Left Atrial Longitudinal Speckle Tracking Echocardiography in Healthy Aging Heart.

BACKGROUND: Left atrial volume (LAV) and function are connected to the left ventricular (LV) haemodynamic patterns. To define the changes of LAV and functions to counterbalance age-related LV diastolic impairment, this study was undertaken.
METHODS: 2D-Left Atrial Speckle Tracking Echocardiography (2D-LASTE) was used to define both LAV and functions in an aged healthy population (group II) respect to adult healthy controls (group I).
RESULTS: Results showed an increasing of left atrial volume indices (LAVI) (maximum, minimum, pre-a) in old subjects in comparison with those obtained in adult healthy controls. On the contrary, LAVI passive emptying unchanged and LAVI passive fraction reduced with advanced age. Finally, LAVI active emptying increased with advancing age to compensate the age-dependent left ventricular diastolic dysfunction. The values of global systolic strain (S); systolic strain rate (SrS); early diastolic strain rate (SrE), and late diastolic strain rate (SrA) were also calculated. With reference to the function, our study confirmed that LA conduit function deteriorates with age while booster pump increases respect to adult controls and reservoir phase is maintained.
CONCLUSIONS: The echocardiographic findings obtained with conventional and tissue Doppler confirmed the connection between LA functions and volumes and age-related LV dysfunction. Conclusively, 2D-LASTE appears to be a reliable tool to evaluate the role of LA to compensate the derangement of left ventricle happening with ageing.

INTRODUCTION

The left atrium (LA) plays a major role in left ventricle performance; hence, LA mechanical function is a surrogate marker of left ventricular (LV) dysfunction.[1] The components of LA function are divided into: Reservoir, conduit, and contractile phase. “Reservoir” corresponds to the difference between maximal and minimum LA volumes occurring in the interval just before the opening mitral valve and just before aortic valve opening. “Conduit” is the early phase of ventricular diastole. The blood is passively transferred into left ventricle just after mitral valve opening. “Contractile” phase or “booster pump” is calculated as the difference between minimum and preatrial contraction. Its role is to augment the stroke volume. The contribution of three phases of the LA function changes according to the diastolic properties of left ventricle. In normal conditions, the contribution of LA to LV filling is 40, 35, and 25%, respectively.[2] In relation to the LV valve movements, LA activity can be divided in four phases: A) Isovolumetric relaxation period occurring between the aortic closure and the opening of the mitral valve; B) LV rapid filling, which begins when LV pressure falls below the atrial pressure and the mitral valve opens; C) Diastasis, this corresponds to the equality between LA and LV pressures; and D) atrial systole, which corresponds to LA contraction and ends at the mitral valve closure.[3] Even though conventional, noninvasive techniques (as echocardiography and tissue velocity imaging derived parameters) are widely used to evaluate LA function; 2D-speckle tracking echocardiography (2D-STE) has emerged as a new, noninvasive method for the assessment of cardiac function. It is an imaging technique that provides accurate and angle-independent informations also regarding LA deformation and motion.[4] Recent reports suggested that 2D LA longitudinal strain obtained with STE (2D-LASTE) is an effective method for quantification of LA function.[56]

It is known that LV diastolic pressure increases with advancing age inducing some changes in LA dimensions and function.[78] To define the changes of atrial function dependent on the ageing, 2D-LASTE study was performed.

MATERIALS AND METHODS

Fifteen consecutive persons (12 males (M) and three females (F)) aged from 70 to 82 years (mean age = 79 ± 5) were chosen among those afferent to our Department of Internal Medicine and Geriatrics between May 2012 and April 2014. These were actually free of any cardiovascular, respiratory, and/or metabolic derangements (Group I).[910] Nineteen healthy adult controls (11 M and eight F) ranging in age from 44 to 61 years (mean age = 55 ± 6 years). None of the enrolled subjects had a history of ischemic heart disease or significant valvular abnormalities, peripheral vascular disease, cerebrovascular disease, systemic hypertension, or diabetes mellitus (Group II).[11] In addition, none of both groups’ subjects received cardioactive drugs at the moment of the study. Exclusion criterion was the inadequate echocardiographic visualization alone. Both groups were in sinus rhythm with heart rate (HR) < 100 beat/min subjects and gave their written informed consent for participation to the study.

Standard transthoracic echocardiography

Echocardiographic examinations were performed by experienced sonographers using a Philips iE33 machine (Eindhoven, NL). All measurements were performed in M- and B-mode in accordance with the American Society of Echocardiography criteria.[12] Ejection fraction% (EF%) was also defined according to the modified Simpson's criteria. The peak of early (E) and late (A) waves of diastolic mitral inflow were measured, and the E/A wave ratio were calculated. Mitral annular plane systolic excursion (MAPSE) was measured using M-mode echocardiography in the mitral annular lateral approach by apical four-chamber view.[13] Left atrial volume (LAV) was determined by the biplane area-length method.[14] Its value was indexed for body surface area in m2 and evaluated in ml/m2.

Pulsed doppler tissue imaging (DTI)

DTI of the mitral annular level was obtained at the lateral position. Values of peak early (e’) and late (a’) diastolic annular velocities were also obtained. Measurements were obtained during end expiration to eliminate respiratory variations and an average of three beats measured. The E/e’ ratio was also calculated.[15] The leading epidemiological, metabolic, and conventional echocardiographic characteristics of two groups are reported in Table 1.

Table 1

Clinical, echocardiography and metabolic aspects of the two enrolled groups

Parameters	Aged	Controls	P-value
n	15 (12 M-3F)	19 (11M-8 F)	NS
Age (years)	79±5	55±6	P<0.001
Heart rate (beats/min)	71±3	72±2	NS
Body surface area (m2)	1.72±0.5	1.74±0.4	NS
Systolic blood pressure (mmHg)	130±9.2	125±10.2	NS
Diastolic blood pressure (mmHg)	83.3±5.7	78±2.3	NS
Fasting blood glucose (mmol/ml)	5.9±0.5	5.7±0.5	NS
Total cholesterol (mmol/ml)	5.81±0.6	5.77±0.7	NS
LVEdV (ml)	107±14	96±17	NS
LVEsV (ml)	47±10	40±13	NS
LVEF (%)	53.3±3.5	58±2.4	NS
MAPSE (mm)	10±0,7	12±0,8	NS
Left atrial diameter (mm)	41.4±3.7	35±4.1	P<0.05
Left atrial volume index (ml/m2)	31±3	23±2.4	P<0.05
Mitral E (cm/s)	74.2±1.4	89.4±2.3	P<0.05
Mitral A (cm/s)	83.3±2.1	75±1.8	P<0.05
Mitral E/A ratio	0.89± 1.8	1.1±2.1	P<0.001
Mitral e’ (cm/s)	7.5±1.4	11±2.6	P<0.001
Mitral a’ (cm/s)	9.8±1.4	8.6±1.8	P<0.05
E/e’ ratio	11.7±1.2	9.1±1.5	P<0.05

LVEdV = Left ventricular end-diastolic volume, LVEsV = left ventricular end-systolic volume, LVEF = left ventricular ejection fraction, MAPSE = mitral annular plane systolic excursion, NS = not significant

STE

Speckle tracking images of the left atrium were obtained both in aged patients (Group I) and in healthy adult controls (Group II) by activating STE on the same echocardiographic machine. Three consecutive cycles were recorded and averaged. Computer-generated LA volume curve during one cardiac cycle. The left atrium endocardial surface was manually traced using a point-and-click approach in apical four-chamber view that automatically allowed a region of interest (ROI). This was manually adjusted, if necessary, to better suit the atrium anatomy. The cardiac cycle was demarcated by indicating QRS onset Afterwards, ROI was divided into six segments (two corresponding to the interatrial septum; two to the lateral wall; and two to the roof of the left atrium). If adequate images quality were not obtained, the records were rejected by the software and excluded from the analysis. By using LA volume, the following LA dynamic volumes were calculated: Maximum LA volume (LAV max), minimal LA volume (LAV min), and the LA volume before the atrial contraction (LAV pre-a). Precisely, LAV max was recorded just before mitral valve opening; LAV min was recorded at mitral valve closure; and LAV pre-a is LAV at onset of atrial systole (P wave). These values were corrected for body surface area as LAV index (LAVI) and consequently, LAVI maximum; LAVI minimum; and LAVI pre-a were obtained.

The phases of left atrial function, corresponding to reservoir, conduit, and booster pump were calculated by the LAVI volumes, as follows:

Left atrial emptying fraction (LAEF) total (corresponding to atrial reservoir function)

LAEF passive (corresponding to atrial conduit function)

LAEF booster (corresponding to atrial contractile booster pump)

In addition, LAVI passive emptying and passive fraction, and LAVI active emptying and active fraction were calculated.[16] In particular, LAVI passive emptying was measured as the difference between LAVI maximum - LAVI pre-a. Passive fraction was obtained as the ratio between LAVI passive emptying/LAVI maximum × 100%. LAVI active emptying is LAVI at onset of atrial systole - LAVI minimum. Active fraction was calculated as LAVI active emptying/LAVI pre-a × 100%.[17] Finally, LA global systolic strain (S); systolic strain rate (SrS); early diastolic strain rate (SrE); and late diastolic rate (SrA) were calculated in both groups. All recordings were obtained during normal respiration.

Statistical analysis

Data referring two-dimensional (2D) echocardiographic finding in both groups were expressed as mean ± standard deviation (SD). Comparison of LAVI (max, min, and pre-a) and the different phases of LAVI in two groups were performed using the Student's t-test for unpaired data. This was also employed to evaluate the differences between two groups referring LAVI passive emptying, passive fraction, LAVI active emptying, and active fraction. The mean values of reservoir, conduit and booster pump phases were also calculated in two groups. Likewise, the values of S, SrS, SrE, and SrA obtained in two groups were compared. A P-value < 0.05 was considered significant. All statistical analyses were performed using the Statistical Package for Social Sciences (SPSS) software. Data were analyzed using analysis of variance to assess for age-related changes in echocardiographic variables applying a Bonferroni's correction between two groups.

RESULTS

Conventional echocardiography

Left ventricular end-diastolic volume (LVEdV; 107 ± 14 ml) and left ventricular end-systolic volume (LVEsV; 47 ± 10 ml) measured in aged subjects were not significantly different from the mean values (LVEdV = 96 ± 17 ml and LVEsV = 40 ± 13 ml) recorded in healthy adults (not significant (NS)). EF% resulted 53 ± 3.5 in aged group and = 58 ± 2.4 in controls. Differences between two values were NS too. On the other hand, left atrial diameter of aged (41.4 ± 3.7 mm) was significantly greater (P < 0.05) than that recorded in healthy controls (35 ± 41 mm). LAVI also increased (P < 0.05) in elderly subjects (31 ± 3 ml/m2) in comparison to healthy adults (23 ± 2.4 ml/m2). E wave recorded in aged group (74.2 ± 1.4 cm/s) was lower (P < 0.05) than that recorded in controls (89.4 ± 2.3 cm/s). On the contrary, A wave significantly (P < 0.05) increased in healthy aged (83.3 ± 2.1 cm/s) compared with the mean value recorded in healthy controls (75 ± 1.8 cm/s). In agreement with these results, E/A waves ratio (P < 0.001) decreased in aged group (0.89 ± 1.8) compared to control group (1.1 ± 2.1). Finally, MAPSE recorded in aged patients (10.0 ± 0.7) was not significantly different (NS) from the mean value obtained in adult controls (12.7 ± 08) [Table 1].

Tissue doppler imaging

The e’ wave velocity decreased (P < 0.001) from 11 ± 2.6 cm/s in controls to 7.5 ± 1.4 cm/s in aged subjects. On the contrary, the a’ wave velocity increased (P < 0.05) from the normal adults (8.6 ± 1.8 cm/s) to the healthy aged (9.8 ± 1.4 cm/s). The E/e’ ratio also decreased (P < 0.05) with advancing age (9.1 ± 1.5 in adults and 11.7 ± 1.2 in healthy aged) [Table 1].

Adequate tracking quality was achieved in 98% of the total segments. LAVI max was = 30.7 ± 7.5 ml/m2, in comparison to 24.7 ± 3.5 ml/m2 found in adult controls (P < 0.05). LAVI minimum recorded in old people was 17.5 ± 6.1 ml/m2 with respect to the controls (13.9 ± 5.2 ml/m2). Differences were significant (P < 0.05). LAVI pre-a recorded in aged group was 26.2 ± 3.8 ml/m2. This value was significantly higher (P < 0.001) then that recorded in healthy adults (17.2 ± 6.7 ml/m2).

In addition; reservoir, conduit, and booster pump fractions of LA were recorded in two groups. Reservoir phase resulted superimposable in two groups (42.9 ± 1.5 in aged subjects vs 43.7 ± 1.8 in controls). On the contrary, conduit fraction reduced (P < 0.001) in oldest individuals (14.6 ± 1.9) in comparison with adult persons (30.3 ± 1.7). Booster pump significantly increased (P < 0.001) in aged group (33.2 ± 2.1) with respect to controls (19.1 ± 1.7) [Table 2]. Finally, LAVI passive emptying was 4.5 ± 0.4 ml/m2 in old subjects and 7.5 ± 0.6 ml/m2 in healthy adults. Differences were significant (P < 0.05). Passive emptying fraction was 14.7 ± 1.7 in aged, whereas its value was 30 ± 1.9 in controls. Differences were significant (P < 0.001). LAVI active emptying calculated in the elderly persons was 9.2 ± 0.7 ml/m2 and 6.3 ± 0.2 ml/m2 in adults without significant differences (NS). Active emptying fraction was 33.2 ± 1.9 in aged individuals, and 25.4 ± 1.3 in healthy adults, with significant differences (P < 0.001) [Table 2].

Table 2

Values of LAVI (maximum, minimum and pre-a), reservoir, conduit and booster pump in two groups. LAVI passive and active emptying fraction recorded in healthy aged and adult controls with statistical significance

Parameters	Elderly	Controls	Statistical significance
LAVI max (ml/m2)	30.7±7.5	24.7±3.5	P<0.05
LAVI min (ml/m2)	17.5±6.1	13.9±5.2	P<0.05
LAVI pre-a (ml/m2)	26.2±3.8	17.2±6.7	P<0.001
Reservoir	42.±1.5	43.7±1.8	NS
Conduit	14.6±1.9	30.3±1.7	P<0.001
Booster pump	33.2±2.1	18.1±1.7	P<0.001
LAVI passive emptying (ml/m2)	4.5±0.4	7.5±0.6	P<0.05
LAVI passive fraction (%)	14.7±1.7	30±1.9	P<0.001
LAVI active emptying (ml/m2)	9.2±0.7	6.3±0.2	NS
LAVI active fraction (%)	33.2±1.9	25.4±1.3	P<0.001
S	39.85±13.7	42.87±15.7	P<0.05
SrS	2.00±0.54	2.04±0.65	NS
SrE	−3.11±0.95	−2.27±0.85	P<0.05
SrA	−1.21±0.65	−2.35±0.78	P<0.01

LAVI = left atrial volume index, SrS = systolic strain rate, SrE = early diastolic strain rate, SrA = late diastolic strain rate, NS = not significant, LAV max = maximum LA volume, LAV min = minimal LA volume, LAV pre-a = LA volume before the atrial contraction

DISCUSSION

The morphological changes happening in the cardiovascular system with advancing age are responsible for increased myocardial stiffness and LV hypertrophy further evolving in LV diastolic dysfunction typical of the elderly.[18] For these changes, in our healthy aged individuals LV systolic function resulted to be within the normal limits with preserved LV volumes and EF% was > 50%, as previously described.[19] Since atrial function is related to increased LV filling, E wave velocity of diastolic mitral flow decreased and A wave significantly increased in comparison to the controls, with the inversion of the E/A wave ratio.[2021] That happens for a strongest atrial contraction in response to the increased end-diastolic LV pressure dependent on increased LV filling pressure (LVFP).[222324] It is known that MAPSE, reflecting longitudinal myocardial shortening, is a simple and sensitive echocardiographic parameter for assessing global longitudinal LV wall function. Its value is a highly accurate predictor of EF.[25] Particularly, MAPSE <8 mm was associated with depressed left ventricular ejection fraction (LVEF; < 50%), whereas mean MAPSE >10 mm was linked with preserved LVEF (> 55%). In our aged persons, we have found a MAPSE value of 10.0 ± 0.7 that is indicative of EF% > 50. In accordance with the study by Tighe et al.,[26] and successively, Munagala et al.,[27] in our healthy aged individuals; we found that the e’ velocity decreases, the a’ velocity increases, and the E/e’ ratio is > 10. It is known that the E/e’ ratio is able to estimate LVFPs in patients with preserved systolic function. Patients with E/e’ > 15 can be classified as having elevated filling pressure. An E/e’ < 8 suggests normal filling pressure. In the range of E/e’ of 8-15 other informations, including systolic function, chambers dimensions, and all Doppler variables must be considered in the analysis of individual patients.[28] Thus, the E/e’ value (11.7 ± 1.2) found in our aged subjects has an uncertain significance; but, considering all other echocardiographic data, the obtained value can be considered indicative of normal LVFPs. 2D-STE is a relatively new technology that tracks speckles in the myocardium frame-by-frame basis throughout the cardiac cycle, resulting in a noninvasive calculation of global and regional velocity, displacement, strain, and strain rate.[293031] In this study, we applied this echocardiographic technique on LA walls of healthy aging hearts to define LA volumes (max, min, and pre-a) indexed for body surface area (LAVI anatomy). LAVI passive/active emptying and functions were also defined in order to evaluate the influence of LAVI function on reservoir, conduit, and booster pump function. In our healthy aged individuals, reservoir is maintained; while conduit phase decreases and booster pump function increases [Figure 1]. These results were further confirmed by strain/strain rate values. Specifically, SrS (correlated with reservoir function) remained unchanged with respect to controls; SrE (correlated with conduit phase) decreased, whereas SrA (corresponding to the contractile function) increased. Changes in LA function occurred in conjunction with age-related changes in LV diastolic function.[32] The preservation of LA function during ventricular systole (reservoir) is important to maintain the cardiac output.[33] LA acts as a conduit during the phase of early LV diastolic filling, evidencing a decline in LA passive emptying fraction. This decline corresponds to the age-related inversion in E/A ratio described in our elderly individuals. Ageing also induces both prolonged LV relaxation and impairment of LV passive properties[2324] responsible for an augmentation of LA contribution to transmitral flow (booster pump).[34] Therefore, the advanced age is associated with depressed LAVI passive emptying function inversely, LAVI active emptying function increases with age[35] in order to maintain systolic ventricular volume, in accordance with a previous report of Rossi et al.,[36] and likewise to the diabetic cardiomyopathy, as we already described.[37] Several studies have reported similar results in healthy aging people.[3839]

Figure 1

(Left) Longitudinal left atrial strain obtained in control. Reservoir, conduit, and booster pump phases are indicated. (Right) Longitudinal atrial strain recorded in aged subject. Reservoir and booster pump increased, while conduit phase decreased. PALS = Peak atrial longitudinal strain

CONCLUSIONS

2D-LASTE represents an easy, noninvasive tool to characterize the morphological and functional LA changes to compensate healthy aged LV diastolic dysfunction. In other words, since LA is directly exposed to LV diastolic dysfunction through the mitral valve, it is evident that the changes of LA function reflects the duration and severity of increased LA pressure following the increased age-related LV diastolic dysfunction. Must be also added that LA dimensions and functions can be better evaluated with a multimodality imaging approach including cardiac magnetic resonance (CMR) and computed tomography (CT). Finally, the recently developed three-dimensional speckle tracking could more easily and simultaneously define the effective motion of speckles in all directions.[40]