Myocardial fibrosis, defined as an excessive accumulation of extracellular matrix (ECM) proteins, results in pathological ventricular remodeling and, eventually leads to heart failure (HF) [1]. Myocardial fibrosis can be divided into several subtypes including: replacement fibrosis, reactive interstitial fibrosis, endomyocardial fibrosis [2] and infiltrative interstitial fibrosis. Reactive interstitial fibrosis is an adaptive, non-specific response distinguished by a scattered microscopic distribution in the myocardium, occasionally accompanied by local peripheral distribution of blood vessels [3], with sustained activation of pro-fibrotic growth factors including transforming growth factor-$\beta$ (TGF-$\beta$) (Fig. 1), fibroblast growth factor-2 and connective tissue growth factor [4]. Such interstitial form of fibrosis is typically secondary to long-term pressure and volume overload, resulting in hyperactive renin-angiotensin-aldosterone system and adrenergic system, as presented in valvular heart disease [5], chronic hypertension [6], and cardiomyopathies such as hypertrophic cardiomyopathy [7], dilated cardiomyopathy [8] and diabetic cardiomyopathy [7], but also in distal non-infarcted myocardium following myocardial infarction [9]. Early mild replacement interstitial fibrosis is reversible with specific treatment [10]. Infiltrative myocardial fibrosis is caused by excessive storage of misfolded, insoluble, aggregated proteins (amyloidosis) [11] or globotriaosylceramide (Fabry disease) [12] in the extracellular matrix. Replacement myocardial fibrosis often occurs after acute myocardial infarction (AMI), where necrotic myocardial cells are replaced by collagen fibers, forming fibrous cardiac scar, which ensures the integrity of the heart from rupture in the early stages of myocardial infarction [13]. On the other hand, if left untreated and overburdened post-AMI, the fibrotic tissue can spread to the non-infarcted myocardium, resulting in decreased tissue compliance and cardiac dysfunction [14]. In addition, the excessive deposition of ECM damages the mechanical-electrical coupling of myocytes, impairing myocardial contractility and raising the incidence of malignant arrhythmias and sudden death [15]. In addition, epidemiological studies and clinical trials have shown that myocardial fibrosis is an independent risk factor of adverse cardiac events such as AMI, HF, arrhythmia and cardiovascular death [7].

Fig. 1.

Cellular process of myocardial fibrosis after myocardial infarction. Repairing macrophages are recruited to engulf apoptotic neutrophils, and release inhibitory transmitters such as transforming growth factor-$\beta$ and other potent inflammatory inhibitors. These factors can induce the activation of resting fibroblasts with higher expression of fibroblast activation protein (FAP), and trigger the profibrotic process. While cardiac scar tissue maintains the structural integrity and pressure-generating capacity, persistent myocardial fibrosis leads to adverse changes in the structure and compliance of the ventricles, resulting in the progression of heart failure (HF).

Myocardial fibrosis can be detected by a variety of methods in clinical practice. Traditionally, endomyocardial biopsy is the gold standard for determining myocardial fibrosis, despite its invasive and inconvenient properties. In addition, the diagnosis of myocardial fibrosis by endomyocardial biopsy can be challenging due to its low diagnostic yield, especially for diffuse myocardial fibrosis [16, 17]. In the past, imaging examination such as electrocardiogram and echocardiography were applied to observe cardiac electrical conduction, cardiac structure, and function. In recent years, novel imaging techniques have provided more evidence for determining the characteristics of myocardial tissue, such as single-photon emission computed tomography (SPECT) and cardiac magnetic resonance (CMR). Different imaging tests have their unique features. Since multimodality imaging plays an important role in the initial assessment and diagnosis of myocardial fibrosis, here we discuss current available noninvasive imaging techniques and their values in guiding clinical treatment and improving patient outcomes.

Echocardiography, based on the principle of ultrasonic ranging, is a preferred non-invasive technique to examine the anatomical structure and function of the heart and great vessels [18]. Echocardiography has outstanding advantages such as convenience, rapidity, and non-invasively bedside use. Fibrosis can be hinted when structural and functional changes such as abnormal myocardial thickening, and systolic or diastolic dysfunction are observed [19]. The strategy of integrated backscatter analysis (IB) in standard 2-dimensional (2D) ultrasound images is the first attempt for noninvasive evaluation of myocardial fibrosis after infarction using echocardiography [20]. It measures two parameters of ultrasonic tissue characterization: the amplitude of the cardiac cycle-dependent variation of the backscatter integral signal (cdv-IB) and the mean value of IB [21]. IB signal calibrated by the backscatter power from the pericardium. Moreover, in patients with dilated or hypertrophic cardiomyopathy, m-IB during a cardiac cycle was reported to correlate with the severity of myocardial fibrosis [22]. The intensity of septal IB signal increases in patients with hypertrophic cardiomyopathy (HCM). As a marker of interstitial fibrosis, it is associated with a progressive increase in Doppler parameters related to ventricular stiffness such as pulmonary venous backward velocities and mitral peak velocity at atrial contraction [23]. Losi and colleagues [23] further showed that in HCM patients, the occurrence of ventricular tachyarrhythmias was significantly associated with higher IB signal rather than septal thickness. In addition, echocardiographic measurements based on backscatter techniques include signal intensity coefficient (SIC), which utilizes the greyscale signal intensity values generated at the myocardium-pericardium interface resulting from interactions between the ultrasound signal and myocardial tissue [24]. SIC produces measurable differences between diseased and healthy myocardium. In populations carrying genetic variants associated with HCM, SIC values significantly correlate with left ventricular (LV) hypertrophy [24]. However, this observation is not applicable in patients with coronary artery disease. Higher calibrated integrated backscatter (cIB) was not confirmed as a marker of increased myocardial fibrosis, but was associated with higher soluble vascular endothelial growth factor receptor-1 (sVEGFR-1) and soluble receptor for advanced glycation end products (RAGE) plasma levels. Meanwhile, correlation between cIB and myocardial fibrosis has not been proven by histological examination and CMR evaluation [25, 26].

CMR has become the preferred imaging modality for evaluating myocardial fibrosis due to its ability in soft tissue characterization. T1-weighted images for scar and T2-weighted images for edema visualization are essential sequences to characterize soft tissue [39]. CMR imaging-derived parameters, particularly by LGE and T1 mapping sequences, are widely used to identify fibrotic myocardium. LGE can depict local replacement myocardial fibrosis as seen in large focal post-infarct scars, while T1 mapping has the potential in detecting and quantifying diffuse myocardial fibrosis, since it evaluates the T1 relaxation time of myocardial tissue [40].

Recently, animal and clinical studies have demonstrated the feasibility of contrast enhanced CT in detecting fibrosis by CT delayed enhancement (CT-DE). The principle of CT-DE is similar to that of CMR LGE [65]. CT-DE allows quantitative assessment of ECV to evaluate fibrosis. CT-based ECV quantification is effective in assessing myocardial fibrosis, showing a strong correlation with CMR findings. CT-ECV also displayed high diagnostic accuracy in distinguishing LGE-positive from LGE-negative segments [65]. Furthermore, previous study indicated that CT was able to assess myocardial fibrosis in cases where CMR is not available, which still requires verification by further large-scale studies [66]. However, despite excellent specificity, the clinical use of CT-DE is limited by its low sensitivity. The study by Bettencourt et al. [65] showed a sensitivity of 53% and a specificity of 98% in 105 patients with suspected coronary artery disease.

Although higher volume of iodinated contrast agents and lower energies improve spatial resolution, the contrast difference between normal and infarcted myocardium detected by CT-DE is suboptimal compared to CMR [67]. To circumvent this limitation, dual-energy CT improves the characterization of tissue composition and image quality by using an X-ray source that emits 2 different spectra or by employing a 2-layer detector to achieve continuous acquisition of CT in different photon spectra [68].

Nuclear medicine a well-established advanced imaging modality for the diagnosis, and evaluation of cardiovascular disease. The combination of radionuclide imaging with biologically targeted molecules provides unique insight into disease mechanisms at the molecular level, which allows an early detection of damaged myocardium before pathological changes occur.

Myocardial fibrosis is recognized as excessive deposition of collagen. The collagen-targeted contrast agent is the first targeted probe for the detection of myocardial fibrosis after a heart attack. ${}^{99m}$Tc-streptavidin-coupled-collagelin and ${}^{99m}$Tc-CBP1495 are two collagen-targeting peptide tracers that have relatively high affinity for collagen. Significantly increased uptake of these tracers was observed in fibrotic tissues of rat models [69]. However, collagen-targeted peptides can only show the late products of myocardial fibrosis, and thus they are not sensitive for fibrosis detection at earlier stages of the disease. It is also not possible to determine whether myocardial fibrosis is ongoing. Velikyan et al. [70] recently reported a ${}^{68}$Ga-labelled collagelin analogue, which showed promises for the detection of early active fibrosis by binding to monomeric collagen before the collagen fibres mature. However, there is still plenty of uncertainty for clinical application.

Molecular targets of activated fibroblasts at early disease stages are predictive of the extent and severity of cardiac fibrosis. In a rat model of myocardial fibrosis, angiotensin II (Ang II) was highly expressed in activated macrophages and myofibroblasts. Through acting on Ang II type 1 receptors (At1R), Ang II induced the expression of TGF-$\beta$, which is the growth factor most closely associated with the development of tissue fibrosis [71]. Positron emission tomography (PET) experiments using ${}^{11}$C-KR31173 in a porcine myocardial infarction model suggested that the radioactive probe detection of At1R is feasible. The application in human were safe, and showed detectable retention of specific myocardial markers, but at lower levels than that in pigs [72]. Cy5.5-Arg-Gly-Asp (RGD) imaging peptide, a targeting marker for myofibroblasts, can also display interstitial changes in myocardial remodeling and assess fibrosis in response to anti-angiotensin therapy [73]. In the early post-myocardial infarction period, the range of tracer uptake measured by 99Tc-RGD imaging after 3 weeks is comparable to that of CMR imaging. The scar size shown by 99Tc-RGD imaging predicts the eventual scar formation after myocardial infarction.

Fibroblast activation protein (FAP) is expressed at high levels in activated fibroblasts and shows low expression in most normal organs [74]. Radioactively labeled fibroblast activation protein inhibitor (FAPI) is developed to detect activated fibroblasts and initially shows great promise in the diagnosis and treatment of cancer patients. The application of FAPI in cardiovascular disease began with the incidental observation of FAPI in cancer patients by PET. A correlation between tracer uptake and reduced ejection fraction has been observed by FAPI-PET imaging in patients with metastatic cancer [75]. FAPI binds to FAP and accumulates strongly in tissue with high fibroblast activation, showing a high bright signal compared to normal myocardium, with a low background signal. By exploiting the molecular characteristics of myocardial fibrosis, where fibroblasts are highly activated to produce collagen fibers, FAPI can be used as a specific target for the management and treatment of cardiovascular disease. More recently, it has been utilized in murine models and in humans for the assessment of myocardial fibrosis following myocardial infarction (MI) [76, 77, 78]. Serial imaging with ${}^{68}$Ga-FAPI in a MI model established by coronary artery ligation showed intense radiotracer uptake around the infarcted area, and the uptake peaked at day 6 [76]. In the study by Zhang et al. [79] aseline uptake volume (UV) was a powerful predictor of LV remodeling at 1 year after STEMI in 26 patients with ST-segment elevation myocardial infarction (STEMI) who underwent Ga-DOTA-FAPI-04 PET (OR = 1.048, p = 0.011). Lyu et al. [80] found that FAPI imaging was able to detect myocardial fibrosis in diabetic, obese and elderly patients, providing additional evidence for early intervention and clinical decision-making in the management of patients at elevated risk of CVD.

In addition, ${}^{68}$Ga-FAPI PET has been used in a rat model of HF to visualize myocardial fibrosis and monitor HF progression [79]. A study by Guokun Wang et al. [81] showed that fibroblast activation in the heart and liver after pressure overload could be monitored using ${}^{68}$Ga-FAPI-04 PET/CT and that this non-invasive technique was a better predictor of subsequent worsening of heart failure. FAP activity is heterogeneously increased in the myocardium of patients with hypertrophic cardiomyopathy, and their PET-measured FAPI uptake is a potential predictor for 5-year risk of sudden death from cardiovascular causes [82].

However, the cost of test, the worries about radiation, and the poor understanding of nuclear medicine have limited its use clinically. In the future, if these obstacles can be overcome, it will open a new era of targeted treatment and management of patients with myocardial fibrosis.

Early detection and targeted treatment of myocardial fibrosis is essential to improve clinical outcomes in patients with cardiovascular diseases. Multimodality non-invasive imaging approaches can directly or indirectly evaluate the presence and severity of cardiac fibrosis, with advantages and disadvantages of each technique summarized in Fig. 2. In summary, CMR is the gold standard for noninvasive detection and quantification of myocardial fibrosis in clinical practice, whereas other techniques show promises as valuable alternatives. Molecular imaging is developing rapidly and has been a promising technique not only for studying pathological mechanisms, but also for investigating the efficacy of individualized therapeutic regimens to meet the growing need for precision medicine. All the progresses made in the development of novel radiopharmaceuticals targeting specific cardiovascular molecules indicated that the revolution in personalized medicine has only just begun.

Fig. 2.

Multimodality imaging assessment of myocardial fibrosis. The multimodality imaging approaches that are able to assess myocardial fibrosis in clinical practice. LGE, CMR-late gadolinium enhancement; PET, positron emission tomography; MRI, magnetic resonance imaging; ECV, extracellular volume; CT, computed tomography; SPECT, single-photon emission computed tomography.

HZ, KWX, YYQ, ZGZ, MJ and JP contributed to the conception and design of this review. HZ and KWX prepared the initial draft. ZGZ, YYQ, MJ and JP were involved in manuscript proofreading and critical revisions. MJ and JP provided thorough and comprehensive guidance. All authors read and approved the final manuscript. All authors have participated sufficiently in the work and agreed to be accountable for all aspects of the work.

Not applicable.

We would like to thank all cardiologists from Renji Hospital, who have devoted their efforts to the diagnosis and treatment of cardiovascular diseases.

MJ received funding support from the National Natural Science Foundation of China (U21A20341, 81971570); Shanghai Academic/Technology Leader Program (21XD1432100) and Shanghai Science and Technology Commission Program (20Y11910500, 22DZ2292400). JP received funding support from the National Natural Science Foundation of China (81930007); National Key Research and Development Program of China (2022YFC3602400); Shanghai Municipal Health Commission (2023ZZ02021 and GWVI-11.1-26). KWX is supported by the National Natural Science Foundation of China (Grant No. 82000634) and Shanghai Sailing Program (Grant No. 20YF1425000). YYQ was supported by Zhejiang Provincial Natural Science Foundation of China (Grant No. LQ21H050002). ZGZ is sponsored by Shanghai Pujiang Program (21PJD039) and the National Natural Science Foundation of China (Grant No. 82300366).

The authors declare no conflict of interest.