Value of low-dose dobutamine stress echocardiography on defining true severe low gradient aortic stenosis in patients with preserved left ventricular ejection fraction

Low-dose dobutamine stress echocardiography (DSE) is a valuable tool to distinguish true-severe (TS) from pseudo-severe (PS) low gradient aortic valve stenosis (LGAS) in patients with reduced left ventricular ejection fraction (LVEF). However, only scanty studies reported the clinical utility of DSE in differentiating TS-LGAS patients with preserved LVEF. We investigated the clinical utility of DSE in LGAS patients with preserved LVEF and the echocardiographic determinants suggestive of TS-LGAS. 130 consecutive LGAS patients [indexed aortic valve area (AVA) ≤ 0.6cm2/m2 and mean trans-aortic pressure gradient (PGmean) < 40mmHg] with preserved (≥ 50%, n = 63) and reduced (< 50%, n = 67) LVEF were included. DSE defined TS-LGAS (projected AVA ≤ 1 cm2) in 61.2% patients with reduced LVEF and in 68.3% patients with preserved LVEF. Multivariate logistic regression analysis showed that baseline AVA was an independent determinant of TS-LGAS both in LVEF ≥ 50% (OR 0.45, P = 0.004) and LVEF < 50% groups (OR 0.55, P = 0.005). Reduced septal and lateral mitral annular plane systolic excursion (MAPSE, OR 0.72 and 0.75, P = 0.013 and 0.016) and septal TDI-s´ were significantly associated with TS-LGAS in patients with LVEF ≥ 50%. Higher systolic pulmonary artery pressure (SPAP, OR 1.43, P = 0.045) was associated with TS-LGAS in patients with LVEF < 50%. DSE is useful to define TS-LGAS also in patients with preserved LVEF. Lower baseline AVA values are linked with TS-LGAS in both patients with reduced and preserved LVEF. Reduced MAPSE and septal TDI-s´ are suggestive of TS-LGAS in patients with preserved LVEF, while higher SPAP is associated with TS-LGAS in patients with reduced LVEF. Electronic supplementary material The online version of this article (10.1007/s10554-018-1416-z) contains supplementary material, which is available to authorized users.


Introduction
Severe aortic stenosis (AS) is usually defined as an aortic valve area (AVA) < 1.0cm 2 , mean transvalvular gradient (PG mean ) ≥ 40 mmHg, and a peak flow velocity ≥ 4.0 m/s [1]. Patients with AVA < 1.0cm 2 and PG mean <40 mmHg are usually defined as low-gradient AS (LGAS). Further examinations are usually required to distinguish true-severe (TS) from pseudo-severe (PS) AS, which is clinically essential for therapeutic decision making because patients with TS-LGAS might benefit, whereas PS-LGAS patients might not benefit from aortic valve replacement (AVR) [2]. LGAS is usually found in patients with reduced LV ejection fraction (LVEF < 50%), but might also be present in patients with preserved LVEF ≥ 50% (so called paradoxical LGAS) [3]. Low-dose dobutamine stress echocardiography (DSE) is Dan Liu and Kai Hu have contributed equally to this work.

Electronic supplementary material
The online version of this article (https ://doi.org/10.1007/s1055 4-018-1416-z) contains supplementary material, which is available to authorized users. 1 3 recommended as a tool to define the severity of aortic stenosis in patients with LGAS and reduced LVEF [4,5]. To date, data regarding the utility of DSE for defining severity of LGAS in patients with preserved LVEF presenting with either paradoxical low flow or normal flow remain scanty. In the present study, we observed if DSE could also be useful to define AS severity in LGAS patients with preserved LVEF. We further explored the conventional echocardiographic markers suggestive of TS-LGAS in LGAS patients with reduced or preserved LVEF.

Study population
A total of 130 consecutive symptomatic low-gradient AS patients (aged 78 ± 8 years, 63.8% male), referred to the University Hospital Würzburg between January 2011 and December 2016, were included in this study. All patients underwent both standard transthoracic echocardiogram and DSE. Enrollment criteria included indexed aortic valve area (AVAi) ≤ 0.6 cm 2 /m 2 and mean trans-aortic pressure gradient (PG mean ) < 40 mmHg as assessed by transthoracic echocardiogram. The study protocol is shown in Fig. 1. The study was conducted in accordance to the Declaration of Helsinki and was approved by the Local Ethics Committee at the University of Würzburg (AZ 11/03 and 60/14). Informed consent was obtained from all patients or their guardians.

Echocardiography
Echocardiographic examination was performed with GE Vingmed Vivid 7 or E9, Horten, Norway. Standard measurements on cardiac structural and functional parameters were made according to the current ASE guideline [6]. LV mass indexed to body surface area (LVMi) was calculated by the suggested formula [7]. End-diastolic and end-systolic volumes and LVEF were measured with the biplane Simpson method in the apical 4-and 2-chamber views. Septal and lateral mitral annular plane systolic excursion (MAPSE) and tricuspid annular plane systolic excursion (TAPSE) were measured by M-mode in the apical 4-chamber view. Systolic pulmonary artery pressure (SPAP) was derived from peak tricuspid regurgitation jet velocity using the simplified Bernoulli equation, in combination with the estimated right atrial pressure. LV diastolic function was assessed according to ASE guidelines for the assessment of diastolic function [8]. Tissue Doppler imaging was performed at the septal and lateral mitral annular sites enabling on-line derivation of myocardial systolic velocity (TDI-s′) and diastolic early velocity (e′) [9]. Speckle tracking imaging analysis was performed using EchoPAC software (GE, Horten, Norway) as described previously [10]. Longitudinal peak systolic strain (LS) of each segment was measured and global LS values were acquired by averaging strain rate and strain values of all 18 segments.

Evaluation of AS severity by standard echocardiography
The diagnosis and classification of AS was made according to recent EAE/ASE recommendations [11]. Left ventricular outflow tract (LVOT) diameter was measured in the parasternal long-axis view focusing on the LVOT at baseline, and this value was also used to calculate the aortic valve area during DSE. The outer edge of the velocity spectrum obtained by continuous wave (CW) Doppler across the aortic valve (AV) was traced to obtain the maximum velocity (V max ), maximum and mean trans-aortic pressure gradient (PG mean and PG max ), and AV velocity time integral (VTI). The subvalvular V max (LVOT V max ), VTI and stroke volume (SV) were obtained by tracing the outer edge of the velocity spectrum at the LVOT by the pulsed wave (PW) Doppler. Aortic valve area by continuity equation VTI [AVA (VTI)] was automatically calculated using the formula: LGAS low gradient aortic stenosis, AVAi indexed aortic valve area, PG mean mean trans-aortic pressure gradient, LVEF left ventricular ejection fraction, TS truesevere, PS pseudo-severe 1 3 Valvuloarterial impedance (Zva) was calculated using the formula:

Low-dose dobutamine stress echocardiography
Dobutamine infusion was started at a dose of 5 µg/kg/min, then increasing to 10, 15, and 20 µg/kg/min at 3-min intervals. Blood pressure and heart rate were recorded at rest and at the end of each stage of dobutamine dosage. Contractile reserve was defined as an increase in stroke volume (SV) of 20% or more [12]. The projected aortic valve area at normal transvalvular flow rate (AVA proj ) was calculated using a simplified method with the following formula [13]: Simplified AVA proj = AVA rest + VC simpl × (250-Q rest ), where VC simpl is the valve compliance and Q was computed by dividing SV by LV ejection time. VC simpl was computed with the formula: where AVA rest and AVA peak are AVA at rest and peak DSE. Q peak and Q rest are Q at rest and peak DSE. TS-LGAS was defined as AVA proj ≤1 cm 2 .

Follow-up
Patients were followed up for a mean of 18 (12-27) months by clinical visit or telephone interview.

Data analysis
Continuous variables are presented as mean ± standard deviation or median (interquartile range, IQR), as appropriate. Normal distribution of all continuous variables was tested by inspecting skewness, kurtosis, and Q-Q plots. Clinical and echocardiographic continuous variables between preserved and reduced LVEF groups and between PS-LGAS and TS-LGAS groups were compared using independent Student´s t test or Mann-Whitney U test as indicated. Categorical variables were expressed as percentages and were compared using a Chi square test or Fisher's exact test, as appropriate. Multivariate logistic regression analysis was conducted to identify markers for differentiating TS-LGAS from PS-LGAS. Odds ratios (OR) with 95% confidence interval (CI) were assessed for indicating diagnostic performance. A twotailed probability value < 0.05 was considered significant. Statistical analysis was performed using IBM SPSS, version 23 for Windows (IBM Corp., New York, USA). Zva mmHg∕ml∕m 2 = systolic blood pressure + PG mean {body surface area indexed SV}.

Baseline echocardiographic characteristics
Mean LVEF was 60 ± 6% in the LVEF ≥ 50% group and 46 ± 4% in the LVEF < 50% group (Table 2). Patients with LVEF < 50% had larger LV, RV and LA cavities, thicker LV walls, lower septal and lateral MAPSE and TAPSE as compared to patients with LVEF ≥ 50% (all P < 0.05). AV Vmax, PG mean , LVOT V max , AV velocity ratio, and SV were significantly lower, while Zva was significantly higher in the LVEF < 50% group than in the LVEF ≥ 50% group (all P < 0.05). AVA VTI was similar between the LVEF LVEF ≥ 50% and < 50% groups (0.86 ± 0.14 vs. 0.84 ± 0.16cm 2 , P = 0.538). LV longitudinal systolic function parameters including septal and lateral TDI-s´as well as global and regional longitudinal strain (LS) at the basal septum were significantly lower in the LVEF < 50% group than that in the LVEF ≥ 50% group.

DSE
DSE results in the patients with LVEF ≥ 50% and LVEF < 50% are shown in Tables 3 and 4. Forty-three out of 63 (68.3%) in the LVEF ≥ 50% group and 41 out of 67 (61.2%) patients in the LVEF < 50% group were diagnosed as TS-LGAS by DSE. AV Vmax, PG mean , LVOT V max , AV velocity ratio, and AVA VTI significantly increased during DSE both in LVEF ≥ 50% and < 50% groups (all P < 0.05, Tables 3, 4). Systolic blood pressure remained unchanged while peak diastolic blood pressure decreased in both groups during DSE.

Discussion
The major findings of the present study are: (1) DSE appears to also be a helpful tool of defining true-severe low gradient aortic stenosis in patients with LVEF ≥ 50%; (2) Baseline AVA derived from transthoracic echocardiography is an independent determinant of TS-LGAS in both patients with LVEF ≥ 50% as well as LVEF < 50%; (3) Reduced septal and lateral MAPSE as well as septal TDI-s´ are associated with TS-LGAS in patients with LVEF ≥ 50% while higher SPAP and lower AV velocity ratio are independently associated with TS-LGAS in patients with LVEF < 50%.

Value of DSE in LGAS patients with reduced and preserved LVEF
Patients with low gradient severe aortic stenosis (LGAS) and preserved LVEF (AVA <1 cm 2 , PG mean <40 mmHg, LVEF >50%) are increasingly recognised in clinical practice. DSE is recommended to define the severity of aortic stenosis in patients with LGAS and reduced LVEF [4,5].
Our results showed that DES is also useful in defining TSfrom PS-LGAS in patients with LVEF < 50% (Table 4). This finding is in line with a recent study, which demonstrated that DSE might serve as a valuable tool to distinguish

TS-LGAS from PS-LGAS in patients with paradoxical
LGAS [14]. There were 20 out of 63 (31.7%) LGAS patients with preserved LVEF presenting with paradoxical low-flow LGAS (SVi ≤ 35 ml/m 2 ) in our cohort. We compared other related parameters as well as the response on DSE between the two subgroups (low-flow vs. normal-flow LGAS). The data show that LV cavity in these patients was smaller than in patients with normal-flow LGAS (left ventricular end-diastolic dimension 43 ± 4 vs. 46 ± 5 mm, P = 0.021). Moreover, DSE seems also to be feasible to identify the TS-AS patients in the low-flow LGAS subgroup as in the normal-flow LGAS subgroup (Supplementary Table 1). Future studies with larger patient cohort are warrant to verify these results.

Echocardiographic determinant of TS-LGAS in patients with preserved LVEF
In line with previous findings from our group [15] and others [16], the present study results demonstrate that reduced MAPSE is suggestive of TS-LGAS in patients with LVEF > 50%. A previous study also showed that MAPSE correlated with aortic valve area (Spearman r = 0.18, P = 0.02) in 205 asymptomatic AS patients with LVEF ≥ 50% [17]. In another study, Rydberg and colleagues found that left atrioventricular plane displacement (MAPSE), but not left ventricular ejection fraction, was influenced by the degree of aortic stenosis [16]. In patients with LGAS and preserved LVEF, LVEF is maintained at normal range at the expense of high LV end-diastolic pressure and volume. In this circumference, LV cannot recruit additional preload because the maximum sarcomere length is already reached and the hypertrophied stiff ventricle could not tolerate additional filling load [18]. This might explain why reduced MAPSE, as a function of LV hemodynamic load, was revealed as the most sensitive marker of TS-LGAS in LGAS patients with LVEF > 50%. Additionally, reduction in TDI-s´, another parameter reflecting LV longitudinal dysfunction, is also suggestive of TS-LGAS in AS patients with LVEF ≥ 50%.

Echocardiographic determinants of TS-LGAS in patients with reduced LVEF
In LGAS patients with reduced LVEF, the disease features include both AS and heart failure. Ventricular remodeling (myocyte hypertrophy and myocardial fibrosis), systolic and diastolic dysfunction, in the setting of valvular stenosis and heart failure scenarios, contributes to the progressively elevated LV filling pressure, and/or left atrial pressure, which in turn would transmit to the pulmonary vasculature, inducing pulmonary venous congestion and pulmonary hypertension, so called WHO Group II -pulmonary hypertension [19,20].
A previous study showed that pulmonary hypertension was presented in the majority of patients with severe aortic stenosis, and ejection fraction serves as one of the correlates of pulmonary hypertension in patients with severe aortic stenosis [21]. Another study indicated that both smaller aortic valve area and lower LVEF were responsible for the presence of pulmonary hypertension in a cohort of 626 patients with severe aortic stenosis [22]. In line with above findings, we found that increased systolic pulmonary pressure is an independent determinant of TS in LGAS patients with reduced LVEF. Another finding of the present study is that lower AV velocity ratio is related to the presence of true severe in LGAS patients with reduced LVEF. This parameter is only recommended in the EAE/ASE recommendations, a value of < 0.25 is suggestive of severe AS [23]. In our cohort, this value ranged from 0.23 at rest and 0.24 during DSE in TS-LGAS patients with LVEF > 50%, and 0.19 at rest and 0.19 during DSE in TS-LGAS patients with reduced LVEF. A previous study reported that velocity ratios could not only define the severity of AS, but also predict outcomes in patients with TS-LGAS and preserved LVEF [24]. Our results suggest AV velocity ratio could be used as an independent predictor for defining TS in LGAS patients with reduced LVEF.

Limitations
This retrospective study was performed monocentric and the number of patients included in the sub-cohorts was relatively small. The results of the present study might thus be affected by patient selection bias. Future studies with large patient cohort are warranted to validate the results obtained from this study. In this study, projected aortic valve area assessed by DSE was used to define the severity of AS [25]. Other imaging modalities may aid in the diagnosis of severe AS. Macroscopic evaluation of the valve at the time of valve replacement or measurement of aortic valve calcium score by multislice computed tomography could provide additional accuracy to differ TS-from PS-AS [26]. Due to the lack of available CT data for AV calcium score in our cohort, the severity of AS was only determined by projected aortic valve area. This study limitation should be considered in the interpretation of results derived from the current study.

Clinical implications
Our study results suggest that DSE enables evaluation of the severity of AS both in LGAS patients with reduced LVEF and in those with preserved LVEF, which is consistent with what has been demonstrated by previous studies [13,26]. In the present study, we further explored the echocardiographic determinants suggestive of TS-LGAS. Reduced MAPSE and TDI-s´ are suggestive of TS in LGAS patients with preserved LVEF, while increased SPAP and lower AV velocity ratio are suggestive of TS in LGAS patients with reduced LVEF. These findings provide incremental information on the diagnosis and therapy decision for LGAS patients, particularly in those patients who cannot tolerate DSE due to the presence of contraindications. Patients with related echocardiographic features should be carefully evaluated for the AS severity with alternative complementary imaging modalities, such as measurement of aortic valve calcium score assessed by multislice computed tomography.

Conclusions
DSE is valuable for staging of LGAS in patients with both reduced and preserved LVEF. Low longitudinal LV function (MAPSE) and septal TDI-s´ are associated with TS-LGAS in patients with preserved LVEF, and high trans-tricuspid pressure gradient (SPAP) and low AV velocity ratio are associated with TS-LGAS in patients with reduced LVEF.