Background
Low-flow low-gradient aortic stenosis (AS) is a predictor of worse outcome compared with normal-flow AS. Although depressed left ventricular ejection fraction (LVEF) is associated with low flow, there is less evidence to support the role of other indices of cardiac structure and function.
Methods
Clinical and echocardiographic data from patients with native AS and valve areas ≤ 1.0 cm 2 were retrospectively analyzed to identify characteristics that are associated with low-flow low-gradient AS.
Results
In total, 941 patients were included. On multivariate analysis, factors independently associated with low flow (stroke volume index < 35 mL/m 2 ) included worse right ventricular systolic function, atrial fibrillation, lower LVEF, and higher left ventricular mass, with moderate or severe mitral regurgitation independently associated with low flow in the 694 patients (74%) with preserved LVEFs.
Conclusions
Right ventricular dysfunction and atrial fibrillation are independently associated with low-flow low-gradient AS, while moderate or severe MR is independently associated with low flow in patients with preserved LVEF. These associations with low flow in AS are clinically important to recognize, to avoid underestimation of AS severity.
Highlights
- •
The presence of depressed RV function, AF, higher LV mass, and lower LVEF is independently associated with low flow in aortic valve stenosis.
- •
Moderate or severe mitral regurgitation was independently associated with low flow in patients with AS and preserved LVEFs.
- •
These associations with low flow are clinically important to recognize, to avoid underestimation of AS severity.
Low-flow low-gradient severe aortic stenosis (LFLGAS) is defined as aortic valve area (AVA) ≤ 1.0 cm 2 with resting mean gradient < 40 mm Hg or peak velocity < 4.0 m/sec. LFLGAS is characterized by low stroke volume index (SVi) (<35 mL/m 2 ), and has been reported in 37% of patients of severe aortic stenosis (AS) with reduced left ventricular (LV) ejection fractions (LVEFs)—classical LFLGAS —and in up to 35% of patients with preserved LVEFs (so-called paradoxical LFLGAS). Identifying patients with LFLGAS is clinically important, because patients who fall into this group have worse outcomes than patients with normal flow. Although various abnormalities in cardiac structure or function, such as right ventricular (RV) dysfunction, mitral regurgitation, and atrial fibrillation (AF), can lead to low SVi, there are few published data pertaining to the relationship between these abnormalities and LFLGAS after taking LVEF into account. Therefore, the objective of this study was to determine which indices of cardiac structure and function are associated with LFLGAS independently of LVEF.
Methods
The institutional research ethics board approved this study. We retrospectively identified all patients with native valve AS referred for transthoracic echocardiography at our institution from 2007 to 2013. Only patients with AVAs ≤ 1.0 cm 2 were included in the analysis. The echocardiogram used for analysis was the first echocardiogram that showed native AVA ≤ 1.0 cm 2 ; any repeat echocardiograms in a given patient were not included in this analysis. The following were exclusion criteria: age < 18 years, prosthetic aortic valves, pregnancy at the time of imaging, supra- or subvalvular AS, and other complex congenital heart disease. Of a total of 1,800 patients with AS identified in the institutional echocardiographic database, 941 patients had AVAs ≤ 1.0 cm 2 and met eligibility criteria. The subsequent analysis was based on these 941 individuals.
Clinical Data
Clinical data were obtained from the electronic medical record. Data recorded included age, sex, history of coronary artery disease (defined as >50% stenosis in at least one major epicardial coronary vessel on angiography or functional evidence of ischemia), myocardial infarction, prior coronary artery bypass grafting, dyslipidemia (defined as low-density lipoprotein ≥ 3 mmol/L or taking lipid-lowering medication), hypertension (by history or taking antihypertensive medication), diabetes mellitus (by history or taking diabetic medication), body mass index, and body surface area (BSA), which was estimated according to the method of DuBois and DuBois as 0.007184 × height (cm) 0.725 × weight (kg) 0.425 .
Echocardiographic Analysis
All echocardiograms at our institution are archived using Image Vault Data system and analyzed using EchoPAC (both from GE Healthcare, Milwaukee, WI) by level III–trained echocardiologists according to standardized laboratory protocols. Echocardiographic data were obtained from final echocardiographic reports. All echocardiographic and Doppler measurements and calculations were performed according to the most recent American Society of Echocardiography guidelines available at the time of reporting.
The following measurements were recorded: LV dimensions, mass, relative wall thickness, and diastolic function; LVEF; left atrial volume; RV size and function (including tricuspid annular plane systolic excursion [TAPSE], as well as visual assessment); valvular stenosis and regurgitation; heart rate, rhythm, and blood pressure at the time of echocardiography; cardiac output; and stroke volume. LV dimensions and mass were indexed to BSA. For purposes of analysis, we divided all valve disease (other than AS) in a binary fashion into less than moderate and moderate or greater, because this level of valve dysfunction was considered hemodynamically significant insofar as effects on SVi were concerned. Similarly, we dichotomized diastolic function into normal or mild (grade I) and moderate or worse (grades II–IV). Stroke volume was measured from pulsed-wave Doppler in the LV outflow tract in the apical five-chamber view and was indexed to BSA. Cardiac output was calculated by multiplying the LV outflow tract pulsed-wave time-velocity integral by heart rate. Systemic arterial compliance was calculated as the ratio of the stroke volume to the arterial pulse pressure. Valvuloarterial impedance ( Z va ), a measure of global LV afterload, was calculated as follows : Z va = (systolic blood pressure + mean aortic valve gradient)/SVi. Z va was excluded from multivariate regression because by definition, it is highly correlated with SVi.
Statistical Analysis
Patients were divided into two groups on the basis of SVi: preserved SVi (≥35 mL/m 2 ) and low SVi (<35 mL/m 2 ). Groups were compared using the Fisher exact test for categorical variables, the Kruskal-Wallis for non–normally distributed continuous variables, and analysis of variance for normally distributed continuous variables. Echocardiographic parameters associated with low SVi were identified using logistic regression. Covariates achieving P values < .20 at the univariate level were included in a multivariate regression, which was performed by stepwise backward elimination. Given that elevated heart rate and low blood pressure may in certain instances be consequences, rather than causes, of low flow, that BSA is the denominator in the SVi calculation, and that male sex should not physiologically explain low transvalvular flow, we then repeated the multivariate regression excluding these four variables. We also performed a sensitivity analysis in which SVi was modeled as a continuous variable and in which continuous covariates were standardized to have a mean value of 0 and SD of 1. We performed another sensitivity analysis that included relative wall thickness as a standardized covariate. Overall α was set at 0.05 to determine statistical significance. The same analysis was then performed stratifying by LVEF (≥50% vs <50%). All statistical analyses were performed with Stata version 13.1 (StataCorp LP, College Station, TX).
Results
Of the 941 patients included, 498 (53%) had preserved SVi and 443 (47%) had low SVi. Patient characteristics are displayed in Table 1 . The median LVEF in the low-SVi group was 52% (interquartile range, 38%–58%), and the median LVEF in the preserved-SVi group was 59% (interquartile range, 55%–63%) ( Table 1 ).
| Whole cohort | LVEF < 50% | LVEF ≥ 50% | |||||||
|---|---|---|---|---|---|---|---|---|---|
| Low | Preserved | P | Low | Preserved | P | Low | Preserved | P | |
| SVi < 35 mL/m 2 ( n = 443) | SVi ≥ 35 mL/m 2 ( n = 498) | SVi < 35 mL/m 2 ( n = 185) | SVi ≥ 35 mL/m 2 ( n = 59) | SVi < 35 mL/m 2 ( n = 256) | SVi ≥ 35 mL/m 2 ( n = 438) | ||||
| Age (y) | 81 (71–87) | 79 (71–86) | .039 | 82 (74–87) | 83 (75–87) | NS | 81 (70–87) | 79 (70–86) | NS |
| Women | 195 (44%) | 288 (58%) | <.001 | 73 (39%) | 34 (58%) | .016 | 122 (48%) | 254 (58%) | .009 |
| BSA (m 2 ) | 1.90 ± 0.272 | 1.81 ± 0.246 | <.001 | 1.89 ± 0.255 | 1.80 ± 0.262 | .030 | 1.90 ± 0.284 | 1.81 ± 0.244 | <.001 |
| Previous CABG | 106 (25%) | 90 (19%) | .024 | 60 (34%) | 13 (22%) | NS | 46 (19%) | 77 (19%) | NS |
| Previous myocardial infarction | 149 (38%) | 100 (22%) | <.001 | 96 (57%) | 22 (42%) | NS | 53 (23%) | 78 (20%) | NS |
| Coronary artery disease | 249 (60%) | 231 (50%) | .002 | 130 (74%) | 38 (66%) | NS | 118 (49%) | 193 (47%) | NS |
| Hypertension | 349 (84%) | 371 (79%) | NS | 141 (82%) | 53 (91%) | NS | 206 (85%) | 317 (78%) | .025 |
| Diabetes | 159 (38%) | 120 (26%) | <.001 | 78 (44%) | 13 (23%) | .005 | 80 (33%) | 106 (26%) | NS |
| Dyslipidemia | 289 (72%) | 310 (68%) | NS | 131 (79%) | 43 (78%) | NS | 157 (67%) | 266 (67%) | NS |
| Rhythm | |||||||||
| Sinus | 228 (51%) | 386 (78%) | <.001 | 99 (58%) | 46 (78%) | .017 | 160 (65%) | 383 (89%) | <.001 |
| AF | 144 (33%) | 49 (10%) | 60 (34%) | 10 (17%) | 82 (33%) | 38 (9%) | |||
| Other (including paced rhythm) | 71 (16%) | 63 (12%) | 13 (8%) | 3 (5%) | 6 (2%) | 10 (2%) | |||
| Systolic blood pressure (mm Hg) | 128 ± 21 | 138 ± 22 | <.001 | 123 ± 23 | 131 ± 22 | NS | 131 ± 20 | 138 ± 22 | <.001 |
| Diastolic blood pressure (mm Hg) | 71 ± 12 | 71 ± 11 | NS | 70 ± 12 | 65 ± 14 | .044 | 72 ± 12 | 72 ± 11 | NS |
| Diastolic function | |||||||||
| Normal/mildly impaired | 137 (56%) | 232 (66%) | .010 | 35 (33%) | 14 (38%) | NS | 102 (73%) | 218 (70%) | NS |
| Moderately impaired or worse | 108 (44%) | 118 (34%) | 71 (67%) | 23 (62%) | 37 (27%) | 95 (30%) | |||
| Heart rate (beats/min) | 75 (64–87) | 65 (59–76) | <.001 | 77 (64–88) | 65 (60–76) | <.001 | 74 (65–85) | 65 (59–76) | <.0001 |
| Z va (mm Hg/mL/m 2 ) | 5.69 (4.96–6.44) | 4.07 (3.63–4.65) | <.001 | 5.84 (4.99–6.83) | 3.95 (3.54–4.60) | <.001 | 5.63 (4.87–6.30) | 4.09 (3.65–4.67) | <.001 |
| Systemic arterial compliance (mL/min/mm Hg) | 0.952 (0.735–1.19) | 1.23 (0.958–1.58) | <.001 | 0.942 (0.692–1.18) | 1.23 (0.968–1.46) | <.001 | 0.853 (0.768–1.24) | 1.23 (0.958–1.59) | <.001 |
| SVi (mL/m 2 ) | 28 ± 5.1 | 44 ± 6.4 | <.001 | 26 ± 5.5 | 42 ± 5.4 | <.001 | 30 ± 4.2 | 44 ± 6.5 | <.001 |
| Cardiac output (L/min) | 3.9 (3.2–4.8) | 5.2 (4.4–5.9) | <.001 | 3.6 (3.0–4.5) | 4.8 (4.4–5.4) | <.001 | 4.1 (3.4–5.0) | 5.2 (4.4–6.0) | <.001 |
| Aortic valve mean gradient (mm Hg) | 32 ± 17 | 41 ± 15 | <.001 | 28 ± 15 | 37 ± 14 | <.001 | 35 ± 17 | 42 ± 15 | <.001 |
| LVEF (%) | 53 (38–57) | 59 (55–63) | <.001 | 38 (28–43) | 43 (38–48) | <.001 | 58 (54–63) | 61 (58–64) | <.001 |
| LVIDD (cm) | 4.8 ± 0.89 | 4.5 ± 0.69 | <.001 | 5.3 ± 1.0 | 5.1 ± 0.85 | NS | 4.5 ± 0.64 | 4.4 ± 0.64 | NS |
| LVISD (cm) | 3.6 ± 1.0 | 3.0 ± 0.74 | <.001 | 4.2 ± 1.1 | 3.9 ± 0.77 | NS | 3.1 ± 0.61 | 2.9 ± 0.65 | .0011 |
| LV mass (g) | 199 ± 69 | 182 ± 60 | <.001 | 216 ± 71 | 222 ± 78 | NS | 186 ± 64 | 176 ± 56 | NS |
| RVIDD (cm) | 3.6 ± 0.80 | 3.3 ± 0.68 | <.001 | 3.9 ± 0.84 | 3.7 ± 0.82 | NS | 3.4 ± 0.71 | 3.3 ± 0.64 | .023 |
| RV function (semiquantitative) | |||||||||
| Normal or mildly impaired | 375 (86%) | 481 (98%) | <.001 | 135 (75%) | 51 (89%) | .017 | 238 (95%) | 429 (99%) | .002 |
| Moderate or severely impaired | 59 (14%) | 11 (2%) | 46 (25%) | 6 (11%) | 13 (5%) | 5 (1%) | |||
| TAPSE (cm) | 1.8 (0.50) | 2.1 (0.47) | <.001 | 1.7 (0.47) | 1.8 (0.42) | NS | 1.9 (0.50) | 2.1 (0.46) | <.001 |
| LAESV (mL) | 82 ± 35 | 72 ± 31 | <.001 | 90 ± 31 | 85 ± 31 | NS | 76 ± 36 | 71 ± 30 | .043 |
| RVSP (mm Hg) | 41 (33–52) | 37 (30–44) | <.001 | 45 (38–55) | 46 (37–53) | NS | 37 (30–43) | 35 (30–43) | NS |
| Mitral regurgitation | |||||||||
| Less than moderate | 390 (88%) | 465 (94%) | .006 | 158 (85%) | 51 (86%) | NS | 230 (90%) | 413 (95%) | .045 |
| Moderate or greater | 52 (12%) | 32 (6%) | 27 (15%) | 8 (14%) | 25 (10%) | 24 (5%) | |||
| Tricuspid regurgitation | |||||||||
| Less than moderate | 385 (87%) | 459 (93%) | .004 | 150 (81%) | 45 (76%) | NS | 233 (91%) | 413 (95%) | NS |
| Moderate or greater | 58 (13%) | 37 (7%) | 35 (19%) | 14 (24%) | 23 (9%) | 23 (5%) | |||
| Aortic regurgitation | |||||||||
| Less than moderate | 408 (93%) | 427 (86%) | .001 | 179 (97%) | 44 (76%) | <.001 | 227 (90%) | 382 (88%) | NS |
| Moderate or greater | 30 (7%) | 68 (14%) | 6 (3%) | 14 (24%) | 24 (10%) | 54 (12%) | |||
| Mitral stenosis | |||||||||
| Less than moderate | 422 (97%) | 457 (94%) | .019 | 180 (99%) | 56 (97%) | NS | 240 (96%) | 400 (94%) | NS |
| Moderate or greater | 12 (3%) | 29 (6%) | 2 (1%) | 2 (3%) | 10 (4%) | 27 (6%) | |||
Factors Associated with Low Flow
On univariate analysis, factors associated with low SVi included older age, male sex, diabetes mellitus, and known coronary artery disease ( Table 1 ). Echocardiographic indices associated with low flow included AF, higher valvuloarterial impedance, higher LV mass, eccentric (rather than concentric) LV remodeling, larger LV dimensions, lower LVEF, larger RV size, worse RV function, and more mitral regurgitation and tricuspid regurgitation, but less aortic insufficiency ( Table 2 ). An example of a patient with low-flow AS is seen in Figure 1 .
