Layer-specific strain and dyssynchrony index alteration in new-onset systemic lupus erythematosus patients without cardiac symptoms.

BACKGROUND

Layer-specific speckle-tracking echocardiography (STE) is a noninvasive approach that assesses subclinical left ventricular dysfunction. We aimed to investigate the (I) alteration of layer-specific STE parameters and the dyssynchrony index; and (II) the disease parameters associated with layer-specific STE change in drug-naïve patients with new-onset systemic lupus erythematosus (SLE) without cardiac symptoms.

METHODS

Thirty-five drug-naïve patients with new-onset SLE and twenty-five healthy controls were enrolled. All individuals received both conventional echocardiographic and two-dimensional STE assessment. The data of layer-specific global longitudinal strain (GLS), global circumferential strain (GCS), and peak systolic dispersion (PSD) were acquired in layer-specific STE.

RESULTS

All patients had a normal left ventricular ejection fraction (LVEF)(mean LVEF: 58%) and conventional echocardiographic parameters were comparable between patients and controls. Decreased layer-specific GLS and elevated PSD were observed in SLE patients (whole layer GLS: -17.6%±3.0% versus -19.3%±2.6%, P=0.02; endocardial GLS: -20.0%±3.2% versus -22.1%±3.0%, P=0.01; epicardial GLS: -15.6%±2.7% versus -16.8%±2.4%, P=0.04; PSD: 41.0±18.9 versus 28.8±10.1 msec, P=0.007). In contrast, there was no difference in layer-specific GCS at three different levels between patients and controls (P>0.05). More severely impaired GLS was observed in patients with higher disease activity, high-risk antiphospholipid antibody (aPL) profile, or renal involvement. The PSD was increased in patients with higher disease activity or a high-risk aPL profile. Correlational analysis showed that GLS at three layers and PSD correlated with high-sensitivity C-reactive protein (hsCRP) levels (whole GLS: r=0.662, P<0.001; endocardial GLS: r=0.637, P<0.001; epicardial GLS: r=0.658, P<0.001; PSD: r=0.390, P=0.021). PSD correlated with epicardial GLS (r=0.360, P=0.047), when treating the hsCRP level, renal involvement, aPL profile, and disease activity as control variables. Multivariate regression showed the hsCRP level and epicardial GLS were predictors of layer-specific GLS impairment and elevated PSD, respectively.

CONCLUSIONS

Drug-naive patients with new-onset SLE are likely to have subclinical GLS impairment and left ventricular dyssynchrony, even in the presence of normal LVEF. SLE-related risk factors are associated with these dysfunctions.