Cardiac MRI parameters for Risk Prediction in Cardiac Sarcoidosis: Insights from a Multicenter Registry

Thursday, February 5, 2026

8:40 AM - 8:50 AM

Location: Asia IV and Hall 8 (3rd Floor)

Presenting Author(s)

Jonathan A. Pan, MD, MSc, MBA

Assistant Professor
University of Virginia Health System
Charlottesville, Virginia, United States

Primary Author(s)

Bishow Paudel, MD

Advanced Cardiovascular Imaging Fellow
University of Virginia
Charlottesville, Virginia, United States

Co-Author(s)

NH

Nisha Hosadurg, MD

Advanced Cardiovascular Imaging Fellow
University of Virginia, United States
RP

Romil Patel, BSc

Medical Student Researcher
University of Virginia School of Medicine, United States
TH

Thomas C. Hilton, Jr., MD

Resident Physician
University of Virginia, United States
CD

Cristiane C. De Carvalho Singulane, MD

Postdoctoral Researcher
University of Virginia Health System
Charlottesville, Virginia, United States
J. Jane J. Cao, MD, MPH

Director of Translational Research, System director of cardiac imaging
St. Francis Hospital, The Heart Center
Roslyn, New York, United States
KP

Karli PIpitone, MPH

Research Analyst
St. Francis Hospital & Heart Center
Huntington, New York, United States
AS

Angel (AJ) Salcedo

Clinical Research Scholar
St. Francis Hospital, The Heart Center, United States
JW

Jonathan Weber, MPH

Research Analyst
St. Francis Hospital, The Heart Center
Greenvale, New York, United States
SQ

Sophia Qiu, BSc

Clinical Scholar
St. Francis Hospital & Heart Center
Roslyn, New York, United States
RY

Ryan R. Yang, MD

Internal Medicine Resident
University of Chicago, Illinois, United States
Jeremy A. Slivnick, MD

Assistant Professor
University of Chicago
Chicago, Illinois, United States
AD

Anthony Deng

College Student
St. Francis Hospital & Heart Center, United States
GL

Gabrielle Liberatoscioli

College Student
St. Francis Hospital, United States
Amit R. Patel, MD, FSCMR

Professor of Medicine
Division of Cardiology, University of Virginia Health System, Charlottesville, Virginia, USA.
Charlottesville, Virginia, United States

Background: Cardiac magnetic resonance (CMR) plays a pivotal role in diagnosing and risk-stratifying cardiac sarcoidosis (CS), particularly through assessment of left ventricular ejection fraction (LVEF) and late gadolinium enhancement (LGE). CS is associated with a high risk of adverse events, including sudden cardiac death, ventricular arrhythmias, and progressive heart failure. This study aimed to investigate the relationship between LVEF, LGE burden, and myocardial strain with clinical outcomes in patients with CS.

Methods: We conducted a retrospective analysis of data from a multicenter cardiac sarcoidosis registry (N = 327) that met the 2016 Japanese Circulation Society (JCS) guidelines or the Heart Rhythm Society (HRS) consensus statement diagnostic criteria. CMR images were analyzed using NeoSoft SuiteDXT (v5.1.1). Volumetric and functional measurements for the left ventricle (LV) were obtained from short-axis and long-axis views. Late gadolinium enhancement (LGE) was defined as myocardial regions with signal intensity >5 standard deviations above normal myocardium as the reference region of interest (ROI), visible in at least two imaging planes. The longitudinal strains were obtained along the long axis, with additional circumferential and radial strains measured along the short axis. The global longitudinal strain value of < -18% and the global circumferential value of < - 16 were used as normal cut-off values; for the radial strain, > 35 % was used as a normal value. The primary composite endpoint included sustained ventricular tachycardia or fibrillation (VT/VF), heart failure hospitalization, all-cause mortality, or left ventricular assist device (LVAD) implantation/heart transplantation. CMR variables analyzed included LVEF, LGE burden, and LV strain parameters. Univariate logistic regression was used to analyze the relationship between CMR parameters and the primary endpoint using SAS version 9.0.

Results: The demographics and extracardiac organ involvement of the patient population are presented in Table 1. Representative segmentations, strain curves, and LGE findings are illustrated in Figure 1. Compared to LVEF >50% patients, those with LVEF 35–50% had an increased risk (OR 2.59), and those with LVEF < 35% had the highest risk (OR 5.77). LGE extent demonstrated a graded risk relationship: 2–5% (OR 2.97), 5–10% (OR 3.91), and >10% (OR 7.48) compared to the < 2% LGE group. LV longitudinal strain >–18% (OR 4.48), LV circumferential strain >–16% (OR 3.28), and radial strain ≤35% (OR 5.69) were each associated with significantly higher risk.

Conclusion: In cardiac sarcoidosis, adverse outcomes are associated with reduced LVEF, greater LGE extent, and abnormal myocardial strain. Further studies involving larger cohorts of cardiac sarcoidosis are necessary to enable multivariable analyses and to assess the additional predictive value of CMR parameters on outcomes.

Table 1: Demographics and extracardiac organ involvement (total subjects Nf 327).

Age Mean (SD)	57(12)
Male N (%)	55 (32%)
BMI mean (SD)	31(7)
Race N (%)	White -194 (59%) African American -18 (5%) Other 16 (5%)
Smoking N (%)	31 (10%)
Hypertension N (%)	142 (43%)
Diabetes N (%)	54 (17%)
HLD N (%)	126(39)
Medications
Beta-Blocker N (%)	79 (24%)
ACE/ARB N (%)	61 (19%)
Steroids use N (%)	94 (29%)
Extracardiac Organ Involvement
Lungs N (%)	157 (48%)
Skin N (%)	20 (6%)
Liver N (%)	13 (4%)
Eye N (%)	8 (2%)
CNS N (%)	7 (2%)
Other N (%)	11 (3%)

BMI= body mass index, N= number, HLD = hyperlipidemia, ACE = Angiotensin converting enzyme, ARB= Angiotensin receptor blocker, CNS = central nervous system. Table 2: Univariate Analysis of Predictors of Adverse Outcomes (Death, LVAD, transplant, HFH, sustained VT/VF)

Category	Variables	Odds Ratio (Unadjusted)	95% CI	P value
LVEF	>50% (ref)	1.00
	35–50%	2.59	1.34–5.02	0.0048
	< 35%	5.77	2.78–11.98	< 0.0001
LGE extent	< 2% (ref)	1.00
	2–5%	2.97	1.48–5.97	0.002
	5–10%	3.91	1.70–9.00	0.0013
	>10%	7.48	3.22–17.34	< 0.0001
LV Peak Strains	LAX global longitudinal Strain (GLS) ≤–18% (ref)	1.00
	GLS >–18%	4.48	1.56–12.85	0.0053
	SAX global circumferential Strain (GCS) ≤–16% (ref)	1.00
	SAX GCS >–16%	3.28	1.78–6.07	0.0002
	SAX global radial Strain (GRS) >35% (ref)	1.00
	SAX GRS ≤35%	5.69	2.66–12.15	< 0.0001

LVAD= left ventricular assist device, HFH= heart failure hospitalization, VT= ventricular tachycardia, VF= ventricular fibrillation, LVEF= left ventricular ejection fraction, LGE= late gadolinium enhancement, LV= left ventricle, LAX= long axis, SAX= short axis, ref= reference level. Figure 1: A-F: Decreased global longitudinal (GLS) and circumferential strain (GCS) with normal LV ejection fraction and LGE < 2% of LV mass. G: Marked LGE of 22% of the LV mass. *A: Short axis segmentations for left ventricle (LV) endocardial and epicardial borders from base, mid to apical segments. B: Long axis segmentation for LV showing mitral valve annular plane (X) and the apex (X) with endocardial and epicardial borders. C: Phase-Sensitive Inversion Recovery showing late gadolinium enhancement (LGE) involving basal inferior and inferior-lateral wall in different planes. D: LV short axis normal radial strain E: LV short axis decreased circumferential strain (Peak strain -13.1%) F: LV long axis decreased peak strain -14.7%.