Cardiovascular magnetic resonance-derived left atrioventricular coupling index for predicting adverse cardiac outcomes in patients with ischemic cardiomyopathy

Abstarct

Background

Risk stratification in ischemic cardiomyopathy (ICM) traditionally relies on left ventricular ejection fraction (LVEF), which may not fully demonstrate the complex interplay between atrium and ventricle. The left atrioventricular coupling index (LACI) has emerged as a promising integrative marker. This study evaluated the prognostic value of cardiovascular magnetic resonance (CMR)-derived LACI in patients with ICM.

Methods

This retrospective cohort study included 276 patients with ICM (mean age 65 ± 12 years and 80% male) who underwent CMR. LACI was calculated as LAEDV/LVEDV. The primary endpoint was a composite of cardiovascular mortality, non-fatal myocardial infarction, heart failure hospitalization (HFH), and major ventricular arrhythmic events. Secondary endpoints included individual components of the primary endpoint and all-cause mortality. Optimal LACI cut-off values were determined using receiver operating characteristic (ROC) analysis.

Results

During a median follow-up of 3.2 years, the primary endpoint occurred in 100 patients (36.2%). The optimal LACI cut-off for predicting the primary outcome was 0.35. Patients with LACI ≥ 0.35 had a significantly higher incidence of the primary outcome compared to those below the threshold (52.3% vs. 31.3%; p = 0.002). In multivariate Cox regression analysis, LACI ≥0.35 was an independent predictor of the primary outcome (aHR 1.83; 95%CI 1.06–3.16; p = 0.029). Among secondary endpoints, LACI ≥ 0.38 was strongly associated with HFH (aHR 3.33; 95%CI 1.58–6.72; p < 0.001), though it did not reach statistical significance for mortality.

Conclusions

CMR-derived LACI is an independent predictor of adverse cardiovascular outcomes, particularly HFH, in patients with ICM.

Graphical abstract

In 276 patients with ischemic cardiomyopathy who underwent cardiovascular magnetic resonance, the LACI predicted adverse outcomes over a median follow-up of 3.2 years. A LACI ≥ 0.35 was associated with a higher rate of the primary composite outcome than a LACI < 0.35 (52.3% vs. 31.3%; P = 0.002) and independently predicted the primary outcome (adjusted HR 1.83; 95% CI 1.06–3.16) and heart failure hospitalization (adjusted HR 3.33; 95% CI 1.58–6.72). A, area; aHR, adjusted hazard ratio; CI, confident interval; CV, cardiovascular; L, length; LA, left atrium; LACI, left atrioventricular coupling index; LAEDV, left atrial end-diastolic volume; LVEDV, left ventricular end-diastolic volume; MI, myocardial infarction; MRI, magnetic resonance imaging.

Highlights

• •This study is the first to establish the prognostic utility of the CMR-derived LACI specifically within an ischemic cardiomyopathy cohort.
• •A LACI threshold of ≥ 0.35 serves as a robust independent predictor of major adverse cardiovascular outcomes, even after adjusting for traditional markers like LVEF.
• •Elevated LACI displays a particularly strong association to adverse outcomes driven primarily by a 3.33-fold increased risk of heart failure hospitalization.

1. Introduction

Ischemic cardiomyopathy (ICM) remains a predominant cause of heart failure and cardiovascular mortality globally, currently affecting over 250 million individuals and being the leading cause of mortality worldwide [1,2]. Paradoxically, substantial advances in the management of acute myocardial infarction (MI) have contributed to this rising prevalence; enhanced post-infarction survival has increased the population of patients living with chronic left ventricular (LV) remodeling and dysfunction [3].

The prognosis for these patients remains unfavorable. Conventionally, risk stratification has relied on LV structural metrics, particularly the left ventricular ejection fraction (LVEF) [4,5]. However, these ventricular-centric measures often fail to fully capture the complex pathophysiology of ischemic heart failure, which involves a continuum of diastolic dysfunction, atrial remodeling, and altered hemodynamic coupling. Consequently, a significant proportion of patients remain at high risk for adverse events, including heart failure hospitalization (HFH), recurrent MI, and death, despite normal LV systolic function [[6], [7], [8], [9]]. Emerging evidence suggests that the dynamic interaction between the left atrium (LA) and the LV, termed left atrioventricular coupling index (LACI), is a critical determinant of global cardiac efficiency [[8], [9], [10], [11]]. Disruption of this coupling can compromise cardiac performance even when LV systolic function is preserved. LACI, defined as the ratio of left atrial end-diastolic volume (LAEDV) to left ventricular end-diastolic volume (LVEDV), has been proposed as an integrative physiological marker [[12], [13], [14]].

Previous investigations have highlighted the prognostic potential of LACI in various cohorts. In the Multi-Ethnic Study of Atherosclerosis (MESA), LACI was independently associated with incident atrial fibrillation (AF: hazard ratio [HR] 1.86), heart failure (HR 1.50), and cardiovascular mortality [8]. Similarly, Torben et al. reported that in patients with acute MI, an elevated LACI was a strong predictor of major adverse cardiovascular events (MACE: HR 3.10), notably stratifying risk even in those with severe LV dysfunction [10]. Furthermore, studies in high-risk populations, such as those with hypertrophic cardiomyopathy (HCM), heart failure, and chronic kidney disease, have confirmed LACI as a robust predictor of mortality and HF [11,[15], [16], [17]].

Despite this promising data, the prognostic utility of LACI specifically in patients with ICM remains undefined. This study aims to investigate the prognostic value of cardiovascular magnetic resonance imaging (CMR)-derived LACI for adverse cardiovascular outcomes, including death, HFH, and malignant ventricular arrhythmias, in patients with ICM.

2. Method

2.1. Study design and population

In this retrospective cohort study, patients aged 18 years or older who were diagnosed with ICM and underwent CMR at King Chulalongkorn Memorial Hospital between 2015 and 2024 were included. ICM was defined as the presence of LV systolic dysfunction accompanied by the presence of severe coronary artery disease including at least one of the following criteria: (i) history of coronary revascularization or acute MI, (ii) > 75% stenosis of the left main stem or left anterior descending artery, (iii) > 75% stenosis in two or more coronary vessels, or (iv) evidence of myocardial ischemia or infarction in the territory of left main stem, left anterior descending artery or at least two coronary territories on CMR [[18], [19], [20], [21]]. Exclusion criteria were patients with (i) acute MI within 3 months before the CMR scan, (ii) co-existing non-ischemic cardiomyopathy, (iii) prior heart transplantation, (iv) a follow-up duration of less than 1 year, (v) inadequate CMR quality for interpretation, and (vi) insufficient medical records.

Baseline characteristics, imaging data, and clinical outcomes were extracted from the hospital's electronic medical database. Mortality outcome was also retrieved from the population census. The first imaging was utilized for patients with multiple CMR studies. The study protocol was reviewed and approved by the Institutional Review Board of the Faculty of Medicine, Chulalongkorn University (IRB number 0117/68, Certificate of Approval number 0494/2025). Informed consent was waived due to the retrospective nature of the study, which utilized unidentified data. The study protocol conforms to the ethical guidelines of the 1975 Declaration of Helsinki as reflected in a priori approval by the institution's human research committee.

2.2. CMR acquisition and analysis

All CMR scans were performed by a 3-T scanner (Magnetom Skyra and Magnetom Vida, Siemens Healthineers, Erlangen, Germany) with an 18-channel phased-array cardiac receiver. The image acquisition included cardiac localizers; cine imaging with steady-state free precession sequence acquired in LV long-axis (2-, 3-, and 4-chamber) and continuous short-axis views (8-mm slices without inter-slice gaps), covering biventricular volume; first-pass perfusion; long inversion time (Ti) sequences; and late gadolinium enhancement (LGE) imaging, with optimal Ti received from Ti scout, which were acquired 5–10 min after the administration of 0.10-0.20 mmol/kg of gadobutrol (Gadovist, Bayer Healthcare, Leverkusen, Germany) or gadoterate meglumine (Dotarem; Guerbet, Villepinte, France) using a spoiled gradient echo with phase-sensitive reconstruction.

Post-processing analysis was conducted with Syngo.via (Siemens Healthineers, Erlangen, Germany). LV volumetric measurements, including LVEDV, LV end-systolic volume (LVESV), LV mass, and LVEF, were automatically quantified from short-axis cine stacks with manual adjustment. LA areas were manually measured at ventricular end-diastole and end-systole. LA volumes were calculated using the biplane area-length method: LA volume = [8 x (A1 x A2)]/3πL, where A1 and A2 are LA areas from the 2- and 4-chamber views, and L is the shortest distance from the mitral annulus plane to the LA roof. Pulmonary veins and the LA appendage were excluded from LA volumes. All volumes were indexed to body surface area. LA ejection fraction (LAEF) was calculated as [(LAESV-LAEDV)/LAESV] x 100.

The LACI was defined as the ratio of the LAEDV to LVEDV (LAEDV/LVEDV). Two independent physicians, blinded to clinical outcomes, evaluated LACI to assess interobserver variability. Intraobserver variability was assessed by repeating LACI measurements at an interval of 1 month.

Scar extent was quantified by software-assisted visual analysis, expressed as a percentage of the total scar volume over the entire LV myocardium volume. Details of scar quantification were specified elsewhere [22].

2.3. Clinical outcomes

The primary endpoint was a composite of cardiovascular mortality, non-fatal MI, HFH, and major ventricular arrhythmic events (including sudden cardiac death, sustained ventricular tachycardia, ventricular fibrillation, and appropriate implantable cardioverter-defibrillator therapy). Secondary endpoints included individual components of the primary endpoint and all-cause mortality.

2.4. Statistical analysis

A sample size of 214 patients was calculated to provide 80% power to detect a significant difference in MACE rates, assuming an event rate of 17.4% in the high-LACI group versus 5.1% in the low-LACI group [10]. Continuous variables were assessed for normality using the Shapiro–Wilk test, then presented as means ± standard deviations (SD) or medians (interquartile ranges [IQR]), as appropriate. Categorical variables are reported as frequencies and percentages. Baseline characteristics were compared between groups using the independent t-test or Mann–Whitney U test for continuous variables, and the Chi-square test for categorical variables. The optimal cut-off value of LACI in predicting adverse outcomes was determined using Youden's index from receiver operating characteristic (ROC) analysis, reporting with the area under the curve (AUC).

Time-to-event analyses were performed using Kaplan–Meier survival curves, with comparisons made using the log-rank test. Cox proportional hazards regression models were used to calculate HR and 95% confidence intervals (CIs) to evaluate the independent prognostic factors of clinical outcomes. LACI and variables with a p-value of less than 0.1 in univariate analysis were included in the multivariate model.

Intraclass correlation coefficients (ICC) were calculated to evaluate interobserver and intraobserver reliability. Statistical analyses were two-sided, and a p-value of less than 0.05 was considered statistically significant. SPSS Statistics software version 29.0.1 (IBM Corp., Armonk, NY, USA) was used for all statistical analyses.

3. Results

Between 2015 and 2024, 344 patients aged ≥ 18 years with ICM underwent CMR. After exclusion (acute MI within 3 months [n = 25], co-existing non-ischemic cardiomyopathy [n = 6], prior heart transplantation [n = 3], follow-up duration < 1 year [n = 17], inadequate CMR quality [n = 3], and insufficient medical records [n = 14]), 276 patients were included in the analysis [Supplementary Fig. 1]. The mean age was 65 ± 12 years and 80.0% were male. Patients who experienced the primary outcome had a significantly higher prevalence of diabetes mellitus (58.0% vs. 39.8%; p = 0.012) and a history of heart failure (67.0% vs. 40.9%; p < 0.001). They also exhibited more advanced heart failure symptoms, with 26.0% classified as New York Heart Association functional class III–IV compared with 9.1% in the no event group (p < 0.001). Renal function was significantly lower in the event group (median estimated glomerular filtration rate 60 vs. 73 mL/min/1.73 m²; p < 0.001). [Table 1].

Table 1.

Baseline characteristics.

	Total (N = 276)	No Event (N = 176)	Event (N = 100)	P-value
Age, years *	65 ± 12	65 ± 11	67 ± 12	0.269
Males	228 (80.0%)	144 (64.6%)	79 (35.4%)	0.568
BMI, kg/m2	23.1 (20.6, 25.5)	23.6 (21.1, 25.4)	22.5 (20.2, 25.5)	0.255
eGFR, mL/min/1.73 m2	70 (56, 81)	73 (61, 83)	60 (47, 76)	< 0.001
NYHA III & IV	42 (14.7%)	16 (9.1%)	26 (26.0%)	< 0.001
Co-morbidities
Diabetes mellitus	133 (46.7%)	70 (39.8%)	58 (58.0%)	0.012
Hypertension	221 (77.5%)	132 (75.0%)	84 (84.0%)	0.081
Dyslipidemia	214 (75.1%)	136 (77.3%)	72 (72.0%)	0.328
Cerebrovascular disease	62 (21.8%)	36 (20.5%)	25 (25.0%)	0.382
Myocardial infarction	172 (60.4%)	99 (56.3%)	68 (68.0%)	0.055
Heart Failure	145 (50.9%)	72 (40.9%)	67 (67.0%)	< 0.001
Atrial fibrillation	40 (14.0%)	24 (13.6%)	15 (15.0%)	0.755
Clinical presentation
Asymptomatic	11 (3.9%)	7 (4.0%)	4 (4.0%)	0.993
Dyspnea	109 (38.2%)	68 (38.6%)	37 (37.0%)	0.788
Angina	143 (50.2%)	88 (50.0%)	49 (49.0%)	0.873
Syncope	13 (4.6%)	10 (5.7%)	3 (3.0%)	0.312
Acute Heart Failure	71 (24.9%)	40 (22.7%)	28 (28.0%)	0.328
Stroke	20 (7.0%)	13 (7.4%)	7 (7.0%)	0.905
Arrhythmias	8 (2.8%)	4 (2.3%)	4 (4.0%)	0.411
Sudden cardiac arrest	8 (2.8%)	3 (1.7%)	5 (5.0%)	0.117
Cardiac MRI parameters
LVEF, %	26 [20, 34]	29 [23, 38]	22 [19, 29]	< 0.001
LVEDVI, ml/m2	137 [108, 172]	132 [105, 161]	152 [113, 179]	0.012
LVESVI, ml/m2	101 [72, 135]	92 [65, 126]	110 [83, 146]	< 0.001
LVSVI, ml/m2	36 [29, 43]	37 [31, 44]	34 [27, 42]	0.019
LVCI, L/min/m2	2.5 [2.0, 2.9]	2.5 [2.1, 2.9]	2.5 [1.9, 2.9]	0.333
LVMI, g/m2	79 [67, 97]	76 [64, 93]	83 [70, 102]	0.049
LV LGE extent, %	22 [13, 33]	23 [14, 32]	21 [11, 33]	0.346
LA diameter, mm *	38 ± 7	38 ± 7	39 ± 7	0.861
LAEDVI, ml/m2	35 [24, 52]	32 [22, 47]	41 [27, 57]	0.003
LAESVI, ml/m2	53 [41, 65]	52 [39, 62]	57 [43, 70]	0.031
LAEF, %	32 [19, 44]	37 [21, 47]	27 [14, 39]	< 0.001
LACI, %	24.4 [18.2, 34.3]	24.2 [18.5, 31.5]	25.0 [17.3, 39.3]	0.228
LACI ≥ 0.35	67 (23.5%)	31 (47.7%)	34 (52.3%)	0.002

* These variables were presented as mean ± standard deviation and analyzed with an independent t-test. Other continuous variables were presented as a median with an interquartile range and compared with the Mann-Whitney U test.

BMI, body mass index; eGFR, estimated glomerular filtration rate; LA, left atrium; LAESVI, left atrial end-diastolic volume index; LAEDVI, left atrial end-diastolic volume index; LAEF, left atrial ejection fraction; LACI, left atrioventricular coupling index; LGE, late gadolinium enhancement; LV, left ventricle; LVCI, left ventricle cardiac index; LVEDVI, left ventricular end-diastolic volume index; LVEF, left ventricular ejection fraction; LVESVI, left ventricular end-systolic volume index; LVMI, left ventricular mass index; LVSVI, left ventricle stroke volume index; NYHA, New York Heart Association functional class.

Patients who developed a primary endpoint exhibited more negative LV remodeling, characterized by significantly larger LV volumes (median LVEDV index: 152 vs. 132 ml/m², p = 0.012 and median LVESV index: 110 vs. 92 ml/m², p < 0.001) and lower LV systolic function (median LVEF: 22% vs. 29%, p < 0.001). The LA volumes were significantly higher in patients with the primary endpoint (median LAEDV index: 41 vs. 32 ml/m², p = 0.003 and median LAESV index: 57 vs. 52 ml/m², p = 0.031). Median LAEF was significantly lower in the event group (27% vs. 37%; p < 0.001) [Table 1]. Detailed coronary anatomy, mode of revascularization, and medications were shown in Supplementary Table 1.

Intra- and inter-observer variation of LACI were excellent with ICC 0.920 (0.824, 0.987; p < 0.001) and 0.886 (0.708, 0.956; p < 0.001).

3.1. Optimal LACI cut-off values

The identified cut-off for the primary outcome was 0.35, yielding a sensitivity and specificity of 34.0% and 82.4%, respectively. LA diameter showed the highest predictive accuracy (AUC 0.564), followed by LACI (AUC 0.544) [Fig. 1]. Optimal LACI cut-off values for all-cause death, cardiovascular death, and HFH were 0.35 (AUC 0.492), 0.35 (AUC 0.468), and 0.38 (AUC 0.642), respectively [Supplementary Figs. 2–4]. Diagnostic performances of these specific cut-off values were shown in the Supplementary Table 2. Optimal cut-off values for non-fatal MI and ventricular arrhythmia were not identified due to low outcome incidence.

Fig. 1.

Receiver operating characteristic curve of LACI and other parameters predicting the primary outcome AUC, area under the curve; LA, left atrium; LACI, left atrioventricular coupling index; LAEF, left atrial ejection fraction; LGE, late gadolinium enhancement; LVEF, left ventricular ejection fraction.

3.2. Adverse outcomes and survival analysis

During a median follow-up period of 3.2 years, the primary composite endpoint occurred in 100 patients (36.2%). When stratified by the LACI cut-off, patients with a LACI ≥ 0.35 had a significantly higher incidence of the primary outcome compared to those below the cut-off (52.3% vs. 31.3%; p = 0.002). (Table 2).

Table 2.

Incidence of primary and secondary outcomes.

	Total (N = 276)	LACI < cut-offs*	LACI ≥ cut-offs*	p-value
Primary outcome	100 (36.2%)	66/211 (31.3%)	34/65 (52.3%)	0.002
Secondary outcomes
All-cause death	31 (11.2%)	29/213 (13.6%)	10/63 (15.9%)	0.184
Cardiovascular death	25 (9.1%)	17/212 (8.0%)	8/64 (12.5%)	0.274
Heart failure hospitalization	53 (19.2%)	30/227 (13.2%)	23/49 (46.9%)	< 0.001
Non-fatal myocardial infarction	13 (4.7%)	-	-	-
Ventricular Arrhythmias	24 (8.7%)	-	-	-

* LACI cut-offs were 0.35 for the primary outcome, 0.35 for all-cause death, 0.35 for cardiovascular death, and 0.38 for heart failure hospitalization.

LACI, left atrioventricular coupling index.

In the multivariate Cox regression model adjusted for potential confounders, LACI ≥ 0.35 remained an independent predictor of the primary outcome (adjusted HR [aHR] 1.83; 95% CI 1.06–3.16; p = 0.029) [Fig. 2A]. Other independent predictors included hypertension (aHR 1.76; 95% CI 1.02–3.05; p = 0.042), history of heart failure (aHR 1.65; 95% CI 1.04–2.62; p = 0.032), eGFR (aHR 0.99; 95% CI 0.98–0.99; p = 0.003), and LVEF (aHR 0.94; 95% CI 0.92-0.97; p < 0.001). (Table 3).

Fig. 2.

Kaplan-Meier curves demonstrate cumulative hazard for the primary outcome (A), all-cause death (B), cardiovascular death (C), and heart failure hospitalization (D) stratified by LACI cut-off. aHR, adjusted hazard ratio; LACI, left atrioventricular coupling index.

Table 3.

Cox proportional hazards regression analysis of the primary outcome.

	Univariate analysis HR (95% CI)	P-value	Multivariate analysis aHR (95% CI)	P-value
Age, years	1.02 (1.00-1.03)	0.084	1.01 (1.00-1.03)	0.145
Males	0.86 (0.53-1.39)	0.527
BMI, kg/m2	0.98 (0.93-1.04)	0.574
eGFR	0.98 (0.97-0.99)	< 0.001	0.99 (0.98-0.99)	0.003
NYHA class III-IV	2.52 (1.61-3.95)	< 0.001	1.43 (0.88-2.33)	0.149
Co-morbidities
Diabetes	1.74 (1.17-2.59)	0.006	1.48 (0.98-2.24)	0.063
Hypertension	1.63 (0.95-2.78)	0.074	1.76 (1.02-3.05)	0.042
Dyslipidemia	0.81 (0.52-1.25)	0.335
Cerebrovascular disease	1.12 (0.71-1.76)	0.634
Myocardial infarction	1.53 (1.00-2.33)	0.048	1.36 (0.87-2.11)	0.174
Heart Failure	2.54 (1.67-3.85)	< 0.001	1.65 (1.04-2.62)	0.032
Atrial fibrillation	1.02 (0.59-1.76)	0.955
Cardiac MRI Parameter
LVEF	0.95 (0.93-0.97)	< 0.001	0.94 (0.92-0.97)	< 0.001
LVMI	1.01 (1.00-1.02)	0.106
LGE extent	1.00 (0.97-1.01)	0.511
LAEF	0.98 (0.97-0.99)	< 0.001	1.01 (0.99-1.03)	0.281
LACI ≥ 0.35	1.90 (1.25-2.87)	0.002	1.83 (1.06-3.16)	0.029

BMI, body mass index; eGFR, estimated glomerular filtration rate; LAEF, left atrial ejection fraction; LACI, left atrioventricular coupling index; LGE, late gadolinium enhancement; LVEF, left ventricular ejection fraction; LVMI, left ventricular mass index; NYHA, New York Heart Association functional class.

All-cause mortality, cardiovascular mortality, and HFH occurred in 11.2%, 9.1%, and 19.2% of patients, respectively. LACI beyond the cut-off value was an independent predictor of HFH (aHR 3.33; 95% CI 1.58–6.72; p < 0.001) but was not for all-cause mortality (aHR 1.38; 95% CI 0.47–4.10; p = 0.558) and cardiovascular mortality (aHR 1.96; 95% CI 0.60–6.46; p = 0.269). (Fig. 2B-D and Supplementary Tables 3–5).

4. Discussion

This study represents the first investigation into the prognostic value of the CMR-derived LACI specifically within a cohort of patients with ICM. Our findings demonstrate that an elevated LACI (≥0.35) is significantly associated with an increased risk of a composite primary outcome comprising cardiovascular death, HFH, non-fatal MI, and ventricular arrhythmias. Notably, after adjusting for potential confounding factors and established imaging parameters such as LVEF, LACI remained an independent predictor of adverse events, driven primarily by its robust association with HFH.

4.1. The pathophysiological significance of LACI in ICM

The transition from simple coronary artery disease to ICM is characterized by chronic LV remodeling, frequently involving both systolic and diastolic impairment [3]. While LVEF has long been the gold standard for risk stratification, it is a relatively late marker of disease and focuses solely on the ventricle [4,5]. Our results support the concept of the left atrioventricular unit as an integrated functional system [9,11]. LACI reflects the disproportionate enlargement of LA relative to LV. In ICM, LA undergoes remodeling in response to a chronic elevation in LV filling pressures, even before overt systolic failure occurs [6]. This atrioventricular mismatch represents a critical point in the continuum of diastolic dysfunction; as the LA fails to compensate for rising LV end-diastolic pressures, the resulting pulmonary venous congestion directly precipitates heart failure symptoms. Consequently, by using a ratio (LAEDV/LVEDV), LACI serves as a sensitive barometer for hemodynamic congestion and atrial-ventricular mismatch better than either volume alone, explaining its particularly robust association with HFH in our cohort [9,12,13].

4.2. Comparison with the previous studies

Our findings align with and extend existing data from other cardiac populations. In the MESA, LACI was found to be a superior predictor of heart failure compared to LVEF in an asymptomatic, multi-ethnic population [9]. Similarly, our identified optimal cut-off of 0.35 is remarkably consistent with the results of Lange et al., who reported that a LACI ≥ 34.7% was a strong predictor of MACE following acute MI [10]. This threshold is further supported by studies in other cardiomyopathies, such as HCM, in which LACI ≥ 36% significantly predicted new-onset AF and stroke in patients with HCM [23]. However, unlike the MESA or post-acute MI cohorts, our study focuses on patients with established ICM. In this higher-risk group, we observed that while LACI was highly predictive of the composite outcome and HFH, it did not reach statistical significance for mortality when analyzed as an individual endpoint. This discrepancy may be due to the relatively small number of death events in our study or the fact that in end-stage ICM, mortality is often driven by multiple extracardiac factors, such as the renal function observed in our baseline characteristics, which may dilute the specific prognostic signal of atrial-ventricular coupling [4].

4.3. LACI vs. other risk markers

In our ROC analysis, linear measures like LA diameter demonstrated comparable, albeit modest, individual predictive accuracy (AUC 0.564 vs. 0.544 for LACI). While these AUC values are relatively low, suggesting limited performance as a standalone diagnostic tool in this high-risk population, LACI's primary strength lies in its performance within multivariate Cox proportional hazard models. Crucially, an elevated LACI ≥ 0.35 remained a robust independent predictor of the primary outcome (aHR 1.83; 95% CI 1.06–3.16; p = 0.029), maintaining its significance even after adjusting for potent traditional risk factors. This suggests that while LACI may not serve as a single predictor for diagnosis, it provides incremental prognostic value by indicating a unique dimension of atrioventricular mismatch that existing metrics do not provide. Furthermore, with excellent interobserver reliability, LACI remains a reproducible, high-yield metric that can be readily integrated into standard CMR reporting.

It is noteworthy that LGE extent was not a significant predictor of the primary outcome in this study, whereas LACI maintained significance. While LGE is a well-established marker of the anatomic substrate in ICM [24], our results suggest that in patients with chronic disease, the hemodynamic consequence of the total disease burden, expressed as atrioventricular mismatch, may be a more dynamic predictor of acute decompensation than the static scar burden alone. Furthermore, the location and transmurality of scars, which were not assessed in this study, are critical determinants of prognosis [25]. For instance, infarcts involving the mitral apparatus can lead to functional mitral regurgitation and disproportionate LA remodeling [26,27]. LACI likely reflects the cumulative functional impact of both the infarcted and remote (non-infarcted) myocardium. By integrating atrial remodeling and ventricular volumes into a single ratio, LACI provides a holistic assessment of global cardiac efficiency and filling pressures that purely anatomic tissue characterization may not fully represent.

4.4. Clinical implications

Identifying high-risk patients with ICM via LACI provides a basis for more tailored clinical management. In this cohort, a LACI ≥ 0.38 was associated with a 3.33-fold increased risk of HFH. Consequently, patients exceeding this threshold may require more aggressive titration of guideline-directed medical therapy or earlier evaluation for advanced heart failure interventions. Given the strength of the association with HFH, LACI may serve as a marker to identify individuals necessitating intensive diuretic optimization and more frequent outpatient monitoring to prevent hospital readmission.

4.5. Limitations

This study has several limitations. First, its retrospective, single-center design may restrict the generalizability of the results to other populations. Second, although LACI was an independent predictor in multivariate models, its AUC in ROC analysis was low. This indicates that LACI serves as an incremental prognostic marker rather than a standalone diagnostic tool, and its clinical utility lies in refining risk when used alongside established metrics. Third, the median follow-up of 3.2 years may be insufficient to fully represent long-term mortality trends in a chronic ICM cohort. Fourth, despite adjusting for multiple confounders, we could not account for all variables, such as socioeconomic status, specific medication dosages, or the exact timing of revascularization. Furthermore, since this cohort duration spanned over the COVID-19 pandemic, the SARS-CoV-2 infection in specific patients could affect the outcomes [28,29]. Fifth, prognostic biomarkers, e.g., serum natriuretic peptide and cardiac marker levels, were not included in the analysis. These biomarkers can further stratify the risk of cardiovascular events in addition to clinical and imaging parameters. Finally, the low incidence of ventricular arrhythmias and non-fatal myocardial infarction limited the ability to determine event-specific LACI cut-offs.

5. Conclusion

In patients with ICM, the CMR-derived LACI is an independent predictor of adverse cardiovascular outcomes, particularly HFH. LACI provides incremental prognostic value beyond traditional markers like LVEF, offering an integrated assessment of atrioventricular remodeling and hemodynamic congestion. While the standalone diagnostic accuracy of LACI is limited by a low AUC, its strong independent association with clinical events suggests utility in refining risk stratification. Further prospective research is required to determine whether incorporating LACI into clinical management can improve long-term outcomes in this population.

CRediT authorship contribution statement

Sedthasith Treewatchareekorn: Writing – original draft, Visualization, Methodology, Formal analysis, Data curation, Conceptualization. Sittinop Titichoatrattana: Validation, Formal analysis, Data curation. Paisit Kosum: Writing – review & editing, Methodology, Data curation. Yossatorn Buranasakulsatid: Investigation, Data curation. Chayachon Okonogi: Investigation, Data curation. Monravee Tumkosit: Supervision, Resources, Investigation. Yongkasem Vorasettakarnkij: Supervision, Resources, Investigation. Pairoj Chattranukulchai: Supervision, Resources, Investigation. Nonthikorn Theerasuwipakorn: Writing – review & editing, Writing – original draft, Visualization, Validation, Supervision, Resources, Methodology, Investigation, Formal analysis, Data curation, Conceptualization.

Availability of data and materials

The data of this study are available from the corresponding author upon reasonable request.

Disclosure

None.

Ethics declarations

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 (as revised in 2013). The study was approved by the institutional ethics committee of the Faculty of Medicine, Chulalongkorn University (IRB No. 0117/68; COA No. 0494/2025) and individual consent for this retrospective analysis was waived.

Sources of funding

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

Appendix A. Supplementary data

The following are the Supplementary data to this article:

figs1.

figs2.

figs3.

figs4.