There are few data on the impact of left ventricular (LV) filling pressure on systolic and diastolic myocardial mechanics in patients with cardiac disease and preserved LV ejection fraction (LVEF) (≥50%).
Patients referred for cardiac catheterization underwent comprehensive echocardiography within 20 minutes of catheterization. Strain and strain rate in longitudinal, radial, and circumferential directions and torsion were measured in systole and diastole. LV preatrial contraction pressure (pre-A) was measured and averaged over 10 cardiac cycles.
Sixty patients were studied (mean age, 55.3 ± 8.9 years). The 30 patients with LV pre-A ≥ 15 mm Hg had significantly lower longitudinal systolic strain and radial, circumferential, and torsional systolic strain rates than the 30 patients with LV pre-A < 15 mm Hg ( P < .05 for all). Similar findings were seen for diastolic variables. There were significant correlations between several systolic and diastolic variables in multiple directions and LV pre-A. On multivariate analysis, the independent predictors of systolic and diastolic speckle-tracking parameters included LVEF and LV pre-A, depending on the specific parameter analyzed.
In patients with preserved LVEF and cardiac disease, several systolic and diastolic myocardial mechanical parameters significantly correlate with LV filling pressure. These data highlight the notion that patients with preserved LVEF and elevated LV filling pressures have significant abnormalities in systolic function as detected by speckle imaging, findings that may challenge the concept of “isolated diastolic dysfunction.” The extent of systolic and diastolic abnormalities in these patients may shed light on the mechanics of heart failure with preserved LVEF.
In general, left ventricular (LV) systolic function has been conventionally defined by LV ejection fraction (LVEF). However, newer echocardiographic techniques have helped identify myocardial mechanical abnormalities in both systole and diastole in patients with cardiac disease and preserved LVEF (≥50%). Specifically, non-Doppler speckle imaging has allowed the non-angle-dependent, multidirectional assessment of ventricular systolic and diastolic function. Consequently, the concept of LV function and dysfunction may be changing. Some investigators have proposed that speckle imaging may provide more comprehensive detection of myocardial abnormalities, particularly subtle ones, not detectable by LVEF alone. However, there are few data on myocardial mechanical function by speckle tracking in multiple directions in patients with preserved LVEF and their relation to invasively measured LV filling pressure. We sought to assess the mechanics of LV function in multiple directions in patients with preserved LVEF and cardiac disease and investigate if these systolic and diastolic mechanics correlated with LV filling pressure.
This study protocol was approved by the Institutional Review Board of the Baylor College of Medicine. Consecutive patients clinically referred to the cardiac catheterization laboratory for coronary angiography were screened for recruitment. Patients with moderate or greater aortic or mitral valve stenosis or regurgitation or prosthetic valves were excluded because the relationship of myocardial mechanics to LVEF and LV filling pressure, not valve disease, was the focus of this investigation. Patients were also excluded if they had nonsinus rhythm because the assessment of myocardial mechanics is more difficult in patients with irregular or high heart rates.
After providing informed consent, patients underwent left-heart catheterization in the supine position via a retrograde approach from the femoral artery. The aortic valve was crossed to determine LV diastolic pressures, and coronary angiography was performed by selective injection of the coronary ostia. LV preatrial contraction pressure (pre-A) and end-diastolic pressure (in millimeters of mercury) were measured and averaged over 10 cardiac cycles. LV pre-A ≥ 15 mm Hg was considered elevated, consistent with prior studies. Although LV end-diastolic pressure was also measured, correlations were performed using LV pre-A because it has been shown to better reflect mean left atrial pressure. The blood pressures of study subjects were measured using fluid-filled catheter measurement of central aortic pressure during cardiac catheterization.
Standard diagnostic views of the left and right coronary anatomy were obtained in patients in the supine position, and lesions ≥70% in diameter in major epicardial arteries represented significant coronary artery disease (CAD). Readings of invasive cardiac hemodynamics and coronary anatomy were performed by an invasive cardiologist blinded to all clinical and echocardiographic data. A comprehensive transthoracic Doppler echocardiographic examination was performed on patients in the left lateral decubitus position immediately after catheterization (all studies commenced within 20 min of catheterization).
Studies were performed on a GE Vivid 7 ultrasonographic machine (GE Healthcare, Milwaukee, WI). Two-dimensional measurements were performed according to the American Society of Echocardiography and included LVEF by the biplane method of disks, maximal left atrial volume by the method of disks, and LV mass by the area-length method ; the latter two variables were indexed to body surface area. Preserved LVEF was defined as ≥50%. Studies were analyzed by echocardiologists blinded to all clinical and catheterization data. Mitral inflow measurements included peak early (E) and peak late (A) velocities, the E/A ratio, and the deceleration time of E. Three cardiac cycles were measured and averaged for all Doppler measurements.
For two-dimensional speckle tracking, the LV myocardium was imaged with a frame rate >40 frames/sec (mean frame rate, 61 ≥ 9 frames/sec). Measurements of two-dimensional strain (S) and strain rate (SR) were performed by offline analysis ( Figures 1-3 ). The endocardial border was manually traced, and a myocardial region of interest was then automatically identified by the software package (EchoPAC Dimension ’06, GE Healthcare). All measured systolic and diastolic SRs were peak SRs. Peak S and systolic SR were measured as well as early diastolic and late diastolic S and SR. For the timing of early and late diastolic S and SR, the times from peak QRS to these events on SR imaging were timed, and the identical intervals were used to identify these diastolic events on S imaging, as previously described. In systole, S and SR were measured in the parasternal short-axis views at the papillary muscle level to determine radial and circumferential S and SR and in the three apical views (which were then averaged) to determine longitudinal S and SR. Similarly in diastole, early and late diastolic S and SR were determined in the radial, circumferential, and longitudinal directions. Myocardial rotation in the parasternal short-axis view was measured at the mitral valve, papillary muscle, and apical levels, from which LV torsion (LV twist normalized for LV diastolic longitudinal length) and LV torsion rate were calculated using the automated function in EchoPAC. Similarly, in diastole, early and late diastolic torsion and torsion rate were measured. All patients were imaged in sinus rhythm. LV diastolic length was measured from the mitral valve annular level to the LV apex, in centimeters, in the apical four-chamber view.
Continuous data are presented as mean ± SD and categorical data as number (percentage). For continuous variables, Student’s t test was performed, and for dichotomous variables, the χ 2 test was used. Linear regression was performed to determine the correlation between continuous variables. Multiple linear regression was used to determine the independent predictors of systolic and diastolic speckle-tracking parameters in preserved and depressed LVEF using factors associated with the presence of heart failure in such patients: presence of CAD, LVEF, LV mass index, and LV pre-A. Interobserver variability was calculated using the same data sets on the same software, using two different echocardiologists at two different times. Intraobserver variability was calculated using the same data sets using the same echocardiologist at two different times, blinded to the patients’ identifying data. A P value < .05 was significant. Analyses were performed with SigmaStat 3.0 (Systat Software, Point Richmond, CA).
Of 138 patients referred for cardiac catheterization screened, 40 were excluded for LVEF < 50%, 21 were excluded for nonsinus rhythm, nine were excluded for mitral stenosis or prosthetic mitral valve, and eight in whom the aortic valve was not crossed at catheterization were excluded. Thus, 60 patients with preserved LVEF were studied. The mean age of the population was 55.3 ± 8.9 years, 27 (45%) were women, 53 (88%) had hypertension, 32 (53%) had diabetes, 14 (23%) were current smokers, and 36 (62%) had significant CAD at catheterization. The baseline, echocardiographic, and hemodynamic characteristics of the population are displayed in Table 1 .
|LV pre-A < 15 mm Hg||LV pre-A ≥ 15 mm Hg|
|Variable||( n = 30)||( n = 30)|
|Age (years)||56.0 ± 9.0||54.6 ± 8.9|
|Women||13 (45%)||14 (43%)|
|Body surface area (m 2 )||1.9 ± 0.3||2.0 ± 0.2|
|Hypertension||26 (87%)||27 (89%)|
|Diabetes mellitus||15 (50%)||17 (56%)|
|Hypercholesterolemia||23 (76%)||22 (75%)|
|Current smokers||7 (24%)||7 (23%)|
|β-blockers||23 (77%)||25 (83%)|
|ACE inhibitors/ARBs||22 (73%)||23 (77%)|
|Calcium channel blockers||4 (13%)||5 (16%)|
|Statins||24 (80%)||26 (87%)|
|Diuretics||10 (33%)||14 (46%)|
|Echocardiography and Doppler|
|Heart rate (beats/min)||69 ± 15||67 ± 14|
|LV diastolic dimension (cm)||4.4 ± 0.7||4.5 ± 0.6|
|LV mass index (g/m 2 )||82.3 ± 38.6||97.1 ± 31.5|
|Left atrial volume index (mL/m 2 )||26.2 ± 13.7 ∗||36.2± 10.8|
|LVEF (%)||62.5 ± 7.3||62.8 ± 6.1|
|Mitral E/A ratio||0.92 ± 0.6 ∗||1.3 ± 0.5|
|Mitral deceleration time (cm/s)||220 ± 52||219 ± 63|
|Pulmonary artery systolic pressure (mm Hg)||26.5 ± 14.3||34.9 ± 12.2|
|Mitral E/Ea ratio (average of annuli)||8.9 ± 5.1 ∗||12.5 ± 6.8|
|Heart rate (beats/min)||71 ± 11 ∗||64 ± 13|
|Systolic blood pressure (mm Hg)||144 ± 23||134 ± 22|
|Diastolic blood pressure (mm Hg)||81 ± 12||76 ± 10|
|LV end-diastolic pressure (mm Hg)||15.5 ± 6.8 ∗||25.7 ± 8.9|
|LV pre-A (mm Hg)||10.8 ± 4.2 ∗||19.3 ± 8.5|
|Significant CAD on angiography||18 (62%)||18 (61%)|
Among the 60 patients with LVEF ≥ 50%, 30 (50%) had LV pre-A ≥ 15 mm Hg, and 30 (50%) had LV pre-A < 15 mm Hg. When treated as continuous variables, several systolic myocardial parameters had significant correlations with LV pre-A ( Figure 4 ), including longitudinal peak S ( R = 0.52, P < .001), longitudinal systolic SR ( R = 0.47, P = .001), radial systolic SR ( R = 0.39, P = .02), circumferential systolic SR ( R = 0.44, P = .0004), torsion ( R = 0.43, P = .005), and torsion rate ( R = 0.40, P = .01). Among systolic variables, longitudinal peak S and radial, circumferential, and torsional SR were more abnormal in patients with LV pre-A ≥ 15 mm Hg compared with <15 mm Hg ( Table 2 ).
|LV pre-A < 15 mm Hg||LV pre-A ≥ 15 mm Hg|
|Variable||( n = 30)||( n = 30)|
|Longitudinal peak S (%)||−19.1 ± 3.0||−17.1 ± 2.4|
|Longitudinal SSR (s −1 )||−0.98 ± 0.2 †||−0.86 ± 1.5|
|Radial peak S (%)||40.3 ± 19.5||38.3 ± 19.5|
|Radial SSR (s −1 )||2.1 ± 0.9 ∗||1.6 ± 0.6|
|Circumferential peak S (%)||−23.0 ± 6.8 †||−20.5 ± 4.5|
|Circumferential SSR (s −1 )||−1.4 ± 0.4 ∗||−1.1 ± 0.3|
|Torsion (%)||16.7 ± 8.7||13.5 ± 7.5|
|Torsion rate (s −1 )||107.3 ± 31.6 ∗||87.4 ± 22.7|
|Longitudinal EDS (%)||−12.7 ± 3.0||−11.8 ± 3.1|
|Longitudinal EDSR (s −1 )||1.1 ± 0.27||1.06 ± 0.3|
|Radial EDS (%)||22.3 ± 11.3||21.4 ± 13.6|
|Radial EDSR (s −1 )||−2.2 ± 0.9 †||−1.8 ± 0.6|
|Circumferential EDS (%)||−14.6 ± 5.6 ∗||−11.3 ± 3.3|
|Circumferential EDSR (s −1 )||1.6 ± 0.7||1.4 ± 0.5|
|End-diastolic torsion (%)||8.9 ± 6.5||8.6 ± 6.4|
|End-diastolic torsion rate (s −1 )||−112.0 ± 50.6 ∗||−69.8 ± 32.9|