CMR-based left atrial strain for predicting left ventricular diastolic dysfunction in asymptomatic hypertensive patients

Introduction

Hypertension (HTN), one of the most prevalent cardiovascular disorders worldwide, affects approximately one-third of adults aged 30–79 years (1). Among HTN-related cardiovascular manifestations, left ventricular diastolic dysfunction (LVDD) is one of the most common and earliest functional impairments (1,2). Its pathogenesis involves multilevel pathological cascades, including pressure overload-induced myocardial remodeling, vascular stiffening, neurohormonal overactivation, and inflammatory cytokine dysregulation (2). As a principal etiological factor in heart failure with preserved ejection fraction (HFpEF), LVDD is closely associated with the intricate pathological progression and adverse prognosis of this condition (3). Clinical evidence indicates a high 5-year all-cause mortality rate for HFpEF (75.3%), which is comparable to that of heart failure with reduced ejection fraction (4). However, current therapeutic strategies remain substantially limited in improving clinical outcomes (5). Consequently, the early identification of subclinical LVDD and timely intervention are pivotal strategies for mitigating the progression from HTN to HFpEF.

Routine evaluation of LVDD primarily relies on echocardiography, a process characterized by multiparametric and complex assessment. The 2016 expert consensus introduced a standardized algorithm for LVDD assessment by estimating left atrial pressure (LAP), significantly reducing the required parameters (6). This algorithm classifies HTN patients into two categories based on LAP levels: no LVDD or LVDD grade I (normal LAP) and LVDD grade II or higher (elevated LAP). Such a classification system enhances diagnostic efficiency while maintaining accuracy validated through invasive methods (7). However, the algorithm still involves at least three parameters, poses diagnostic challenges for some patients, and is inapplicable to cases with pulmonary HTN, arrhythmias, or mitral valve disease. Numerous studies have demonstrated that speckle tracking echocardiography (STE) effectively identifies LAP elevation and LVDD by quantifying left atrial (LA) strain parameters (8-10), and its potential for optimizing LVDD assessment workflows has garnered expert endorsement (8,11). Nevertheless, STE requires additional imaging protocols and faces limitations in acoustic window quality and spatial resolution, hindering its clinical adoption.

The recently developed cardiac magnetic resonance feature tracking (CMR-FT) technique enables the assessment of LA strain from routine cine sequences without additional scans. CMR-FT demonstrates better accuracy and reproducibility than STE (12,13). Conventional cardiac magnetic resonance (CMR), however, lacks valvular hemodynamic parameters, which limits the assessment of diastolic function. In contrast, CMR-FT-based LA strain analysis may provide critical complementary metrics (14). To date, studies investigating CMR-FT-derived LA strain in HTN patients remain scarce, and the dynamic changes of these parameters in LVDD have not been well characterized. This study aimed to systematically characterize CMR-derived LA parameters in patients with essential HTN and explore their potential for the early identification of LVDD associated with elevated LAP. We present this article in accordance with the STROBE reporting checklist (available at https://qims.amegroups.com/article/view/10.21037/qims-2025-1128/rc).

Methods

Study population

This study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. The study protocol was approved by the Institutional Ethics Committee of Ganzhou People’s Hospital (No. MR-36-25-039496), and the requirement for written informed consent was waived for this retrospective study. We consecutively enrolled patients with arterial HTN who underwent diagnostic or therapeutic evaluations at our center between January 2019 and January 2024. HTN was defined as systolic blood pressure (SBP) ≥140 mmHg or diastolic blood pressure (DBP) ≥90 mmHg on at least two separate office measurements. The inclusion criteria were preserved left ventricular ejection fraction (LVEF >50%), sinus rhythm on electrocardiography, and no overt clinical symptoms. Exclusion criteria included secondary HTN, history of myocardial infarction, atrial fibrillation, other known cardiac diseases, significant valvular pathologies, prior cardiac surgery, renal failure, and poor imaging quality. A total of 152 eligible hypertensive patients were included. Additionally, 60 age- and sex-matched healthy controls with no history of cardiovascular disease were enrolled. All controls demonstrated normal findings on echocardiography and electrocardiography. All participants underwent echocardiography and CMR within a 7-day period. Baseline laboratory tests, imaging findings, and medication histories were retrieved from electronic medical records.

Transthoracic echocardiography (TTE)

Echocardiographic examinations were performed using a Philips EPIQ ultrasound system (Philips Healthcare, Amsterdam, The Netherlands) equipped with an S5-1 cardiac transducer (1–5 MHz frequency range) by certified cardiac sonographers. In standardized apical four-chamber views, early diastolic mitral inflow velocity (E) was measured via pulsed-wave Doppler, while tissue Doppler imaging was used to acquire early diastolic mitral annular velocities (e’) at septal and lateral sites, with the E/e’ ratio calculated as the average of both measurements. Peak tricuspid regurgitation velocity (TRV) was assessed by continuous-wave Doppler when detectable. LA maximal volume was quantified via the biplane Simpson method and indexed to body surface area (LAVI). According to current guideline criteria (6), patients were categorized into two subgroups: (I) HTN with normal LAP (no LVDD or grade I LVDD); and (II) HTN with elevated LAP (grade II or higher LVDD).

CMR protocol

All CMR examinations were conducted on a 3.0 T MRI scanner (Skyra, Siemens Healthineers, Erlangen, Germany) equipped with an 18-channel cardiac phased-array coil and vectorcardiographic gating. Participants were positioned supine and scanned during end-expiratory breath-holds. Standard cardiac cine imaging was acquired using a balanced steady-state free precession (bSSFP; TrueFISP) sequence with asymmetric echo sampling. This protocol acquired long-axis views (two-chamber, three-chamber, and four-chamber; slice thickness, 5 mm; no gap) and a contiguous short-axis stack covering the entire left ventricle (slice thickness, 8 mm; no gap). Key acquisition parameters were as follows: field of view, 208–300 × 180–256 mm²; matrix, 208–300 × 144–224 pixels; repetition time, 2.9–3.5 ms; echo time, 1.3–1.5 ms; flip angle, 42°–55°; and temporal resolution, 35–55 ms per cardiac phase.

CMR image analysis

All acquired images were transferred to a dedicated cardiac post-processing workstation (CVI 42 v5.17.2, Circle Cardiovascular Imaging, Calgary, Canada) for analysis. Two cardiovascular imaging specialists, each with >5 years of dedicated cardiac MRI experience and blinded to clinical data, performed offline analysis. Within the ventricular function module, short-axis cine stacks were used to manually adjust endocardial and epicardial contours, with corrections to automated software tracings, enabling calculation of LV functional indices: ejection fraction (EF), end-diastolic volume index (EDVI), end-systolic volume index (ESVI), stroke volume index (SVI), cardiac output index (CI), left ventricular mass index (LVMI), and maximal wall thickness (MWT). In the strain analysis module, predefined short-axis contours were applied, and long-axis contours were manually adjusted to derive three-dimensional LV myocardial strain parameters: global radial strain (GRS), global circumferential strain (GCS), and global longitudinal strain (GLS). Details of these strain measurements are provided in Figure S1.

In the biplane long-axis view, two-chamber and four-chamber cine images were analyzed. The software automatically tracked LA contours, which were manually adjusted to calculate LA volumetric parameters throughout the cardiac cycle. After normalization to body surface area (BSA), three indices were derived: maximum left atrial volume index (LAVImax), minimum left atrial volume index (LAVImin), and pre-contraction left atrial volume index (LAVIpre). LA functional metrics were calculated as follows: LA total EF = (LAVImax − LAVImin)/LAVImax × 100%, LA passive EF = (LAVImax − LAVIpre)/LAVImax × 100%, and LA booster EF = (LAVIpre − LAVImin)/LAVIpre × 100%.

LA strain analysis was performed using the strain analysis module. Endocardial and epicardial contours were manually delineated on two- and four-chamber cine images at maximal and minimal LA volumes, with exclusion of the LA appendage and pulmonary veins. The software automatically propagated these contours throughout the cardiac cycle, and manual adjustments were applied when necessary to generate strain and strain rate curves. Key parameters derived from the analysis included reservoir strain (εs), conduit strain (εe), and booster pump strain (εa), along with corresponding strain rate parameters: peak positive strain rate (SRs), peak early negative strain rate (SRe), and peak late negative strain rate (SRa). The measurement details were illustrated in Figure 1.

Figure 1 Left atrial strain analysis by CMR feature tracking in an HTN patient with normal LAP. (A,B) The LA tracking contours at the end-diastolic phase in two-chamber and four-chamber views, respectively. (C,D) The LA strain and strain rate curves. CMR, cardiac magnetic resonance; HTN, hypertension; LA, left atrial; LAP, left atrial pressure; SRa, peak late negative strain rate; SRe, peak early negative strain rate; SRs, peak positive strain rate; εa, booster strain; εe, conduit strain; εs, reservoir strain.

Reproducibility

Reproducibility analysis was performed by randomly selecting images from 30 hypertensive patients and 10 healthy controls. LA strain parameters were assessed by two blinded cardiologists to evaluate inter-observer variability. Intra-observer variability was assessed by having one cardiologist remeasure the strain parameters from the same images after a one-month interval.

Statistical analysis

Continuous variables are presented as mean ± SD or median (IQR) for normally and non-normally distributed data, respectively. Group comparisons across control, HTN (normal LAP), and HTN (elevated LAP) groups used one-way ANOVA with Bonferroni post-hoc test for normal variables, Kruskal-Wallis test for non-normal variables, and Chi-squared or Fisher’s exact test for categorical variables [expressed as numbers (%)]. Bland-Altman analysis was performed to evaluate the agreement of LAVImax measurements between TTE and CMR. Linear regression identified factors influencing LA strain parameters, incorporating variables with P<0.1 in univariate analysis and no multicollinearity in the multivariate model. Receiver operating characteristic (ROC) curves assessed the ability of CMR-derived LA parameters to differentiate HTN (elevated LAP) from HTN (normal LAP) patients; the area under the curve (AUC) was calculated and compared using DeLong’s test. A combined LA parameter diagnostic model was developed via binary logistic regression, with performance evaluated using AUC. The intraclass correlation coefficient (ICC) assessed the reproducibility of CMR-FT-derived LA strain measurements. All analyses were performed using IBM SPSS 26.0; statistical significance was defined as two-tailed P<0.05.

Results

Baseline characteristics

This study included three groups for analysis: healthy controls (n=60; 50.42±13.67 years; 56.7% male), HTN (normal LAP) group (n=96; 52.49±12.69 years; 66.7% male), and HTN (elevated LAP) group (n=56; 55.44±11.84 years; 71.4% male); clinical characteristics are summarized in Table 1. Compared with healthy controls, both HTN groups (with or without elevated LAP) showed significantly higher SBP, DBP, BSA, and BMI (all P<0.05), whereas no significant differences were observed in sex distribution, age, heart rate, smoking history, or drinking history among the three groups (all P>0.05).

Comparison of CMR and echocardiographic parameters between hypertensive patients and healthy controls

As shown in Table 2, compared with the control group, HTN patients exhibited significantly higher LVMI and MWT values, with significant differences also observed between hypertensive subgroups (P<0.05); EDVI and ESVI were lower in the HTN (normal LAP) group than in the control group (P<0.05), whereas no significant differences were detected among other subgroups. Parameters including LVEF, SVI, and CI showed no significant intergroup differences. LV myocardial strain analysis revealed a progressive decline in GRS, GCS, and GLS values across the control, HTN (normal LAP), and HTN (elevated LAP) groups; however, no statistically significant differences in GRS or GCS were observed between the two HTN subgroups (P>0.05).

Compared with controls, patients in the HTN (normal LAP) group showed no significant difference in LAVImax, whereas it was markedly reduced in the HTN (elevated LAP) group. Both LAVImin and LAVIpre demonstrated a stepwise reduction across the control, HTN (normal LAP), and HTN (elevated LAP) groups, with significant pairwise differences among all groups (all P<0.05). Relative to controls, HTN patients (with or without elevated LAP) exhibited significantly impaired LA reservoir function (εs, SRs, LA total EF) and conduit function (εe, SRe, LA passive EF). With the exception of LA passive EF, all these parameters showed significantly greater impairment in the HTN (elevated LAP) group than in the HTN (normal LAP) group (all P<0.05; Table 2, Figure 2). In contrast, LA booster pump function (εa, SRa, LA booster EF) did not differ significantly between controls and the HTN (normal LAP) group but was significantly reduced in the HTN (elevated LAP) group.

Figure 2 Half violin plots depicting LA strain parameters among the three groups: Healthy controls, HTN with normal LAP, and HTN with elevated LAP. Illustrated parameters: (A) εs, (B) εe, (C) εa, (D) SRe, (E) SRs, and (F) SRa. *, P<0.05 versus Healthy controls; ^#, P<0.05 versus HTN (normal LAP). HTN, hypertension; LA, left atrial; LAP, left atrial pressure; SRa, peak late negative strain rate; SRe, peak early negative strain rate; SRs, peak positive strain rate; εa, booster strain; εe, conduit strain; εs, reservoir strain.

Echocardiography demonstrated significantly reduced mitral E-wave velocity and e’ in hypertensive patients compared with controls (all P<0.05). The E-wave showed no significant difference between HTN subgroups (P=0.99). The E/e’ ratio exhibited a progressive increase across all three study groups (all P<0.05). Notably, the prevalence of TRV >2.8 m/s was significantly higher in HTN patients with elevated LAP (P<0.05). Additionally, our data revealed mild underestimation of LAVImax by TTE versus CMR, with a mean bias of +2.22 mL/m² (95 % CI: 1.85–2.59; Figure 3, Table S1). Agreement between the two techniques was excellent (ICC =0.86, Table S1).

Figure 3 Bland-Altman analysis of LAVImax measurements: TTE vs. CMR. CMR, cardiac magnetic resonance; LAVImax, maximum left atrial volume index; TTE, transthoracic echocardiography.

Factors affecting LA strain parameters

The multivariable regression model (Table 3) included parameters significantly associated with εs, εe, and εa in univariate analysis: e’, E/e’ ratio, LAVImin, GRS, GLS, BMI, age, LVMI, SBP, and DBP. LAVImin was selected for the model over LAVImax because it demonstrated greater intergroup differences and due to strong collinearity between the two measures. All included variables had variance inflation factors (VIF) ranging from 1.177 to 2.903, indicating no severe multicollinearity.

The analysis revealed that εs was independently and positively associated with e’ (β=0.340), GRS (β=0.215), and GLS (β=0.468), and inversely associated with LAVImin (β=-0.231) and LVMI (β=−0.063) (all P<0.05). For εe, it was independently positively correlated with e’ (β=0.670) and GLS (β=0.380), whereas it was negatively correlated with LAVImin (β=−0.136) and LVMI (β=−0.042) (all P<0.05). Finally, εa exhibited an independent positive correlation with GRS (β=0.167) and independent negative correlations with the E/e’ ratio (β=−0.192) and BMI (β=−0.122) (all P<0.05).

Diagnostic capability of LA strain parameters

ROC curve analysis revealed the discriminative capacity of multiple parameters in differentiating HTN (elevated LAP) from HTN (normal LAP) (Table 4, Figure 4). All LA strain parameters demonstrated strong discriminatory performance. Notably, εs exhibited superior diagnostic performance, with an AUC of 0.896 (95% CI: 0.846–0.945; optimal cutoff: 29.55%; sensitivity: 84.4%; specificity: 82.1%), and significantly outperformed conventional LVDD indices such as e’ (AUC =0.800) and E/e’ ratio (AUC =0.802), as well as GLS (AUC =0.709). Although εe showed a numerically higher AUC than εs, no statistically significant difference was observed between them (P=0.329). Furthermore, LAVImin demonstrated stronger discriminative power than LAVImax (AUC =0.770 vs. 0.711, P=0.0069). The combined model integrating εs and LAVImin achieved exceptional diagnostic accuracy, yielding an AUC of 0.919 (sensitivity: 89.58%; specificity: 80.36%).

Figure 4 ROC analysis for discriminating the HTN (elevated LAP) group from the HTN (normal LAP) group. (A) LA strain and conventional diastolic function parameters. (B) Combination of εs and LAVImin. AUC, area under the curve; CI, confidence interval; E, early diastolic mitral inflow velocity; e’, average early diastolic mitral annular velocity (septal and lateral); GLS, global longitudinal strain; LA, left atrial; LAP, left atrial pressure; LAVImax, maximum left atrial volume index; LAVImin, minimum left atrial volume index; ROC, receiver operating characteristic; εe, conduit strain; εs, reservoir strain.

Reproducibility of LA strain parameters

As shown in Table 5, the intra-observer ICC of LA strain parameters ranged from 0.879 to 0.941, and the inter-observer ICC ranged from 0.813 to 0.935. Both fell within the good to excellent range.

Discussion

In this study, we evaluated LA function in HTN patients with or without elevated LAP using CMR-FT. Key findings include: (I) HTN patients demonstrated impaired LA reservoir and conduit function compared with controls, even in those with normal LAP, whereas LA pump function was preserved at this early stage. CMR-FT sensitively detected these functional alterations; (II) LAVImin demonstrated superior performance to the guideline-recommended volumetric parameter LAVImax for evaluating LVDD; (III) LA strain parameters exhibited significant independent associations with traditional diastolic indices (e’, E/e’ ratio, LA volume indices) and LV strain parameters; (IV) CMR-derived LA strain parameters—particularly εs—demonstrated robust diagnostic performance in distinguishing HTN patients with elevated LAP from those with normal LAP.

The LA is directly connected to the LV, dynamically regulating ventricular filling through its reservoir, conduit, and booster pump functions (15). Owing to this intimate LA-LV coupling, adaptive changes in LA function frequently precede detectable LV dysfunction (15,16), with elevated LAP serving as a central compensatory mechanism in LVDD. When diastolic function deteriorates, reduced ventricular compliance leads to progressive increases in LV end-diastolic pressure, thereby impairing diastolic filling. Driven by retrograde pressure transmission and the need to sustain LV filling demands, compensatory LAP elevation ensues. Thus, elevated LAP is recognized as both a critical pathophysiological consequence of LVDD and a key biomarker for its identification and staging, having been integrated into guideline-endorsed LVDD diagnostic criteria (6).

LAVI remains the only routinely accessible parameter for assessing LAP elevation-associated LVDD in standard CMR protocols, as it requires no additional sequences. Parameters such as E and e’ are excluded from routine evaluations due to their reliance on specialized acquisitions and post-processing. Although guideline-endorsed for LVDD diagnosis and cardiovascular risk stratification, LAVI has limited sensitivity: it reflects chronic LAP elevation through passive geometric adaptation but fails to capture phase-specific LA functional dynamics. In contrast, LA strain parameters overcome this limitation by quantifying myocardial deformation across cardiac phases, with validated sensitivity for detecting subclinical dysfunction in multiple studies (17-19). Our findings demonstrate that, compared to controls, HTN patients without elevated LAP show comparable LAVI but exhibit significantly reduced LA reservoir and conduit strain on CMR-FT. These results confirm that LA functional impairment precedes structural remodeling (16,17,20) and indicate that LA reservoir and conduit dysfunction in HTN emerges before guideline-defined LVDD, with CMR-FT enabling sensitive detection of these subclinical changes.

Further analysis revealed progressive impairment of LA strain parameters in HTN patients with increasing LAP. Interestingly, booster pump function parameters (LA booster EF, εa, SRa) remained preserved in HTN patients with normal LAP but exhibited significant decline in those with elevated LAP and LVDD—a pattern consistent with previous studies (21,22). This divergence may stem from the fact that reservoir and conduit functions, governed by atrioventricular compliance and LV filling pressures (15,23,24), are impaired early in hypertensive pathology. In contrast, booster pump function, primarily dependent on intrinsic contractility, initially compensates through preserved or mildly enhanced contraction to sustain LV filling despite reservoir and conduit dysfunction. With disease progression, however, sustained LA dilation exceeds compensatory reserves, triggering rapid functional deterioration (25). Additionally, our study revealed that LAVImin demonstrated superior sensitivity compared to the guideline-recommended parameter LAVImax, showing significant differences even between HTN patients with normal LAP and healthy controls. This phenomenon may be attributed to the direct exposure of the LA to LV pressure during LV end-diastole, when the mitral valve is fully open. Consequently, LAVImin is more closely associated with LV filling pressure and exhibits a stronger correlation with LVDD severity (26,27). Given its equivalent measurement feasibility to LAVImax in standard imaging protocols, LAVImin could serve as a practical marker for detecting subclinical LA remodeling.

Previous studies have preliminarily investigated correlations between CMR-derived LA strain parameters and indirect markers of LV diastolic dysfunction, such as LV peak diastolic strain rate (28) and mitral E/A ratio <1 (21). However, their associations with guideline-recommended diastolic indices (e’, LAVI, E/e’ ratio) remained unreported. Linear regression analyses demonstrated that LA reservoir strain and conduit strain independently correlated with e’ and LAVI, while booster pump strain independently correlated with the E/e’ ratio. Mechanistic analysis revealed significant associations of GLS and LVMI with reservoir and conduit strains, and of GRS with reservoir and booster pump strains, confirming LA-LV biomechanical coupling (21,29). These findings support LA strain parameters as complementary biomarkers for LV diastolic evaluation. Additionally, the inverse correlation between age and LA conduit strain aligns with population-level evidence (30,31).

Our analysis of LV strain parameters revealed a progressive worsening trend across the study subgroups. Among the three parameters, only GLS demonstrated statistically significant differences between patients with normal and elevated LAP, whereas GCS and GRS showed no significant variations. Consistent with this, Zhou et al. (16) reported that CMR-FT-derived LA strain parameters outperformed LV GLS in detecting early-stage dysfunction, suggesting superior sensitivity of LA metrics. These collective findings indicate that LA strain parameters may be more sensitive than LV strain for assessing diastolic dysfunction, particularly in subclinical hypertensive populations.

ROC analysis confirmed the robust discriminative capacity of CMR-derived LA strain parameters for distinguishing hypertensive patients with elevated versus normal LAP. Notably, εs demonstrated exceptional performance with an AUC of 0.896 (optimal cutoff: 29.55%), surpassing traditional diastolic indices and GLS. Furthermore, LAVImin showed significantly superior discriminative capacity than the guideline-recommended LAVImax, and the combination of εs and LAVImin provided the highest diagnostic accuracy. While LAVImax remains the only volumetric parameter recommended in current guidelines—supported by comprehensive normative data (adult upper limit: 34 mL/m²) and validated risk thresholds (6,32,33)—recent studies indicate that LAVImin more effectively detects early diastolic dysfunction and predicts the risk of heart failure and atrial fibrillation (26,27,34). Although some echocardiographic studies have reported lower reproducibility for LAVImin compared to LAVImax (35) and standardized reference values are still lacking, the growing adoption of CMR is expected to increase recognition of the clinical value of LA strain and LAVImin. In this study, CMR-derived LA strain parameters demonstrated good-to-excellent reproducibility, and CMR-derived LAVImax showed high agreement with TTE.

There are several limitations in this study. First, as a single-center study with a small sample size, selection bias is inevitable. Second, the duration of HTN and variations in antihypertensive medications may influence LA strain parameters, but the lack of stratified analysis of these factors could bias the interpretation. Third, in this study, elevated LAP diagnoses relied exclusively on retrospectively collected echocardiographic indices. Substantial evidence confirms that hemodynamic parameters acquired through dedicated CMR sequences (e.g., E, e’, and TRV) reliably assess diastolic function with high concordance to echocardiographic measures (36,37). Future prospective studies should validate comprehensive multiparametric CMR protocols for LVDD evaluation. Finally, while CMR-FT has clinical value, its application is limited by specialized post-processing software, increased workload, and vendor-dependent measurement variability (13). Promisingly, emerging AI tools with automated myocardial contouring and rapid strain quantification (38,39) may address these limitations.

Conclusions

CMR-derived LA strain parameters demonstrated high discriminatory power for identifying cardiac dysfunction in patients with HTN. The combination of reservoir strain (εs) and LAVImin showed strong ability to detect LVDD associated with elevated LAP, supporting their potential use in early identification of diastolic impairment. Based on these findings, we propose integrating LA reservoir strain and LAVImin as supplementary CMR parameters for diastolic function evaluation in hypertensive patients.

Acknowledgments

None.

Footnote

Reporting Checklist: The authors have completed the STROBE reporting checklist. Available at https://qims.amegroups.com/article/view/10.21037/qims-2025-1128/rc

Data Sharing Statement: Available at https://qims.amegroups.com/article/view/10.21037/qims-2025-1128/dss

Funding: This study was supported by the Guiding Science and Technology Program of Ganzhou Science and Technology Bureau (2025ZSFCE0557) and the Clinical Research Center for Medical Imaging in Jiangxi Province (20223BCG7400199).

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://qims.amegroups.com/article/view/10.21037/qims-2025-1128/coif). The authors have no conflicts of interest to declare.

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. This study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. The study protocol was approved by the Institutional Ethics Committee of Ganzhou People’s Hospital (No. MR-36-25-039496), and the requirement for written informed consent was waived for this retrospective study.

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. Mancia G, Kreutz R, Brunström M, Burnier M, Grassi G, Januszewicz A, et al. 2023 ESH Guidelines for the management of arterial hypertension The Task Force for the management of arterial hypertension of the European Society of Hypertension: Endorsed by the International Society of Hypertension (ISH) and the European Renal Association (ERA). J Hypertens 2023;41:1874-2071. [Crossref] [PubMed]
2. Ottosen CI, Nadruz W, Inciardi RM, Johansen ND, Fudim M, Biering-Sørensen T. Diastolic dysfunction in hypertension: a comprehensive review of pathophysiology, diagnosis, and treatment. Eur Heart J Cardiovasc Imaging 2024;25:1525-36. [Crossref] [PubMed]
3. Nadruz W, Shah AM, Solomon SD. Diastolic Dysfunction and Hypertension. Med Clin North Am 2017;101:7-17. [Crossref] [PubMed]
4. Shah KS, Xu H, Matsouaka RA, Bhatt DL, Heidenreich PA, Hernandez AF, Devore AD, Yancy CW, Fonarow GC. Heart Failure With Preserved, Borderline, and Reduced Ejection Fraction: 5-Year Outcomes. J Am Coll Cardiol 2017;70:2476-86. [Crossref] [PubMed]
5. Anker SD, Usman MS, Anker MS, Butler J, Böhm M, Abraham WT, et al. Patient phenotype profiling in heart failure with preserved ejection fraction to guide therapeutic decision making. A scientific statement of the Heart Failure Association, the European Heart Rhythm Association of the European Society of Cardiology, and the European Society of Hypertension. Eur J Heart Fail 2023;25:936-55. [Crossref] [PubMed]
6. Nagueh SF, Smiseth OA, Appleton CP, Byrd BF 3rd, Dokainish H, Edvardsen T, Flachskampf FA, Gillebert TC, Klein AL, Lancellotti P, Marino P, Oh JK, Popescu BA, Waggoner AD. Recommendations for the Evaluation of Left Ventricular Diastolic Function by Echocardiography: An Update from the American Society of Echocardiography and the European Association of Cardiovascular Imaging. J Am Soc Echocardiogr 2016;29:277-314. [Crossref] [PubMed]
7. Andersen OS, Smiseth OA, Dokainish H, Abudiab MM, Schutt RC, Kumar A, Sato K, Harb S, Gude E, Remme EW, Andreassen AK, Ha JW, Xu J, Klein AL, Nagueh SF. Estimating Left Ventricular Filling Pressure by Echocardiography. J Am Coll Cardiol 2017;69:1937-48. [Crossref] [PubMed]
8. Singh A, Addetia K, Maffessanti F, Mor-Avi V, Lang RM. LA Strain for Categorization of LV Diastolic Dysfunction. JACC Cardiovasc Imaging 2017;10:735-43. [Crossref] [PubMed]
9. Xu Y, Zhou JW, Su B, Sun YJ, Zhou J, Liu Y, Zhou BY, Zou C. Left atrial strain parameters to predicting elevated left atrial pressure in patients with atrial fibrillation. Echocardiography 2024;41:e15876. [Crossref] [PubMed]
10. Seo J, Yu HT, Kim IS, Kim TH, Joung B, Pak HN, Cho I, Shim CY, Hong GR, Ha JW. Novel algorithm for non-invasive estimation of left atrial pressure in patients with atrial fibrillation. Eur Heart J Cardiovasc Imaging 2025;26:414-21. [Crossref] [PubMed]
11. Robinson S, Ring L, Oxborough D, Harkness A, Bennett S, Rana B, Sutaria N, Lo Giudice F, Shun-Shin M, Paton M, Duncan R, Willis J, Colebourn C, Bassindale G, Gatenby K, Belham M, Cole G, Augustine D, Smiseth OA. The assessment of left ventricular diastolic function: guidance and recommendations from the British Society of Echocardiography. Echo Res Pract 2024;11:16. [Crossref] [PubMed]
12. van Everdingen WM, Zweerink A, Nijveldt R, Salden OAE, Meine M, Maass AH, Vernooy K, De Lange FJ, van Rossum AC, Croisille P, Clarysse P, Geelhoed B, Rienstra M, Van Gelder IC, Vos MA, Allaart CP, Cramer MJ. Comparison of strain imaging techniques in CRT candidates: CMR tagging, CMR feature tracking and speckle tracking echocardiography. Int J Cardiovasc Imaging 2018;34:443-56. [Crossref] [PubMed]
13. Pathan F, Zainal Abidin HA, Vo QH, Zhou H, D'Angelo T, Elen E, Negishi K, Puntmann VO, Marwick TH, Nagel E. Left atrial strain: a multi-modality, multi-vendor comparison study. Eur Heart J Cardiovasc Imaging 2021;22:102-10. [Crossref] [PubMed]
14. Wen J, Zhang X, Li S, Tao X, Fang Q, Zhou S, Xia L, Gong L. Identification of heart failure with preserved ejection fraction in patients with hypertension: a left atrial myocardial strain cardiac magnetic resonance study. Quant Imaging Med Surg 2023;13:2881-94. [Crossref] [PubMed]
15. Thomas L, Marwick TH, Popescu BA, Donal E, Badano LP. Left Atrial Structure and Function, and Left Ventricular Diastolic Dysfunction: JACC State-of-the-Art Review. J Am Coll Cardiol 2019;73:1961-77. [Crossref] [PubMed]
16. Zhou D, Yang W, Yang Y, Yin G, Li S, Zhuang B, Xu J, He J, Wu W, Jiang Y, Sun X, Wang Y, Sirajuddin A, Zhao S, Lu M. Left atrial dysfunction may precede left atrial enlargement and abnormal left ventricular longitudinal function: a cardiac MR feature tracking study. BMC Cardiovasc Disord 2022;22:99. [Crossref] [PubMed]
17. Liu J, Li J, Xia C, He W, Li X, Wang Y, Shen S, Tong N, Peng L. Diastolic dysfunction in adults with uncomplicated obesity evaluated with left atrial and left ventricular tissue tracking and ventricular volume-time curve: a prospective cardiac magnetic resonance study. Quant Imaging Med Surg 2024;14:5040-56. [Crossref] [PubMed]
18. Shao G, Cao Y, Cui Y, Han X, Liu J, Li Y, Li N, Liu T, Yu J, Shi H. Early detection of left atrial and bi-ventricular myocardial strain abnormalities by MRI feature tracking in normotensive or hypertensive T2DM patients with preserved LV function. BMC Cardiovasc Disord 2020;20:196. [Crossref] [PubMed]
19. Yang Y, Yin G, Jiang Y, Song L, Zhao S, Lu M. Quantification of left atrial function in patients with non-obstructive hypertrophic cardiomyopathy by cardiovascular magnetic resonance feature tracking imaging: a feasibility and reproducibility study. J Cardiovasc Magn Reson 2020;22:1. [Crossref] [PubMed]
20. Stefani LD, Trivedi SJ, Ferkh A, Emerson P, Marschner S, Gan G, Altman M, Thomas L. Left atrial mechanics evaluated by two-dimensional strain analysis: alterations in essential hypertension. J Hypertens 2024;42:274-82. [Crossref] [PubMed]
21. Li L, Chen X, Yin G, Yan W, Cui C, Cheng H, Lu M, Zhao S. Early detection of left atrial dysfunction assessed by CMR feature tracking in hypertensive patients. Eur Radiol 2020;30:702-11. [Crossref] [PubMed]
22. Jarasunas J, Aidietis A, Aidietiene S. Left atrial strain - an early marker of left ventricular diastolic dysfunction in patients with hypertension and paroxysmal atrial fibrillation. Cardiovasc Ultrasound 2018;16:29. [Crossref] [PubMed]
23. Inoue K, Khan FH, Remme EW, Ohte N, García-Izquierdo E, Chetrit M, Moñivas-Palomero V, Mingo-Santos S, Andersen ØS, Gude E, Andreassen AK, Wang TKM, Kikuchi S, Stugaard M, Ha JW, Klein AL, Nagueh SF, Smiseth OA. Determinants of left atrial reservoir and pump strain and use of atrial strain for evaluation of left ventricular filling pressure. Eur Heart J Cardiovasc Imaging 2021;23:61-70. [Crossref] [PubMed]
24. Hoit BD. Left Atrial Reservoir Strain: Its Time Has Come. JACC Cardiovasc Imaging 2022;15:392-4. [Crossref] [PubMed]
25. Mehrzad R, Rajab M, Spodick DH. The three integrated phases of left atrial macrophysiology and their interactions. Int J Mol Sci 2014;15:15146-60. [Crossref] [PubMed]
26. Nguyen J, Weber J, Hsu B, Mulyala RR, Wang L, Cao JJ. Comparing left atrial indices by CMR in association with left ventricular diastolic dysfunction and adverse clinical outcomes. Sci Rep 2021;11:21331. [Crossref] [PubMed]
27. Ben-Arzi A, Hazanov E, Ghanim D, Rozen G, Marai I, Grosman-Rimon L, Kachel E, Amir O, Carasso S. Left atrial minimal volume: association with diastolic dysfunction and heart failure in patients in sinus rhythm or atrial fibrillation with preserved ejection fraction. BMC Med Imaging 2021;21:76. [Crossref] [PubMed]
28. Song Y, Li L, Chen X, Shao X, Lu M, Cheng J, Zhao S. Early Left Ventricular Diastolic Dysfunction and Abnormal Left Ventricular-left Atrial Coupling in Asymptomatic Patients With Hypertension: A Cardiovascular Magnetic Resonance Feature Tracking Study. J Thorac Imaging 2022;37:26-33. [Crossref] [PubMed]
29. Han PL, Li K, Jiang Y, Jiang L, Tang X, Guo YK, Li Y, Yang ZG. Left atrioventricular coupling and left atrial abnormality in patients with acute myocardial infarction with and without hyperglycemia assessed with 3.0T cardiac magnetic resonance imaging feature tracking. Quant Imaging Med Surg 2025;15:2347-61. [Crossref] [PubMed]
30. Eckstein J, Körperich H, Paluszkiewicz L, Burchert W, Piran M. Multi-parametric analyses to investigate dependencies of normal left atrial strain by cardiovascular magnetic resonance feature tracking. Sci Rep 2022;12:12233. [Crossref] [PubMed]
31. Truong VT, Palmer C, Wolking S, Sheets B, Young M, Ngo TNM, Taylor M, Nagueh SF, Zareba KM, Raman S, Mazur W. Normal left atrial strain and strain rate using cardiac magnetic resonance feature tracking in healthy volunteers. Eur Heart J Cardiovasc Imaging 2020;21:446-53. [Crossref] [PubMed]
32. Pieske B, Tschöpe C, de Boer RA, Fraser AG, Anker SD, Donal E, et al. How to diagnose heart failure with preserved ejection fraction: the HFA-PEFF diagnostic algorithm: a consensus recommendation from the Heart Failure Association (HFA) of the European Society of Cardiology (ESC). Eur J Heart Fail 2020;22:391-412. [Crossref] [PubMed]
33. Almeida P, Rodrigues J, Lourenço P, Maciel MJ, Bettencourt P. Left atrial volume index is critical for the diagnosis of heart failure with preserved ejection fraction. J Cardiovasc Med (Hagerstown) 2018;19:304-9. [Crossref] [PubMed]
34. Bertelsen L, Diederichsen SZ, Haugan KJ, Brandes A, Graff C, Krieger D, Kronborg C, Køber L, Højberg S, Vejlstrup N, Svendsen JH. Left atrial volume and function assessed by cardiac magnetic resonance imaging are markers of subclinical atrial fibrillation as detected by continuous monitoring. Europace 2020;22:724-31. [Crossref] [PubMed]
35. Badano LP, Miglioranza MH, Mihăilă S, Peluso D, Xhaxho J, Marra MP, Cucchini U, Soriani N, Iliceto S, Muraru D. Left Atrial Volumes and Function by Three-Dimensional Echocardiography: Reference Values, Accuracy, Reproducibility, and Comparison With Two-Dimensional Echocardiographic Measurements. Circ Cardiovasc Imaging 2016;9:e004229. [Crossref] [PubMed]
36. Fujikura K, Sathya B, Acharya T, Benovoy M, Jacobs M, Sachdev V, Hsu LY, Arai AE. CMR provides comparable measurements of diastolic function as echocardiography. Sci Rep 2024;14:11658. [Crossref] [PubMed]
37. Lamy J, Xiang J, Shah N, Kwan JM, Kim Y, Upadhyaya K, Reinhardt SW, Meadows J, McNamara RL, Baldassarre LA, Peters DC. Diastolic dysfunction evaluation by cardiovascular magnetic resonance derived E, a, e': Comparison to echocardiography. Physiol Rep 2024;12:e70078. [Crossref] [PubMed]
38. Ghadimi S, Auger DA, Feng X, Sun C, Meyer CH, Bilchick KC, Cao JJ, Scott AD, Oshinski JN, Ennis DB, Epstein FH. Fully-automated global and segmental strain analysis of DENSE cardiovascular magnetic resonance using deep learning for segmentation and phase unwrapping. J Cardiovasc Magn Reson 2021;23:20. [Crossref] [PubMed]
39. Zhao K, Zhu Y, Chen X, Yang S, Yan W, Yang K, et al. Machine Learning in Hypertrophic Cardiomyopathy: Nonlinear Model From Clinical and CMR Features Predicting Cardiovascular Events. JACC Cardiovasc Imaging 2024;17:880-93. [Crossref] [PubMed]