Background/Purpose: The timing and etiology of diastolic impairment in pediatric-onset systemic lupus erythematosus (pSLE) are poorly understood, and the role of screening echocardiography remains unclear. We compared left ventricular diastolic function at pSLE diagnosis with controls and determined the prevalence of abnormal echocardiographic findings.

Methods: Echocardiograms of children with pSLE ages 5-18 years performed within 1 year of diagnosis and age- and sex-matched controls with structurally normal hearts (evaluated for benign murmurs or chest pain during the same period) were re-read by two blinded cardiologists. Baseline characteristics, SLE disease activity index (SLEDAI), and cardiovascular symptoms/signs were abstracted by chart review. Diastolic indices (E/A ratio, e’, E/e’, and isovolumetric relaxation time (IVRT)) were compared using linear mixed effects models adjusted for systemic hypertension. Other abnormalities, including pericardial effusion and valvular disease, were also evaluated. Pearson’s correlation was used to identify factors associated with worse diastolic indices.

Results: 85 children with incident pSLE had baseline echocardiograms, of which 61% were for screening, 15% were for cardiovascular symptoms/signs, and 23% were for other indications. Median time from SLE diagnosis to echocardiogram was 6 days (interquartile range 1-70). Prior glucocorticoid exposure was minimal (Table 1). Diastolic indices were significantly worse in pSLE cases compared to controls (with lower E/A, lower e’, higher E/e’ and longer IVRT) even after adjustment for hypertension (Table 2). 6 pSLE cases (7%) met cutoffs for abnormally low e’ and 32 (47%) had prolonged IVRT, though none met international criteria for Grade I diastolic dysfunction. Mild to moderate pericardial effusions, aortic and mitral insufficiency were present in 15%, 17%, and 6% of all pSLE cases, respectively, including 8%, 15%, and 3% of asymptomatic pSLE cases (n=61). SLE disease activity, but not presence of effusion, was correlated with worse E/e’ (ρ 0.31, p=0.02), septal e’ (-0.30, p<0.01), and IVRT (0.50, p<0.01).

Conclusion: Subclinical echocardiographic findings were prevalent in a group of children with incident SLE, with worse diastolic indices at diagnosis compared to peers without SLE, independent of long-term glucocorticoid use or hypertension. Future longitudinal studies using echocardiography will determine whether these measures of diastolic function worsen over time and if they are prognostic of future cardiac complications.