In brief, we conducted a case-control study of MS patients and otherwise healthy subjects, all with normal EF. Patients underwent 2D-STE exam. Longitudinal strain and GLS were measured and compared between the two groups. The absolute average value for GLS was lower in MS patients than in controls but did not reach statistical significance. On the other hand, the absolute strain values for basal and mid-myocardial segments were significantly lower in the MS subjects.

It is estimated that MS has a prevalence of 51.21 per 100,000 in developing countries [21]. It has been suggested that a subset of MS patients, despite normal EF, may have some degree of LV dysfunction [22]. This myocardial dysfunction can be detected in STE by a lower absolute value of segmental and global strain [13,14,15]. Gerede et al. proposed that MS patients with a GLS value of under − 16.98 or a GLS rate of under − 1.45 had a faster disease progression [23]. This highlights the need for closer follow-up in MS patients with lower strain rates.

In a study by Sengupta et al., 57 patients with severe MS and 19 healthy controls underwent 2D speckle tracking-based GLS measurement. Patients with severe MS had lower LVEF and absolute GLS values than the control group. GLS in 48 (84.2%) patients with severe MS was below the 25th percentile of controls. After BMV, GLS was significantly improved (before: -14.6 ± 3.3% vs. after: -17.8 ± 3.5%) [24]. In a similar study on 72 patients with isolated MS and 31 controls, no significant difference was found in LVEF and LV end-systolic or diastolic dimensions between the two groups. Absolute GLS in patients with MS was significantly lower than in the control group. Interestingly, no significant differences in GLS measurements among MS sub-groups (mild, moderate, and severe) were reported [25].

In our study, MS patients had lower absolute strain values compared to controls in the basal and middle segments (1–4, 6, 8, 10, and 12). While some studies have had similar results, others have reported differences in strain values of all segments. For instance, Simsek et al. study on 32 patients with isolated MS and 25 healthy controls demonstrated that absolute values of peak systolic strain and strain rate of all heart segments in patients with MS were significantly lower than in healthy subjects. This study used a different method of heart segmentation with 12 segments [26]. On the other hand, Ozdemir et al. showed that patients with severe MS had a reduced longitudinal peak strain and strain rate in all basal and some mid-segments of the left ventricle [27]. In Roushdy et al. investigation, left and right ventricular GLS measurements were performed on 32 patients with MS and 30 healthy subjects. It noted that left and right ventricular absolute GLS values in patients with MS were lower than the control group (-16.5 ± 2.7% vs. -21.0 ± 1.5 and − 18.3 ± 4.7 vs. -19.8 ± 1.3, respectively). The strain measurements in basal and septal segments and RV free wall were significantly decreased. After BMV, examinations were repeated within 24 h and three months later in patients with MS. Compared with baseline values, left and right ventricular absolute GLS were significantly increased within 24 h after BMV. This increase was maintained after three months [22].

LV dysfunction is frequently observed in MS patients [28]. Although LV dysfunction can present with reduced EF, some MS patients may have subclinical LV dysfunction with preserved EF. The exact underlying mechanisms are still up for debate. Some studies, like Sengupta et al., suggested that LV dysfunction is mainly due to hemodynamic abnormality of the heart and can be reversible after valvulotomy [24]. A possible explanation is valvulotomy increases the preload, thus increasing LV end-diastolic volume (LVED). Increased LVED is then believed to improve LV function. Other studies suggest that there may be an intrinsic LV abnormality in MS that may not be completely reversible with valvulotomy [29]. McKay et al. showed that there was either no change or minimal increase in LVED volume after balloon valvuloplasty in MS patients [30]. These findings contradict the reduced LVED theory [28, 31]. Regardless of the mechanism, LV-GLS measurement can aid in detecting subclinical LV dysfunction in rheumatic MS patients with preserved EF.

There is also some pathological evidence on the topic. Waller et al. described the histology of excised stenotic mitral valves and reported evidence of Aschoff bodies in the interstitial tissue of the myocardium [32]. Although the clinical significance of Aschoff bodies is unknown, Waller et al. suggested that they may represent previous fibrinoid necrosis in the heart. This hypothesis may also be reasonable from a pathophysiological point of view. Rheumatic MS is a delayed complication of acute rheumatic fever attack involving all three heart layers (endocardium, myocardium, and pericardium). Although the attack is acute, the inflammation appears to continue and cause fibrosis and scarring of the mitral apparatus. It may be presumable that similar mechanisms also result in endocardium and myocardium fibrosis, as Klein et al. suggested [28]. It has been hypothesized that this fibrosis may extend from the mitral apparatus to the adjacent myocardium, explaining the difference in strain values of basal segments between MS patients and the normal population [22, 27, 33]. Still, there is no conclusive evidence confirming these assumptions. Another hypothesis is that the fibrotic mitral valve can cause poster-basal wall motion abnormality by having a tethering effect [34].

It is noteworthy that segments 13–15 and 17 showed slightly higher absolute strains in severe MS patients than in controls. While the difference in the strain value of these segments was mainly negligible, the strain value for segment 13 was notably higher in the cases. We believe this contradictory finding in this single myocardial segment may be due to some technical error in accurate image acquisition of the apical-anterior segment in the apical two-chamber view and should be assessed more in future studies. Another explanation is that there may be inaccuracies in tracking because of poor image quality or the operator missing control of the tracking.

A major limitation of the current study was the non-matched case-control design. ANOVA-ANCOVA statistical adjustment was used to overcome the important differences in baseline characteristics, and the final results were mainly expectable and compatible with the pathophysiology of rheumatic MS. In addition, we showed that while the absolute strain value was lower in the basolateral segments in the cases compared to controls, in some apical segments, such as segment 13, it might be higher among the cases. Thus, it is presumable that the GLS index alone may not be accurate enough to identify myocardial dysfunction, and longitudinal strain of other segments is required to provide a complete picture.

Another limitation was the rather small sample size of the study, which stems from the rarity of severe rheumatic MS patients. However, this was a common shortcoming among similar studies on this topic. Additionally, intraobserver variability was not measured in the current study. Finally, the imaging artifacts due to respiration may have affected strain measurements’ quality. To solve this issue, we performed image acquisitions at the end of expiration with three registered cycles.