Atrial mechanical function facilitates the transition between the almost continuous flow through the venous circulation and the intermittent filling of the ventricles and exerts a profound effect on overall cardiovascular performance.1–3 Many clinical conditions are associated with atrial remodeling and dilatation and can affect the passive and active components of atrial function.2,4 Two of them, mitral regurgitation and atrial fibrillation, are of particular interest in horses.
Chronic mitral regurgitation results in chronic volume overload and progressive LA dilatation that may eventually result in an increase in LA pressures, LA contractile dysfunction, and a reduction in LA emptying fraction.2 The degree of LA enlargement is related to the chronicity and severity of regurgitation, the extent of fluid retention, and the degree of ventricular dysfunction and plays a role as a prognostic indicator in mitral regurgitation in humans and animals.5,6
Atrial fibrillation leads to a loss of atrial pump function, associated with a decrease in cardiac output during high levels of exercise. Atrial enlargement is thought to be a major independent risk factor for development of atrial fibrillation in patients with mitral regurgitation and is an important prognostic indicator, predicting restoration of sinus rhythm and risk of atrial fibrillation recurrence after cardioversion.4,7 Conversely, atrial fibrillation itself may cause progressive atrial enlargement, most likely as a consequence of contractile dysfunction of atrial myocytes and an increase in atrial compliance.1,2,7,a In other species, including humans, atrial enlargement and atrial contractile dysfunction persist for a certain period after conversion of atrial fibrillation to sinus rhythm. This condition has been termed atrial stunning and has been attributed to atrial fibrillation–induced atrial remodeling, consisting of electrical and structural changes in the atrial myocardium.8,9 The clinical relevance of atrial stunning in horses is unknown to date. However, active pump function is particularly important during exercise, when diastole shortens and ventricular filling time is limited, and atrial mechanical dysfunction can substantially reduce cardiac output and impair exercise capacity in human athletes.2,10 Therefore, it seems likely that atrial stunning may have a negative impact on performance in athletic horses. Furthermore, the degree of atrial contractile dysfunction and the time course of recovery may have prognostic relevance and may predict maintenance of sinus rhythm after cardioversion.2,11-13
Despite the clinical relevance of atrial enlargement and atrial mechanical dysfunction in horses, structural and functional characteristics of the equine atria are incompletely studied to date. Pressure-volume loops are considered the gold standard of assessment of LA mechanical function.1,2,14 However, this approach requires invasive measures, which precludes its use in clinical practice. A variety of 2-dimensional, flow Doppler, and tissue Doppler echocardiographic variables have been used to assess LA size and mechanical function noninvasively in dogs and people.1,11,15-20 In horses, assessment of LA size has traditionally been limited to subjective evaluation and measurement of the LA diameter.6,21 However, the methods commonly used may not provide accurate measurements of the LA chamber at its widest dimensions and may not accurately reflect changes in LA geometry and actual LA size.3,21 Moreover, LA mechanical function is usually not specifically assessed during routine examination.
Accordingly, goals of the study reported here were to evaluate the feasibility, describe the techniques, and determine the reliability of transthoracic echocardiography for characterization of LA size and mechanical function in horses by use of 2-dimensional, flow Doppler, and TDI methods. Our final goal was to formulate preliminary recommendations for echocardiographic assessment of LA size and mechanical function in horses.
Materials and Methods
Study population—Six horses (4 geldings, 2 mares; 3 Standardbreds, 3 Thoroughbreds) aged (mean ± SD) 9.8 ± 2.2 years and with a body weight of 548 ± 32 kg were studied prospectively. All horses were part of the hospital teaching herd and were considered healthy on the basis of findings on physical examination, cardiac auscultation, electrocardiogram, and routine echocardiographic examination. None of the horses was in athletic condition, and none of them received medications during the 2 weeks preceding entry into the study. The Institutional Laboratory Animal Care and Use Committee of The Ohio State University approved the studies.
Echocardiography—All studies were conducted in unsedated horses standing in a quiet room, restrained by an experienced handler. Transthoracic echocardiographyb was performed with a phased-array transducerc at a frequency of 1.9/4.0 MHz (octave harmonics). A single-lead electrocardiogram was recorded simultaneously. Recordings were stored as still frames or cine-loops in digital raw format for offline analyses.d Three representative nonconsecutive cardiac cycles were measured and averaged for each variable. Cycles immediately following a sinus pause or second-degree atrioventricular block were precluded from analysis. The minimal resolution of measurements, hence the smallest possible increments of measurement given by the software, was 0.07 cm for 2-dimensional spatial measurements; 0.07 cm/s (velocity) and 4 milliseconds (time intervals) for PW tissue Doppler recordings; 0.03 cm/s and 1 millisecond for color tissue Doppler recordings; and 0.01 m/s and 4 milliseconds for PW Doppler recordings of transmitral flow velocity profiles.
Routine transthoracic 2-dimensional, M-mode, and color Doppler echocardiography were performed to assess cardiac structures, valvular competence, chamber dimensions, and left ventricular systolic function by use of standard right parasternal long-axis and shortaxis views.6,21,22 The main attention was then directed to the assessment of LA size and mechanical function. The echocardiographic variables used in this study were chosen on the basis of published data from the medical literature and on the basis of our own preliminary studies (data not shown). Details on machine settings, imaging planes, measurements, variables, calculations, and abbreviations are summarized (Appendices 1–4). Briefly, the left atrium was imaged in a right parasternal 4-chamber view, optimized to obtain an image of the entire atrium at its maximal dimensions (Figure 1). The LA linear dimensions, area, and volume, respectively, were measured or calculated during maximal atrial filling, immediately prior to mitral valve opening, and indexed to the size of the aorta.3,15,16,23 For this purpose, the aortic annulus diameter was determined from a right parasternal long-axis view by measuring the inner distance between the opened aortic valve leaflets during peak systole. Left atrial passive and active emptying were characterized by use of calculated ejection-phase indices, including passive, active, and total LA fractional shortening, fractional area change, and emptying fraction.15 Active emptying was further assessed by calculating the ratio of active to total area change and emptying volume, respectively. Finally, the reservoir function was assessed by calculating the LA reservoir indices, derived from the LA area and volume measurements, respectively.15,17
Right parasternal 4-chamber echocardiographic view of the left atrium used to obtain an image of the entire atrium at its largest dimensions. The mitral annular diameter (MAD), LA diameter (LAD), and LA length (LAL) were measured, and the LA area was traced as shown. The LA volume was calculated automatically on the basis of the traced area. All measurements were performed in zoom mode. The scale indicates the depth of penetration in centimeters. The small triangle indicates the focal zone, and the V indicates the dorsal direction. LV = Left ventricle. See Appendix 1 for detailed measurement guidelines.
Citation: American Journal of Veterinary Research 68, 7; 10.2460/ajvr.68.7.735
Right parasternal 4-chamber echocardiographic view of the left atrium used to obtain an image of the entire atrium at its largest dimensions. The mitral annular diameter (MAD), LA diameter (LAD), and LA length (LAL) were measured, and the LA area was traced as shown. The LA volume was calculated automatically on the basis of the traced area. All measurements were performed in zoom mode. The scale indicates the depth of penetration in centimeters. The small triangle indicates the focal zone, and the V indicates the dorsal direction. LV = Left ventricle. See Appendix 1 for detailed measurement guidelines.
Citation: American Journal of Veterinary Research 68, 7; 10.2460/ajvr.68.7.735
Right parasternal 4-chamber echocardiographic view of the left atrium used to obtain an image of the entire atrium at its largest dimensions. The mitral annular diameter (MAD), LA diameter (LAD), and LA length (LAL) were measured, and the LA area was traced as shown. The LA volume was calculated automatically on the basis of the traced area. All measurements were performed in zoom mode. The scale indicates the depth of penetration in centimeters. The small triangle indicates the focal zone, and the V indicates the dorsal direction. LV = Left ventricle. See Appendix 1 for detailed measurement guidelines.
Citation: American Journal of Veterinary Research 68, 7; 10.2460/ajvr.68.7.735
The left atrium, LA appendage, and aorta were also imaged in a right parasternal short-axis view. Atrial linear dimensions and atrial cross-sectional area were measured at end systole and indexed to the linear dimension and area of the aortic root (Appendix 2; Figure 2).16,18 Functional indices were not determined in the short-axis plane.
Right parasternal short-axis echocardiographic view of the left atrium, LA appendage, and aorta. Left atrial linear dimensions (LAsxD(1) and LAsxD(2)) and aortic diameter (AoD) were measured, and the LA area and aortic area were traced as shown. Measurements were performed at end systole. The V indicates the cranial direction. See Figure 1 for remainder of key and Appendix 2 for detailed measurement guidelines.
Citation: American Journal of Veterinary Research 68, 7; 10.2460/ajvr.68.7.735
Right parasternal short-axis echocardiographic view of the left atrium, LA appendage, and aorta. Left atrial linear dimensions (LAsxD(1) and LAsxD(2)) and aortic diameter (AoD) were measured, and the LA area and aortic area were traced as shown. Measurements were performed at end systole. The V indicates the cranial direction. See Figure 1 for remainder of key and Appendix 2 for detailed measurement guidelines.
Citation: American Journal of Veterinary Research 68, 7; 10.2460/ajvr.68.7.735
Right parasternal short-axis echocardiographic view of the left atrium, LA appendage, and aorta. Left atrial linear dimensions (LAsxD(1) and LAsxD(2)) and aortic diameter (AoD) were measured, and the LA area and aortic area were traced as shown. Measurements were performed at end systole. The V indicates the cranial direction. See Figure 1 for remainder of key and Appendix 2 for detailed measurement guidelines.
Citation: American Journal of Veterinary Research 68, 7; 10.2460/ajvr.68.7.735
Transmitral flow velocity profiles were recorded from a left parasternal long-axis view with the PW Doppler cursor positioned between the opened tips of the mitral leaflets (Figure 3).24,25 The transducer was positioned as ventral as possible and angled dorsally to improve alignment with blood flow. No angle correction was used. Left atrial mechanical function was characterized by peak velocity and velocity time integral during atrial contraction (A wave), atrial systolic time intervals, A-wave acceleration, and LA ejection force (Appendix 3).1,11,15,19,20,26-33
Transmitral flow velocity profile recorded from a left parasternal long-axis echocardiographic view with the PW Doppler cursor positioned between the open tips of the mitral valve leaflets. No angle correction was used. The horizontal scale indicates the time in seconds. The vertical scale indicates the velocity in meters per second. Emax = Peak E wave velocity. EVTI = Velocity-time integral E wave. Amax = Peak A wave velocity. AVTI = Velocity-time integral A wave. tAmax = Time to peak A wave. PEPA = LA pre-ejection period. ETA = LA ejection time. ATA = A wave acceleration time. See Appendix 3 for detailed measurement guidelines.
Citation: American Journal of Veterinary Research 68, 7; 10.2460/ajvr.68.7.735
Transmitral flow velocity profile recorded from a left parasternal long-axis echocardiographic view with the PW Doppler cursor positioned between the open tips of the mitral valve leaflets. No angle correction was used. The horizontal scale indicates the time in seconds. The vertical scale indicates the velocity in meters per second. Emax = Peak E wave velocity. EVTI = Velocity-time integral E wave. Amax = Peak A wave velocity. AVTI = Velocity-time integral A wave. tAmax = Time to peak A wave. PEPA = LA pre-ejection period. ETA = LA ejection time. ATA = A wave acceleration time. See Appendix 3 for detailed measurement guidelines.
Citation: American Journal of Veterinary Research 68, 7; 10.2460/ajvr.68.7.735
Transmitral flow velocity profile recorded from a left parasternal long-axis echocardiographic view with the PW Doppler cursor positioned between the open tips of the mitral valve leaflets. No angle correction was used. The horizontal scale indicates the time in seconds. The vertical scale indicates the velocity in meters per second. Emax = Peak E wave velocity. EVTI = Velocity-time integral E wave. Amax = Peak A wave velocity. AVTI = Velocity-time integral A wave. tAmax = Time to peak A wave. PEPA = LA pre-ejection period. ETA = LA ejection time. ATA = A wave acceleration time. See Appendix 3 for detailed measurement guidelines.
Citation: American Journal of Veterinary Research 68, 7; 10.2460/ajvr.68.7.735
Finally, TDI techniques were applied to evaluate active LA wall motion. Wall motion patterns were recorded by use of PW TDI and color-coded TDI techniques, respectively. The color-coded TDI recordings were analyzed by use of the system-integrated Q-analysis softwared that calculates the mean velocity of the wall segment covered by the sample area, plots the velocity data over time, and allows quantitative analyses of the wall motion patterns (Figure 4). On the basis of the results of our preliminary studies, LA wall motion was quantified by measuring the following variables: maximum and minimum radial wall motion velocity at the midpoint of the LA free wall in long- and short-axis views, time intervals from the onset of the electrocardiographic P wave to the onset and to the peak of the Am wave, and the duration of the Am wave (Appendix 4).17,34-37 When the spectral Doppler tracing of the wall motion velocity was multiphasic (characterized by an early positive or negative wave, followed by a late positive-negative wave), only the more consistent late wave was considered for measurements. When the velocity curve did not allow clear distinction of the onset of the second wave, time from P wave to onset of Am wave and duration of Am wave were not measured (this was the case in 39 out of a total of 126 measured cardiac cycles from the short-axis color TDI recordings of the LA wall).
Pulsed-wave TDI (top) and postprocessed color-coded TDI (bottom) of LA wall motion, quantified by measuring the maximum (vmax) and minimum (vmin) radial wall motion velocity at the midpoint of the LA free wall in long-axis (lx) and short-axis (sx). Time intervals from the onset of the electrocardiographic P wave to the onset (tAm) and the peak of the Am wave (tvmax), respectively, and the duration of the Am wave (dAm) were determined. Only the more consistent late wave was considered for measurements. The horizontal scales of the spectral tracings indicate the time in seconds. The vertical scales of the spectral tracings indicate the velocity in meters per second (top) and centimeters per second (bottom),. See Appendix 4 for detailed measurement guidelines.
Citation: American Journal of Veterinary Research 68, 7; 10.2460/ajvr.68.7.735
Pulsed-wave TDI (top) and postprocessed color-coded TDI (bottom) of LA wall motion, quantified by measuring the maximum (vmax) and minimum (vmin) radial wall motion velocity at the midpoint of the LA free wall in long-axis (lx) and short-axis (sx). Time intervals from the onset of the electrocardiographic P wave to the onset (tAm) and the peak of the Am wave (tvmax), respectively, and the duration of the Am wave (dAm) were determined. Only the more consistent late wave was considered for measurements. The horizontal scales of the spectral tracings indicate the time in seconds. The vertical scales of the spectral tracings indicate the velocity in meters per second (top) and centimeters per second (bottom),. See Appendix 4 for detailed measurement guidelines.
Citation: American Journal of Veterinary Research 68, 7; 10.2460/ajvr.68.7.735
Pulsed-wave TDI (top) and postprocessed color-coded TDI (bottom) of LA wall motion, quantified by measuring the maximum (vmax) and minimum (vmin) radial wall motion velocity at the midpoint of the LA free wall in long-axis (lx) and short-axis (sx). Time intervals from the onset of the electrocardiographic P wave to the onset (tAm) and the peak of the Am wave (tvmax), respectively, and the duration of the Am wave (dAm) were determined. Only the more consistent late wave was considered for measurements. The horizontal scales of the spectral tracings indicate the time in seconds. The vertical scales of the spectral tracings indicate the velocity in meters per second (top) and centimeters per second (bottom),. See Appendix 4 for detailed measurement guidelines.
Citation: American Journal of Veterinary Research 68, 7; 10.2460/ajvr.68.7.735
Reliability of echocardiographic variables—All horses underwent repeated echocardiographic examinations by 2 experienced examiners, according to the previously determined imaging guidelines. One echocardiographer (CCS) examined each horse 3 times at an interval of 2 days. On 1 occasion, a second, independent echocardiographer (KES) examined each horse immediately before (3 horses) or after (3 horses) the other echocardiographer. All recordings were labeled with random codes, allowing subsequent offline measurements in a blinded fashion.
Intraobserver measurement variability was determined by a single observer (CCS) measuring the same 6 studies (1 study of each horse) repeatedly on 3 days, thereby averaging the same set of 3 cardiac cycles for each variable. For determination of interobserver measurement variability, a second observer (KES) measured the same cardiac cycles on the same 6 studies, independently of the first observer. For determination of interobserver within-day variability, 1 observer (CCS) measured the 2 studies of each horse that were recorded consecutively on the same day by the 2 observers. Intraobserver between-day variability was determined by 1 blinded observer (CCS) measuring the 3 studies of each horse that were recorded by CCS on different days. Interobserver between-day variability was determined by 1 blinded observer (CCS) measuring 2 studies of each horse that were recorded by the 2 observers 2 days apart; the studies were chosen so that 3 horses were examined by CSS first and 3 horses were examined by KES first. All measurements were performed with the stored recordings in random order and with the observers blinded to signalment and previous measurements.
The test reliability was quantified by use of the within-subject variance for repeated measurements (residual mean square) determined by a 1-way ANO-VA with horses as the groups.38 The within-subject standard deviation was calculated as the square root of the residual mean square. Measurement variability and recording variability were reported in 2 ways. Firstly, the within-subject CV expressed as a percent value was calculated as CV = sw/mean × 100, where sw is the within-subject standard deviation.38 The degree of variability was arbitrarily defined as follows: CV < 5%, very low variability; 5% to 15%, low variability; 15% to 25%, moderate variability; > 25%, high variability. Secondly, in addition to the CV, the absolute value below which the difference between 2 measurements will lie with 95% probability was estimated following the British Standards Institution recommendations as follows: 1.96 × <2 × sw = 2.77 × sw.38 Summary statistics (mean ± SD) of each variable were calculated on the basis of the first study of each horse (n = 6) and are reported for comparison. All statistical and graphical analyses were performed by use of standard computer software.e,f
Results
Echocardiographic assessment of LA size and mechanical function by 2-dimensional echocardiography and TDI was possible in all horses by use of right parasternal long- and short-axis views. Transmitral flow velocity profiles could be recorded in all horses by PW Doppler echocardiography in a left parasternal long-axis view.
Reliability data of all echocardiographic variables of LA size and mechanical function were summarized (Tables 1–4). Briefly, most 2-dimensional variables of LA size had very low to low variability in long- and short-axis imaging planes, respectively. Exceptions were the indexed measures of LA area and volume in long-axis view. Among the 2-dimensional indices of LA mechanical function, the fractional shortening (on the basis of linear measurements of the LA diameter) had the highest variability for all 3 components (passive, active, and total). Conversely, variability of area- and volume-based variables, respectively, was considerably lower. Generally, variability was higher for the indices of active LA function, compared with the indices of passive function and reservoir function. Variables of LA function derived from PW Doppler imaging of transmitral blood flow were characterized by very low to low measurement variability, low to moderate within-day variability, and low to high between-day variability. The TDI variables were characterized by very low to low measurement variability and low to high within-day and between-day variability. Wall motion velocities were more variable than TDI-derived time intervals.
Reliability of 2-dimensional echocardiographic variables used for assessment of LA size on the basis of findings in 6 horses.
| Variables | Mean ± SD | Measurement variability | Within-day variability | Between-day variability | |||||||
|---|---|---|---|---|---|---|---|---|---|---|---|
| Intraobserver | Interobserver | Interobserver | Intraobserver | Interobserver | |||||||
| CV (%) | BSI* | CV (%) | BSI* | CV (%) | BSI* | CV (%) | BSI* | CV (%) | BSI* | ||
| Right parasternal 4-chamber view of the left atrium† | |||||||||||
| LADmax(cm)§ | 12.26 ± 1.10 | 1.6 | 0.53 | 2.7 | 0.93 | 4.5 | 1.5 | 3.6 | 1.2 | 4.0 | 1.4 |
| LADmax/AAD§ | 1.97 ± 0.22 | 1.8 | 0.094 | 1.8 | 0.098 | 7.5 | 0.41 | 4.8 | 0.26 | 11.1 | 0.61 |
| LALmax(cm) | 9.09 ± 0.69 | 1.9 | 0.47 | 4.5 | 1.2 | 4.7 | 1.2 | 2.9 | 0.74 | 3.6 | 0.91 |
| LALmax/AAD | 1.46 ± 0.12 | 2.0 | 0.078 | 4.0 | 0.16 | 7.2 | 0.29 | 3.6 | 0.15 | 10.9 | 0.44 |
| MADmax(cm) | 9.77 ± 0.94 | 0.9 | 0.25 | 3.1 | 0.83 | 3.7 | 1.0 | 3.1 | 0.84 | 3.7 | 1.0 |
| MADmax/AAD | 1.57 ± 0.18 | 1.3 | 0.057 | 3.3 | 0.14 | 5.8 | 0.25 | 3.9 | 0.17 | 7.6 | 0.34 |
| LAAmax(cm2)§ | 91.3 ± 11.8 | 1.7 | 4.2 | 6.8 | 18.0 | 5.0 | 13.0 | 4.7 | 12.0 | 4.9 | 13.0 |
| LAAmax/AAD2§ | 2.35 ± 0.36 | 2.6 | 0.17 | 5.8 | 0.39 | 11.7 | 0.77 | 6.4 | 0.42 | 20.4 | 1.4 |
| LAVmax(cm3) | 771 ± 155 | 2.5 | 54.0 | 9.3 | 211.0 | 7.4 | 164.0 | 7.3 | 158.0 | 7.2 | 161.0 |
| LAVmax/AAD3 | 3.20 ± 0.74 | 4.1 | 0.36 | 8.1 | 0.75 | 17.4 | 1.6 | 9.6 | 0.84 | 28.8 | 2.7 |
| Right parasternal short-axis view of the aorta and left atrium‡ | |||||||||||
| LAsxDmax(1)(cm) | 10.40 ± 1.05 | 2.0 | 0.58 | 8.1 | 2.2 | 4.2 | 1.3 | 5.7 | 1.7 | 5.2 | 1.6 |
| LAsxDmax(1)/AosxD | 1.50 ± 0.19 | 3.1 | 0.13 | 11.2 | 0.44 | 5.4 | 0.23 | 8.8 | 0.38 | 5.9 | 0.25 |
| LAsxDmax(2)(cm) | 10.71 ± 0.92 | 1.1 | 0.33 | 2.8 | 0.83 | 2.6 | 0.78 | 2.5 | 0.75 | 5.4 | 1.62 |
| LAsxDmax(2)/AosxD | 1.54 ± 0.17 | 3.2 | 0.13 | 5.7 | 0.24 | 4.2 | 0.18 | 5.9 | 0.25 | 6.0 | 0.25 |
| LA sxAmax (cm)2)§ | 113.6 ± 18.8 | 1.7 | 5.2 | 3.5 | 11.0 | 7.4 | 25.0 | 7.4 | 24.0 | 12.8 | 16.0 |
| LAsxAmax /AosxA§ | 2.58 ± 0.46 | 2.4 | 0.17 | 12.3 | 0.82 | 5.7 | 0.43 | 8.9 | 0.67 | 7.2 | 0.54 |
Absolute value below which the difference between 2 measurements will lie with 95% probability following the recommendations of the BSI.
Timing (max) of measurements was as follows: max= 1 frame before opening of the mitral valve (ie, maximum size).
Timing (max) of measurements was as follows: max = atrial size at end systole, 1 frame after closure of the aortic valve (ie, maximum size).
Variables recommended for assessment of LA size and mechanical function.
BSI = British Standards Institution. LAD = LA diameter. AAD = Aortic annular diameter. LAD / AAD = LAD indexed to AAD. LAL = LA length. LALmax/AAD = LAL indexed to AAD. MAD = Mitral annular diameter. MADmax/AAD = MAD indexed to AAD. LAA = LA area. LAAmaxmax/AAD2 = LAA indexed to AADmax2. LAV = LA volume. LAVmax/AAD3 = LAV indexed to AAD3. sx = Short-axis view. LAsxDmax(1) = LA diameter (1). LAsxDmax(2)max = LA diameter (2). AosxD = Aortic diameter. LAsxDmax(1)/AosxD = LAsxDmax(1) indexed to AosxD. LAsxDmax(2)/AosxD = LAsxDmax(2) indexed to AosxD. LAsxAmax = LA area. AosxA = Aortic area. LAsxsxAmax/AosxA = LAsxmax(1)sxAmax indexed to AosxA See Appendices 1 and 2 for detailed explanations of variables.
Reliability of 2-dimensional echocardiographic variables from the right parasternal 4-chamber view used for assessment of LA mechanical function on the basis of findings in 6 horses.
| Variables†‡ | Mean ± SD | Measurement variability | Within-day variability | Between-day variability | |||||||
|---|---|---|---|---|---|---|---|---|---|---|---|
| Intraobserver | Interobserver | Interobserver | Intraobserver | Interobserver | |||||||
| CV (%) | BSI* | CV (%) | BSI* | CV (%) | BSI* | CV (%) | BSI* | CV (%) | BSI* | ||
| Passive LA-FS (%) | 4.6 ± 2.0 | 24.2 | 3.3 | 42.6 | 5.0 | 34.5 | 5.5 | 31.2 | 4.5 | 35.6 | 6.2 |
| Active LA-FS(%) | 7.8 ± 4.9 | 29.4 | 5.4 | 31.4 | 6.5 | 48.1 | 8.7 | 57.0 | 8.9 | 51.2 | 11.0 |
| Total LA-FS (%) | 12.0 ± 5.8 | 18.8 | 5.8 | 27.0 | 8.6 | 34.4 | 11.0 | 34.7 | 10.0 | 35.4 | 13.0 |
| Passive LA-FAC(%)§ | 21.7 ± 4.0 | 7.4 | 4.4 | 15.4 | 8.4 | 12.0 | 8.0 | 11.5 | 7.3 | 15.8 | 9.8 |
| Active LA-FAC(%)§ | 19.1 ± 6.1 | 13.0 | 6.8 | 14.9 | 7.9 | 24.3 | 13.0 | 23.8 | 12.0 | 25.2 | 14.0 |
| Total LA-FAC(%)§ | 36.8 ± 5.6 | 6.5 | 6.6 | 9.9 | 9.6 | 4.6 | 4.9 | 7.5 | 7.6 | 11.8 | 12.0 |
| Passive LA-EF(%) | 25.9 ± 3.8 | 7.6 | 5.3 | 18.5 | 12.0 | 11.3 | 8.9 | 8.9 | 6.6 | 11.9 | 8.8 |
| Active LA-EF(%) | 24.4 ± 9.1 | 13.3 | 9.0 | 13.9 | 9.5 | 23.1 | 15.0 | 22.6 | 14.0 | 29.8 | 21.0 |
| Total LA-EF(%) | 44.3 ± 7.1 | 6.5 | 7.8 | 9.6 | 11.0 | 6.9 | 8.6 | 8.5 | 10.0 | 13.1 | 16.0 |
| Active:total AC§ | 0.40 ± 0.10 | 10.6 | 0.12 | 15.5 | 0.19 | 24.3 | 0.25 | 22.7 | 0.23 | 22.4 | 0.25 |
| Active:total EV | 0.40 ± 0.12 | 11.6 | 0.13 | 18.6 | 0.23 | 22.1 | 0.22 | 17.9 | 0.18 | 25.4 | 0.28 |
| LA-RI(Area)(%)§ | 60.0 ± 13.5 | 10.2 | 17.0 | 14.2 | 22.0 | 8.0 | 14.0 | 11.7 | 19.0 | 18.1 | 31.0 |
| LA-R(Vol)(%) | 83.5 ± 21.6 | 12.0 | 27.0 | 16.5 | 35.0 | 13.4 | 32.0 | 16.1 | 36.0 | 21.3 | 51.0 |
Timing (min, a, max) of measurements was as follows: max = 1 frame before opening of the mitral valve (ie, maximum size), a = onset of the P wave (ie, size at onset of active contraction), and min = closure of the mitral valve (ie, minimum size).
Passive = (max - a)/max. Active = (a -min)/a. Total = (max-min)/max.
Variables recommended for assessment of LA size and mechanical function.
LA-FS = LA fractional shortening. LA-FAC = LA fractional area change. LA-EF= LA emptying fraction. Active:total AC = Ratio of active to total area change. Active:total EV = Ratio of active to total emptying volume. LA-RI(Area) = Area-based LA reservoir index. LA-RI(Vol) = Volume-based LA reservoir index.
See Table 1 for remainder of key and Appendix 1 for detailed explanations of variables.
Reliability of PW Doppler variables of transmitral blood flow from the left parasternal angled long-axis view used for assessment of LA mechanical function of 6 horses.
| Variables | Mean ± SD | Measurement variability | Within-day variability | Between-day variability | |||||||
|---|---|---|---|---|---|---|---|---|---|---|---|
| Intraobserver | Interobserver | Interobserver | Intraobserver | Interobserver | |||||||
| CV (%) | BSI* | CV (%) | BSI* | CV (%) | BSI* | CV (%) | BSI* | CV (%) | BSI* | ||
| Emax(Cm/s)† | 58.5 ± 18.4 | 1.8 | 3.0 | 3.5 | 5.8 | 13.8 | 22.0 | 13.8 | 23.0 | 26.3 | 44.0 |
| EVTI(cm) | 11.5 ± 3.7 | 4.8 | 1.6 | 3.8 | 1.2 | 16.0 | 5.7 | 16.9 | 5.9 | 28.7 | 10.0 |
| Amax (cm/s)† | 37.1 ± 8.8 | 1.6 | 1.7 | 4.4 | 4.6 | 8.4 | 7.8 | 18.8 | 18.0 | 22.7 | 22.0 |
| AVTI(cm) | 4.6 ± 1.0 | 2.3 | 0.29 | 8.4 | 1.0 | 11.0 | 1.4 | 19.6 | 2.4 | 20.0 | 2.6 |
| Emax:Amax† | 1.62 ± 0.42 | 2.3 | 0.10 | 1.9 | 0.086 | 12.9 | 0.64 | 18.3 | 0.92 | 24.1 | 1.2 |
| EVTI:AVTI | 2.58 ± 0.81 | 3.5 | 0.25 | 3.9 | 0.29 | 21.9 | 1.8 | 27.0 | 2.2 | 30.1 | 2.4 |
| FAmax(%)† | 39 ± 7 | 1.3 | 1.4 | 1.1 | 1.2 | 8.1 | 8.2 | 12.1 | 12.0 | 15.7 | 16.0 |
| FAVTII(%) | 30 ± 6 | 2.7 | 2.1 | 3.4 | 2.8 | 14.6 | 11.0 | 19.1 | 14.0 | 19.5 | 15.0 |
| tAmax (ms)† | 170 ± 27 | 4.4 | 23.1 | 7.4 | 38.7 | 8.4 | 44.9 | 15.4 | 82.1 | 7.4 | 39.0 |
| PEPA(ms)† | 89 ± 20 | 4.9 | 12.0 | 10.5 | 27.0 | 20.4 | 53.0 | 19.5 | 53.0 | 24.0 | 63.0 |
| ETA (ms)† | 207 ± 19 | 3.4 | 20.0 | 11.3 | 61.0 | 4.4 | 26.0 | 6.0 | 36.0 | 6.7 | 39.0 |
| ATA (ms)† | 92 ± 17 | 5.1 | 13.0 | 7.1 | 19.0 | 10.4 | 29.0 | 10.6 | 28.0 | 16.9 | 46.0 |
| dv/dtA (cm/s2) | 434 ± 162 | 3.8 | 46.0 | 3.4 | 40.0 | 15.7 | 153.0 | 26.2 | 272.0 | 34.8 | 359.0 |
| PEPA:ETA† | 0.44 ± 0.12 | 8.3 | 0.10 | 18.7 | 0.26 | 21.7 | 0.27 | 19.1 | 0.25 | 18.7 | 0.26 |
| LA ejection force (kilodynes) | 53.3 ± 20.9 | 4.6 | 7.1 | 6.4 | 9.7 | 10.8 | 14.0 | 33.8 | 44.0 | 6.4 | 9.7 |
Variables recommended for assessment of LA mechanical function.
Emax = Peak E wave velocity. EVTI = Velocity-time integral E wave. Amax = Peak A wave velocity. AVTI = Velocity-time integral A wave. Emax:Amax = Ratio of Emax to Amax. EVTI:AVTI = Ratio of EVTIVTI to AVTI. FAmax = Fractional active emptying velocity. FAmaxVTIVTI = Fractional active emptying velocity-time integral. tAmax = Time to peak A wave. PEPA = LA pre-ejection period. ETA = LA ejection time. ATA = Acceleration time of A wave. dv/dtA = Acceleration rate of A wave. PEPA:ETA = Ratio of PEPA to ETAA.
See Table 1 for remainder of key and Appendix 3 for detailed explanations of variables.
Reliability of tissue Doppler variables used for assessment of LA mechanical function of 6 horses.
| Variables | Mean ± SD | Measurement variability | Within-day variability | Between-day variability | |||||||
|---|---|---|---|---|---|---|---|---|---|---|---|
| Intraobserver | Interobserver | Interobserver | Intraobserver | Interobserver | |||||||
| CV (%) | BSI* | CV (%) | BSI* | CV (%) | BSI* | CV (%) | BSI* | CV (%) | BSI* | ||
| Right parasternal 4-chamberviewof the left atrium | |||||||||||
| PWTDI | |||||||||||
| vmax (cm/s) | 5.4 ± 2.9 | 3.2 | 0.47 | 8.2 | 1.3 | 24.4 | 2.9 | 45.5 | 5.7 | 41.5 | 5.9 |
| vmin (cm/s) | 4.9 ± 1.8 | 2.1 | 0.29 | 12.5 | 1.9 | 20.6 | 2.3 | 33.2 | 4.0 | 33.2 | 4.5 |
| tvamax (ms)† | 190 ± 21 | 1.4 | 7.1 | 2.9 | 16.0 | 6.7 | 33.0 | 12.3 | 65.0 | 12.0 | 63.0 |
| tAm (ms)† | 147 ± 12 | 2.8 | 11.0 | 5.9 | 24.0 | 7.0 | 27.0 | 12.5 | 52.0 | 13.0 | 52.0 |
| dAm (ms)† | 126 ± 18 | 3.5 | 12.0 | 4.9 | 18.0 | 6.2 | 23.0 | 11.3 | 40.0 | 13.3 | 48.0 |
| tAm:dAm† | 1.19 ± 0.14 | 6.3 | 0.21 | 11.7 | 0.37 | 10.9 | 0.32 | 19.6 | 0.66 | 18.6 | 0.59 |
| Color TDI | |||||||||||
| vmax (cm/s) | 4.0 ± 1.9 | 3.4 | 0.38 | 13.8 | 1.7 | 28.3 | 2.1 | 34.1 | 3.1 | 58.1 | 5.5 |
| vmin (cm/s) | 3.8 ± 1.7 | 4.8 | 0.49 | 13.3 | 1.5 | 20.3 | 1.4 | 41.7 | 3.5 | 42.7 | 3.9 |
| tvmax (ms) | 209 ± 22 | 1.8 | 10.0 | 2.2 | 13.0 | 11.6 | 63.0 | 9.2 | 53.0 | 15.4 | 85.0 |
| tAm (ms) | 154 ± 10 | 1.8 | 7.6 | 5.7 | 25.0 | 8.1 | 33.0 | 25.7 | 107.0 | 45.5 | 185.0 |
| dAm(ms) | 116 ± 14 | 3.3 | 11.0 | 5.4 | 17.0 | 6.3 | 22.0 | 24.2 | 83.0 | 43.8 | 150.0 |
| tAm :dAm | 1.35 ± 0.14 | 4.2 | 0.16 | 9.0 | 0.34 | 11.4 | 0.38 | 15.4 | 0.52 | 10.8 | 0.36 |
| Right parasternal short-axis view of the left atrium | |||||||||||
| PWTDI | |||||||||||
| vmax (cm/s) | 6.5 ± 3.3 | 2.1 | 0.38 | 4.8 | 0.89 | 7.5 | 1.0 | 40.9 | 6.1 | 45.6 | 7.0 |
| vmin (cm/s) | 6.1 ± 2.4 | 8.9 | 1.4 | 9.1 | 1.6 | 6.5 | 0.86 | 32.2 | 4.6 | 38.2 | 5.6 |
| tvmax (ms) | 190 ± 15 | 1.8 | 9.3 | 1.6 | 8.3 | 8.0 | 39.0 | 9.5 | 50.0 | 18.0 | 91.0 |
| tAm (ms) | 143 ± 10 | 1.9 | 7.4 | 2.9 | 11.0 | 12.1 | 44.0 | 11.8 | 47.0 | 26.2 | 99.0 |
| dAm(ms) | 132 ± 13 | 2.5 | 9.1 | 3.8 | 14.0 | 10.6 | 39.0 | 6.2 | 23.0 | 11.0 | 39.0 |
| tAm:dAm | 1.00 ± 0.13 | 3.3 | 0.10 | 5.1 | 0.15 | 16.3 | 0.45 | 13.8 | 0.42 | 32.2 | 0.98 |
| Color TDI | |||||||||||
| vmax (cm/s) | 3.7 ± 1.8 | 2.9 | 0.30 | 2.8 | 0.28 | 10.0 | 0.71 | 35.7 | 3.1 | 51.3 | 4.5 |
| vmin (cm/s) | 3.4 ± 1.4 | 3.5 | 0.34 | 3.8 | 0.36 | 11.1 | 0.85 | 26.3 | 2.4 | 48.7 | 4.4 |
| tvmax (ms) | 201 ± 9 | 1.7 | 9.4 | 1.4 | 8.0 | 10.4 | 52.0 | 11.2 | 61.0 | 14.3 | 73.0 |
| tAm (ms) | 152 ± 13 | 2.0‡ | 8.4‡ | 2.6‡ | 11.0‡ | 5.9‡ | 23.0‡ | 16.7‡ | 68.0‡ | 5.9‡ | 24.0‡ |
| dAm(ms) | 118 ± 7 | 3.7‡ | 12.0‡ | 4.8‡ | 16.0‡ | 8.6‡ | 27.0‡ | 5.1‡ | 17.0‡ | 9.1‡ | 29.0‡ |
| tAm:d Am | 1.29 ± 0.13 | 3.3‡ | 0.12‡ | 4.8‡ | 0.17‡ | 4.9‡ | 0.17‡ | 15.9‡ | 0.55‡ | 5.9‡ | 0.21‡ |
Variables recommended for assessment of LA mechanical function.
Values affected by selection bias (39 of 126 cycles could not be measured because the velocity curve did not allow clear distinction of the onset or the end of the second velocity wave) leading to falsely high reliability (low variability) values.
Vmax = Peak positive wall motion velocity. Vmin = Peak negative wall motion velocity. tvmax = Time to peak positive velocity. tAm = Time from P wave to onset of Am wave. dAm = Duration of Am wave. tAm:dAm = Ratio of tAm to dAm.
See Table 1 for remainder of key and Appendix 4 for detailed explanations of variables.
Discussion
Results of our study indicate that LA size and mechanical function can be assessed by transthoracic echocardiography in standing unsedated adult horses. Furthermore, our results provide data on imaging techniques and reliability by use of a variety of standard 2-dimensional and transmitral flow Doppler measurements as well as TDI variables.
Assessment of LA dimensions by measurement of LA diameter is part of any routine echocardiogram in horses.6,21,22 However, the 2 methods traditionally used to measure the LA diameter have substantial limitations. M-mode recordings performed in a right parasternal short-axis view at the level of the heart base largely depend on the placement of the cursor line and often provide a measure of the LA appendage rather than the LA body.21 Two-dimensional echocardiography in a left parasternal long-axis view often does not allow imaging of the left atrium in its entirety as a result of interference with the ventral lung border; this often results in measurements of the LA diameter that are not parallel to the mitral valve annulus or that are made too close to the annulus and thereby underestimate the true maximal atrial diameter.21 Conversely, the right parasternal window used in our study offers a view of the entire left atrium and provides sufficient anatomic landmarks for consistent measurements of the largest LA dimensions in any direction and at any instance of the cardiac cycle.
One-dimensional (linear) measurements such as the LA diameter are easy to obtain. However, in humans, they tend to be less sensitive in the detection of changes in LA size, compared with 2- or 3-dimensional measurements, because LA enlargement may not occur in a uniform fashion and LA geometry may change over time.3,4,39 Anatomic limitations in adult horses (ie, inability to obtain left apical views) prevent the use of the standard biplane measures of LA volume that are recommended in people.3,40 Therefore, in addition to LA diameter and length, we chose to measure LA area in right parasternal long- and short-axis views. We also calculated the LA volume by use of a single-plane method of disks in the long-axis plane. However, single-plane estimates of LA volume assume that the atrium has a circular cross-section in the third dimension, so that part of the left atrium and the entire left auricle are not included in the volume estimate. Therefore, the calculated volumes used in our study may not provide substantial additional information, compared with area measurements, and may actually be less reliable as the result of amplification of small measurement errors.
The timing of LA measurements in horses is often not specified in the literature, leading to uncertainties when comparing results from different studies. According to the recommendations of the American Society and the European Association of Echocardiography, we measured the LA size at the end of the ventricular systole, 1 frame prior to mitral valve opening, when the LA chamber was at its greatest dimension.3 To account for potential differences in body size, measures of LA size were indexed to the dimensions of the aortic root according to the principles of allometric scaling, assuming that the aortic size correlates with body mass.23,41 We did not attempt to prove that these aortic ratios were actually independent of body size. Rather, we intended to investigate the influence on test reliability when indexing a variable to another independently measured reference variable.
Atrial mechanical function is determined by the complex interrelation between atrial loading conditions, atrial and ventricular inotropic state, the rate and extent of atrial and ventricular relaxation, atrial and ventricular compliance, heart rate, and intra-atrial electrical conduction.2 Noninvasive echocardiographic variables are unlikely to allow distinct separation and quantification of the individual factors causing alterations in LA mechanical function. Rather, they provide an overall assessment of the resulting atrial mechanical performance.2 Nevertheless, atrial mechanical function can be divided into 3 distinct phases that, to a certain degree, can be assessed separately by means of echocardiography.1,2,15 During ventricular systole, the atria act as a reservoir to receive blood from venous return. During early and mid-diastole, the atria serve as a conduit, allowing emptying of the stored blood into the ventricles and continued passage of venous return from the large veins into the ventricles. During late diastole, the atria actively contract and serve as boosterpumps that establish the final ventricular end-diastolic volume.2,42-44
In our study, reservoir function was assessed by use of the area- and volume-based LA reservoir index.15,17 Conduit function was not specifically assessed because of the lack of accurate methods for volumetric measurements of LA inflow and outflow during diastole.15 Hence, most variables used in our study were related to active LA pump function, including 2-dimensional ejection-phase variables, PW Doppler variables of transmitral flow velocity, and tissue Doppler variables of LA wall motion. Because of technical limitations and anatomic restrictions leading to poor alignment with blood flow and insufficient image quality, we elected not to investigate pulmonary venous flow and LA appendage flow profiles. Also, tissue Doppler-based strain and strain rate, which have been used in humans to quantify LA mechanical function, were not studied because the image quality and the atrial wall thickness were insufficient to properly derive these variables from the radial (as opposed to longitudinal) velocity data recorded in the imaging planes available in horses.17,45
Two-dimensional indices of LA mechanical function are easily calculated from linear, area, or volume measurements of LA dimensions at different time points during the cardiac cycle.15,27,46 The active fractional shortening, fractional area change, and emptying fraction, respectively, directly quantify the active booster-pump function, while the active-to-total area change and volume change ratios relate the active emptying to the total emptying.15
Transmitral blood flow velocities, determined by PW Doppler echocardiography, are commonly used for assessment of LV diastolic function and evaluation of LV filling pressures.47 However, the mitral flow velocity profile is also strongly influenced by LA mechanical function. Most importantly, LA active pump function (together with LV compliance) is a major determinant of the maximal transmitral A wave velocity.15 The velocity time integral of the A wave correlates well with LA stroke volume in people,2 whereas the mitral ratios of peak early to late diastolic filling velocity (peak E wave velocity-to-peak A wave velocity ratio and velocity-time integral E wave–to–velocity-time integral A wave ratio) and the fractional active emptying (fractional active emptying velocity, and fractional active emptying velocity-time integral) provide information on relative contribution of atrial pump function to ventricular filling. The LA systolic time intervals (left atrial pre-ejection period, left atrial ejection time, and left atrial preejection period-to-left atrial ejection time ratio) may be useful indices of LA systolic function, with a lower left atrial pre-ejection period-to-left atrial ejection time ratio indicating improved systolic function.15,20,26,27 Similarly, time from P wave to peak A wave, acceleration time of the A wave, and A wave acceleration rate have been used to characterize atrial electromechanical activation in humans27–29 and may serve the same purpose in horses. The LA ejection force is a composite variable based on the peak A wave velocity and the mitral valve area, calculated from a linear measurement of the mitral valve diameter. The LA ejection force provides a physiologic assessment of the strength of the atrial contraction and has been widely used as a surrogate for LA systolic function in people.11,32,33,37 The major limitation of transmitral Doppler flow recordings in adult horses is the lack of adequate alignment with the direction of transmitral blood flow. However, although the absolute measurements may not be accurate, velocity ratios, time intervals, and serial comparisons of peak A wave velocity over time may still be valid, provided that they can be reliably and consistently measured.
Recently, novel TDI techniques have been used to evaluate global and regional atrial contractile function in humans by use of a variety of imaging planes and imaging modalities.17,35-37 However, the clinical value of this technique remains unclear to date. In humans, longitudinal wall motion recorded from apical imaging planes is usually studied. These views are not available in adult horses. Therefore, we elected to study radial motion of the LA free wall in a right parasternal long- and short-axis view by use of PW and color TDI techniques, assuming that peak positive wall motion velocity and peak negative wall motion velocity directly reflect atrial contraction and relaxation, whereas TDI time intervals reflect global atrial systolic function in a similar fashion as systolic time intervals derived from transmitral Doppler flow profiles.
Similar to other diagnostic methods, echocardiographic measurements are subject to different sources of variability. Normal biological variability can affect echocardiographic recordings in the short term (ie, beat to beat) and long term (ie, hour to hour or day to day) and is influenced by physiologic effects, environmental factors leading to stress responses, and behavioral reactions.48 Recording variability is caused by differences within and between observers in transducer placement, imaging planes, and machine settings. Measurement variability largely depends on image quality and is related to the ability to identify anatomic landmarks (for spacial measurements) and temporal events (for measurement of time intervals), machine settings, minimal resolution of measurements provided by the software, and adherence of operators to measurement guidelines. With contemporary equipment, calibration error is largely eliminated by the ability to store and analyze raw data in digital format.
In our study, we attempted to minimize the sources of variability by performing the examinations in a quiet environment, by use of high-quality echocardiographic equipment with digital raw data storage, standardizing machine settings for recording and measurements, implementing strict imaging and measurement guidelines, and optimizing image planes to achieve adequate image quality.g However, to avoid selection bias, we deliberately refrained from selecting small-frame young horses in athletic condition that were more likely to provide ideal images. Instead, our study population consisted of untrained horses of typical size and varying body condition that were more likely to represent a typical patient population.
The reliability of a diagnostic test is the extent to which the test yields the same results on repeated trials.8 Hence, test reliability is directly related to the variability of the test results. Reliability data are important to assess the usefulness of echocardiographic variables to measure serial changes within an individual horse or differences between small groups of horses related to disease progression, treatment, age, exercise, training, or other factors. Many different methods have been used to quantify reliability of echocardiographic measurements. Unfortunately, no general consensus exists on how to assess reliability of echocardiographic variables. Results of different studies may vary depending on study design and statistical analyses. Furthermore, the terminology and format used to report results are not standardized. Therefore, direct comparison between studies has been difficult. In our study, we quantified reliability by calculating the SD for repeated measurements on the same horses.38 Results were then reported in 2 ways, as within-subjects CV and as the absolute value below which the difference between 2 measurements will lie with 95% probability.38 The CV has the advantage that it is standardized to the mean and therefore independent of the absolute values and the units of measurement. This feature simplifies comparison between different variables within a study. In our study, echocardiographic variables and indices were classified on the basis of the magnitude of the CV. However, for clinical use, rigid cutoffs may be misleading, as they do not account for the magnitude of changes that are seen in clinical practice and that are thought to be clinically relevant. Therefore, we also provided absolute values, below which the differences between 2 measurements will lie with 95% probability if caused by normal variability. These values allow, on a case-by-case basis, comparison with measured changes in echocardiographic variables to decide whether the observations represent true changes or normal variability.
The intraobserver and interobserver measurement variability was determined to assess the ability of observers to reliably measure the respective variables. In general, echocardiographic variables are unlikely to be useful if image quality and image planes do not allow reliable measurement of variables. Our results indicated that most variables had a low to very low measurement variability, with the exception of the 2-dimensional variables of LA mechanical function that were characterized by low to moderate variability (area- and volume-based variables) and high variability (fractional shortening).
Determination of the within-day interobserver variability allowed assessment of the error introduced by a second operator performing the examination while minimizing (but not excluding) the biological variability (by performing the examinations consecutively on the same day) and measurement error (by having 1 observer measure all recordings). For a variable to be clinically useful, imaging guidelines and adequate training should allow 2 independent observers to record standardized images of adequate quality to reliably measure the variable. Our results indicated that this was true for most variables used. Overall, variables of LA size were characterized by a very low to low recording variability, and most variables of LA mechanical function were characterized by a low to moderate recording variability. Exceptions were the LA fractional shortening and peak positive wall motion velocity determined by color TDI from a right parasternal long-axis view.
Finally, we determined the between-day intraobserver and interobserver variability, respectively, to assess the day-to-day variability, including biological, recording, and measurement variability. These 2 measures of test reliability may be most useful for clinical applications. In general, true alterations in echocardiographic variables caused by disease or interventions in an individual patient would have to be larger than the potential changes caused by day-to-day variability. Because of differences in patient population, equipment, observer experience, implementation of imaging guidelines, and image quality, data of our study may not be directly applicable to all clinical situations. However, in situations where 1 experienced operator performs serial echocardiographic examinations on the same patient by use of contemporary equipment and adequate imaging guidelines, the data for the intraobserver between-day variability may be helpful as a guide to decide whether observed alterations in echocardiographic variables represent true changes. In instances where 2 observers perform serial examinations on the same horse, it is generally advisable that 1 observer measures both recordings for direct comparison of variables.48 In this situation, data for the interobserver between-day variability can be used, provided that both observers are equally experienced and trained, use the same equipment, and follow standardized imaging guidelines. For many less standardized situations, variability of echocardiographic measurements may be higher than reported here.48
Our results indicate that LA diameter, length, area, and volume can be reliably measured from a right parasternal long- and short-axis view, respectively. The 2 linear measurements performed in the short-axis plane may be somewhat redundant to the diameter measured in long-axis, as they are oriented in a similar direction. Reliability data do not suggest superior performance of the short-axis measurements over the long-axis measurements or vice versa. Therefore, measurements during routine examination may be limited to LA diameter and length from a long-axis view. However, it will have to be determined in future studies whether one of the short-axis measurements is superior in horses with a dilated left atrium. In addition to the traditional linear measurements, we suggest measuring the cross-sectional LA area in a right parasternal long- and short-axis view to assess changes of LA size in horses. Calculation of the LA volume on the basis of a single-plane method may not provide substantial additional information and might actually be slightly less reliable, compared with the long-axis area measurements. Indexed measures of LA size were generally more variable than the native measurements, most likely as a result of the cumulative errors of the 2 independent measurements. Therefore, correction to body weight (or body weight raised to the nth power) may be preferable because indexing to a constant would not affect reliability of the variable.23,41,49
The area- and volume-based 2-dimensional indices of LA reservoir function and LA pump function performed similarly in regards to reliability of measurements, while all components of the LA fractional shortening were highly variable. Following our recommendations for assessment of LA size, we suggest using the area-based variables for assessment of LA reservoir function and LA pump function in horses. The active fractional area change and the active-to-total area change ratio were characterized by moderate day-to-day variability and may therefore not be very reliable indicators to detect minor alterations in LA pump function. The total fractional area change was characterized by low variability and may therefore be a useful index for clinical applications, although it does not allow separation of the active pump function from reservoir and conduit function.
Variables derived from maximal transmitral flow velocities appeared to be slightly more reliable than integral variables. Overall, our results revealed a moderate to high day-to-day variability for these variables, suggesting that their accuracy to detect minor changes in individual patients may be limited. Among the atrial systolic time intervals, left atrial pre-ejection period-to-left atrial ejection time ratio, left atrial ejection time, acceleration time of the A wave, and time from P wave to peak A wave had low to moderate variability, indicating a potential clinical value for assessment of LA mechanical function in horses. Conversely, the use of the A wave acceleration rate is hampered by the high between-day variability. Similarly, the LA ejection force may not be useful for detection and serial evaluation of LA mechanical dysfunction because of its high intra-observer between-day variability.
Generally, recordings of adequate quality could be recorded in long- and short-axis views and with PW and color TDI methods. The exception was the color TDI short-axis recordings that often did not allow distinct identification of the beginning and the end of the wall motion velocity wave, in which case time from P wave to onset of Am wave and duration of Am wave could not be measured. The results indicated poor day-to-day reliability for all velocity measures. Therefore, in individual patients, wall motion velocity may at best serve to detect presence or absence of wall motion but probably will not allow accurate quantification of minor changes in velocity over time. Among the TDI-derived time intervals, reliability was adequate only for the PW and color TDI measurements in right parasternal long-axis views (disregarding the short-axis color TDI measurements that were affected by selection bias). We believe that these time intervals may be a clinically useful tool for quantifying atrial electromechanical activation in horses.34 However, it remains unknown how much additional information is gained by measuring TDI time intervals in addition to 2-dimensional and transmitral flow velocity variables.
In conclusion, we were able to determine that LA size and LA mechanical function can be evaluated noninvasively by use of 2-dimensional echocardiography, transmitral flow velocity profiles, and TDI-based wall motion analysis. Our imaging guidelines and recommendations will have to be validated in clinical practice. Further studies will be required to establish reference range values, considering potential differences related to breed, age, sex, body weight, and athletic condition. Finally, and most importantly, the clinical value of these variables to assess disease-related alterations in LA function and their relation to severity of disease, exercise capacity, and prognosis will have to be established.
ABBREVIATIONS
| LA | Left atrial |
| TDI | Tissue Doppler imaging |
| PW | Pulsed wave |
| CV | Coefficient of variation |
Van Loon G, Deprez P, Duytschaever M, et al. Effect of experimental chronic atrial fibrillation in equines. In: Van Loon G. Atrial pacing and experimental atrial fibrillation in equines. PhD dissertation, Department of Large Animal Internal Medicine, Faculty of Veterinary Medicine, University of Gent, Belgium, 2001;161–206.
GE Vivid 7 ultrasound system, GE Medical Systems, Milwaukee, Wis.
M3S phased array transducer, GE Medical Systems, Milwaukee, Wis.
EchoPAC, version 3.1.3, GE Medical Systems, Milwaukee, Wis.
Microsoft Office Excel 2003, Microsoft Corp, Redmond, Wash.
SigmaStat, version 3.01, SPSS Inc, Chicago, Ill.
Merriam-Webster's collegiate dictionary, electronic edition, version 2.5, Springfield, Mass: Merriam-Webster Inc, 2000.
References
- 1
Stefanadis C, Dernellis J, Toutouzas P. A clinical appraisal of left atrial function. Eur Heart J 2001;22:22–36.
- 2↑
Pagel PS, Kehl F, Gare M, et al. Mechanical function of the left atrium: new insights based on analysis of pressure-volume relations and Doppler echocardiography. Anesthesiology 2003;98:975–994.
- 3↑
Lang RM, Bierig M, Devereux RB, et al. Recommendations for chamber quantification. Eur J Echocardiogr 2006;7:79–108.
- 4
Abhayaratna WP, Seward JB, Appleton CP, et al. Left atrial size: physiologic determinants and clinical applications. J Am Coll Cardiol 2006;47:2357–2363.
- 5
Rossi A, Vassanelli C. Left atrium: no longer neglected. Ital Heart J 2005;6:881–885.
- 6
Reef VB. Cardiovascular ultrasonography. In:Reef VB, ed.Equine diagnostic ultrasound. Philadelphia: WB Saunders, 1998;215–272.
- 7
Schotten U, Neuberger HR, Allessie MA. The role of atrial dilatation in the domestication of atrial fibrillation. Prog Biophys Mol Biol 2003;82:151–162.
- 8↑
Allessie M, Ausma J, Schotten U. Electrical, contractile and structural remodeling during atrial fibrillation. Cardiovasc Res 2002;54:230–246.
- 9
Schotten U, Duytschaever M, Ausma J, et al. Electrical and contractile remodeling during the first days of atrial fibrillation go hand in hand. Circulation 2003;107:1433–1439.
- 10
Trikas AG, Tentolouris KA, Tousoulis D, et al. Left atrial function under exercise. Eur Heart J Suppl 2000;2:K34–K37.
- 11
Manning WJ, Silverman DI, Katz SE, et al. Atrial ejection force: a noninvasive assessment of atrial systolic function. J Am Coll Cardiol 1993;22:221–225.
- 12
Ökçün B, Yigit Z, Arat A, et al. Stunning of the left atrium after conversion of atrial fibrillation: predictor for maintenance of sinus rhythm? Echocardiography 2005;22:402–407.
- 13
Harjai KJ, Mobarek SK, Cheirif J, et al. Clinical variables affecting recovery of left atrial mechanical function after cardioversion from atrial fibrillation. J Am Coll Cardiol 1997;30:481–486.
- 14
Hoit BD, Shao Y, Gabel M, et al. In vivo assessment of left atrial contractile performance in normal and pathological conditions using a time-varying elastance model. Circulation 1994;89:1829–1838.
- 15↑
Piotrowski G, Goch A, Wlazlowski R, et al. Non-invasive methods of atrial function evaluation in heart diseases. Med Sci Monit 2000;6:827–839.
- 16
Rishniw M, Erb HN. Evaluation of four 2-dimensional echocardiographic methods of assessing left atrial size in dogs. J Vet Intern Med 2000;14:429–435.
- 17
Di Salvo G, Caso P, LoPiccolo R, et al. Atrial myocardial deformation properties predict maintenance of sinus rhythm after external cardioversion of recent-onset lone atrial fibrillation. A color Doppler myocardial imaging and transthoracic and transesophageal echocardiographic study. Circulation 2005;112:387–395.
- 18
Hansson K, Haggstrom J, Kvart C, et al. Left atrial to aortic root indices using two-dimensional and M-mode echocardiography in cavalier King Charles spaniels with and without left atrial enlargement. Vet Radiol Ultrasound 2002;43:568–575.
- 19
Okamoto M, Tsubokura T, Tsuchioka Y, et al. Determinants of left atrial systolic time intervals—assessment by pulsed Doppler echocardiography. Jpn Circ J 1991;55:232–237.
- 20
Okamoto M, Tsubokura T, Morishita K, et al. Effects of volume loading on left atrial systolic-time intervals. J Clin Ultrasound 1991;19:405–411.
- 21↑
Patteson M. Two-dimensional and M-mode echocardiography. In:Marr CM, ed.Cardiology of the horse. London: WB Saunders, 1999;93–116.
- 22
Boon JA. Evaluation of size, function, and hemodynamics. In:Boon JA, ed.Manual of veterinary echocardiography. Baltimore: The Williams & Wilkins Co, 1998;151–260.
- 23
Brown DJ, Rush JE, MacGregor J, et al. M-mode echocardiographic ratio indices in normal dogs, cats, and horses: a novel quantitative method. J Vet Intern Med 2003;17:653–662.
- 24
Reef VB, Lalezari K, DeBoo J, et al. Pulsed-wave Doppler evaluation of intracardiac blood flow in 30 clinically normal Standardbred horses. Am J Vet Res 1989;50:75–83.
- 25
Blissitt KJ, Bonagura JD. Pulsed wave Doppler echocardiography in normal horses. Equine Vet J Suppl 1995;19:38–46.
- 26
Abe H, Yokouchi M, Deguchi F, et al. Measurement of left atrial systolic time intervals in hypertensive patients using Doppler echocardiography: relation to fourth heart sound and left ventricular wall thickness. J Am Coll Cardiol 1988;11:800–805.
- 27
Wang YP, Takenaka K, Sakamoto T, et al. Effects of volume loading, propranolol, and heart rate changes on pump function and systolic time intervals of the left atrium in open-chest dogs. Acta Cardiol 1993;48:245–262.
- 28
Mizuno R, Fujimoto S, Nakano H, et al. Atrial conduction abnormalities in patients with systemic progressive sclerosis. Eur Heart J 1997;18:1995–2001.
- 29
Voon WC, Sheu SH, Hwang YY. Doppler study of transmitted transmitral A wave in patients with a fourth heart sound. Echocardiography 1997;14:243–250.
- 30
Manning WJ, Leeman DE, Gotch PJ, et al. Pulsed Doppler evaluation of atrial mechanical function after electrical cardioversion of atrial fibrillation. J Am Coll Cardiol 1989;13:617–623.
- 31
Manning WJ, Silverman DI, Katz SE, et al. Impaired left atrial mechanical function after cardioversion: relation to the duration of atrial fibrillation. J Am Coll Cardiol 1994;23:1535–1540.
- 32
Mattioli AV, Castelli A, Bastia E, et al. Atrial ejection force in patients with atrial fibrillation: comparison between DC shock and pharmacological cardioversion. Pacing Clin Electrophysiol 1999;22:33–38.
- 33
Mattioli AV, Castelli A, Sternieri S, et al. Doppler sonographic evaluation of left atrial function after cardioversion of atrial fibrillation. J Ultrasound Med 1999;18:289–294.
- 34↑
Quintana M, Lindell P, Saha SK, et al. Assessment of atrial regional and global electromechanical function by tissue velocity echocardiography: a feasibility study on healthy individuals. Cardiovasc Ultrasound 2005;3: 4.
- 35
Thomas L, Levett K, Boyd A, et al. Changes in regional left atrial function with aging: evaluation by Doppler tissue imaging. Eur J Echocardiogr 2003;4:92–100.
- 36
Gurlertop Y, Yilmaz M, Acikel M, et al. Tissue Doppler properties of the left atrial appendage in patients with mitral valve disease. Echocardiography 2004;21:319–324.
- 37
Khankirawatana B, Khankirawatana S, Peterson B, et al. Peak atrial systolic mitral annular velocity by Doppler tissue reliably predicts left atrial systolic function. J Am Soc Echocardiogr 2004;17:353–360.
- 38↑
Bland M. Clinical measurement. In:Bland M, ed.An introduction to medical statistics. Oxford, England: Oxford University Press, 2000;268–293.
- 39
Tsang TSM, Abhayaratna WP, Barnes ME, et al. Prediction of cardiovascular outcomes with left atrial size: is volume superior to area or diameter? J Am Coll Cardiol 2006;47:1018–1023.
- 40
Lester SJ, Ryan EW, Schiller NB, et al. Best method in clinical practice and in research studies to determine left atrial size. Am J Cardiol 1999;84:829–832.
- 41
Sluysmans T, Colan SD. Theoretical and empirical derivation of cardiovascular allometric relationships in children. J Appl Physiol 2005;99:445–457.
- 42
Rossi A, Zardini P, Marino P. Modulation of left atrial function by ventricular filling impairment. Heart Fail Rev 2000;5:325–331.
- 43
Kihara Y, Sasayama S, Miyazaki S, et al. Role of the left atrium in adaptation of the heart to chronic mitral regurgitation in conscious dogs. Circ Res 1988;62:543–553.
- 44
Barbier P, Solomon SB, Schiller NB, et al. Left atrial relaxation and left ventricular systolic function determine left atrial reservoir function. Circulation 1999;100:427–436.
- 45
Marwick TH. Measurement of strain and strain rate by echocardiography: ready for prime time? J Am Coll Cardiol 2006;47:1313–1327.
- 46
Mattioli AV, Sansoni S, Lucchi GR, et al. Serial evaluation of left atrial dimension after cardioversion for atrial fibrillation and relation to atrial function. Am J Cardiol 2000;85:832–836.
- 47↑
Otto CM. Ventricular diastolic filling and function. In:Otto CM, ed.Textbook of clinical echocardiography. 3rd ed. Philadelphia: Elsevier Saunders, 2004;166–195.
- 48↑
Gardin JM. How reliable are serial echocardiographic measurements in detecting regression in left ventricular hypertrophy and changes in function? J Am Coll Cardiol 1999;34:1633–1636.
- 49
Cornell CC, Kittleson MD, DellaTorre P, et al. Allometric scaling of M-mode cardiac measurements in normal adult dogs. J Vet Intern Med 2004;18:311–321.
Appendix 1
Two-dimensional echocardiographic variables used for assessment of LA size and mechanical function.
Appendix 2
Two-dimensional echocardiographic variables used for assessment of LA size.
Appendix 3
Pulsed-wave Doppler variables of transmitral blood flow used for assessment of LA mechanical function.
Appendix 4
Tissue Doppler imaging variables used for assessment of LA mechanical function.
