Author + information
- Received March 29, 2013
- Revision received April 30, 2013
- Accepted May 21, 2013
- Published online August 27, 2013.
- Florent Le Ven, MD∗,
- Mélanie Freeman, MD†,
- John Webb, MD†,
- Marie-Annick Clavel, DVM, PhD∗,
- Miriam Wheeler, MD†,
- Éric Dumont, MD∗,
- Chris Thompson, MD†,
- Robert De Larochellière, MD∗,
- Robert Moss, MD†,
- Daniel Doyle, MD∗,
- Henrique B. Ribeiro, MD∗,
- Marina Urena, MD∗,
- Luis Nombela-Franco, MD∗,
- Josep Rodés-Cabau, MD∗ and
- Philippe Pibarot, DVM, PhD∗∗ ()
- ∗Institut Universitaire de Cardiologie et de Pneumologie de Québec/Québec Heart and Lung Institute, Laval University, Québec City, Québec, Canada
- †St. Paul's Hospital, University of British Columbia, Vancouver, British Columbia, Canada
- ↵∗Reprint requests and correspondence:
Dr. Philippe Pibarot, Institut Universitaire de Cardiologie et de Pneumologie de Québec, 2725 Chemin Sainte-Foy, Québec, Québec G1V-4G5, Canada.
Objectives This study sought to assess the impact of baseline left ventricular (LV) outflow, LV ejection fraction (LVEF), and transvalvular gradient on outcomes following transcatheter aortic valve replacement (TAVR) in patients with severe aortic stenosis (AS).
Background Low flow (i.e., reduced stroke volume index [SVi]) can occur with both reduced and preserved LVEF. Low flow is often associated with low gradient despite severe stenosis and with worse outcomes following surgical aortic valve replacement. However, there are few data about the impact of low flow on outcomes following TAVR.
Methods We retrospectively analyzed the clinical, Doppler-echocardiographic, and outcome data prospectively collected in 639 patients who underwent TAVR for symptomatic severe AS in 2 Canadian centers.
Results In this cohort, 334 (52.3%) patients had a low flow (SVi <35 ml/m2) and these patients had increased 30-day mortality (11.4 vs. 5.9%, p = 0.01), 2-year all-cause mortality (35.3 vs. 30.9%, p = 0.005), and 2-year cardiovascular mortality (25.7 vs. 16.8%, p = 0.01) compared with patients with normal flow. Reduced flow was an independent predictor of 30-day mortality (odds ratio: 1.94, p = 0.026), cumulative all-cause mortality (hazard ratio: 1.27 per 10 ml/m² SVi decrease, p = 0.016), and cumulative cardiovascular mortality (hazard ratio: 1.29 per 10 ml/m² decrease, p = 0.04). Despite significant association in univariable analyses, low LVEF and low mean gradient were not found to be independent predictors of outcomes in multivariable analyses.
Conclusions Low flow but not low LVEF or low gradient is an independent predictor of early and late mortality following TAVR in high-risk patients with severe AS. SVi should be integrated in the risk stratification process of these patients.
A low-flow state (i.e., reduced left ventricular [LV] stroke volume index [SVi]) can occur with both reduced (i.e., classical low flow) or preserved (i.e., paradoxical low flow [LF]) LV ejection fraction (LVEF) (1). Patients with classical (low LVEF) LF, low gradient (LG) generally have a poor prognosis with medical treatment and a high operative mortality when undergoing surgical aortic valve replacement (1). Recent studies also suggest that patients with paradoxical (normal LVEF), LF, LG have better prognosis with surgery compared with medical therapy despite higher operative mortality compared to patients with normal flow (NF) and high gradient (HG) (1–3). Transcatheter aortic valve replacement (TAVR) may provide an alternative to surgical aortic valve replacement in these high-risk patients with classical or paradoxical LF aortic stenosis (AS). However, the impact of LF on outcomes following TAVR is unknown. Furthermore, recent studies reported that the presence of a low transvalvular gradient prior to procedure is associated with increased mortality following TAVR, whereas the analyses with respect to impact of LVEF on post-procedural outcomes yielded to conflicting results (4–7).
The objective of this study was thus to examine the respective impact of pre-procedural flow (i.e., SVi), LVEF, and transvalvular gradient on mortality following TAVR.
We retrospectively analyzed the clinical and Doppler-echocardiographic data prospectively collected in 775 consecutive patients who underwent TAVR with a balloon-expandable valve for symptomatic severe AS between January 2005 and April 2012 at the Québec Heart and Lung Institute and St. Paul's Hospital. We excluded patients who had: 1) “valve-in-valve” procedure; 2) TAVR for other indication than severe AS; and 3) incomplete pre-procedural Doppler-echocardiographic data. Finally, 639 patients (82.5%) were included in the study. The study protocol was performed in accordance with the institutional ethics committees of each center, and all patients gave informed written consent for the procedures and the research study.
Parameters of LV and aortic valve function were measured by Doppler echocardiography prior to the procedure, as previously described (8). Stroke volume was measured by pulsed wave Doppler in the LV outflow tract and was indexed for body surface area (SVi). The patients were first separated into 4 groups according to their SVi and mean gradient: 1) NF (SVi >35 ml/m²) and HG (≥40 mm Hg) (NF-HG); 2) NF and LG (<40 mm Hg) (NF-LG); 3) LF (SVi <35 ml/m²) and HG (LF-HG); and 4) LF and LG (LF-LG). The LF-LG group was then further separated into 2 groups according to LVEF, that is, classical LF-LG with low LVEF (<50%) (LF-LG-low ejection fraction [LEF]) and paradoxical LF-LG with preserved LVEF (≥50%) (LF-LG-normal EF [NEF]).
The primary endpoint of this study was all-cause mortality and the secondary endpoints were cardiovascular mortality and 30-day mortality defined according to the Valve Academic Research Consortium-2 recommendations (9).
Continuous variables were tested for distribution normality with the Shapiro-Wilk test and expressed as mean ± SD or median and interquartile range. Because EuroSCORE and Society of Thoracic Surgeons score were not normally distributed, a natural log transformation was used for these variables. Differences between groups were assessed using analysis of variance for continuous variables with subsequent Tukey HSD pairwise comparisons, and the chi-square test or Fisher exact test for categorical variables as appropriate. Survival curves were presented as Kaplan-Meier curves, and the log-rank test was used for comparison between groups. The effect of the clinical and Doppler echocardiographic variables on survival was assessed with Cox proportional hazards regression models for cumulative all-cause and cardiovascular mortality and with backward stepwise logistic regression models for 30-day mortality. Clinically relevant variables with a p value ≤0.1 on individual analysis were included in the multivariable models. Age was forced into the models. A p value ≤0.05 was considered statistically significant.
Table 1 shows the comparison of the patients' baseline characteristics between the 5 study groups.
Impact of flow, LVEF, and gradient on 30-day mortality
Patients with LF had higher 30-day mortality compared with those in the NF group (38 of 334 [11%] vs. 18 of 305 [6%], p = 0.01) (Fig. 1). Among patients with LF-LG, 30-day mortality was not statistically different (p = 0.5) between those with NEF (9%) and those with LEF (12%) (Fig. 1). The independent predictors of 30-day mortality were male gender (odds ratio [OR]: 1.90 [95% confidence interval (CI): 1.07 to 3.46], p = 0.03), diabetes (OR: 2.2 [95% CI: 1.11 to 4.73], p = 0.032), estimated glomerular filtration rate (OR: 1.14 [95% CI: 1.01 to 1.30] per 10 ml/min decrease, p = 0.037), pulmonary hypertension (OR: 2.26 [95% CI: 1.25 to 4.04], p = 0.008), and SVi <35 ml/m² (OR: 1.94 [95% CI: 1.08 to 3.59], p = 0.026).
Impact of flow, LVEF, and gradient on cumulative mortality
Overall, there were 207 deaths during a median follow-up of 12 (interquartile range: 1.5 to 24) months and 125 were of cardiovascular cause. Patients with LF had increased all-cause mortality (hazard ratio [HR]: 1.48 [95% CI: 1.13 to 1.97], p = 0.005) and cardiovascular mortality (HR: 1.57 [95% CI: 1.10 to 2.27], p = 0.01) compared with those with NF (Fig. 2). Mortality was also increased in patients with LEF compared with those with NEF (all-cause mortality HR: 1.61 [95% CI: 1.19 to 2.14], p = 0.002; cardiovascular mortality HR: 1.73 [95% CI: 1.18 to 2.49], p = 0.005) and in patients with LG compared with those with HG (all-cause mortality HR: 1.45 [95% CI: 1.10 to 1.91], p = 0.008; cardiovascular mortality HR: 1.55 [95% CI: 1.09 to 2.22], p = 0.016).
Patients in NF-HG, LF-HG, and NF-LG groups had similar outcomes (all-cause mortality p = 0.18; cardiovascular mortality p = 0.21), whereas all-cause mortality was significantly higher in LF-LG group compared with NF-HG (HR: 1.96 [95% CI: 1.36 to 2.83], p = 0.0003), NF-LG (HR: 1.61 [95% CI: 1.06 to 2.48], p = 0.02), and LF-HG (HR: 1.52 [95% CI: 1.06 to 2.19], p = 0.02) groups (Fig. 3). Cardiovascular mortality was also increased in the LF-LG group compared with the NF-HG group (HR: 2.18 [95% CI: 1.36 to 3.53], p = 0.001). After further dichotomization of the LF-LG group according to LVEF, no significant difference was observed between LF-LG-NEF and LF-LG-LEF subgroups with respect to all-cause (p = 0.21) or cardiovascular (p = 0.33) mortality (Fig. 4).
The predictors of all-cause and cardiovascular mortality are shown in Tables 2 and 3, respectively. In multivariable analysis, SVi was an independent predictor of both all-cause mortality (HR: 1.27 [95% CI: 1.04 to 1.55] per 10 ml/m² decrease, p = 0.016) and cardiovascular mortality (HR: 1.29 [95% CI: 1.01 to 1.66] per 10 ml/m² decrease, p = 0.04). Although they were significant predictors in univariable analysis, LVEF and mean gradient were not independent predictors in multivariable analysis (Tables 2 and 3).
The main findings of this study are: 1) about half of the patients undergoing TAVR are in LF state; 2) LF but not low LVEF or LG is an independent predictor of cumulative all-cause and cardiovascular mortality as well as 30-day mortality following TAVR; 3) patients with LF-LG had worse outcomes following TAVR compared with those with NF-HG; and 4) patients with paradoxical LF-LG-NEF had similar outcomes compared with those with classical LF-LG-LEF.
Numerous studies have reported that low LVEF is a powerful independent risk factor for increased early and late mortality following surgical aortic valve replacement, and this factor is included in the operative risk scores (10,11). In the context of TAVR, some studies reported an association between low baseline LVEF and increased mortality (4,7) but others did not (5,6). In the PARTNER-I (Placement of AoRTic TraNscathetER Valves) trial (5), reduced LVEF was not a predictor of mortality in both the surgical aortic valve replacement and TAVR arms. However, patients with a LVEF <20% were excluded from this trial. In the present study, low LVEF was associated with increased mortality in univariable analysis but after adjustment for other risk factors including SVi, this association was no longer significant. On the other hand, reduced SVi was independently associated with increased early and late mortality. LVEF may underestimate the extent of myocardial systolic dysfunction in presence of concentric remodeling, such as generally observed in patients with severe AS, whereas SVi is a direct measure of the efficiency of the cardiac pump function and its ability to meet tissue perfusion and metabolic demand.
Previous studies (2,3) as well as the recent ESC-EACTS (European Society of Cardiology-European Association for Cardio-Thoracic Surgery) guidelines (12) proposed a cut-point value of 35 ml/m2 to define LF. The present study corroborates this cut-point in the context of TAVR but also suggests that the mortality increases continuously with further reduction in SVi.
Very low pre-operative mean gradient (<20 mm Hg) has been associated with increased mortality following surgical aortic valve replacement in patients with classical LF-LG AS (11). Several studies also reported an association between low pre-procedural mean gradient (<40 mm Hg) and increased mortality following TAVR (5,6). However, in the present study, the association between mean gradient and mortality was no longer significant after adjustment for SVi. These findings suggest that the inverse relationship between baseline gradient and mortality following TAVR is probably, in large part, due to the presence of LF.
Data were prospectively collected in consecutive series of patients from 2 large volume centers and there were no exclusion criteria with regard to low LVEF or LG. However, the data were retrospectively queried and there was no Core Laboratory for the analysis of echocardiograms. The multivariable model for 30-day mortality may be overfitted. Doppler-echocardiographic estimation of SVi may be subject to measurement errors, particularly in the presence of poor image quality, elliptic shape of the LV outflow tract, or atrial fibrillation (13).
LF but not low LVEF or LG is an independent predictor of early and late mortality in high-risk patients with severe AS undergoing TAVR. Assessment of SVi should be included in the risk stratification of these patients. Future randomized studies are needed to assess the impact of therapy in patients with LF AS.
The authors would like to thank Isabelle Fortin, Jacinthe Aubé, and Mélanie Côté for their assistance in the collection of the data.
This study was funded by research grants (MOP-57745 and MOP-126072) from the Canadian Institutes of Health Research (Ottawa, Ontario, Canada). Dr. Le Ven is supported by a clinical and research fellowship from the “Fédération Française de Cardiologie.” Drs. Webb, Dumont, and Rodés-Cabau are consultants for Edwards Lifesciences. Drs. Rodés-Cabau and De Larochellière are consultants for St. Jude Medical. Dr. Clavel is supported by a postdoctoral fellowship from the Canadian Institutes of Health Research (CIHR). Dr. Thompson has received honoraria from Edwards Lifesciences and Abbott Vascular. Dr. Moss has received a teaching stipend from Edwards Lifesciences. Dr. Pibarot holds the Canada Research Chair in Valvular Heart Disease supported by CIHR and has received a research grant from Edwards Lifesciences. All other authors have reported that they have no relationships relevant to the contents of this paper to disclose.
- Abbreviations and Acronyms
- aortic stenosis
- confidence interval
- high gradient
- hazard ratio
- low ejection fraction
- low flow
- low gradient
- left ventricle/ventricular
- left ventricular ejection fraction
- mean gradient
- normal ejection fraction
- normal flow
- odds ratio
- stroke volume index
- transcatheter aortic valve replacement
- Received March 29, 2013.
- Revision received April 30, 2013.
- Accepted May 21, 2013.
- American College of Cardiology Foundation
- Pibarot P.,
- Dumesnil J.G.
- Hachicha Z.,
- Dumesnil J.G.,
- Bogaty P.,
- Pibarot P.
- Clavel M.A.,
- Dumesnil J.G.,
- Capoulade R.,
- Mathieu P.,
- Sénéchal M.,
- Pibarot P.
- Fraccaro C.,
- Al-Lamee R.,
- Tarantini G.,
- et al.
- Clavel M.A.,
- Webb J.G.,
- Rodés-Cabau J.,
- et al.
- Kappetein A.P.,
- Head S.J.,
- Genereux P.,
- et al.
- Connolly H.M.,
- Oh J.K.,
- Orszulak T.A.,
- et al.
- Levy F.,
- Laurent M.,
- Monin J.L.,
- et al.
- Vahanian A.,
- Alfieri O.,
- Andreotti F.,
- et al.