Dynamic monitoring of left atrial strain: a novel paradigm for early detection of cardiotoxicity in colorectal cancer patients undergoing chemotherapy

Introduction

Colorectal cancer ranks as the third most prevalent malignancy globally (1). Fluorouracil-based chemotherapy remains a cornerstone treatment for advanced gastrointestinal malignancies, including colorectal cancer, gastric cancer, and small bowel adenocarcinoma (2). Nevertheless, its clinical utility is limited by cardiovascular toxicity and variable responses to targeted agents (3). Cardiotoxicity associated with this regimen often presents with asymptomatic or nonspecific manifestations, potentially leading to underdiagnosis in clinical practice (4). Chemotherapy-induced cardiotoxicity has emerged as a critical determinant of long-term outcomes in cancer patients (5), necessitating vigilant cardiac monitoring during treatment.

LA strain has been established as a sensitive marker for subclinical or early-stage diastolic dysfunction (6). Growing evidence demonstrates its superior accuracy over conventional Doppler-derived parameters in assessing left ventricular (LV) filling pressures (7). Functioning as the hemodynamic nexus between pulmonary and systemic circulation, LA strain may better reflect LV diastolic performance by modulating ventricular loading conditions (8). Notably, expert consensus indicates that diastolic dysfunction typically precedes systolic impairment in chemotherapy-related cardiotoxicity (9). While left ventricular global longitudinal strain (LVGLS) shows high sensitivity for detecting LV dysfunction in cancer patients (10), the role of LA strain remains insufficiently characterized.

This study systematically evaluates the temporal evolution of LA strain parameters during the course of chemotherapy in colorectal cancer patients, addressing three pivotal questions: (I) the sequential impact of chemotherapy on LA mechanical efficiency and its relationship with LVGLS changes; (II) heterogeneous patterns of LA functional impairment and associated risk factors; and (III) the predictive value of LA strain parameters for diastolic function deterioration. We present this article in accordance with the STROBE reporting checklist (available at https://qims.amegroups.com/article/view/10.21037/qims-2025-1909/rc).

Methods

Study population

This prospective study consecutively recruited all eligible patients at The Second Hospital of Lanzhou University between January 2023 and January 2024. The study initially enrolled 145 patients scheduled for fluorouracil-based combination chemotherapy at our institution. Inclusion criteria: (I) histologically confirmed colorectal adenocarcinoma requiring fluorouracil-containing regimens; (II) age ≥18 years; (III) normal hepatic/renal function [alanine aminotransferase (ALT)/aspartate aminotransferase (AST) ≤2.5×ULN, estimated glomerular filtration rate (eGFR) ≥60 mL/min] with expected survival ≥6 months. Exclusion criteria: (I) prior mediastinal/thoracic radiotherapy or anthracycline exposure (e.g., doxorubicin/epirubicin); (II) resting electrocardiogram (ECG) abnormalities: ST depression ≥0.1 mV or T-wave inversion; heart rate >100 bpm; frequent PVCs (>5/min) or complex arrhythmias; (III) cardiac structural/functional abnormalities: left ventricular ejection fraction (LVEF) <50%; moderate-severe aortic stenosis; New York Heart Association (NYHA) class III–IV heart failure; (IV) suboptimal echocardiographic image quality (≥2 non-assessable views) or inability to complete ≥3 chemotherapy cycles; (V) patients who failed to complete the entire follow-up or had incomplete data. Treatment regimens: all enrolled patients received standardized fluoropyrimidine-based chemotherapy, including: FOLFOX (oxaliplatin 85 mg/m² + leucovorin 400 mg/m² intravenous infusion + 5-fluorouracil 400 mg/m² intravenous bolus followed by 2,400 mg/m² continuous infusion over 46 hours, repeated every 2 weeks), XELOX (oxaliplatin 130 mg/m² intravenous infusion + capecitabine 1,000 mg/m² orally twice daily, repeated every 3 weeks), or capecitabine monotherapy (1,250 mg/m² orally twice daily, repeated every 3 weeks). The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. The study protocol was approved by the Ethics Committee of The Second Hospital of Lanzhou University (No. 2023A-436). Informed consent was obtained from all individual participants prior to their enrollment in the study.

Baseline data

Detailed demographic characteristics (age, sex, body mass index), cardiovascular risk factors (smoking history, hypertension, diabetes, dyslipidemia), treatment history [prior cancer treatment regimens and use of cardioprotective medications—including angiotensin-converting enzyme (ACE) inhibitors, angiotensin receptor blockers (ARBs), β-blockers, and statins], and laboratory tests (complete blood count and biochemical parameters) were collected through clinical medical records and patient interviews. All patients underwent standardized cardiac assessments (12-lead electrocardiography and echocardiography). New-onset symptoms during the entire chemotherapy cycle were documented for all patients.

Echocardiography

A standardized echocardiographic follow-up protocol was implemented in this study, with assessments performed at four key time points over 12 months: baseline before chemotherapy (T0), after the first cycle (T1), mid-chemotherapy (T2), and at the end of chemotherapy (T3). All examinations strictly adhered to the standardized operating procedures outlined in the joint guidelines of the American Society of Echocardiography and the European Association of Cardiovascular Imaging [2015] (11). Examinations were conducted using a GE E95 ultrasound system equipped with an M5S transducer (3.5 MHz) by cardiac sonographers with at least 5 years of experience. Key assessment parameters included: (I) ventricular function assessment: the biplane Simpson’s method was applied in apical four-chamber and two-chamber views to measure left ventricular end-diastolic volume (LVEDV), left ventricular end-systolic volume (LVESV), and LVEF. (II) Cardiac structural measurements: parasternal long-axis view: interventricular septal thickness at end-diastole (IVSd), left ventricular posterior wall thickness at end-diastole (LVPWd), and left atrial anteroposterior diameter (LAD). (III) Apical four-chamber view: left atrial volume (LAV, measured by the area-length method) and body surface area-indexed left atrial volume (LAVI). (IV) Myocardial mass calculation: The Devereux-modified formula was used: LVM (g) =0.8×1.04×[(IVSD + LVPWd + LVIDd)³ − LVIDd³]+0.6. (V) Diastolic function assessment: Pulsed-wave Doppler: mitral inflow E-wave and A-wave velocities, E/A ratio. Tissue Doppler: average mitral annular e’ velocity (septal and lateral walls), E/e’ ratio. Continuous-wave Doppler: tricuspid regurgitation peak velocity (TR).

Strain analysis

All patients were positioned in the standard left lateral decubitus position with continuous ECG monitoring. Dynamic images of apical four-, three-, and two-chamber views were acquired (frame rate ≥70 fps, with ≥3 cardiac cycles stored). Gain settings were adjusted to ensure clear endocardial visualization without excessive reverberation. Three consecutive cardiac cycles were stored and analyzed using dedicated software (EchoPAC PC).

For GLS, images were imported into EchoPAC PC software, where manual marking of the mitral annulus and apical points defined the LV region of interest (ROI). The software automatically tracked myocardial speckle motion and generated 17-segment strain curves, ultimately providing the average LVGLS.

For LA strain analysis, a dedicated “LA strain” module was used. The endocardial border was manually traced at end-diastole (R-wave peak), delineating the LA endocardium from the mitral annulus to the opposite annulus while avoiding pulmonary vein ostia and the LA appendage. The system automatically generated an ROI, and automatic tracking was accepted unless manual correction was required (displacement error >2 mm) (12). Key parameters obtained included: LA reservoir strain (LASr): peak positive strain during ventricular systole. LA conduit strain (LAScd): strain decline at the E-wave phase. LA contractile strain (LASct): negative strain at P-wave onset. Peak values for each parameter (LASr-P, LAScd-P, LASct-P) were also recorded (Figure 1). To ensure objective assessment, Speckle-tracking analysis and conventional diastolic parameter assessments were performed independently by different researchers, each of whom remained blinded to the other’s results as well as to the patients’ clinical information.

Figure 1 Acquisition of GLS and LA strain. (A) The process for obtaining GLS. (B) The process for obtaining LA strain. GLS, global longitudinal strain; LA, left atrial.

Statistical analysis

Continuous variables were expressed as mean ± standard deviation (normally distributed) or median (interquartile range) (non-normally distributed) based on normality tests. Categorical variables were described as counts (percentages). For between-group comparisons, normally distributed data were analyzed using one-way analysis of variance (ANOVA), while non-normally distributed data were assessed using the Friedman test. Categorical variables were compared using the chi-square test or Fisher’s exact test, as appropriate.

The relative change rate of strain parameters from baseline was calculated and compared [change rate = (follow-up value – baseline value) / baseline value × 100%]. Correlation matrix analysis and association analyses were performed to explore the relationship between LA strain parameters and diastolic function parameters. Forest plots were used to illustrate subgroup differences, and interaction tests were conducted to assess subgroup effects. Multivariable linear regression covariates were preselected based on clinical relevance and biological plausibility. Variable selection did not rely on univariate P-values, but employed a prespecified hierarchical modeling approach to systematically control for confounding factors. Receiver operating characteristic (ROC) curve analysis was employed to evaluate the predictive value of diastolic function and LA strain parameters for diastolic dysfunction. Scatter plots were generated to analyze the correlation between LVGLS and LA strain parameters. Patients were stratified into two groups based on a >15% decline in ΔGLS from baseline during chemotherapy for comparative analysis of baseline characteristics.

A linear mixed-effects model was applied to analyze the dynamic changes in LASr, with time (T0–T3) as a fixed effect, adjusted for age and sex, and incorporating a random intercept. Additionally, latent class growth analysis (LCGA) was performed to identify distinct trajectories of LASr changes. Parameter estimation was conducted using restricted maximum likelihood (REML), with the optimal classification determined based on the Bayesian Information Criterion (BIC) and entropy values (>0.7).

Two experienced cardiac sonographers independently analyzed the echocardiographic images. A random subset of 20 cases was selected for blinded remeasurement (with a 2-week interval) to assess LVGLS and LA strain parameters (LASr, LAScd, LASct, and their peak values). Intra- and inter-observer reliability was evaluated using the intraclass correlation coefficient (ICC), with ICC >0.90 considered excellent and 0.75–0.90 considered good agreement.

All statistical analyses were performed using R 3.3.2 (http://www.R-project.org, The R Foundation) and Free Statistics software version 1.9.1. A two-sided P<0.05 was considered statistically significant.

Results

Patient characteristics

A total of 145 patients met the inclusion criteria (Figure S1). Of these, 24 were excluded after enrollment (22 switched to alternative chemotherapy regimens, 2 withdrew consent). Twelve were excluded due to poor echocardiographic image quality, 11 did not complete the full chemotherapy course, and 4 were lost to follow-up. Thus, the final study population comprised 94 colorectal cancer patients (57.4% male, mean age 54.0±13.0 years). Most received the FOLFOX regimen (61.7%), and hypertension was the most prevalent cardiovascular risk factor (23.4%). Key laboratory findings demonstrated normal myocardial enzymes (CK 58.0 U/L, CK-MB 11.0 U/L) and hemoglobin levels (125.0±21.4 g/dL). Baseline cardiac parameters were within normal ranges (SBP 117.6±12.7 mmHg, heart rate 72.3±10.3 bpm) (Table 1).

Temporal changes in echocardiographic parameters during chemotherapy

Longitudinal analysis revealed progressive deterioration in myocardial deformation parameters, whereas conventional echocardiographic measures remained largely stable. Most structural parameters (LV volumes, dimensions, wall thickness) and traditional functional indices (LVEF, FS, E/e′) exhibited no significant changes. LVGLS demonstrated marked progressive worsening from baseline (−21.9%±1.9%) to T3 (−19.1%±2.1%, P<0.001), accompanied by significant declines in LASr (40.3%±4.1% to 35.2%±4.6%, P<0.001) and LAScd (22.7%±5.1% to 18.4%±5.5%, P<0.001). A modest increase in LV mass index was observed (88.1±14.9 to 96.4±17.1 g/m², P=0.003) (Table 2).

Analysis of decline rates showed progressive deterioration in all functional parameters (P<0.05). The reduction in GLS escalated from 7.2% at T1 to 12.2% at T3, while LASr declined from 7.3% to 12.3%. The most pronounced deterioration occurred in LAScd (14.2% to 18.8%). Notably, LASct exhibited no significant decline until the end of chemotherapy (6.4%) (Table S1).

Correlation matrix analysis between LA strain parameters and conventional diastolic function indices

The correlation matrix analysis revealed significant associations between conventional echocardiographic parameters and LA strain indices (Figure 2). At all time points (T0–T3), LASr demonstrated strong negative correlations with E/e’ (−0.85 to −0.77), TR velocity (−0.81 to −0.78), and LAVI (−0.84 to −0.78), while showing positive correlations with e’ (0.35 to 0.46). LAScd exhibited moderate negative correlations with E/e’ (−0.58 to −0.46), whereas LASct showed weaker associations. Excellent consistency was observed among strain parameters (LASr/LASr-P: 0.93–0.96; LAScd/LAScd-P: 0.95–0.98; LASct/LASct-P: 0.90–0.95), while LAScd and LASct were negatively correlated (−0.58 to −0.54). The E/A ratio generally showed weak correlations with other parameters (|r|<0.3) (Table S2, Figure 2).

Figure 2 Correlation matrix plot between conventional echocardiographic parameters and left atrial strain parameters. (A) T0, baseline period. (B) T1, after the first cycle of chemotherapy. (C) T2, mid-chemotherapy period. (D) T3, end of chemotherapy. LAScd, left atrial conduit strain; LAScd-p, left atrial conduit strain-peak; LASct, left atrial contractile strain; LASct-p, left atrial contractile strain-peak; LASr, left atrial reservoir strain; LASr-p, left atrial reservoir strain-peak; LAVI, left atrial volume index; TR, tricuspid regurgitation.

Association analysis between LA strain and diastolic function parameters

To further investigate the relationship between LA strain and diastolic function parameters, we performed linear regression analyses. Conventional diastolic parameters (E/A, e’-sep, e’-lat, E/e’, LAVI, TR) were standardized and combined into a composite conventional diastolic parameter (CDP). Univariate analysis demonstrated strong correlations between CDP and LA strain measurements, particularly with LASr [β=1 (0.89, 1.1), P<0.001] and LAScd [β=0.85 (0.62, 1.08), P<0.001]. Diastolic dysfunction markers (E/e’, LAVI) showed negative correlations with LA strain (all P<0.001), while tissue Doppler velocities (e’-sep, e’-lat) exhibited positive correlations (P≤0.003). LASr-P/ LAScd-P demonstrated stronger associations than non-peak values (Tables S3,S4).

Multivariate linear regression analysis (Table 3) revealed positive associations between conventional diastolic function and LA strain parameters in all models (all P<0.05). The strongest associations were observed for LASr-P (coefficient =1.07, 95% CI: 0.92–1.22 in crude model) and LASr (coefficient =1, 95% CI: 0.89–1.11). These correlations remained robust after adjusting for age and sex (Model I), with slightly strengthened coefficients following additional adjustment for treatment regimen and cardiovascular risk factors (Model II). In the fully adjusted model incorporating cardiac structural and functional parameters (Model III), LASr and LASr-P maintained strong associations (coefficients =0.95 and 0.93 respectively; both P<0.001), while LAScd and LAScd-P exhibited attenuated correlations (coefficients =0.57 and 0.65; P=0.038 and 0.017 respectively).

Subgroup analyses (Table S5, Figure 3) demonstrated significant CDP-LASr associations across all subgroups (P<0.001). Age ≥60 years (β=1.22 vs. 0.88, P=0.01) and ACE inhibitor use (β=0.65) significantly modified these associations, while treatment regimens, risk factors, and ΔGLS showed no significant effects (all P>0.05). No significant differences were observed among subgroups stratified by chemotherapy regimen, cardiovascular risk factors, or cardiac functional status (ΔGLS >15%).

Figure 3 Association between CDP and left atrial function across subgroups. ACEI, angiotensin-converting enzyme inhibitor; CDP, conventional diastolic parameters; CI, confidence interval; GLS, global longitudinal strain.

ROC curve analysis of LA strain parameters versus conventional diastolic function parameters

ROC curve analysis demonstrated that LASr exhibited superior diagnostic performance [area under the curve (AUC): 0.792, sensitivity: 75.5%, specificity: 91.5%] compared to conventional diastolic parameters (AUC: 0.503–0.603) and other strain measurements. These findings position LASr as an advanced marker for diastolic function assessment (Figure S2, Table S6).

Model 1 (conventional diastolic parameters) included: E/A + e’-sep + e’-lat + E/e’ + LAVI + TR.

Model 2 (LA strain parameters) included: LASr + LAScd + LASct + LASr-P + LAScd-P + LASct-P. Compared to Model 1, Model 2 showed significantly improved diagnostic performance with:

• Higher AUC (0.836 vs. 0.617, P<0.001); improved sensitivity (84.04% vs. 64.89%); enhanced specificity (70.21% vs. 55.32%); superior Youden index (1.5426); better positive predictive value (89.43%); higher accuracy rate (80.59%); reduced false-negative rate by nearly 20 percentage points (15.96% vs. 35.11%). These comprehensive metrics demonstrate the clear advantage of the strain parameter model over conventional assessment (Figure 4, Table 4).

Figure 4 ROC comparison between conventional diastolic function and left atrial strain parameter models. Model 1 incorporated the following parameters: E/A ratio, e’-sep, e’-lat, E/e’ ratio, LAVI and TR. Model 2 included comprehensive left atrial strain parameters (LASr, LAScd, LASct) and their peak values (LASr-P, LAScd-P, LASct-P). AUC, area under the curve; CI, confidence interval; ROC, receiver operating characteristic.

Heterogeneous trajectories of LA strain decline during chemotherapy

Longitudinal analysis revealed progressive decline in LASr throughout chemotherapy (Table S7). After covariate adjustment, baseline LASr averaged 39.15%, with significant reductions of 3.18%, 4.28%, and 5.08% at T1, T2, and T3 respectively (all P<0.001). Neither sex nor age significantly influenced this pattern, though substantial inter-individual baseline variability was observed with consistent decline rates (random effects distribution shown in Figure S3).

To characterize LASr progression patterns, we employed LCGA modeling LASr trajectories across T0-T3 timepoints, adjusted for sex and age. The 2-class solution demonstrated optimal fit with: lowest BIC (1,423.86); highest entropy (0.80); all posterior probabilities ≥0.86. The final model identified two distinct trajectories (Table 5, Figure 5): Class 1 (stable/slow-declining): 75.5% of participants showing mean LASr decline of 2%; Class 2 (rapid-declining): 24.5% of participants demonstrating mean LASr reduction of 6%. These trajectories differed significantly (P<0.001, Figure S4), revealing clinically meaningful subpopulations with divergent LA reservoir function evolution during chemotherapy.

Figure 5 Comparison of individual weighted prediction trajectories across class 2–5 models. (A) 2-class model; (B) 3-class model; (C) 4-class model; (D) 5-class model. The figure illustrates the differential evolving patterns of left atrial function across distinct patient subgroups during chemotherapy.

Exploration of the relationship between LVGLS and LA strain parameters

Pearson correlation analysis was performed to examine the associations between LVGLS and LA strain/diastolic function parameters. At baseline (T0), LVGLS showed no significant correlations with any parameters. During chemotherapy, significant negative correlations emerged between LVGLS and both E/e’ (T1–T3: r=−0.35 to −0.38) and LAVI (r=−0.36 to −0.42) (P<0.05), while positive correlations were observed with LASr (r=0.38 to 0.39, P<0.001). LAScd demonstrated weak correlations with LVGLS (r=0.20 to 0.26), whereas LASct showed no significant association. The strongest negative correlation between TR and LVGLS occurred at mid-chemotherapy (T2: r=−0.39). LASr-P exhibited similar correlations with LVGLS as LASr (r=0.33 to 0.35) (Table S8, Figure 6).

Figure 6 Correlation between GLS and left atrial strain. (A) The scatter plot of LASr versus LVGLS. (B) The scatter plot of LAScd versus LVGLS. (C) The scatter plot of LASct versus LVGLS. (D) The scatter plot of LASr-P versus LVGLS. (E) The scatter plot of LAScd-P versus LVGLS. (F) The scatter plot of LASct-P versus LVGLS. GLS, global longitudinal strain; LAScd, left atrial conduit strain; LAScd-p, left atrial conduit strain-peak; LASct, left atrial contractile strain; LASct-p, left atrial contractile strain-peak; LASr, left atrial reservoir strain; LVGLS, left ventricular global longitudinal strain.

Comparative analysis of baseline characteristics between patients with different magnitudes of GLS decline (ΔGLS >15% vs. ≤15%) revealed that the >15% group had significantly higher prevalence of hypertension (40.7% vs. 16.4%, P=0.007) and cardiovascular medication use (P=0.001). While conventional cardiac function parameters showed no intergroup differences, the key finding was significantly reduced LA contractile function in the >15% group (LASct: 16.1% vs. 18.2%, P=0.012) (Table S9). These findings suggest the existence of distinct subpopulations with heterogeneous progression patterns in LA reservoir function.

Reproducibility analysis

Both intra- and inter-observer reproducibility for LA strain parameters (LASr, LAScd, LASct) and LVGLS demonstrated excellent agreement (all ICC >0.85), indicating high measurement reliability (Table S10).

Discussion

This prospective study evaluating subclinical cardiac functional changes throughout chemotherapy in colorectal cancer patients reveals the unique value of LA strain parameters in early detection of chemotherapy-associated subclinical cardiac injury. The main findings include: (I) significant correlations between LA strain and conventional diastolic function parameters, with LASr demonstrating superior diagnostic performance; (II) heterogeneous temporal patterns of LA strain decline during chemotherapy (slow-declining vs. rapid-declining trajectories); (III) moderate association between LA strain and LVGLS, suggesting baseline LA function may influence the degree of post-chemotherapy myocardial mechanical impairment in patients developing subclinical cardiotoxicity. These findings support the incorporation of LA strain assessment in routine cardiac monitoring during chemotherapy.

First, we identified significant negative correlations between LA strain parameters (LASr, LAScd) and conventional diastolic function indices (E/e’, LAVI), which remained robust after adjusting for multiple confounders including age, sex, and treatment regimens. These findings align with previous studies (13), although our analysis further revealed this association to be more pronounced in elderly patients, potentially attributable to their reduced myocardial reserve and increased susceptibility to chemotherapy toxicity (14). Subgroup analysis in our study suggested a weaker association between conventional diastolic parameters and LASr in ACEI users (adjusted coefficient: 0.65 vs. 1.06). Although not statistically significant (interaction P=0.179), this finding indicates that ACEIs may potentially mitigate chemotherapy-induced atrial dysfunction through cardiac remodeling improvement. However, this observation requires further validation due to sample size limitations. Notably, the diagnostic model incorporating LA strain parameters (LASr, LAScd, LASct and their peak values) demonstrated superior performance compared to the conventional diastolic function model (LAVI, e’-lat, e’-sep, E/e’, E/A, TR) (AUC 0.836 vs. 0.617, P<0.001), with substantially improved sensitivity. LASr exhibited significantly better diagnostic accuracy for diastolic dysfunction than conventional parameters (15-18), with its high sensitivity and specificity establishing it as an ideal marker for early detection of cardiac mechanical injury.

Secondly, our longitudinal analysis revealed unique dynamic evolution patterns of LA strain. Unlike previous studies (19), we observed dose-dependent declines in LASr during chemotherapy, with the most pronounced reduction occurring during the first cycle (20), accompanied by significant interindividual variability. LCGA identified two distinct progression trajectories: approximately 75% of patients exhibited gradual decline, while 25% demonstrated rapid deterioration. This observed heterogeneity may reflect underlying genetic predisposition and pharmacometabolic variations (21), suggesting that routine LASr monitoring could identify high-risk subgroups requiring more aggressive intervention - a finding warranting further investigation. Notably, LAScd showed the most marked deterioration, exceeding LVGLS impairment, while LASct changes appeared relatively delayed. This temporal hierarchy suggests particular susceptibility of LA function to chemotherapy-induced mechanical injury (19,22), with diastolic abnormalities potentially preceding systolic dysfunction (23). The early manifestation is predominantly impaired passive filling function, while active contractile function remains unaffected until advanced stages of the disease (24). The underlying mechanisms may involve direct cardiovascular toxicity of fluoropyrimidine drugs, with endothelial dysfunction serving as the core initiating factor. These drugs disrupt microcirculatory homeostasis by damaging vascular endothelial cells, subsequently inducing coronary microvascular ischemia. The LA wall, being thin and metabolically active, is particularly vulnerable to ischemia, making its diastolic function one of the earliest to be compromised. Furthermore, subclinical forms of coronary vasospasm may contribute to this process by causing myocardial stunning (25-27). Collectively, these mechanisms suggest that fluoropyrimidines may preferentially affect the function of the left atrium, a structure highly sensitive to blood supply.

Our study elucidates important relationships between LVGLS and LA function in chemotherapy patients. While conventional perspectives suggest atrial dysfunction follows ventricular injury (19), we demonstrated that post-chemotherapy LVGLS positively correlates with LASr and negatively correlates with diastolic parameters (E/e’, LAVI). This indicates that impaired ventricular contraction may affect atrial function through increased afterload (7,28). These findings underscore the necessity of monitoring both atrial and ventricular function during chemotherapy. Notably, patients with >15% LVGLS reduction exhibited higher hypertension prevalence and more pronounced LA contractile dysfunction (2.1% reduction in LASct) (29,30), suggesting baseline cardiovascular status may influence susceptibility to chemotherapy-induced cardiotoxicity. These results support the clinical utility of strain imaging for early cardiotoxicity detection and suggest: (I) intensified cardiac monitoring for patients with cardiovascular risk factors; (II) incorporation of baseline LA function assessment into risk stratification protocols.

Although this study primarily focused on imaging biomarkers, the early decline in LA reservoir strain was closely associated with important clinical outcomes. A reduction in LASr may predict subsequent progression of diastolic dysfunction, potentially leading to clinical decisions such as chemotherapy regimen modification or interruption. Additionally, impaired LA function is associated with an increased risk of symptomatic heart failure (particularly heart failure with preserved ejection fraction) and atrial arrhythmias. While the current study is limited by sample size and follow-up duration, the early changes in LASr provide a basis for identifying high-risk patients requiring close monitoring. Future prospective studies are warranted to validate its predictive value for clinical endpoints.

Limitations and future directions

This study has several limitations: (I) the single-center design may affect the generalizability of our findings; (II) the relatively small sample size restricts the power of subgroup analyses; (III) the lack of long-term follow-up data precludes assessment of the prognostic value of strain parameters; (IV) the absence of systematic cardiac biomarker monitoring limited two aspects: comparative analysis of diagnostic performance against strain parameters, and comprehensive assessment of cardiotoxicity; (V) the speckle-tracking echocardiography used in this study has the following technical limitations: inter-vendor algorithmic differences affect the comparability of measurements (31); LA strain is susceptible to cardiac loading conditions (32); insufficient image frame rates may lead to loss of motion details; and sternal depression may introduce bias in basal segment strain measurements (33). These factors represent potential sources of variation in the measurement results.

Future studies should address the following: (I) investigation of strain-based personalized intervention strategies; and (II) elucidation of the association between strain changes and long-term cardiovascular outcomes; (III) integrate serial biomarker measurements with strain imaging to establish a more comprehensive cardiotoxicity assessment system.

Conclusions

This study demonstrates that LA strain parameters, particularly LASr, serve as sensitive indicators for assessing diastolic function with superior diagnostic value. The identification of distinct trajectory patterns underscores variations in inherent individual susceptibility. These findings support the incorporation of LA strain monitoring into routine cardiac evaluation protocols for chemotherapy patients.

Acknowledgments

The authors thank all team members at the Ultrasound Medical Center, The Second Hospital of Lanzhou University for their helpful cooperation.

Footnote

Reporting Checklist: The authors have completed the STROBE reporting checklist. Available at https://qims.amegroups.com/article/view/10.21037/qims-2025-1909/rc

Data Sharing Statement: Available at https://qims.amegroups.com/article/view/10.21037/qims-2025-1909/dss

Funding: This study has received funding from Natural Science Foundation of Gansu Province, China (No. 24JRRA1096), Cuiying Scientific and Technological Innovation Program of The Second Hospital & Clinical Medical School, Lanzhou University (No. CY2024-QN-B04), and Gansu Province Clinical Research Center for Ultrasonography (No. 2020-0411-SFC-0021).

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://qims.amegroups.com/article/view/10.21037/qims-2025-1909/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. The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. The study protocol was approved by the Ethics Committee of The Second Hospital of Lanzhou University (No. 2023A-436). Informed consent was obtained from all individual participants prior to their enrollment in the 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. Ionescu VA, Diaconu CC, Gheorghe G, Mihai MM, Diaconu CC, Bostan M, Bleotu C. Gut Microbiota and Colorectal Cancer: A Balance Between Risk and Protection. Int J Mol Sci 2025;26:3733. [Crossref] [PubMed]
2. Li X, Ma Y, Wu J, Ni M, Chen A, Zhou Y, Dai W, Chen Z, Jiang R, Ling Y, Yao Q, Chen W. Thiol oxidative stress-dependent degradation of transglutaminase2 via protein S-glutathionylation sensitizes 5-fluorouracil therapy in 5-fluorouracil-resistant colorectal cancer cells. Drug Resist Updat 2023;67:100930. [Crossref] [PubMed]
3. Dell'Aquila E, Zeppola T, Stellato M, Pantano F, Scartozzi M, Madaudo C, Pietrantonio F, Cremolini C, Aprile G, Vincenzi B, Moretto R, Puzzoni M, Garattini SK, Lobefaro R, Tonini G, Santini D. Anti-EGFR Therapy in Metastatic Small Bowel Adenocarcinoma: Myth or Reality? Clin Med Insights Oncol 2020;14:1179554920946693. [Crossref] [PubMed]
4. Lestuzzi C, Stolfo D, De Paoli A, Banzato A, Buonadonna A, Bidoli E, Tartuferi L, Viel E, De Angelis G, Lonardi S, Innocente R, Berretta M, Bergamo F, Guglielmi A, Sinagra G, Herrmann J. Cardiotoxicity from Capecitabine Chemotherapy: Prospective Study of Incidence at Rest and During Physical Exercise. Oncologist 2022;27:e158-67. [Crossref] [PubMed]
5. Molnár AÁ, Birgés K, Surman A, Merkely B. The Complex Connection Between Myocardial Dysfunction and Cancer Beyond Cardiotoxicity: Shared Risk Factors and Common Molecular Pathways. Int J Mol Sci 2024;25:13185. [Crossref] [PubMed]
6. Nagueh SF, Khan SU. Left Atrial Strain for Assessment of Left Ventricular Diastolic Function: Focus on Populations With Normal LVEF. JACC Cardiovasc Imaging 2023;16:691-707. [Crossref] [PubMed]
7. 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]
8. Vos JL, Leiner T, van Dijk APJ, Pedrizzetti G, Alenezi F, Rodwell L, van der Wegen CTPM, Post MC, Driessen MMP, Nijveldt R. Cardiovascular magnetic resonance-derived left ventricular intraventricular pressure gradients among patients with precapillary pulmonary hypertension. Eur Heart J Cardiovasc Imaging 2022;24:78-87. [Crossref] [PubMed]
9. 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]
10. Stokke TM, Hasselberg NE, Smedsrud MK, Sarvari SI, Haugaa KH, Smiseth OA, Edvardsen T, Remme EW. Geometry as a Confounder When Assessing Ventricular Systolic Function: Comparison Between Ejection Fraction and Strain. J Am Coll Cardiol 2017;70:942-54. [Crossref] [PubMed]
11. Lang RM, Badano LP, Mor-Avi V, Afilalo J, Armstrong A, Ernande L, Flachskampf FA, Foster E, Goldstein SA, Kuznetsova T, Lancellotti P, Muraru D, Picard MH, Rietzschel ER, Rudski L, Spencer KT, Tsang W, Voigt JU. Recommendations for cardiac chamber quantification by echocardiography in adults: an update from the American Society of Echocardiography and the European Association of Cardiovascular Imaging. J Am Soc Echocardiogr 2015;28:1-39.e14. [Crossref] [PubMed]
12. Voigt JU, Mălăescu GG, Haugaa K, Badano L. How to do LA strain. Eur Heart J Cardiovasc Imaging 2020;21:715-7. [Crossref] [PubMed]
13. Emerson P, Deshmukh T, Stefani L, Mahendran S, Hogg M, Brown P, Panicker S, Altman M, Gottlieb D, Thomas L. Left atrial strain in cardiac surveillance of bone marrow transplant patients with prior anthracycline exposure. Int J Cardiol 2022;354:68-74. [Crossref] [PubMed]
14. Bak M, Park H, Lee SH, Lee N, Ahn MJ, Ahn JS, Jung HA, Park S, Cho J, Kim J, Park SJ, Chang SA, Lee SC, Park SW, Kim EK. The Risk and Reversibility of Osimertinib-Related Cardiotoxicity in a Real-World Population. J Thorac Oncol 2025;20:167-76. [Crossref] [PubMed]
15. Goyal A, Abbasi HQ, Yakkali S, Khan AM, Tariq MD, Sohail AH, Khan R. Left Atrial Strain as a Predictor of Early Anthracycline-Induced Chemotherapy-Related Cardiac Dysfunction: A Pilot Systematic Review and Meta-Analysis. J Clin Med 2024;13:3904. [Crossref] [PubMed]
16. 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]
17. Potter EL, Ramkumar S, Kawakami H, Yang H, Wright L, Negishi T, Marwick TH. Association of Asymptomatic Diastolic Dysfunction Assessed by Left Atrial Strain With Incident Heart Failure. JACC Cardiovasc Imaging 2020;13:2316-26. [Crossref] [PubMed]
18. Zuckerberg JC, Matsubara D, Kauffman HL, Chang JC, Calderon-Anyosa R, Patel C, Hogarty AN, Falkensammer CB, Mercer-Rosa LM, Quartermain MD, Wang Y, Banerjee A. Left atrial stiffness and strain are novel indices of left ventricular diastolic function in children: validation followed by application in multisystem inflammatory syndrome in children due to COVID-19. Eur Heart J Cardiovasc Imaging 2023;24:1241-51. [Crossref] [PubMed]
19. Gu J, Wang D, Jiang L, Huang Y, Ding L, Chen X, He Y, Zhou Z, Pu D. Assessment of Global Cardiac Function Using AutoSTRAIN Automatic Strain Quantitative Technology in Patients With Breast Cancer Undergoing Anthracycline-Based Chemotherapy. Ultrasound Med Biol 2023;49:368-74. [Crossref] [PubMed]
20. Vater LB, Lefebvre B, Turk A, Clasen SC. Fluoropyrimidine Cardiotoxicity: Incidence, Outcomes, and Safety of Rechallenge. Curr Oncol Rep 2022;24:943-50. [Crossref] [PubMed]
21. Lu Y, Pan W, Deng S, Dou Q, Wang X, An Q, Wang X, Ji H, Hei Y, Chen Y, Yang J, Zhang HM. Redefining the Incidence and Profile of Fluoropyrimidine-Associated Cardiotoxicity in Cancer Patients: A Systematic Review and Meta-Analysis. Pharmaceuticals (Basel) 2023;16:510. [Crossref] [PubMed]
22. Emerson P, Stefani L, Boyd A, Richards D, Hui R, Altman M, Thomas L. Alterations in Left Atrial Strain in Breast Cancer Patients Immediately Post Anthracycline Exposure. Heart Lung Circ 2024;33:684-92. [Crossref] [PubMed]
23. Yu C, Pathan F, Tan TC, Negishi K. The Utility of Advanced Cardiovascular Imaging in Cancer Patients-When, Why, How, and the Latest Developments. Front Cardiovasc Med 2021;8:728215. [Crossref] [PubMed]
24. Rusali CA, Lupu IC, Rusali LM, Cojocaru L. Left Atrial Strain-Current Review of Clinical Applications. Diagnostics (Basel) 2025;15:1347. [Crossref] [PubMed]
25. Hammond ST, Baumfalk DR, Parr SK, Butenas ALE, Scheuermann BC, Turpin VG, Behnke BJ, Hashmi MH, Ade CJ. Impaired microvascular reactivity in patients treated with 5-fluorouracil chemotherapy regimens: Potential role of endothelial dysfunction. Int J Cardiol Heart Vasc 2023;49:101300. [Crossref] [PubMed]
26. Fabin N, Bergami M, Cenko E, Bugiardini R, Manfrini O. The Role of Vasospasm and Microcirculatory Dysfunction in Fluoropyrimidine-Induced Ischemic Heart Disease. J Clin Med 2022;11:1244. [Crossref] [PubMed]
27. Lamberti M, Porto S, Marra M, Zappavigna S, Grimaldi A, Feola D, Pesce D, Naviglio S, Spina A, Sannolo N, Caraglia M. 5-Fluorouracil induces apoptosis in rat cardiocytes through intracellular oxidative stress. J Exp Clin Cancer Res 2012;31:60. [Crossref] [PubMed]
28. Di Lisi D, Moreo A, Casavecchia G, Cadeddu Dessalvi C, Bergamini C, Zito C, Madaudo C, Madonna R, Cameli M, Novo G. Atrial Strain Assessment for the Early Detection of Cancer Therapy-Related Cardiac Dysfunction in Breast Cancer Women (The STRANO STUDY: Atrial Strain in Cardio-Oncology). J Clin Med 2023;12:7127. [Crossref] [PubMed]
29. Inoue K, Machino-Ohtsuka T, Nakazawa Y, Iida N, Sasamura R, Bando H, Chiba S, Tasaka N, Ishizu T, Murakoshi N, Xu D, Sekine I, Tajiri K. Early Detection and Prediction of Anthracycline-Induced Cardiotoxicity - A Prospective Cohort Study. Circ J 2024;88:751-9. [Crossref] [PubMed]
30. Chen J, Cheng C, Fan L, Xu X, Chen J, Feng Y, Tang Y, Yang C. Assessment of left heart dysfunction to predict doxorubicin cardiotoxicity in children with lymphoma. Front Pediatr 2023;11:1163664. [Crossref] [PubMed]
31. Farsalinos KE, Daraban AM, Ünlü S, Thomas JD, Badano LP, Voigt JU. Head-to-Head Comparison of Global Longitudinal Strain Measurements among Nine Different Vendors: The EACVI/ASE Inter-Vendor Comparison Study. J Am Soc Echocardiogr 2015;28:1171-1181, e2.
32. Rösner A, Barbosa D, Aarsæther E, Kjønås D, Schirmer H, D'hooge J. The influence of frame rate on two-dimensional speckle-tracking strain measurements: a study on silico-simulated models and images recorded in patients. Eur Heart J Cardiovasc Imaging 2015;16:1137-47. [Crossref] [PubMed]
33. Sonaglioni A, Fagiani V, Nicolosi GL, Lombardo M. The influence of pectus excavatum on biventricular mechanics: a systematic review and meta-analysis. Minerva Cardiol Angiol 2024; Epub ahead of print. [Crossref]