Our study reveals numerous observations with significant novelty. First, we demonstrate that LVMD is significantly increased in patients with SLE, with the highest values being observed in individuals with concomitant APS, followed by those with renal involvement. The heterogeneous contraction pattern of the left ventricle observed for the patients with SLE with normal LVEF and normal GLS indicates sub-clinical myocardial disease. Second, our findings show that aberrant LVMD is associated with more-severe atherosclerosis and higher levels of complement protein C4. Third, SLE subjects with APS exhibited impaired LV diastolic function, as reflected by a slightly higher E/e’ ratio, and mild right ventricular impairment, as represented by lower TAPSE and RV FWS, in both APS and LN, suggestive of bi-ventricular affection.

Heterogeneous myocardial contraction, as evidenced by increased LVMD, is associated with myocardial fibrosis [36,37,38], and has recently attracted increased interest, as it has been found to predict adverse events, in particular ventricular arrhythmias, in patients with ischemic heart disease and various cardiomyopathies [14]. Little is known about LVMD in the general population, but increased LVMD is associated with older age, conduction disturbances and a more-severe CVD risk profile [39]. Indeed, patients with SLE carry a high CVD risk profile based on both traditional and non-traditional risk factors for atherosclerosis [40]. Our observation of a more-heterogeneous contraction pattern in SLE is similar to a previous study that included 35 subjects with SLE, reporting higher LVMD in patients who had a high-risk antiphospholipid antibody (aPL) profile despite exhibiting normal conventional echocardiographic parameters [41]. However, in contrast to our findings, He and colleagues reported increased LVMD in LN compared with other SLE phenotypes [21]. Notably, however, the latter study did not account for the aPL profile, and classified all other phenotypes as extra-renal SLE, potentially influencing the comparison.

The clinical relevance of increased LVMD in patients with SLE remains an area of active investigation [17, 22, 42]. In coronary artery disease and cardiomyopathies, LVMD is associated with worse clinical outcomes [14, 43]. However, whether this translates to a higher CVD risk in SLE requires further investigation. Two recent studies have shown a higher degree of LVMD among patients with SLE who have higher disease activity, positioning LVMD as a potential marker of sub-clinical myocardial dysfunction in patients with SLE [42, 44]. These findings are consistent with the results of our present study, which in a multivariable analysis identifies a significant association between SLE disease activity, assessed by SLEDAI-2K, and LVMD. In the context of clinical relevance, it is important to note that LVMD is easily analysed by semi-automatic and commercially available software.

A unique aspect of our study was the use of high-frequency ultrasound imaging of the carotids and the central arteries, including the aortic arch, which revealed a higher prevalence of atherosclerotic plaques, and higher IMT in the internal carotid artery and aortic arch in patients with increased LVMD. To the best of our knowledge, this association has not been previously explored. Previous studies utilised carotid ultrasound to assess atherosclerosis in SLE versus HC, demonstrating accelerated progression of carotid plaques in SLE [6, 45,46,47]. Our observations suggest that atherosclerosis contributes to sub-clinical myocardial dysfunction, as reflected by greater LVMD, even in the early stages when conventional echocardiographic parameters appear normal. The results warrant increased awareness among clinicians, leading to effective strategies for prevention of CVD events in subjects with SLE, e.g. lifestyle modification, control of disease activity and use of anti-malarials [40].

Previous studies have established an association between APS and LN with an increased burden of atherosclerosis, which is driven by endothelial dysfunction, immune cell dysregulation, autoantibodies, immune complex deposition, complement activation, and cytokine-mediated inflammation [48]. In our multivariable regression analysis, complement C4 emerged as a significant predictor of LVMD, which is a novel finding. Nevertheless, previous studies have examined the associations between CVD and the complement system [49]. A population-based prospective study from southern Sweden reported correlations between complement C3/C4 and CVD risk factors, with only C4 remaining independently predictive of CVD incidence [50]. Yet another relevant study identified baseline serum C4 levels as an independent predictor of stroke in patients undergoing coronary angiography [51].

Hypothetically, high C4 levels reflect inflammation in the atherosclerotic plaques in both the macro- and micro-vasculature, subsequently leading to myocardial dysfunction. This finding may appear paradoxical, as C3 and C4 levels are typically reduced in active SLE due to complement consumption [52]. However, in the present study, it was primarily subjects in remission or with low disease activity on stable immunosuppressive therapy who were included, which may explain the observed C4 levels above the lower reference limit. Similar results were seen in a study conducted by Mulvihill et al., who investigated complement components in childhood-onset SLE, finding higher C4 concentrations in patients with hypertension [53]. The mechanistic interplay between complement activation, atherosclerosis, and myocardial dysfunction warrants further investigation [54]. Several complement proteins, including C3, C5, and C1q, have been implicated in atherogenesis, suggesting a potential avenue for targeted therapies to mitigate CVD risk in the general population and specifically in patients with SLE [55]. To note, a great majority of the included patients in our study were prescribed anti-malarial agents which, apart from reducing thrombotic events, also have been ascribed cardioprotective effects in recent review articles [40, 56].

To our knowledge, this is the first study to assess systematically LVMD in well-characterised SLE sub-groups, including patients with APS, LN and skin and joint manifestations, all of which have distinct CVD risk profiles. The use of advanced STE enables sensitive detection of sub-clinical myocardial dysfunction, even in patients with preserved LVEF and GLS. Another strength of this study is the extended protocol of vascular ultrasound for plaques and IMT measurements.

Several limitations should be acknowledged. First, the relatively small sample size limits the generalisability of our findings and may have reduced the power to detect sub-group differences. Second, the cross-sectional design precludes causal inferences regarding the relationships between vascular and cardiac abnormalities. Third, whereas IMT and the presence of plaques serve as surrogate markers for atherosclerosis, the study includes no characterisation of the coronary circulation. The study focused on sub-clinical cardiac dysfunction and long-term follow-up is needed to assess whether the patients with SLE with increased LVMD are at higher risk of clinical cardiac events, such as arrhythmias and heart failure. Finally, we acknowledge as a limitation that data on some relevant CVD risk factors, e.g. family history and physical inactivity, were not available to us.