Prognostic and functional importance of both overt and subclinical left ventricular systolic dysfunction in systemic sclerosis

Objectives: To quantify the frequency and clinical implications of systemic sclerosis (SSc)-associated left ventricular function (LV) impairment.

Methods: Australian Scleroderma Cohort Study participants meeting ACR/EULAR criteria for SSc with ≥1 echocardiographic LVEF measurement were included. Overt LV dysfunction was indicated by reduced LV ejection fraction (LVEF) and subclinical LV dysfunction was measured using impaired LV global longitudinal strain (LV-GLS>-16 %). Those with secondary causes of LV dysfunction (myocardial ischaemia, valvulopathy and pulmonary arterial hypertension) were excluded. Chi-squared tests, two-sample t-tests or Wilcoxon rank-sum tests were used for between-group comparison as appropriate. Generalised estimating equations(GEE) were used to model longitudinal data. Kaplan-Meier and Cox proportional hazard models were used for survival analyses.

Results: Of 1141 participants with no co-morbid cardiac disease, 2.4 % ever recorded a LVEF<50 %, while only 0.6 % ever recorded a LVEF≤40 %. LV-GLS data were available for 90 % of participants at one centre (n = 218). Impaired LV-GLS was detected in 21 % despite LVEF≥50 %. Those with a LVEF<50 % were more frequently male (p = 0.01) with dcSSc (p < 0.01), higher inflammatory markers (p < 0.02) and skeletal muscle disease (p < 0.05). In multivariable analyses, recording a LVEF<50 % was associated with increased mortality (HR2.3, 95 %CI1.0-4.8, p = 0.04). Impaired LV-GLS was also associated with poorer survival in univariable analyses (HR3.4, 95 %CI1.0-11.8, p = 0.05). Those with a LVEF<50 % more frequently recorded WHO Class III/IV dyspnoea (OR3.5, 95 %CI1.6-7.7, p < 0.01), with shorter six-minute walk distance (p = 0.01), higher Health Assessment Questionnaire-Disability Index scores (p < 0.01) and lower Short Form-36 Physical Component Summary scores (p = 0.02). Increased dyspnoea (WHO Class III/IV dyspnoea; OR3.6, 95 %CI1.4-9.2, p < 0.01) was also seen in those with impaired LV-GLS.

Conclusions: Both overt and subclinical SSc-associated LV dysfunction are associated with worse survival and impaired physical function. The frequency of abnormal LV-GLS in those with consistently normal LVEF suggests an under-appreciated burden of subtle LV systolic dysfunction in SSc that has a significant impact on patient symptomatology.