Original Article |
Corresponding author: Radostina Ilieva ( radostilieva@yahoo.com ) © 2024 Radostina Ilieva, Elena Kinova, Boris Slavchev, Petar Kalaydzhiev, Desislava Somleva, Assen Goudev.
This is an open access article distributed under the terms of the Creative Commons Attribution License (CC BY 4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Citation:
Ilieva R, Kinova E, Slavchev B, Kalaydzhiev P, Somleva D, Goudev A (2024) Clinical and echocardiographic characteristics of patients with atrial cardiomyopathy and their impact on prognosis. Folia Medica 66(5): 608-617. https://doi.org/10.3897/folmed.66.e135893
|
Introduction: Patients with diverse demographic and clinical characteristics and comorbidities are included in the category of atrial cardiomyopathy (ACM).
Aim: Our study aims to evaluate the demographic, clinical, laboratory, and echocardiographic parameters of patients with ACM and to assess their impact on prognosis.
Materials and methods: Only 200 of the 724 consecutively evaluated patients with dilated left atrium who met the criteria for advanced ACM were included in the analysis. Forty age- and sex-matched controls with normal left atrial volume were also included. On enrollment, all patients received a detailed echocardiography with volumetric and speckle tracking analysis, and they were followed for 36 months for cardiovascular outcomes, including mortality.
Results: The mean age of the ACM population was 73.91±9.74 years, with 58% being women. Hypertension was found in 93% of them, 79% had atrial fibrillation, 60% had heart failure, 37% were obese, and 26% had diabetes. Over a median follow-up of 20.6 months, 35 deaths were registered in the ACM group compared to 1 death in the control group (17.5% vs. 2%, p=0.011). The presence of heart failure (HR 5.2, p=0.004), cancer (HR 3.7, p=0.007), severe tricuspid regurgitation (TR) (HR 5.4, p<0.001), high NT-proBNP (HR 1.4, p<0.001), and low right ventricular free wall strain (RVFWLS) (HR 1.2, p=0.006) were predictors of poor outcome.
Conclusion: In patients with ACM, the most prevalent comorbidities are hypertension, atrial fibrillation, heart failure, obesity, and diabetes. ACM is associated with high mortality with the best echocardiographic predictors – the presence of severe TR and RVFWLS >−17 %.
atrial cardiomyopathy, atrial fibrillation, heart failure, NT-proBNP, right ventricular strain
Though the relationship between atrial dilation and atrial arrhythmias has been known for a long time, atrial cardiomyopathy (ACM) has been introduced as a new term in recent years due to the wider use of the new imaging modalities in cardiology – strain echocardiography and cardiac magnetic resonance. Several definitions have been proposed for atrial cardiomyopathy[
The etiological factors of ACM are not entirely clarified. Still, the following comorbidities are essential in remodeling the atria: atrial fibrillation (AF), arterial hypertension, diabetes mellitus, obesity, obstructive sleep apnea, and aging. Structural and functional changes in the atria result also from congestive heart failure, valvular diseases, cardiac amyloidosis, genetic diseases, myocarditis, and other conditions.[
ACM includes patients with heterogenous demographic and clinical characteristics and comorbidities. Our study aims to evaluate the demographic, clinical, laboratory and echocardiographic parameters of patients with atrial cardiomyopathy and to assess their impact on prognosis.
Consecutive patients with advanced atrial cardiomyopathy who were hospitalized in the Cardiology Department of our hospital between September 2020 and May 2023 were included in the study. We defined advanced atrial cardiomyopathy as a severely dilated left atrium with volume index (LAVI) ≥48 ml/m2, preserved left ventricular (LV) systolic function – ejection fraction (EF) ≥50%, without a primary valvular or ventricular disease. Exclusion criteria were the presence of ventricular cardiomyopathy (dilated, hypertrophic or infiltrative), moderate or severe left ventricular hypertrophy (interventricular septum and posterior wall thickness of LV >13 mm), presence of primary valvular diseases (mitral stenosis, mitral or tricuspid regurgitation, aortic stenosis), acute coronary syndrome or pulmonary embolism, congenital heart disease, and constrictive pericarditis. Patients with pacemakers and cancer who did not fulfil the exclusion criteria were included in the study.
The study also included 40 control patients with normal left atrial volume (LAVI <34 ml/m2). The patients from the control group were age- and sex-matched to the patients with atrial cardiomyopathy but with structurally normal hearts. They were recruited from the patients to our hospital’s outpatient clinic.
All demographic and clinical data of the study group and the controls were collected from the hospital database (GlobalHis) and included age, sex, blood pressure and heart rate, smoking status, body mass index (BMI), the presence of comorbidities such as atrial fibrillation, heart failure (HF), arterial hypertension, diabetes mellitus, obesity, obstructive sleep apnea (OSA), coronary artery disease (CAD), sinoatrial (SA) or atrio-ventricular (AV) block, a pacemaker, stroke, and cancer, and the current medication. Collected laboratory data included hemoglobin and creatinine level, N-terminal pro-B-type natriuretic peptide (NT-proBNP), and C-reactive protein (CRP). The estimated glomerular filtration rate (eGFR) was calculated using the 2021 CKD-EPI Creatinine formula.
All patients (study group and controls) underwent detailed two-dimensional echocardiography on a Philips Epiq 7 machine with Matrix X5-1 transducer, including volumetric and speckle tracking analysis performed by a single operator. For the analysis, 32 echocardiographic parameters were used including strain of the left and right atrium, left and right ventricle. All measurements were performed according to the recommendations for cardiac chamber quantification of the European Association of Cardiovascular Imaging.[
The primary outcome of the study was all-cause death. Cardiovascular death was defined as death directly related to cardiovascular diseases, mainly congestive heart failure, sudden death, or an embolic event. The patients from the study group and the controls were followed for a median of 20.6 months by a visit to the clinic or a telephone call for the occurrence of the primary outcome. When contact with the patient was impossible, information was gathered from his relatives, or his vital status was verified from the National Health Institute records of Bulgaria.
The study was approved by the Research Ethics Committee of the Medical University of Sofia. All patients signed informed consent before inclusion.
Continuous variables, expressed as means and standard deviation, and the differences between the groups were assessed using the one-way analysis of variance (ANOVA) and Kruskal-Wallis one-way ANOVA, respectively. Categorical variables were expressed as counts and percentages, and differences between the groups were assessed using the chi-squared test (or Fisher exact test, when appropriate).
Differences in terms of mortality by group membership were evaluated using a log-rank test and drafted according to Kaplan-Meier curves. Stepwise proportional Cox regression analysis was used to determine the predictors of mortality.
In our main analysis, when evaluating predictors of mortality, NT-proBNP and right ventricular free wall longitudinal strain (RVFWLS) were separately assessed as the best predictors of mortality. Using these continuous variables, the accuracy of their prediction of outcomes was assessed by generating receiver‐operating characteristic curves (ROC) and reporting the area under the curve (AUC) using parametric methods, and reporting sensitivity and specificity. Recognizing a lack of accepted cutoff(s) for RV strain for assessing RV dysfunction or predicting outcome, we determined the RV strain threshold that maximized the index: sensitivity + specificity − 1 for classifying all‐cause mortality in our cohort.
All analyses were performed using IBM SPSS v. 29.0. All statistical significance levels were two-sided, and significant differences were expressed as p<0.05.
Of 724 consecutive patients with dilated left atrium, only 200 met the inclusion criteria for advanced atrial cardiomyopathy. The population under study, comprising 58% women, had a mean age of 73.9±9.7 years (46–100). Hypertension was found in 93% of them, diabetes in 26%, 79% had AF, 60% of them had heart failure, 20% had cancer, and 24% had SA or AV block. The mean BMI was 27.5±5.3, and 37% of them were obese.
The ACM group, compared to the control group, had significantly higher rates of hypertension (93% vs. 75%, p=0.001), congestive heart failure (60% vs. 0%, p<0.001), diabetes (26% vs. 8%, p=0.013, OSA (11% vs. 0%, p=0.032), SA or AV block (24% vs. 0%, p=0.001), and implanted pacemakers (15% vs. 0%, p=005). The ACM patients also had a significantly higher prevalence of coronary artery disease (9% vs. 0%, p=0.049), stroke (17% vs. 3%, p=0.020) and AF (79% vs. 5%, p<0.001).
In terms of laboratory parameters, the ACM patients, compared to the control group, had lower eGFR (67.4±21.4 vs. 78.6±18.7 ml/min/1.73 m2, p=0.010) and higher levels of NT-proBNP (1674.8±2053.4 vs. 88±32.7 pg/ml, p<0.001). Baseline demographic, clinical and laboratory characteristics are presented in Table
Demographic, clinical and laboratory parameters of ACM patients and controls
Atrial cardiomyopathy n=200 | Controls n=40 | p-value | |
Age (years) | 73.9±9.7 | 72.9±7.5 | 0.907 |
Female sex (%) | 58 | 59 | 0.936 |
Hypertension (%) | 93 | 75 | 0.001 |
Diabetes mellitus (%) | 26 | 8 | 0.013 |
CHF (%) | 60 | 0 | <0.001 |
Obesity (%) | 37 | 25 | 0.148 |
OSA (%) | 11 | 0 | 0.032 |
Cancer (%) | 20 | 8 | 0.060 |
CAD (%) | 9 | 0 | 0.049 |
SA or AV block (%) | 24 | 0 | 0.001 |
Pacemaker (%) | 15 | 0 | 0.005 |
Stroke (%) | 17 | 3 | 0.020 |
AF (%) | 79 | 5 | <0.001 |
Paroxysmal AF (%) | 29 | 5 | 0.063 |
Persistent/permanent AF (%) | 51 | 0 | <0.001 |
BMI (kg/m2) | 27.5±5.3 | 27.1±4.0 | 0.700 |
Smoking (%) | 32 | 21 | 0.169 |
SBP mmHg | 129±18 | 127±13 | 0.576 |
DBP mmHg | 79±11 | 80±9 | 0.543 |
HR | 75±16 | 71±11 | 0.239 |
CRP (mg/dl) | 1.5±2.1 | 0.9±0.9 | 0.177 |
Hemoglobin (g/l) | 131.6±20.0 | 135.6±12.4 | 0.320 |
Creatinine (mcmol/l) | 94.1±38.2 | 77.4±18.1 | 0.027 |
eGRF (ml/min/1.73 m2) | 67.4±21.4 | 78.6±18.7 | 0.010 |
NT-proBNP (pg/ml) | 1 674.8±2053.4 | 88±32.7 | <0.001 |
Fifty-eight percent of ACM patients were treated with diuretics, 65% with angiotensin-converting enzyme inhibitors (ACE) or angiotensin receptor blockers (ARB), 79% with beta-blockers, 34% with mineralocorticoid receptor antagonists (MRA), and 14% with sodium-glucose cotransporter-2 inhibitors (SGLT2i). Eighteen percent of ACM patients received digoxin, 35% antiarrhythmic drugs, and 72% – anticoagulation agents. Compared to the control group, ACM patients had significantly higher rates of diuretic, beta-blocker, MRA, SGLT2i, digoxin, antiarrhythmic, and anticoagulant treatment (Table
Atrial cardiomyopathy (n= 200) | Controls (n= 40) | p-value | |
Diuretics (%) | 58 | 0 | <0.001 |
ACEi/ARB (%) | 65 | 63 | 0.764 |
Beta blockers (%) | 79 | 38 | <0.001 |
Statin (%) | 34 | 38 | 0.628 |
MRA (%) | 34 | 3 | <0.001 |
SGLT2i (%) | 14 | 3 | 0.036 |
Digoxin (%) | 18 | 0 | 0.004 |
Antiarrhythmic drug (%) | 35 | 3 | <0.001 |
Anticoagulation agents (%) | 72 | 3 | <0.001 |
Anticoagulation – DOAC (%) | 61 | 3 | <0.001 |
Anticoagulation – Vitamin K antagonist (%) | 11 | 0 | 0.032 |
The mean ejection fraction of the LV of ACM patients was 56.1±4.9%, the mean LAVI was 54.7±8.5 ml/m2, and mean RAVI was 36.0±15.9 ml/m2. Their global longitudinal strain of LV was 16.0±4.9%, right ventricular free wall strain - −20.1±5.9%, reservoir strain of LA – 17.1±9.4%, and of RA – 18.8±9.7%. There was a significant difference in all studied echocardiographic parameters between the ACM and control group (Table
Atrial cardiomyopathy n= 200 | Controls n= 40 | p-value | |
LV EF (%) | 56.1±4.9 | 59.8±4.7 | <0.001 |
GLS LV (%) | −16.0±3.9 | −19.4±3.2 | <0.001 |
E/e׳ m | 14.3±4.9 | 10.1±2.6 | <0.001 |
E/e׳ l | 11.5±3.8 | 8.3±2.4 | <0.001 |
RVFWLS (%) | −20.1±5.9 | −24.6±5.9 | <0.001 |
RVGLS (%) | −16.9±5.8 | −19.1±8.3 | 0.044 |
LAVI ml/m2 | 54.7±8.5 | 25.9±5.9 | <0.001 |
LASr (%) | 17.1±9.4 | 39.7±11.2 | <0.001 |
LAScd (%) | −11.7±5.5 | −22.3±6.4 | <0.001 |
LASct (%) | −5.2±5.4 | −17.4±7.8 | <0.001 |
RAVI ml/m2 | 36.0±15.9 | 18.7±6.0 | <0.001 |
RASr (%) | 18.8±9.7 | 35.7±10.2 | <0.001 |
RAScd (%) | −12.6±7.1 | −22.7±8.2 | <0.001 |
RASct (%) | −5.4±5.2 | −13.4±5.7 | <0.001 |
PASP mmHg | 40.1±13.4 | 27.7±6.4 | <0.001 |
MR >mild (%) | 35 | 8 | <0.001 |
Severe MR (%) | 3.5 | 0 | 0.232 |
TR >mild (%) | 46 | 8 | <0.001 |
Severe TR (%) | 20 | 0 | 0.002 |
Over a median follow-up of 20.6±9.6 (1-39) months, 35 deaths were registered in the ACM group compared to 1 death in the control group (17.5% vs. 2%, p=0.011). Most deaths (21) in the ACM group were of cardiovascular origin, mainly due to heart failure. The causes of death of ACM patients are presented in Table
Causes of death in ACM | Number of patients |
Cardiovascular death | 21 |
Heart failure | 16 |
Thromboembolism | 3 |
Sudden cardiac death | 2 |
Noncardiovascular death | 12 |
Cancer | 5 |
Bleeding | 1 |
Infection | 4 |
Other | 2 |
Unknown | 2 |
Kaplan-Meier curves showing the difference in the mid-term mortality between ACM patients and the control group.
Multivariable stepwise Cox proportional regression analysis was employed to identify predictors of mortality. The model included the following clinical variables: sex, age, atrial fibrillation, arterial hypertension, diabetes mellitus, heart failure, obesity, obstructive sleep apnea, cancer, coronary artery disease, SA or AV block, implanted pacemakers, stroke, BMI, smoking, and systolic and diastolic blood pressure, as well as heart rate (Table
Multivariable Cox proportional hazards regression analysis of the clinical variables
Variable | Hazard ratio | 95% Confidence interval | p-value |
Age | 1.03 | 0.98÷1.09 | 0.243 |
Sex, female | 2.25 | 0.77÷6.61 | 0.140 |
Arterial hypertension | 2.39 | 0.41÷13.89 | 0.332 |
Atrial fibrillation | 1.46 | 0.76÷3.08 | 0.178 |
Diabetes mellitus | 1.12 | 0.96÷1.39 | 0.216 |
Heart failure | 5.2 | 1.67÷16.32 | 0.004 |
Obesity | 0.83 | 0.69÷1.95 | 0.274 |
ОСА | 2.98 | 0.99÷9.36 | 0.056 |
Cancer | 3.69 | 1.43÷9.49 | 0.007 |
CAD | 1.39 | 0.35÷5.60 | 0.636 |
SA or AV block | 1.06 | 0.18÷6.37 | 0.952 |
Pacemakers | 0.98 | 0.12÷8.08 | 0.983 |
Stroke | 1.02 | 0.78÷1.53 | 0.253 |
BMI | 1.05 | 0.92÷1.19 | 0.485 |
Smoking | 3.07 | 0.99÷11.39 | 0.060 |
Systolic blood pressure | 0.99 | 0.96÷1.02 | 0.533 |
Diastolic blood pressure | 0.97 | 0.93÷1.02 | 0.241 |
Heart rate | 1.02 | 0.98÷1.04 | 0.282 |
The same analysis was applied to the following laboratory parameters: hemoglobin, creatinine, eGFR, CRP, and NT-proBNP. Among these, NT-proBNP (HR 1.4, CI 1.2÷1.6, p<0.001) emerged as the strongest predictor of mortality (Table
Multivariable Cox proportional hazards regression analysis of the laboratory parameters
Variable | Hazard ratio | 95% Confidence interval | p-value |
CRP | 1.11 | 0.93÷1.35 | 0.239 |
Hemoglobin | 0.98 | 0.95÷1.00 | 0.121 |
Creatinine | 1.01 | 0.99÷1.02 | 0.745 |
eGRF | 0.96 | 0.91÷1.01 | 0.159 |
NT-proBNP | 1.40 | 1.20÷1.60 | <0.001 |
From all studied echocardiographic parameters, the following variables were included in the proportional Cox regression analysis to identify predictors of mortality: ejection fraction and global longitudinal strain of the left ventricle, E/e׳ of the medial and lateral mitral annulus, free wall global longitudinal strain and global strain of the right ventricle, indexed left and right atrial volume, the three components of left and right atrial strain (reservoir, conduit, and contractile), pulmonary artery systolic pressure, and the presence of more than mild, and severe mitral and tricuspid regurgitation (Table
Multivariable Cox proportional hazards regression analysis of the echocardiographic parameters
Variable | Hazard ratio | 95% Confidence interval | p-value |
LV EF | 0.97 | 0.88÷1.06 | 0.485 |
GLS LV | 1.09 | 0.75÷1.07 | 0.246 |
E/e׳ m | 1.25 | 0.91÷2.02 | 0.073 |
E/e׳ l | 1.35 | 0.98÷2.05 | 0.055 |
RVFWLS | 1.20 | 1.05÷1.37 | 0.006 |
RVGLS | 1.10 | 1.02÷1.39 | 0.019 |
LAVI | 1.05 | 0.99÷1.01 | 0.057 |
LASr | 1.10 | 1.01÷1.19 | 0.018 |
LAScd | 1.04 | 0.91÷1.14 | 0.171 |
LASct | 1.04 | 0.96÷1.12 | 0.353 |
RAVI | 1.09 | 0.95÷1.05 | 0.880 |
RASr | 1.08 | 1.03÷1.14 | 0.010 |
RAScd | 1.04 | 1.00÷1.09 | 0.052 |
RASct | 1.09 | 0.98÷1.20 | 0.106 |
PASP | 1.02 | 0.98÷1.06 | 0.459 |
MR >mild | 1.23 | 0.56÷2.70 | 0.613 |
Severe MR | 2.9 | 0.64÷13.37 | 0.167 |
TR >mild | 1.96 | 0.93÷4.19 | 0.082 |
Severe TR | 5.4 | 2.63÷11.31 | <0.001 |
Since NT-proBNP and right ventricular free wall strain (RVFWLS) were identified as the best predictors of mortality, we selected them for ROC analysis to calculate the area under the curve (AUC). NT-proBNP correctly classified death with an AUC value of 0.812 (p<0.001). NT-proBNP value of 1200 pg/ml was determined to discriminate best mortality with 82% sensitivity and 71% specificity (Fig.
Receiver operating characteristic for NT-proBNP as a predictor of mortality in ACM patients.
In our study, we assessed the clinical and echocardiographic characteristics of a cohort of patients with advanced atrial cardiomyopathy and explored their impact on prognosis. These are the four major findings: Firstly, the most common comorbidities of ACM patients were hypertension, atrial fibrillation, congestive heart failure, obesity, and diabetes. Secondly, patients with ACM had relatively high levels of NT-proBNP and were treated mainly with beta-blockers, ACEi/ARB, anticoagulants, and diuretics. Third, ACM is characterized by unfavorable echocardiographic parameters. Finally, ACM is an entity with a high mortality rate, and there are distinct clinical, laboratory, and echocardiographic characteristics that are associated with poor outcome.
Our findings of the most common comorbidities of ACM patients are in line with the literature that high-incidence pathologies like congestive heart failure[
The second important finding in our study is the presence of elevated levels of NT-proBNP in patients with ACM. Increased NT-proBNP levels serve as an important diagnostic and prognostic biomarker. They show a significant correlation with echocardiographic parameters related to AF remodeling and dysfunction and are associated with the burden of AF.[
Regarding the treatment of ACM patients, we found that most patients were treated with beta-blockers, anticoagulants, diuretics, and ACE inhibitors or ARB. The recommended strategies for the pharmacological management of ACM are based on stroke prevention, cardiac rhythm therapy, and rate control.[
Taking into account the important roles of angiotensin II, aldosterone, and inflammation in the pathophysiology of ACM, therapy with ACE inhibitors (ACEi) or angiotensin receptor blockers (ARB) shows promising results in reducing the incidence of AF in patients with cardiovascular diseases. Additionally, ACEi/ARB therapy has an additive effect when combined with spironolactone and statins.[
Some recent reports on the pleiotropic, cardiovascular protective effects of sodium-glucose cotransporter-2 (SGLT2) inhibitors show promising results on atrial remodeling.[
Our study included a comprehensive evaluation of the echocardiographic characteristics of the ACM patients with 32 parameters studied. We also used advanced echocardiographic methods – a two-dimensional strain of the left and right atrium, and of the left and right ventricle. The most extensively studied echocardiographic parameters in ACM are LA size and volume and, recently, LA strain. It is established that low LA reservoir strain values predict recurrences of AF, especially after pulmonary vein isolation, stroke and other embolic events, heart failure, and poor outcome.[
The poor prognosis of advanced atrial cardiomyopathy that we established is confirmed in a large study by Masuda et al.[
Regarding the predictors of outcome, most studies so far have been performed on AF patients and only a few in ACM. AF is the most important etiological factor of ACM, therefore, results from AF studies are extrapolated on ACM. In our study, we demonstrated that the strongest predictors of poor outcome are the presence of heart failure, cancer, high NT-proBNP values, RVFWLS >−17% and severe tricuspid regurgitation. The high proportion of patients with heart failure in our cohort confirms the pivotal role of this comorbidity as the leading cause of death in patients with AF, even more important than ischemic stroke.[
Though the prognostic significance of NT-proBNP is widely studied in patients with heart failure, to our knowledge, this issue has not been previously explored in the ACM population. It has been established that a value of NT-proBNP >1000 pg/ml predicts an increased risk of mortality and hospitalizations in patients with chronic stable heart failure.[
Although the prognostic role of RV strain in atrial cardiomyopathy is not yet established, it is known to be an important predictor of mortality in heart failure, pulmonary hypertension, and valvular heart disease.[
In patients with atrial cardiomyopathy, we identified that the most prevalent comorbidities are hypertension, atrial fibrillation, heart failure, obesity, and diabetes. These patients are characterized by unfavorable echocardiographic parameters and high mortality rates. The presence of heart failure, cancer, severe tricuspid regurgitation, NT-proBNP >1200 pg/ml, and RVFWLS >−17% are associated with worse outcome. Our results provide new insights into the clinical complexity of atrial cardiomyopathy and its outcome.
The authors have no support to report.
The authors have no funding to report.
The authors have declared that no competing interests exist.
Our study has several limitations, and our results should be interpreted with caution. One is the small sample size we studied. Secondly, patients in our study were with advanced form of ACM and were from hospital setting – older with more comorbidities, thus, our results may not be valid for younger and healthier outpatient population with more subtle forms of atrial cardiomyopathy. Thirdly, our conclusions apply only to the Bulgarian population. Further studies are warranted to analyze and better understand the heterogeneity of atrial cardiomyopathy.