Pulmonary artery hypertension (PAH) increases mortality in patients with severe aortic stenosis. We hypothesized that left atrial (LA) dysfunction would be related to PAH in patients with severe aortic stenosis complicated by left ventricular (LV) systolic dysfunction. The data from 70 patients with severe aortic stenosis and LV systolic dysfunction were analyzed. From the transmitral flow, the peak early (E) and late (A) diastolic velocities were obtained. From the pulmonary vein flow, the peak S-wave, D-wave, and reversed atrial wave velocities were determined. The right ventricular systolic pressure was measured in 50 patients and averaged 38 ± 13 mm Hg. Patients with PAH (n = 19) presented with greater LV diameters, E/A ratio, E-wave velocity, LV mass index, reversed atrial wave velocity, and LA volume (p <0.05) and lower S/D ratio and total and active LA emptying fractions (p <0.05). Simple linear regression analysis revealed that the LA volumes and total and active LA emptying fractions displayed the strongest correlations with the right ventricular systolic pressure. Multiple regression analysis revealed that the minimum LA volume (r = 0.61, p = 0.0001) independently correlated with the right ventricular systolic pressure, irrespective of the aortic valve (AV) area or gradient. In patients who underwent an echocardiographic examination ≥1 month after AV replacement, LA function had improved significantly. The degree of improvement was related to the degree of recovery of the LV diastolic function and diameter. In conclusion, in patients with severe aortic stenosis and concomitant LV systolic dysfunction, the LA function parameters displayed the strongest correlations with the right ventricular systolic pressure, irrespective of the AV area or gradient and were impaired in patients with PAH. LA function recovered after AV replacement. Additional studies are warranted to determine the prognostic significance of LA function in this setting.
The left atrial (LA) size is a recognized marker of increased left ventricular (LV) filling pressure and is increased in patients with severe aortic stenosis. LA function, expressed as the total LA emptying fraction, is also related to pulmonary artery hypertension (PAH) in patients with heart failure or mitral regurgitation. In patients with heart failure, the maximum LA volume has also been linked to PAH and was shown to be an independent predictor of pulmonary artery systolic pressure (PASP). In the presence of mitral regurgitation, the active LA emptying fraction correlated independently with the PASP. Therefore, we hypothesized that the LA volume and function could be potential markers of PAH in patients with aortic stenosis and concomitant heart failure. To test this hypothesis, we analyzed the LA function and its relation to PASP in patients with isolated aortic stenosis who were scheduled to undergo aortic valve (AV) replacement. We also analyzed LA function long term after surgery to assess the reversibility of LA dysfunction. Furthermore, understanding the mechanisms responsible for PAH in patients with aortic stenosis could have physiopathologic and therapeutic relevance.
Methods
The study population was selected from 1,508 consecutive patients who had undergone isolated AV replacement from January 2002 to June 2006 at Cleveland Clinic (Cleveland, Ohio). The selected patients met the following inclusion criteria: (1) severe aortic stenosis, defined as an AV area of ≤1.0 cm 2 using the continuity equation; and (2) LV systolic dysfunction, defined as a LV ejection fraction of <50%. Of the 1,508 patients, 132 were selected. Patients with significant aortic regurgitation (greater than grade 2; n = 14), more than moderate mitral regurgitation (n = 6), severe tricuspid regurgitation (n = 1), known coronary artery disease (n = 7), pacing rhythm at the preoperative examination (n = 4), permanent or paroxysmal atrial fibrillation (n = 20), and inadequate imaging quality (n = 10) were excluded. The data presented were abstracted from our echocardiography and surgical databases, which have been approved by the Institutional Review Board for Clinical Research at Cleveland Clinic. The final population of the study consisted of 70 patients. Of the 70 patients, we identified 28 who had undergone an echocardiographic examination at our institution >1 month after AV replacement. Follow-up echocardiograms were performed according to the recommendation by the patients’ physicians.
Studies were performed through several commercially available, ultrasound systems. Cardiac dimensions were measured in accordance with the American Society of Echocardiography recommendations. M-mode echocardiography was used to measure the LA diameter and LV end-diastolic and end-systolic diameters. The LV mass index was calculated according to previously published formulas. The LV and LA volumes were determined using the modified Simpson rule with images obtained from apical 4- and 2-chamber views. Pulsed wave Doppler was obtained in the apical 4-chamber view, positioned at the mitral leaflet tips. From the transmitral recordings, the peak early (E) and late (A) diastolic filling velocities, E/A ratio, and E-wave deceleration time were obtained. The peak and mean transaortic valve gradients were calculated using the simplified Bernoulli equation. The AV area was calculated by continuity equation using the velocity-time integral of the aortic and LV outflow tract flows. Pulmonary venous flow recordings were obtained from the 4-chamber view with the 5-mm sample volume positioned 1 to 2 cm into the right upper pulmonary vein, and the following measurements were taken: peak S-wave inflow velocity during ventricular systole, peak D-wave inflow velocity during the early phase of ventricular diastole and the corresponding S/D ratio, and the peak reversed atrial wave (Ar) velocity during LA contraction. PASP was derived from continuous wave Doppler interrogation of tricuspid regurgitation. The right atrial pressure was estimated from the inferior vena cava size and inspiratory collapse, in accordance with the American Society of Echocardiography recommendations and previously published reports.
The following indexes of LA function were calculated according to a previous study. The total LA stroke volume was calculated as the maximum LA volume minus the minimum LA volume. The active LA stroke volume was calculated as the precontraction LA volume minus the minimum LA volume. The passive LA stroke volume was calculated as the maximum LA volume minus the precontraction LA volume. The total LA emptying fraction was calculated as (total LA stroke volume/maximum LA volume) × 100. The active LA emptying fraction was calculated as (active LA stroke volume/precontraction LA volume) × 100. The passive LA emptying fraction was calculated as (passive LA stroke volume/maximum LA volume) × 100.
The echocardiograms were stored digitally and reviewed off-line with software (Prosolv Cardiovascular Analyzer, Problem Solving Concepts, Indianapolis, Indiana).
The calculations were done using commercially available statistical software (GraphPad Prism, version 3.02, La Jolla, California, and MedCalc, version 9.2.0.2, Mariakerke, Belgium). Continuous variables are expressed as the mean ± SD and discrete variables as percentages. Comparisons between patients with and without PAH were performed using the unpaired Student t test. Each variable was tested for correlation with PASP using simple linear regression analysis (Pearson’s correlation). All variables with a significant univariate association with PASP were entered in a multivariate stepwise regression analysis, with PASP as the dependent variable. The pre- and postoperative LA function parameters were compared using the paired Student t test. The null hypothesis was rejected at p <0.05.
Results
The clinical characteristics and surgical procedures of the studied patients are listed in Table 1 . All patients underwent isolated AV replacement without serious complications.
| Variable | Subjects (n = 70) |
|---|---|
| Clinical characteristic | |
| Age (years) | 69 ± 12 |
| Men/women | 50/20 |
| Body surface area (m 2 ) | 2.0 ± 0.3 |
| Coronary angiography | |
| Normal | 35 (50%) |
| Mild stenosis (<25%) | 27 (39%) |
| Moderate stenosis (26–50%) | 8 (11%) |
| Surgical procedure | |
| Prosthesis valve | |
| Bioprosthesis valve | 62 (89%) |
| Carpentier-Edwards | 61 |
| 3F | 1 |
| Mechanical valve | 5 (7%) |
| St. Jude | 1 |
| Carbomedics | 4 |
| Homograft | 3 (4%) |
| Prosthesis valve size (mm) | 23 ± 2 |
| 19 | 8 (11%) |
| 21 | 13 (19%) |
| 23 | 19 (27%) |
| 25 | 22 (31%) |
| 27 | 7 (10%) |
| 29 | 1 (1%) |
The echocardiographic characteristics are listed in Table 2 . Of the 70 patients, 24 (34.3%) had mild LV systolic dysfunction, 24 (34.3%) had moderate, and 22 (31.4%) had severe LV systolic dysfunction. Of the 70 patients, 38 (54%) had a mean AV gradient >40 mm Hg, 27 (39%) had a mean AV gradient of 20 to 40 mm Hg, and 5 (7%) had a mean AV gradient of <20 mm Hg. Finally, 32 patients (46%) had bicuspid valves, 37 (53%) had tricuspid valves, and 1 (1%) had a unicuspid valve. The LA volumes were increased. The PASP was measured in 50 patients and averaged 38 ± 13 mm Hg; 19 patients had elevated PASP (>40 mm Hg; 51 ± 11 mm Hg). No significant difference was found in the gender proportion between patients with PAH (13 men and 6 women) or without PAH (20 men and 11 women; p = NS). Also, no significant difference was found in age (71 ± 12 vs 69 ± 13 years, respectively) between patients with and without PAH. Patients with PAH had a larger LA size, LV diameters, LV mass index, E/A ratio, E-wave velocity, and Ar velocity and lower S/D ratio than patients without PAH. The lower LV ejection fraction and AV area in patients with PAH did not reach statistical significance ( Table 3 ).
| Variable | Subjects (n = 70) |
|---|---|
| Left ventricle | |
| End-diastolic diameter (cm) | 5.6 ± 0.7 |
| End-systolic diameter (cm) | 4.3 ± 0.8 |
| End-diastolic volume (ml/m 2 ) | 79 ± 27 |
| End-systolic volume (ml/m 2 ) | 51 ± 24 |
| Ejection fraction (%) | 38 ± 10 |
| Mass (g/m 2 ) | 158 ± 41 |
| Left atrium | |
| Diameter (cm) | 4.4 ± 0.7 |
| Area (cm 2 ) | 24 ± 6 |
| Maximum volume (ml/m 2 ) | 40 ± 17 |
| Minimum volume (ml/m 2 ) | 21 ± 12 |
| Precontraction volume (ml/m 2 ) | 30 ± 12 |
| Total emptying fraction (%) | 51 ± 15 |
| Active emptying fraction (%) | 33 ± 19 |
| Passive emptying fraction (%) | 26 ± 12 |
| Aortic valve | |
| Valve area (cm 2 ) | 0.70 ± 0.19 |
| Peak gradient (mm Hg) | 74 ± 25 |
| Mean gradient (mm Hg) | 44 ± 16 |
| Transmitral inflow | |
| E-wave velocity (cm/s) | 89 ± 33 |
| A-wave velocity (cm/s) | 82 ± 32 |
| E/A ratio | 1.3 ± 0.8 |
| E-wave deceleration time (ms) | 208 ± 93 |
| Pulmonary vein flow | |
| Peak reversed A-wave velocity (cm/s) | 32 ± 7.0 |
| Peak S-wave/peak D-wave velocities ratio | 0.94 ± 0.47 |
| Characteristic | Normal PASP (n = 31) | High PASP (n = 19) |
|---|---|---|
| Left ventricle | ||
| End-diastolic diameter (cm) | 5.3 ± 0.7 | 5.9 ± 0.7 ⁎ |
| End-systolic diameter (cm) | 4.0 ± 0.8 | 4.6 ± 0.9 † |
| End-diastolic volume (ml/m 2 ) | 74 ± 28 | 89 ± 30 |
| End-systolic volume (ml/m 2 ) | 47 ± 25 | 59 ± 27 |
| Ejection fraction (%) | 39 ± 10 | 35 ± 11 |
| Mass (g/m 2 ) | 147 ± 36 | 182 ± 39 † |
| Left atrium | ||
| Diameter (cm) | 4.2 ± 0.6 | 4.7 ± 0.7 † |
| Area (cm 2 ) | 22 ± 6 | 27 ± 5 ⁎ |
| Maximum volume (ml/m 2 ) | 36 ± 13 | 51 ± 20 † |
| Minimum volume (ml/m 2 ) | 16 ± 9 | 29 ± 14 ⁎ |
| Precontraction volume (ml/m 2 ) | 26 ± 11 | 37 ± 14 ⁎ |
| Total emptying fraction (%) | 57 ± 13 | 44 ± 11 ⁎ |
| Active emptying fraction (%) | 39 ± 18 | 24 ± 15 † |
| Passive emptying fraction (%) | 30 ± 11 | 25 ± 10 |
| Aortic valve | ||
| Valve area (cm 2 ) | 0.71 ± 0.19 | 0.60 ± 0.19 |
| Peak gradient (mm Hg) | 75 ± 24 | 83 ± 26 |
| Mean gradient (mm Hg) | 44 ± 15 | 51 ± 16 |
| Transmitral inflow | ||
| E-wave velocity (cm/s) | 78 ± 27 | 108 ± 41 ⁎ |
| A-wave velocity (cm/s) | 88 ± 28 | 82 ± 47 |
| E/A ratio | 1.0 ± 0.5 | 1.7 ± 1.2 ⁎ |
| E-wave deceleration time (ms) | 224 ± 98 | 177 ± 79 |
| Pulmonary vein flow | ||
| Peak reversed A-wave velocity (cm/s) | 31 ± 6 | 37 ± 9 ⁎ |
| Peak S-wave/peak D-wave velocities ratio | 1.15 ± 0.43 | 0.66 ± 0.50 ⁎ |
All LA volumes evaluated were larger in the patients with PAH than in those without PAH. The total and active LA emptying fractions were lower in patients with PAH than in those without PAH, but the passive LA emptying fraction was similar between the 2 groups ( Figure 1 and Table 3 ). Therefore, the LA contractile and reservoir functions were depressed in patients with PAH.
Simple linear regression analysis revealed significant positive correlations between PASP and the LA diameter and area, LV diameters, E-wave velocity, E/A ratio, LV mass index, Ar velocity and negative correlations between PASP and LV ejection fraction and S/D ratio. No significant relation to age was found. The correlations between PASP and the LA volume and function parameters were stronger than with the previously described echocardiographic parameters. PASP correlated positively with the maximum, minimum, and precontraction LA volumes and negatively with the total and active LA emptying fractions ( Figure 1 and Table 4 ).
| Variable | r | p Value |
|---|---|---|
| Left ventricle | ||
| End-diastolic diameter | 0.37 | 0.009 |
| End-systolic diameter | 0.35 | 0.01 |
| End-diastolic volume index | 0.22 | NS |
| End-systolic volume index | 0.25 | NS |
| Ejection fraction | −0.29 | 0.04 |
| Mass index | 0.37 | 0.01 |
| Left atrium | ||
| Diameter | 0.29 | 0.04 |
| Area | 0.48 | 0.0004 |
| Maximum volume index | 0.51 | 0.0002 |
| Minimum volume index | 0.63 | <0.0001 |
| Precontraction volume index | 0.52 | 0.0002 |
| Total emptying fraction | −0.59 | <0.0001 |
| Active emptying fraction | −0.50 | 0.0003 |
| Passive emptying fraction | −0.19 | NS |
| Aortic valve | ||
| Valve area | −0.14 | NS |
| Peak gradient | 0.07 | NS |
| Mean gradient | 0.09 | NS |
| Transmitral inflow | ||
| E-wave velocity | 0.44 | 0.002 |
| A-wave velocity | 0.04 | NS |
| E/A ratio | 0.34 | 0.03 |
| E-wave deceleration time | −0.22 | NS |
| Pulmonary vein flow | ||
| Peak reversed A-wave velocity | 0.33 | 0.03 |
| Peak S-wave/peak D-wave velocities ratio | −0.36 | 0.02 |
Multiple stepwise regression analysis, including all variables with a significant univariate association with PASP, revealed that only the minimum LA volume (p = 0.0001) independently correlated with the PASP.
A subgroup of 28 patients underwent follow-up echocardiography at our institution ≥1 month after AV replacement (22 ± 15 months). A significant reduction in the minimum and precontraction LA volumes was seen, with improvement of the total, passive, and active LA emptying fractions after AV replacement ( Figure 2 ). The reduction in the maximum LA volume was not significant. Also, a significant reduction was seen in the LV diameters, volumes, and mass, with improvement in the LV ejection fraction, after AV replacement. The mean and peak AV gradients decreased significantly after AV replacement. The E/A ratio did not change significantly, but an increase occurred in the E-wave deceleration time after AV replacement. The Ar velocity decreased significantly after AV replacement ( Table 5 ).
