Background
The aim of the present study was to assess and compare the disease progression of aortic stenosis (AS) subtypes from nonsevere to severe disease on the basis of measures of gradient and flow.
Methods
Seventy-seven patients with AS (mean aortic valve area, 1.3 ± 0.3 cm 2 at baseline) underwent echocardiographic examination, including two-dimensional speckle-tracking strain measurements. Patients were retrospectively grouped according to mean transvalvular pressure gradient (40 mm Hg) into low-gradient (LG/AS) and high-gradient (HG/AS) groups. The LG/AS group was further subdivided into low-flow (LF/LG; i.e., stroke volume index < 35 mL/m 2 ) and normal-flow (NF/LG) groups. For subanalysis, the LF/LG group was split into two groups: “paradoxical” (P-LF/LG; ejection fraction > 50%) and “classical” LF/LG (C-LF/LG; ejection fraction < 50%). Follow-up echocardiography was performed in patients with severe AS after 3.3 ± 1.7 years. Survival status was ascertained after 5.0 ± 2.0 years.
Results
Coronary artery disease was more frequent in LG/AS than HG/AS patients. Already at baseline, LF/LG patients showed reduced left ventricular global systolic strain and reduced systemic arterial compliance compared with HG/AS patients (HG/AS, 1.0 ± 0.4 mL · mm Hg− 1 · m −2 ; NF/LG, 0.9 ± 0.2 mL · mm Hg− 1 · m −2 ; LF/LG, 0.6 ± 0.2 mL · mm Hg −1 · m −2 ; P < .001). The initially elevated valvuloarterial impedance increased significantly more in LG/AS than in the other groups (HG/AS, 2.2 ± 0.9 mm Hg · mL− 1 · m −2 ; NF/LG, 2.2 ± 0.5 mm Hg · mL− 1 · m −2 ; LF/LG, 3.2 ± 0.8 mm Hg · mL −1 · m− 2 ; P < .001), while aortic valve area decreased by 42% in HG/AS versus 34% in NF/LG and 32% in LF/LG ( P < .001). At follow-up, global systolic strain was significantly reduced in C-LF/LG (7.7 ± 2.5 vs 13.5 ± 2.9 in P-LF/LG, P < .001). In P-LF/LG, mitral E/E′ ratio increased significantly from 8.9 ± 4.0 to 26.4 ± 9.2 ( P < .05).
Conclusions
In patients with AS with high-gradient physiology, the valve constitutes the primary problem. By contrast, low-gradient AS is a systemic disease with valvular, vascular, and myocardial components, resulting in a slower progression of transvalvular gradient, but worse clinical outcome. In C-LF/LG, impaired systolic function leads to an LG flow pattern, whereas the pathophysiology in P-LF/LG is predominantly a diastolic dysfunction.
Highlights
- •
Adding insight on natural disease progression of low-gradient AS.
- •
Adding insight on hemodynamic characteristics of AS subgroups.
- •
Highlighting differences in subgroups of AS.
- •
Specific characteristics of subgroups will help clinicians in tailoring treatment.
- •
Raising awareness to detect the condition of low-gradient AS.
In the late 1980’s, aortic stenosis (AS) was commonly considered a unidirectional degenerative disease, limited solely to the valve itself. Since then, thanks to extensive research efforts, it is now widely agreed that a complex multifactorial pathophysiology underlies AS, involving both the myocardium and the vascular system. In 1980, Carabello et al . observed that among patients with AS undergoing aortic valve replacement, some failed to recover from compromised left ventricular (LV) function. These patients featured a particular profile; that is, they had severe AS, and their mean transvalvular pressure gradient (PG mean ) was low. This subgroup has been intensely studied in recent years, and the condition has now been termed low-gradient AS (LG/AS; PG mean < 40 mm Hg). Of note, further studies proved that when the LG/AS group was split into a low-flow group (LF/LG; indexed LV stroke volume [SVi] < 35 mL/m 2 ) and a normal-flow group (NF/LG), differences in mortality became apparent. Looking more closely into the LF/LG flow patterns, a subgroup with “paradoxically” preserved LV ejection fraction could be detected, raising questions about the pathophysiology of this subtype.
Carabello et al . wondered if it an underlying cardiomyopathy led to these remarkable differences in clinical outcomes, representing “two distinct groups, rather than opposite ends of a spectrum.” The vast majority of studies conducted so far have focused on severe AS or compared patients with moderate and severe AS. However, the course of disease progression, that is, from nonsevere to severe AS, in the aforementioned different subgroups of AS has not yet been specifically examined. We therefore hypothesized that distinct echocardiographic and clinical profiles are suited to indicate which patient will progress to LG/AS. In the present study, we addressed the following questions: (1) Can differences between AS subgroups already be detected in patients with nonsevere AS? and (2) Are there any differences in disease progression between the subgroups of AS allowing better tailoring of treatment and clinical decision making?
Methods
We analyzed data from patients who were referred to our clinic and listed in our AS registry between 2005 and 2013. Of 1,305 patients in the registry, 77 met the inclusion criterion of having documented progression from nonsevere (baseline) to severe AS by transthoracic echocardiographic examinations performed ≥6 months apart. Exclusion criteria were bicuspid AS, history of aortic valve replacement, and significant concomitant valvular disease at baseline ( Figure 1 ). Hypertension was presumed to be present when antihypertensive medication was prescribed or office blood pressure was >140/90 mm Hg. Significant coronary artery disease was presumed to be present when >50% stenosis in at least one coronary artery bed was recorded or a history of myocardial infarction, percutaneous intervention, or coronary artery bypass graft was documented. Obesity was defined as body mass index ≥ 30 kg/m 2 , hypercholesterolemia as total cholesterol > 200 mg/dL, and renal insufficiency as estimated glomerular filtration rate < 60 mL/min. European System for Cardiac Operative Risk Evaluation I score was assessed in all patients at echocardiographic follow-up using the online risk calculator. Clinical data at the time of echocardiography were recorded and documented for baseline and final echocardiography. All subjects gave informed consent, and approval of the local ethics committee was obtained.
A 3.5-MHz transducer was used for data acquisition, and data were stored on dedicated workstations (Vivid 7 and E9; GE Vingmed Ultrasound AS, Horten, Norway). These data were reanalyzed offline by experienced cardiologists after the routine examination and expanded by two-dimensional speckle-tracking strain analysis (EchoPAC version 112; GE Vingmed Ultrasound AS). For the echocardiographic examination, patients were in the supine left lateral position. The diameter of the LV outflow tract was obtained in B mode from the parasternal long-axis view in midsystole, measuring the distance between the ventricular base of the aortic cusps hinge points. Measurements of LV dimensions were performed at end-diastole at the height of the papillary muscles tips (ventricular and septal diameter, posterior wall thickness) in the parasternal long-axis view. These dimensions were integrated into the corrected formula for LV mass of the American Society of Echocardiography and indexed to body surface area. Left atrial diameter was measured anterior-posterior at end-systole in the parasternal long-axis view. By measuring diastolic mitral valve inflow at the height of the mitral valve tips in the apical four-chamber view, early (E) and atrial (A) wave patterns were documented. An E/A ratio < 1 was defined as impaired relaxation or first-degree diastolic impairment, an E/A ratio > 2 as restrictive filling pattern or third-degree diastolic impairment. Values between 1 and 2 were considered normal, if during a Valsalva maneuver the E/A ratio did not reach values <1, which was considered a pseudonormal filling pattern, or second-degree diastolic impairment. The tissue Doppler–derived value for the velocity of the septal and lateral portions of the mitral annulus in the apical four-chamber view were averaged and denoted E′. The E/E′ was used as an estimate of LV end-diastolic filling pressures. Using M-mode imaging, mitral annular plane systolic excursion (MAPSE) and tricuspid annular plane systolic excursion were measured. Systolic pulmonary artery pressure was estimated by conversion of measurements of tricuspid regurgitation jet via the Bernoulli equation and its summation with estimated central venous pressure by measuring the breathing variability of the inferior caval vein at the height of the liver veins’ insertion. Hemodynamic parameters were calculated as follows: the ratio of the velocity-time integral of both the LV outflow tract and the transaortic jet yields the dimensionless velocity index. Measurement of stroke volume in the LV outflow tract was performed using pulsed-wave Doppler at the same time as brachial blood pressure was obtained using an arm-cuff sphygmomanometer. Systemic arterial compliance (SAC) was calculated as the ratio of SVi and the difference between systolic and diastolic blood pressures. Multiplying mean arterial pressure by 80 and dividing this term by cardiac output computed systemic vascular resistance. The sum of systolic artery pressure and PG mean divided by SVi yields valvuloarterial impedance. The mean observation time from first to last echocardiographic examination was 3.3 ± 1.7 years.
At the end of the study, telephone follow-up was conducted to assess whether aortic valve replacement had been performed and if patients were still alive or, if not, if death was due to cardiovascular causes. The mean time period from the first echocardiographic examination to telephone follow-up (telephone follow-up time) was 5.0 ± 2.0 years.
Patients were grouped retrospectively on the basis of the echocardiographic follow-up examination, when the severe stage of AS was first documented ( Figure 2 ). AS was defined as severe if aortic valve area (AVA) was <1 cm 2 and as nonsevere if AVA was ≥1 cm 2 , according to guidelines. Furthermore, patients were divided into two groups according to PG mean : high-gradient AS (HG/AS) if PG mean was ≥40 mm Hg and LG/AS if PG mean was < 40 mm Hg. As recently proposed, an SVi < 35 mL/m 2 was used to further subcategorize the LG/AS group into LF/LG and NF/LG groups. By this approach, three groups were formed and used in the main analyses : group 1, HG/AS; group 2, NF/LG; and group 3, LF/LG. For subanalysis, the LF/LG group was divided into a “classical” LF/LG (C-LF/LG) subgroup with reduced LV ejection fraction (<50%) and a “paradoxical” LF/LG (P-LF/LG) subgroup with preserved LV ejection fraction. Differences between these two subgroups were tested for every parameter of the main analysis. Significant differences are reported in the results section of this article; the complete subanalysis can be accessed in the Online Data Supplement .
Categorical data are reported as counts and percentages and were analyzed using χ 2 or Fisher exact tests, as appropriate. Continuous data were assessed using one-way analysis of variance and are reported as mean ± SD. Normally distributed groups (determined by Kolmogorov-Smirnov test) were compared using the Tukey honestly significantly different or Games-Howell test, as appropriate. When distribution was not normal, the Kruskal-Wallis test was used. Differences in continuous data between baseline and follow-up within groups were compared using paired Student’s t tests or paired-sample Wilcoxon signed rank tests for key echocardiographic variables reflecting AS severity, systolic function, and hemodynamics, as indicated in the respective table or figure legend. P values < .05 were considered to indicate statistical significance. For all statistical analysis, SPSS (SPSS, Inc, Chicago, IL) was used.
Results
Echocardiographic follow-up time (3.3 ± 1.7 years) as well as telephone follow-up time (5.0 ± 2.0 years) did not differ significantly between the three groups ( Table 1 ).
| Variable | HG/AS ( n = 31) | NF/LG ( n = 18) | LF/LG ( n = 28) | P |
|---|---|---|---|---|
| Telephone follow-up (y) | 5.3 ± 1.8 | 5.2 ± 2.4 | 4.6 ± 2.0 | .363 |
| Aortic valve replacement | 21 (68%) | 5 (28%) ∗ | 14 (50%) | .026 |
| Death from any cause | 10 (32%) | 5 (28%) | 18 (64%) ∗ † | .016 |
| Echocardiographic follow-up (y) | 3.2 ± 1.5 | 3.6 ± 2.0 | 3.2 ± 1.7 | .621 |
| Decrease of two or more NYHA classes | 1 (3%) | 5 (28%) ∗ | 9 (32%) ∗ | .006 |
| Hospitalization for heart failure | 7 (23%) | 7 (42%) | 21 (78%) ∗ † | <.001 |
| EuroSCORE (%) | 12 ± 10 | 19 ± 15 | 32 ± 17 ∗ † | <.001 |
∗ P < .05, HG/AS versus NF/LG and LF/LG.
Clinical Characteristics at Baseline
At baseline, patients were 74 ± 9 years old, and 62% were men. As shown in Table 2 , all three groups shared a similar demographic background. The HG/AS group consisted of 31 patients with PG mean ≥ 40 mm Hg at echocardiographic follow-up. The LG/AS group consisted of 46 patients. Of those, 18 were allocated into the NF/LG group and 28 to the LF/LG group using an SVi cutoff of 35 mL/m 2 . The NF/LG and LF/LG groups more often had coronary as well as peripheral artery disease compared with the HG/AS group. This finding was consistent with a higher proportion of patients with histories of percutaneous intervention in the LF/LG group. All patients in the LG/AS group and 94% in the HG/AS group had hypertension.
| Variable | HG/AS ( n = 31) | NF/LG ( n = 18) | LF/LG ( n = 28) | P |
|---|---|---|---|---|
| Men | 19 (61%) | 11 (61%) | 18 (64%) | .965 |
| Age (y) | 71.8 ± 10 | 74.8 ± 10 | 75.5 ± 7 | .245 |
| Body surface area (m 2 ) | 1.94 ± 0.2 | 1.83 ± 0.2 | 1.9 ± 0.2 | .168 |
| Systolic blood pressure (mm Hg) | 137 ± 17 | 136 ± 20 | 149 ± 23 | .080 |
| Diastolic blood pressure (mm Hg) | 78 ± 9 | 71 ± 12 | 81 ± 12 † | .034 |
| Hypertension | 29 (94%) | 18 (100%) | 28 (100%) | .218 |
| Obesity | 23 (74%) | 12 (71%) | 22 (82%) | .679 |
| Hypercholesterolemia | 17 (55%) | 15 (83%) | 20 (71%) | .104 |
| Diabetes | 12 (39%) | 6 (33%) | 13 (46%) | .659 |
| Smoking | 6 (19%) | 4 (22%) | 6 (21%) | .966 |
| Atrial fibrillation | 6 (20%) | 4 (22%) | 10 (36%) | .359 |
| Left bundle branch block | 4 (16%) | 2 (13%) | 3 (13%) | .743 |
| Renal insufficiency: eGFR < 60 mL/min | 7 (23%) | 9 (50%) | 8 (30%) | .133 |
| Coronary artery disease | 9 (29%) | 11 (61%) ∗ | 22 (79%) ∗ | <.001 |
| PCI/stent | 3 (10%) | 3 (18%) | 11 (39%) ∗ | .021 |
| History of myocardial infarction | 4 (13%) | 6 (33%) | 11 (39%) | .061 |
| Peripheral artery disease | 2 (7%) | 6 (33%) ∗ | 7 (25%) | .047 |
| NYHA class I/II | 25 (81%) | 12 (67%) | 21 (75%) | .549 |
| NYHA class III/IV | 6 (19%) | 6 (33%) | 7 (25%) | .549 |
| Hospitalization for heart failure | 4 (12.9%) | 2 (11%) | 6 (22%) | .514 |
| Angina pectoris | 10 (32%) | 4 (22%) | 12 (42%) | .401 |
| Syncope | 5 (16%) | 2 (11%) | 3 (11%) | .544 |
| Pulmonary hypertension | 8 (30%) | 4 (22%) | 11 (50%) | .191 |
| Moderate AI or MI | 1 (3%) | 1 (6%) | 0 (0%) | .492 |
∗ P < .05, HG/AS versus NF/LG and LF/LG.
AS Severity and Its Progression
AVA, indexed AVA, and dimensionless velocity index were similar among all three groups at baseline ( Table 3 ). Nevertheless, in the LG/AS group, mean transvalvular flow velocity and peak transvalvular flow velocity (V max ) as well as PG mean and peak transvalvular pressure were significantly lower compared with the HG/AS group ( P < .001 for all). During echocardiographic follow-up, AVA decreased by 0.6 ± 0.4 cm 2 in the HG/AS group, by 0.5 ± 0.2 cm 2 in the NF/LG group, and by 0.4 ± 0.2 cm 2 in the LF/LG group ( P = .044) ( Figure 3 ). At follow-up, dimensionless velocity index was highest in the NF/LG group ( P = .002). Absolute and relative increases in V max and V max per year were highest in the HG/AS group ( Table 4 ). Fifty-seven percent of patients in the HG/AS group showed “fast progress,” defined as an increase of V max > 0.3 m/sec/y, whereas only 25% of patients in the NF/LG group and 13% of those in the LF/LG group showed such progress ( P < .001 for both). Similar results were found for mean transvalvular flow velocity and its annual progression. PG mean increased by 28 ± 15.5 mm Hg in the HG/AS group, by 15 ± 6.9 mm Hg in the NF/LG group, and by 6.5 ± 5.9 mm Hg in the LF/LG group ( P < .001) ( Figure 3 ). This resulted in a fivefold higher annual progression of PG mean in the HG/AS group compared with the LF/LG group.
| Variable | HG ( n = 31) | NF/LG ( n = 18) | LF/LG ( n = 28) | P |
|---|---|---|---|---|
| AS severity | ||||
| Indexed AVA § (cm 2 · m −2 ) | ||||
| BL | 0.7 ± 0.3 | 0.7 ± 0.1 | 0.6 ± 0.2 | .321 |
| FU | 0.4 ± 0.1 ‡ | 0.5 ± 0.1 ∗ ‡ | 0.4 ± 0.1 † ‡ | <.001 |
| Dimensionless velocity index § | ||||
| BL | 0.35 ± 0.1 | 0.38 ± 0.07 | 0.32 ± 0.09 | .126 |
| FU | 0.21 ± 0.05 ‡ | 0.25 ± 0.06 ∗ ‡ | 0.20 ± 0.04 † ‡ | .002 |
| Mean transvalvular velocity || (m · sec −1 ) | ||||
| BL | 2.6 ± 0.5 | 2.0 ± 0.4 ∗ | 2.1 ± 0.4 ∗ | <.001 |
| FU | 3.6 ± 0.5 ‡ | 2.7 ± 0.3 ∗ ‡ | 2.4 ± 0.3 ∗ ‡ | <.001 |
| Peak transvalvular velocity || (m · sec −1 ) | ||||
| BL | 3.5 ± 0.6 | 2.7 ± 0.4 ∗ | 2.9 ± 0.5 ∗ | <.001 |
| FU | 4.6 ± 0.6 ‡ | 3.7 ± 0.3 ∗ ‡ | 3.3 ± 0.5 ∗ ‡ | <.001 |
| Peak transvalvular pressure gradient § (mm Hg) | ||||
| BL | 49.9 ± 16.7 | 32.2 ± 9.8 ∗ | 33.1 ± 11.9 ∗ | .001 |
| FU | 85.9 ± 23.6 ‡ | 53.2 ± 8.4 ∗ ‡ | 43.9 ± 13.1 ∗ ‡ | <.001 |
| LVOT diameter (mm) | ||||
| BL | 22.0 ± 1.2 | 21.8 ± 1.8 | 21.9 ± 1.6 | .825 |
| Systolic parameters | ||||
| LV ejection fraction § [%] | ||||
| BL | 62.0 ± 7.1 | 54.7 ± 11.1 ∗ | 52.7 ± 11 ∗ | .001 |
| FU | 58.3 ± 8.5 ‡ | 53.5 ± 10.5 | 40.2 ± 13.7 ∗ † ‡ | <.001 |
| LVOT peak velocity (m · s −1 ) | ||||
| BL | 1.1 ± 0.2 | 1.0 ± 0.1 | 0.9 ± 0.1 ∗ † | .001 |
| FU | 0.9 ± 0.2 | 0.8 ± 0.1 | 0.7 ± 0.1 ∗ † | <.001 |
| Stroke volume (mL) | ||||
| BL | 98.9 ± 18.2 | 95.9 ± 14.1 | 73.2 ± 16.1 ∗ † | <.001 |
| FU | 89.4 ± 20.0 | 79.1 ± 10.2 | 51.6 ± 9.7 ∗ † | <.001 |
| SVi (mL · m −2 ) | ||||
| BL | 51.1 ± 11.1 | 51.5 ± 6.6 | 38 ± 9.8 ∗ † | <.001 |
| FU | 46.3 ± 10.4 | 43.4 ± 6.6 | 26.6 ± 4.8 ∗ † | <.001 |
| Cardiac index (L · min −1 · m −2 ) | ||||
| BL | 3.5 ± 0.8 | 3.2 ± 0.6 | 2.6 ± 0.6 ∗ | <.001 |
| FU | 3.2 ± 0.8 | 2.9 ± 0.5 | 2.2 ± 0.5 ∗ † | <.001 |
| MAPSE § (mm) | ||||
| BL | 12.0 ± 1.9 | 10.9 ± 1.8 | 10.2 ± 2.7 ∗ | .014 |
| FU | 10.9 ± 2.1 ‡ | 8.6 ± 1.7 ∗ ‡ | 6.5 ± 1.8 ∗ † ‡ | <.001 |
| Tricuspid annular plane systolic excursion (mm) | ||||
| BL | 21.7 ± 4.2 | 21.3 ± 5.5 | 19.9 ± 4.4 | .391 |
| FU | 21.2 ± 5.2 | 20.1 ± 5.8 | 16.1 ± 4.1 ∗ † | <.001 |
| Systolic pulmonary artery pressure (mm Hg) | ||||
| BL | 28.6 ± 7.3 | 29.8 ± 9.7 | 29.6 ± 6.7 | .897 |
| FU | 37.7 ± 17.9 | 29.5 ± 10.9 | 42.0 ± 13.5 † | .030 |
| Diastolic parameters | ||||
| Mitral E/A ratio | ||||
| BL | 0.9 ± 0.5 | 1.0 ± 0.5 | 0.9 ± 0.4 | .973 |
| FU | 1.2 ± 0.7 | 1.7 ± 2.8 | 1.6 ± 0.6 | .607 |
| Mitral E/E′ ratio | ||||
| BL | 20.9 ± 10.0 | 18.4 ± 6.8 | 17.2 ± 7.9 | .519 |
| FU | 25.3 ± 10.1 | 21.9 ± 10.0 | 26.8 ± 10.7 | .316 |
| Diastolic impairment, first degree/second degree/third degree | ||||
| BL | 15/2/0 | 9/3/1 | 10/3/0 | .568 |
| FU | 16/6/2 | 8/5/1 | 4/8/2 | .220 |
| Duration of diastole (msec) | ||||
| BL | 503 ± 166 | 608 ± 145 | 528 ± 132 | .126 |
| FU | 522 ± 181 | 529 ± 133 | 439 ± 99 † | .049 |
| LV and left atrial geometry | ||||
| LV end-diastolic diameter (mm) | ||||
| BL | 47.1 ± 5.4 | 48.2 ± 6.6 | 51 ± 7.4 | .074 |
| FU | 48.8 ± 5.9 | 51.1 ± 9.2 | 53.4 ± 8.7 | .087 |
| Septal diastolic thickness (mm) | ||||
| BL | 12.3 ± 1.4 | 12.5 ± 1.4 | 12.1 ± 2.0 | .788 |
| FU | 13.4 ± 1.3 | 13.3 ± 1.6 | 12.5 ± 2.4 | .137 |
| End-diastolic posterior wall thickness (mm) | ||||
| BL | 12.1 ± 1.6 | 12.0 ± 1.2 | 11.6 ± 1.7 | .419 |
| FU | 13.1 ± 1.7 | 12.9 ± 1.6 | 12.0 ± 2.0 | .068 |
| Indexed LV mass (g · m −2 ) | ||||
| BL | 111.4 ± 27.8 | 125.9 ± 28.3 | 115.9 ± 29.4 | .246 |
| FU | 134.2 ± 25.7 | 152.0 ± 42.7 | 137.1 ± 36.3 | .198 |
| Systolic left atrial diameter (mm) | ||||
| BL | 42.8 ± 5.6 | 41.3 ± 4.9 | 42.4 ± 6.4 | .693 |
| FU | 44.0 ± 5.4 | 42.8 ± 5.6 | 45.6 ± 5.9 | .241 |
| Hemodynamics | ||||
| Systemic vascular resistance § (mm Hg · min · L −1 ) | ||||
| BL | 1,153 ± 305 | 1,274 ± 274 | 1,702 ± 421 ∗ † | <.001 |
| FU | 1,306 ± 492 | 1,403 ± 416 | 1,774 ± 421 ∗ † | <.001 |
| Valvular resistance § (dyne · sec · cm −5 ) | ||||
| BL | 142 ± 67 | 94 ± 31 ∗ | 123 ± 45 | .041 |
| FU | 297 ± 79 ‡ | 194 ± 45 ∗ ‡ | 206 ± 53 ∗ | <.001 |
∗ P < .05, HG/AS versus NF/LG and LF/LG.
† P < .05, LF/LG versus NF/LG.
§ Progression between BL and FU compared by Wilcoxon signed rank test.
|| Progression between BL and FU compared by paired Student’s t test.
| Variable | HG ( n = 31) | NF/LG ( n = 18) | LF/LG ( n = 28) | P |
|---|---|---|---|---|
| Δ AVA (cm 2 ) | −0.6 ± 0.4 (−42 ± 14%) | −0.5 ± 0.2 (−34 ± 10%) | −0.4 ± 0.2 (−32 ± 11% ∗ ) | .044 (.017) |
| Δ Indexed AVA (cm 2 · m −2 ) | −0.3 ± 0.2 (−41 ± 14%) | −0.3 ± 0.1 (−34 ± 10%) | −0.2 ± 0.1 (−32 ± 12% ∗ ) | .068 (.023) |
| Δ Indexed AVA per year (cm 2 · m −2 · y −1 ) | −0.12 ± 0.10 (−17 ± 11%) | −0.10 ± 0.07 (−14 ± 11%) | −0.09 ± 0.05 (−14 ± 8%) | .350 (.534) |
| Δ Peak transvalvular velocity (m · sec −1 ) | 1.3 ± 0.9 (37 ± 28%) | 0.9 ± 0.3 (33 ± 15%) | 0.4 ± 0.5 ∗ † (16 ± 17%) | <.001 (.006) |
| Δ Peak transvalvular velocity per year (m · sec −1 · y −1 ) | 0.4 ± 0.3 (13 ± 10%) | 0.3 ± 0.3 (12 ± 7%) | 0.1 ± 0.3 (5 ± 10% ∗ † ) | <.001 (.005) |
| Fast progress >0,3 (m · s −1 · y −1 ) | 17 (57%) | 4 (25%) | 3 (13%) ∗ | <.001 |
| Δ Mean transvalvular velocity (m · sec −1 ) | 1.1 ± 0.7 (42 ± 28%) | 0.7 ± 0.4 ∗ (38 ± 27%) | 0.4 ± 0.3 ∗ † (18 ± 17%) | <.001 (.006) |
| Δ Mean transvalvular velocity per year (m · s −1 · y −1 ) | 0.4 ± 0.3 (16 ± 14%) | 0.2 ± 0.2 (12 ± 7%) | 0.1 ± 0.2 (6 ± 9% ∗ ) | <.001 (.006) |
| Δ PG mean (mm Hg) | 28 ± 15.5 (110 ± 79%) | 15 ± 6.9 ∗ (96 ± 62%) | 6.5 ± 5.9 ∗ † (39 ± 33%) | <.001 (<.001) |
| Δ PG mean per year (mm Hg · y −1 ) | 11 ± 8.1 (41 ± 32%) | 5.5 ± 4.2 (29 ± 16%) | 2.1 ± 3.4 (14 ± 18%) | <.001 (<.001) |
| Δ PG max (mm Hg) | 39 ± 24.8 (94 ± 77%) | 22 ± 8.6 (75 ± 45%) | 10 ± 14.0 (38 ± 38%) | <.001 (.010) |
| Δ Ejection fraction (percentage points) | −3.9 ± 9.0 (−6 ± 14%) | −1.2 ± 8.0 ∗ (−1 ± 14%) | −13.0 ± 11.8 ∗ † (−24 ± 22%) | <.001 (<.001) |
| Δ SVi (mL · m −2 ) | −5.1 ± 9.8 (−8 ± 18%) | −9.7 ± 8.5 (−17 ± 15%) | −11.6 ± 8.0 ∗ (28 ± 15% ∗ ) | .044 (<.001) |
| Δ MAPSE (mm) | −1.3 ± 2.0 (−10 ± 17%) | −2.1 ± 1.4 (−20 ± 11%) | −3.8 ± 2.5 ∗ † (−34 ± 18% ∗ † ) | <.001 (<.001) |
| Δ Global longitudinal strain (percentage points) | −3.3 ± 4.3 (−16 ± 21%) | −3.0 ± 4.4 (−15 ± 27%) | −3.5 ± 4.3 (−22 ± 29%) | .925 (.653) |
| Δ Valvuloarterial impedance (mm Hg · mL −1 · m 2 ) | 2.2 ± 0.9 (103 ± 47%) | 2.2 ± 0.5 (136 ± 49%) | 3.2 ± 0.8 ∗ † (119 ± 45%) | <.001 (.200) |
| Δ Systemic vascular resistance (mm Hg · min · L −1 ) | 115 ± 331 (11 ± 31%) | 118 ± 338 (28 ± 50%) | 324 ± 368 (6 ± 29%) | .057 (.302) |
| Δ Systemic arterial compliance (mL · mm Hg −1 · m −2 ) | −0.04 ± 0.08 (−6 ± 31%) | −0.07 ± 0.06 (−14 ± 20%) | −0.09 ± 0.07 (−8 ± 27%) | .036 (.748) |
∗ P < .05, HG/AS versus NF/LG and LF/LG.
Morphologic and Functional Left-Heart Assessment
LV and left atrial dimensions did not change significantly from baseline values during follow-up ( Table 3 ). LV longitudinal function differed between groups at baseline: although still preserved, the LG groups had significantly lower ejection fractions than the HG/AS group. The LF/LG group showed reduced MAPSE as well as reduced global systolic strain compared with the HG/AS group ( P = .014 and P = .003, respectively) ( Figure 4 ). In the course of echocardiographic follow-up, ejection fraction decreased by the most in the LF/LG group ( P < .001) ( Table 4 ). This reduction was mirrored in a reduction of long-axis function, reflected by MAPSE and global systolic strain ( Table 3 , Figure 4 ). The flow reduction was already apparent at baseline: in the LF/LG group, the cardiac index was reduced compared with the HG/AS group; stroke volume, SVi, and peak velocity in the LV outflow tract were reduced in the LF/LG group compared with the other groups ( P < .001 for all). All three groups showed mild diastolic impairment, whereas pseudonormal filling patterns were rare at baseline. In the LF/LG group, diastole was significantly shorter at echocardiographic follow-up, compared with the NF/LG group. The LG/AS group displayed reduced right ventricular function and elevated pulmonary artery pressure.
Hemodynamics and Afterload
Because SAC was significantly reduced ( Figure 4 ), systemic vascular resistance was significantly elevated in the LF/LG group compared with both other groups at baseline ( P < .001). In the LF/LG group, the increase in systemic vascular resistance doubled compared with both the NF/LG and HG/AS groups ( P = .057). Conversely, SAC decreased most in the LF/LG group ( P = .036). At baseline, valvuloarterial impedance was significantly higher in the LF/LG group compared with the NF/LG and HG/AS groups and showed a significantly higher absolute increase during echocardiographic follow-up ( P < .001 for both). In the HG/AS group, valvular resistance at baseline was highest ( P = .041) and increased considerably more than in both other groups ( P < .001).
Follow-Up and Outcomes
At the last echocardiographic examination, patients in the NF/LG and LF/LG groups were in worse New York Heart Association (NYHA) functional classes than those in the HG/AS group ( P < .001) ( Figure 5 , Table 1 ). A deterioration of two or more NYHA classes from baseline to echocardiographic follow-up was found in 28% of NF/LG patients and 32% of LF/LG patients versus 3% of HG/AS patients ( P = .006). Hospitalization for heart failure occurred most often in LF/LG patients ( P < .001). European System for Cardiac Operative Risk Evaluation score differed not only between the HG/AS and LG/AS groups but also within the LG/AS group at echocardiographic follow-up ( P < .001). Until telephone follow-up, only 28% of the NF/LG group underwent aortic valve replacement, compared with 50% in the LF/LG group and 68% in the HG/AS group ( P = .026). At study end, all-cause mortality was highest among LF/LG patients, with 64% deceased, in contrast with 28% in the NF/LG group and 32% in the HG/AS group ( P = .016) ( Figure 6 ).
Subanalysis of the LF/LG Group: P-LF/LG versus C-LF/LG
There were no differences in demographic and clinical data at baseline between the P-LF/LG and C-LF/LG groups, except for higher systolic blood pressure in the P-LF/LG group (167 ± 16 vs 140 ± 20 mm Hg in the C-LF/LG group, P = .003). Hospitalization for heart failure occurred more often in the C-LF/LG group (90%) than in the P-LF/LG group (50%) ( P = .044). Forty-five percent of the C-LF/LG group had decreases of two or more NYHA classes from baseline to follow-up (vs 0% in the P-LF/LG group) ( P = .044).
At follow-up, the dimensionless velocity index was below the cutoff for severe AS in both the P-LF/LG and C-LF/LG groups ( Table 5 ). Values for transvalvular flow and gradient increased significantly in both groups until follow-up ( Table 5 ). Interestingly, the values for LV ejection fraction did not differ significantly between the P-LF/LG and C-LF/LG at baseline, but there was a significant decrease in LV ejection fraction until follow-up in the C-LF/LG group ( Table 5 ). Similar observations could be made for other parameters of systolic function, such as MAPSE (P-LF/LG, −23.4 ± 16.0%; C-LF/LG, −40.0 ± 17.2% of the respective baseline value; P = .047) and global systolic strain, leading to significantly lower values at follow-up. Stroke volume and SVi were significantly lower in the C-LF/LG group, but their reduction over the course of echocardiographic follow-up could be noted in both the P-LF/LG and C-LF/LG groups. With more dilated left ventricles, baseline values for mitral E/E′ ratio in C-LF/LG patients were significantly higher than in P-LF/LG patients. Of note, mitral E/E′ ratio almost tripled in the P-LF/LG group during follow-up ( Table 5 ). Regarding hemodynamic parameters, no differences between the two subgroups could be noted, although both valvuloarterial impedance and valvular resistance increased significantly over time ( Table 5 ).
| Variable | P-LF/LG ( n = 8) | C-LF/LG ( n = 20) | P |
|---|---|---|---|
| AS severity | |||
| AVA † (cm 2 ) | |||
| BL | 1.3 ± 0.3 | 1.2 ± 0.2 | .201 |
| FU | 0.8 ± 0.1 ‡ | 0.8 ± 0.1 ‡ | .290 |
| Indexed AVA † (cm 2 · m −2 ) | |||
| BL | 0.7 ± 0.1 | 0.6 ± 0.2 | .748 |
| FU | 0.4 ± 0.1 ‡ | 0.4 ± 0.1 ‡ | .798 |
| Dimensionless velocity index ∗ | |||
| BL | 0.38 ± 0.07 | 0.28 ± 0.07 | .007 |
| FU | 0.22 ± 0.03 ‡ | 0.19 ± 0.04 ‡ | .044 |
| Mean transvalvular velocity ∗ (m · sec −1 ) | |||
| BL | 2.0 ± 0.4 | 2.2 ± 0.5 | .520 |
| FU | 2.5 ± 0.3 ‡ | 2.4 ± 0.4 ‡ | .835 |
| Peak transvalvular velocity ∗ (m · sec −1 ) | |||
| BL | 2.8 ± 0.4 | 2.9 ± 0.5 | .657 |
| FU | 3.3 ± 0.4 ‡ | 3.3 ± 0.5 ‡ | .787 |
| PG mean † (mm Hg) | |||
| BL | 18.6 ± 6.3 | 21.2 ± 8.7 | .468 |
| FU | 27.3 ± 5.6 ‡ | 26.8 ± 6.9 ‡ | .871 |
| Peak transvalvular pressure gradient ∗ (mm Hg) | |||
| BL | 31.9 ± 10.0 | 33.7 ± 13.0 | .736 |
| FU | 44.1 ± 9.8 ‡ | 43.9 ± 14.4 ‡ | .961 |
| LVOT diameter (mm) | |||
| BL | 21.3 ± 1.5 | 22.2 ± 1.6 | .227 |
| Systolic parameters | |||
| LV ejection fraction ∗ (%) | |||
| BL | 59.3 ± 5.1 | 50.2 ± 11.8 | .062 |
| FU | 56.8 ± 3.9 | 33.6 ± 10.1 ‡ | <.001 |
| LVOT peak velocity ∗ (m · sec −1 ) | |||
| BL | 0.9 ± 0.1 | 0.8 ± 0.1 | .180 |
| FU | 0.8 ± 0.1 ‡ | 0.6 ± 0.1 ‡ | .023 |
| Stroke volume ∗ (mL) | |||
| BL | 83.4 ± 13.8 | 68.1 ± 15.0 | .036 |
| FU | 58.1 ± 6.9 ‡ | 49.0 ± 9.4 ‡ | .020 |
| SVi ∗ (mL · m −2 ) | |||
| BL | 41.6 ± 6.3 | 36.0 ± 11.0 | .235 |
| FU | 29.4 ± 3.6 ‡ | 25.4 ± 4.8 ‡ | .043 |
| Cardiac index ∗ (L · min −1 · m −2 ) | |||
| BL | 2.7 ± 0.3 | 2.7 ± 0.7 | .758 |
| FU | 2.5 ± 0.5 | 2.0 ± 0.4 ‡ | .057 |
| MAPSE † (mm) | |||
| BL | 10.9 ± 1.6 | 9.9 ± 3.1 | .434 |
| FU | 8.1 ± 1.0 ‡ | 5.8 ± 1.5 ‡ | .001 |
| Tricuspid annular plane systolic excursion † (mm) | |||
| BL | 18.1 ± 4.9 | 20.8 ± 4.0 | .209 |
| FU | 17.8 ± 3.0 | 15.5 ± 4.3 ‡ | .190 |
| Systolic pulmonary artery pressure ∗ (mm Hg) | |||
| BL | 31.0 ± 7.0 | 28.8 ± 6.8 | .577 |
| FU | 48.1 ± 15.6 ‡ | 39.8 ± 12.4 | .163 |
| Global systolic strain ∗ (%) | |||
| BL | 15.4 ± 3.3 | 12.3 ± 4.8 | .143 |
| FU | 13.5 ± 2.9 | 7.7 ± 2.5 ‡ | <.001 |
| Diastolic parameters | |||
| Mitral E/A ratio ∗ | |||
| BL | 0.7 ± 0.9 | 1.1 ± 0.5 | .188 |
| FU | 1.2 ± 0.3 | 1.8 ± 0.7 ‡ | .118 |
| Mitral E/E′ ratio † | |||
| BL | 8.9 ± 4.0 | 22.0 ± 4.8 | .001 |
| FU | 26.4 ± 9.2 ‡ | 26.9 ± 11.5 | .924 |
| Diastolic impairment, first degree/second degree/third degree | |||
| BL | 4/1/0 | 6/2/0 | .685 |
| FU | 2/2/0 | 2/6/2 | .417 |
| Duration of diastole ∗ (msec) | |||
| BL | 574.9 ± 143.0 | 504.9 ± 125.3 | .264 |
| FU | 445.0 ± 139.1 | 436.8 ± 83.0 | .887 |
| LV and left atrial geometry | |||
| LV end-diastolic diameter ∗ (mm) | |||
| BL | 46.3 ± 7.6 | 52.8 ± 6.7 | .045 |
| FU | 47.2 ± 6.9 | 55.8 ± 8.2 ‡ | .016 |
| Septal diastolic thickness ∗ (mm) | |||
| BL | 12.4 ± 1.6 | 12.0 ± 2.2 | .666 |
| FU | 13.8 ± 2.6 ‡ | 12.0 ± 2.2 | .076 |
| End-diastolic posterior wall thickness ∗ (mm) | |||
| BL | 12.0 ± 1.7 | 11.4 ± 1.7 | .434 |
| FU | 12.4 ± 2.3 | 11.9 ± 1.9 ‡ | .582 |
| Indexed LV mass ∗ (g · m −2 ) | |||
| BL | 107.7 ± 29.9 | 119.3 ± 29.4 | .391 |
| FU | 124.6 ± 36.7 ‡ | 142.6 ± 35.7 ‡ | .239 |
| Systolic left atrial diameter ∗ (mm) | |||
| BL | 42.1 ± 8.7 | 42.5 ± 5.5 | .916 |
| FU | 42.9 ± 5.8 | 46.8 ± 5.6 ‡ | .115 |
| Hemodynamics | |||
| Valvuloarterial impedance ∗ (mm Hg · mL −1 · m 2 ) | |||
| BL | 2.5 ± 0.2 | 2.9 ± 0.9 | .150 |
| FU | 5.8 ± 1.0 ‡ | 5.9 ± 1.1 ‡ | .836 |
| Systemic arterial compliance (mL · mm Hg −1 · m −2 ) | |||
| BL | 0.5 ± 0.1 | 0.6 ± 0.2 | .285 |
| FU | 0.5 ± 0.1 | 0.6 ± 0.2 | .389 |
| Systemic vascular resistance (mm Hg · min · L −1 ) | |||
| BL | 1,694.2 ± 313.4 | 1,707.1 ± 412.2 | .952 |
| FU | 1,722.0 ± 375.2 | 1,798.0 ± 412.2 | .681 |
| Valvular resistance (dyne · sec · cm −5 ) | |||
| BL | 107.1 ± 45.9 | 130.7 ± 44.2 | .269 |
| FU | 188.3 ± 31.7 ‡ | 213.5 ± 58.5 ‡ | .264 |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
