A Depression in Left Ventricular Diastolic Filling following Prolonged Strenuous Exercise is Associated with Changes in Left Atrial Mechanics

A Depression in Left Ventricular Diastolic Filling following Prolonged Strenuous Exercise is Associated with Changes in Left Atrial Mechanics David Oxborough, MSc, Greg Whyte, PhD, Mathew Wilson, MPhil, Rory O’Hanlon, MRCPI, Karen Birch, PhD, Robert Shave, PhD, Gillian Smith, MSc, Richard Godfrey, PhD, Sanjay Prasad, MRCP, and Keith George, PhD, Leeds, Liverpool, Walsall, and London, United Kingdom

Background: Standard marathon running can result in a depression of left ventricular (LV) diastolic function during early recovery. Left atrial (LA) mechanics are integral in maintaining an early diastolic pressure gradient as well as being responsive to changes in LV diastolic function, and therefore the detailed assessment of LA mechanics may provide further insight. The aim of this study was to determine the impact of prolonged strenuous exercise on mechanics of the left atrium and the association with changes in LV diastolic function. Methods: Myocardial speckle-tracking echocardiograms of the left atrium and left ventricle were obtained prior to, immediately after, and 6 hours after the completion of a marathon (42.2 km) in 17 healthy adult men. Speckle tracking was used to determine peak atrial deformation, early diastolic deformation rate, and contractile function, including atrial activation time. LA volumes throughout the cardiac cycle were also assessed to provide reservoir, conduit, and booster pump volumes. Diastolic assessment of the left ventricle included peak early diastolic strain rate, late diastolic strain rate, and standard indices. Temporal assessment of LV ‘‘twist’’ and ‘‘untwist’’ was also evaluated. Results: All 17 subjects completed the marathon (mean finishing time, 209 6 19 minutes; range, 173-241 minutes). Although contractile function was significantly increased, there was a reduction in early diastolic deformation rate that was correlated with reduced atrial deformation. Atrial activation time was significantly increased after the race. All LV indices of diastolic function were reduced in early diastole, whereas late diastolic function was increased after the race. LV torsion was significantly reduced at end-systole and significantly elevated in the isovolumic period and early diastole, after the race. All indices returned toward baseline at 6 hours after exercise. Conclusions: This study demonstrates transient changes in LV diastolic relaxation following prolonged exercise that appear to have a direct impact on subsequent LA deformation. The impact of reduced LA preload on these findings and the delay in LA activation time requires further exploration. (J Am Soc Echocardiogr 2010;23:968-76.) Keywords: Atrial function, Diastolic function, Exercise, Strain imaging

There is growing evidence of a reduction in left ventricular (LV) systolic and diastolic function during recovery from prolonged exercise.1-4 A recent meta-analysis suggested that postexercise changes From the Faculty of Medicine and Health (D.O.) and the Faculty of Biological Science (K.B.), University of Leeds, Leeds, United Kingdom; the Research Institute for Sport and Exercise Sciences, Liverpool John Moores University, Liverpool, United Kingdom (G.W., K.G.); the Research Centre of Sport and Exercise Performance, University of Wolverhampton, Walsall, United Kingdom (M.W.); the Department of Cardiac Magnetic Resonance Imaging, Royal Brompton and Harefield NHS Trust, London, United Kingdom (R.O., G.S., S.P.); and the Centre for Sport Medicine and Human Performance, Brunel University, Uxbridge, London, United Kingdom (R.S., R.G.). Reprint requests: David Oxborough, MSc, School of Healthcare, University of Leeds, Leeds, LS2 9UT, United Kingdom (E-mail: [email protected]). 0894-7317/$36.00 Copyright 2010 by the American Society of Echocardiography. doi:10.1016/j.echo.2010.06.002

968

in LV diastolic function were consistent across exercise mode and duration.3 Whether the alteration in LV diastolic filling is due to reduced preload5 or to changes in intrinsic myocardial function4,6-8 is controversial. Surprisingly, little attention has been paid to the role of the left atrium during recovery from prolonged exercise and whether this is influenced by changes in LV diastolic filling. During systole, the left atrium acts as a reservoir, dependent on myocardial compliance, preload, and descent of the LV base, while in early diastole, it acts as a conduit delivering blood into the left ventricle while ‘‘pulling’’ blood from the pulmonary veins. In late diastole, the left atrium acts as a contractile unit to ‘‘top up’’ LV volume.9 Diastolic filling is a complex interaction involving the effects of preload, afterload, and myocardial function on both the left atrium and left ventricle. The rate and volume of flow in early diastole are driven by the rate of LV relaxation and ‘‘untwist’’ in conjunction with the rise and maintenance of pressure within the left atrium. The development of this pressure gradient between the left atrium and left ventricle during diastole is therefore an important determinant of early diastolic

Oxborough et al 969

Journal of the American Society of Echocardiography Volume 23 Number 9

function.10 Although the contribution of LV relaxation and moA = Atrial flow velocity tion to this process has been assessed,11,12 the interrogation Aact = Atrial activation time of left atrial (LA) mechanics in ASRa = Indices of response to prolonged exercise deformation during atrial would provide additional insight contraction and direction in this complex topic. ASRe = Indices of deformation during ventricular To the best of our knowledge, diastole no studies have comprehensively assessed LA mechanics folASRr = Indices of lowing prolonged strenuous deformation during ventricular exercise. In one study, the indisystole rect assessment of the left atrium, E = Early flow velocity via pulmonary venous flow interrogation, suggested reduced sucE’ = Peak early diastolic tion through the atrial conduit LA = Left atrial after prolonged exercise,7 but LAED = LA volume at endthis could reflect changes in LA diastole mechanics, an alteration in LV performance, or a combination LAES = LA volume at endof both. LV performance followsystole ing prolonged strenuous exercise LV = Left ventricular has previously been assessed using myocardial or twoLVSRa = LV late diastolic SR dimensional (2D) speckle trackLVSRe = LV early diastolic ing.2,4,12-14 This has allowed SR myocardial strain (e) and strain RV = Right ventricular rate (SR) to be measured in circumferential, radial, and RVFAC = RV fractional area longitudinal planes15 and has change facilitated the measurement of SR = Strain rate LV rotation and torsion.16-19 Recently, speckle tracking of SRa = Late SR the left atrium in providing SRe = Early SR deformation data was reported to be feasible and 2D = Two-dimensional reproducible,20-22 and thus its application in studies of prolonged exercise may be insightful. In view of (1) evidence of a consistent depression in diastolic function during recovery from marathon running, (2) evidence of the importance of LA function in LV relaxation and filling, and (3) the recent development of 2D speckle tracking to assess e and SR in the left atrium, the primary aim of this study was to assess LA mechanics during recovery from prolonged exercise. A secondary aim was to investigate the relationship between changes in LV diastolic function and alterations in LA structure or e and SR immediately and 6 hours after the completion of a marathon race. These broad aims allowed the following hypothesis to be assessed: a transient depression in LV diastolic filling following prolonged strenuous exercise is temporally associated with transient changes in LA mechanics. Abbreviations

Figure 1 Apical 4-chamber view demonstrating the speckletracking region of interest at the left atrium. disease and were not currently taking any form of prescribed medication. All subjects provided written informed consent to participate, and the study was granted ethics approval by the ethics committee of the Brompton, Harefield and the National Heart and Lung Institute. Design Echocardiography was performed approximately 24 hours prior to starting the marathon. Postrace echocardiographic studies were completed within 1 hour of race completion, and a further acquisition was made 6 hours after marathon completion. All subjects refrained from training for 48 hours and avoided alcohol and caffeine 24 hours prior to the race. During the marathon, participants were permitted to consume food and fluid ad libitum. On the day of the marathon, start times ranged from 5 to 7 AM. Maximum air temperature reached 10 C, with approximately 40% humidity at sea level. Procedures After a full explanation of procedures, subjects lay supine, and after a 5-minute resting period, the echocardiographic examination was initiated. Heart rate was taken from electrocardiography, inherent to the ultrasound system. All echocardiographic images were acquired using a commercially available ultrasound system (Vivid 7; GE Vingmed Ultrasound AS, Horten, Norway) with a 1.5-MHz to 4-MHz phased-array transducer. All images were acquired by a single experienced sonographer with the subject lying in the left lateral decubitus position and recorded to DVD in a raw Digital Imaging and Communications in Medicine format. Acquisition and analysis were performed by the same single experienced sonographer. All data were analyzed offline using commercially available software (EchoPAC version 6.0; GE Vingmed Ultrasound AS), and a minimum of 3 cardiac cycles were averaged for all peak indices. Conventional Echocardiography

METHODS Subjects Seventeen male recreational runners (mean age, 33 6 6 years; range, 26-44 years) volunteered to participate in the study and run a marathon (42.2 km). All subjects were healthy and free from known cardiovascular disease and any early family history of cardiovascular

Standard 2D, pulsed-wave Doppler, and pulsed tissue Doppler echocardiographic parameters were obtained from parasternal and apical acoustic windows. All settings were optimized to obtain maximum signal-to-noise ratio and 2D images to provide optimal endocardial delineation. LV end-diastolic dimension, LV end-diastolic volume, LV end-systolic volume, LA anterior-posterior diameter, and LA volume at end-systole (LAES) were measured in accordance with

970 Oxborough et al

Journal of the American Society of Echocardiography September 2010

Figure 2 Offline analysis demonstrating LA e from the lateral wall.

Figure 3 Offline analysis demonstrating LA SR from the lateral wall. recommendations from the American Society of Echocardiography.23 LA volume was also measured prior to atrial contraction and at end-diastole (LAED). From these data, LA reservoir volume was calculated as LAES  LAED, LA conduit volume

as the difference between LV stroke volume and LA reservoir volume, and LA booster pump volume as LA volume prior to atrial contraction  LAED, as previously described.24 All volumes were derived using Simpson’s biplane methodology, and LV ejection fraction was

Oxborough et al 971

Journal of the American Society of Echocardiography Volume 23 Number 9

calculated. LV diastolic function was assessed using pulsed-wave Doppler recordings from apical 4-chamber and 3-chamber orientations. For the assessment of diastolic function, a 4-mm sample volume was placed at the tips of the mitral leaflets in diastole, and transmitral flow was acquired to obtain peak early (E) and atrial (A) flow velocities.25 To assess isovolumic relaxation time, the 4-mm pulsed-wave sample was positioned between the LV inflow and outflow from an apical 3-chamber orientation. For the tissue Doppler assessment of E0 , the apical 4-chamber orientation was used, and a 2-mm sample volume was positioned at the septal aspect of the mitral valve annulus, ensuring the best alignment between wall motion and the ultrasound beam. The high-pass filter was bypassed, and gains were set to minimal values to obtain the best signal-to-noise ratio. The Nyquist limit was set between 10 and 35 cm/s. Peak early diastolic (E0 ) tissue myocardial velocity was recorded and the E/E0 ratio calculated. Right ventricular (RV) function was assessed using 2D images to provide a RV fractional area change (RVFAC). Two-Dimensional Myocardial Speckle Tracking For the acquisition of LA functional data, the apical 4-chamber orientation was used, with the region of interest placed around the endocardial surface of the atrial lateral wall, superior wall, and atrial septum (see Figure 1). The image was optimized with the focal point positioned at the mid atrial wall, and frame rates were maintained as high as possible, but no greater than 90 frames/s. Offline analysis allowed the assessment of peak data of the left atrium and was assessed from the lateral wall, only providing indices of deformation during ventricular systole (Ae and ASRs), during ventricular diastole (ASRe), and during atrial contraction (ASRa) (see Figures 2 and 3). Atrial activation time (Aact) was also measured and defined as the time from the onset of the electrocardiographic P wave to the onset of mechanical atrial contraction. LV radial and circumferential e and SR and rotational data were derived from a parasternal short-axis view at the base, defined as at the tips of the mitral valve. The calculation of torsion also required the acquisition of an apical short-axis view at the level immediately above the level of systolic cavity obliteration.19 The focal point was positioned close to the center of the LV cavity to provide optimal beam width while reducing the effects of divergence. Longitudinal e and SR data were obtained from an apical 4-chamber view only. The focal point was positioned at the level of the mitral valve. In both orientations, frame rates were maximized with an acceptable level >40 and <90 frames/s. All images were optimized with gain, compression, and dynamic range to enhance myocardial definition, while depth, frequency, and insonation angle were standardized from all acquisitions to reduce the impact of variability.26 For the left ventricle, parasternal short-axis segments provided data from the inferoseptum, anteroseptum, and anterior, lateral, posterior, and inferior walls. The apical 4-chamber orientation provided data from the basal, mid, and apical septum and basal, mid, and apical lateral walls. All myocardial segments were considered acceptable and included in the analysis. Peak rotation and rotation rate were obtained, and torsion was calculated as the net difference between apical and basal rotation. Peak LV circumferential, radial, and longitudinal diastolic function was assessed from early diastolic SR (LVSRe) and late diastolic SR (LVSRa) directly from the EchoPAC software. A number of data points, dependent on the maximum frame rate of the 2D image, from LV rotation and torsion and e curves were stored from the offline workstation and exported to a spreadsheet (Excel 2007; Microsoft Corporation, Redmond, WA). The spreadsheet was then used to select

Table 1 Standard 2D and Doppler indices Parameter

Before exercise

Immediately after exercise

6 hours after exercise

Heart rate 57 (53-62) 80 (74-86)*† 68 (62-73) (beats/min) E (m/s) 0.75 (0.68-0.83) 0.70 (0.65-0.75)† 0.76 (0.7-0.82) A (m/s) 0.46 (0.4-0.51) 0.67 (0.58-0.76)*† 0.54 (0.48-0.60) 1.5 (1.3-1.7) E/A 1.7 (1.5-1.9) 1.1 (0.9-1.3)*† IVRT (ms) 76 (66-86) 74 (64-82) 66 (57-75) LVd (mm) 54 (52-56) 52 (50-55)* 53 (51-55) EDV (mL) 154 (139-170) 142 (125-159) 147 (130-163) ESV (mL) 52 (46-59) 53 (45-61) 51 (45-57) EF (%) 66 (64-68) 63 (60-65) 65 (63-67) 13 (11 to15) 11 (10-13)* 12 (11-14) E0 (cm/s) 6.2 (5.4-7.1) 6.6 (5.5-7.7) 6.3 (5.7-6.9) E/E0 LAd (mm) 40 (38-42) 38 (36-41)* 40 (37-42) RVFAC (%) 49 (46-52) 44 (41-49)* 47 (43-52) EDV, End-diastolic volume; EF, ejection fraction; ESV, end-systolic volume; IVRT, isovolumic relaxation time; LAd, LA anteriorposterior diameter; LVd, LV end-diastolic dimension. Data are expressed as mean (95% confidence interval). *P < .05, immediately after versus before exercise. †P < .05, immediately after versus 6 hours after exercise.

data points from each subject in 10% increments of both systole and diastole, essentially providing 20 data points from each cardiac cycle. The mean values of all 20 data points from the 17 subjects were then calculated, allowing the graphical representation of the complete cardiac cycle, before, immediately after, and 6 hours after the marathon. Previous data collected in our laboratory on 20 subjects from two separate acquisitions 2 days apart revealed coefficients of variation for peak atrial e, ASRe, ASRa, and Aact of 8.9%, 7.3%, 6.2%, and 3.4% respectively. The coefficients of variation for peak LV radial, circumferential, and longitudinal peak early diastolic LVSRe were 5.3%, 7.1%, and 10.1% and for late diastolic LVSRa were 4.5%, 6.7%, and 8.1%, respectively.27 Statistical Analysis Echocardiographic data related to LA and LV structure and function assessed before, immediately after, and 6 hours after exercise were analyzed using repeated-measures analysis of variance and post hoc Tukey tests as appropriate. For LA and LV functional indices that were significantly altered after the race, bivariate correlational analysis was used to establish any relationship to exercise-related changes in heart rate and preload (end-diastolic volume). Further changes in LV function were correlated with changes in LA mechanics.

RESULTS All 17 subjects completed the marathon (mean finishing time, 209 6 19 minutes; range, 173-241 minutes). Postrace body mass was reduced (80 6 9.2 to 78.8 6 8.6 kg, P < .001), but heart rate was increased (57 6 8 to 80 6 12 beats/min, P < .001) and remained elevated 6 hours after the marathon (68 6 11 beats/min, P < .001). LV end-diastolic volume was reduced after the race but did not reach statistical significance (154 6 30 to 142 6 33 mL, P = .07; 146 6 32 mL at 6 hours).

972 Oxborough et al

Journal of the American Society of Echocardiography September 2010

Table 2 LA functional data Parameter

Before exercise

Immediately after exercise

6 hours after exercise

Aact (TS-onset) (ms) Aact (TS-peak) (ms) Ae (%) ASRs (L/s) ASRe (L/s) ASRa (L/s) LAES (mL) LApreA (mL) LAED (mL) Reservoir volume (mL) Conduit volume (mL) Booster volume (mL)

33.5 (25.6 to 41.4) 108.8 (97.9 to 119.7) 53.1 (42.8 to 63.2) 2.79 (2.35 to 3.25) 4.37 (5.12 to 3.63) 2.95 (3.56 to 2.33) 65 (43 to 79) 47 (32 to 64) 37 (26 to 48) 28 (16 to 37) 74 (50 to 112) 10 (5 to 17)

53.5 (42.4 to 64.6)* 118.3 (106.3 to 130.2) 44.2 (37.1 to 51.1)* 2.62 (2.15 to 3.09) 3.24 (3.84 to 2.64)* 3.24 (4.02 to 2.54)* 57 (43 to 76)* 49 (37 to 65) 35 (27 to 46)* 23 (10 to 47)* 66 (36 to 103)* 14 (4 to 29)*

51.1 (41.5 to 60.8)† 114.2 (104.7 to 123.6) 51.7 (43.1 to 60.4)† 2.77 (2.24 to 3.3) 3.90 (4.66 to 3.14) 2.88 (3.66 to 2.10) 67 (45 to 85)‡ 49 (35 to 67) 38 (28 to 48)‡ 29 (14 to 47)‡ 66 (35 to 103) 11 (4 to 25)

LApreA, LA volume prior to atrial contraction. Data are expressed as mean (95% confidence interval). *P < .05, immediately after versus before exercise. †P < .05, before versus 6 hours after exercise. ‡P < .05, immediately after versus 6 hours after exercise.

Conventional Echocardiography All standard 2D and Doppler findings are presented in Table 1. Ejection fraction was not significantly altered after the race. Transmitral E/A ratio was reduced by 35% after the race as a consequence of reduction in peak E velocity and a compensatory increase in peak A velocity. The ratio had not completely returned to baseline at 6 hours. There was no significant change in the isovolumic relaxation time or early diastolic deceleration time. Early diastolic annular velocity (E0 ) decreased slightly (P < .05), and thus E/E0 was also unchanged. RVFAC was reduced by 9% after the race but had returned to the baseline level at 6 hours following race completion. LA Deformation and Volumes Peak Ae and peak ASRe were reduced after the race (17% and 26%, respectively), with both variables returning to baseline at 6 hours after exercise. There was a 10% increase in atrial contractility rate immediately after the race, which returned to baseline at 6 hours after exercise. The 38% increase in Aact after the race remained present at 6 hours after the race. LA anterior-posterior diameter, LAES, and LAED were significantly reduced after the race (5%, 13%, and 5%, respectively) but had returned to baseline levels at 6 hours after the race. Reservoir and conduit volume also reduced after the race by 18% and 11%, respectively, while atrial booster pump volume increased by 40% (see Table 2). LV Torsion and e Diastolic Function Torsion remained significantly elevated in early diastole (20%, 30%, and 40% of the diastolic interval) when postrace data were compared with prerace, data suggesting a reduced rate of diastolic untwisting. This was confirmed by basal and apical LV rotation data, which were still elevated during early diastole after the race (see Figure 4). All values returned close to baseline at 6 hours after the race. Changes in peak LV e and SR data (see Table 3) provide support for the torsion and untwist data. Peak LVSRe was reduced in all planes of relaxation. Circumferential, radial, and longitudinal SRe/SRa was significantly reduced after the race as a result of reduced LVSRe and an

increased LVSRa. Peak circumferential and longitudinal early diastolic SR occurred 10% later in diastole immediately after the race. At 6 hours after the race, this had returned to baseline. Regional data for radial and circumferential SRe/SRa were reduced from all myocardial segments, but longitudinal SRe was significantly reduced in the septum but not the lateral wall. Interestingly, SRa increased in both septal and lateral walls, causing a concomitant reduction in SRe/SRa at both sites after the race. Correlation Analysis The depression in Doppler E/A, circumferential SRe/SRa, and radial SRe/SRa could be partially explained by an increased heart rate at the postrace assessment (r = 0.67, P = .003; r = 0.76, P < .001; and r = 0.49, P = .046, respectively). Peak Ae correlated with LA reservoir volume (r = 0.47, P = .045), while ASRa correlated with LA booster pump volume (r = 0.57, P = .032). The passive deformation of the atrial myocardium (ASRe) demonstrated a moderate correlation with the absolute e value (Ae) (r = 0.58, P = .014). The change in Doppler E/A ratio was also correlated with postrace changes in LAES and reservoir volume (r = 0.61, P = .009, and r = 0.57, P = .017, respectively). The changes in circumferential SRe/SRa correlated significantly with LAES and reservoir volume (r = 0.68, P = .003, and r = 0.73, P = .001, respectively). Changes in RVFAC correlated with longitudinal SRe/SRa (r = 0.54, P = .024). With regard to LV function, the small changes in peak e from each myocardial plane did not correlate with changes in the SRe/ SRa ratio from the same muscle layer; longitudinal (r = 0.074, P > .05), radial (r = 0.10, P > .05), and circumferential (r = 0.09, P > .05). This supports the interdependence of these diastolic parameters and the possible physiologic nature of these changes.

DISCUSSION This is the first study of its kind to investigate, in depth, the relationship between LA structure and deformation with LV diastolic function following prolonged exercise. The key data suggest postexercise reductions in both LA and LV deformation characteristics during early

Journal of the American Society of Echocardiography Volume 23 Number 9

Figure 4 Temporal distribution of torsion and rotation from the LV base and apex.

diastole. LV filling during early diastole is generated by an adequate pressure gradient,28 with factors that reduce either LA pressure or delay the drop in LV pressure influencing the velocity and propagation of blood into the left ventricle. A reduced pressure gradient in early diastole following prolonged exercise has been demonstrated previously using standard Doppler indices and flow propagation velocity.5,29 Our data support these changes in LV diastolic function

Oxborough et al 973

(extending to e, SRs, torsion, and untwist) and further add complex analysis of LA mechanics. After the race, we observed a reduction in peak LA early diastolic deformation (ASRe) as well as a reduction in parameters of LV early diastolic function. This is further supported with a reduced LA conduit volume after exercise. The gradient to the left ventricle is directly related to the rate of LV relaxation (suction), and in view of this, it is very likely that reduced LV relaxation has a direct impact on the observed reduction in LA early diastolic deformation. The forward pressure gradient is also driven by the rate of untwisting of circumferential fibers during the isovolumic relaxation period and the initial stages of early diastole.10 This study demonstrates a reduced rate of untwist during the early stages of diastole following slightly reduced peak systolic torsion. LV relaxation during this period of the cardiac cycle is both a passive and an active process, dependent on generated systolic tension30 and, at a cellular level, the ability for Ca2+ release from the troponin complex and subsequent reuptake into the sarcoplasmic reticulum.31 The isovolumic relaxation phase is independent of LA function, and consistent with findings presented by Nottin et al,12 this suggests that a delay in untwist is more likely a consequence of an intrinsic reduction in LV relaxation. Peak positive LA deformation (Ae) during ventricular systole was also significantly reduced after the race and was complemented by a reduction in LA reservoir volume. LA distension is dependent on a number of factors: (1) LA relaxation, (2) LV longitudinal contraction and hence rate of the descent of the base, (3) LA compliance or chamber stiffness, and (4) RV systole through pulmonary venous inflow and hence LA preload.32 It is thought that LA filling is initially brought about by relaxation of the myocardial fibers, and this relaxation is dependent on the rate of myocardial lengthening and the extent of shortening from the previous atrial contraction.33 We observed no change in the rate of LA myocardial relaxation (ASRs) but demonstrated increases in contractility (ASRa) and booster pump volume. This increase in contractility may be a consequence of reduced LA afterload, secondary to reduced LV filling, and therefore both these findings support more of a change in LA distension rather than a reduction in intrinsic atrial myocardial relaxation. Other factors influencing LA distension (Ae) were also assessed. Compliance and stiffness of the left atrium are determined by the descent of the LV base,32 and this was reduced when measured using peak LV longitudinal e (18 to 15.1 after the race, P < .05). On the basis of this and a positive correlation between ASRe and Ae, we can speculate that the impact of LV function on LA distension also influences the subsequent LA early diastolic deformation. RV stroke volume may be reduced when pulmonary pressures are elevated,34 in the presence of impaired RV myocardial contractility or when there is a reduction in total blood volume and preload. A reduction in RVFAC was observed, and we can assume that this is also a potential factor in the generation of a reduced LA volume and reservoir volume in this study. In view of the limitations of echocardiographic assessment of RV function, it is apparent that a more comprehensive and quantitative assessment is required, warranting further research. Changes in RVFAC correlated with longitudinal diastolic indices of LV function, and this may be further evidence of a series or parallel impact on LV diastolic filling. This relationship between LA filling and LV filling is also confirmed by a positive correlation of transmitral E/A ratio and LA volumes at end-systole and reservoir volume. Interestingly, there was a lack of correlation of these LA volume indices with LV volume. Other studies have demonstrated reductions

974 Oxborough et al

Journal of the American Society of Echocardiography September 2010

Table 3 Peak LV e and SR data Parameter

Before exercise

Immediately after exercise

6 hours after exercise

Circ e Circ SRs (L/s) Circ SRe (L/s) Circ SRa (L/s) Circ SRe/SRa Circ basal rotation ( ) Circ apical rotation ( ) Circ torsion ( ) Circ torsion/length ( /cm) Circ basal untwist rate ( /s) Circ apical untwist rate ( /s) Rad e Rad SRs (L/s) Rad SRe (L/s) Rad SRa (L/s) Rad SRe/SRa Long e Long SRs (L/s) Long SRe (L/s) Long SRa (L/s) Long SRe/SRa

20.6 (22 to 19.2) 1.37 (1.47 to 1.27) 1.9 (1.73 to 2.06) 0.64 (0.54 to 0.74) 3.2 (2.7 to 3.6) 4.9 (6.1 to 3.8) 7.3 (5.7 to 9.5) 10.9 (8.7 to 11.4) 1.29 (1.08 to 1.51) 68.9 (57.9 to 79) 54.8 (67.9 to 41.7) 53.6 (45.5 to 61.6) 1.7 (1.53 to 1.86) 1.77 (1.98 to 1.56) 0.821 (1.05 to 0.58) 2.8 (2 to 3.7) 18.5 (19.5 to 17.4) 1.07 (1.13 to 1.01) 1.02 (0.89 to 1.11) 0.42 (0.36 to 0.52) 1.9 (1.6 to 2.2)

18.5 (19.8 to 17.2)† 1.41 (1.49 to 1.32) 1.68 (1.5 to 1.86)† 0.96 (0.78 to 1.13)* 2 (1.6 to 2.5)*† 5.5 (6.8 to 4.2) 7.6 (6.3 to 8.9) 9.8 (7.2 to 10.7) 1.36 (1.16 to 1.56) 66.4 (55.3 to 77.4)† 54.4 (65.3 to 43.5) 45.7 (37.9 to 53.4)* 1.79 (1.61 to 1.98) 1.79 (2.1 to 1.48) 1.45 (1.76 to 1.14)*† 1.4 (1.1 to 1.7)*† 15.1 (16.2 to 13.2)* 1.1 (1.17 to 1.02) 0.95 (0.88 to 1.07) 0.65 (0.50 to 0.77)* 1.4 (1.2 to 1.6)*

20.7 (22.2 to 19.2) 1.48 (1.56 to 1.41) 1.98 (1.82 to 2.16) 0.73 (0.59 to 0.87) 3.1 (2.4 to 3.8) 5.6 (6.5 to 4.8) 7.1 (5.5 to 8.6) 10.5 (8.9 to 11.1) 1.30 (1.1 to 1.5) 77.9 (66.9 to 88.9) 50 (57.2 to 42.8) 52.3 (45.9 to 58.7) 1.84 (1.7 to 2) 1.92 (2.2 to 1.64) 1.05 (1.28 to 0.83) 2.2 (1.6 to 2.8) 17.9 (18.9 to 16.8) 1.15 (1.23 to 1.06) 1.01 (0.86 to 1.06) 0.56 (0.46 to 0.63) 1.6 (1.4 to 1.9)

Circ, circumferential; long, longitudinal; rad, radial. Data are expressed as mean (95% confidence interval). *P < .05, immediately after versus before exercise. †P < .05, immediately after versus 6 hours after exercise.

in RV diastolic35 and systolic2 function following prolonged exercise, whereas other investigators have speculated that changes are due to a reduction in total blood volume.36 It is important to note that Aact increased after exercise, whereas one would expect a reduction with higher heart rates during recovery, particularly in view of reduced LAES.37 We also observed an increase in atrial contractility, and although this could be a consequence of reduced atrial afterload secondary to reduced LV filling, it could also be caused by an increase in intrinsic atrial myocardial contractility. It is difficult to speculate the mechanism(s) for this increase in activation time, but with a recent publication demonstrating delayed electromechanical association in the left ventricle following prolonged exercise,38 these findings may suggest some degree of intrinsic reduction in mechanical activation, although there is limited support from any of the other atrial indices. It is also important to discuss the positive correlations of some of the diastolic indices with heart rate. These indicate that some of the changes may be partially explained by a reduced diastolic filling time. None of the LA deformation parameters correlate with heart rate, suggesting that the influence of heart rate is only a small part of the physiologic process. At this stage, it is also relevant to highlight the changes observed by Giannaki et al,39 who observed an increase in E0 velocity with an increased heart rate in recovery from exercise, the opposite of what we observed in this study. Clinical Implications The assessment of diastolic function is recommended on standard clinical echocardiography,40 particularly with many patients suffering from clinical heart failure having normal ejection fractions.41 This study highlights the importance of considering the effect of LV relax-

ation on LA filling and also the potential impact of reduced preload. A reduced E/A ratio is often seen and assumed to be a consequence of primarily reduced LV relaxation. It is important not to overlook the contribution of LA preload, particularly in patients in a hypovolemic state or when there is suspicion of reduced RV stroke volume, both of which could occur in a postexercise setting. This study supports the findings of Kim et al20 in the feasibility of assessing LA function using myocardial speckle tracking and equally its ability to provide comprehensive data related to filling of the left ventricle. This technique has been demonstrated when applied to the assessment of longitudinal function of the right ventricle, although in limited applications.42 It is likely that to enhance our understanding of the impact of prolonged exercise, future studies should apply this to discern the nature of any reductions in preload. The findings of this study also support Nottin et al12 in the belief that intrinsic changes in LV relaxation, independent of changes in preload, occur during the recovery from prolonged strenuous exercise. The acute impact on health from running a marathon is not known, but previous studies have suggested a persistent change in function when assessed 3 to 4 weeks after the race.4 In view of the small transient changes observed in this study, we are unable to support Neilan et al4 in their findings on the adverse long-term effects of marathon running. Limitations This study was carried out in a relatively small sample of only male subjects completing a marathon in a narrow range of finishing times. Therefore, further work may aim to assess the impact of gender and training levels on the magnitude of observed changes. The small sample size may also have an impact on the statistical power of some of the relationships.

Journal of the American Society of Echocardiography Volume 23 Number 9

We have attempted to use quantitative echocardiography to assess LA mechanics, including stiffness and to some degree function. It is apparent that a comprehensive assessment would be most accurate during diastasis, when the left atrium and left ventricle are not interacting. With postrace heart rates of approximately 80 beats/min, the diastatic period was largely absent, and unfortunately this specific analysis was not feasible in this study. Also the use of echocardiography to establish a noninvasive measure of pressure-volume relationship for the left atrium would possibly provide further functional value and would therefore benefit from exploration in future work. Our primary aim was to assess the relationship between LA mechanics and LV diastolic function, and therefore the assessment of RV function was constrained to the use of standard 2D echocardiography. It is clear that these findings have highlighted the need to provide a comprehensive assessment of the right ventricle, in this way assessing any role it has on LA and LV dysfunction after prolonged exercise.

CONCLUSIONS This study demonstrated, for the first time, a reduction in LA deformation during filling while demonstrating increased deformation during atrial contraction following prolonged exercise. Changes in LV diastolic function following prolonged strenuous exercise appear to have a direct impact on both LA diastolic and systolic peak deformation. It is also apparent that reduced LA preload may also influence this process. Reduced atrial preload can be directly related to RV stroke volume and pulmonary artery pressure, and further studies are required to establish whether this is a consequence of reduced central blood volume and preload or, similar to the left ventricle, intrinsic reductions in RV function. REFERENCES 1. Douglas PS, O’Toole ML, Hiller WD, Hackney K, Reichek N. Cardiac fatigue after prolonged exercise. Circulation 1987;76:1206-13. 2. La Gerche A, Connelly KA, Mooney DJ, MacIsaac AI, Prior DL. Biochemical and functional abnormalities of left and right ventricular function after ultra-endurance exercise. Heart 2008;94:860-6. 3. Middleton N, Shave R, George K, Whyte G, Hart E, Atkinson G. Left ventricular function immediately following prolonged exercise: a meta-analysis. Med Sci Sports Exerc 2006;38:681-7. 4. Neilan TG, Yoerger DM, Douglas PS, Marshall JE, Halpern EF, Lawlor D, et al. Persistent and reversible cardiac dysfunction among amateur marathon runners. Eur Heart J 2006;27:1079-84. 5. Hart E, Shave R, Middleton N, George K, Whyte G, Oxborough D. Effect of preload augmentation on pulsed wave and tissue Doppler echocardiographic indices of diastolic function after a marathon. J Am Soc Echocardiogr 2007;20:1393-9. 6. Douglas PS, O’Toole ML, Woolard J. Regional wall motion abnormalities after prolonged exercise in the normal left ventricle. Circulation 1990; 82:2108-14. 7. George K, Oxborough D, Forster J, Whyte G, Shave R, Dawson E, et al. Mitral annular myocardial velocity assessment of segmental left ventricular diastolic function after prolonged exercise in humans. J Physiol 2005;569: 305-13. 8. George K, Shave R, Oxborough D, Whyte G, Dawson E. Longitudinal and radial systolic myocardial tissue velocities after prolonged exercise. Appl Physiol Nutr Metab 2006;31:256-60. 9. Abhayaratna WP, Seward JB, Appleton CP, Douglas PS, Oh JK, Tajik AJ, et al. Left atrial size: physiologic determinants and clinical applications. J Am Coll Cardiol 2006;47:2357-63.

Oxborough et al 975

10. Burns AT, La Gerche A, Prior DL, MacIsaac AI. Left ventricular untwisting is an important determinant of early diastolic function. JACC Cardiovasc Img 2009;2:709-16. 11. Esch BT, Scott JM, Warburton DER, Thompson R, Taylor D, Baron JC, et al. Left ventricular torsion and untwisting during exercise in heart transplant recipients. J Physiol 2009;587:2375-86. 12. Nottin S, Doucende G, Schuster I, Tanguy S, Dauzat M, Obert P. Alteration in left ventricular strains and torsional mechanics after ultralong duration exercise in athletes. Circ Cardiovasc Imaging 2009;2: 323-30. 13. Dawson EA, Whyte GP, Black MA, Jones H, Hopkins N, Oxborough D, et al. Changes in vascular and cardiac function after prolonged strenuous exercise in humans. J Appl Physiol 2008;105:1562-8. 14. George K, Shave R, Oxborough D, Cable T, Dawson E, Artis N, et al. Left ventricular wall segment motion after ultra-endurance exercise in humans assessed by myocardial speckle tracking. Eu J Echocardiogr 2009;10: 238-43. 15. Korinek J, Wang J, Sengupta PP, Miyazaki C, Kjaergaard J, McMahon E, et al. Two-dimensional strain—a Doppler-independent ultrasound method for quantitation of regional deformation: validation in vitro and in vivo. J Am Soc Echocardiogr 2005;18:1247-53. 16. Borg AN, Harrison JL, Argyle RA, Ray S. Left ventricular torsion in primary chronic mitral regurgitation. Heart 2008;94:597-603. 17. Helle-Valle T, Crosby J, Edvardsen T, Lyseggen E, Amundsen BH, Smith HJ, et al. New noninvasive method for assessment of left ventricular rotation: speckle tracking echocardiography. Circulation 2005;112:3149-56. 18. Notomi Y, Lysyansky P, Setser RM, Shiota T, Popovic ZB, MartinMiklovic MG, et al. Measurement of ventricular torsion by twodimensional ultrasound speckle tracking imaging. J Am Coll Cardiol 2005; 45:2034-41. 19. Opdahl A, Helle-Valle T, Remme EW, Vartdal T, Pettersen E, Lunde K, et al. Apical rotation by speckle tracking echocardiography: a simplified bedside index of left ventricular twist. J Am Soc Echocardiogr 2008;21: 1121-8. 20. Kim DG, Lee KJ, Lee S, Jeong S-Y, Lee YS, Choi YJ, et al. Feasibility of twodimensional global longitudinal strain and strain rate imaging for the assessment of left atrial function: a study in subjects with a low probability of cardiovascular disease and normal exercise capacity. Echocardiography 2009;26:1179-87. 21. Vianna-Pinton R, Carlos AM, Christy MB, Kwan SL, Teresa SMT, Christopher PA. Two-dimensional speckle-tracking echocardiography of the left atrium: feasibility and regional contraction and relaxation differences in normal subjects. J Am Soc Echocardiogr 2009;22:299-305. 22. Saraiva RM, Demirkol S, Buakhamsri A, Greenberg N, Popovic ZB, Thomas JD, et al. Left atrial strain measured by two-dimensional speckle tracking represents a new tool to evaluate left atrial function. J Am Soc Echocardiogr 2010;23:172-80. 23. Lang RM, Bierig M, Devereux RB, Flachskampf FA, Foster E, Pellikka PA, et al. Recommendations for chamber quantification: a report from the American Society of Echocardiography’s Guidelines and Standards Committee and the Chamber Quantification Writing Group, developed in conjunction with the European Association of Echocardiography, a branch of the European Society of Cardiology. J Am Soc Echocardiogr 2005;18: 1440-63. 24. Ogawa K, Hozumi T, Sugioka K, Iwata S, Otsuka R, Takagi Y, et al. Automated assessment of left atrial function from time-left atrial volume curves using a novel speckle tracking imaging method. J Am Soc Echocardiogr 2009;22:63-9. 25. Garcia MJ, Thomas JD, Klein AL. New doppler echocardiographic applications for the study of diastolic function. J Am Coll Cardiol 1998;32: 865-75. 26. Sivesgaard K, Christensen SD, Nygaard H, Hasenkam JM, Sloth E. Speckle tracking ultrasound is independent of insonation angle and gain: an in vitro investigation of agreement with sonomicrometry. J Am Soc Echocardiogr 2009;22:852-8. 27. Oxborough D, Batterham AM, Shave R, Artis N, Birch KM, Whyte G, et al. Interpretation of two-dimensional and tissue Doppler-derived strain and

976 Oxborough et al

28.

29.

30.

31. 32.

33. 34. 35.

strain rate data: is there a need to normalize for individual variability in left ventricular morphology? Eur J Echocardiogr 2009;10:677-82. Khouri SJ, Maly GT, Suh DD, Walsh TE. A practical approach to the echocardiographic evaluation of diastolic function. J Am Soc Echocardiogr 2004;17:290-7. Middleton N, Shave R, George K, Whyte G, Forster J, Oxborough D, et al. Novel application of flow propagation velocity and ischaemia-modified albumin in analysis of postexercise cardiac function in man. Exp Physiol 2006;91:511-9. Esch B, Warburton D. Left ventricular torsion and recoil: implications for exercise performance and cardiovascular disease. J Appl Physiol 2009; 106:362-9. Bers DM. Calcium fluxes involved in control of cardiac myocyte contraction. Circ Res 2000;87:275-81. Barbier P, Solomon SB, Schiller NB, Glantz SA. Left atrial relaxation and left ventricular systolic function determine left atrial reservoir function. Circulation 1999;100:427-36. Tsakiris AG, Padiyar R, Gordon DA, Lipton I. Left atrial size and geometry in the intact dog. Am J Physiol Heart Circ Physiol 1977;232:H167-72. Henning RJ. Effects of positive end-expiratory pressure on the right ventricle. J Appl Physiol 1986;61:819-26. Oxborough D, Shave R, Middleton N, Whyte G, Forster J, George K. The impact of marathon running upon ventricular function as assessed by 2D, Doppler, and tissue-Doppler echocardiography. Echocardiography 2006; 23:635-41.

Journal of the American Society of Echocardiography September 2010

36. Alshaher M, El-Mallakh R, Dawn B, Siddiqui T, Longaker RA, Stoddard MF. Cardiac manifestations of exhaustive exercise in nonathletic adults: does cardiac fatigue occur? Echocardiography 2007;24:237-42. 37. Merckx KL, De Vos CB, Palmans A, Habets J, Cheriex EC, Crijns HJGM, et al. Atrial activation time determined by transthoracic Doppler tissue imaging can be used as an estimate of the total duration of atrial electrical activation. J Am Soc Echocardiogr 2005;18:940-4. 38. Chan-Dewar F, Oxborough D, Shave R, Gregson W, Whyte G, Noakes T, et al. Evidence of increased electro-mechanical delay in the left and right ventricle following prolonged exercise. Eur J Appl Physiol 2010;108:581-7. 39. Giannaki C, Oxborough D, George K. Diastolic Doppler flow and tissueDoppler velocities during, and in recovery from, low intensity supine exercise. Appl Physiol Nutr Metab 2008;33:896-902. 40. Rakowski H, Appleton C, Chan KL, Dumesnil JG, Honos G, Jue J, et al. Canadian consensus recommendations for the measurement and reporting of diastolic dysfunction by echocardiography: from the investigators of consensus on diastolic dysfunction by echocardiography. J Am Soc Echocardiogr 1996;9:736-60. 41. Sanderson JE, Fraser AG. Systolic dysfunction in heart failure with a normal ejection fraction: echo-Doppler measurements. Prog Cardiovasc Dis 2006;49:196-206. 42. Kalogeropoulos AP, Georgiopoulou VV, Howell S, Pernetz M-A, Fisher MR, Lerakis S, et al. Evaluation of right intraventricular dyssynchrony by two-dimensional strain echocardiography in patients with pulmonary arterial hypertension. J Am Soc Echocardiogr 2008;21:1028-34.