If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Department of Cardiology, Herlev and Gentofte University Hospital, Copenhagen, DenmarkDepartment of Biomedical Sciences, Faculty of Health and Medical Sciences, University of Copenhagen, Denmark
Prognostic value of left atrial strain obtained by speckle tracking echocardiography.
•
Prognostic value of strain on atrial fibrillation following myocardial infarction.
•
Significantly lower left atrial strain in patient with atrial fibrillation.
•
Left atrial reservoir strain is a significant predictor of atrial fibrillation.
Abstract
Background
Atrial fibrillation (AF) is the most common arrythmia following ST-segment elevation myocardial infarction (STEMI) and can lead to stroke and other heart-related diseases. This study aimed to determine the prognostic value of left atrial (LA) strain, obtained by speckle tracking echocardiography (STE), in predicting incident AF outcomes following STEMI treated with primary percutaneous coronary intervention (pPCI).
Methods and results
This prospective study comprised of 392 STEMI patients treated with pPCI. The patients had an echocardiography performed at a median of two days after their STEMI. Along with conventional measures, LA strain was obtained by speckle tracking from two apical projections. The outcome was new-onset atrial fibrillation. LA reservoir, contractile and conduit strain were measurable from echocardiograms of 303 included patients. At a median follow-up time of 5.6 years (IQR: 5.0–6.1 years), 18 patients (6,3%) developed incident AF. Mean age was 62.0 years ±11.5 and follow-up was 100%. Significantly lower LA strain values were observed in patients who experienced AF during follow-up as compared to patients who didn't. Both reservoir, contractile and conduit strain were significant univariable predictors. In the multivariable model, only LA reservoir strain remained a significant independent predictor of AF.
Conclusion
Left atrial reservoir strain obtained by two-dimensional speckle tracking echocardiography is an independent predictor of incident AF following STEMI.
Atrial fibrillation (AF) is a supraventricular tachyarrhythmia characterized by uncoordinated atrial depolarizations. AF has a high prevalence in the population (1–2% of the general population [
Guidelines for the management of atrial fibrillationThe Task Force for the Management of Atrial Fibrillation of the European Society of Cardiology (ESC).
]) and is the most common non-physiological arrhythmia. AF is associated with a variability of cardiovascular diseases, such as stroke and other and thromboembolic events, progressive heart failure with dysfunction of the left ventricle besides low quality of life and high mortality [
]. Patients who have suffered from an ST-segment elevation myocardial infarction (STEMI) and subsequently have been treated with a primary percutaneous coronary intervention (pPCI) are at risk of developing AF. This is due to formation of scar tissue and remodeling of the left atrial wall resulting from the infarction. This may lead to interruption of the normal conduction pathway of the heart and spontaneous depolarisations.
Echocardiography is an important, accessible and non-invasive examination post pPCI. Over the past decade more attention has been paid to the left atrium (LA). Estimates of AF risk have previously been based on the structure of the LA with echocardiographic measurements such as LA volume indexed to body surface area (LAVI) which has been shown to be a useful predictor of AF [
]. Newer echocardiographic parameters with focus on LA function have been introduced. These include the left atrial emptying fraction (LAEF) and the LA strain analysis obtained by two-dimensional speckle tracking echocardiography (STE). Recent studies have found that these STE-derived parameters are applicable measurements when it comes to the study of LA and AF [
] and lead to improved patient outcome and possibly provide a socioeconomic benefit in healthcare.
Before these measures can be used for individualized risk stratification, the prognostic value must be tested and confirmed. However there seems to be a lack of studies which have investigated the clinic value of these measures in this group of patients which are at high risk of developing AF. This study aimed to examine the prognostic value of LA strain obtained by STE in predicting AF outcomes following STEMI-patients treated with pPCI.
2. Material and methods
This study represents a cohort study existing of a total of 392 STEMI patients prospectively included in the period September 2006 to December 2008. The patients were treated with a pPCI at Gentofte Hospital, University of Copenhagen, Denmark.
Inclusion criteria was a STEMI diagnosis (definition of STEMI is described elsewhere [
5012 Layer specific global longitudinal strain obtained by speckle tracking echocardiography for predicting heart failure and cardiovascular death following STEMI treated with primary PCI.
Eur. Heart J. [Internet].2017 Aug 1; 38 ([cited 2020 Jul 6]. Available from:)
]). A written informed consent was obtained from all patients. Of the initially 392 included patients, 14 were excluded due to known history with atrial fibrillation. Another 75 patients were also excluded due to echocardiographic measures with poor imaging quality for LA strain analysis. This left 303 patients to be included in the present study. Baseline data was obtained from all patients via their electronic health records. This data and the study population have been presented and elaborated in detail elsewhere [
A novel echocardiographic risk stratification scheme for predicting heart failure hospitalization and cardiovascular mortality after ST-segment elevation myocardial infarction.
The definition of diabetes was a fasting plasma glucose level > 7 mmol/L or a non-fasting plasma glucose level > 11 mmol/L or the use of anti-diabetic medication. The definition of hypertension was a use of anti-hypertensive medication ahead of the STEMI, a systolic blood pressure equal to or exceeding 140 mmHg or a diastolic blood pressure equal to or exceeding 90 mmHg during hospitalization.
The study was performed in accordance with the 2nd declaration of Helsinki and approved by the regional ethics committee.
2.1 Reperfusion and management
All patients were treated with pPCI according to contemporary guidelines. 300 mg acetylsalicylic acid, 600 mg clopidogrel and 10.000 international units of unfractionated heparin were administered before the coronary intervention was completed. The use of glycoprotein inhibitors was left to the discretion of the operator. Following pPCI, the patients were treated with anti-platelet, anti- thrombotic and beta-antagonistic medication as directed in contemporary guidelines [
ACC/AHA Guidelines for the Management of Patients with ST-elevation myocardial infarction: a Report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (Committee to Revise the 1999 Guidelines for the Management of Patients With Acute Myocardial Infarction).
The outcome was incident AF. Endpoints were obtained by usage of International Classification of Disease codes (ICD-10) from the Danish Board of Health's National Patient Registry. Follow-up was 100%. The ICD-10 codes include all hospital contacts for all Danish citizens including both in- and outpatient contacts. As it varies whether a specific type of atrial fibrillation has been coded in the register (e.g. paroxysmal atrial fibrillation) or a more unspecific diagnosis code was used (e.g. atrial fibrillation and atrial flutter, unspecified) a composite atrial fibrillation outcome was used for the present study. A sub-analysis was conducted with new-onset stroke post AF as the outcome. This endpoint was also extracted from the patient register.
2.3 Echocardiographic examinations
The echocardiographic examinations were performed by experienced clinicians or sonographers using a GE Vivid 7 ultrasound apparatus (GE Healthcare, Horten, Norway) with a 3.5 MHz probe. The patients underwent echocardiography at a median of two days [IQR 1;3 days] after the STEMI. Echocardiographic measures, including conventional two-dimensional echocardiography and tissue Doppler imaging were performed on all patients. Following the examination data were analyzed offline using commercially available software (EchoPac, GE Healthcare) by an experienced investigator blinded to patient data.
2.4 Conventional echocardiography and Tissue Doppler Imaging
Two-dimensional parasternal long axis images were used to measure left ventricle (LV) dimensions at the end of a diastole. These measurements include LV internal diameter (LVIDd), LV posterior wall thickness (LVPWd) and the interventricular septum diameter (IVSd). Left ventricular ejection fraction (LVEF) was calculated using the modified Simpson's biplane method. The left atrial volume (LAV) from the apical 4- and 2-chamber views was determined using the biplane area-length method. Following the maximal LAV was indexed to the body surface area to estimate LAVI. The left atrial emptying fraction (LAEF) was calculated ((end systolic LA volume – end diastolic LA volume)/end systolic LA volume).
Pulse-wave tissue Doppler imaging was utilized on the 4-chamber view in the mitral annulus in order to determine the early diastolic mitral annular velocity (e’) of the septal and lateral wall. The peak velocity blood flow from relaxation in the left ventricle in early diastole (E) and the peak velocity blood flow in the late diastole (A) were also measured using pulsed-wave Doppler. The average of E and e’ was subsequently used to calculate the E/e’ ratio.
2.5 Speckle tracking echocardiography
Two-dimensional speckle tracking (2DSTE) was performed with a semi-automated function which defined a region of interest. To quantify LV deformation, 2DSTE analysis of the LV was performed in the apical 4-chamber, 2-chamber, and 3-chamber views. LA 2DSTE analysis were performed in the apical 4-chamber and 2-chamber views. The endocardial border was manually defined. The software automatically traced the entire myocardium and thereby generated a region of interest (ROI). The ROI could subsequently be manually readjusted in cases of inaccurate tracing. The software divided the ROI into 6 segments for a total of 18 and 12 segments for the LV and LA, respectively. Inadequate segments for 2DSTE analysis could be excluded by the investigator if deemed necessary. All strain values were reported as absolute values.
2.6 Statistical analysis
Gaussian-distributed variables are represented as means with standard deviations (SD) and the comparison between the two groups, the group with incident AF and without, was made using student's t-test. Non-Gaussian distributed variables are represented as medians with interquartile range (IQR) and compared using the Mann-Whitney U test. Comparison between categorical variables was made using the chi-squared test.
In order to examine the prognostic value of LA strain in predicting the outcome, survival analysis was conducted. To obtain univariable analysis and multivariable models, Cox proportional hazard regression models were used. The extent of the multivariable adjustment was limited by the number of events which was 18. Model 1 is adjusted for age, gender and mean arterial pressure (MAP), all variables which are known to be correlated with the endpoint. Model 2 is an extended version of Model 1 with the addition of the biochemical finding peak Troponin I (TnI). In Model 3, 4, 5 and 6 both clinical, biochemical findings and echocardiographic findings were included. These models are adjusted for the same variables as Model 2 with the addition of GLS, LAVi, LVEF and E/e’, respectively.
In Fig. 2, restrictive cubic spline models are displayed to visualize the relation between the outcome and the strain results. The incident rates were obtained using a Poisson model. In Fig. 3, Kaplan-Meier curves are illustrated. These show the endpoint in relation to each strain measurement with the corresponding tertiles. (See Fig. 1.)
Fig. 1Example of 2DSTE analysis performed on the LA.
Caption: The endocardial border of the LA was manually defined. The entire myocardium was automatically contoured by the software. The curves represent the atrial deformation during a cardiac cycle including the reservoir, conduit and contractile phase.
For all analyses, p-value ≤0.05 in two-sided tests was considered statistically significant. All statistical results were obtained using STATA v.14 (StataCorp, LP, College Station, TX).
Reproducibility analysis for LA reservoir strain and contractile strain was performed on 20 randomly selected patients from the study cohort. Both intra- and interobserver analyses were conducted and reported as mean difference ± 1.96 standard deviation (SD).
To further investigate the clinical usefulness of LA strain parameters, a sub-analysis was carried out with new-onset stroke post AF as the outcome. A Cox proportional hazard regression model was used to assess the association between left atrial strain parameters and new-onset stroke after the diagnosis AF.
3. Results
A total of 303 STEMI patients were included in this study. During a median follow-up time of 5.6 years (IQR: 5.0–6.1 years) 18 patients (6.3%) developed new-onset AF. Mean age was 62.0 years ±11.5 and 77% of the patients were male. Baseline characteristics, stratified by endpoint, are displayed in Table 1.
Table 1Baseline Clinical Characteristics for patients stratified by endpoint AF.
Clinical Characteristics
No AF n = 285
AF n = 18
p-value
Age (years)
61.6 (11.4)
67.61 (12.1)
0.032
Male
220 (77.2%)
13 (72.2%)
0.63
Heart rate BPM
76.9 (40.3)
81.1 (20.0)
0.66
BMI (kg/m2)
26.8 (4.5)
27.79 (4.2)
0.35
Diabetes
24 (8.4%)
3 (16.7%)
0.23
Current smoker
156 (54.7%)
7 (38.9%)
0.19
Systolic blood pressure (mmHg)
135.2 (25.2)
144.6 (31.5)
0.13
Diastolic blood pressure (mmHg)
81.4 (16.6)
88.5 (20.7)
0.086
MAP, mmHg
99.4 (18.3)
107.2 (22.9)
0.084
Hypertension
85 (29.8%)
7 (38.9%)
0.42
Hypercholesterolemia
46 (16.1%)
2 (22.2%)
0.5
Previous MI
16 (5.6%)
0 (0.0%)
0.3
Previous stroke
7 (2.8%)
0 (0.0%)
0.48
Heart failure
76.9 (6.5)
81.1 (20.0)
0.66
Angiographic findings
TIMI Flow following pPCI
0.25
TIMI 0
9 (3.2%)
0 (0,0%)
TIMI 1
11 (3.9%)
2 (11,1%)
TIMI 2
23 (8.2%)
0 (0,0%)
TIMI 3
237 (84.6%)
16 (88,9%)
Multivessel occlusion
72 (25.3%)
8 (44.4%)
0.073
LAD lesion
129 (45.3%)
8 (44.4%)
0.95
RCA lesion
105 (36.8%)
8 (44.4%)
0.52
Biochemical findings
TnI, μg/L
110 (27.8–223)
189.5 (62.6–445)
0.065
eGFR, mL/min
74,5 (20.6)
64.3 (20.5)
0.044
CRP mg/L
3 (1–8)
3 (2–9)
0.42
Echocardiography
LVEF, %
38.2 (11.7)
35.3 (11.8)
0.33
LVIDd mm/m2
2.51 (0.33)
2.65 (0.46)
0.09
LVMI, g/m2
94.5 (27.0)
98.7 (48.4)
0.54
IVSd, mm/ m2
5.3 (1,1)
4.7 (1.0)
0.034
LVPWd, mm/ m2
5.3 (0.9)
5.1 (1.0)
0.33
E, cm/s
77.4 (19.2)
79.7 (19.1)
0.62
e’, cm/s
7.5 (2.1)
6.7 (2.2)
0.17
E/e’
14.9 (5.7)
18.4 (12.7)
0.026
Abnormal E/e’ (E/e’ > 14)
46 (16.1%)
7 (38.9%)
0.014
GLS, %
−12.4 (3.5)
−10.5 (4.4)
0.023
LAVI, mL/m2
25.3 (6.5)
26.8 (7.5)
0.35
Dilated LA (LAVI > 34)
29 (10.3%
3 (16.7%)
0.40
LA reservoir, %
33.4 (15.0)
24.1 (12.4)
0.011
LA conduit, %
−17.4 (9.5)
−12.0 (9.0)
0.020
LA contractile, %
16.0 (7.8)
12.1 (6.5)
0.041
BMI: Body Mass Index, MI: Myocardial infarction, TnI: peak troponin I, LVEF: left ventricular ejection fraction, LVIDd: Left Ventricular Internal Dimension at end-distole, LVMI: Left Ventricular Mass Index, IVSd: Interventricular septal end diastole, LVPWd: Left ventricular posterior wall thickness at end-diastole, E: peak transmitral early diastolic inflow velocity e’: early ventricular myocardial relaxation velocity, GLS: Global longitudinal strain, LAVI: Left Atrial Volume Index.
Patients developing AF were significantly older but did not differ on any other clinical characteristics. Furthermore, no angiographic differences were observed. Biochemical findings showed that patients developing AF had lower estimated glomerular filtration rate (eGFR) than patients who did not develop AF. No significant association was found between peak TnI levels and new-onset AF. Several echocardiographic variables were associated with new-onset AF. Patients reaching outcome had higher LVIDd and E/e’ and lower GLS and IVSd. The LA strains, reservoir, conduit and contractile strain, respectively, were all significantly lower in patients with new-onset AF.
3.1 Prediction of outcome
In the univariable unadjusted analysis both LA reservoir strain, LA conduit strain and LA contractile strain were significant predictors of new-onset AF. Univariable and multivariable Cox regression assessing the relationship between LA strain parameters and incident AF are depicted in Table 2. The incidence rate of AF according to LA reservoir, conduit and contractile strain are illustrated in Fig. 2. The study population was stratified into tertiles of LA reservoir, contractile and conduit strain, respectively, as displayed in Fig. 3. Lower tertiles of LA strain were associated with an increased risk of developing AF. The lowest tertiles displayed a 2-fold to 7-fold increased risk of developing AF compared to the highest tertile (LA reservoir: 1. tertile vs. 3. tertile: HR 4.8 95%CI(1.4–16.8), P < 0.015; LA contractile: 1. tertile vs. 3. tertile: HR 2.6 95%CI(0.8–8.4), P < 0.103; LA conduit: 1. tertile vs. 3. tertile: HR 6.8 95%CI(1.5–30.5), P < 0.012).
Table 2Prediction of incident AF using Cox regressions.
Unadjusted
Hazard Ratio
P-value
LA reservoir
0.9, 95% CI 0.9–1.9
0.006
LA condiut
1.1, 95% CI 1.0–1.2
0,012
LA contractile
0.9, 95% CI 0.9–1.0
0.036
Model 1
Hazard Ratio
P-value
LA reservoir
0.9, 95% CI 0.9–1.0
0.009
LA conduit
1.1, 95% CI 1.0–1.2
0.029
LA contractile
0.9, 95% CI 0.9–1.0
0.025
Model 2
Hazard Ratio
P-value
LA reservoir
1.0, 95% CI 0.9–1.0
0.026
LA conduit
1.1, 95% CI 1.0–1.2
0.096
LA contractile
1.9, 95% CI 0.9–1.0
0.034
Model 3
Hazard Ratio
P-value
LA reservoir
1.0, 95% CI 0.9–1.0
0.047
LA conduit
1.1, 95% CI 1.0–1.2
0.179
LA contractile
0.9, 95% CI 0.9–1.0
0.067
Model 4
Hazard Ratio
P-value
LA reservoir
1.0, 95% CI 0.9–1.0
0.043
LA conduit
1.1, 95% CI 1.0–1.2
0.110
LA contractile
0.9, 95% CI 0.9–1.0
0.058
Model 5
Hazard Ratio
P-value
LA reservoir
0.9, 95% CI 0.9–1.0
0.015
LA conduit
1.1, 95% CI 1.0–1.2
0.101
LA contractile
0.9, 95% CI 0.9–1.0
0.015
Model 6
Hazard Ratio
P-value
LA reservoir
1.0, 95% CI 0.9–1.0
0.041
LA conduit
1.1, 95% CI 1.0–1.2
0.143
LA contractile
0.9, 95% CI 0.9–1.0
0.048
Model 1 is adjusted for age, gender and MAP.
Model 2 is adjusted for age, gender, MAP and TnI.
Model 3 is adjusted for the same variables as Model 2 and additionally for GLS.
Model 4 is adjusted for the same variables as Model 2 and additionally for LAVi.
Model 5 is adjusted for the same variables as Model 2 and additionally for EF.
Model 6 is adjusted for the same variables as Model 2 and additionally for E/e’ > 14.
Caption: Incidence rate of endpoint incident AF per 1000 patient years in relation to three LA strain measurements. The curves are cubic spine regressions. The incidence rates are estimated with a Poisson model. 95% confidence intervals are represented by the shaded areas.
Caption: Displayed horizontally is time from inclusion which is STEMI treated with pPCI and following echocardiography. Displayed vertically is the probability of event-free survival. Three tertiles for each LA strain measurement are depicted. 1st tertile represents the highest LA strain values, and 3rd tertile represents the lowest LA strain values.
In the multivariable model, after adjusting for age, gender and MAP, both LA reservoir, conduit and contractile strain remained significantly associated with the endpoint (Table 2, Model 1). Only reservoir and contractile LA strain parameters remained independent predictors of incident AF when also adjusting for peak TnI (Table 2, Model 2).
Further multivariable analysis showed that when adjusting for age, gender, MAP, peak TnI and separately for GLS, LAVi or EF (Model 3, 4, 5), LA reservoir strain was the only measurement to remain a significant predictor of the endpoint in all models. LA contractile strain revealed to be a significant predictor when adjusting for EF (Model 5). In these models (Model 3, 4, 5), LA conduit strain was not significantly associated with incident AF. LA reservoir and contractile strain remained independent predictors of AF after adjustment for E/e’ > 14 (Model 6).
Reproducibility was found to be good when testing intra- and interobserver variability for LA reservoir strain and LA contractile strain with a small bias (LA reservoir strain: mean difference ± 1.96 SDs was 2.4 ± 5.8 for the intraobserver analysis and − 2.5 ± 9.6 for the interobserver analysis, LA contractile strain: 0.9 ± 4.2 for the intraobserver analysis and − 1.4 ± 6.6 for the interobserver analysis).
Of the 18 patients who developed new-onset AF, 4 had a stroke after the diagnosis AF was given. Neither reservoir, conduit nor contractile strain were found to be significantly associated with development of stroke after new-onset AF (LA reservoir: HR 0.95 95%CI(0.87, 1.05), p = 0.32, per 1% decrease; LA conduit: HR 1.19 95%CI(0.96, 1.47), p = 0.11, per 1% decrease; LA contractile: HR 1.00 95%CI(0.88, 1.14), p = 0.99, per 1% decrease).
4. Discussion
To our knowledge this is the first prospective study to investigate the clinical value of LA reservoir, contractile and conduit strain for predicting AF after STEMI treated with pPCI. We found that after multivariable adjustments, only LA reservoir strain remained significantly associated with the outcome of developing AF, while LA conduit strain and contractile strain did not. This study therefore suggests that LA reservoir strain add incremental prognostic value in the echocardiographic risk stratification for development of AF after STEMI treated with pPCI.
The measure of reservoir strain was superior when compared to contractile and conduit strain, and may be the most important LA strain parameter [
]. This may be because this phase depends on both the compliance of the atrial wall in addition to the longitudinal LV function. Hence, the measure entails information regarding the health of the LA and LV in one measure. In addition, LA reservoir strain is a measure of the peak deformation. During the reservoir phase, the LA fills and stretches to its maximal value. LA peak value may be a more accessible measurement compared to contractile and conduit strain measurements.
In a meta-analysis, including a total of 2542 healthy subjects with the age 25–68 years, Pathan et al. determined reference ranges for atrial function. This was approximately 39.4% for LA reservoir [
]. In comparison, our mean LA reservoir strain values were 33.4% in patients without AF and 24.1% in patients with incident AF. This elucidates the differences in LA reservoir function between a healthy control group and patients with STEMI.
Previous studies have found that lower LA reservoir strain is an independent predictor of cardiovascular morbidity and mortality in the general population [
]. Former research has primarily focused on volumetric measurements for assessment of LA function and development of AF. Tsang et al. showed that maximal LA volume is an independent predictor of incident AF in older men and women [
Echocardiographic predictors for progression to persistent or permanent atrial firbrillation in patients with paroxysmal atrial fibrillation: E6P study.
]. Previous studies have found results in line with ours concerning the use of LA strain for AF risk stratification in other study populations. An association between reduced LA reservoir and contractile function and paroxysmal AF has been demonstrated [
Relationship between left atrial strain, diastolic dysfunction and subclinical atrial fibrillation in patients with cryptogenic stroke: the SURPRISE echo substudy.
]. They found that LA reservoir strain remained an independent predictor of AF. They concluded that LA reservoir strain could provide prognostic value in this patient group. In another retrospective study of patients with ischemic stroke, 187 patients were enrolled [
]. A strong association between LA reservoir strain and paroxysmal atrial fibrillation (PAF) was present in this patient group. As in our study, LA reservoir strain remained significant even after adjusting for GLS and LAV. This is interesting to note because GLS and LAVI have been demonstrated to predict incident AF independently of each other and of other risk factors [
Left ventricular systolic dysfunction by longitudinal strain is an independent predictor of incident atrial fibrillation: a community-based cohort study.
]. Previously it has been shown by Ersbøll et al. that the prognostic value of LA reservoir strain, in relation to prediction of heart failure and cardiovascular death, was dependent on GLS and LAV [
The prognostic value of left atrial peak reservoir strain in acute myocardial infarction is dependent on left ventricular longitudinal function and left atrial size.
]. Since GLS indirectly expresses the LA reservoir function, it has been discussed whether LA strain measurements were able to provide further details beyond GLS and LAV. Our findings add importance to this discussion since our results suggest that LA reservoir strain is independent of GLS and LAV in a population of STEMI patients.
Iona et al. presented a strong association between LA strain and new-onset AF in hypertensive patients [
]. They found significantly lower values of LA strain in hypertensive patients who had developed AF than in hypertensive patients without AF. This was a study with 180 prospectively enrolled hypertensive patients in which 41 of the patients had experienced a recent AF episode which was converted to sinus rhythm. This was different from our study where known history with AF was an exclusion criterion. Also, no significant association was found between hypertension and AF in our baseline characteristics.
Despite the fact that the cohorts of patients differed between the different studies, it seems that LA strain is a useful predictor of AF in the majority of studies. Structural remodeling and increased formation of fibrous tissue in the LA wall are important risk factors in the development of AF. An inverse linear relation between LA reservoir strain and the amount of fibrosis in the LA wall has been revealed [
Left atrial strain and strain rate in patients with paroxysmal and persistent atrial fibrillation: relationship to left atrial structural remodeling detected by delayed- enhancement MRI.
]. The compliance of the LA is reduced due to the remodeling and fibrosis restricting the reservoir function of the atrium. This indicates that LA reservoir strain could be a sensitive marker of LA fibrosis which contributes to an increased risk of developing AF [
Although LA 2DSTE has been proven useful in several studies and shows great potential in future risk stratification, the technique itself has limitations [
]. The technique is frame-dependent and if the patient's 2D echocardiography has suboptimal image quality it cannot be used. The thin wall of the LA compared to e.g., the LV can make it more challenging to perform an optimal speckle tracking analysis. This has also previously been reported in another recent study using cardiac MR to measure LA strain [
]. However, the reproducibility of LA strain derived from 2D speckle tracking in the present study was found to be overall good with a small bias suggesting limited influence of the results of the present study. During a cardiac cycle the atria and ventricles move in opposite directions. This interplay between atrial and ventricular function can complicate the analysis [
]. Despite this, our results are similar to those found in several other studies which have also indicated the presence of a strong association between LA reservoir function and incident AF.
4.1 Study limitations
This study is subject to certain limitations. The extent of the multivariable analysis was limited by the low number of events. The fact that the cohort only included patients treated with pPCI limited the ability to relate the results to other patients with myocardial infarction. Further studies are needed to investigate whether there is a relation between strain parameters and AF in patient populations with other cardiovascular diseases than STEMI. Endpoints were collected from ICD-10 codes from the Danish Board of Health's National Patient Register. Regular ECGs or Holter monitoring following the pPCI were not performed as part of this study. We unfortunately did not have information on the type of atrial fibrillation for all patients. As the diagnosis of AF was obtained through a register there is a risk of missing potential AF endpoints. However, the validity of the AF diagnosis in the Danish National Patient Register has previously been tested and found to be accurate [
Left atrial reservoir strain is an independent predictor of incident AF after STEMI treated with pPCI. This measure can add prognostic information and assist in risk stratification for AF. LA conduit and contractile strain were not independent predictors of incident AF after multivariable adjustment.
Funding
This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.
Declaration of Competing Interest
Nothing to Disclose.
References
Camm A.J.
Kirchhof P.
Lip G.Y.H.
Schotten U.
Savelieva I.
Ernst S.
et al.
Guidelines for the management of atrial fibrillationThe Task Force for the Management of Atrial Fibrillation of the European Society of Cardiology (ESC).
5012 Layer specific global longitudinal strain obtained by speckle tracking echocardiography for predicting heart failure and cardiovascular death following STEMI treated with primary PCI.
Eur. Heart J. [Internet].2017 Aug 1; 38 ([cited 2020 Jul 6]. Available from:)
A novel echocardiographic risk stratification scheme for predicting heart failure hospitalization and cardiovascular mortality after ST-segment elevation myocardial infarction.
ACC/AHA Guidelines for the Management of Patients with ST-elevation myocardial infarction: a Report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (Committee to Revise the 1999 Guidelines for the Management of Patients With Acute Myocardial Infarction).
Echocardiographic predictors for progression to persistent or permanent atrial firbrillation in patients with paroxysmal atrial fibrillation: E6P study.
Relationship between left atrial strain, diastolic dysfunction and subclinical atrial fibrillation in patients with cryptogenic stroke: the SURPRISE echo substudy.
Left ventricular systolic dysfunction by longitudinal strain is an independent predictor of incident atrial fibrillation: a community-based cohort study.
The prognostic value of left atrial peak reservoir strain in acute myocardial infarction is dependent on left ventricular longitudinal function and left atrial size.
Left atrial strain and strain rate in patients with paroxysmal and persistent atrial fibrillation: relationship to left atrial structural remodeling detected by delayed- enhancement MRI.
00736-7&id=gr1.jpg){kind=link}
00736-7&id=gr2.jpg){kind=link}
00736-7&id=gr3.jpg){kind=link}