Transcatheter aortic valve implantation : is an acute improvement in left ventricular ejection fraction as assessed by 3D echocardiography associated to further functional improvement at follow-up?

Transcatheter aortic valve implantation (TAVI) has demonstrated excellent immediate and long-term results in high-risk patients with severe aortic stenosis (AS), similar to surgical aortic valve replacement (AVR) [1]. When severe AS is associated with left ventricular (LV) dysfunction, AVR still results in significant improvement of LV function and survival [2] and [3]. Similarly, significant improvement in LV ejection fraction (EF) has been demonstrated after TAVI regardless of its baseline value [4] and [5]. Since patients with baseline LV dysfunction were at higher risk of major adverse cardiovascular events, accurate evaluation of LVEF and understanding the effect of afterload reduction on acute and mid-term LVEF changes are mandatory. However, the correlation between acute and mid-term LVEF has not been investigated. Aims of this study were to evaluate in TAVI patients whether 1) acute changes in LVEF predict mid-term LVEF improvement; 2) the co-existence of coronary artery disease (CAD) and/or other clinical variables may influence LVEF changes and/or preclude its improvement. A total of 143 consecutive patients with symptomatic severe AS undergoing TAVI with Edwards-SAPIEN valve were enrolled in this study. The institutional review board approved the study. Severe AS was defined as an aortic valve area < 1 cm2 or a mean transaortic pressure gradient ≥ 40 mm Hg [6]. All procedures were monitored by intraoperative three-dimensional transesophageal echocardiography (3DTEE) and offline volumetric analysis was performed. Conventional 2D and Doppler transthoracic echocardiography was obtained at baseline, before hospital discharge and at 6 months follow-up. Forty-three patients were excluded because of death within 6 months after procedure (11 patients), loss at follow-up (6 patients) or more than mild post-TAVI aortic regurgitation (26 patients). The final study sample was divided in Group 1 composed of 48 patients with 3D-LVEF < 50% at baseline and Group 2 composed of 52 patients with 3D-LVEF > 50%. A cut-off value of ≥ 5% has been defined as significant LVEF improvement on the basis of inter- and intra-observer variability. Group 1 had significant higher logistic EuroSCORE and had more frequently a NYHA functional class IV (Table 1). No differences between the two groups were found as concerns the coexistence of CAD, systemic arterial pressures, and transfemoral vs. transapical approaches. Table 2 reports 2D and 3D echocardiographic characteristics of the population at baseline and at 6 months. The lower LVEF at baseline in Group 1 was associated with larger LV and left atrial volumes and greater LV mass. Group 1 showed lower mean and peak transaortic gradients associated with higher PASP, higher E/e′ ratio and lower mitral deceleration time. At 6 months, prosthetic valve performance, PASP and diastolic parameters were similar in the two groups, while LVEF in Group 1 was still lower than in Group 2, but reached a mean value ≥ 50%. Table 1. Baseline clinical characteristics of the study population. Overall (n = 100) Group 1 -------------------------------------------------------------------------------- Group 2 (n = 52) p1 p2 All (n = 48) No acute LVEF improvement (n = 18) Acute LVEF improvement (n = 30) Male, n (%) 34 (34%) 23 (48%) 8 (44%) 15 (50%) 11 (21%) 0.005 0.709 Age (years) 82 ± 6 82 ± 7 83 ± 6 81 ± 7 82 ± 6 0.928 0.436 Body surface area (m2) 1.7 ± 0.2 1.7 ± 0.2 1.7 ± 0.2 1.7 ± 0.2 1.7 ± 0.2 0.062 1.000 Logistic EuroSCORE (%) 20.8 ± 12.1 25.7 ± 13.8 24.2 ± 12.4 26.6 ± 14.7 16.4 ± 8.1 < 0.001 0.690 NYHA class 0.010 0.191 I 0 (0%) 0 (0%) 0 (0%) 0 (0%) 0 (0%) II 23 (23%) 6 (13%) 2 (11%) 4 (13%) 17 (33%) III 53 (53%) 25 (52%) 7 (39%) 18 (60%) 28 (54%) IV 24 (24%) 17 (35%) 9 (50%) 8 (27%) 7 (13%) Transfemoral approach, n (%) 76 (76%) 38 (79%) 13 (72%) 25 (83%) 38 (73%) 0.476 0.359 Prosthesis size 0.036 0.399 23 42 (42%) 15 (31%) 6 (33%) 9 (30%) 27 (52%) 26 58 (58%) 33 (69%) 12 (67%) 21 (70%) 25 (48%) Angina, n (%) 35 (35%) 14 (29%) 6 (44%) 8 (27%) 21 (40%) 0.198 0.522 Dyspnea, n (%) 90 (90%) 47 (98%) 18 (100%) 29 (97%) 43 (82%) 0.022 0.634 Syncope, n (%) 20 (20%) 5 (10%) 2 (11%) 3 (10%) 15 (29%) 0.023 0.467 High blood pressure, n (%) 90 (90%) 42 (88%) 13 (72%) 29 (97%) 48 (92%) 0.442 0.056 CAD, n (%) 56 (56%) 30 (62%) 9 (50%) 21 (70%) 26 (50%) 0.242 0.194 Peripheral vascular disease, n (%) 46 (46%) 18 (38%) 6 (33%) 12 (40%) 28 (54%) 0.119 0.673 p1: for comparison between Group 1 and Group 2. p2: for comparison between no acute vs. acute LVEF improvement patients. EuroSCORE = European System for Cardiac Operative Risk Evaluation; NYHA = New York Heart Association; BP = blood pressure; CAD = coronary artery disease. Full-size table Table optionsView in workspaceDownload as CSV Table 2. Comparison of echocardiographic parameters at baseline and 6 months follow-up. Baseline -------------------------------------------------------------------------------- 6 Months -------------------------------------------------------------------------------- Overall (n = 100) Group 1 (n = 48) Group 2 (n = 52) Overall (n = 100) Group 1 (n = 48) Group 2 (n = 52) 2D transthoracic echocardiography 2D-LVEDV index (ml/m2) 63 ± 22 76 ± 23 51 ± 14⁎ 61 ± 20 72 ± 21 50 ± 13⁎ 2D-LVESV index (ml/m2) 32 ± 19 46 ± 17 18 ± 8⁎ 26 ± 14 35 ± 15† 19 ± 8⁎ 2D-LVEF (%) 52.7 ± 15.1 39.7 ± 8.3 64.8 ± 8.4⁎ 58.4 ± 9.7 52.9 ± 8.5† 63.5 ± 7.7⁎ LA volume (ml) 100 ± 40 111 ± 37 91 ± 41⁎ 93 ± 40 107 ± 40 82 ± 36⁎‡ LV mass (g) 259 ± 79 283 ± 93 240 ± 59⁎ 222 ± 89 254 ± 108† 198 ± 64⁎‡ AVA index (cm2/m2) 0.36 ± 0.09 0.35 ± 0.10 0.37 ± 0.08 1.16 ± 0.30 1.11 ± 0.28† 1.20 ± 0.33‡ Mean gradient (mm Hg) 50 ± 15 44 ± 14 55 ± 13⁎ 11 ± 4 10 ± 4† 11 ± 4‡ Peak gradient (mm Hg) 81 ± 23 72 ± 22 88 ± 21⁎ 20 ± 7 19 ± 6† 21 ± 7‡ PASP (mm Hg) 42 ± 11 46 ± 12 39 ± 10⁎ 36 ± 8 36 ± 8† 36 ± 8‡ Mitral E/A ratio 1.0 ± 0.6 1.2 ± 0.8 0.9 ± 0.4 0.9 ± 0.5 1.0 ± 0.7† 0.8 ± 0.3 Mitral E/e′ 18.8 ± 6.7 21.0 ± 7.8 17.3 ± 5.3⁎ 15.9 ± 4.5 15.7 ± 4.3† 16.1 ± 4.7 Mitral DT (ms) 241 ± 95 203 ± 93 271 ± 85⁎ 260 ± 76 262 ± 86† 258 ± 69 3D transesophageal echocardiography 3D-LVEDV index (ml/m2) 65 ± 23 79 ± 25 51 ± 11⁎ – – – 3D-LVESV index (ml/m2) 34 ± 21 50 ± 20 19 ± 6⁎ – – – 3D-LVEF (%) 51 ± 15 37.6 ± 7.1 64.2 ± 6.3⁎ – – – LV = left ventricular; EDV = end-diastolic volume; ESV = end-systolic volume; EF = ejection fraction; LA = left atrial; AVA = aortic valve area; PASP = pulmonary artery systolic pressure. ⁎p < 0.05: Group 1 vs. Group 2. †p < 0.05: 6 months vs. baseline in Group 1. ‡p < 0.05: 6 months vs. baseline in Group 2. Full-size table Table optionsView in workspaceDownload as CSV Group 1 showed a significant improvement in the 3D-LVEF immediately after TAVI procedure and at 6 months. At 6 months, a subgroup of 28 patients out of 48 reached a 2D-LVEF greater than 50%. Group 2 patients did not show any significant variation in LVEF. Regarding the immediate intraoperative evaluation, 30 patients out of 48 in Group 1 had an improvement > 5% and 18 patients did not show a meaningful improvement. At 6 months, further LVEF improvement occurred in 19 patients who had already had an immediate increase in LVEF and in 12 patients who did not show any acute LVEF improvement (Fig. 1). Fig. 1. Flowchart scheme showing the number and percentage of cases with acute 3D-LVEF improvement or no acute improvement in patients with low 3D-LVEF at baseline (Group 1) and the changes at 6 months. Figure optionsDownload full-size imageDownload high-quality image (104 K)Download as PowerPoint slide No clinical variable significantly discriminated patients with or without acute improvement in 3D-LVEF including the presence of CAD (Table 1). Patients with an acute improvement in LVEF had a peak and mean transaortic gradients at baseline significantly higher than those with no LVEF changes (Table 3). Interestingly, only cases in Group 1 with an acute improvement of LVEF > 5% showed a significant reduction in LV mass at follow-up. Table 3. Baseline and 6 months echocardiographic characteristics of patients with and without acute LVEF. Baseline -------------------------------------------------------------------------------- 6 Months -------------------------------------------------------------------------------- No acute LVEF improvement (n = 18) Acute LVEF improvement (n = 30) No acute LVEF improvement (n = 18) Acute LVEF improvement (n = 30) 2D-LVEDV index (ml/m2) 78 ± 24 75 ± 23 72 ± 22 72 ± 21 2D-LVESV index (ml/m2) 49 ± 18 45 ± 17 37 ± 17† 33 ± 14‡ 2D-LVEF (%) 37.0 ± 6.4 41.3 ± 9.0 50.3 ± 8.4† 54.5 ± 8.4‡ LA volume (ml) 97 ± 36 119 ± 35 90 ± 31 118 ± 41⁎ LV mass (g) 270 ± 85 290 ± 98 260 ± 88 249 ± 123‡ AVA index (cm2/m2) 0.34 ± 0.11 0.36 ± 0.09 1.12 ± 0.31† 1.10 ± 0.26‡ Mean gradient (mm Hg) 38 ± 12 48 ± 15⁎ 11 ± 5† 11 ± 3‡ Peak gradient (mm Hg) 63 ± 20 78 ± 23⁎ 19 ± 8† 19 ± 5‡ PASP (mm Hg) 43 ± 8 47 ± 14 38 ± 10 35 ± 7‡ Mitral E/A ratio 1.0 ± 0.4 1.0 ± 0.7 0.9 ± 0.6 1.0 ± 0.7 Mitral E/e′ 19.3 ± 6.5 16.5 ± 6.0 16.0 ± 4.7† 15.6 ± 4.1‡ Mitral DT (ms) 225 ± 63 219 ± 56 262 ± 93 263 ± 83‡ Improvement belonging to Group 1. LV = left ventricular; EDV = end-diastolic volume; ESV = end-systolic volume; EF = ejection fraction; LA = left atrial; AVA = aortic valve area; PASP = pulmonary artery systolic pressure. ⁎p < 0.05: no acute vs. acute LVEF improvement patients. †p < 0.05: 6 months vs. baseline in no acute LVEF improvement patients. ‡p < 0.05: 6 months vs. baseline in acute LVEF improvement patients. Full-size table Table optionsView in workspaceDownload as CSV The main findings of our study are two-fold: a) in patients with LV systolic dysfunction undergoing TAVI, acute hemodynamic effects (mainly afterload reduction) are related to immediate increase in LVEF. Normalization of LVEF occurred at 6 months in 58% of cases with baseline LV systolic dysfunction. Even though acute increase in LVEF is associated with further LVEF improvement > 5% in 62.5% of cases, a significant LVEF improvement at mid-term follow-up may occur also in patients without acute LVEF increase. b) Clinical status and the coexistence of CAD do not influence LVEF changes and/or preclude LV systolic function improvement. Concentric hypertrophy in AS generally maintains LV chamber size and LVEF within normal limits, but in a late stage of the disease LV function can deteriorate as preload reserve is lost [7]. Even when LVEF is markedly reduced (< 25%), it can improve to normal after AVR, suggesting that afterload mismatch rather than irreversibly depressed myocardial contractility was responsible for LV failure. Patients with severe LV dysfunction showed significant improvement in LVEF after TAVI [4] and [5]. However, the rate and extent of postoperative changes in LVEF are largely variable and have not yet been entirely defined [8]. LV function may display substantial changes in the immediate post-operative period due to acute decrease of afterload and also intermediate and long-term changes related to several chronic adaptations including regression of LV hypertrophy and increase in myocardial blood flow. This latter mechanism has been also described in the immediate post-operative phases of AVR [9] and TAVI [10], as related to decrease in hemodynamic load, increase in diastolic perfusion, in systolic component of coronary flow and in subendocardium perfusion. In our study, 62.5% of patients with LVEF < 50% had an immediate intraoperative improvement greater than 5%. Afterload was one of the major determinants of LV dysfunction even in patients with associated CAD. In fact, patients with higher baseline gradients and LV dysfunction had immediate increase in LVEF regardless the presence of CAD. Interestingly, the trend was toward a normalization of LVEF, 28 out of 48 patients reached an LVEF ≥ 50% at 6 months, and 19 out of 30 patients with immediate increase in LVEF > 5% had a further increase of 5% in the follow-up. Conversely, 18 out of 48 patients with baseline LVEF < 50% had no acute LVEF improvement. However, 12 of them had an increase of LVEF at 6 months. Therefore, even though acute reduction of afterload due to the abolishment of aortic gradient and optimal prosthetic valve performance accounts for the immediate increase in LV systolic function, the absence of immediate increase of LVEF does not preclude late recovery of systolic function. We may postulate that the immediate increase of LVEF is mainly determined by acute unloading of the ventricle while several other factors may contribute to late normalization of LVEF. Indeed, patients with an acute improvement in LVEF had baseline transaortic gradients significantly higher than those without improvement in LVEF thus reinforcing the concept that the immediate increase in LV function is related to acute unloading. In addition to the important role of echocardiography in the evaluation of TAVI patients at follow-up, immediate periprocedural echocardiographic findings may give important clinical information and LVEF acute improvement indicates a high probability of normalization or trend to normalization of LVEF. References [1]S.K. Kodali, M.R. Williams, C.R. Smith et al. Two-year outcomes after transcatheter or surgical aortic-valve replacement N Engl J Med, 366 (2012), pp. 1686–1695 View Record in Scopus | Cited By in Scopus (363) [2]M.E. Halkos, E.P. Chen, E.L. Sarin et al. Aortic valve replacement for aortic stenosis in patients with left ventricular dysfunction Ann Thorac Surg, 88 (2009), pp. 746–751 Article| PDF (314 K)| View Record in Scopus | Cited By in Scopus (17) [3]J.B. Goldberg, J.P. Desimone, R.S. Kramer et al. Impact of preoperative left ventricular ejection fraction on long-term survival after aortic valve replacement for aortic stenosis Circ Cardiovasc Qual Outcomes, 6 (2013), pp. 35–41 View Record in Scopus| Full Text via CrossRef | Cited By in Scopus (2) [4]S.H. Ewe, N. Ajmone Marsan, M. Pepi et al. Impact of left ventricular systolic function on clinical and echocardiographic outcomes following transcatheter aortic valve implantation for severe aortic stenosis Am Heart J, 160 (2010), pp. 1113–1120 Article| PDF (359 K)| View Record in Scopus | Cited By in Scopus (37) [5]C. Fraccaro, R. Al-Lamee, G. Tarantini et al. Transcatheter aortic valve implantation in patients with severe left ventricular dysfunction: immediate and mid-term results, a multicenter study Circ Cardiovasc Interv, 5 (2012), pp. 253–260 View Record in Scopus| Full Text via CrossRef | Cited By in Scopus (15) [6]A. Vahanian, O. Alfieri, F. Andreotti et al. Guidelines on the management of valvular heart disease (version 2012) Eur Heart J, 33 (2012), pp. 2451–2496 View Record in Scopus | Cited By in Scopus (376) [7]J. Ross Jr. Afterload mismatch in aortic and mitral valve disease: implications for surgical therapy J Am Coll Cardiol, 5 (1985), pp. 811–826 Article| PDF (1835 K)| View Record in Scopus | Cited By in Scopus (108) [8]E.S. Monrad, O.M. Hess, T. Murakami et al. Time course of regression of left ventricular hypertrophy after aortic valve replacement Circulation, 77 (1988), pp. 1345–1355 View Record in Scopus| Full Text via CrossRef | Cited By in Scopus (131) [9]G. Tamborini, P. Barbier, E. Doria et al. Influences of aortic pressure gradient and ventricular septal thickness with systolic coronary flow in aortic valve stenosis Am J Cardiol, 78 (1996), pp. 1303–1306 Article| PDF (403 K)| View Record in Scopus | Cited By in Scopus (10) [10]I. Ben-Dor, S.A. Goldstein, R. Waksman et al. Effects of percutaneous aortic valve replacement on coronary blood flow assessed with transesophageal Doppler echocardiography in patients with severe aortic stenosis Am J Cardiol, 104 (2009), pp. 850–855 Article| PDF (688 K)| View Record in Scopus | Cited By in Scopus (9) Corresponding author at: Centro Cardiologico Monzino IRCCS, Via Parea 4, 20138 Milan, Italy. Tel.: + 39 0258002011; fax: + 39 0258002287. 1This author takes responsibility for all aspects of the reliability and freedom from bias of the data presented and their discussed interpretation.