Impact of low stroke volume on mortality in patients with severe aortic stenosis and preserved left ventricular ejection fraction

Impact of low stroke volume on mortality in patients with severe aortic stenosis and preserved... Abstract Aims In patients with severe aortic stenosis (AS) and preserved left ventricular ejection fraction (LVEF), low flow (LF) is currently defined using Doppler-echocardiography by a stroke volume index (SVi)<35 mL/m2. However, the relationship between LF and outcome remains unclear as data on normal reference values defining LF are scarce, and previous studies did not explore the risk associated with other SVi cut-points. We analysed the relationship between LF and mortality in severe AS to establish prognostic LF values associated with mortality risk. Methods and results This study included 1450 consecutive patients with severe AS (aortic valve area <1 cm2 and/or <0.6 cm2/m2) and preserved LVEF and 1645 controls with normal echocardiograms. Patients were stratified in three groups: (i) SVi > 35 mL/m2 or SV > 70 mL; (ii) SVi 30–35 mL/m2 or SV 55–70 mL; and (iii) SVi < 30 mL/m2 or SV < 55 mL. Mortality with medical and surgical management was analysed. Five-year survival was low for SVi < 30 mL/m2 and SV < 55 mL compared to the other groups (all P-values <0.001). After adjustment for outcome predictors, including aortic valve replacement, mortality risk was considerable with SVi < 30 mL/m2 vs. >35 mL/m2 [adjusted hazard ratio (HR) 1.60 (1.17–2.18)] and SV < 55 mL vs. >70 mL [adjusted HR 1.84 (1.32–2.58)]. Similar mortality risk was observed for SVi 30–35 mL/m2 vs. >35 mL/m2 [adjusted HR 1.05 (0.78–1.41)], and for SV 55–70 mL vs. >70 mL [adjusted HR 1.22 (0.94–1.58)]. The prognostic impact of SVi < 30 mL/m2 and SV < 55 mL was consistent in subgroups, including asymptomatic patients and patients with low-gradient severe AS. Conclusion Low flow defined as SVi < 30 mL/m2 or SV < 55 mL is an important outcome predictor in severe AS with preserved LVEF under medical and surgical management. Further studies are needed to prospectively test these values for risk stratification and decision making. Aortic stenosis, Stroke volume, Stroke volume index, Low flow, Doppler-echocardiography, Outcome Introduction Current guidelines define severe aortic stenosis (AS) as an aortic valve area (AVA) < 1 cm2 (>0.6 cm/m2) and a peak aortic jet velocity (Vmax) ≥4 m/s, or a mean pressure gradient ≥40 mmHg in patients with preserved (≥50%) left ventricular ejection fraction (LVEF).1,2 In patients with AS and preserved LVEF, low flow (LF) is currently defined1,2 by a stroke volume index (SVi)  < 35 mL/m2 based on several studies that have arbitrarily used this cut-off value.3–5 Previous studies have suggested that LF, defined as SVi < 35 mL/m2 is associated with poor prognosis in asymptomatic and symptomatic patients with low-gradient (LG) AS and preserved LVEF.3,5–8 The Doppler-derived measurement of stroke volume (SV) at the aortic annulus has shown good correlation with invasive calculations9–11 and is widely used in clinical practice and recommended by guidelines.12 However, published normal reference values in healthy individuals are scarce.13–15 The relationship between SV and mortality has not been described across the whole spectrum of patients with severe AS and, therefore, the SVi value delineating a subgroup of LF severe AS at high risk of death remains unclear. Moreover, non-indexed SV values associated with high mortality risk have never been reported. This study analyses the relationship between SV and SVi measured at the time of AS diagnosis and all-cause mortality during follow-up. We enrolled in two tertiary centres (Amiens, and Lille, France) patients with severe AS and preserved LVEF and aimed to evaluate the predictive value of SV and SVi on outcome with medical and surgical management and establish prognostic LF values associated with mortality risk. Methods Patient population Between 2000 and 2015, patients ≥18 years of age diagnosed with ≥mild AS (aortic leaflet calcification with reduction in systolic movements and Vmax > 2.5 m/s) were prospectively identified and included in an electronic database. We excluded: (i) >mild aortic and/or mitral regurgitation; (ii) prosthetic valves, congenital heart disease, supravalvular or subvalvular AS, or dynamic LV outflow tract obstruction; (iii) mitral stenosis; and (iv) patients who refused to participate in the study. This analysis included 1450 patients with severe AS [defined as AVA < 1 cm2 and/or AVA normalized to body surface area (BSA) <0.6 cm2] and preserved LVEF. Eighty-one patients were excluded because of missing data. Clinical and demographic baseline characteristics were collected.16 An index summating the patient’s individual comorbidities was calculated.16 Controls Using the echocardiography databases, we retrospectively identified between 2014 and 2016, 1645 consecutive individuals ≥18 years of age with normal echocardiograms. These individuals had normal blood pressure, and no personal history of cardiovascular disease. All echocardiograms were validated as normal by physicians experienced in transthoracic echocardiography. We obtained institutional review board authorizations prior to conducting the study. The study was conducted in accordance with institutional policies, national legislation, and the revised Helsinki declaration. Patients gave informed written consent prior to participation in the study. Echocardiography All patients underwent a comprehensive Doppler-echocardiography study, using commercially available ultrasound systems. Aortic flow was recorded using continuous-wave Doppler, systematically in several acoustic windows (apical 5-chamber, right parasternal, suprasternal, epigastric).16 Stroke volume was calculated by multiplying the LV outflow tract area with the LV outflow tract time-velocity integral.3,17 The LV outflow tract diameter was measured in zoomed parasternal long-axis views in early systole at the level of aortic cusp insertion.18 The LV outflow tract time-velocity integral was recorded from the apical 5-chamber view, with the sample volume positioned about 5 mm proximal to the aortic valve.17 Aortic valve area and SV were indexed to BSA. When patients were in sinus rhythm, three cardiac cycles were averaged for all measures. For patients in atrial fibrillation, five cardiac cycles were averaged. Treatment decision and follow-up The majority of patients were followed in the outpatient clinics of the two tertiary centres. The others were followed in public hospitals or private practices by referring cardiologists working together with the tertiary centres. Information on follow-up was obtained by direct patient interview or by repeated follow-up letters and questionnaires to physicians, patients and (if necessary) next of kin. Ninety-three per cent of patients were followed up to 2 years or death. Follow-up was complete up to death or to the end of the study in 1295 patients (89%). The endpoint was overall survival after diagnosis with medical and surgical treatment. Clinical decisions regarding medical management and indications for surgery (presence of symptoms, LVEF impairment, or abnormal exercise test for asymptomatic patients) were made by the heart team with the approval of the patient’s referring cardiologist in accordance with guideline recommendations.16 Statistical analysis Flow across the aortic valve was analysed as variable normalized to BSA (SVi) as well as non-indexed variable (SV). For each of the two variables (SVi and SV), patients were classified in three flow groups: (i) SVi > 35 mL/m2 or SV > 70 mL; (ii) SVi 30–35 mL/m2 or SV 55–70 mL; and (iii) SVi < 30 mL/m2 or SV < 55 mL. Continuous variables were expressed as median (25th and 75th percentiles), and categorical variables were summarized as frequency percentages and counts. Baseline continuous variables were compared across flow groups using the Kruskal–Wallis tests and categorical variables were compared by the Pearson’s χ2 statistics or Fisher’s exact tests. The significance between the highest group and the others was examined if there was a significant difference across groups. Individual differences were compared with Mann–Whitney U tests (with Bonferroni correction for multiple comparisons). Estimated survival rates and 95% confidence intervals (95% CIs) were estimated according to the Kaplan–Meier method and compared with two-sided log-rank tests. Univariate and multivariable analyses of all-cause mortality were performed using Cox proportional hazards models. We did not use model-building techniques and entered in the models covariates of potential prognostic impact on an epidemiological basis. These covariates were: age, sex, body mass index, Charlson comorbidity index (not including age), symptoms (New York Heart Association Class II–IV dyspnoea, angina, or syncope), history of hypertension, coronary artery disease, atrial fibrillation, systolic blood pressure at baseline, Vmax, AVA, LVEF, and indexed LV mass. The effect of aortic valve replacement (AVR) on outcome was analysed as a time-dependent covariate using the entire follow-up. Age, body mass index, comorbidity index, systolic blood pressure, Vmax, AVA, LVEF, and indexed LV mass were used as continuous variables. The proportional hazards assumption was confirmed using statistics and graphs based on the Schoenfeld residuals. To show the additive value of the flow measurements, we computed the likelihood ratios of the following models: Model 1 including clinical factors, Model 2 including clinical factors, Vmax, AVA, LVEF, and indexed LV mass, and Models 3 and 4 including clinical factors, Vmax, AVA, LVEF, indexed LV mass, and flow quantification (SVi or SV). We compared the models using the global χ2 statistic, the Akaike Information Criterion and the Harrell’s C concordance statistic. Additionally, we evaluated the goodness-of-fit of the models using the Gronnesby–Brogan test. We conducted subgroup analyses to determine the homogeneity of the association of SVi and SV and mortality. First, we estimated the effect of SVi and SV on mortality in each subgroup using a Cox univariate model and then formally tested for first-order interactions entering interaction terms, separately for each subgroup. A significance level of 0.05 was assumed for all tests. All P-values are results of two-tailed tests. Data were analysed with SPSS (v 18.0; IBM Corp, Armonk, NY, USA) and STATA (version 12, StataCorp LP, College Station, TX, USA). Estimates of sensitivity and specificity for various SVi/SV cut-points were computed from time-dependent receiver operating curves (ROC) using the ‘timeROC’ package in R (R project for Statistical Computing, version 3.3.3, https://www.r-project.org/, 25 November 2017). Results Stroke volume by Doppler echocardiography in normal individuals In the cohort of 1645 consecutive individuals [age: 65 (54–77) years, 53% males] with normal echocardiograms, median BSA, LV outflow tract diameter, and LV outflow tract time-velocity integral were 1.88 (1.72–2.02) m2, 22 (20–24) mm, and, respectively, 18.5 (16.2–21.1) cm. Median LVEF was 62.8 (55.7–71.5) %. Median SVi was 37.3 (32.0–43.8) mL/m2 and median SV 70.0 (58.4–83.3) mL. Baseline characteristics of patients with aortic stenosis according to stroke volume Patients with AS were stratified in three flow groups: (i) SVi > 35 mL/m2 or SV > 70 mL; (ii) SVi 30–35 mL/m2 or SV 55–70 mL; and (iii) SVi < 30 mL/m2 or SV < 55 mL. The baseline characteristics of the 1450 patients, according to SVi are presented in Table 1. Patients with SVi < 30 mL/m2 were more often diabetic, with greater BSA and body mass index, and lower systolic blood pressure compared to those with SVi > 35 mL/m2. Almost 40% of patients with SVi < 30 mL/m2 were in atrial fibrillation compared to 27% of patients with SVi > 35 mL/m2 (Table 1). Low SV groups tended to have lower blood pressure and less frequent history of hypertension than higher SV groups due to the inclusion of patients with LF high-gradient severe AS. Low SV groups include both patients with low (<40 mmHg) and high gradient. In the subgroup of patients with LG/LF severe AS, the frequency of history of hypertension was higher, of 73%, as expected. As regard to echo-Doppler parameters, patients with SVi < 30 mL/m2 had significantly lower Vmax and smaller AVA compared to those with SVi > 35 mL/m2. Left ventricular ejection fraction and LV mass were greater in patients with SVi > 35 mL/m2 vs. SVi < 30 mL/m2 (Table 1). The baseline characteristics of the study patients stratified according to non-indexed SV are presented in the Supplementary material online, Table S1. Table 1 Baseline demographic, clinical, and echo-Doppler characteristics of patients according to stroke volume index groups Variable SVi < 30 mL/m2 SVi 30 to 35 mL/m2 SVi > 35 mL/m2 P-value (n = 190) (n = 221) (n = 1039) Demographic and clinical characteristics  Age (years) 78.5 (71.0–84.0) 78.2 (72.1–83.0) 77.5 (70.0–82.6) 0.07  Male sex (%, n) 43.7 (83) 48 (106) 49.1 (510) 0.39  Body surface area (m2) 1.9 (1.7–2.0)a 1.89 (1.72–2.0)a 1.82 (1.70–1.99) 0.001  Body mass index (kg/m2) 27.3 (24.2–32.5)a 27.7 (24.5–31.3)b 26.6 (23.7–29.7) <0.001  Systolic blood pressure (mmHg) 130.0 (120.0–145.0)a 140.0 (120.0–150.0) 140.0 (124.0–150.0) 0.004  New York Heart Association class (%, n) 0.53   I–II 73.2 (139) 74.7 (165) 76.6 (796)   III–IV 26.8 (51) 25.3 (56) 23.4 (243)  Angina (%, n) 20.0 (38) 18.1 (40) 23.8 (247) 0.13  Syncope (%, n) 10.0 (19) 13.6 (30) 11.5 (119) 0.51  History of hypertension (%, n) 68.4 (130) 72.4 (160) 74.5 (774) 0.21  Diabetes mellitus (%, n) 40.5 (77)b 33.5 (74) 26.9 (280) <0.001  Coronary artery disease (%, n) 46.8 (89) 52.9 (117) 52.2 (542) 0.36  History of atrial fibrillation (%, n) 39.5 (75)b 33 (73)a 27.4 (285) 0.002  Charlson comorbidity index 2.0 (1.0–4.0) 2.0 (1.0–3.0) 2.0 (1.0–3.0) 0.38 Echocardiography and Doppler parameters  Aortic valve area (cm2) 0.65 (0.50–0.76)b 0.65 (0.53–0.80)b 0.76 (0.64–0.87) <0.001  Indexed aortic valve area (cm2/m2) 0.33 (0.28–0.39)b 0.35 (0.29–0.42)b 0.41 (0.35–0.47) <0.001  Peak aortic jet velocity (m/s) 4.1 (3.3–4.5)b 4.2 (3.7–4.7)b 4.4 (4.0–5.0) <0.001  Transaortic mean pressure gradient (mmHg) 42.0 (29.8–50.2)b 45.0 (33.0–56.0)b 50.0 (41.0–62.0) <0.001  LV outflow tract diameter (mm) 20.0 (19.0–22.0)b 21.0 (20.0–22.0)b 22.0 (20.4–23.0) <0.001  LV outflow tract velocity time integral (cm) 15.0 (13.0–18.0)b 18.0 (16.0–20.0)b 22.0 (20.0–25.0) <0.001  Stroke volume (mL) 50.1 (43.1–56.4)b 60.8 (56.7–66.0)b 81.4 (72.2–91.3) <0.001  Stroke volume index (mL/m2) 27.3 (24.5–29.2)b 33.0 (31.6–34.0)b 43.9 (39.7–49.5) <0.001  LV end-diastolic diameter (mm) 47.1 (42.0–54.0) 48.0 (43.0–53.0) 49.0 (44.0–53.0) 0.084  LV end-systolic diameter (mm) 30.5 (25.0–35.0) 31.0 (27.0–35.0)a 30.0 (26.0–34.0) 0.035  Ejection fraction (%) 60.0 (54.0–66.0)b 62.0 (56.0–66.0)b 65.0 (60.0–70.0) <0.001  Indexed LV mass (g/m2) 112.2 (92.5–144.1)b 127.0 (94.0–152.0)a 128.1 (106.3–156.0) <0.001 Variable SVi < 30 mL/m2 SVi 30 to 35 mL/m2 SVi > 35 mL/m2 P-value (n = 190) (n = 221) (n = 1039) Demographic and clinical characteristics  Age (years) 78.5 (71.0–84.0) 78.2 (72.1–83.0) 77.5 (70.0–82.6) 0.07  Male sex (%, n) 43.7 (83) 48 (106) 49.1 (510) 0.39  Body surface area (m2) 1.9 (1.7–2.0)a 1.89 (1.72–2.0)a 1.82 (1.70–1.99) 0.001  Body mass index (kg/m2) 27.3 (24.2–32.5)a 27.7 (24.5–31.3)b 26.6 (23.7–29.7) <0.001  Systolic blood pressure (mmHg) 130.0 (120.0–145.0)a 140.0 (120.0–150.0) 140.0 (124.0–150.0) 0.004  New York Heart Association class (%, n) 0.53   I–II 73.2 (139) 74.7 (165) 76.6 (796)   III–IV 26.8 (51) 25.3 (56) 23.4 (243)  Angina (%, n) 20.0 (38) 18.1 (40) 23.8 (247) 0.13  Syncope (%, n) 10.0 (19) 13.6 (30) 11.5 (119) 0.51  History of hypertension (%, n) 68.4 (130) 72.4 (160) 74.5 (774) 0.21  Diabetes mellitus (%, n) 40.5 (77)b 33.5 (74) 26.9 (280) <0.001  Coronary artery disease (%, n) 46.8 (89) 52.9 (117) 52.2 (542) 0.36  History of atrial fibrillation (%, n) 39.5 (75)b 33 (73)a 27.4 (285) 0.002  Charlson comorbidity index 2.0 (1.0–4.0) 2.0 (1.0–3.0) 2.0 (1.0–3.0) 0.38 Echocardiography and Doppler parameters  Aortic valve area (cm2) 0.65 (0.50–0.76)b 0.65 (0.53–0.80)b 0.76 (0.64–0.87) <0.001  Indexed aortic valve area (cm2/m2) 0.33 (0.28–0.39)b 0.35 (0.29–0.42)b 0.41 (0.35–0.47) <0.001  Peak aortic jet velocity (m/s) 4.1 (3.3–4.5)b 4.2 (3.7–4.7)b 4.4 (4.0–5.0) <0.001  Transaortic mean pressure gradient (mmHg) 42.0 (29.8–50.2)b 45.0 (33.0–56.0)b 50.0 (41.0–62.0) <0.001  LV outflow tract diameter (mm) 20.0 (19.0–22.0)b 21.0 (20.0–22.0)b 22.0 (20.4–23.0) <0.001  LV outflow tract velocity time integral (cm) 15.0 (13.0–18.0)b 18.0 (16.0–20.0)b 22.0 (20.0–25.0) <0.001  Stroke volume (mL) 50.1 (43.1–56.4)b 60.8 (56.7–66.0)b 81.4 (72.2–91.3) <0.001  Stroke volume index (mL/m2) 27.3 (24.5–29.2)b 33.0 (31.6–34.0)b 43.9 (39.7–49.5) <0.001  LV end-diastolic diameter (mm) 47.1 (42.0–54.0) 48.0 (43.0–53.0) 49.0 (44.0–53.0) 0.084  LV end-systolic diameter (mm) 30.5 (25.0–35.0) 31.0 (27.0–35.0)a 30.0 (26.0–34.0) 0.035  Ejection fraction (%) 60.0 (54.0–66.0)b 62.0 (56.0–66.0)b 65.0 (60.0–70.0) <0.001  Indexed LV mass (g/m2) 112.2 (92.5–144.1)b 127.0 (94.0–152.0)a 128.1 (106.3–156.0) <0.001 LV, left ventricular; SVi, stroke volume index. Continuous variables are expressed as median (interquartile range) and categorical variables as percentages and counts. a P < 0.05 individual category vs. ‘SVi >35 mL/m2’. b P < 0.001 individual category vs. ‘SVi >35 mL/m2’. Table 1 Baseline demographic, clinical, and echo-Doppler characteristics of patients according to stroke volume index groups Variable SVi < 30 mL/m2 SVi 30 to 35 mL/m2 SVi > 35 mL/m2 P-value (n = 190) (n = 221) (n = 1039) Demographic and clinical characteristics  Age (years) 78.5 (71.0–84.0) 78.2 (72.1–83.0) 77.5 (70.0–82.6) 0.07  Male sex (%, n) 43.7 (83) 48 (106) 49.1 (510) 0.39  Body surface area (m2) 1.9 (1.7–2.0)a 1.89 (1.72–2.0)a 1.82 (1.70–1.99) 0.001  Body mass index (kg/m2) 27.3 (24.2–32.5)a 27.7 (24.5–31.3)b 26.6 (23.7–29.7) <0.001  Systolic blood pressure (mmHg) 130.0 (120.0–145.0)a 140.0 (120.0–150.0) 140.0 (124.0–150.0) 0.004  New York Heart Association class (%, n) 0.53   I–II 73.2 (139) 74.7 (165) 76.6 (796)   III–IV 26.8 (51) 25.3 (56) 23.4 (243)  Angina (%, n) 20.0 (38) 18.1 (40) 23.8 (247) 0.13  Syncope (%, n) 10.0 (19) 13.6 (30) 11.5 (119) 0.51  History of hypertension (%, n) 68.4 (130) 72.4 (160) 74.5 (774) 0.21  Diabetes mellitus (%, n) 40.5 (77)b 33.5 (74) 26.9 (280) <0.001  Coronary artery disease (%, n) 46.8 (89) 52.9 (117) 52.2 (542) 0.36  History of atrial fibrillation (%, n) 39.5 (75)b 33 (73)a 27.4 (285) 0.002  Charlson comorbidity index 2.0 (1.0–4.0) 2.0 (1.0–3.0) 2.0 (1.0–3.0) 0.38 Echocardiography and Doppler parameters  Aortic valve area (cm2) 0.65 (0.50–0.76)b 0.65 (0.53–0.80)b 0.76 (0.64–0.87) <0.001  Indexed aortic valve area (cm2/m2) 0.33 (0.28–0.39)b 0.35 (0.29–0.42)b 0.41 (0.35–0.47) <0.001  Peak aortic jet velocity (m/s) 4.1 (3.3–4.5)b 4.2 (3.7–4.7)b 4.4 (4.0–5.0) <0.001  Transaortic mean pressure gradient (mmHg) 42.0 (29.8–50.2)b 45.0 (33.0–56.0)b 50.0 (41.0–62.0) <0.001  LV outflow tract diameter (mm) 20.0 (19.0–22.0)b 21.0 (20.0–22.0)b 22.0 (20.4–23.0) <0.001  LV outflow tract velocity time integral (cm) 15.0 (13.0–18.0)b 18.0 (16.0–20.0)b 22.0 (20.0–25.0) <0.001  Stroke volume (mL) 50.1 (43.1–56.4)b 60.8 (56.7–66.0)b 81.4 (72.2–91.3) <0.001  Stroke volume index (mL/m2) 27.3 (24.5–29.2)b 33.0 (31.6–34.0)b 43.9 (39.7–49.5) <0.001  LV end-diastolic diameter (mm) 47.1 (42.0–54.0) 48.0 (43.0–53.0) 49.0 (44.0–53.0) 0.084  LV end-systolic diameter (mm) 30.5 (25.0–35.0) 31.0 (27.0–35.0)a 30.0 (26.0–34.0) 0.035  Ejection fraction (%) 60.0 (54.0–66.0)b 62.0 (56.0–66.0)b 65.0 (60.0–70.0) <0.001  Indexed LV mass (g/m2) 112.2 (92.5–144.1)b 127.0 (94.0–152.0)a 128.1 (106.3–156.0) <0.001 Variable SVi < 30 mL/m2 SVi 30 to 35 mL/m2 SVi > 35 mL/m2 P-value (n = 190) (n = 221) (n = 1039) Demographic and clinical characteristics  Age (years) 78.5 (71.0–84.0) 78.2 (72.1–83.0) 77.5 (70.0–82.6) 0.07  Male sex (%, n) 43.7 (83) 48 (106) 49.1 (510) 0.39  Body surface area (m2) 1.9 (1.7–2.0)a 1.89 (1.72–2.0)a 1.82 (1.70–1.99) 0.001  Body mass index (kg/m2) 27.3 (24.2–32.5)a 27.7 (24.5–31.3)b 26.6 (23.7–29.7) <0.001  Systolic blood pressure (mmHg) 130.0 (120.0–145.0)a 140.0 (120.0–150.0) 140.0 (124.0–150.0) 0.004  New York Heart Association class (%, n) 0.53   I–II 73.2 (139) 74.7 (165) 76.6 (796)   III–IV 26.8 (51) 25.3 (56) 23.4 (243)  Angina (%, n) 20.0 (38) 18.1 (40) 23.8 (247) 0.13  Syncope (%, n) 10.0 (19) 13.6 (30) 11.5 (119) 0.51  History of hypertension (%, n) 68.4 (130) 72.4 (160) 74.5 (774) 0.21  Diabetes mellitus (%, n) 40.5 (77)b 33.5 (74) 26.9 (280) <0.001  Coronary artery disease (%, n) 46.8 (89) 52.9 (117) 52.2 (542) 0.36  History of atrial fibrillation (%, n) 39.5 (75)b 33 (73)a 27.4 (285) 0.002  Charlson comorbidity index 2.0 (1.0–4.0) 2.0 (1.0–3.0) 2.0 (1.0–3.0) 0.38 Echocardiography and Doppler parameters  Aortic valve area (cm2) 0.65 (0.50–0.76)b 0.65 (0.53–0.80)b 0.76 (0.64–0.87) <0.001  Indexed aortic valve area (cm2/m2) 0.33 (0.28–0.39)b 0.35 (0.29–0.42)b 0.41 (0.35–0.47) <0.001  Peak aortic jet velocity (m/s) 4.1 (3.3–4.5)b 4.2 (3.7–4.7)b 4.4 (4.0–5.0) <0.001  Transaortic mean pressure gradient (mmHg) 42.0 (29.8–50.2)b 45.0 (33.0–56.0)b 50.0 (41.0–62.0) <0.001  LV outflow tract diameter (mm) 20.0 (19.0–22.0)b 21.0 (20.0–22.0)b 22.0 (20.4–23.0) <0.001  LV outflow tract velocity time integral (cm) 15.0 (13.0–18.0)b 18.0 (16.0–20.0)b 22.0 (20.0–25.0) <0.001  Stroke volume (mL) 50.1 (43.1–56.4)b 60.8 (56.7–66.0)b 81.4 (72.2–91.3) <0.001  Stroke volume index (mL/m2) 27.3 (24.5–29.2)b 33.0 (31.6–34.0)b 43.9 (39.7–49.5) <0.001  LV end-diastolic diameter (mm) 47.1 (42.0–54.0) 48.0 (43.0–53.0) 49.0 (44.0–53.0) 0.084  LV end-systolic diameter (mm) 30.5 (25.0–35.0) 31.0 (27.0–35.0)a 30.0 (26.0–34.0) 0.035  Ejection fraction (%) 60.0 (54.0–66.0)b 62.0 (56.0–66.0)b 65.0 (60.0–70.0) <0.001  Indexed LV mass (g/m2) 112.2 (92.5–144.1)b 127.0 (94.0–152.0)a 128.1 (106.3–156.0) <0.001 LV, left ventricular; SVi, stroke volume index. Continuous variables are expressed as median (interquartile range) and categorical variables as percentages and counts. a P < 0.05 individual category vs. ‘SVi >35 mL/m2’. b P < 0.001 individual category vs. ‘SVi >35 mL/m2’. Outcome impact of stroke volume Clinical management and follow-up Median (25th, 75th percentile) overall follow-up was 38 (17–69) months. The total number of deaths recorded during follow-up was 440. AVR was performed in 1072 patients (74%) by surgical (n = 970, 90.5%) or percutaneous (n = 102, 9.5%) techniques. Aortic bioprostheses were used in 879 patients (82%). One hundred and sixty patients had at least one associated coronary artery bypass graft at the time of AVR. AVR rates were lower in patients with LF (76% for SVi > 35 mL/m2, 72% for SVi 30–35 mL/m2, and 62% for SVi < 30 mL/m2, P-value 0.001). Outcome impact of stroke volume index Estimated 5-year survival was 72 (69–75)% for SVi > 35 mL/m2, 69 (62–76)% for SVi 30–35 mL/m2, and 53 (44–60)% for SVi < 30 mL/m2 (overall P-value 0.001, Figure 1A). On multivariable analysis, after adjustment for age, sex, body mass index, Charlson comorbidity index, systolic blood pressure, symptoms, hypertension, coronary artery disease, atrial fibrillation, LVEF, Vmax, AVA, and indexed LV mass, the risk of death associated with SVi < 30 mL/m2 was significant [adjusted hazard ratio (HR) 1.59 (1.17–2.18)] compared to SVi > 35 mL/m2, while risk was similar for SVi 30–35 mL/m2 and >35 mL/m2 (Table 2; Supplementary material online, Table S2). Further adjustment for surgery did not influence the strength of the association between SVi < 30 mL/m2 and mortality [adjusted HR 1.60 (1.17–2.18); Table 2; Figure 1B], while there was no excess risk associated with SVi 30–35 mL/m2 vs. >35 mL/m2 (Table 2; Figure 1B). Five-year survival of patients with SVi < 30 mL/m2 was significantly lower than that of patients with SVi ≥ 30 mL/m2 (53 ± 4% vs. 71 ± 2%; P < 0.001; Figure 2A). Compared to SVi ≥ 30 mL/m2, the adjusted mortality risk of patients with SVi < 30 mL/m2 was considerable (Table 2; Figure 2B). The addition of SVi < 30 mL/m2 to a multivariable model including clinical factors and echo-Doppler parameters improved significantly the measures of model performance (Supplementary material online, Table S3). Table 2 Relative risk of all-cause death associated with stroke volume All-cause death HR (95% CI) P-value SVi  Multivariable model without AVRa   >35 mL/m2 Referent   30 to 35 mL/m2 0.98 (0.73–1.32) 0.89   <30 mL/m2 1.59 (1.17–2.18) 0.003   ≥30 mL/m2 Referent   <30 mL/m2 1.60 (1.20–2.14) 0.001  Multivariable model with AVRb   >35 mL/m2 Referent   30 to 35 mL/m2 1.05 (0.78–1.41) 0.76   <30 mL/m2 1.60 (1.17–2.18) 0.003   ≥30 mL/m2 Referent   <30 mL/m2 1.56 (1.17–2.10) 0.002 SV  Multivariable model without AVRa   >70 mL Referent   55 to 70 mL 1.20 (0.93–1.55) 0.16   <55 mL 1.81 (1.30–2.52) <0.001   ≥55 mL/m2 Referent   <55 mL/m2 1.60 (1.26–2.12) 0.001  Multivariable model with AVRb   >70 mL Referent   55 to 70 mL 1.22 (0.94–1.58) 0.12   <55 mL 1.84 (1.32–2.58) <0.001   ≥55 mL/m2 Referent   <55 mL/m2 1.61 (1.21–2.14) 0.001 All-cause death HR (95% CI) P-value SVi  Multivariable model without AVRa   >35 mL/m2 Referent   30 to 35 mL/m2 0.98 (0.73–1.32) 0.89   <30 mL/m2 1.59 (1.17–2.18) 0.003   ≥30 mL/m2 Referent   <30 mL/m2 1.60 (1.20–2.14) 0.001  Multivariable model with AVRb   >35 mL/m2 Referent   30 to 35 mL/m2 1.05 (0.78–1.41) 0.76   <30 mL/m2 1.60 (1.17–2.18) 0.003   ≥30 mL/m2 Referent   <30 mL/m2 1.56 (1.17–2.10) 0.002 SV  Multivariable model without AVRa   >70 mL Referent   55 to 70 mL 1.20 (0.93–1.55) 0.16   <55 mL 1.81 (1.30–2.52) <0.001   ≥55 mL/m2 Referent   <55 mL/m2 1.60 (1.26–2.12) 0.001  Multivariable model with AVRb   >70 mL Referent   55 to 70 mL 1.22 (0.94–1.58) 0.12   <55 mL 1.84 (1.32–2.58) <0.001   ≥55 mL/m2 Referent   <55 mL/m2 1.61 (1.21–2.14) 0.001 Results of multivariable analyses. Charlson comorbidity index does not include age. AVR, aortic valve replacement; CI, confidence interval; HR, hazard ratio; SV, stroke volume; SVi, stroke volume index. a Model is adjusted for age, gender, body mass index, history of hypertension, coronary artery disease, symptoms, Charlson comorbidity index, atrial fibrillation, systolic blood pressure, peak aortic jet velocity, aortic valve area, left ventricular ejection fraction and indexed left ventricular mass. b Model is adjusted for covariates included in the model without AVR and AVR as time-dependent covariate. Table 2 Relative risk of all-cause death associated with stroke volume All-cause death HR (95% CI) P-value SVi  Multivariable model without AVRa   >35 mL/m2 Referent   30 to 35 mL/m2 0.98 (0.73–1.32) 0.89   <30 mL/m2 1.59 (1.17–2.18) 0.003   ≥30 mL/m2 Referent   <30 mL/m2 1.60 (1.20–2.14) 0.001  Multivariable model with AVRb   >35 mL/m2 Referent   30 to 35 mL/m2 1.05 (0.78–1.41) 0.76   <30 mL/m2 1.60 (1.17–2.18) 0.003   ≥30 mL/m2 Referent   <30 mL/m2 1.56 (1.17–2.10) 0.002 SV  Multivariable model without AVRa   >70 mL Referent   55 to 70 mL 1.20 (0.93–1.55) 0.16   <55 mL 1.81 (1.30–2.52) <0.001   ≥55 mL/m2 Referent   <55 mL/m2 1.60 (1.26–2.12) 0.001  Multivariable model with AVRb   >70 mL Referent   55 to 70 mL 1.22 (0.94–1.58) 0.12   <55 mL 1.84 (1.32–2.58) <0.001   ≥55 mL/m2 Referent   <55 mL/m2 1.61 (1.21–2.14) 0.001 All-cause death HR (95% CI) P-value SVi  Multivariable model without AVRa   >35 mL/m2 Referent   30 to 35 mL/m2 0.98 (0.73–1.32) 0.89   <30 mL/m2 1.59 (1.17–2.18) 0.003   ≥30 mL/m2 Referent   <30 mL/m2 1.60 (1.20–2.14) 0.001  Multivariable model with AVRb   >35 mL/m2 Referent   30 to 35 mL/m2 1.05 (0.78–1.41) 0.76   <30 mL/m2 1.60 (1.17–2.18) 0.003   ≥30 mL/m2 Referent   <30 mL/m2 1.56 (1.17–2.10) 0.002 SV  Multivariable model without AVRa   >70 mL Referent   55 to 70 mL 1.20 (0.93–1.55) 0.16   <55 mL 1.81 (1.30–2.52) <0.001   ≥55 mL/m2 Referent   <55 mL/m2 1.60 (1.26–2.12) 0.001  Multivariable model with AVRb   >70 mL Referent   55 to 70 mL 1.22 (0.94–1.58) 0.12   <55 mL 1.84 (1.32–2.58) <0.001   ≥55 mL/m2 Referent   <55 mL/m2 1.61 (1.21–2.14) 0.001 Results of multivariable analyses. Charlson comorbidity index does not include age. AVR, aortic valve replacement; CI, confidence interval; HR, hazard ratio; SV, stroke volume; SVi, stroke volume index. a Model is adjusted for age, gender, body mass index, history of hypertension, coronary artery disease, symptoms, Charlson comorbidity index, atrial fibrillation, systolic blood pressure, peak aortic jet velocity, aortic valve area, left ventricular ejection fraction and indexed left ventricular mass. b Model is adjusted for covariates included in the model without AVR and AVR as time-dependent covariate. Figure 1 View largeDownload slide (A) The Kaplan–Meier survival curves according to stroke volume index groups. (B) Adjusted survival curves according to stroke volume index groups. Curves are adjusted for age, gender, body mass index, history of hypertension, coronary artery disease, symptoms, Charlson comorbidity index, atrial fibrillation, systolic blood pressure, peak aortic jet velocity, aortic valve area, left ventricular ejection fraction, indexed left ventricular mass, and surgery as time-dependent variable. SVi, stroke volume index. Figure 1 View largeDownload slide (A) The Kaplan–Meier survival curves according to stroke volume index groups. (B) Adjusted survival curves according to stroke volume index groups. Curves are adjusted for age, gender, body mass index, history of hypertension, coronary artery disease, symptoms, Charlson comorbidity index, atrial fibrillation, systolic blood pressure, peak aortic jet velocity, aortic valve area, left ventricular ejection fraction, indexed left ventricular mass, and surgery as time-dependent variable. SVi, stroke volume index. Figure 2 View largeDownload slide (A) The Kaplan–Meier survival curves of patients with stroke volume index < 30 mL/m2 vs. ≥30 mL/m2. (B) Adjusted survival curves of patients with stroke volume index <30 mL/m2 vs. ≥30 mL/m2. Covariates used for multivariable analysis are the same as in Figure 1B. SVi, stroke volume index. Figure 2 View largeDownload slide (A) The Kaplan–Meier survival curves of patients with stroke volume index < 30 mL/m2 vs. ≥30 mL/m2. (B) Adjusted survival curves of patients with stroke volume index <30 mL/m2 vs. ≥30 mL/m2. Covariates used for multivariable analysis are the same as in Figure 1B. SVi, stroke volume index. When SVi was further stratified above the 35 mL/m2 cut-point (Supplementary material online, Figure S1), on multivariable analysis, SVi < 30 mL/m2 remained associated with increased mortality risk compared to SVi ≥ 45 mL/m2 [adjusted HR 1.59 (1.10–2.31)] while mortality risk was similar for SVi 30–35 mL/m2 and ≥45 mL/m2 [adjusted HR 1.09 (0.92–1.52)] and for SVi 35–45 mL/m2 and ≥45 mL/m2 [adjusted HR 1.01 (0.85–1.31)]. By time-dependent ROC analysis, SVi at a cut-point of 35 mL/m2 had a sensitivity of 58.7 ± 3.3% and a specificity of 25.6 ± 1.4% for 5-year mortality prediction. The 30 mL/m2 cut-point had 79.4 ± 2.1% sensitivity and 15.3 ± 1.0% specificity while for the 45 mL/m2 cut-point sensitivity and specificity were 21.2 ± 2.7% and, respectively, 67.0 ± 2.2%. The association of SVi < 30 mL/m2 and mortality risk was consistent in subgroups of patients with severe AS (Figure 3A). There were no significant interactions between SVi < 30 mL/m2 and any of the subgroups. There was no interaction between small (<1.6 m2) or large (>2 m2) BSA and the prognostic effect of SVi < 30 mL/m2 (both P for interaction ≥0.10). In patients with LG (<40 mmHg) severe AS (n = 395), on multivariable analysis, the risk of death associated with SVi < 30 mL/m2 was significant compared to the referent group (>35 mL/m2), while mortality risk was similar for SVi 30–35 mL/m2 and >35 mL/m2 (Supplementary material online, Table S4). Figure 3 View largeDownload slide (A) Hazard ratio and 95% confidence interval for risk of events associated with stroke volume index <30 mL/m2 in subgroups of patients with severe aortic stenosis. (B) Hazard ratio and 95% confidence interval for risk of events associated with stroke volume < 55 mL in subgroups of patients with severe aortic stenosis. CI, confidence interval; HR, hazard ratio. Figure 3 View largeDownload slide (A) Hazard ratio and 95% confidence interval for risk of events associated with stroke volume index <30 mL/m2 in subgroups of patients with severe aortic stenosis. (B) Hazard ratio and 95% confidence interval for risk of events associated with stroke volume < 55 mL in subgroups of patients with severe aortic stenosis. CI, confidence interval; HR, hazard ratio. Outcome impact of stroke volume Estimated 5-year survival was 75 (72–79)% for SV > 70 mL, 69 (64–74)% for SV 55–70 mL, and 49 (39–55)% for SV < 55 mL (overall P-value <0.001, Figure 4A). On multivariable analysis, after adjustment for covariates of prognostic importance, SV < 55 mL was strongly and independently associated with all-cause death [adjusted HR 1.81 (1.30–2.52); Table 2]. The risk of death associated with SV 55–70 mL was comparable to that of patients with SV > 70 mL (Table 2; Supplementary material online, Table S2). After further adjustment for surgery, SV < 55 mL remained independently predictive of mortality [adjusted HR 1.84 (1.32–2.58]; Table 2; Figure 4B], while the adjusted mortality risk of patients with SV 55–70 mL and with SV > 70 mL was similar (Table 2; Figure 4B). Five-year survival of patients with SV < 55 mL was significantly lower than that of patients with SV ≥ 55 mL (49 ± 4% vs. 72 ± 2%; P < 0.001; Figure 5A). Compared to SV ≥ 55 mL, the adjusted mortality risk of patients with SV < 55 mL was considerable (Table 2; Figure 5B). Model performance improved significantly when SV < 55 mL was added to a multivariable model including clinical factors and echo-Doppler parameters (Supplementary material online, Table S3). Figure 4 View largeDownload slide (A) The Kaplan–Meier survival curves according to stroke volume groups. (B) Adjusted survival curves according to stroke volume groups. Covariates used for multivariable analysis are the same as in Figure 1B. SV, stroke volume. Figure 4 View largeDownload slide (A) The Kaplan–Meier survival curves according to stroke volume groups. (B) Adjusted survival curves according to stroke volume groups. Covariates used for multivariable analysis are the same as in Figure 1B. SV, stroke volume. Figure 5 View largeDownload slide (A) The Kaplan–Meier survival curves of patients with stroke volume <55 mL vs. ≥55 mL. (B) Adjusted survival curves of patients with stroke volume index <55 mL vs. ≥55 mL. Covariates used for multivariable analysis are the same as in Figure 1B. HR, hazard ratio; SV, stroke volume. Figure 5 View largeDownload slide (A) The Kaplan–Meier survival curves of patients with stroke volume <55 mL vs. ≥55 mL. (B) Adjusted survival curves of patients with stroke volume index <55 mL vs. ≥55 mL. Covariates used for multivariable analysis are the same as in Figure 1B. HR, hazard ratio; SV, stroke volume. When SV was further stratified above the 70 mL cut-point (Supplementary material online, Figure S1), on multivariable analysis, SV < 55 mL remained associated with increased mortality compared to SV ≥ 85 mL [adjusted HR 1.66 (1.12–2.48)] while mortality risk was similar for SV 55–70 mL/m2 and ≥85 mL [adjusted HR 1.14 (0.83–1.57)] and for SV 70–85 mL and ≥85 mL [adjusted HR 1.04 (0.78–1.22)]. The 70 mL SV cut-point had a sensitivity of 46.7 ± 2.7% and a specificity of 35.1 ± 2.7% for predicting 5-year mortality. The 55 mL cut-point had 77.7 ± 2.3% sensitivity and 13.1 ± 1.5% specificity while for the 85 mL cut-point sensitivity and specificity were 20.7 ± 2.2% and, respectively, 69.7 ± 2.2%. We further explored the association of SV < 55 mL and mortality risk in subgroups of patients with severe AS (Figure 3B). The only significant interaction was with BSA > 2.0 m2 (Figure 3B). There was no interaction between small (<1.6 m2) BSA and the prognostic effect of SV < 55 mL (P for interaction 0.38). In the subgroup with LG severe AS, mortality risk was similar for SV 55–70 mL and >70 mL, while SV < 55 mL was significantly associated with excess mortality (Supplementary material online, Table S4). Discussion The present study based on a large registry of patients with severe AS and preserved LVEF managed in routine clinical practice demonstrates that baseline SVi and SV assessed by Doppler-echocardiography are major independent determinants of long-term outcome under medical and surgical management. First, the effect of LF on mortality is powerful and remains valid after adjustment for factors known as major determinants of outcome such as age, comorbidity, symptoms, blood pressure, atrial fibrillation, coronary artery disease, LVEF, Vmax, and surgery. Second, LF defined by SVi < 30 mL/m2 or non-indexed SV < 55 mL (observed in 13% and respectively 14% of the study population) is associated with more than 50% increase of the risk of all-cause mortality during follow-up, irrespective of baseline characteristics such as symptoms, pressure gradients across the aortic valve and management type. Our analysis challenges the current 35 mL/m2 value used to define LF in patients with AS and preserved LVEF, showing that 30 mL/m2 and 55 mL are turning points of mortality. Conversely, above these values outcome is not affected by further flow stratification. Flow in the intermediate range (SVi 30 to 35 mL/m2 and SV 55 to 70 mL) is more frequent (25% of patients) but does not imply excess mortality risk compared to SVi > 35 mL/m2 and SV > 70 mL. Finally, we report for the first time the link between non-indexed SV and mortality, and show that the 55 mL value can be reliably used in the vast majority of patients (with the exception of those with large body size). Thus, SVi < 30 mL/m2 and SV < 55 mL identify patients with severe AS and preserved LVEF at high risk of death and should be used in clinical practice for risk stratification proposes. The assessment of the SV is routinely performed by Doppler-echocardiography at the level of the LV outflow tract.12 The Doppler-derived measurement of SV has shown good correlation with invasive calculations9–11 and is widely used in clinical practice and recommended by guidelines.12 Despite the widespread use of Doppler-derived SV, published normal reference values in healthy individuals are scarce.13–15 In the report by Chin et al.,13 mean SV was 70 mL and mean SVi was 38 mL/m2 among control individuals. In a longitudinal population study of 2524 individuals (mean age 46 years) free from cardiovascular disease, mean SVi was 39 mL/m2, and 31% had SVi < 35 mL/m2.14 In a healthy population of 584 males aged 70, mean SVi was 38 mL/m2.15 In our control population of 1645 individuals with normal echocardiograms, median Doppler-derived SVi was 37.3 (32.0–43.8) mL/m2 and median SV was 70.0 (58.4–83.3) mL. Moreover, almost 40% of these individuals had SVi < 35 mL/m2, suggesting that Doppler-derived LF state as currently defined1,2 is frequent in the healthy general population. The SVi < 35 mL/m2 value is used to define LF in severe AS, although this cut-point has been arbitrarily defined.3–5 According to the data presented above and to previous reports,13–15 the 35 mL/m2 value is in the normal range. Therefore, the 35 mL/m2 value is in our opinion, too high, and a lower value (i.e. 30 mL/m2) should be used for risk stratification. Previous studies have reported that SVi < 35 mL/m2 is associated with poor prognosis in asymptomatic and symptomatic patients with LG AS and preserved LVEF.3–6,19 SVi < 35 mL/m2 was identified as a strong predictor of outcome in a retrospective study including patients with at least mild AS7 and in patients with mild-to-moderate AS included in the Simvastatin Ezetimibe in Aortic Stenois (SEAS) study.8 In the Placement of Aortic Transcatheter Valves (PARTNER) trial and registry, LF (defined as SVi < 35 mL/m2) at baseline was independently predictive of mortality,20 while severe LF at discharge (defined as the lowest tertile of discharge SVi; mean value 23.1 mL/m2) was a strong predictor of poor outcome after transcatheter AVR despite the observed overall beneficial effect of the technique.21 The present study establishes different prognostic LF values based on their association with mortality in a large spectrum of patients with severe AS, irrespective of pressure gradients across the aortic valve. With SVi < 30 mL/m2, mortality risk was important, while SVi 30–35 mL/m2 and 35–45 mL/m2 were not associated with poor outcome. Our study shows that LF can also be reliably defined based on non-indexed SV. In our opinion, both non-indexed SV and SVi are important prognosticators in severe AS, the latter being preferred in patients with large body size. Strengths and limitations Information on follow-up was retrospectively obtained, and therefore this study has the inherent limitations of such analyses. The specific indications for AVR during follow-up were not collected in our database. The elliptical shape of the LV outflow tract22 and the uncertainties regarding the best site for LV outflow tract diameter measurement in patients with severe valve calcification might lead to underestimation of the LV outflow tract cross-sectional area and Doppler-derived SV. Furthermore, Doppler-derived SV values are lower compared to SV assessed by magnetic resonance imaging translating into different SV cut-points associated with mortality. However, the Doppler-derived SV measurement, despite its limitations, is part of any routine echocardiographic examination and, in patients with AS, is systematically used to calculate the AVA by quantitative Doppler-echocardiography.12 This study did not use hemodynamic parameters which might have contributed to more precise identification of patients with LF. We acknowledge that treatments were not standardized and that AVR was less frequently performed in patients with low SV. However, multivariable models were adjusted for AVR using a time-dependent methodology. As numerous parameters were associated with SV, the unadjusted Kaplan–Meier estimates might be confounded and need to be carefully interpreted. Finally, we cannot exclude a certain degree of inclusion bias, especially for the control group. Clinical implication This analysis of a large registry of severe AS with preserved EF in routine clinical practice shows that LF is a valuable parameter in predicting prognosis with medical and surgical management and identifying patients at high risk of death. Our findings challenge the current 35 mL/m2 SVi value used to define LF in severe AS with preserved EF and establish new LF values based on their association with outcome. We demonstrate that SV < 30 mL/m2 or SV < 55 mL are strong independent predictors of mortality irrespective of clinical presentation, mean pressure gradient, and management type. Stroke volume index between 30 and 35 mL/m2 or SV between 55 and 70 mL do not delineate a subgroup with increased mortality compared to SVi > 35 mL/m2 or SV > 70 mL. Therefore, SVi < 30 mL/m2 and SV < 55 mL identify a subgroup at high risk of death under medical and surgical management and should define LF in severe AS with preserved LVEF. Further studies are needed to prospectively test in clinical practice these values for risk stratification and decision making. Supplementary material Supplementary material is available at European Heart Journal online. Conflict of interest: none declared. References 1 Nishimura RA , Otto CM , Bonow RO , Carabello BA , Erwin JP , Guyton RA , O’gara PT , Ruiz CE , Skubas NJ , Sorajja P , Sundt TM , Thomas JD. 2014 AHA/ACC Guideline for the management of patients with valvular heart disease. A Report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines . J Am Coll Cardiol 2014 ; 63 : e57 – 185 . Google Scholar CrossRef Search ADS PubMed 2 Baumgartner H , Falk V , Bax JJ , De Bonis M , Hamm C , Holm PJ , Iung B , Lancellotti P , Lansac E , Rodriguez Muñoz D , Rosenhek R , Sjögren J , Tornos Mas P , Vahanian A , Walther T , Wendler O , Windecker S , Zamorano JL. 2017 ESC/EACTS guidelines for the management of valvular heart disease . Eur Heart J 2017 ; 38 : 2739 – 2791 . Google Scholar CrossRef Search ADS PubMed 3 Hachicha Z , Dumesnil JG , Bogaty P , Pibarot P. Paradoxical low-flow, low-gradient severe aortic stenosis despite preserved ejection fraction is associated with higher afterload and reduced survival . Circulation 2007 ; 115 : 2856 – 2864 . Google Scholar CrossRef Search ADS PubMed 4 Dumesnil JG , Pibarot P , Carabello B. Paradoxical low flow and/or low gradient severe aortic stenosis despite preserved left ventricular ejection fraction: implications for diagnosis and treatment . Eur Heart J 2010 ; 31 : 281 – 289 . Google Scholar CrossRef Search ADS PubMed 5 Clavel MA , Dumesnil JG , Capoulade R , Mathieu P , Sénéchal M , Pibarot P. Outcome of patients with aortic stenosis, small valve area, and low-flow, low-gradient despite preserved left ventricular ejection fraction . J Am Coll Cardiol 2012 ; 60 : 1259 – 1267 . Google Scholar CrossRef Search ADS PubMed 6 Eleid MF , Sorajja P , Michelena HI , Malouf JF , Scott CG , Pellikka PA. Flow-gradient patterns in severe aortic stenosis with preserved ejection fraction: clinical characteristics and predictors of survival . Circulation 2013 ; 128 : 1781 – 1789 . Google Scholar CrossRef Search ADS PubMed 7 Capoulade R , Le Ven F , Clavel MA , Dumesnil JG , Dahou A , Thébault C , Arsenault M , O'Connor K , Bédard É , Beaudoin J , Sénéchal M , Bernier M , Pibarot P. Echocardiographic predictors of outcomes in adults with aortic stenosis . Heart 2016 ; 102 : 934 – 942 . Google Scholar CrossRef Search ADS PubMed 8 Lønnebakken MT , De Simone G , Saeed S , Boman K , Rossebø AB , Bahlmann E , Gohlke-Bärwolf C , Gerdts E. Impact of stroke volume on cardiovascular risk during progression of aortic valve stenosis . Heart 2017 ; 103 : 1443 – 1448 . Google Scholar CrossRef Search ADS PubMed 9 Huntsman LL , Stewart DK , Barnes SR , Franklin SB , Colocousis JS , Hessel EA. Noninvasive Doppler determination of cardiac output in man. Clinical validation . Circulation 1983 ; 67 : 593 – 602 . Google Scholar CrossRef Search ADS PubMed 10 Ihlen H , Amlie JP , Dale J , Forfang K , Nitter-Hauge S , Otterstad JE , Simonsen S , Myhre E. Determination of cardiac output by Doppler echocardiography . Br Heart J 1984 ; 51 : 54 – 60 . Google Scholar CrossRef Search ADS PubMed 11 Dericbourg C , Tribouilloy C , Kugener H , Avinee P , Rey JL , Lesbre JP. Noninvasive measurement of cardiac output by pulsed Doppler echocardiography. Correlation with thermodilution . Arch Mal Coeur Vaiss 1990 ; 83 : 237 – 244 . Google Scholar PubMed 12 Baumgartner H , Hung J , Bermejo J , Chambers JB , Edvardsen T , Goldstein S , Lancellotti P , LeFevre M , Miller F Jr , Otto CM. Recommendations on the echocardiographic assessment of aortic valve stenosis: a focused update from the European Association of Cardiovascular Imaging and the American Society of Echocardiography . J Am Soc Echocardiogr 2017 ; 30 : 372 – 392 . Google Scholar CrossRef Search ADS PubMed 13 Chin CW , Khaw HJ , Luo E , Tan S , White AC , Newby DE , Dweck MR. Echocardiography underestimates stroke volume and aortic valve area: implications for patients with small-area low-gradient aortic stenosis . Can J Cardiol 2014 ; 30 : 1064 – 1072 . Google Scholar CrossRef Search ADS PubMed 14 Chirinos JA , Rietzschel ER , De Buyzere ML , De Bacquer D , Gillebert TC , Gupta AK , Segers P. Arterial load and ventricular-arterial coupling: physiologic relations with body size and effect of obesity . Hypertension 2009 ; 54 : 558 – 566 . Google Scholar CrossRef Search ADS PubMed 15 Andrén B , Lind L , Hedenstierna G , Lithell H. Left ventricular hypertrophy and geometry in a population sample of elderly males . Eur Heart J 1996 ; 17 : 1800 – 1807 . Google Scholar CrossRef Search ADS PubMed 16 Tribouilloy C , Rusinaru D , Maréchaux S , Castel AL , Debry N , Maizel J , Mentaverri R , Kamel S , Slama M , Lévy F. Low-gradient, low-flow severe aortic stenosis with preserved left ventricular ejection fraction: characteristics, outcome, and implications for surgery . J Am Coll Cardiol 2015 ; 65 : 55 – 66 . Google Scholar CrossRef Search ADS PubMed 17 Quiñones MA , Otto CM , Stoddard M , Waggoner A , Zoghbi WA. Recommendations for quantification of Doppler echocardiography: a report from the Doppler Quantification Task Force of the Nomenclature and Standards Committee of the American Society of Echocardiography . J Am Soc Echocardiogr 2002 ; 15 : 167 – 184 . Google Scholar CrossRef Search ADS PubMed 18 Oh JK , Taliercio CP , Holmes DR Jr , Reeder GS , Bailey KR , Seward JB , Tajik AJ. Prediction of the severity of aortic stenosis by Doppler aortic valve area determination: prospective Doppler-catheterization correlation in 100 patients . J Am Coll Cardiol 1988 ; 11 : 1227 – 1234 . Google Scholar CrossRef Search ADS PubMed 19 Eleid MF , Sorajja P , Michelena HI , Malouf JF , Scott CG , Pellikka PA. Survival by stroke volume index in patients with low-gradient normal EF severe aortic stenosis . Heart 2015 ; 101 : 23 – 29 . Google Scholar CrossRef Search ADS PubMed 20 Herrmann HC , Pibarot P , Hueter I , Gertz ZM , Stewart WJ , Kapadia S , Tuzcu EM , Babaliaros V , Thourani V , Szeto WY , Bavaria JE , Kodali S , Hahn RT , Williams M , Miller DC , Douglas PS , Leon MB. Predictors of mortality and outcomes of therapy in low-flow severe aortic stenosis: a Placement of Aortic Transcatheter Valves (PARTNER) trial analysis . Circulation 2013 ; 127 : 2316 – 2326 . Google Scholar CrossRef Search ADS PubMed 21 Anjan VY , Herrmann HC , Pibarot P , Stewart WJ , Kapadia S , Tuzcu EM , Babaliaros V , Thourani VH , Szeto WY , Bavaria JE , Kodali S , Hahn RT , Williams M , Miller DC , Douglas PS , Leon MB. Evaluation of flow after transcatheter aortic valve replacement in patients with low-flow aortic stenosis: a secondary analysis of the PARTNER randomized clinical trial . JAMA Cardiol 2016 ; 1 : 584 – 592 . Google Scholar CrossRef Search ADS PubMed 22 Maes F , Pierard S , de Meester C , Boulif J , Amzulescu M , Vancraeynest D , Pouleur AC , Pasquet A , Gerber B , Vanoverschelde JL. Impact of left ventricular outflow tract ellipticity on the grading of aortic stenosis in patients with normal ejection fraction . J Cardiovasc Magn Reson 2017 ; 19 : 37. Google Scholar CrossRef Search ADS PubMed Published on behalf of the European Society of Cardiology. All rights reserved. © The Author(s) 2018. For permissions, please email: journals.permissions@oup.com. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices) http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png European Heart Journal Oxford University Press

Impact of low stroke volume on mortality in patients with severe aortic stenosis and preserved left ventricular ejection fraction

Loading next page...
 
/lp/ou_press/impact-of-low-stroke-volume-on-mortality-in-patients-with-severe-0JEnygWDxS
Publisher
Oxford University Press
Copyright
Published on behalf of the European Society of Cardiology. All rights reserved. © The Author(s) 2018. For permissions, please email: journals.permissions@oup.com.
ISSN
0195-668X
eISSN
1522-9645
D.O.I.
10.1093/eurheartj/ehy123
Publisher site
See Article on Publisher Site

Abstract

Abstract Aims In patients with severe aortic stenosis (AS) and preserved left ventricular ejection fraction (LVEF), low flow (LF) is currently defined using Doppler-echocardiography by a stroke volume index (SVi)<35 mL/m2. However, the relationship between LF and outcome remains unclear as data on normal reference values defining LF are scarce, and previous studies did not explore the risk associated with other SVi cut-points. We analysed the relationship between LF and mortality in severe AS to establish prognostic LF values associated with mortality risk. Methods and results This study included 1450 consecutive patients with severe AS (aortic valve area <1 cm2 and/or <0.6 cm2/m2) and preserved LVEF and 1645 controls with normal echocardiograms. Patients were stratified in three groups: (i) SVi > 35 mL/m2 or SV > 70 mL; (ii) SVi 30–35 mL/m2 or SV 55–70 mL; and (iii) SVi < 30 mL/m2 or SV < 55 mL. Mortality with medical and surgical management was analysed. Five-year survival was low for SVi < 30 mL/m2 and SV < 55 mL compared to the other groups (all P-values <0.001). After adjustment for outcome predictors, including aortic valve replacement, mortality risk was considerable with SVi < 30 mL/m2 vs. >35 mL/m2 [adjusted hazard ratio (HR) 1.60 (1.17–2.18)] and SV < 55 mL vs. >70 mL [adjusted HR 1.84 (1.32–2.58)]. Similar mortality risk was observed for SVi 30–35 mL/m2 vs. >35 mL/m2 [adjusted HR 1.05 (0.78–1.41)], and for SV 55–70 mL vs. >70 mL [adjusted HR 1.22 (0.94–1.58)]. The prognostic impact of SVi < 30 mL/m2 and SV < 55 mL was consistent in subgroups, including asymptomatic patients and patients with low-gradient severe AS. Conclusion Low flow defined as SVi < 30 mL/m2 or SV < 55 mL is an important outcome predictor in severe AS with preserved LVEF under medical and surgical management. Further studies are needed to prospectively test these values for risk stratification and decision making. Aortic stenosis, Stroke volume, Stroke volume index, Low flow, Doppler-echocardiography, Outcome Introduction Current guidelines define severe aortic stenosis (AS) as an aortic valve area (AVA) < 1 cm2 (>0.6 cm/m2) and a peak aortic jet velocity (Vmax) ≥4 m/s, or a mean pressure gradient ≥40 mmHg in patients with preserved (≥50%) left ventricular ejection fraction (LVEF).1,2 In patients with AS and preserved LVEF, low flow (LF) is currently defined1,2 by a stroke volume index (SVi)  < 35 mL/m2 based on several studies that have arbitrarily used this cut-off value.3–5 Previous studies have suggested that LF, defined as SVi < 35 mL/m2 is associated with poor prognosis in asymptomatic and symptomatic patients with low-gradient (LG) AS and preserved LVEF.3,5–8 The Doppler-derived measurement of stroke volume (SV) at the aortic annulus has shown good correlation with invasive calculations9–11 and is widely used in clinical practice and recommended by guidelines.12 However, published normal reference values in healthy individuals are scarce.13–15 The relationship between SV and mortality has not been described across the whole spectrum of patients with severe AS and, therefore, the SVi value delineating a subgroup of LF severe AS at high risk of death remains unclear. Moreover, non-indexed SV values associated with high mortality risk have never been reported. This study analyses the relationship between SV and SVi measured at the time of AS diagnosis and all-cause mortality during follow-up. We enrolled in two tertiary centres (Amiens, and Lille, France) patients with severe AS and preserved LVEF and aimed to evaluate the predictive value of SV and SVi on outcome with medical and surgical management and establish prognostic LF values associated with mortality risk. Methods Patient population Between 2000 and 2015, patients ≥18 years of age diagnosed with ≥mild AS (aortic leaflet calcification with reduction in systolic movements and Vmax > 2.5 m/s) were prospectively identified and included in an electronic database. We excluded: (i) >mild aortic and/or mitral regurgitation; (ii) prosthetic valves, congenital heart disease, supravalvular or subvalvular AS, or dynamic LV outflow tract obstruction; (iii) mitral stenosis; and (iv) patients who refused to participate in the study. This analysis included 1450 patients with severe AS [defined as AVA < 1 cm2 and/or AVA normalized to body surface area (BSA) <0.6 cm2] and preserved LVEF. Eighty-one patients were excluded because of missing data. Clinical and demographic baseline characteristics were collected.16 An index summating the patient’s individual comorbidities was calculated.16 Controls Using the echocardiography databases, we retrospectively identified between 2014 and 2016, 1645 consecutive individuals ≥18 years of age with normal echocardiograms. These individuals had normal blood pressure, and no personal history of cardiovascular disease. All echocardiograms were validated as normal by physicians experienced in transthoracic echocardiography. We obtained institutional review board authorizations prior to conducting the study. The study was conducted in accordance with institutional policies, national legislation, and the revised Helsinki declaration. Patients gave informed written consent prior to participation in the study. Echocardiography All patients underwent a comprehensive Doppler-echocardiography study, using commercially available ultrasound systems. Aortic flow was recorded using continuous-wave Doppler, systematically in several acoustic windows (apical 5-chamber, right parasternal, suprasternal, epigastric).16 Stroke volume was calculated by multiplying the LV outflow tract area with the LV outflow tract time-velocity integral.3,17 The LV outflow tract diameter was measured in zoomed parasternal long-axis views in early systole at the level of aortic cusp insertion.18 The LV outflow tract time-velocity integral was recorded from the apical 5-chamber view, with the sample volume positioned about 5 mm proximal to the aortic valve.17 Aortic valve area and SV were indexed to BSA. When patients were in sinus rhythm, three cardiac cycles were averaged for all measures. For patients in atrial fibrillation, five cardiac cycles were averaged. Treatment decision and follow-up The majority of patients were followed in the outpatient clinics of the two tertiary centres. The others were followed in public hospitals or private practices by referring cardiologists working together with the tertiary centres. Information on follow-up was obtained by direct patient interview or by repeated follow-up letters and questionnaires to physicians, patients and (if necessary) next of kin. Ninety-three per cent of patients were followed up to 2 years or death. Follow-up was complete up to death or to the end of the study in 1295 patients (89%). The endpoint was overall survival after diagnosis with medical and surgical treatment. Clinical decisions regarding medical management and indications for surgery (presence of symptoms, LVEF impairment, or abnormal exercise test for asymptomatic patients) were made by the heart team with the approval of the patient’s referring cardiologist in accordance with guideline recommendations.16 Statistical analysis Flow across the aortic valve was analysed as variable normalized to BSA (SVi) as well as non-indexed variable (SV). For each of the two variables (SVi and SV), patients were classified in three flow groups: (i) SVi > 35 mL/m2 or SV > 70 mL; (ii) SVi 30–35 mL/m2 or SV 55–70 mL; and (iii) SVi < 30 mL/m2 or SV < 55 mL. Continuous variables were expressed as median (25th and 75th percentiles), and categorical variables were summarized as frequency percentages and counts. Baseline continuous variables were compared across flow groups using the Kruskal–Wallis tests and categorical variables were compared by the Pearson’s χ2 statistics or Fisher’s exact tests. The significance between the highest group and the others was examined if there was a significant difference across groups. Individual differences were compared with Mann–Whitney U tests (with Bonferroni correction for multiple comparisons). Estimated survival rates and 95% confidence intervals (95% CIs) were estimated according to the Kaplan–Meier method and compared with two-sided log-rank tests. Univariate and multivariable analyses of all-cause mortality were performed using Cox proportional hazards models. We did not use model-building techniques and entered in the models covariates of potential prognostic impact on an epidemiological basis. These covariates were: age, sex, body mass index, Charlson comorbidity index (not including age), symptoms (New York Heart Association Class II–IV dyspnoea, angina, or syncope), history of hypertension, coronary artery disease, atrial fibrillation, systolic blood pressure at baseline, Vmax, AVA, LVEF, and indexed LV mass. The effect of aortic valve replacement (AVR) on outcome was analysed as a time-dependent covariate using the entire follow-up. Age, body mass index, comorbidity index, systolic blood pressure, Vmax, AVA, LVEF, and indexed LV mass were used as continuous variables. The proportional hazards assumption was confirmed using statistics and graphs based on the Schoenfeld residuals. To show the additive value of the flow measurements, we computed the likelihood ratios of the following models: Model 1 including clinical factors, Model 2 including clinical factors, Vmax, AVA, LVEF, and indexed LV mass, and Models 3 and 4 including clinical factors, Vmax, AVA, LVEF, indexed LV mass, and flow quantification (SVi or SV). We compared the models using the global χ2 statistic, the Akaike Information Criterion and the Harrell’s C concordance statistic. Additionally, we evaluated the goodness-of-fit of the models using the Gronnesby–Brogan test. We conducted subgroup analyses to determine the homogeneity of the association of SVi and SV and mortality. First, we estimated the effect of SVi and SV on mortality in each subgroup using a Cox univariate model and then formally tested for first-order interactions entering interaction terms, separately for each subgroup. A significance level of 0.05 was assumed for all tests. All P-values are results of two-tailed tests. Data were analysed with SPSS (v 18.0; IBM Corp, Armonk, NY, USA) and STATA (version 12, StataCorp LP, College Station, TX, USA). Estimates of sensitivity and specificity for various SVi/SV cut-points were computed from time-dependent receiver operating curves (ROC) using the ‘timeROC’ package in R (R project for Statistical Computing, version 3.3.3, https://www.r-project.org/, 25 November 2017). Results Stroke volume by Doppler echocardiography in normal individuals In the cohort of 1645 consecutive individuals [age: 65 (54–77) years, 53% males] with normal echocardiograms, median BSA, LV outflow tract diameter, and LV outflow tract time-velocity integral were 1.88 (1.72–2.02) m2, 22 (20–24) mm, and, respectively, 18.5 (16.2–21.1) cm. Median LVEF was 62.8 (55.7–71.5) %. Median SVi was 37.3 (32.0–43.8) mL/m2 and median SV 70.0 (58.4–83.3) mL. Baseline characteristics of patients with aortic stenosis according to stroke volume Patients with AS were stratified in three flow groups: (i) SVi > 35 mL/m2 or SV > 70 mL; (ii) SVi 30–35 mL/m2 or SV 55–70 mL; and (iii) SVi < 30 mL/m2 or SV < 55 mL. The baseline characteristics of the 1450 patients, according to SVi are presented in Table 1. Patients with SVi < 30 mL/m2 were more often diabetic, with greater BSA and body mass index, and lower systolic blood pressure compared to those with SVi > 35 mL/m2. Almost 40% of patients with SVi < 30 mL/m2 were in atrial fibrillation compared to 27% of patients with SVi > 35 mL/m2 (Table 1). Low SV groups tended to have lower blood pressure and less frequent history of hypertension than higher SV groups due to the inclusion of patients with LF high-gradient severe AS. Low SV groups include both patients with low (<40 mmHg) and high gradient. In the subgroup of patients with LG/LF severe AS, the frequency of history of hypertension was higher, of 73%, as expected. As regard to echo-Doppler parameters, patients with SVi < 30 mL/m2 had significantly lower Vmax and smaller AVA compared to those with SVi > 35 mL/m2. Left ventricular ejection fraction and LV mass were greater in patients with SVi > 35 mL/m2 vs. SVi < 30 mL/m2 (Table 1). The baseline characteristics of the study patients stratified according to non-indexed SV are presented in the Supplementary material online, Table S1. Table 1 Baseline demographic, clinical, and echo-Doppler characteristics of patients according to stroke volume index groups Variable SVi < 30 mL/m2 SVi 30 to 35 mL/m2 SVi > 35 mL/m2 P-value (n = 190) (n = 221) (n = 1039) Demographic and clinical characteristics  Age (years) 78.5 (71.0–84.0) 78.2 (72.1–83.0) 77.5 (70.0–82.6) 0.07  Male sex (%, n) 43.7 (83) 48 (106) 49.1 (510) 0.39  Body surface area (m2) 1.9 (1.7–2.0)a 1.89 (1.72–2.0)a 1.82 (1.70–1.99) 0.001  Body mass index (kg/m2) 27.3 (24.2–32.5)a 27.7 (24.5–31.3)b 26.6 (23.7–29.7) <0.001  Systolic blood pressure (mmHg) 130.0 (120.0–145.0)a 140.0 (120.0–150.0) 140.0 (124.0–150.0) 0.004  New York Heart Association class (%, n) 0.53   I–II 73.2 (139) 74.7 (165) 76.6 (796)   III–IV 26.8 (51) 25.3 (56) 23.4 (243)  Angina (%, n) 20.0 (38) 18.1 (40) 23.8 (247) 0.13  Syncope (%, n) 10.0 (19) 13.6 (30) 11.5 (119) 0.51  History of hypertension (%, n) 68.4 (130) 72.4 (160) 74.5 (774) 0.21  Diabetes mellitus (%, n) 40.5 (77)b 33.5 (74) 26.9 (280) <0.001  Coronary artery disease (%, n) 46.8 (89) 52.9 (117) 52.2 (542) 0.36  History of atrial fibrillation (%, n) 39.5 (75)b 33 (73)a 27.4 (285) 0.002  Charlson comorbidity index 2.0 (1.0–4.0) 2.0 (1.0–3.0) 2.0 (1.0–3.0) 0.38 Echocardiography and Doppler parameters  Aortic valve area (cm2) 0.65 (0.50–0.76)b 0.65 (0.53–0.80)b 0.76 (0.64–0.87) <0.001  Indexed aortic valve area (cm2/m2) 0.33 (0.28–0.39)b 0.35 (0.29–0.42)b 0.41 (0.35–0.47) <0.001  Peak aortic jet velocity (m/s) 4.1 (3.3–4.5)b 4.2 (3.7–4.7)b 4.4 (4.0–5.0) <0.001  Transaortic mean pressure gradient (mmHg) 42.0 (29.8–50.2)b 45.0 (33.0–56.0)b 50.0 (41.0–62.0) <0.001  LV outflow tract diameter (mm) 20.0 (19.0–22.0)b 21.0 (20.0–22.0)b 22.0 (20.4–23.0) <0.001  LV outflow tract velocity time integral (cm) 15.0 (13.0–18.0)b 18.0 (16.0–20.0)b 22.0 (20.0–25.0) <0.001  Stroke volume (mL) 50.1 (43.1–56.4)b 60.8 (56.7–66.0)b 81.4 (72.2–91.3) <0.001  Stroke volume index (mL/m2) 27.3 (24.5–29.2)b 33.0 (31.6–34.0)b 43.9 (39.7–49.5) <0.001  LV end-diastolic diameter (mm) 47.1 (42.0–54.0) 48.0 (43.0–53.0) 49.0 (44.0–53.0) 0.084  LV end-systolic diameter (mm) 30.5 (25.0–35.0) 31.0 (27.0–35.0)a 30.0 (26.0–34.0) 0.035  Ejection fraction (%) 60.0 (54.0–66.0)b 62.0 (56.0–66.0)b 65.0 (60.0–70.0) <0.001  Indexed LV mass (g/m2) 112.2 (92.5–144.1)b 127.0 (94.0–152.0)a 128.1 (106.3–156.0) <0.001 Variable SVi < 30 mL/m2 SVi 30 to 35 mL/m2 SVi > 35 mL/m2 P-value (n = 190) (n = 221) (n = 1039) Demographic and clinical characteristics  Age (years) 78.5 (71.0–84.0) 78.2 (72.1–83.0) 77.5 (70.0–82.6) 0.07  Male sex (%, n) 43.7 (83) 48 (106) 49.1 (510) 0.39  Body surface area (m2) 1.9 (1.7–2.0)a 1.89 (1.72–2.0)a 1.82 (1.70–1.99) 0.001  Body mass index (kg/m2) 27.3 (24.2–32.5)a 27.7 (24.5–31.3)b 26.6 (23.7–29.7) <0.001  Systolic blood pressure (mmHg) 130.0 (120.0–145.0)a 140.0 (120.0–150.0) 140.0 (124.0–150.0) 0.004  New York Heart Association class (%, n) 0.53   I–II 73.2 (139) 74.7 (165) 76.6 (796)   III–IV 26.8 (51) 25.3 (56) 23.4 (243)  Angina (%, n) 20.0 (38) 18.1 (40) 23.8 (247) 0.13  Syncope (%, n) 10.0 (19) 13.6 (30) 11.5 (119) 0.51  History of hypertension (%, n) 68.4 (130) 72.4 (160) 74.5 (774) 0.21  Diabetes mellitus (%, n) 40.5 (77)b 33.5 (74) 26.9 (280) <0.001  Coronary artery disease (%, n) 46.8 (89) 52.9 (117) 52.2 (542) 0.36  History of atrial fibrillation (%, n) 39.5 (75)b 33 (73)a 27.4 (285) 0.002  Charlson comorbidity index 2.0 (1.0–4.0) 2.0 (1.0–3.0) 2.0 (1.0–3.0) 0.38 Echocardiography and Doppler parameters  Aortic valve area (cm2) 0.65 (0.50–0.76)b 0.65 (0.53–0.80)b 0.76 (0.64–0.87) <0.001  Indexed aortic valve area (cm2/m2) 0.33 (0.28–0.39)b 0.35 (0.29–0.42)b 0.41 (0.35–0.47) <0.001  Peak aortic jet velocity (m/s) 4.1 (3.3–4.5)b 4.2 (3.7–4.7)b 4.4 (4.0–5.0) <0.001  Transaortic mean pressure gradient (mmHg) 42.0 (29.8–50.2)b 45.0 (33.0–56.0)b 50.0 (41.0–62.0) <0.001  LV outflow tract diameter (mm) 20.0 (19.0–22.0)b 21.0 (20.0–22.0)b 22.0 (20.4–23.0) <0.001  LV outflow tract velocity time integral (cm) 15.0 (13.0–18.0)b 18.0 (16.0–20.0)b 22.0 (20.0–25.0) <0.001  Stroke volume (mL) 50.1 (43.1–56.4)b 60.8 (56.7–66.0)b 81.4 (72.2–91.3) <0.001  Stroke volume index (mL/m2) 27.3 (24.5–29.2)b 33.0 (31.6–34.0)b 43.9 (39.7–49.5) <0.001  LV end-diastolic diameter (mm) 47.1 (42.0–54.0) 48.0 (43.0–53.0) 49.0 (44.0–53.0) 0.084  LV end-systolic diameter (mm) 30.5 (25.0–35.0) 31.0 (27.0–35.0)a 30.0 (26.0–34.0) 0.035  Ejection fraction (%) 60.0 (54.0–66.0)b 62.0 (56.0–66.0)b 65.0 (60.0–70.0) <0.001  Indexed LV mass (g/m2) 112.2 (92.5–144.1)b 127.0 (94.0–152.0)a 128.1 (106.3–156.0) <0.001 LV, left ventricular; SVi, stroke volume index. Continuous variables are expressed as median (interquartile range) and categorical variables as percentages and counts. a P < 0.05 individual category vs. ‘SVi >35 mL/m2’. b P < 0.001 individual category vs. ‘SVi >35 mL/m2’. Table 1 Baseline demographic, clinical, and echo-Doppler characteristics of patients according to stroke volume index groups Variable SVi < 30 mL/m2 SVi 30 to 35 mL/m2 SVi > 35 mL/m2 P-value (n = 190) (n = 221) (n = 1039) Demographic and clinical characteristics  Age (years) 78.5 (71.0–84.0) 78.2 (72.1–83.0) 77.5 (70.0–82.6) 0.07  Male sex (%, n) 43.7 (83) 48 (106) 49.1 (510) 0.39  Body surface area (m2) 1.9 (1.7–2.0)a 1.89 (1.72–2.0)a 1.82 (1.70–1.99) 0.001  Body mass index (kg/m2) 27.3 (24.2–32.5)a 27.7 (24.5–31.3)b 26.6 (23.7–29.7) <0.001  Systolic blood pressure (mmHg) 130.0 (120.0–145.0)a 140.0 (120.0–150.0) 140.0 (124.0–150.0) 0.004  New York Heart Association class (%, n) 0.53   I–II 73.2 (139) 74.7 (165) 76.6 (796)   III–IV 26.8 (51) 25.3 (56) 23.4 (243)  Angina (%, n) 20.0 (38) 18.1 (40) 23.8 (247) 0.13  Syncope (%, n) 10.0 (19) 13.6 (30) 11.5 (119) 0.51  History of hypertension (%, n) 68.4 (130) 72.4 (160) 74.5 (774) 0.21  Diabetes mellitus (%, n) 40.5 (77)b 33.5 (74) 26.9 (280) <0.001  Coronary artery disease (%, n) 46.8 (89) 52.9 (117) 52.2 (542) 0.36  History of atrial fibrillation (%, n) 39.5 (75)b 33 (73)a 27.4 (285) 0.002  Charlson comorbidity index 2.0 (1.0–4.0) 2.0 (1.0–3.0) 2.0 (1.0–3.0) 0.38 Echocardiography and Doppler parameters  Aortic valve area (cm2) 0.65 (0.50–0.76)b 0.65 (0.53–0.80)b 0.76 (0.64–0.87) <0.001  Indexed aortic valve area (cm2/m2) 0.33 (0.28–0.39)b 0.35 (0.29–0.42)b 0.41 (0.35–0.47) <0.001  Peak aortic jet velocity (m/s) 4.1 (3.3–4.5)b 4.2 (3.7–4.7)b 4.4 (4.0–5.0) <0.001  Transaortic mean pressure gradient (mmHg) 42.0 (29.8–50.2)b 45.0 (33.0–56.0)b 50.0 (41.0–62.0) <0.001  LV outflow tract diameter (mm) 20.0 (19.0–22.0)b 21.0 (20.0–22.0)b 22.0 (20.4–23.0) <0.001  LV outflow tract velocity time integral (cm) 15.0 (13.0–18.0)b 18.0 (16.0–20.0)b 22.0 (20.0–25.0) <0.001  Stroke volume (mL) 50.1 (43.1–56.4)b 60.8 (56.7–66.0)b 81.4 (72.2–91.3) <0.001  Stroke volume index (mL/m2) 27.3 (24.5–29.2)b 33.0 (31.6–34.0)b 43.9 (39.7–49.5) <0.001  LV end-diastolic diameter (mm) 47.1 (42.0–54.0) 48.0 (43.0–53.0) 49.0 (44.0–53.0) 0.084  LV end-systolic diameter (mm) 30.5 (25.0–35.0) 31.0 (27.0–35.0)a 30.0 (26.0–34.0) 0.035  Ejection fraction (%) 60.0 (54.0–66.0)b 62.0 (56.0–66.0)b 65.0 (60.0–70.0) <0.001  Indexed LV mass (g/m2) 112.2 (92.5–144.1)b 127.0 (94.0–152.0)a 128.1 (106.3–156.0) <0.001 Variable SVi < 30 mL/m2 SVi 30 to 35 mL/m2 SVi > 35 mL/m2 P-value (n = 190) (n = 221) (n = 1039) Demographic and clinical characteristics  Age (years) 78.5 (71.0–84.0) 78.2 (72.1–83.0) 77.5 (70.0–82.6) 0.07  Male sex (%, n) 43.7 (83) 48 (106) 49.1 (510) 0.39  Body surface area (m2) 1.9 (1.7–2.0)a 1.89 (1.72–2.0)a 1.82 (1.70–1.99) 0.001  Body mass index (kg/m2) 27.3 (24.2–32.5)a 27.7 (24.5–31.3)b 26.6 (23.7–29.7) <0.001  Systolic blood pressure (mmHg) 130.0 (120.0–145.0)a 140.0 (120.0–150.0) 140.0 (124.0–150.0) 0.004  New York Heart Association class (%, n) 0.53   I–II 73.2 (139) 74.7 (165) 76.6 (796)   III–IV 26.8 (51) 25.3 (56) 23.4 (243)  Angina (%, n) 20.0 (38) 18.1 (40) 23.8 (247) 0.13  Syncope (%, n) 10.0 (19) 13.6 (30) 11.5 (119) 0.51  History of hypertension (%, n) 68.4 (130) 72.4 (160) 74.5 (774) 0.21  Diabetes mellitus (%, n) 40.5 (77)b 33.5 (74) 26.9 (280) <0.001  Coronary artery disease (%, n) 46.8 (89) 52.9 (117) 52.2 (542) 0.36  History of atrial fibrillation (%, n) 39.5 (75)b 33 (73)a 27.4 (285) 0.002  Charlson comorbidity index 2.0 (1.0–4.0) 2.0 (1.0–3.0) 2.0 (1.0–3.0) 0.38 Echocardiography and Doppler parameters  Aortic valve area (cm2) 0.65 (0.50–0.76)b 0.65 (0.53–0.80)b 0.76 (0.64–0.87) <0.001  Indexed aortic valve area (cm2/m2) 0.33 (0.28–0.39)b 0.35 (0.29–0.42)b 0.41 (0.35–0.47) <0.001  Peak aortic jet velocity (m/s) 4.1 (3.3–4.5)b 4.2 (3.7–4.7)b 4.4 (4.0–5.0) <0.001  Transaortic mean pressure gradient (mmHg) 42.0 (29.8–50.2)b 45.0 (33.0–56.0)b 50.0 (41.0–62.0) <0.001  LV outflow tract diameter (mm) 20.0 (19.0–22.0)b 21.0 (20.0–22.0)b 22.0 (20.4–23.0) <0.001  LV outflow tract velocity time integral (cm) 15.0 (13.0–18.0)b 18.0 (16.0–20.0)b 22.0 (20.0–25.0) <0.001  Stroke volume (mL) 50.1 (43.1–56.4)b 60.8 (56.7–66.0)b 81.4 (72.2–91.3) <0.001  Stroke volume index (mL/m2) 27.3 (24.5–29.2)b 33.0 (31.6–34.0)b 43.9 (39.7–49.5) <0.001  LV end-diastolic diameter (mm) 47.1 (42.0–54.0) 48.0 (43.0–53.0) 49.0 (44.0–53.0) 0.084  LV end-systolic diameter (mm) 30.5 (25.0–35.0) 31.0 (27.0–35.0)a 30.0 (26.0–34.0) 0.035  Ejection fraction (%) 60.0 (54.0–66.0)b 62.0 (56.0–66.0)b 65.0 (60.0–70.0) <0.001  Indexed LV mass (g/m2) 112.2 (92.5–144.1)b 127.0 (94.0–152.0)a 128.1 (106.3–156.0) <0.001 LV, left ventricular; SVi, stroke volume index. Continuous variables are expressed as median (interquartile range) and categorical variables as percentages and counts. a P < 0.05 individual category vs. ‘SVi >35 mL/m2’. b P < 0.001 individual category vs. ‘SVi >35 mL/m2’. Outcome impact of stroke volume Clinical management and follow-up Median (25th, 75th percentile) overall follow-up was 38 (17–69) months. The total number of deaths recorded during follow-up was 440. AVR was performed in 1072 patients (74%) by surgical (n = 970, 90.5%) or percutaneous (n = 102, 9.5%) techniques. Aortic bioprostheses were used in 879 patients (82%). One hundred and sixty patients had at least one associated coronary artery bypass graft at the time of AVR. AVR rates were lower in patients with LF (76% for SVi > 35 mL/m2, 72% for SVi 30–35 mL/m2, and 62% for SVi < 30 mL/m2, P-value 0.001). Outcome impact of stroke volume index Estimated 5-year survival was 72 (69–75)% for SVi > 35 mL/m2, 69 (62–76)% for SVi 30–35 mL/m2, and 53 (44–60)% for SVi < 30 mL/m2 (overall P-value 0.001, Figure 1A). On multivariable analysis, after adjustment for age, sex, body mass index, Charlson comorbidity index, systolic blood pressure, symptoms, hypertension, coronary artery disease, atrial fibrillation, LVEF, Vmax, AVA, and indexed LV mass, the risk of death associated with SVi < 30 mL/m2 was significant [adjusted hazard ratio (HR) 1.59 (1.17–2.18)] compared to SVi > 35 mL/m2, while risk was similar for SVi 30–35 mL/m2 and >35 mL/m2 (Table 2; Supplementary material online, Table S2). Further adjustment for surgery did not influence the strength of the association between SVi < 30 mL/m2 and mortality [adjusted HR 1.60 (1.17–2.18); Table 2; Figure 1B], while there was no excess risk associated with SVi 30–35 mL/m2 vs. >35 mL/m2 (Table 2; Figure 1B). Five-year survival of patients with SVi < 30 mL/m2 was significantly lower than that of patients with SVi ≥ 30 mL/m2 (53 ± 4% vs. 71 ± 2%; P < 0.001; Figure 2A). Compared to SVi ≥ 30 mL/m2, the adjusted mortality risk of patients with SVi < 30 mL/m2 was considerable (Table 2; Figure 2B). The addition of SVi < 30 mL/m2 to a multivariable model including clinical factors and echo-Doppler parameters improved significantly the measures of model performance (Supplementary material online, Table S3). Table 2 Relative risk of all-cause death associated with stroke volume All-cause death HR (95% CI) P-value SVi  Multivariable model without AVRa   >35 mL/m2 Referent   30 to 35 mL/m2 0.98 (0.73–1.32) 0.89   <30 mL/m2 1.59 (1.17–2.18) 0.003   ≥30 mL/m2 Referent   <30 mL/m2 1.60 (1.20–2.14) 0.001  Multivariable model with AVRb   >35 mL/m2 Referent   30 to 35 mL/m2 1.05 (0.78–1.41) 0.76   <30 mL/m2 1.60 (1.17–2.18) 0.003   ≥30 mL/m2 Referent   <30 mL/m2 1.56 (1.17–2.10) 0.002 SV  Multivariable model without AVRa   >70 mL Referent   55 to 70 mL 1.20 (0.93–1.55) 0.16   <55 mL 1.81 (1.30–2.52) <0.001   ≥55 mL/m2 Referent   <55 mL/m2 1.60 (1.26–2.12) 0.001  Multivariable model with AVRb   >70 mL Referent   55 to 70 mL 1.22 (0.94–1.58) 0.12   <55 mL 1.84 (1.32–2.58) <0.001   ≥55 mL/m2 Referent   <55 mL/m2 1.61 (1.21–2.14) 0.001 All-cause death HR (95% CI) P-value SVi  Multivariable model without AVRa   >35 mL/m2 Referent   30 to 35 mL/m2 0.98 (0.73–1.32) 0.89   <30 mL/m2 1.59 (1.17–2.18) 0.003   ≥30 mL/m2 Referent   <30 mL/m2 1.60 (1.20–2.14) 0.001  Multivariable model with AVRb   >35 mL/m2 Referent   30 to 35 mL/m2 1.05 (0.78–1.41) 0.76   <30 mL/m2 1.60 (1.17–2.18) 0.003   ≥30 mL/m2 Referent   <30 mL/m2 1.56 (1.17–2.10) 0.002 SV  Multivariable model without AVRa   >70 mL Referent   55 to 70 mL 1.20 (0.93–1.55) 0.16   <55 mL 1.81 (1.30–2.52) <0.001   ≥55 mL/m2 Referent   <55 mL/m2 1.60 (1.26–2.12) 0.001  Multivariable model with AVRb   >70 mL Referent   55 to 70 mL 1.22 (0.94–1.58) 0.12   <55 mL 1.84 (1.32–2.58) <0.001   ≥55 mL/m2 Referent   <55 mL/m2 1.61 (1.21–2.14) 0.001 Results of multivariable analyses. Charlson comorbidity index does not include age. AVR, aortic valve replacement; CI, confidence interval; HR, hazard ratio; SV, stroke volume; SVi, stroke volume index. a Model is adjusted for age, gender, body mass index, history of hypertension, coronary artery disease, symptoms, Charlson comorbidity index, atrial fibrillation, systolic blood pressure, peak aortic jet velocity, aortic valve area, left ventricular ejection fraction and indexed left ventricular mass. b Model is adjusted for covariates included in the model without AVR and AVR as time-dependent covariate. Table 2 Relative risk of all-cause death associated with stroke volume All-cause death HR (95% CI) P-value SVi  Multivariable model without AVRa   >35 mL/m2 Referent   30 to 35 mL/m2 0.98 (0.73–1.32) 0.89   <30 mL/m2 1.59 (1.17–2.18) 0.003   ≥30 mL/m2 Referent   <30 mL/m2 1.60 (1.20–2.14) 0.001  Multivariable model with AVRb   >35 mL/m2 Referent   30 to 35 mL/m2 1.05 (0.78–1.41) 0.76   <30 mL/m2 1.60 (1.17–2.18) 0.003   ≥30 mL/m2 Referent   <30 mL/m2 1.56 (1.17–2.10) 0.002 SV  Multivariable model without AVRa   >70 mL Referent   55 to 70 mL 1.20 (0.93–1.55) 0.16   <55 mL 1.81 (1.30–2.52) <0.001   ≥55 mL/m2 Referent   <55 mL/m2 1.60 (1.26–2.12) 0.001  Multivariable model with AVRb   >70 mL Referent   55 to 70 mL 1.22 (0.94–1.58) 0.12   <55 mL 1.84 (1.32–2.58) <0.001   ≥55 mL/m2 Referent   <55 mL/m2 1.61 (1.21–2.14) 0.001 All-cause death HR (95% CI) P-value SVi  Multivariable model without AVRa   >35 mL/m2 Referent   30 to 35 mL/m2 0.98 (0.73–1.32) 0.89   <30 mL/m2 1.59 (1.17–2.18) 0.003   ≥30 mL/m2 Referent   <30 mL/m2 1.60 (1.20–2.14) 0.001  Multivariable model with AVRb   >35 mL/m2 Referent   30 to 35 mL/m2 1.05 (0.78–1.41) 0.76   <30 mL/m2 1.60 (1.17–2.18) 0.003   ≥30 mL/m2 Referent   <30 mL/m2 1.56 (1.17–2.10) 0.002 SV  Multivariable model without AVRa   >70 mL Referent   55 to 70 mL 1.20 (0.93–1.55) 0.16   <55 mL 1.81 (1.30–2.52) <0.001   ≥55 mL/m2 Referent   <55 mL/m2 1.60 (1.26–2.12) 0.001  Multivariable model with AVRb   >70 mL Referent   55 to 70 mL 1.22 (0.94–1.58) 0.12   <55 mL 1.84 (1.32–2.58) <0.001   ≥55 mL/m2 Referent   <55 mL/m2 1.61 (1.21–2.14) 0.001 Results of multivariable analyses. Charlson comorbidity index does not include age. AVR, aortic valve replacement; CI, confidence interval; HR, hazard ratio; SV, stroke volume; SVi, stroke volume index. a Model is adjusted for age, gender, body mass index, history of hypertension, coronary artery disease, symptoms, Charlson comorbidity index, atrial fibrillation, systolic blood pressure, peak aortic jet velocity, aortic valve area, left ventricular ejection fraction and indexed left ventricular mass. b Model is adjusted for covariates included in the model without AVR and AVR as time-dependent covariate. Figure 1 View largeDownload slide (A) The Kaplan–Meier survival curves according to stroke volume index groups. (B) Adjusted survival curves according to stroke volume index groups. Curves are adjusted for age, gender, body mass index, history of hypertension, coronary artery disease, symptoms, Charlson comorbidity index, atrial fibrillation, systolic blood pressure, peak aortic jet velocity, aortic valve area, left ventricular ejection fraction, indexed left ventricular mass, and surgery as time-dependent variable. SVi, stroke volume index. Figure 1 View largeDownload slide (A) The Kaplan–Meier survival curves according to stroke volume index groups. (B) Adjusted survival curves according to stroke volume index groups. Curves are adjusted for age, gender, body mass index, history of hypertension, coronary artery disease, symptoms, Charlson comorbidity index, atrial fibrillation, systolic blood pressure, peak aortic jet velocity, aortic valve area, left ventricular ejection fraction, indexed left ventricular mass, and surgery as time-dependent variable. SVi, stroke volume index. Figure 2 View largeDownload slide (A) The Kaplan–Meier survival curves of patients with stroke volume index < 30 mL/m2 vs. ≥30 mL/m2. (B) Adjusted survival curves of patients with stroke volume index <30 mL/m2 vs. ≥30 mL/m2. Covariates used for multivariable analysis are the same as in Figure 1B. SVi, stroke volume index. Figure 2 View largeDownload slide (A) The Kaplan–Meier survival curves of patients with stroke volume index < 30 mL/m2 vs. ≥30 mL/m2. (B) Adjusted survival curves of patients with stroke volume index <30 mL/m2 vs. ≥30 mL/m2. Covariates used for multivariable analysis are the same as in Figure 1B. SVi, stroke volume index. When SVi was further stratified above the 35 mL/m2 cut-point (Supplementary material online, Figure S1), on multivariable analysis, SVi < 30 mL/m2 remained associated with increased mortality risk compared to SVi ≥ 45 mL/m2 [adjusted HR 1.59 (1.10–2.31)] while mortality risk was similar for SVi 30–35 mL/m2 and ≥45 mL/m2 [adjusted HR 1.09 (0.92–1.52)] and for SVi 35–45 mL/m2 and ≥45 mL/m2 [adjusted HR 1.01 (0.85–1.31)]. By time-dependent ROC analysis, SVi at a cut-point of 35 mL/m2 had a sensitivity of 58.7 ± 3.3% and a specificity of 25.6 ± 1.4% for 5-year mortality prediction. The 30 mL/m2 cut-point had 79.4 ± 2.1% sensitivity and 15.3 ± 1.0% specificity while for the 45 mL/m2 cut-point sensitivity and specificity were 21.2 ± 2.7% and, respectively, 67.0 ± 2.2%. The association of SVi < 30 mL/m2 and mortality risk was consistent in subgroups of patients with severe AS (Figure 3A). There were no significant interactions between SVi < 30 mL/m2 and any of the subgroups. There was no interaction between small (<1.6 m2) or large (>2 m2) BSA and the prognostic effect of SVi < 30 mL/m2 (both P for interaction ≥0.10). In patients with LG (<40 mmHg) severe AS (n = 395), on multivariable analysis, the risk of death associated with SVi < 30 mL/m2 was significant compared to the referent group (>35 mL/m2), while mortality risk was similar for SVi 30–35 mL/m2 and >35 mL/m2 (Supplementary material online, Table S4). Figure 3 View largeDownload slide (A) Hazard ratio and 95% confidence interval for risk of events associated with stroke volume index <30 mL/m2 in subgroups of patients with severe aortic stenosis. (B) Hazard ratio and 95% confidence interval for risk of events associated with stroke volume < 55 mL in subgroups of patients with severe aortic stenosis. CI, confidence interval; HR, hazard ratio. Figure 3 View largeDownload slide (A) Hazard ratio and 95% confidence interval for risk of events associated with stroke volume index <30 mL/m2 in subgroups of patients with severe aortic stenosis. (B) Hazard ratio and 95% confidence interval for risk of events associated with stroke volume < 55 mL in subgroups of patients with severe aortic stenosis. CI, confidence interval; HR, hazard ratio. Outcome impact of stroke volume Estimated 5-year survival was 75 (72–79)% for SV > 70 mL, 69 (64–74)% for SV 55–70 mL, and 49 (39–55)% for SV < 55 mL (overall P-value <0.001, Figure 4A). On multivariable analysis, after adjustment for covariates of prognostic importance, SV < 55 mL was strongly and independently associated with all-cause death [adjusted HR 1.81 (1.30–2.52); Table 2]. The risk of death associated with SV 55–70 mL was comparable to that of patients with SV > 70 mL (Table 2; Supplementary material online, Table S2). After further adjustment for surgery, SV < 55 mL remained independently predictive of mortality [adjusted HR 1.84 (1.32–2.58]; Table 2; Figure 4B], while the adjusted mortality risk of patients with SV 55–70 mL and with SV > 70 mL was similar (Table 2; Figure 4B). Five-year survival of patients with SV < 55 mL was significantly lower than that of patients with SV ≥ 55 mL (49 ± 4% vs. 72 ± 2%; P < 0.001; Figure 5A). Compared to SV ≥ 55 mL, the adjusted mortality risk of patients with SV < 55 mL was considerable (Table 2; Figure 5B). Model performance improved significantly when SV < 55 mL was added to a multivariable model including clinical factors and echo-Doppler parameters (Supplementary material online, Table S3). Figure 4 View largeDownload slide (A) The Kaplan–Meier survival curves according to stroke volume groups. (B) Adjusted survival curves according to stroke volume groups. Covariates used for multivariable analysis are the same as in Figure 1B. SV, stroke volume. Figure 4 View largeDownload slide (A) The Kaplan–Meier survival curves according to stroke volume groups. (B) Adjusted survival curves according to stroke volume groups. Covariates used for multivariable analysis are the same as in Figure 1B. SV, stroke volume. Figure 5 View largeDownload slide (A) The Kaplan–Meier survival curves of patients with stroke volume <55 mL vs. ≥55 mL. (B) Adjusted survival curves of patients with stroke volume index <55 mL vs. ≥55 mL. Covariates used for multivariable analysis are the same as in Figure 1B. HR, hazard ratio; SV, stroke volume. Figure 5 View largeDownload slide (A) The Kaplan–Meier survival curves of patients with stroke volume <55 mL vs. ≥55 mL. (B) Adjusted survival curves of patients with stroke volume index <55 mL vs. ≥55 mL. Covariates used for multivariable analysis are the same as in Figure 1B. HR, hazard ratio; SV, stroke volume. When SV was further stratified above the 70 mL cut-point (Supplementary material online, Figure S1), on multivariable analysis, SV < 55 mL remained associated with increased mortality compared to SV ≥ 85 mL [adjusted HR 1.66 (1.12–2.48)] while mortality risk was similar for SV 55–70 mL/m2 and ≥85 mL [adjusted HR 1.14 (0.83–1.57)] and for SV 70–85 mL and ≥85 mL [adjusted HR 1.04 (0.78–1.22)]. The 70 mL SV cut-point had a sensitivity of 46.7 ± 2.7% and a specificity of 35.1 ± 2.7% for predicting 5-year mortality. The 55 mL cut-point had 77.7 ± 2.3% sensitivity and 13.1 ± 1.5% specificity while for the 85 mL cut-point sensitivity and specificity were 20.7 ± 2.2% and, respectively, 69.7 ± 2.2%. We further explored the association of SV < 55 mL and mortality risk in subgroups of patients with severe AS (Figure 3B). The only significant interaction was with BSA > 2.0 m2 (Figure 3B). There was no interaction between small (<1.6 m2) BSA and the prognostic effect of SV < 55 mL (P for interaction 0.38). In the subgroup with LG severe AS, mortality risk was similar for SV 55–70 mL and >70 mL, while SV < 55 mL was significantly associated with excess mortality (Supplementary material online, Table S4). Discussion The present study based on a large registry of patients with severe AS and preserved LVEF managed in routine clinical practice demonstrates that baseline SVi and SV assessed by Doppler-echocardiography are major independent determinants of long-term outcome under medical and surgical management. First, the effect of LF on mortality is powerful and remains valid after adjustment for factors known as major determinants of outcome such as age, comorbidity, symptoms, blood pressure, atrial fibrillation, coronary artery disease, LVEF, Vmax, and surgery. Second, LF defined by SVi < 30 mL/m2 or non-indexed SV < 55 mL (observed in 13% and respectively 14% of the study population) is associated with more than 50% increase of the risk of all-cause mortality during follow-up, irrespective of baseline characteristics such as symptoms, pressure gradients across the aortic valve and management type. Our analysis challenges the current 35 mL/m2 value used to define LF in patients with AS and preserved LVEF, showing that 30 mL/m2 and 55 mL are turning points of mortality. Conversely, above these values outcome is not affected by further flow stratification. Flow in the intermediate range (SVi 30 to 35 mL/m2 and SV 55 to 70 mL) is more frequent (25% of patients) but does not imply excess mortality risk compared to SVi > 35 mL/m2 and SV > 70 mL. Finally, we report for the first time the link between non-indexed SV and mortality, and show that the 55 mL value can be reliably used in the vast majority of patients (with the exception of those with large body size). Thus, SVi < 30 mL/m2 and SV < 55 mL identify patients with severe AS and preserved LVEF at high risk of death and should be used in clinical practice for risk stratification proposes. The assessment of the SV is routinely performed by Doppler-echocardiography at the level of the LV outflow tract.12 The Doppler-derived measurement of SV has shown good correlation with invasive calculations9–11 and is widely used in clinical practice and recommended by guidelines.12 Despite the widespread use of Doppler-derived SV, published normal reference values in healthy individuals are scarce.13–15 In the report by Chin et al.,13 mean SV was 70 mL and mean SVi was 38 mL/m2 among control individuals. In a longitudinal population study of 2524 individuals (mean age 46 years) free from cardiovascular disease, mean SVi was 39 mL/m2, and 31% had SVi < 35 mL/m2.14 In a healthy population of 584 males aged 70, mean SVi was 38 mL/m2.15 In our control population of 1645 individuals with normal echocardiograms, median Doppler-derived SVi was 37.3 (32.0–43.8) mL/m2 and median SV was 70.0 (58.4–83.3) mL. Moreover, almost 40% of these individuals had SVi < 35 mL/m2, suggesting that Doppler-derived LF state as currently defined1,2 is frequent in the healthy general population. The SVi < 35 mL/m2 value is used to define LF in severe AS, although this cut-point has been arbitrarily defined.3–5 According to the data presented above and to previous reports,13–15 the 35 mL/m2 value is in the normal range. Therefore, the 35 mL/m2 value is in our opinion, too high, and a lower value (i.e. 30 mL/m2) should be used for risk stratification. Previous studies have reported that SVi < 35 mL/m2 is associated with poor prognosis in asymptomatic and symptomatic patients with LG AS and preserved LVEF.3–6,19 SVi < 35 mL/m2 was identified as a strong predictor of outcome in a retrospective study including patients with at least mild AS7 and in patients with mild-to-moderate AS included in the Simvastatin Ezetimibe in Aortic Stenois (SEAS) study.8 In the Placement of Aortic Transcatheter Valves (PARTNER) trial and registry, LF (defined as SVi < 35 mL/m2) at baseline was independently predictive of mortality,20 while severe LF at discharge (defined as the lowest tertile of discharge SVi; mean value 23.1 mL/m2) was a strong predictor of poor outcome after transcatheter AVR despite the observed overall beneficial effect of the technique.21 The present study establishes different prognostic LF values based on their association with mortality in a large spectrum of patients with severe AS, irrespective of pressure gradients across the aortic valve. With SVi < 30 mL/m2, mortality risk was important, while SVi 30–35 mL/m2 and 35–45 mL/m2 were not associated with poor outcome. Our study shows that LF can also be reliably defined based on non-indexed SV. In our opinion, both non-indexed SV and SVi are important prognosticators in severe AS, the latter being preferred in patients with large body size. Strengths and limitations Information on follow-up was retrospectively obtained, and therefore this study has the inherent limitations of such analyses. The specific indications for AVR during follow-up were not collected in our database. The elliptical shape of the LV outflow tract22 and the uncertainties regarding the best site for LV outflow tract diameter measurement in patients with severe valve calcification might lead to underestimation of the LV outflow tract cross-sectional area and Doppler-derived SV. Furthermore, Doppler-derived SV values are lower compared to SV assessed by magnetic resonance imaging translating into different SV cut-points associated with mortality. However, the Doppler-derived SV measurement, despite its limitations, is part of any routine echocardiographic examination and, in patients with AS, is systematically used to calculate the AVA by quantitative Doppler-echocardiography.12 This study did not use hemodynamic parameters which might have contributed to more precise identification of patients with LF. We acknowledge that treatments were not standardized and that AVR was less frequently performed in patients with low SV. However, multivariable models were adjusted for AVR using a time-dependent methodology. As numerous parameters were associated with SV, the unadjusted Kaplan–Meier estimates might be confounded and need to be carefully interpreted. Finally, we cannot exclude a certain degree of inclusion bias, especially for the control group. Clinical implication This analysis of a large registry of severe AS with preserved EF in routine clinical practice shows that LF is a valuable parameter in predicting prognosis with medical and surgical management and identifying patients at high risk of death. Our findings challenge the current 35 mL/m2 SVi value used to define LF in severe AS with preserved EF and establish new LF values based on their association with outcome. We demonstrate that SV < 30 mL/m2 or SV < 55 mL are strong independent predictors of mortality irrespective of clinical presentation, mean pressure gradient, and management type. Stroke volume index between 30 and 35 mL/m2 or SV between 55 and 70 mL do not delineate a subgroup with increased mortality compared to SVi > 35 mL/m2 or SV > 70 mL. Therefore, SVi < 30 mL/m2 and SV < 55 mL identify a subgroup at high risk of death under medical and surgical management and should define LF in severe AS with preserved LVEF. Further studies are needed to prospectively test in clinical practice these values for risk stratification and decision making. Supplementary material Supplementary material is available at European Heart Journal online. Conflict of interest: none declared. References 1 Nishimura RA , Otto CM , Bonow RO , Carabello BA , Erwin JP , Guyton RA , O’gara PT , Ruiz CE , Skubas NJ , Sorajja P , Sundt TM , Thomas JD. 2014 AHA/ACC Guideline for the management of patients with valvular heart disease. A Report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines . J Am Coll Cardiol 2014 ; 63 : e57 – 185 . Google Scholar CrossRef Search ADS PubMed 2 Baumgartner H , Falk V , Bax JJ , De Bonis M , Hamm C , Holm PJ , Iung B , Lancellotti P , Lansac E , Rodriguez Muñoz D , Rosenhek R , Sjögren J , Tornos Mas P , Vahanian A , Walther T , Wendler O , Windecker S , Zamorano JL. 2017 ESC/EACTS guidelines for the management of valvular heart disease . Eur Heart J 2017 ; 38 : 2739 – 2791 . Google Scholar CrossRef Search ADS PubMed 3 Hachicha Z , Dumesnil JG , Bogaty P , Pibarot P. Paradoxical low-flow, low-gradient severe aortic stenosis despite preserved ejection fraction is associated with higher afterload and reduced survival . Circulation 2007 ; 115 : 2856 – 2864 . Google Scholar CrossRef Search ADS PubMed 4 Dumesnil JG , Pibarot P , Carabello B. Paradoxical low flow and/or low gradient severe aortic stenosis despite preserved left ventricular ejection fraction: implications for diagnosis and treatment . Eur Heart J 2010 ; 31 : 281 – 289 . Google Scholar CrossRef Search ADS PubMed 5 Clavel MA , Dumesnil JG , Capoulade R , Mathieu P , Sénéchal M , Pibarot P. Outcome of patients with aortic stenosis, small valve area, and low-flow, low-gradient despite preserved left ventricular ejection fraction . J Am Coll Cardiol 2012 ; 60 : 1259 – 1267 . Google Scholar CrossRef Search ADS PubMed 6 Eleid MF , Sorajja P , Michelena HI , Malouf JF , Scott CG , Pellikka PA. Flow-gradient patterns in severe aortic stenosis with preserved ejection fraction: clinical characteristics and predictors of survival . Circulation 2013 ; 128 : 1781 – 1789 . Google Scholar CrossRef Search ADS PubMed 7 Capoulade R , Le Ven F , Clavel MA , Dumesnil JG , Dahou A , Thébault C , Arsenault M , O'Connor K , Bédard É , Beaudoin J , Sénéchal M , Bernier M , Pibarot P. Echocardiographic predictors of outcomes in adults with aortic stenosis . Heart 2016 ; 102 : 934 – 942 . Google Scholar CrossRef Search ADS PubMed 8 Lønnebakken MT , De Simone G , Saeed S , Boman K , Rossebø AB , Bahlmann E , Gohlke-Bärwolf C , Gerdts E. Impact of stroke volume on cardiovascular risk during progression of aortic valve stenosis . Heart 2017 ; 103 : 1443 – 1448 . Google Scholar CrossRef Search ADS PubMed 9 Huntsman LL , Stewart DK , Barnes SR , Franklin SB , Colocousis JS , Hessel EA. Noninvasive Doppler determination of cardiac output in man. Clinical validation . Circulation 1983 ; 67 : 593 – 602 . Google Scholar CrossRef Search ADS PubMed 10 Ihlen H , Amlie JP , Dale J , Forfang K , Nitter-Hauge S , Otterstad JE , Simonsen S , Myhre E. Determination of cardiac output by Doppler echocardiography . Br Heart J 1984 ; 51 : 54 – 60 . Google Scholar CrossRef Search ADS PubMed 11 Dericbourg C , Tribouilloy C , Kugener H , Avinee P , Rey JL , Lesbre JP. Noninvasive measurement of cardiac output by pulsed Doppler echocardiography. Correlation with thermodilution . Arch Mal Coeur Vaiss 1990 ; 83 : 237 – 244 . Google Scholar PubMed 12 Baumgartner H , Hung J , Bermejo J , Chambers JB , Edvardsen T , Goldstein S , Lancellotti P , LeFevre M , Miller F Jr , Otto CM. Recommendations on the echocardiographic assessment of aortic valve stenosis: a focused update from the European Association of Cardiovascular Imaging and the American Society of Echocardiography . J Am Soc Echocardiogr 2017 ; 30 : 372 – 392 . Google Scholar CrossRef Search ADS PubMed 13 Chin CW , Khaw HJ , Luo E , Tan S , White AC , Newby DE , Dweck MR. Echocardiography underestimates stroke volume and aortic valve area: implications for patients with small-area low-gradient aortic stenosis . Can J Cardiol 2014 ; 30 : 1064 – 1072 . Google Scholar CrossRef Search ADS PubMed 14 Chirinos JA , Rietzschel ER , De Buyzere ML , De Bacquer D , Gillebert TC , Gupta AK , Segers P. Arterial load and ventricular-arterial coupling: physiologic relations with body size and effect of obesity . Hypertension 2009 ; 54 : 558 – 566 . Google Scholar CrossRef Search ADS PubMed 15 Andrén B , Lind L , Hedenstierna G , Lithell H. Left ventricular hypertrophy and geometry in a population sample of elderly males . Eur Heart J 1996 ; 17 : 1800 – 1807 . Google Scholar CrossRef Search ADS PubMed 16 Tribouilloy C , Rusinaru D , Maréchaux S , Castel AL , Debry N , Maizel J , Mentaverri R , Kamel S , Slama M , Lévy F. Low-gradient, low-flow severe aortic stenosis with preserved left ventricular ejection fraction: characteristics, outcome, and implications for surgery . J Am Coll Cardiol 2015 ; 65 : 55 – 66 . Google Scholar CrossRef Search ADS PubMed 17 Quiñones MA , Otto CM , Stoddard M , Waggoner A , Zoghbi WA. Recommendations for quantification of Doppler echocardiography: a report from the Doppler Quantification Task Force of the Nomenclature and Standards Committee of the American Society of Echocardiography . J Am Soc Echocardiogr 2002 ; 15 : 167 – 184 . Google Scholar CrossRef Search ADS PubMed 18 Oh JK , Taliercio CP , Holmes DR Jr , Reeder GS , Bailey KR , Seward JB , Tajik AJ. Prediction of the severity of aortic stenosis by Doppler aortic valve area determination: prospective Doppler-catheterization correlation in 100 patients . J Am Coll Cardiol 1988 ; 11 : 1227 – 1234 . Google Scholar CrossRef Search ADS PubMed 19 Eleid MF , Sorajja P , Michelena HI , Malouf JF , Scott CG , Pellikka PA. Survival by stroke volume index in patients with low-gradient normal EF severe aortic stenosis . Heart 2015 ; 101 : 23 – 29 . Google Scholar CrossRef Search ADS PubMed 20 Herrmann HC , Pibarot P , Hueter I , Gertz ZM , Stewart WJ , Kapadia S , Tuzcu EM , Babaliaros V , Thourani V , Szeto WY , Bavaria JE , Kodali S , Hahn RT , Williams M , Miller DC , Douglas PS , Leon MB. Predictors of mortality and outcomes of therapy in low-flow severe aortic stenosis: a Placement of Aortic Transcatheter Valves (PARTNER) trial analysis . Circulation 2013 ; 127 : 2316 – 2326 . Google Scholar CrossRef Search ADS PubMed 21 Anjan VY , Herrmann HC , Pibarot P , Stewart WJ , Kapadia S , Tuzcu EM , Babaliaros V , Thourani VH , Szeto WY , Bavaria JE , Kodali S , Hahn RT , Williams M , Miller DC , Douglas PS , Leon MB. Evaluation of flow after transcatheter aortic valve replacement in patients with low-flow aortic stenosis: a secondary analysis of the PARTNER randomized clinical trial . JAMA Cardiol 2016 ; 1 : 584 – 592 . Google Scholar CrossRef Search ADS PubMed 22 Maes F , Pierard S , de Meester C , Boulif J , Amzulescu M , Vancraeynest D , Pouleur AC , Pasquet A , Gerber B , Vanoverschelde JL. Impact of left ventricular outflow tract ellipticity on the grading of aortic stenosis in patients with normal ejection fraction . J Cardiovasc Magn Reson 2017 ; 19 : 37. Google Scholar CrossRef Search ADS PubMed Published on behalf of the European Society of Cardiology. All rights reserved. © The Author(s) 2018. For permissions, please email: journals.permissions@oup.com. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices)

Journal

European Heart JournalOxford University Press

Published: Mar 13, 2018

There are no references for this article.

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off