Because of diastolic coupling between the left atrium and left ventricle, we hypothesized that left atrial (LA) function mirrors the diastolic function of left ventricle. The aims of this study were to assess whether LA volume parameters can be good indexes of left ventricular diastolic dysfunction. Six hundred fifty-nine patients underwent cardiac catheterization and measurements of left ventricular filling pressure (LVFP). Echocardiographic examinations including tissue Doppler and LA volumes were also assessed. Ratio of early diastolic mitral inflow velocity to early diastolic mitral annular velocity and LVFP tended to increase after progression of diastolic dysfunction. The inverse phenomenon existed in LA ejection and LA distensibility. LA distensibility was superior to LA ejection fraction and early diastolic mitral inflow velocity/early diastolic mitral annular velocity for identifying LVFP >15 mm Hg (areas under receiver operating characteristic curve 0.868, 0.834, and 0.759, respectively) and for differentiating pseudonormal from normal diastolic filling (areas under receiver operating characteristic curve 0.962, 0.907, and 0.741, respectively). Multivariate logistic regression showed that LA ejection fraction and LA distensibility were associated significantly with the presence of pseudonormal/restrictive ventricular filling. In conclusion, LA volume parameters can identify LVFP >15 mm Hg and differentiate among patterns of ventricular diastolic dysfunction. For assessing diastolic function LA parameters offer better performance than even tissue Doppler.
Diastolic dysfunction has a major role in producing signs and symptoms in patients presenting with heart failure. The gold standard for diagnosing diastolic dysfunction is the time constant of relaxation obtained by direct invasive measurement. However, this is not feasible in daily clinical practice. There are several noninvasive parameters that can assess different properties of diastolic dysfunction including Doppler mitral inflow velocity-derived variables, pulmonary venous flow velocity, color M-mode flow propagation velocity, and tissue Doppler imaging (TDI). However, many diastolic parameters are influenced by physiologic factors, particularly by alternations in preload. TDI has emerged as a new technique that is less affected by loading conditions and provides a strong complementary role in the assessment of diastolic function. Left atrial (LA) volume parameters are potentially useful tools for assessing diastolic dysfunction, but their roles need further investigation. The objective of the present study was to use LA volume parameters for grading diastolic dysfunction compared to TDI.
Methods
From December 2007 through March 2010, 1,128 patients who underwent cardiac catheterization were recruited prospectively. Exclusion criteria were the (1) presence of mitral stenosis, prosthetic mitral valve, or mitral regurgitation with greater than mild severity, (2) any abnormality of the atrial septum (e.g., atrial septal defect or aneurysm), (3) acute coronary syndrome, (4) rhythm other than sinus rhythm, (5) inadequate image quality (particularly the pattern of pulmonary venous flow), and (6) lack of informed consent. In total 659 patients were enrolled for final analysis; reasons for catheterization were peripheral artery obstructive disease (n = 26), survey of congestive heart failure (n = 102), and coronary angiography/management (n = 531). Creatinine clearance was estimated for each participant according to the Cockcroft–Gault equation. Renal dysfunction was defined as an estimated creatinine clearance <60 ml/min at first blood sampling at hospitalization. The protocol was approved by the institutional review board (VGHKS99-015). After a detailed explanation patients were invited to participate in this study and written informed consents were obtained from all enrolled participants before cardiac catheterization.
Selective angiography was performed using standard techniques. After catheterization, left ventricular filling pressure (LVFP) measurements were performed using a fluid-filled pigtail catheter placed into the left ventricle. The fourth intercostal space in the midaxillary line was used as the 0 level. LVFP was recorded continuously (50 mm/s) by a 6Fr pigtail catheter placed at the mid LV cavity using fluoroscopic screening. Measured pressure recordings included LV systolic pressure and LV pressure before the A wave, which was defined as pressure at onset of atrial contraction. The average of LV pressure before the A wave over 5 cardiac cycles defined LVFP. LVFP was considered increased at >15 mm Hg.
Echocardiography (iE33 System, Philips Medical System, Andover, Massachusetts) was performed immediately after LVFP measurements with patients still on the cardiac catheter laboratory table in a slightly left decubitus position. Transmitral flow profiles including peak early diastolic flow velocity (E), late diastolic flow velocity (A), and mitral early deceleration time were assessed. LV ejection fraction was calculated using the Simpson biplane technique. In addition, pulmonary venous flow velocities were obtained. LV mass was calculated by the formula described by Devereux and Reichek. LV mass was indexed to a patient’s body surface area. Pulse-wave TDI was performed using spectral pulse-wave Doppler signal filters and minimum optimal gain. In apical views a pulse-wave Doppler sample volume was placed at the level of the mitral annulus over the septal and lateral borders. Pulse-wave TDI results were characterized by a myocardial systolic wave, an early diastolic wave (e′), and an atrial contraction diastolic wave. The pulse-wave TDI tracing was recorded over 5 cardiac cycles at a sweep speed of 100 mm/s and used for off-line calculations. The average e′ of the septal and lateral mitral annuluses was chosen to estimate LVFP by the E/e′ method. All LA volume measurements were calculated using the biplane area–length method. LA volumes were measured at 3 points: (1) immediately before the mitral valve opening (maximal LA volume), (2) at the onset of the P wave on electrocardiogram (preatrial contraction volume), and (3) at the mitral valve closure (minimal LA volume). LA distensibility was calculated as (maximal LA volume minus minimal LA volume) multiplied by 100% divided by minimal LA volume. LA ejection fraction was calculated as (preatrial contraction volume minus minimal LA volume) multiplied by 100% divided by preatrial contraction volume. LA volumes were indexed to body surface area.
Diastolic dysfunction was classified by mitral inflow pattern. Presence of mitral E/A <0.75 or deceleration time >240 ms was considered evidence of impaired relaxation. In a more severe stage of diastolic dysfunction with pseudonormal LV filling, transmitral flow characteristics were similar to those in patients with normal diastolic function. In this study pseudonormal and normal LV fillings were defined by the presence of a mitral E/A of 0.75 to 1.50 and a deceleration time of 151 to 240 ms and differentiated by (1) E/e′ >15 or (2) E/e′ ≥11 in association with atrial reversal duration of pulmonary venous flow exceeding the mitral A-wave duration by ≥30 ms. Restrictive diastolic filling was the most severe form of diastolic dysfunction with markedly increased LVFP. Presence of mitral E/A >1.5 or deceleration time ≤150 ms was considered evidence of this abnormality.
In the first 100 enrolled cases maximal LA volume, minimal LA volume, and preatrial contraction volume were measured by 2 independent observers. Interobserver variability was calculated as the difference between values obtained by the 2 observers divided by the mean. Interobserver difference and variability of maximal LA volume were 3.2 ± 5.4 ml and 5.4 ± 8.7%, respectively. Interobserver variabilities and differences were 6.1 ± 8.9% and 2.9 ± 4.1 ml for minimal LA volume and 5.7 ± 7.4% and 2.5 ± 3.4 ml for preatrial contraction volume, respectively. Therefore, interobserver variabilities in LA distensibility and LA ejection fraction measurements were 5.8 ± 7.6% and 3.5 ± 4.2%. Interobserver difference and variability of mitral E velocity were 3.1 ± 5.8 cm/s and 4.2 ± 7.4%. Interobserver variabilities and differences were 2.7 ± 5.2% and 0.2 ± 0.4 cm/s for myocardial systolic wave, 2.2 ± 5.1% and 0.2 ± 0.4 cm/s for e′, and 2.8 ± 5.6% and 0.3 ± 0.5 cm/s for atrial contraction diastolic wave, respectively. Thus, the interobserver variability in E/e′ was 6.3 ± 7.4%.
SPSS (SPSS, Inc., Chicago, Illinois) was used for all statistical analyses. All continuous variables were presented as mean ± SD. Analysis of variance and post hoc test for unpaired data were used to estimate group differences. Comparison of clinical characteristics was performed by chi-square analysis for categorical variables. Bivariate analysis, simple correlation, and linear regression were used when appropriate. Area under the receiver operating characteristic curve was used to evaluate the sensitivity and specificity for predicting increased LVFP and differentiating pseudonormal from normal ventricular filling. Accuracy of receiver operating characteristic curve analysis was measured by area under the receiver operating characteristic curve, which indicated which parameters provided superior or inferior performance. Univariate and multivariate logistic regression analyses were performed on potential variables for predictions of increased LVFP and presence of pseudonormal ventricular filling. Variables with a potential association with multivariate modeling were selected by univariate analyses and subsequently selected with entry and retention in the model set at a significance level of 0.05. Multivariate logistic regression analysis was performed to assess the correlation between clinical parameters and advanced diastolic dysfunction (pseudonormal/restrictive filling).
Results
Table 1 presents the basic characteristics and echocardiographic parameters according to diastolic function. Seventy-four of 659 patients (11.2%) were classified as having normal filling, 317 (48.1%) as having impaired relaxation, 175 (26.6%) as having pseudonormal filling, and 93 (14.1%) as having restrictive filling. Three hundred seventy-six patients (57.1%) had normal systolic function, defined as LV ejection fraction >50%.
| Variable | Normal | Impaired Relaxation | Pseudonormal | Restrictive | p Value |
|---|---|---|---|---|---|
| (n = 74) | (n = 317) | (n = 175) | (n = 93) | ||
| Age (years) | 60 ± 14 | 67 ± 13 | 64 ± 15 | 66 ± 14 | 0.082 ⁎ |
| Women/men | 15/64 | 67/316 | 66/172 | 38/94 | <0.0001 |
| Diabetes mellitus | 10 (12.7%) | 126 (32.9%) | 90 (37.8%) | 52 (39.4%) | <0.0001 |
| Hypertension | 29 (36.7%) | 259 (67.6%) | 143 (60.1%) | 70 (53%) | <0.0001 |
| Current smoker | 43 (54.4%) | 199 (52%) | 122 (51.3%) | 56 (42.4%) | 0.003 |
| Body weight (kg) | 68 ± 10 | 68 ± 11 | 65 ± 13 | 65 ± 13 | 0.008 † ‡ ∥ |
| Renal dysfunction (creatine clearance <60 ml/min) | 18 (22.8%) | 225 (58.7%) | 135 (56.7%) | 88 (66.7%) | <0.0001 |
| Left ventricular mass index (g/m 2 ) | 119 ± 28 | 146 ± 33 | 153 ± 34 | 168 ± 40 | <0.0001 ⁎ † ‡ § ∥ ¶ |
| Heart rate (beats/min) | 70 ± 12 | 73 ± 12 | 79 ± 15 | 82 ± 17 | <0.0001 † ‡ § ∥ ¶ |
| Mitral early diastolic flow velocity (cm/s) | 83 ± 21 | 64 ± 18 | 88 ± 26 | 103 ± 30 | <0.0001 ⁎ † ‡ ∥ ¶ |
| Mitral late diastolic flow velocity (cm/s) | 68 ± 18 | 86 ± 18 | 75 ± 22 | 58 ± 25 | <0.0001 ⁎ † ‡ § ∥ ¶ |
| Deceleration time (ms) | 183 ± 25 | 231 ± 57 | 188 ± 33 | 146 ± 41 | <0.0001 ⁎ † ‡ ∥ ¶ |
| Systolic velocity of pulmonary venous flow (cm/s) | 54 ± 12 | 50 ± 12 | 46 ± 13 | 42 ± 17 | <0.0001 ⁎ † ‡ § ∥ ¶ |
| Diastolic velocity of pulmonary venous flow (cm/s) | 44 ± 10 | 48 ± 10 | 53 ± 14 | 57 ± 17 | <0.0001 † ‡ § ∥ ¶ |
| Atrial reverse velocity of pulmonary venous flow (cm/s) | 30 ± 6 | 31 ± 8 | 32 ± 6 | 35 ± 7 | 0.001 † ∥ ¶ |
| Left ventricular ejection fraction (%) | 56 ± 7 | 50 ± 9 | 45 ± 9 | 41 ± 12 | <0.0001 ⁎ † ‡ § ∥ ¶ |
| Septal mitral annulus (cm/s) | |||||
| Systolic velocity of mitral annulus | 7.3 ± 1.5 | 6.5 ± 1.6 | 6.0 ± 1.8 | 5.3 ± 1.8 | <0.0001 ⁎ † ‡ § ∥ ¶ |
| Early diastolic velocity of mitral annulus | 7.6 ± 1.9 | 6.0 ± 1.6 | 5.8 ± 2.2 | 5.3 ± 2.5 | <0.0001 ⁎ † ‡ § ¶ |
| Late diastolic velocity of mitral annulus | 8.3 ± 1.8 | 8.2 ± 1.8 | 6.9 ± 1.9 | 5.7 ± 2.1 | <0.0001 † ‡ § ∥ ¶ |
| Lateral mitral annulus (cm/s) | |||||
| Systolic velocity of mitral annulus | 9.3 ± 2.4 | 8.1 ± 2.3 | 7.3 ± 2.4 | 6.2 ± 2.5 | <0.0001 ⁎ † ‡ § ∥ ¶ |
| Early diastolic velocity of mitral annulus | 9.7 ± 2.7 | 7.7 ± 2.3 | 7.3 ± 2.8 | 7.1 ± 2.1 | <0.0001 ⁎ † § |
| Late diastolic velocity of mitral annulus | 9.7 ± 2.2 | 9.0 ± 2.3 | 8.3 ± 2.6 | 6.5 ± 2.6 | <0.0001 † ‡ § ∥ ¶ |
| Mitral early diastolic flow velocity/early diastolic velocity of septal mitral annulus | 11.7 ± 4.7 | 13.5 ± 5.2 | 17.2 ± 7.3 | 22.9 ± 9.8 | <0.0001 ⁎ † ‡ § ∥ ¶ |
| Mitral early diastolic flow velocity/early diastolic velocity of lateral mitral annulus | 9.4 ± 4.1 | 10.5 ± 4.6 | 14.1 ± 6.4 | 17.1 ± 7.4 | <0.0001 † ‡ § ∥ ¶ |
| Mitral early diastolic flow velocity/average early diastolic velocity of mitral annulus | 10.0 ± 3.7 | 11.5 ± 4.3 | 15.0 ± 5.5 | 20.1 ± 7.4 | <0.0001 ⁎ † ‡ § ∥ ¶ |
| Maximal indexed left atrial volume (ml/m 2 ) | 23.9 ± 4.5 | 30.7 ± 11.8 | 40.1 ± 16.1 | 48.0 ± 19.8 | <0.0001 ⁎ † ‡ § ∥ ¶ |
| Left atrial volume index before P wave (ml/m 2 ) | 15.6 ± 9.2 | 23.7 ± 10.0 | 33.2 ± 13.1 | 40.4 ± 18.9 | <0.0001 ⁎ † ‡ § ∥ ¶ |
| Minimal indexed left atrial volume (ml/m 2 ) | 9.1 ± 6.1 | 15.7 ± 6.9 | 26.0 ± 11.7 | 35.0 ± 19.0 | <0.0001 ⁎ † ‡ ∥ § ¶ |
| Left atrial ejection fraction (%) | 39 ± 8 | 34 ± 9 | 22 ± 9 | 14 ± 8 | <0.0001 ⁎ † ‡ § ∥ ¶ |
| Left atrial distensibility (%) | 198 ± 61 | 107 ± 44 | 61 ± 28 | 40 ± 19 | <0.0001 ⁎ † ‡ § ∥ ¶ |
| Left ventricular filling pressure (mm Hg) | 10.4 ± 3.1 | 14.3 ± 4.3 | 20.8 ± 6.3 | 26.1 ± 7.2 | <0.0001 ⁎ † ‡ § ∥ ¶ |
⁎ Normal group versus impaired relaxation group (p <0.05, post hoc analysis).
† Normal group versus restrictive group (p <0.05, post hoc analysis).
‡ Impaired relaxation group versus pseudonormal group (p <0.05, post hoc analysis).
§ Normal group versus pseudonormal group (p <0.05, post hoc analysis).
∥ Impaired relaxation group versus restrictive group (p <0.05, post hoc analysis).
¶ Pseudonormal group versus restrictive group (p <0.05, post hoc analysis).
Although LV diastolic function progressively worsened, the downward trend of LA distensibility and LA ejection fraction corresponded with an upward trend of E/e′. These 3 diastolic parameters were applied to differentiate pseudonormal from normal diastolic filling in all patients (249 patients) with the presence of mitral E/A of 0.75 to 1.50 and deceleration time of 151 to 240 ms. In accordance to receiver operating characteristic curve analysis, LA distensibility <93% precisely identified pseudonormal diastolic ventricular filling (area under receiver operating characteristic curve 0.962, sensitivity 92%, and specificity 91%) but the sensitivity/specificity of LA ejection fraction <31.6% and E/e′ >11.2 were only 85%/82% and 69%/67% ( Table 2 ). For differentiating pseudonormal from normal diastolic filling in patients with preserved LV systolic function and normal-like ventricular filling, LA distensibility was also better than LA ejection fraction and E/e′ (areas under receiver operating characteristic curve 0.915, 0.842, and 0.685, respectively; Table 2 ).
| AUC | 95% CI | Cut-Off Points | Sensitivity | Specificity | |
|---|---|---|---|---|---|
| Predict pseudonormal ventricular filling | |||||
| All patients (n = 659) | |||||
| Early diastolic flow velocity/average early diastolic velocity of mitral annulus | 0.741 | 0.680–0.802 | >11.2 | 69% | 67% |
| Left atrial ejection fraction (%) | 0.907 | 0.873–0.940 | <31.6% | 85% | 82% |
| Left atrial distensibility (%) | 0.962 | 0.943–0.982 | <94% | 92% | 91% |
| Normal left ventricular ejection fraction (n = 376) | |||||
| Early diastolic flow velocity/average early diastolic velocity of mitral annulus | 0.685 | 0.595–0.774 | >9.9 | 66% | 64% |
| Left atrial ejection fraction (%) | 0.842 | 0.777–0.908 | <31% | 76% | 80% |
| Left atrial distensibility (%) | 0.915 | 0.863–0.967 | <95% | 84% | 87% |
| Predict left ventricular filling pressure >15 mm Hg | |||||
| All patients (n = 659) | |||||
| Early diastolic flow velocity/average early diastolic velocity of mitral annulus | 0.759 | 0.727–0.791 | >11.7 | 70% | 68% |
| Left atrial ejection fraction (%) | 0.834 | 0.807–0.861 | <29% | 78% | 75% |
| Left atrial distensibility (%) | 0.868 | 0.844–0.892 | <80% | 79% | 78% |
| Normal left ventricular ejection fraction (n = 376) | |||||
| Early diastolic flow velocity/average early diastolic velocity of mitral annulus | 0.719 | 0.666–0.772 | >11.0 | 66% | 64% |
| Left atrial ejection fraction (%) | 0.779 | 0.729–0.829 | <31% | 70% | 69% |
| Left atrial distensibility (%) | 0.828 | 0.785–0.871 | <87% | 76% | 74% |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
