Study Subjects

The study group consisted of 268 subjects in normal sinus rhythm (mean age, 64 ± 12 years; 171 men), including 85 healthy subjects who were recruited from hospital employees and their relatives. All healthy subjects were aged > 40 years and did not have histories of hypertension, diabetes, and/or cardiovascular disease. None of the subjects were taking cardiac medications. In addition, 183 consecutive patients, who were referred for clinically indicated transthoracic echocardiography (iE33; Philips Medical Systems, Andover, MA), were studied. Exclusion criteria included inadequate image quality of the LA wall, presence of cardiac arrhythmias, mitral valvular disease, and rapid heart rates (usually >90 beats/min) resulting in fusion of the E and A waves. The institutional review board of the hospital approved the study, and all subjects provided written informed consent before participation.

Echocardiography

Two-dimensional echocardiographic examinations were performed using a broadband transducer connected to a commercially available ultrasound machine (S5-1 probe, iE33). From the apical four-chamber and two-chamber views, LV end-diastolic and end-systolic volumes and ejection fraction were calculated using the biplane Simpson’s method. Each subject underwent pulsed-wave Doppler examination of mitral inflow obtained in the apical four-chamber view before and during the Valsalva maneuver. Pulmonary venous flow was also obtained. Doppler tissue imaging, used to measure septal mitral annular velocity, was performed by placing the sample volume in the septal corner of the mitral annulus in the apical four-chamber view. LV outflow tract velocities were obtained from the apical long-axis view. Isovolumic relaxation time was calculated as the time interval between aortic valve closure and mitral valve opening using mitral inflow and LV outflow tract velocities.

Assessment of LV Diastolic Function Grade

From mitral inflow velocities, the E-wave and A-wave velocities, E-wave deceleration time, and E/A velocity ratio were measured. Peak diastolic annular velocity during early diastolic rapid filling (E′) was also measured to calculate the E/E′ ratio in all subjects. All Doppler measurements were averaged from three consecutive beats.

LV diastolic function was categorized as normal, mild (defined as impaired relaxation without evidence of increased filling pressures), moderate (defined as impaired relaxation associated with moderate elevation of filling pressures or pseudonormal filling), or severe (defined as advanced reduction in compliance or reversible or fixed restrictive filling), as previously described by Redfield et al. ( Table 1 ). Subjects were required to meet at least two Doppler criteria consistent with mild, moderate, or severe diastolic dysfunction to be so classified. Subjects with one criterion for diastolic dysfunction or those whose parameters were borderline and suggestive of but not definitive of diastolic dysfunction were classified as indeterminate.

Assessment of LA Volume Curve Versus Time Using 2DSTE

Three consecutive beats were acquired in the apical four-chamber view (frame rate, 61 ± 14 frames/sec) using second harmonic mode. Two-dimensional endocardial tracking of the LA wall was performed using speckle-tracking software (QLAB version 7.0, TMQ; Philips Medical Systems). Initialization of the end-diastolic frame was completed with three points (septal and lateral corner of the mitral annulus and the roof of the left atrium). The software provides six tracking regions of interest on the LA wall and automatically performs speckle-tracking analysis in nine regions throughout the cardiac cycle. Manual adjustments of the LA tracking points were performed when required. LA volumes were calculated using the single-plane area-length formula. Computer-generated LA volume curves throughout the cardiac cycle were obtained, from which maximal and minimal LA volumes and LA volume before atrial contraction were measured. These values were indexed to body surface area (LAVImax, LAVImin, and LAVIpre-a, respectively).

Thereafter, the following LA emptying volume indices were calculated :

• 1.
  

  LA total emptying volume index = LAVImax − LAVImin,

  
  
• 2.
  

  LA passive emptying volume index = LAVImax − LAVIpre-a, and

  
  
• 3.
  

  LA active emptying volume index = LAVIpre-a − LAVImin.

LA reservoir function was assessed using two indices:

• 1.
  

  expansion index = [(LAVImax − LAVImin)/LAVImin] × 100, and

  
  
• 2.
  

  diastolic emptying index = [(LAVImax − LAVImin)/LAVImax] × 100.

LA conduit function was assessed using two indices:

• 1.
  

  passive emptying percentage of total emptying = [(LAVImax − LAVIpre-a)/(LAVImax − LAVImin)] × 100, and

  
  
• 2.
  

  passive emptying index = (LAVImax − LAVIpre-a)/LAVImax × 100.

Booster pump function was assessed using two indices:

• 1.
  

  active emptying percentage of total emptying = [(LAVIpre-a − LAVImin)/(LAVImax − LAVImin)] × 100, and

  
  
• 2.
  

  active emptying index = (LAVIpre-a − LAVImin)/LAVIpre-a × 100.

Statistical Analysis

Continuous data are expressed as mean ± SD. In each subject, data points describing LA volume according to time were exported to spreadsheet software (Microsoft Excel; Microsoft Corporation, Redmond, WA) to calculate LA volume waveforms. To adjust for intersubject differences in heart rate, time sequences were normalized to the percentage of systolic duration (i.e., at end-systole, t was 100%) as well as diastolic duration (i.e., at end-diastole, t was 100%). Data were analyzed using analysis of variance with post hoc analysis of Bonferroni’s correction between diastolic dysfunction grade groups. Also, linear or polynomial regression analysis was performed to determine LA functional indices and traditional diastolic parameters. A P value < .05 was considered significant. Intraobserver and interobserver variability for the LA volume measurements using 2DSTE (8%-13% and 9%-14%, respectively) were reported in our previous study.