Echocardiographic evaluation of left atrial stiffness significantly improves the noninvasive detection of heart failure with preserved ejection fraction in patients presenting with paroxysmal atrial fibrillation, according to a clinical investigation conducted at Peking Union Medical College Hospital. Reporting in Echo Research and Practice, the investigators demonstrated that an echocardiographic surrogate of left atrial stiffness—calculated as the ratio of early diastolic filling pressure estimate to three-dimensional left atrial emptying fraction—discriminated between patients with high and low probabilities of heart failure with preserved ejection fraction. By incorporating noninvasive left atrial mechanical indices alongside ventricular filling parameters, the prospective observational study provides clinicians with a practical echocardiographic approach to assist diagnostic decision-making without immediately requiring invasive intracardiac catheterization.
Previous observations established that atrial fibrillation is the most common sustained cardiac arrhythmia globally, carrying an age-standardized prevalence of roughly 1.6 percent among Chinese adults. Pathophysiologically, the primary hemodynamic disturbance in heart failure with preserved ejection fraction involves impaired left ventricular relaxation and elevated filling pressures, which chronically transmit backward to the left atrium. This prolonged hemodynamic overload causes structural remodeling, progressive chamber stiffening, and mechanical dysfunction. Clinically, clinicians frequently attribute exertional symptoms entirely to rhythm irregularities; however, restoring sinus rhythm does not always resolve symptoms, indicating underlying atrial myopathy and occult heart failure. Although invasive hemodynamic catheterization during exercise represents the diagnostic benchmark, its substantial procedural costs, technical complexity, and invasiveness limit regular clinical deployment, underscoring the urgent need for validated, noninvasive echocardiographic markers.
The prospective study recruited consecutive patients diagnosed with paroxysmal atrial fibrillation between September 2020 and September 2022 for a discovery cohort, subsequently evaluating an independent validation cohort screened between May 2023 and December 2023. Eligible participants were aged between 18 and 80 years, met standard European Society of Cardiology electrocardiographic criteria for paroxysmal atrial fibrillation with documented episodes terminating within seven days, and had a preserved left ventricular ejection fraction of 50 percent or higher. The study also excluded individuals presenting with valvular atrial fibrillation, moderate-to-severe valvular disease, hyperthyroidism, active cardiomyopathies, congenital anomalies, unstable coronary syndromes, recent revascularization, or previous catheter ablation procedures.
To evaluate diagnostic categorization rigorously, the investigators determined the probability of heart failure with preserved ejection fraction using the validated H2FPEF scoring algorithm. The scoring system integrates clinical and echocardiographic components: heavy body mass index, treatment with two or more antihypertensive medications, presence of atrial fibrillation, echocardiographic pulmonary artery systolic pressure, advanced age, and Doppler-derived mitral filling ratios. Patients accumulating a score between 6 and 9 were classified into the high-probability group, whereas those with lower composite scores were designated into the low-probability reference group.
Comprehensive transthoracic echocardiography was performed on all participants using a specialized ultrasound platform equipped with a 1.6 to 3.2 megahertz transducer, adhering strictly to American Society of Echocardiography recommendations. All acoustic recordings were obtained while patients remained in stable sinus rhythm—either spontaneously present or achieved via successful cardioversion—to minimize beat-to-beat hemodynamic variability. Left atrial volumes were recorded at maximum end-systolic volume prior to mitral opening, at minimum end-diastolic volume at mitral closure, and immediately preceding atrial contraction via three-dimensional imaging. The left atrial emptying fraction was derived directly from these minimal and maximal volumetric datasets. Concurrently, two-dimensional speckle-tracking software tracked the endocardial border to measure global phasic longitudinal deformation, while tissue Doppler imaging quantified septal and lateral mitral annular early diastolic peak velocities, from which left atrial stiffness was formulated as the ratio of mitral inflow velocity over annular velocity divided by three-dimensional emptying fraction.
Baseline comparisons within the discovery cohort demonstrated substantial clinical and anatomical differences between the two risk groups. Patients classified with a high probability of heart failure were significantly older, had higher body mass index values, and included a greater proportion of women compared with the low-probability group. These high-probability patients also exhibited elevated rates of vascular comorbidities, including essential hypertension, diabetes mellitus, coronary atherosclerotic heart disease, and hyperlipidemia, alongside higher baseline circulating concentrations of N-terminal pro-brain natriuretic peptide and high-sensitivity C-reactive protein. Conventional echocardiographic measurements showed that left ventricular mass, chamber dimensions, left ventricular ejection fraction, and left ventricular global longitudinal strain did not differ significantly between cohorts. In contrast, markers reflecting left ventricular diastolic relaxation and left atrial mechanics diverged sharply between the groups, underscoring that functional mechanical impairment localized primarily to filling dynamics rather than resting systolic ejection.
Specifically, individuals in the high-probability heart failure group demonstrated distinctly impaired diastolic profiles, evidenced by lower average mitral annular early diastolic velocities of 6.78 centimeters per second compared to 8.40 centimeters per second in the low-probability group. The ratio of early mitral inflow to annular velocity was markedly elevated in high-probability patients, yielding a median value of 12.30 versus 8.40 among low-probability controls. Regarding atrial dimensions and deformation, minimum left atrial volume and anteroposterior diameter were significantly elevated in high-probability cases, whereas maximal volume showed no statistically significant difference. Left atrial functional metrics were substantially reduced, with left atrial emptying fraction averaging 49.52 percent in high-probability patients compared to 56.47 percent in controls, and reservoir strain falling to 27.4 percent versus 33.2 percent. Consequently, noninvasive left atrial stiffness was markedly higher in high-probability individuals, displaying a median index of 0.25 compared with 0.15 in low-probability participants.
Receiver-operating characteristic analyses established that left atrial stiffness delivered exceptional diagnostic discrimination in the discovery cohort, achieving an area under the curve of 0.862 with a 95 percent confidence interval spanning 0.801 to 0.922. Utilizing the optimal Youden index threshold of 0.21, the stiffness index yielded a diagnostic sensitivity of 77.27 percent, a specificity of 84.62 percent, a positive predictive value of 60.71 percent, and a negative predictive value of 92.37 percent. When applying an established age-adjusted normal threshold of 0.27 derived from European normative population studies, discriminatory accuracy remained robust. In multivariate logistic regression models adjusting for non-overlapping clinical variables, elevated left atrial stiffness remained an independent predictor of high-probability heart failure, whereas higher mitral annular early diastolic velocity served as a consistent protective factor against disease classification.
The researchers observed the strongest diagnostic performance when combining left atrial mechanics with ventricular relaxation velocities rather than relying on solitary measurements. Patients exhibiting both a reduced mitral annular early diastolic velocity of 7.275 centimeters per second or less and a left atrial stiffness ratio of 0.27 or higher demonstrated an exceptional positive likelihood ratio of 12.18 in the discovery cohort. This combined diagnostic approach substantially outperformed any individual echocardiographic parameter assessed in isolation. To confirm reproducibility, two independent cardiologists blinded to clinical status reassessed left atrial stiffness measurements across ten randomly chosen subjects, yielding high intra-observer and inter-observer intraclass correlation coefficients of 0.885 and 0.891, respectively. These high reproducibility metrics confirm that quantitative calculation of the stiffness ratio remains reliable across repeated offline ultrasound evaluations.
Validation testing conducted in the separate temporal cohort of 83 patients—comprising 32 high-probability and 51 low-probability individuals—further substantiated the diagnostic relevance of these noninvasive parameters. Consistent with discovery findings, left atrial stiffness remained significantly elevated among high-probability patients in the validation group, averaging 0.28 compared to 0.19 in low-probability controls, generating an area under the receiver-operating characteristic curve of 0.765. When applying the combined diagnostic rule requiring both depressed annular velocity and elevated stiffness, the assessment achieved a positive likelihood ratio of 7.18 and a specificity of 96.08 percent. The authors noted that while the combined criteria exhibited lower diagnostic sensitivity in the validation sample, its elevated positive predictive value and likelihood ratio position the metric as a highly dependable noninvasive rule-in tool to support clinical diagnosis.
To verify that findings were not driven solely by the heavy mathematical contribution of atrial fibrillation to the H2FPEF scoring model, the researchers performed a prespecified sensitivity analysis using an alternative anatomical and symptomatic definition. In this secondary evaluation of 23 symptomatic patients who simultaneously exhibited elevated natriuretic peptides and documented structural heart disease criteria, left atrial stiffness retained high discriminatory accuracy with an area under the curve of 0.859. Furthermore, combining reduced mitral annular velocity with elevated atrial stiffness in this subgroup generated a positive likelihood ratio of 10.69, confirming that the relationship reflects authentic atrial and ventricular pathophysiological remodeling rather than an artifact of algorithmic scoring overlap.
The study authors highlighted several methodological limitations that frame the clinical interpretation of their results. The investigation was conducted as a single-center cross-sectional observational study, which inherently prevents establishing causal relationships and introduces potential referral selection bias. Additionally, heart failure status was established using clinical scoring algorithms rather than invasive exercise right heart catheterization, meaning diagnostic classification errors cannot be completely ruled out. Calculating stiffness incorporates early diastolic filling velocity ratios, creating mathematical overlap with components embedded in diagnostic scores, although incorporating three-dimensional emptying fractions introduces independent physiological information. Furthermore, excluding patients with severe atrial enlargement exceeding 50 millimeters and restricting analysis to tracings recorded during sinus rhythm limits generalizability to patients with advanced atrial remodeling or permanent arrhythmia, warranting broader multicenter evaluation.
Subject of Research: Echocardiographic assessment of left atrial stiffness for diagnosing heart failure with preserved ejection fraction in paroxysmal atrial fibrillation
Article Title: The diagnostic value of left atrial stiffness for heart failure with preserved ejection fraction in patients with paroxysmal atrial fibrillation
Article References: Lai, J., Qin, X., Zhang, D., Wang, J., Yu, J., Sun, K., Yang, D., Fan, J., Zhang, L., Cheng, Z., Cheng, K., Gao, P., Deng, H., Fang, L., Chen, T., Fang, Q., & Liu, Y. (2026). The diagnostic value of left atrial stiffness for heart failure with preserved ejection fraction in patients with paroxysmal atrial fibrillation. Echo Research & Practice, 13(1), Article 6. https://doi.org/10.1186/s44156-026-00107-5
Image Credits: AI Generated
DOI: 10.1186/s44156-026-00107-5
Keywords: Atrial Fibrillation, Heart Failure with Preserved Ejection Fraction, Left Atrial Stiffness, Echocardiography, Diastolic Function, Tissue Doppler Imaging, Cardiovascular Research, Clinical Diagnostics, diagnostic, value, left, atrial
