Early diastolic dysfunction in sarcoidosis without overt cardiac involvement

Introduction

Cardiac involvement in sarcoidosis may remain clinically silent and is frequently undetectable even with normal cardiac MRI or PET imaging. Echocardiographic indices of diastolic function, particularly E/e′, may reveal early functional abnormalities. This study aimed to compare diastolic function between patients with sarcoidosis without objective cardiac involvement and healthy controls, and to determine whether sarcoidosis is independently associated with higher left ventricular filling pressures.

Purpose

To assess whether patients with sarcoidosis without cardiac involvement on advanced imaging exhibit subclinical diastolic dysfunction, as evaluated by echocardiographic E/e′, compared with healthy controls.

Methods

We conducted a retrospective case–control study including patients with biopsy-confirmed sarcoidosis and no evidence of cardiac involvement on MRI or PET, matched 1:1 to healthy controls. Baseline demographic characteristics and cardiovascular risk factors were collected. Echocardiographic assessment included tissue Doppler–derived e′ velocities and the E/e′ ratio. Multivariable linear regression was performed to evaluate the independent association between sarcoidosis and E/e′, adjusting for age, sex, hypertension, diabetes, dyslipidemia, obesity, atrial fibrillation, chronic kidney disease, prior stroke, sleep apnea and left ventricular (LV) ejection fraction.

Results

A total of 72 individuals were included (36 sarcoidosis and 36 controls). The two groups were comparable in baseline characteristics, with no significant differences in age (50.5 years vs 51 years, p=0.875) or the prevalence of hypertension (27.8% vs 36.1%, p=0.448), diabetes mellitus (13.9% vs 5.6%, p=0.233), dyslipidemia (38.9% vs 36.1%, p=0.808), obesity (13.9% vs 27.8%, p=0.147), atrial fibrillation (2.8% vs 5.6%, p=0.555), chronic kidney disease (5.6% vs 8.3%, p=0.643), prior stroke (2.8% vs 5.6%, p=0.555) and obstructive sleep apnea (19.4% vs 8.3%, p=0.173). Median LV ejection fraction was also similar between groups (60.8% vs 60.5%, p=0.505).

In the unadjusted analysis, E/e′ was higher in patients with sarcoidosis compared with controls, although the difference did not reach statistical significance (p = 0.063).

In the multivariable linear regression model, sarcoidosis was independently associated with higher E/e′ (p = 0.016). No issues of multicollinearity were detected.

Conclusion

In this study, patients with sarcoidosis without cardiac involvement on CMR or PET showed higher E/e′ after adjustment for cardiovascular risk factors, indicating subtly increased left ventricular filling pressures. These findings suggest that early diastolic impairment may already be present despite normal advanced imaging.