 1.8% and 3.8%, respectively. While no single baseline factor independently predicted periprocedural neurological events, AF was an independent predictor of early neurological events, with such patients experiencing worse longer term survival [18]. Additional studies have similarly shown that preprocedure AF is a marker for increased long-term overall and cardiovascular mortality as well as worsened New York Heart Association functional class and increased heart failure rehospitalizations [6]. AF may both increase the risk for neurological events and act as a marker for more advanced cardiac disease, resulting in increased long-term cardiac morbidity.
In the landmark TAVR trials, a number of exclusion criteria were imposed, among which were intracardiac thrombi [1–4]. Subsequent studies analysing outcomes of ‘real-world’ patients who do not fulfil these criteria demonstrated acceptable, if not improved, results [5]. However, no study has assessed the outcomes of TAVR in patients with LAT or SEC. These findings are associated with an increased baseline risk of neurological and other embolic events, and inadvertent catheter manipulation may theoretically accelerate this process. In the only report of a patient with AF and LAT undergoing TAVR, frequent wire prolapse through the mitral valve into the left atrium was noted. Embolization of thrombus to the aortic valve was noted, and percutaneous embolectomy was performed successfully with an otherwise uneventful postprocedural course [19]. Interestingly, in a separate report of TAVR in the presence of left ventricular thrombus utilizing a cerebral embolic protection device, embolization did not occur [20].
The 2006 AHA/ACC/HRS Guideline for the Management of Patients with Atrial Fibrillation considers it a Class IIA indication in patients with LAT being considered for cardioversion to anticoagulate with warfarin to an international normalized ration (INR) of 2–3 for 3 weeks prior to and 4 weeks after cardioversion [21]. In our institutional practice, if LAT was identified intraoperatively, in most cases the procedure was performed and anticoagulation instituted/continued postoperatively if tolerated. If LAT was identified preoperatively, anticoagulation was initiated if the patient was not already on it and the procedure delayed 4–6 weeks. In most cases, a patient in AF with LAT is already on anticoagulation and thus options for further treatment are limited. In the case of LAT in a patient undergoing SAVR, a concomitant maze procedure (if the patient has AF) would be performed and the left atrial appendage ligated with extirpation of LAT.
In the current study, no adverse neurological or other embolic events were noted in patients with LAT or SEC undergoing TAVR. In all cases with LAT, thrombus was still present postprocedure. At an average long-term follow-up of up to ∼1.6 years, no neurological events were noted in patients with LAT or SEC. However, when independent predictors of 30-day and 1-year mortality were assessed, while neither LAT nor SEC was predictive at 30 days, LAT was a predictor at 1 year, and AF was a predictor of both. Long-term outcome was influenced by the presence of LAT. It is likely that patients with LAT represent a subgroup of AF patients with more advanced cardiac disease and more long-standing AF which is reflected in their poorer long-term survival. However, there was no evidence of procedure-related embolic events in patients with LAT and thus no influence on short-term 30-day survival.
The 1-year survival rate of patients with LAT was 50%. This is worse than traditional estimates of the natural survival of patients with severe symptomatic AS managed medically, although the patients in this study had a greater burden of comorbidity than historical controls. Regardless, one must wonder whether TAVR should be attempted in such patients, even in the absence of any embolic risk, given their poor overall prognosis. It is also unclear whether strategies aimed at eliminating AF and LAT would be beneficial in such patients undergoing TAVR. These points require further study.
Limitations
Limitations of this study include those inherent to a retrospective analysis utilizing chart review, including incomplete data, potential inaccuracies in data and potential for selection bias. A major limitation was the small sample size, in particular the small number of patients with LAT and SEC which form the basis of this study. In addition, due to differences in clinical practice across centres, extrapolation of results may be of limited value. Evidence suggesting improved left atrial and left atrial appendage function after TAVR suggests that resolution of thrombus may be accelerated with anticoagulation [22]. Data regarding long-term follow-up with TOE to assess for resolution of LAT and SEC were not available, which is another limitation of the current study.
CONCLUSION
In conclusion, although sample sizes were limited, the current study suggests that TAVR may be performed in patients with LAT and SEC with reasonably safe outcomes from an embolic and neurological risk point of view. However, LAT and SEC exist predominantly in association with AF. As a result, their effect on mortality is primarily a function of the effect of AF and the associated cardiac dysfunction, which are known markers of reduced long-term mortality. Larger, randomized studies are needed to better study this phenomenon.
Funding
This work was supported in part by the National Institutes of Health [grant T35EB006732].
Conflict of interest: Arash Salemi is a clinical proctor for Medtronic, Inc. and Edwards Lifesciences. All other authors have no conflicts to declare.
REFERENCES
- 1. Smith CR, Leon MB, Mack MJ, Miller DC, Moses JW, Svensson LG. et al. Transcatheter versus surgical aortic-valve replacement in high-risk patients. N Engl J Med 2011;364:2187–98. [DOI] [PubMed] [Google Scholar]
- 2. Leon MB, Smith CR, Mack M, Miller DC, Moses JW, Svensson LG. et al. Transcatheter aortic-valve implantation for aortic stenosis in patients who cannot undergo surgery. N Engl J Med 2010;363:1597–607. [DOI] [PubMed] [Google Scholar]
- 3. Adams DH, Popma JJ, Reardon MJ, Yakubov SJ, Coselli JS, Deeb GM. et al. Transcatheter aortic-valve replacement with a self-expanding prosthesis. N Engl J Med 2014;370:1790–8. [DOI] [PubMed] [Google Scholar]
- 4. Popma JJ, Adams DH, Reardon MJ, Yakubov SJ, Kleiman NS, Heimansohn D. et al. Transcatheter aortic valve replacement using a self-expanding bioprosthesis in patients with severe aortic stenosis at extreme risk for surgery. J Am Coll Cardiol 2014;63:1972–81. [DOI] [PubMed] [Google Scholar]
- 5. Sedaghat A, Al-Rashid F, Sinning J-M, Wendt D, Thielmann M, Grube E. et al. Outcome in TAVI patients with symptomatic aortic stenosis not fulfilling PARTNER study inclusion criteria. Catheter Cardiovasc Interv 2015;86:1097–104. [DOI] [PubMed] [Google Scholar]
- 6. Chopard R, Teiger E, Meneveau N, Chocron S, Gilard M, Laskar M. et al. Baseline characteristics and prognostic implications of pre-existing and new-onset atrial fibrillation after transcatheter aortic valve implantation: results from the FRANCE-2 Registry. JACC Cardiovasc Interv 2015;8:1346–55. [DOI] [PubMed] [Google Scholar]
- 7. Saric M, Armour AC, Arnaout MS, Chaudhry FA, Grimm RA, Kronzon I. et al. Guidelines for the use of echocardiography in the evaluation of a cardiac source of embolism. J Am Soc Echocardiogr 2016;29:1–42. [DOI] [PubMed] [Google Scholar]
- 8. Lenders GD, Paelinck BP, Wouters K, Claeys MJ, Rodrigus IE, Van Herck PL. et al. Transesophageal echocardiography for cardiac thromboembolic risk assessment in patients with severe, symptomatic aortic valve stenosis referred for potential transcatheter aortic valve implantation. Am J Cardiol 2013;111:1470–4. [DOI] [PubMed] [Google Scholar]
- 9. Rodés-Cabau J, Dumont E, Boone RH, Larose E, Bagur R, Gurvitch R. et al. Cerebral embolism following transcatheter aortic valve implantation: comparison of transfemoral and transapical approaches. J Am Coll Cardiol 2011;57:18–28. [DOI] [PubMed] [Google Scholar]
- 10. Athappan G, Gajulapalli RD, Sengodan P, Bhardwaj A, Ellis SG, Svensson L. et al. Influence of transcatheter aortic valve replacement strategy and valve design on stroke after transcatheter aortic valve replacement: a meta-analysis and systematic review of literature. J Am Coll Cardiol 2014;63:2101–10. [DOI] [PubMed] [Google Scholar]
- 11. Blackstone EH, Suri RM, Rajeswaran J, Babaliaros V, Douglas PS, Fearon WF. et al. Propensity-matched comparisons of clinical outcomes after transapical or transfemoral transcatheter aortic valve replacement: a placement of aortic transcatheter valves (PARTNER)-I trial substudy. Circulation 2015;131:1989–2000. [DOI] [PubMed] [Google Scholar]
- 12. Abdel-Wahab M, Neumann F-J, Mehilli J, Frerker C, Richardt D, Landt M. et al. 1-year outcomes after transcatheter aortic valve replacement with balloon-expandable versus self-expandable valves: results from the CHOICE randomized clinical trial. J Am Coll Cardiol 2015;66:791–800. [DOI] [PubMed] [Google Scholar]
- 13. Van Mieghem NM, El Faquir N, Rahhab Z, Rodríguez-Olivares R, Wilschut J, Ouhlous M. et al. Incidence and predictors of debris embolizing to the brain during transcatheter aortic valve implantation. JACC Cardiovasc Interv 2015;8:718–24. [DOI] [PubMed] [Google Scholar]
- 14. Bijuklic K, Haselbach T, Witt J, Krause K, Hansen L, Gehrckens R. et al. Increased risk of cerebral embolization after implantation of a balloon-expandable aortic valve without prior balloon valvuloplasty. J Am Coll Cardio Interv 2015;8:1608–13. [DOI] [PubMed] [Google Scholar]
- 15. Holmes DR, Nishimura RA, Grover FL, Brindis RG, Carroll JD, Edwards FH. et al. Annual outcomes with transcatheter valve therapy: from the STS/ACC TVT registry. J Am Coll Cardiol 2015;66:2813–23. [DOI] [PubMed] [Google Scholar]
- 16. Beohar N, Kirtane AJ, Blackstone E, Waksman R, Holmes D, Minha S. et al. Trends in complications and outcomes of patients undergoing transfemoral transcatheter aortic valve replacement: experience from the PARTNER continued access registry. JACC Cardiovasc Interv 2016;9:355–63. [DOI] [PubMed] [Google Scholar]
- 17. Walther T, Hamm CW, Schuler G, Berkowitsch A, Kotting J, Mangner N. et al. Perioperative results and complications in 15,964 transcatheter aortic valve replacements: prospective data from the GARY registry. J Am CollCardiol 2015;65:2173–80. [DOI] [PubMed] [Google Scholar]
- 18. Bosmans J, Bleiziffer S, Gerckens U, Wenaweser P, Brecker S, Tamburino C. et al. The incidence and predictors of early- and mid-term clinically relevant neurological events after transcatheter aortic valve replacement in real-world patients. J Am Coll Cardiol 2015;66:209–17. [DOI] [PubMed] [Google Scholar]
- 19. Williams PD, Belder MA, de Maredia N, Muir DF.. Embolization of left atrial appendage thrombus during transcatheter aortic valve replacement: a potential mechanism of periprocedural stroke. J Am Coll Cardidiol 2015;8:1770–1. [DOI] [PubMed] [Google Scholar]
- 20. Grover PM, O’Neill BP, Velazquez O, Heldman AW, O’Neill WW, Cohen MG.. Cerebral protection against left ventricular thrombus during transcatheter aortic valve replacement in a patient with critical aortic stenosis. Tex Heart Inst J 2013;40:477–80. [PMC free article] [PubMed] [Google Scholar]
- 21. Fuster V, Ryden LE, Cannom DS, Crijns AB, Curtis AB, Ellenbogen KA. et al. ACC/AHA/ESC 2006 guidelines for the management of patients with atrial fibrillation: a report of the American College of Cardiology/American Heart Association task force on practice guidelines and the European Society of Cardiology Committee for Practice Guidelines (Writing Committee to Revise the 2001 Guidelines for the Management of Patients With Atrial Fibrillation): developed in collaboration with the European Heart Rhythm Association and the Heart Rhythm Society. Circulation 2006;114:e257–354. [DOI] [PubMed] [Google Scholar]
- 22. Aslan S, Gul M, Cakmak HA, Ozturk D, Celik O, Satilmisoglu M. et al. Short-term effects of transcatheter aortic valve implantation on left atrial appendage function. Cardiol J 2015;22:527–34. [DOI] [PubMed] [Google Scholar]