Diagnostic challenges in isolated cardiac sarcoidosis

Abstract

Isolated cardiac sarcoidosis (iCS), defined by granulomatous inflammation limited to the myocardium, represents the most diagnostically challenging and prognostically adverse form of sarcoidosis. As it is fundamentally a diagnosis of exclusion, iCS diagnosis relies on the absence of extracardiac sarcoid and the integration of multimodality imaging, endomyocardial biopsy, and molecular testing, each with inherent limitations. Cardiac magnetic resonance and positron emission tomography provide complementary assessment of inflammation, fibrosis, and ventricular function, enhancing diagnostic confidence. Electroanatomic mapping-guided biopsy may improve histologic yield, whereas genetic testing helps exclude phenocopies such as arrhythmogenic, hypertrophic, or dilated cardiomyopathies. Circulating biomarkers remain non-specific but may complement imaging-based algorithms. Future research should focus on harmonized imaging protocols, non-FDG radiotracers, and molecular tissue profiling to refine activity assessment and guide therapy. Multimodal, probability-based frameworks represent the most promising approach for earlier, more accurate diagnosis and risk stratification in iCS.

Graphical Abstract

Graphical Abstract.

*With negative AFB (acid-fast bacilli) and GMS (fungal) stains. Abbreviations: AV, atrioventricular; CMR, cardiac magnetic resonance; CT, computed tomography; EAM, electroanatomic mapping; EMB, endomyocardial biopsy; FDG, fluorodeoxyglucose; iCS, isolated cardiac sarcoidosis; LGE, late gadolinium enhancement; LVEF, left ventricular ejection fraction; PET, positron emission tomography; RV, right ventricular.

Introduction

Sarcoidosis is a multisystem granulomatous disease resulting from an exaggerated immune response to environmental or infectious antigens in genetically susceptible individuals.^1–3 Population studies suggest the prevalence of systemic sarcoidosis ranges from ∼10 to 40 per 100 000 in the USA and Europe, with higher prevalence in women compared to men, and African Americans compared to White populations.^4–9 Among its manifestations, cardiac sarcoidosis (CS) is particularly serious, accounting for most sarcoidosis-related mortality through its association with conduction abnormalities, ventricular arrhythmias, and heart failure.^10,11 Clinically apparent cardiac involvement occurs in approximately 2%-7% of patients with systemic sarcoidosis, while autopsy and advanced imaging studies suggest subclinical myocardial involvement in up to one-third of cases.^11–13 Patients with systemic sarcoidosis should therefore be monitored for cardiac manifestations, because the absence of cardiac involvement at a single time point does not exclude future development of CS.¹⁴ Although evidence is insufficient to support fixed-interval screening, changes in clinical status or new cardiac symptoms should prompt re-evaluation.^15,16

Isolated CS (iCS), defined by granulomatous inflammation confined to the myocardium without evidence of extracardiac disease, and poses the greatest diagnostic and therapeutic challenge. The reported prevalence among patients with CS ranges from 3% to 60%.¹⁷ It usually presents in middle age and is associated with more severe LV dysfunction, conduction block, and malignant ventricular arrhythmias than systemic CS, leading to higher hospitalization and mortality.¹⁷ These observations underscore that iCS is not a benign or early stage of systemic sarcoidosis but rather a distinct, high-risk phenotype often recognized only after advanced cardiac involvement has occurred.

Unlike systemic sarcoidosis with secondary cardiac involvement, iCS lacks extracardiac clues such as lymphadenopathy, pulmonary infiltrates, or elevated systemic biomarkers, leaving clinicians to rely heavily on cardiac imaging, electrophysiologic findings, and, when feasible, endomyocardial biopsy (EMB). However, each of these modalities is limited: imaging findings may be subtle or non-specific, biopsy is constrained by sampling error, and clinical presentation often mimics other inflammatory or arrhythmogenic cardiomyopathies. Consequently, iCS frequently eludes timely recognition, with many patients first diagnosed following life-threatening arrhythmias or advanced heart failure.

Recent multi-societal consensus statements have moved from binary diagnostic thresholds towards probability-based frameworks that integrate multimodality imaging, histologic data, and clinical context to define diagnostic certainty (Figure 1).^14,18,19 Concurrent advances, such as electroanatomic mapping (EAM)-guided biopsy, hybrid imaging, and emerging molecular or genetic markers, are reshaping the diagnostic landscape. Yet, major uncertainties remain regarding how to refine iCS diagnostic algorithms to account for heterogeneity across clinical, geographical, and technological domains.

Figure 1.

Diagnostic overview for isolated cardiac sarcoidosis. Abbreviations: AV, atrioventricular; CMR, cardiac magnetic resonance; CT, computed tomography; FDG, fluorodeoxyglucose; iCS, isolated cardiac sarcoidosis; LGE, late gadolinium enhancement; LVEF, left ventricular ejection fraction; PET, positron emission tomography. Created in BioRender. *with negative AFB (acid-fast bacilli) and GMS (fungal) stains

This review focuses on the diagnostic challenges of iCS, summarizing current evidence and expert consensus on its definition, diagnostic approach, and differential diagnosis. We highlight the biological and clinical features that distinguish iCS from systemic sarcoidosis and discuss emerging imaging and molecular tools that may enable earlier and more reliable detection of this elusive entity. A detailed discussion of therapeutic strategies and treatment monitoring is beyond the scope of the present review.

Definition

The presence of iCS denotes CS occurring without demonstrable extracardiac involvement after thorough systemic evaluation. Definitions differ across major societies. The Japanese Circulation Society (JCS 2016) uniquely allows the diagnosis of iCS even in the absence of extracardiac histology, either through myocardial biopsy demonstrating non-caseating granulomas or, when biopsy is negative or unfeasible, through a predefined combination of major and minor cardiac imaging or clinical criteria.¹⁸ In contrast, the Heart Rhythm Society (HRS 2014) recognizes iCS only when myocardial tissue shows granulomatous inflammation; imaging or clinical features alone are insufficient when extracardiac disease cannot be documented.¹⁹ The American Heart Association (AHA 2024) adopts a probability-based model integrating cardiac magnetic resonance (CMR), fluorine-18-fluorodeoxyglucose positron emission tomography (FDG-PET), electrophysiology, and biopsy results, allowing cases without extracardiac disease to be classified as probable or highly probable iCS, while reserving ‘definite’ status for biopsy-proven disease.¹⁴ Across societies, the essential elements remain the exclusion of extracardiac sarcoidosis and the demonstration (histologic or multimodal) of compatible myocardial involvement.

Clinical suspicion

Unlike systemic sarcoidosis, which often provides extracardiac diagnostic clues, iCS typically manifests through acute or unexplained cardiac events. Suspicion should be high in adults with high-grade atrioventricular block, sustained ventricular tachyarrhythmias, new left ventricular systolic dysfunction, or ventricular aneurysm in the absence of coronary disease or clear myocarditis (Table 1).^17,20,21

Table 1.

Common cardiac manifestations of cardiac sarcoidosis

Clinical manifestation	Frequency in reported cohortsa	Notes
Palpitations	40%-60%	Often due to ventricular arrhythmias or conduction abnormalities
(Pre-)Syncope	20%-50%	Frequently related to high-grade AV block or ventricular tachycardia
Dyspnea/fatigue	20%-40%	May reflect heart failure or restrictive cardiomyopathy
Chest pain	10%–30%	Can mimic ischemic heart disease; often non-obstructive
Signs of heart Failure	10%–20%	Usually later in disease course
Sudden cardiac death	10%–20%	May be the first manifestation, particularly in younger patients
High-grade AV block (≥ type 2)	23%–30%	One of the most common conduction disturbances in CS
Monomorphic VT	10%–20%	Typically scar-related, often presenting with palpitations or syncope
Polymorphic VT/VF	5%–10%	Less frequent but associated with high risk of sudden cardiac death

AV, atrioventricular; VF, ventricular fibrillation; VT, ventricular tachycardia.

^aFrequencies vary across published cohorts^18,22,23

In observational cohorts, more than 90% of patients with iCS are hospitalized for AV block, heart failure, or ventricular tachycardia, often with echocardiographic abnormalities despite no evidence of systemic disease.²⁴ Ventricular tachycardia occurs in about one-third of cases, sometimes progressing to ventricular fibrillation, while LV dysfunction (LVEF <50%) is more frequent than in systemic CS.^22,24

Because clinically silent extracardiac sarcoidosis is common, iCS can be over- or misclassified if a comprehensive extracardiac assessment is not actively pursued.^22,24 Nonetheless, many patients truly present without systemic involvement, and the diagnosis of iCS is often established only after a major cardiac event triggers detailed multimodality evaluation.

Biomarkers

Since the diagnostic trajectory of iCS remains complex, a reliable circulating biomarker could greatly aid early detection, treatment guidance, and follow-up. To date, no circulating biomarker has been validated as a reliable indicator of myocardial granulomatous activity or response to immunosuppression in iCS.²⁵ Still, several routinely available markers may provide complementary diagnostic and prognostic information.²⁶

Angiotensin-converting enzyme (ACE), the most widely used biomarker in systemic sarcoidosis, is elevated in approximately two-thirds of patients and correlates with LGE on CMR, but it lacks specificity, as it may also rise in other granulomatous conditions such as tuberculosis or fungal infection.^26–29 Soluble IL-2 receptor (sIL-2R), a marker of T-cell activation, has been associated with atrioventricular block, ventricular arrhythmias, heart failure hospitalization, and all-cause death in CS and may be more sensitive than ACE for detecting systemic disease (sensitivity 88%, specificity 85%).^30–33 However, its role in iCS remains uncertain.

B-type natriuretic peptide (BNP) and N-terminal pro-BNP (NT-proBNP) reflect haemodynamic stress rather than granulomatous activity. Although frequently elevated in CS,^{27–29,34,35} and associated with cardiac involvement in one study on sarcoidosis (AUC 0.85),³⁶ they likely reflect ventricular dysfunction rather than disease-specific activity. By contrast, high-sensitivity cardiac troponins T (hs-cTnT) and I (hs-cTnI) have been associated with active myocardial injury, worse response to corticosteroid therapy, and cardiovascular events even in patients without overt cardiac involvement at baseline.^37,38 In a recent cohort, troponin T, NT-proBNP, and creatinine independently predicted major adverse outcomes (LVAD implantation, transplantation, or death), even after adjustment for ejection fraction.²⁶ Persistently elevated troponin levels may therefore signal ongoing myocardial inflammation or injury despite normal imaging findings.

Overall, biomarkers remain adjunctive and cannot replace imaging or tissue-based evaluation in iCS. Their future value will likely lie in multi-marker approaches integrated with clinical and imaging data for diagnosis and longitudinal monitoring.

Imaging

Multimodality imaging is central to the diagnostic pathway of CS. In the 2024 AHA scientific statement, the proposed algorithm, CMR with gadolinium, is recommended as the first-line test following clinical suspicion, while FDG-PET is recommended when CMR is inconclusive or normal but suspicion remains high.¹⁴ The 2025 European Society of Cardiology (ESC) guidelines, while recognizing that the diagnostic pathway in iCS is challenging, state that typical multimodality imaging patterns are usually sufficient.³⁹ Importantly, each imaging modality provides complementary strengths and limitations (Table 2).

Table 2.

Overview of imaging modalities and their advantages and disadvantages in diagnosing cardiac sarcoidosis

Modality	Advantages	Disadvantages	Accuracya	Current use
TTE	Wide accessibility Low cost No radiation	Low sensitivity Low specificity	SE∼25–50% SP∼70–80%	Screening Follow-up
FDG-PET	High sensitivity Prognostic value of FDG	Lower specificity (other cardiomyopathies, recent cardiac procedures) Fasting protocol needed	SE∼80–90% SP∼60–75%	Rule-in Assess disease activity Monitor treatment response
CMR	Prognostic value of LGE No dietary preparation needed No radiation	Early stages may precede detectable LGE No evidence to support its use as a surrogate treatment target	SE∼75–100% SP∼70–85%	Rule-out Prognostic stratification Evaluating competing causes
EMB	‘Gold standard’ Diagnostic certainty	Low-yield (sampling errors) Risk of complications	SE∼20–30% SP∼100%	If iCS and diagnostic uncertainty
Voltage-guided EMB	Potentially increased sensitivity compared to EMB	Time consuming Requires supplementary equipment and training Lacks validation	SE∼25–50% SP∼70–80%	Selected cases

CMR, cardiac magnetic resonance; EMB, endomyocardial biopsy; FDG-PET, fluorodeoxyglucose positron emission tomography; iCS, isolated cardiac sarcoidosis; LGE, late gadolinium enhancement; SE, sensitivity; SP, specificity; TTE, transthoracic echocardiography.

^aReported sensitivity and specificity ranges are derived from heterogeneous cohorts of cardiac sarcoidosis and may vary depending on disease stage, imaging protocols, and reference standards.

Echocardiography

Despite limited sensitivity and specificity, transthoracic echocardiography can be useful for initial screening for patients with iCS manifestations, because of its wide accessibility and low cost. It may show reduced LVEF, regional wall aneurysm, basal septal thinning, and abnormal global longitudinal strain. In rare cases, massive inflammation can lead to wall thickening.^40,41

Positron emission tomography

FDG-PET, combined with computed tomography (PET/CT) or CMR (PET/CMR), is the most widely used technique to identify inflammatory diseases.^42–45 The 2025 ESC guidelines recommend FDG-PET for the diagnostic work-up, including detection of myocardial inflammation, as well as for follow-up and assessment of therapeutic response in patients with CS (I, Level B).³⁹

Since FDG uptake can reflect glucose metabolism in activated macrophages within granulomas,⁴⁶ proper patient preparation is essential to suppress physiological myocardial glucose uptake. Current guidelines recommend preparation with a fat-enriched diet lacking carbohydrates for 12–24 h prior to the PET scan, prolonged fasting (12–18 h), and/or the use of intravenous heparin approximately 15 min prior to FDG injection;⁴⁴ however, a fat-enriched diet/prolonged fasting may more effectively suppress physiology FDG uptake.⁴⁷ Moderate- (or higher-) intensity exercise that may promote gluconeogenesis should be avoided. Use of β-hydroxybutyrate to confirm ketosis can also help verify adequate physiological FDG suppression.⁴⁸

Abnormal focal or focal-on-diffuse uptake indicates CS, whereas diffuse homogeneous uptake is typically non-specific (Table 3).⁴⁴ Meta-analyses report a pooled sensitivity of 75–89% and a specificity of 78–83%, whereby inadequate preparation markedly reduces accuracy, and adding perfusion imaging can enhance diagnostic precision.⁴⁶ FDG-PET alone is insufficient to establish the CS diagnosis, as various other disease entities can cause FDG uptake.^49,50

Table 3.

Practical interpretation of myocardial ¹⁸F-FDG uptake patterns and key differentials in suspected cardiac sarcoidosis

Interpretive step	Typical FDG pattern	Differential	Suggested clue or next step
1. Exclude physiologic uptake	Diffuse homogeneous uptake	Inadequate dietary suppression	Repeat scan with high-fat/low-carbohydrate preparation
2. Identify focal inflammatory patterns	Focal or focal-on-diffuse uptake	(ICI-) Myocarditis, Cardiac sarcoidosis	Correlate with clinical context, ICI exposure, extracardiac findings, perfusion imaging, CMR
	Focal, focal-on-diffuse, or multifocal myocardial uptake; LV, RV, or biventricular involvement may occur	‘Hot’ phases of ACM/ARVC/ALVC	Correlate with CMR and genetic testing
	Usually absent or non-specific uptake	Cardiac amyloidosis	Correlate with amyloid risk scores, 99mTc-SPECT/CT ± CMR
3. Evaluate perfusion–metabolism mismatch	Reduced perfusion with increased FDG uptake	Active cardiac sarcoidosis or hibernating myocardium	Assess coronary anatomy/ischemic history, integrate with CMR
4. Uptake in infarct territory	Focal uptake in coronary distribution	Subacute myocardial infarction-related inflammation	Correlate with recent infarction history, CMR
5. Mass-like intense uptake	Nodular focal uptake corresponding to mass	Cardiac tumour or metastasis	Confirm with CMR or CT

ACM, arrhythmogenic cardiomyopathy; ALVC, arrhythmogenic left ventricular cardiomyopathy; ARVC, arrhythmogenic right ventricular cardiomyopathy; CMR, cardiac magnetic resonance; CT, computed tomography; FDG, fluorodeoxyglucose; ICI, immune-checkpoint inhibitor; LV, left ventricle/left ventricular; MI, myocardial infarction; PET, positron emission tomography; RV, right ventricle/right ventricular; SPECT/CT, single photon emission tomography.

Cardiac magnetic resonance

CMR is pivotal in suspected CS and is recommended by the 2025 ESC guidelines for assessment of myocardial involvement using tissue characterization (I, Level B).³⁹ CMR provides high-resolution images of the myocardium, tissue inflammation, and fibrosis (Figure 2).⁵¹ Late gadolinium enhancement (LGE) imaging is able to detect areas of fibrosis with high sensitivity and is most often distributed in a patchy, non-ischaemic pattern, but subendocardial and even transmural involvement is possible.^52,53 A ‘hook sign’ of septal LGE extending into the right ventricular free wall has been coined as an imaging biomarker for CS, but an identical pattern can be seen in giant cell myocarditis.^54,55 Myocardial oedema and active inflammation can be assessed with T2-weighted imaging and T2 mapping.⁵⁶

Figure 2.

Two examples of isolated cardiac sarcoidosis (iCS). Case 1 (left): A 41-year-old Asian male with biopsy-confirmed iCS presenting with atypical right bundle branch block. A) Parasternal long-axis echocardiography demonstrating basal septal wall thickening; B) CMR short-axis late gadolinium enhancement showing a hook-shaped septal pattern with right ventricular involvement; C) FDG-PET demonstrating intense septal uptake (SUVmax > 8); D) Electroanatomic voltage map (Case 1) revealing a large low-voltage region on the right-ventricular aspect of the interventricular septum; green points mark endomyocardial biopsy sites containing non-caseating granulomas. Case 2 (right): A 45-year-old Caucasian female with biopsy-confirmed iCS presenting with fast, sustained ventricular tachycardia. E) FDG-PET demonstrating intense septal uptake (SUVmax > 8); F) CMR short-axis late gadolinium enhancement showing a hook-shaped septal pattern with right-ventricular involvement; G) and H) endomyocardial biopsy histopathology showing extensive non-caseating granulomas with multinucleated giant cells and surrounding fibrosis. Abbreviations: CMR, cardiac magnetic resonance; FDG-PET, fluorodeoxyglucose positron emission tomography; iCS, isolated cardiac sarcoidosis; SUVmax, maximum standardized uptake value

Together with lesion localization, CMR contributes to risk stratification, since the absence of LGE is considered to be a good prognostic marker, although it might not rule out the presence of CS entirely.⁵⁷ Diagnostic performance estimates vary because a robust gold standard is often lacking (outside of small autopsy studies), and discordance with histopathology has been reported in advanced cases.^58–60 Accordingly, most consensus statements emphasize a complementary (additive) role of CMR with FDG-PET, particularly when inflammatory activity or treatment response assessment is required.^14,61 Limitations include reduced interpretability in patients with implantable devices and the lack of validated CMR-based strategies for monitoring response to immunosuppression.^61–63 Finally, novel artificial intelligence (AI)-based ‘virtual native enhancement’ techniques may allow gadolinium-free imaging, though validation in CS is pending.⁶⁴

Endomyocardial biopsy

EMB remains the diagnostic reference standard for CS when non-caseating granulomas are identified in myocardial tissue (Figure 3), and is particularly important for confirming iCS in the absence of extracardiac disease.^18,19,65 However, EMB has limited sensitivity due to the patchy distribution of granulomatous inflammation and preferential involvement of the left ventricular free wall or basal septum, while routine sampling is typically performed in the right ventricle. As a result, diagnostic yield is modest (∼30%–40%), and a negative biopsy does not exclude CS, especially in late or fibrotic disease.^11,66–69 Diagnostic yield may improve when biopsies are pursued in patients with high probability features, including ventricular tachycardia or fibrillation, LVEF <45%, elevated cardiac troponins, and LGE of the mid-apical septum.⁶⁷

Figure 3.

A) HE-staining (x200) of endomyocardial biopsy in a CS patient with sarcoidal granulomas (x200). B) Azan staining (x200) of endomyocardial biopsy in CS patient with sarcoidal granulomas and fibrosis (blue). The images were provided by the Institute of Cardiac Diagnostic and Therapy (IKDT) Berlin, Germany. Used with permission

To mitigate the limitations of conventional sampling, advanced tissue-based and image-guided approaches are emerging. Myocardial gene-expression profiling has revealed distinct molecular signatures capable of differentiating CS from giant-cell myocarditis, even when granulomas are absent, and may serve as a complementary diagnostic tool when histology is inconclusive.^70,71 EAM-guided biopsy represents another procedural advance.^72,73 Notably, EAM has been demonstrated to guide EMB effectively, enabling the retrieval of myocardial samples from areas of abnormal voltage in the ventricular wall, thereby reducing sampling error and enhancing EMB sensitivity.^74–76 Following the development of 3D-EAM systems, some centres have introduced a novel technique for EMB execution in patients with ventricular arrhythmias. This approach involves performing EMB in ventricular segments displaying electrical abnormalities identified by EAM. Early versions of this method relied on catheter–bioptome colocalization, whereas later refinements electrically connected the bioptome to the mapping system, creating an electric dipole.⁷² This allows a real-time visualization of the bioptome within the 3D map and enables precise, live targeting of the low-voltage area.^77–80

Recent reviews demonstrate that this EAM-EMB approach, compatible with CARTO and Ensite NAVx mapping systems, improves both diagnostic accuracy and procedural safety compared with conventional fluoroscopy-guided EMB, particularly in patients with ventricular arrhythmias.^77,79 Further improvement was possible with the use of high definition voltage maps, which enable an even more precise sampling of low voltage areas.⁷⁹ The details of the technical setup are highlighted in Figure 4. Careful tissue handling and processing remain essential to maximize diagnostic yield. The 2025 ESC guidelines recommend considering voltage-guided or imaging-guided EMB in patients with suspected CS when standard sampling is non-diagnostic (IIa, Level C).³⁹ However, evidence remains largely observational, and the magnitude of sensitivity gain and its cost-effectiveness are still uncertain. Nevertheless, in specialized centres, targeted EMB integrated with multimodality imaging represents a promising strategy to improve the diagnostic yield and safety of myocardial tissue sampling in iCS.^{72,74,75,77–79}

Figure 4.

Voltage-guided endomyocardial biopsy using the NavX Abbott, USA. EAM included a 3D reconstruction of the left (LV) and/or right ventricle (RV) as well as intracardiac ECG recording using the Advisor HD Grid mapping-catheter (Abbott, USA). The electroanatomic map was used to localize areas of low voltage representing diseased myocardium. Endomyocardial samples were obtained through the right femoral vein through a disposable bioptome (Bipal, Biosense Webster) using a steerable sheath (Agilis NxT, St Jude Medical)

Differential diagnosis

iCS shares several clinical and morphologic features with other non-ischaemic cardiomyopathies that may involve both ventricles and present with arrhythmogenic phenotypes.⁸¹ Recognizing these overlaps is crucial, as a correct diagnosis of CS enables the timely initiation of immunosuppressive therapy and can alter prognosis.¹⁹ Moreover, differential diagnoses must be considered if patients with postulated iCS do not improve, or even deteriorate, despite high-dose immunosuppression. The most frequent differential considerations include arrhythmogenic cardiomyopathy (ACM), hypertrophic cardiomyopathy (HCM), and Fabry disease.

ACM

iCS is among the most important mimickers of genetically determined ACM.⁸² Indeed, iCS can fulfil the 2010 Task Force Criteria for arrhythmogenic right ventricular cardiomyopathy (ARVC), presenting with major right ventricular dysfunction and ventricular arrhythmias that may initially lead to a misdiagnosis of ARVC.⁸² Histologic evidence of non-caseating granulomas on EMB remains the only definitive discriminator. When patients labelled as ARVC undergo systematic evaluation, including biopsy, CS is identified in up to 15% of cases.⁸³ Electrocardiographically, CS tends to show longer PR intervals, higher-grade atrioventricular block, and broader QRS complexes compared with ARVC.⁸⁴ On imaging, CS more frequently exhibits biventricular or left-dominant dysfunction, late gadolinium enhancement (LGE) in the septal or basal LV regions, and focal FDG uptake on PET, features rarely seen in pure desmosomal ACM.^84,85 Interestingly, such differences were recorded in patients who otherwise fulfilled definite 2010 Task Force Criteria, thus highlighting the limit of these criteria when it comes to differentiate between CS and genetically-determined ACM in clinical setting.

HCM

Although no prospective or registry data directly compare CS and HCM, numerous case reports describe striking phenotypic similarities. CS may manifest with septal or apical wall thickening mimicking HCM, particularly in early disease stages. In contrast, HCM typically exhibits diffuse or asymmetric hypertrophy with patchy, amorphous LGE within hypertrophic segments that may evolve towards thinning and a dilated phenotype over time.⁸⁶ In CS, CMR more often demonstrates basal septal subepicardial or mid-myocardial LGE extending towards the right ventricular insertion,⁸⁷ asymmetric septal involvement leading to left ventricular outflow obstruction,⁸⁸ apical involvement with local LGE.⁸⁹ In patients with known systemic sarcoidosis, the presence of asymmetric focal wall thickening should prompt evaluation for CS with CMR and FDG-PET.

Fabry disease

Fabry disease is an X-linked lysosomal storage disorder caused by deficient α-galactosidase A activity, leading to the accumulation of globotriaosylceramide in multiple organs.⁹⁰ Cardiac manifestations include concentric or asymmetric left ventricular hypertrophy, papillary muscle thickening, and progressive myocardial fibrosis, predisposing to malignant ventricular arrhythmias. Although the phenotype is generally more severe in males, heterozygous females may also be affected due to variable X-chromosome inactivation.⁹⁰ Typical imaging features include concentric LV hypertrophy with reduced global longitudinal strain and right ventricular involvement.⁹¹ On CMR, the basal inferolateral wall exhibits characteristic LGE in up to half of patients.⁹² T1 relaxation times, elevated in CS-related fibrosis, are reduced in AFD due to sphingolipid accumulation, a distinction that can enable early diagnosis even before hypertrophy develops.⁹³ Importantly, the onset of myocardial fibrosis in late disease stages might compensate for sphingolipid accumulation, thus resulting in normal T1 times. In advanced disease, superimposed fibrosis may normalize T1 values, complicating interpretation. Confirmation requires an enzymatic assay and GLA gene testing.⁹⁰

Other cardiomyopathies

Beyond these primary mimickers, several additional conditions may reproduce clinical or imaging features of iCS. Viral myocarditis can present with regional wall motion abnormalities, ventricular arrhythmias, elevated cardiac biomarkers, and patchy mid-myocardial or subepicardial LGE, sometimes indistinguishable from CS without histologic confirmation.⁹⁴ Autoimmune and connective tissue diseases, such as systemic lupus erythematosus, rheumatoid arthritis, and systemic sclerosis, can also produce myocardial inflammation together with extracardiac findings that complicate diagnosis.⁹⁵ Chagas disease, endemic in South America but increasingly encountered worldwide, may mimic CS through basal or inferolateral wall motion abnormalities, ventricular aneurysms, conduction disease, and a propensity for malignant ventricular arrhythmias.⁹⁶ In all these entities, multimodality imaging alone may be insufficient, and integration of epidemiologic context, serologic testing, and, when feasible, EMB is essential to avoid misclassification.

Genetic testing

Although CS itself is not linked to a single causative mutation, genetic susceptibility may modulate immune response and disease risk. Genome-wide association studies have identified HLA-region polymorphisms associated with sarcoidosis in European and North American populations, supporting a model of aberrant antigen presentation and autoimmune hyperactivation.⁹⁷ Clinically, these observations are paralleled by reports describing overlap between CS and other genetic-related cardiomyopathies.⁹⁸

Given the substantial diagnostic uncertainty surrounding iCS, particularly when EMB is non-diagnostic, clinicians should maintain a low threshold for genetic testing to exclude inherited cardiomyopathies with overlapping presentations. Red flags that may trigger genetic evaluation are atypical clinical or imaging features for iCS (e.g. predominant right ventricular involvement or arrhythmia-dominant presentations with relatively preserved LV function), a suggestive family history, and a mismatch between imaging findings and the overall clinical phenotype (such as marked fibrosis with minimal inflammatory activity). Comprehensive gene panels targeting desmosomal, intermediate filament, sarcomeric, and metabolic pathways can identify occult forms of ARVC, HCM, and rarer phenotypes such as laminopathies or desmin-related disease. Identifying a pathogenic variant has direct therapeutic and prognostic implications, i.e. guiding family screening, informing device decisions, and preventing inappropriate long-term immunosuppression. In this context, genetic evaluation is not an alternative but rather a complementary diagnostic step in patients with suspected iCS, serving to distinguish inflammatory from genetic myocardial disease when histology and imaging alone remain inconclusive. Additionally, genetic testing should be reconsidered if patients with iCS demonstrate poor or absent response to immunosuppressive therapy.

Future perspectives

Despite recent technological progress, major gaps remain in achieving accurate non-invasive diagnosis of iCS, as current imaging tools only partially capture the underlying pathobiology. Ideally, imaging should mirror histopathological activity, guide therapy, and inform prognosis. Conventional CMR primarily detects ventricular dysfunction and fibrosis through cine imaging and LGE, which together provide only modest prognostic value for arrhythmias or heart failure and have limited sensitivity for early or subclinical disease.⁹⁹ Emerging CMR-based techniques aim to extend beyond visual assessment. Radiomics, which extracts voxel-level information from routine images, has shown preliminary promise in improving the detection of active inflammation when added to standard CMR markers, although clinical validation remains limited.^100,101

Metabolic imaging techniques, including phosphorus-31 MR spectroscopy and hyperpolarized MR, offer the potential to interrogate myocardial energetics and macrophage-driven inflammatory activity, but currently remain experimental and require further validation in CS.^102–105 Given the higher sensitivity of PET for detecting active inflammation, combined PET/CMR represents a promising strategy in iCS. In a prospective cohort of patients with suspected CS, simultaneous assessment of FDG uptake, LGE, and quantitative mapping significantly improved diagnostic performance compared with single-modality imaging, and colocalization of FDG uptake with LGE predicted adverse clinical outcomes.⁶⁰ However, limited availability, higher costs, and restricted access to PET/CMR constrain widespread implementation, underscoring the need for prospective outcome-driven trials (e.g. NCT05954507).¹⁰⁶

Since FDG-PET has inherent pitfalls, including physiological myocardial uptake, dietary preparation, and timing constraints, alternative radiotracers are being developed. Hypoxia imaging with ¹⁸F-fluoromisonidazole avoids physiological myocardial uptake and has shown feasibility in early studies, while ¹⁸F-fluorothymidine targets proliferative inflammatory tissue but may offer weaker signal intensity.^107,108 Fibroblast activation protein inhibitor (FAPI) targets activated fibroblasts involved in inflammation and early fibrosis.⁴³ While most experience derives from post-infarction remodelling, early anecdotal observations suggest potential feasibility in CS.^109,110 The results of future studies, such as the ongoing FLAIR trial (NCT07083011), will be critical to define the clinical utility and potential role of FAPI-PET in iCS.

Recently, artificial intelligence (AI)-enabled ECG analysis has emerged as a low-cost, scalable screening tool that may enrich pre-test probability and optimize downstream imaging utilization. A convolutional neural network trained on raw ECG waveforms demonstrated strong discrimination between CS and controls, although validation in larger and iCS-specific cohorts is required.¹¹¹ Moreover, AI methods may refine the characterization of myocardial infiltration and inflammation across imaging modalities such as echocardiography, CMR, and PET, improving both sensitivity and reproducibility.^112–114

Finally, the advanced diagnostic modalities and workflows outlined in this review may not be universally available. Access to these technologies, as well as to specialized multidisciplinary expertise, varies considerably across healthcare systems and geographic regions. In such settings, pragmatic diagnostic pathways incorporating clinical assessment, electrocardiography, echocardiography, and referral to specialized centres may be required. Efforts to standardize imaging protocols and expand access to advanced cardiac imaging will be important to improve the timely recognition and management of iCS.

Conclusions

Isolated CS is a high-risk, diagnostically elusive phenotype in which timely recognition hinges on probability-based frameworks that integrate multimodality imaging (CMR+FDG-PET), targeted EAM-guided biopsy, where feasible, and a low threshold for genetic testing when histology is negative to avoid missing inherited cardiomyopathies. From a clinical perspective, diagnostic assessment should be interpreted in conjunction with arrhythmic risk stratification and broader management considerations, rather than relying on any single test or binary criterion. Looking ahead, progress will depend on prospective, biopsy-anchored studies with harmonized imaging protocols and endpoints, alongside evaluation of non-FDG tracers and molecular tissue profiling to distinguish active inflammation from scar and to monitor treatment response. Such multimodal, probabilistic strategies offer the clearest path to earlier diagnosis and more precise, outcome-relevant care in iCS.

Declarations

Disclosure of Interest

M.R.G. has received educational/travel support from Abbott. All other authors declare no conflict of interest related to this manuscript.

Data Availability

No data were generated or analysed for this manuscript.

Funding

All authors declare no funding for this contribution.