Systematic Review and Meta-Analysis of Cardiac MRI T1 and ECV Measurements in Pre-Heart Failure Populations

Abstract

Background/Objectives: Heart failure (HF) often develops from a prolonged asymptomatic phase where early detection could prevent progression. Pre-heart failure (pre-HF) populations—those with risk factors (Stage A) or subclinical myocardial changes (Stage B)—are critical for intervention. Cardiac magnetic resonance (CMR) with T1 and extracellular volume (ECV) mapping offers a non-invasive approach to detect early myocardial changes in these groups. This systematic review evaluates the role of T1 and ECV mapping in pre-HF populations, focusing on their diagnostic and prognostic utility. Methods: A systematic search of PubMed, EMBASE, and Cochrane was conducted up to April 2025, identifying 17 studies that met inclusion criteria. Data was extracted directly into Excel, and methodological quality was assessed using the Newcastle–Ottawa Scale (NOS) for cohort and cross-sectional studies and AMSTAR-2 for systematic reviews and meta-analyses. A meta-analysis was performed using Review Manager (RevMan) to compare T1 and ECV values between pre-HF and control groups. Results: Studies consistently reported elevated T1 (989.6–1415.41 milliseconds) and ECV (25.7–42.81%) in pre-HF groups compared to controls (T1: 967–1310.63 ms, ECV: 23.5–29.9%). Meta-analysis showed a significant increase in T1 (MD: 27.62 ms, 95% CI: 8.04–47.19, p < 0.006) and ECV (MD: 2.97%, 95% CI: 1.88–4.06, p < 0.00001) in pre-HF groups. RQS scores ranged from 17.2% to 77.8% (mean: 37.9%), and NOS scores ranged from 5 to 8 (mean: 6.2), reflecting variability in study quality. The AMSTAR-2 rating for the systematic review was moderate. Conclusions: T1 and ECV mapping enhance CMR-based detection of early myocardial changes in pre-HF, offering a promising non-invasive approach to predict HF risk. However, variability in study quality, small sample sizes, and methodological inconsistencies limit generalisability. Future research should focus on standardised protocols, prospective designs, and multi-center studies to integrate these techniques into clinical practice, potentially guiding preventive therapies such as SGLT2is and tafamidis.

1. Introduction

Heart failure (HF) remains a global health challenge, affecting over 64 million individuals worldwide and imposing a substantial burden on healthcare systems [1]. The preclinical phase, termed pre-heart failure (pre-HF), offers a critical window for early intervention to prevent progression. Pre-HF includes individuals with risk factors like type 2 diabetes mellitus (T2DM), hypertension, genetic predispositions (Stage A), or those with subclinical myocardial changes (Stage B), as defined by the American College of Cardiology/American Heart Association guidelines [2]. These conditions drive myocardial remodeling, such as fibrosis, before symptoms emerge, making non-invasive detection vital [3].

Cardiac magnetic resonance (CMR) with T1 and extracellular volume (ECV) mapping is a powerful tool for detecting early myocardial changes. Native T1 mapping measures tissue relaxation time, reflecting fibrosis, edema, or protein deposition, while ECV quantifies extracellular matrix expansion, a proxy for diffuse fibrosis [4,5]. Unlike echocardiography, which often misses subtle changes, or late gadolinium enhancement (LGE), which detects focal scars, T1 and ECV capture diffuse processes, ideal for pre-HF evaluation [6,7]. Normal ranges (e.g., native T1 ~950–1000 ms, ECV ~25–26% at 1.5 T) provide benchmarks for pathological deviations [8].

This systematic review synthesizes evidence on T1 and ECV mapping in pre-HF populations, including T2DM, hypertension, and genetic cohorts like hypertrophic cardiomyopathy (HCM). We aim to evaluate their diagnostic and prognostic utility as biomarkers, addressing variability in study protocols and populations. By focusing on asymptomatic individuals, we seek to clarify how these techniques can guide early interventions, such as sodium-glucose cotransporter-2 inhibitors or angiotensin receptor–neprilysin inhibitors, to improve outcomes [9,10].

2. Materials and Methods

2.1. Study Design and Reporting Guidelines

This study is a systematic review of cohort, cross-sectional, and meta-analytic studies, following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) reporting guidelines [11].

2.2. Search Strategy

In April 2025, we conducted searches across PubMed (MEDLINE), EMBASE, and Cochrane. The search strategy included terms such as “T1 mapping,” “extracellular volume,” “cardiac magnetic resonance,” “pre-heart failure,” “type 2 diabetes,” “hypertension,” “hypertrophic cardiomyopathy,” and “ischemic risk,” with filters for English language and publication dates from January 2010 to April 2025. The final search date was 1 April 2025 (A detailed search strategy is provided in Supplementary Materials).

2.3. Inclusion/Exclusion Criteria

Inclusion criteria include the following:

• Studies investigating T1 and/or ECV mapping on CMR in pre-HF populations (asymptomatic individuals with risk factors or subclinical changes, without overt HF).
• Patients aged 18 years and above.
• Prospective, retrospective, cohort, cross-sectional studies, systematic reviews, or meta-analyses.

Exclusion criteria include the following:

• Studies solely on established HF (e.g., HFrEF, HFpEF) without a pre-HF subgroup.
• Studies on chronic kidney disease (CKD) populations (e.g., uremic cardiomyopathy).
• Non-human studies.
• Case series/reports, consensus statements, or conference abstracts.
• Studies not reporting T1/ECV data or not using CMR.

2.4. Study Selection, Data Extraction, and Quality Assessment

A database was compiled using EndNote X9™ version X9.3.3. Two researchers independently reviewed search results. Duplicates were removed, and titles were screened for relevance. Abstracts of potentially relevant studies were assessed for eligibility based on the inclusion/exclusion criteria. Full texts of eligible abstracts were further evaluated. Discrepancies were resolved through discussion with a third reviewer.

Data were extracted directly into Excel by two reviewers independently, collecting information on study details, design, population, CMR parameters, T1/ECV values, and outcomes. Methodological quality was assessed using the Newcastle–Ottawa Scale (NOS) [12] for cohort and cross-sectional studies, with AMSTAR-2 [13] for systematic reviews and meta-analyses.

2.5. Systematic Review Registration

This review was registered on PROSPERO in April 2025 (ID: CRD420251030475).

2.6. Statistical Analysis

A meta-analysis was conducted on studies reporting T1 and ECV values for pre-HF and control groups. The mean difference (MD) was calculated to compare T1 and ECV between groups, using a random-effects model due to expected heterogeneity, which accounted for potential variations in study populations and methodologies. Heterogeneity was assessed using Cochran’s Q test and the I² statistic to quantify the degree of inconsistency across studies. The analysis was performed using Review Manager (RevMan) 5.4, and forest plots were generated to visualize the results, providing a clear graphical representation of the effect sizes and their confidence intervals. Sensitivity analyses were also conducted to evaluate the robustness of the findings by systematically excluding studies with high risk of bias or outlier results, ensuring the stability and reliability of the pooled estimates. Additionally, subgroup analyses were explored to investigate potential sources of heterogeneity, such as differences in CMR field strength (e.g., 1.5 T vs. 3 T), population characteristics (e.g., T2DM vs. hypertension), and study design (e.g., prospective vs. retrospective), further refining the interpretation of the meta-analysis outcomes. These comprehensive approaches ensured a thorough evaluation of the data, maximizing the validity of the conclusions drawn from the meta-analysis.

3. Results

3.1. Search Results

The literature search resulted in 356 studies from databases and 2 additional studies identified through citation searching, totaling 358 records. After removing 48 duplicates, 308 records remained for screening. Initial screening led to the exclusion of 183 records based on title and abstract irrelevance. Of the 125 reports sought for retrieval, all were retrieved, and none were excluded due to unavailability. Full-text review resulted in the exclusion of 110 studies: 55 did not focus on pre-HF populations, 39 involved CKD populations, and 16 did not report T1 or ECV measurements. This left 15 studies, totaling 17 with the addition of 2 found via citation searching (see Figure 1).

3.2. Methodological Characteristics and Quality of Studies

Of the 17 studies, 16 were cross-sectional or cohort studies (11 prospective, 5 retrospective), and 1 was a systematic review/meta-analysis. Sample sizes ranged from 65 to 1176 patients, with a total of 3353 participants across all studies. In total, 2 studies reported only ECV values [14,15], 3 reported only T1 values [16,17,18], with 12 reporting both T1 and ECV values.

Table 1. Methodological characteristics and quality assessment of included studies.

Study	Study Design	Population	Sample Size	CMR Protocol	Field Strength	Sequence	Contrast Agent
Wong et al. (2014) [14]	Prospective cohort	T2DM, controls	1176	ECV only	1.5 T	MOLLI	Gadoteridol
Snel et al. (2022) [15]	Prospective cross-sectional	Overweight, HTN, controls	126	T1, ECV	3 T	MOLLI	Gadoteric acid
Fontana et al. (2014) [16]	Cross-sectional	ATTR amyloidosis, controls	270	T1 only	1.5 T	ShMOLLI	Gadoterate meglumine
Mohamed et al. (2024) [17]	Prospective observational	HTN (ethnic groups), controls	110	T1 only	1.5 T	MOLLI	Unspecified gadolinium-based contrast
Hwang et al. (2025) [18]	Retrospective observational	AL amyloidosis, controls	300	T1 only	3 T	MOLLI	Gadobutrol
Cerne et al. (2023) [19]	Prospective observational	PH (PrePH, IpcPH), controls	73	T1, ECV, LGE	1.5 T	MOLLI	Gadobutrol
Alabed et al. (2021) [20]	Systematic review/meta-analysis	PAH, controls	606	T1, ECV	1.5, 3 T	MOLLI, ShMOLLI	Varying gadolinium-based contrasts
Gao et al. (2019) [21]	Prospective observational	T2DM, controls	100	T1, ECV	3 T	MOLLI	Gadobenate dimeglumine
Kuruvilla et al. (2015) [22]	Cross-sectional observational	HTN (LVH, non-LVH), controls	65	T1, ECV	1.5 T	MOLLI	Gadopentetate dimeglumine
Laohabut et al. (2021) [23]	Retrospective cohort	T2DM, controls (CAD suspected)	739	T1, ECV	3 T	MOLLI	Unspecified gadolinium-based contrast
Shu et al. (2024) [24]	Prospective observational	T2DM, controls	65	T1, ECV	1.5 T	MOLLI	Gadolinium-based contrast
Liu et al. (2022) [25]	Prospective cross-sectional	T2DM, controls	122	T1, ECV	3 T	MOLLI	Gadobenate dimeglumine
Li et al. (2023) [26]	Cross-sectional	T2DM with preserved EF, controls	114	T1, ECV	1.5 T	MOLLI	Gadobutrol
Cao et al. (2018) [27]	Prospective observational	T2DM, controls	82	T1, ECV	1.5 T	MOLLI	Gadolinium-diethylenetriamine pentaacetic acid
GA Boros et al. (2024) [28]	Prospective observational	T2DM, controls (CAD)	155	T1, ECV	1.5 T	ShMOLLI	Gadoterate meglumine
AS Bojer et al. (2022) [29]	Cross-sectional observational	T2DM, controls	264	ECV only	3 T	MOLLI	Gadobutrol
Shi et al. (2021) [30]	Retrospective	HCM, HHD, controls	146	T1, ECV	3 T	MOLLI	Gadopentetate dimeglumine

Among the 17 included studies, 9 utilized 1.5 T field strength, 7 used 3 T, and 1 incorporated both 1.5 T and 3 T due to its meta-analytic nature [20]. For sequences, 14 studies employed MOLLI, while 3 used ShMOLLI. Contrast agents varied widely, with gadobutrol used in four studies, gadobenate dimeglumine in two, gadoterate meglumine in two, and others including gadoteridol, gadolinium-diethylenetriamine pentaacetic acid, gadoteric acid, gadopentetate dimeglumine, unspecified gadolinium-based contrasts, or varying agents in the meta-analysis. These differences in field strength, sequences, and contrast agents likely contribute to measurement variability, as 3 T scanners generally yield higher T1 values than 1.5 T, and sequence variations (e.g., MOLLI vs. ShMOLLI) can affect precision and bias in T1/ECV quantification.

summarizes the methodological characteristics and quality scores of the included studies. RQS scores ranged from 17.2% to 77.8% (mean: 37.9%), reflecting variability in methodological rigor, with higher scores for prospective studies with large sample sizes and clear validation [14,19]. NOS scores for the 16 cohort/cross-sectional studies ranged from 5 to 8 (mean: 6.2), indicating moderate to high quality on average, with lower scores for studies with limited follow-up or controls [14]. The AMSTAR-2 rating for the systematic review [20] was moderate, reflecting a comprehensive search and risk of bias assessment.

3.3. T1 and ECV Findings and Meta-Analysis Results

T1 and ECV values in pre-HF populations varied across studies, reflecting differences in risk factors and populations. Values ranged widely across the 17 studies, with T1 ranging from 989.6 to 1415.41 ms in pre-HF groups and 967 to 1310.63 ms in controls, while ECV ranged from 25.7% to 42.81% in pre-HF groups and 23.5% to 29.9% in controls.

Table 2. T1 and ECV results and key findings of included studies.

Study	Population	Sample Size	T1 (ms ± SD) Pre-HF	T1 (ms ± SD) Controls	T1 p-Value	ECV (% ± SD) Pre-HF	ECV (% ± SD) Controls	ECV p-Value	Key Findings/Outcomes
Wong et al. (2014) [14]	T2DM, controls	1176	-	-	-	30.2 ± 4.3	28.1 ± 3.8	p < 0.001	ECV > 30% predicted HF admission (HR: 1.52, p < 0.01)
Snel et al. (2022) [15]	Overweight, HTN, controls	126	1152.6 ± 35.23	1147 ± 30	p > 0.05	23.45 ± 2.21	24.7 ± 2.1	p < 0.01	ECV lower in overweight/HTN
Fontana et al. (2014) [16]	ATTR amyloidosis, controls	270	1088.7 ± 68.9	967 ± 34	p < 0.001	-	-	-	T1 elevation tracks amyloid burden
Mohamed et al. (2024) [17]	HTN (ethnic groups), controls	110	1003.65 ± 57.65	-	-	-	-	-	No ethnic differences in ECV
Hwang et al. (2025) [18]	AL amyloidosis, controls	300	1328.1 ± 64.4	-	-	-	-	-	T1 and ECV diagnostic for amyloidosis
Cerne et al. (2023) [19]	PH (PrePH, IpcPH), controls	73	1050.9 ± 33.8	1012.9 ± 29.4	p < 0.05	31.0 ± 4.1	28.2 ± 3.7	p < 0.05	PrePH had higher septal T1
Alabed et al. (2021) [20]	PAH, controls	606	1038.94 ± 72.54	987.36 ± 28.46	p < 0.05	32.23 ± 3.53	26.6 ± 3.39	p < 0.003	T1 and ECV elevated in PAH
Gao et al. (2019) [21]	T2DM, controls	100	1285.46 ± 68.82	1279.83 ± 121.85	p > 0.017	36.23 ± 4.62	29.73 ± 2.28	p < 0.001	ECV increased with HbA1c levels
Kuruvilla et al. (2015) [22]	HTN (LVH, non-LVH), controls	65	989.6 ± 33.3	967.4 ± 35.0	p < 0.05	28.0 ± 3.0	26.0 ± 2.0	p < 0.05	ECV linked to strain impairment
Laohabut et al. (2021) [23]	T2DM, controls (CAD suspected)	739	1335 ± 75	1311 ± 58	p = 0.516	30.0 ± 5.9	28.8 ± 4.7	p = 0.004	ECV predicted CV outcomes (p = 0.004)
Shu et al. (2024) [24]	T2DM, controls	65	1044.8 ± 55.9	1053.0 ± 23.4	p = 0.264	32.1 ± 3.2	26.2 ± 1.6	p < 0.001	Non-contrast T1ρ mapping feasible
Liu et al. (2022) [25]	T2DM, controls	122	1290.41 ± 39.29	1293.65 ± 59.70	p < 0.05	33.27 ± 2.68	29.90 ± 2.35	p < 0.05	ECV correlated with diastolic dysfunction
Li et al. (2023) [26]	T2DM with preserved EF, controls	114	1057.49 ± 41.24	1035.02 ± 26.65	p < 0.05	30.37 ± 4.295	26.33 ± 2.81	p < 0.05	ECV linked to HbA1c levels
Cao et al. (2018) [27]	T2DM, controls	82	1026.9 ± 30.0	1011.8 ± 26.0	p = 0.022	27.4 ± 2.5	24.6 ± 2.2	p < 0.001	ECV linked to systolic strain impairment
GA Boros et al. (2024) [28]	T2DM, controls (CAD)	155	1015.5 ± 46.0	1003.8 ± 42.8	p = 0.10	25.7 ± 2.6	23.5 ± 2.3	p < 0.01	ECV increased in T2DM with CAD
AS Bojer et al. (2022) [29]	T2DM, controls	264	-	-	-	28.8 ± 3.2	27.4 ± 2.1	p < 0.004	ECV associated with ischemic heart disease
Shi et al. (2021) [30]	HCM, HHD, controls	146	1295.78 ± 80.10	1233.45 ± 35.58	p < 0.001	29.28 ± 5.25	25.96 ± 2.96	p < 0.0001	T1 and ECV diagnostic for HCM/HHD

details the results, including associations with outcomes. HF events were reported as outcomes in only three studies. Wong et al. [14] observed 38 events (21 HF hospitalizations, 24 deaths) over a median of 1.3 years in diabetic individuals, with ECV associated with these events (HR: 1.52, 95% CI: 1.21–1.89 per 3% ECV increase). Hwang et al. [18] reported an HF hospitalization HR of 4.247 (95% CI: 1.215–14.851, p = 0.024) for automated ECV ≥ 40% in AL-CA patients (revised Mayo stage III or IV), with 20 out of 39 patients (51.3%) experiencing HF hospitalization during a mean 22-month follow-up. Laohabut et al. [23] included HF hospitalization in a composite cardiovascular outcome (HR: 2.41, 95% CI: 1.17–4.98 associated with T2DM), with 33 out of 739 patients (4.5%) experiencing HF hospitalization.

Meta-Analysis of T1 Values (15 Studies):

The forest plot for T1 values (Figure 2) included 15 studies that reported T1 data, 13 of which included control groups. The meta-analysis showed a significant increase in T1 in pre-HF groups compared to controls, with a mean difference (MD) of 27.62 ms (95% CI: 8.04–47.19, p < 0.006). Heterogeneity was high (I² = 96%, p < 0.00001). The largest effect was observed in Fontana et al. [16], with an MD of 121.70 ms (95% CI: 107.86–135.54) in ATTR amyloidosis patients, reflecting significant amyloid deposition. In contrast, Snel et al. [15] showed a smaller, non-significant difference (MD: 5.60 ms, 95% CI: −6.31 to 17.51) in overweight and hypertensive young adults, suggesting potentially less pronounced myocardial changes in this population.

Meta-Analysis of ECV Values (14 Studies):

The forest plot for ECV values (Figure 3) included 12 studies that reported ECV data for both pre-HF and control groups, with 2 studies lacking a control group. The meta-analysis showed a significant increase in ECV in pre-HF groups compared to controls, with an MD of 2.97% (95% CI: 1.88–4.06, p < 0.00001). Heterogeneity was high (I² = 94%, p < 0.00001), reflecting variability in study populations and methods. The largest effect was in Gao et al. [21], with an MD of 6.50% (95% CI: 5.08–7.92) in T2DM patients with high HbA1c, indicating substantial interstitial expansion.

4. Discussion

This systematic review synthesizes evidence from 17 studies to evaluate the role of T1 and ECV mapping in CMR for detecting early myocardial changes in pre-HF populations. Pre-HF, encompassing Stage A (risk factors like T2DM, hypertension, or genetic predispositions) and Stage B (subclinical changes), lacks a universal definition but is characterized by conditions driving myocardial remodeling [2]. Studies have defined pre-HF variably, often including Stage B patients with detectable subclinical alterations, as Stage A patients may lack CMR-detectable changes. The studies consistently reported elevated T1 and ECV in pre-HF groups compared to controls, reflecting interstitial expansion and fibrosis. The meta-analysis confirmed significant increases in both T1 (MD: 27.62 ms, 95% CI: 8.04–47.19, p < 0.006) and ECV (MD: 2.97%, 95% CI: 1.88–4.06, p < 0.00001) in pre-HF groups, with high heterogeneity (I² = 96% for T1, 94% for ECV). T2DM, AL amyloidosis, and pulmonary hypertension showed the most pronounced elevations (e.g., T1 1415.41 ms in AL amyloidosis [18]; ECV 36.23% in T2DM with high HbA1c [21]). These findings underscore T1 and ECV’s potential as non-invasive biomarkers for identifying pre-HF, offering a window for early intervention to prevent HF progression.

4.1. T1 Meta-Analysis Findings

The meta-analysis of T1 values included 15 studies, demonstrating a significant overall increase in T1 in pre-HF groups compared to controls (MD: 27.62 ms, 95% CI: 8.04–47.19, p < 0.006). A 27.62 ms increase is clinically meaningful, as deviations of 20–50 ms from normal T1 (~950–1000 ms at 1.5 T) often indicate a significant pathology, such as fibrosis or amyloid deposition, potentially warranting closer monitoring or intervention [8]. However, the high heterogeneity (I² = 96%, p < 0.00001) indicates substantial variability across studies, which warrants a detailed examination. This heterogeneity may partly stem from differences in CMR field strength (e.g., higher T1 values at 3 T vs. 1.5 T), sequences (MOLLI vs. ShMOLLI), and contrast agents, which can influence relaxation times and measurement accuracy. The overall MD of 27.62 ms, while statistically significant, is relatively modest compared to some individual study effects, suggesting that T1’s utility may vary depending on the underlying condition. For instance, Fontana et al. [16] reported the largest T1 mean difference of 121.70 ms (95% CI: 107.86–135.54) in ATTR amyloidosis patients, reflecting the profound impact of amyloid deposition on myocardial tissue composition. This large effect size is consistent with the pathophysiology of amyloidosis, where extracellular amyloid protein infiltration markedly prolongs T1 relaxation times [16]. Similarly, Shi et al. [30] showed a substantial T1 increase (MD: 62.33 ms, 95% CI: 43.53–81.13) in patients with hypertrophic cardiomyopathy (HCM) and hypertensive heart disease (HHD), likely due to diffuse fibrosis and myocyte disarray characteristic of these conditions.

In contrast, several studies reported smaller or non-significant T1 differences, contributing to the overall heterogeneity. For example, Snel et al. [15] found a minimal T1 difference (MD: 5.60 ms, 95% CI: −6.31 to 17.51, p = 0.36) in overweight and hypertensive young adults, suggesting that early-stage disease in this population may not yet cause significant T1 prolongation. Similarly, Laohabut et al. [23] reported a non-significant T1 difference (MD: 4.00 ms, 95% CI: −7.76 to 15.76, p = 0.43) in T2DM patients with suspected CAD, indicating that T1 may be less sensitive to early myocardial changes in some T2DM cohorts. The variability in T1 effects is likely influenced by differences in disease severity, patient demographics, and CMR protocols. For instance, studies using 3 T scanners [15] reported higher baseline T1 values (1152.6 ms in pre-HF) compared to 1.5 T scanners (1050.9 ms in pre-HF [19]), reflecting field strength effects on T1 relaxation times [31]. Additionally, the use of different sequences (e.g., MOLLI [19] vs. ShMOLLI [16]) may contribute to measurement variability.

4.2. ECV Meta-Analysis Findings

The meta-analysis of ECV values included 12 studies, showing a significant overall increase in ECV in pre-HF groups compared to controls (MD: 2.97%, 95% CI: 1.88–4.06, p < 0.00001). An ECV increase of 2.97% is clinically significant, as elevations of 1–3% above normal (~25–26%) are associated with adverse outcomes, such as HF admission, and may justify preventive therapies [14]. The high heterogeneity (I² = 94%, p < 0.00001) again indicates substantial variability, which we explore in detail. Similar to T1, this heterogeneity could arise from variations in field strength, sequences, and contrast agents, which affect ECV calculations through differences in contrast dynamics and signal-to-noise ratios. The overall MD of 2.97% suggests a consistent but modest increase in interstitial expansion across pre-HF populations, which is clinically meaningful given that normal ECV values typically range from 25% to 26% [5]. The largest ECV difference was observed in Gao et al. [21], with an MD of 6.50% (95% CI: 5.08–7.92) in T2DM patients with high HbA1c levels, reflecting severe interstitial expansion. This large effect size aligns with the study’s focus on T2DM patients with poor glycemic control, where chronic hyperglycemia drives significant extracellular matrix remodeling [21]. Similarly, Shu et al. [24] reported a substantial ECV increase (MD: 5.90%, 95% CI: 4.70–7.10) in T2DM patients, further emphasizing ECV’s sensitivity to diabetic myocardial changes.

Other studies showed more moderate ECV differences, contributing to the overall heterogeneity. For instance, Alabed et al. [20] reported an MD of 5.63% (95% CI: 4.96–6.30) in PAH patients, reflecting increased interstitial expansion due to right ventricular pressure overload in PAH. In contrast, Snel et al. [15] showed a negative MD (−1.25%, 95% CI: −2.05 to −0.45), suggesting lower ECV values in overweight/hypertensive young adults compared to controls, which may reflect early-stage disease with minimal fibrosis. However, this finding could also be influenced by the lack of indexing of ECV to body surface area (BSA), as unadjusted ECV values might underestimate interstitial expansion in overweight individuals due to variations in body composition. This effect may be compounded by higher hematocrit levels in overweight patients, often due to hypoventilation, potentially further underestimating ECV by altering contrast dynamics [15]. This potential influence of BSA indexing on ECV measurements may be the subject of future studies, which could explore whether adjusting ECV to BSA improves its accuracy in detecting interstitial expansion across diverse populations. AS Bojer et al. [29] reported a smaller ECV difference (MD: 1.40%, 95% CI: 0.50–2.30) in T2DM patients, possibly due to a less severe disease phenotype or differences in patient selection (RQS 25.0%, NOS 5).

4.3. T1 and ECV as Biomarkers in Pre-HF

T1 and ECV mapping demonstrated sensitivity for detecting subclinical myocardial changes across diverse pre-HF populations. The meta-analysis results highlight ECV’s robustness as a biomarker, with a consistent increase (MD: 2.97%) across studies, driven by large effects in T2DM (e.g., MD: 6.50% [21]; MD: 5.90% [24]). This aligns with prior evidence linking interstitial fibrosis to diabetic cardiomyopathy, even in asymptomatic individuals [14]. In contrast, T1’s smaller overall effect (MD: 27.62 ms) suggests that it may be less sensitive to early changes in some pre-HF populations, though it is highly effective in conditions like amyloidosis (very high T1 values in Hwang et al. [18]). Hypertensive cohorts showed increased ECV (26.0–29.28% vs. 24.6–26.0% in controls), correlating with strain impairment in Kuruvilla et al. (p < 0.05, RQS 33.3%, NOS 5) [22], but T1 differences were smaller (e.g., Kuruvilla et al., MD: 22.20 ms, 95% CI: 4.51–39.89), indicating that ECV may be a more reliable marker for early fibrosis in hypertension.

4.4. Prognostic Implications

The meta-analysis supports ECV’s prognostic utility, with a consistent increase (MD: 2.97%) across studies. Wong et al. [14] found that ECV > 30% predicted HF admission (HR: 1.52, p < 0.01) in T2DM patients (RQS 77.8%, NOS 8) [14], while Laohabut et al. [23] linked ECV to cardiovascular outcomes (p = 0.004) [23]. The large ECV effect in Gao et al. [21] (2019, MD: 6.50%) suggests that T2DM patients with poor glycemic control are at particularly high risk, potentially benefiting from early interventions like SGLT2 inhibitors [9]. T1’s prognostic value is less clear, with a smaller overall effect (MD: 27.62 ms) and slightly higher heterogeneity (I² = 96%). While T1 is highly sensitive in amyloidosis [18], its utility in T2DM and hypertension is less consistent, as seen in smaller effects (e.g., Snel et al. [15] MD: 5.60 ms). The lack of longitudinal data in many studies limits long-term outcome assessment, highlighting the need for prospective studies with robust follow-up. Only three studies reported HF events linked to T1/ECV, with incidences ranging from 4.5% to 51.3% over follow-up periods and associations primarily with elevated ECV (e.g., HR 1.52–4.247). This scarcity of data on HF incidence directly tied to T1/ECV values underscores the need for future research to investigate these associations more comprehensively, including links to hard outcomes like mortality, which were rarely addressed in the included studies, with only Wong et al. [14] reporting deaths as part of events.

4.5. Clinical Applications

T1 and ECV mapping have tangible clinical applications, particularly in risk stratification and surveillance. For example, in hypertrophic cardiomyopathy (HCM), T1 and ECV may improve risk stratification for implantable cardioverter-defibrillator (ICD) implantation. Current guidelines use extensive late gadolinium enhancement (LGE) to estimate ventricular tachycardia risk, but T1 and ECV provide a more global assessment of myocardial fibrosis, potentially offering superior or additive prognostic value [32]. A study by Avanesov et al. found that extracellular volume by CMR predicted the 5-year risk of sudden cardiac death and syncope or non-sustained ventricular tachycardia in HCM, supporting its potential role in ICD decision making [33].

Dedicated randomized controlled trials, such as the PARABLE trial investigating sacubitril/valsartan in pre-HF populations with preserved ejection fraction, highlight the growing focus on early intervention in pre-HF, where T1 and ECV could guide patient selection for such therapies [10].

In surveillance, T1 and ECV could optimize screening for inherited cardiac conditions like HCM or transthyretin (TTR) amyloidosis. For HCM, first-degree relatives require lifelong echocardiography every 3–5 years to detect preclinical disease [6]. T1 and ECV, which detect fibrosis before hypertrophy, could identify at-risk individuals earlier, enabling timely interventions like beta-blockers. Normal T1 and ECV values may allow longer screening intervals, reducing healthcare burden. Similarly, in TTR amyloidosis, gene-positive, phenotype-negative individuals could benefit from T1/ECV monitoring to detect early myocardial changes, guiding therapies like tafamidis [34]. These applications highlight T1 and ECV’s potential to address clinical challenges, though cost-effectiveness studies are needed. Studies indicate that early detection strategies for pre-heart failure (pre-HF), such as natriuretic peptide (NP) testing and AI-enhanced ECG screening, are cost-effective, with NP-based approaches reducing cardiovascular disease and HF prevention costs [35], and AI-ECG screening at age 65 costing USD 43,351 per QALY gained [36]. Screening in high-risk groups like type 2 diabetes patients is also cost-effective at common willingness-to-pay thresholds [37]. Regarding cardiac magnetic resonance (CMR), costs range from USD 350 to USD 3065 per scan in the US, while heart failure (HF) hospitalization costs range from USD 10,737 to USD 17,830 per event [38]. CMR-guided strategies show potential long-term savings, such as 42% to 52% in low-risk patients [39] or 3% reduction (EUR 23 per patient) after 10 years [40], by averting HF events and reducing unnecessary procedures. However, more comprehensive cost-effectiveness studies are needed to fully evaluate CMR in pre-HF detection across diverse populations.

4.6. Methodological Variability and Quality

The high heterogeneity in both T1 (I² = 96%) and ECV (I² = 94%) meta-analyses reflects methodological variability, as indicated by RQS scores (17.2–77.8%, mean: 37.9%). Higher RQS scores were associated with prospective designs, large sample sizes, and clear validation (e.g., RQS 77.8% [14]; RQS 52.8% [19]), while retrospective, single-center studies with small samples scored lower (e.g., RQS 22.2% [30]; RQS 25.0% [23]). NOS scores (5–8, mean: 6.2) indicated moderate to high quality, with lower scores for studies with limited follow-up or controls (e.g., Fontana et al. [16], NOS 5). The AMSTAR-2 rating for Alabed et al. [20] was moderate (RQS 36.1%). Variability in CMR protocols (e.g., field strength, sequences) likely contributed to heterogeneity, with 3 T scanners (e.g., Snel et al. [15]) showing higher T1 values than 1.5 T scanners (e.g., Cerne et al. [19]) [31]. Differences in contrast agents across studies may also exacerbate heterogeneity by altering gadolinium kinetics and ECV calculations.

4.7. Clinical Implications and Future Directions

T1 and ECV mapping offer a non-invasive approach to detect pre-HF, with the meta-analysis confirming their utility (T1 MD: 27.62 ms, ECV MD: 2.97%). ECV’s consistent effect and large increases in T2DM (e.g., Gao et al. [21]) suggest that it may be a more reliable biomarker for early risk stratification, particularly in high-risk groups. T1’s utility is more condition-specific, with larger effects in amyloidosis (e.g., Hwang et al. [18]). CMR’s ability to detect architectural distortion before functional impairment, unlike echocardiography-based strain imaging, supports its role in prevention rather than treatment [6]. Integration into clinical practice faces challenges: high costs, limited CMR availability, and labor-intensive segmentation [31]. Due to these limitations, alternative imaging modalities such as speckle tracking echocardiography (STE) should be considered for detecting subclinical myocardial dysfunction in pre-HF cohorts, as it provides a more accessible, cost-effective, and comprehensive noninvasive evaluation of left ventricular mechanics. Recent meta-analyses and studies have demonstrated that STE effectively identifies impaired left ventricular global longitudinal strain (GLS) in amyloidosis patients (e.g., average GLS magnitude significantly reduced compared to controls) and metabolic syndrome cohorts (e.g., subtle systolic dysfunction with reduced GLS and global strain rate), highlighting its potential for early detection and risk stratification in pre-HF populations [41,42]. AI-based automated segmentation, as explored in Hwang et al. [18], could enhance feasibility [43]. The high heterogeneity and variability in study quality underscore the need for prospective, multi-center studies with standardized CMR protocols, as advocated by the SCMR [6]. High-quality studies like Wong et al. [14] provide a model for future research, and initiatives like the UK Biobank CMR study [44] could validate T1 and ECV thresholds, potentially guiding preventive therapies.

4.8. Risk Stratification

We propose a multi-tiered risk-stratification algorithm for pre-heart failure (pre-HF) populations, incorporating initial low-cost screening tools (e.g., AI-enhanced ECG or natriuretic peptide testing) before proceeding to cardiac magnetic resonance (CMR) with T1 and extracellular volume (ECV) mapping. This approach aims to optimize resource use, improve detection accuracy, and personalize management. The algorithm integrates clinical factors, biomarkers, and reassessment protocols for dynamic risk adjustment.

(1) Identify and Initially Screen High-Risk Individuals: Target asymptomatic adults (age ≥ 45) with Stage A pre-HF risk factors (e.g., type 2 diabetes mellitus [T2DM] with HbA1c > 7%, hypertension [BP > 140/90 mmHg], genetic predispositions like hypertrophic cardiomyopathy [HCM] family history, obesity [BMI > 30 kg/m²], or coronary artery disease). Perform low-cost initial screening with AI-ECG (sensitivity ~80% for asymptomatic left ventricular dysfunction [ALVD]) or natriuretic peptide (NP) testing (e.g., NT-proBNP > 125 pg/mL). If negative, routine follow-up (e.g., annual clinical assessment). If positive, proceed to confirmatory speckle tracking echocardiography (STE) or CMR with T1/ECV mapping.

(2) Perform STE or CMR and Classify Risk Based on GLS/Strain Rate or T1/ECV Values:

• Low-Risk: GLS > −18% (normal range) or T1 < 1000 ms (1.5 T) or <1200 ms (3 T) AND ECV < 26%. Indicates minimal myocardial changes. Recommend standard preventive care (lifestyle modifications: diet, exercise, smoking cessation) with reassessment every 3–5 years via repeat screening (AI-ECG/NP preferred over CMR for cost).
• Moderate-Risk: GLS −16% to −18% or T1 1000–1100 ms (1.5 T) or 1200–1300 ms (3 T) OR ECV 26–30%. Suggests early interstitial expansion/fibrosis. Initiate enhanced monitoring (e.g., quarterly clinical visits, home BP/glucose tracking) and intensified lifestyle interventions (e.g., structured exercise programs, weight loss targets). Consider adjunctive therapies like ACE inhibitors/ARBs if hypertension is present. Reassess with AI-ECG/NP, STE, or CMR every 1–2 years.
• High-Risk: GLS < −16% or T1 > 1100 ms (1.5 T) or >1300 ms (3 T) OR ECV > 30%. Indicates significant subclinical changes with high HF progression risk. Trigger evidence-based preventive pharmacotherapy (e.g., SGLT2 inhibitors for T2DM/hypertension) and frequent surveillance (e.g., biannual AI-ECG/NP or STE, annual CMR). Refer to cardiology for a comprehensive evaluation.

(3) Integrate Multimodal Data for Personalized Management and Reassessment: Combine STE (e.g., GLS, strain rate) or T1/ECV results with clinical factors (e.g., HbA1c > 8%, elevated NP levels, abnormal global longitudinal strain on echocardiography [<−16%], age > 65, multiple comorbidities). Use a scoring system (e.g., weighted: T1/ECV 40%, biomarkers 30%, clinical risks 30%) to refine the risk category if borderline. Tailored interventions may be considered, e.g., add ARNI for moderate risk with strain abnormalities. Reassess dynamically: every 6–12 months for high risk (adjust based on therapy response, e.g., repeat STE or CMR if GLS improves >10% or T1/ECV improves >10%); annually for moderate risk; extend to 3–5 years for low-risk stable patients. Monitor for progression to Stage B (structural changes) or C (symptomatic HF), and adjust accordingly. This algorithm may be validated in prospective studies for efficacy and cost-effectiveness.

4.9. Limitations

This review is limited by variability in study quality (RQS: 17.2–77.8%, NOS: 5–8), small sample sizes (e.g., Shu et al., 2024 [24]), and high heterogeneity (I² = 96% for T1, 94% for ECV), reflecting differences in populations, CMR protocols, and study designs. Heterogeneity is further amplified by inconsistencies in field strength (1.5 T vs. 3 T), sequences (MOLLI vs. ShMOLLI), and contrast agents, which can lead to disparate T1/ECV measurements and complicate pooled analyses. The lack of histological validation challenges the specificity of T1 and ECV for fibrosis versus other processes like edema [31]. Variability in CMR protocols and inconsistent pre-HF definitions further complicates comparisons.

5. Conclusions

This systematic review and meta-analysis demonstrates that T1 and ECV mapping in CMR can detect early myocardial changes in pre-HF populations, with elevated T1 and ECV values compared to controls indicating interstitial expansion and fibrosis. The meta-analysis confirms significant increases, supporting their utility in pre-HF detection. These techniques offer a promising non-invasive approach for risk stratification in T2DM, hypertension, and other high-risk groups, with ECV showing particular prognostic value for HF events. Clinical applications, such as ICD risk stratification in HCM and surveillance for inherited conditions, highlight their potential to improve patient care [32,33]. High heterogeneity, small sample sizes, and methodological inconsistencies (e.g., CMR protocols, study designs) limit clinical applicability. Future research should prioritize standardized CMR protocols, prospective multi-center studies, and automated analysis tools to enhance reliability and integrate T1 and ECV mapping into routine practice, ultimately improving HF prevention strategies.