ABSTRACT
Background Interstitial fibrosis as quantified by cardiac magnetic resonance (CMR) has been demonstrated in arrhythmic mitral valve prolapse (MVP), a condition with known female predominance. However, prior studies included only MVP cases with significant mitral regurgitation (MR) or mitral annular disjunction (MAD). We sought to evaluate the association between interstitial fibrosis and complex ventricular ectopy (ComVE) in MVPs unselected for MAD or severe MR, and to investigate the contribution of sex to this association.
Methods We performed contrast CMR in consecutive individuals with MVP between 2020 and 2022. Extracellular volume fraction (ECV%), a surrogate marker for interstitial fibrosis, was quantified using T1 mapping. Replacement fibrosis was assessed using late gadolinium enhancement (LGE). ComVE, defined as frequent premature ventricular contractions and/or non-sustained/sustained ventricular tachycardia (VT), was detected using ambulatory ECG monitoring.
Results We identified 59 MVP cases without severe MR (49% women, 80% with mild or less MR) and available ECV% measurement. Among these, 23 (39%) had ComVE, including a case of aborted ventricular fibrillation (VF) and one with sudden arrhythmic death, both females. Global ECV% was significantly greater in ComVE versus non-ComVE (31%[27-33] vs 27%[23-30], p=0.002). In MVP-ComVE, higher segmental ECV% was not limited to the inferolateral/inferior LV wall, but was also demonstrated in atypical segments including the anterior/anterolateral wall (p<0.05). The association between ComVE and ECV% was driven by female sex (32%[30-33] vs 28%[26-30], p=0.003 in females; 31%[25-33] vs 26%[23-30], p=0.22 in males). ECV% remained independently associated with an increased risk of ComVE, including VT/VF, after adjustment for cardiovascular risk factors, MAD, and LGE (p<0.01).
Conclusion In MVP without significant MR, interstitial fibrosis by CMR is associated with an increased risk of ComVE, suggesting a primary myopathic process. The stronger association between interstitial fibrosis and ComVE in females may explain why severe arrhythmic complications are more prevalent among women.
INTRODUCTION
Mitral valve prolapse (MVP) is a common heritable condition affecting 2% to 3% of individuals worldwide.1 MVP without significant mitral regurgitation (MR) is generally considered a benign condition. However, a small but non-negligible subset of MVP subjects (0.14% to 1.8% yearly) are at risk of life-threatening arrhythmic events and sudden cardiac death (SCD).2, 3 This “malignant” form of MVP has been commonly associated with female sex, younger age, bileaflet involvement with multi-segmental myxomatous disease, replacement fibrosis in the papillary muscles and adjacent left ventricular (LV) wall, and complex ventricular ectopy (ComVE).3–5 ComVE, defined as non-sustained or sustained ventricular tachycardia (VT) and/or frequent pleomorphic PVCs, may be found even in MVP without significant MR and is associated with excess mortality.5–7
The role of myocardial fibrosis as a substrate for ComVE and SCD in MVP is supported by seminal histologic and imaging studies.8–11 Indeed, MVP-related abnormal valvular-myocardial mechanics related to greater prolapse, mitral annular disjunction (MAD), excessive displacement of papillary muscles, and curling of the basal myocardium may induce myocardial fibrosis.12 Replacement fibrosis, typically involving the papillary muscles and LV inferolateral wall, can be detected by late gadolinium enhancement (LGE) on cardiac magnetic resonance (CMR) and relates to arrhythmic MVP.9, 13–15 However, ComVE may be present even in the absence of replacement myocardial fibrosis.10, 16 In addition to LGE, CMR can detect diffuse or interstitial fibrosis using the extracellular volume fraction (ECV%) calculated from T1 mapping.17, 18 As shown by our group in a post-mortem investigation19, diffuse interstitial myocardial fibrosis could represent an alternative substrate for life-threatening arrhythmia in MVP20, regardless of MR severity or replacement fibrosis. Prior studies investigating the role of T1 mapping/ECV% in arrhythmic MVP have included mostly cases with severe MR, limiting our understanding about the presence of a primary myopathic process in MVP.20, 21 Furthermore, it remains uncertain whether sex-related differences, which may impact not only the mitral valve apparatus, but also myocardial structure and function, could influence the contribution of diffuse myocardial fibrosis to ComVE in MVP.22
We therefore hypothesized that among MVPs without significant MR: 1) interstitial fibrosis as quantified by global ECV% on CMR is increased in those with ComVE, suggesting a primary myopathy in MVP; and 2) the relationship between interstitial fibrosis and ComVE is influenced by female sex.
METHODS
Study Population
We identified 113 consecutive MVPs recruited as part of the University of California San Francisco (UCSF) MVP Registry from 2020 to 2022 with adequate contrast CMR/T1 mapping and without the following: i) severe MR and/or prior mitral valve (MV) intervention; ii) ischemic or non-ischemic cardiomyopathy (including sarcoidosis); iii) LV ejection fraction <35%; iv) complex congenital heart disease; v) >mild aortic valve disease; or vi) prior defibrillator implantation. Eleven studies were excluded because of the lack of comprehensive ambulatory ECG monitoring to ascertain ventricular arrhythmia. All remaining 59 MVPs had complete demographic/ clinical variables within a calculated Charlson comorbidities index,23 and laboratory, ambulatory ECG and CMR data. Ethics Committee of UCSF approved the study and participants gave informed consent. All data produced in the present study are available upon reasonable request to the authors.
Ascertainment of Ventricular Arrhythmia
Ventricular arrhythmia data were ascertained through 48-hour Holter monitors. ComVE was defined as: ≥5% burden of pleomorphic premature ventricular contractions (PVCs) (isolated and/or in couplets/triplets), presence of sustained or non-sustained ventricular tachycardia (NSVT), or ventricular fibrillation (VF).7, 16 Arrhythmic cause of death was ascertained using the National Death Index for SCD cases. PVC burden and origin (i.e., papillary muscles, mitral valve/annulus, LV/RV outflow tract, or other) were also recorded. Pleomorphic ventricular ectopy was defined as PVCs of at least 2 different morphologies. NSVT was defined as > 3 PVCs with a rate >100 bpm lasting <30 seconds.16
Cardiac Magnetic Resonance Imaging
Acquisition
Imaging was performed using the UCSF Core using 3.0T scanners Discovery MR750w or SIGNA Premier (GE Healthcare, Milwaukee, WI).24 Briefly, standard long- and short-axis cine images covering the entire LV were obtained using a balanced steady-state free precession sequence. Typical parameters were TR/TE 2.8/1.3ms, flip angle 45°. LGE was performed 5-15 min following injection of 0.1 mmol/kg gadobutrol, and images were obtained using a high-resolution breath-hold two-dimensional sequence at three separate levels in the short-axis plane (basal, mid, and apical). An inversion-recovery fast gradient-echo sequence was performed in two phase-encoding directions to differentiate true late enhancement from artefact.25, 26 The inversion time was optimized to achieve satisfactory nulling of the myocardium. T1 mapping data were acquired using a Modified Look-Locker Inversion-recovery (MOLLI) sequence with built-in motion correction and an acquisition scheme of 3(3)-3(3)-5 both before and following contrast agent administration. Two-dimension (2D) phase-contrast (through-plane), velocity-encoded, imaging approach through the aortic valve was performed to quantify MR.
Image Analysis
Image analysis was performed offline using a standardized approach (cvi42 version 5.14.0, Circle Cardiovascular Imaging Inc., Calgary, Canada). LV volume and mass were indexed to body surface area. Papillary muscles were included when measuring LV mass and excluded when measuring volumes.26 MVP was defined as superior displacement of one or both mitral leaflets >2 mm beyond the mitral annulus on long axis view, as previously described.8 The MVP phenotype was further subdivided as: bileaflet or single-leaflet (anterior or posterior). MR volume was calculated as the difference between LV stroke volume and the forward aortic flow volume. MR fraction was obtained by dividing MR volume by the LV stroke volume. MR severity was assessed using on a qualitative and/or quantitative approach, with visual assessment or 2D phase-contrast imaging when available.
The presence of MAD (inferolateral vs inferior, anterior, or anterolateral) was assessed qualitatively on cine long-axis views as a separation between the left atrial wall at the level of MV junction and the LV free wall.27 In addition, LV “curling”, defined as an abnormal or exaggerated basal myocardial systolic motion, was qualitatively and quantitatively assessed as previously described.12, 14
On dedicated images, LGE was quantified in a semi-automated manner using a signal intensity threshold of >5 standard deviations above the mean value in a region of normal myocardium. Areas of inversion artefact or signal contamination by epicardial fat or blood pool were manually excluded. LGE quantification was expressed as percentage of LV mass.
T1 mapping data were generated from a short-axis cine stack of the LV, and the regions of the endocardium and epicardium on the native T1 maps were manually contoured to avoid partial-volume or cardiac-motion-related blurring of the myocardium-blood boundary, where blood signal may contaminate myocardial data. The LV mid-section was selected due to its greater reproducibility and use in prior MVP-based CMR studies.21, 28 A 10% offset was also applied to avoid signal contamination from the blood pool and epicardial fat. The pre-contrast contours were then copied onto the corresponding post-contrast maps with minor adjustments made to avoid partial volume effects. Regions of interest were then drawn in the blood, avoiding myocardium and papillary muscles. ECV fraction was calculated according to the following formula: ECV%= partition coefficient x [1-hematocrit], where partition coefficient = [ΔR1myocardium/ΔR1blood-pool] and ΔR1 = (1/post-contrast T1-1/native T1).29 Hematocrit was measured as per protocol immediately before CMR study. Native myocardial T1 and ECV% were also recorded for all myocardial segments, including the LV inferolateral wall, a location typically subject to abnormal traction by the prolapsing leaflets.14, 21, 30
Statistical Analysis
Continuous variables were expressed as mean ± SD or median (interquartile range [IQR]), and were tested for normality of distribution and homogeneity of variances with the Shapiro-Wilk and Levene tests, respectively. Continuous variables with normal distribution were compared between groups with Student’s t test. Continuous variables non-normally distributed were compared with Wilcoxon-Mann-Whitney test. Categorical variables were presented as frequencies and percentages and were compared with 2 test or Fisher’s exact test as appropriate. Univariable and multivariable linear regression analyses were performed to identify the factors significantly associated with diffuse myocardial fibrosis (i.e. ECV%). The multivariable model included all clinically relevant factors and those with a p-value <0.20 in univariable analysis to predict ECV% (i.e. age, inverted lateral T waves, MR severity, LV end-systolic volume index, LGE, and hematocrit). Results were presented as standardized beta coefficient (β) with standard error. A two-tailed p value ≤0.05 was considered significant. Statistical analyses were performed with Stata software version 14.2 (StataCorp, College Station, Texas).
RESULTS
Study Population
Demographic, arrhythmic, and CMR characteristics of the study population are presented in Table 1. Of the 59 MVPs included in this analysis, a total of 23 (39%) cases (48% women, and 30% men) had ComVE. Among these, one later experienced sudden arrhythmic death, and another had an aborted VF arrest treated with an implantable cardiac defibrillator (ICD). Both were females.
PVC burden was significantly higher in the ComVE group, and by definition all cases in this group had PVC burden ≥5% (Table 1). PVCs were pleomorphic in all cases with PVCs burden ≥5%, and originated from typical MVP sites such as the papillary muscles and MV annulus31 (Table 1). The proportion of T-wave inversion, including inverted inferolateral T-waves, was comparable between groups (Table 1).
CMR characteristics, including MR severity (mild or less versus moderate), LV volumes, presence of MAD (inferolateral: 48% vs 54%, p=0.63) and LGE in the LV inferolateral segment (13% vs 16%, p=1.00) and papillary muscles (13% vs 8%, p=0.66), were comparable between groups (Table 1). However, the ComVE group had lower LV ejection fraction (59±7% vs 62±5%, p=0.05) and a trend toward more bileaflet involvement (78% vs 55%, p=0.07) compared to the non-ComVE group.
Ventricular Arrhythmia, Diffuse Interstitial Fibrosis and Sex-Based Differences
Native T1 was greater in the ComVE than non-ComVE group (1284±10 vs 1247±43 ms, p=0.004), while post-contrast T1 was slightly lower in the ComVE group (554±72 vs 580±57 ms, p=0.13). Mean hematocrit value was similar between groups (42.5±3.1 vs 42.6±3.9 ms, p=0.94).
Median global ECV% was significantly greater in MVP with ComVE compared to non-ComVE (31.0% [27.0-33.0] vs 27.5% [23.0-30.0], p=0.006) (Figure 2). Moreover, exclusion of MVPs with LGE (n=19) provided similar results (31.0% [27.0-33.0] vs 27.0% [23.0-30.0], p=0.01). Regardless of ComVE, women demonstrated greater ECV% compared to men (30.0% [27.0-32.0] vs 27.0% [23.0-32.0]), although the comparison was not statistically significant (p=0.10). Importantly, women with ComVE had significantly greater ECV% than those without ComVE (31.5% [30.0-33.0] vs 28.0% [26.0-30.0], p=0.003) (Figure 2). In men, there was no significant difference in ECV% between ComVE and non-ComVE (31.0% [25.0-33.0] vs 27.0% [23.0-30.0], p=0.32) (Figure 2).
Figure 3 shows the distribution of ECV% across the mid LV segments between ComVE and non-ComVE. Median ECV% was significantly greater in ComVE versus non-ComVE across all segments (all p≤0.05), except for the anteroseptal segment (Figure 3A). Furthermore, in MVPs with ComVE, median ECV% was significantly increased in the inferolateral segment compared to the other segments (all p≤0.05), with the exception of the inferior segment (Figure 3A). MVPs without ComVE had a different distribution of ECV%, with no increase in median ECV% in the inferolateral segment compared to other myocardial segments (Figure 3A). Similarly, women with ComVE had significantly greater ECV% across all segments compared to those without ComVE (all p≤0.05), except for the anteroseptal and anterior segments (Figure 3B). Moreover, median ECV% in the inferolateral segment was markedly increased compared to other myocardial segments in women with ComVE, but not in those without ComVE (Figure 3B). In the subset of men, there were no significant differences in median ECV% across all myocardial segments between ComVE versus non-ComVE (Figure 3C). Moreover, in the ComVE group, median ECV% in the inferolateral segment was significantly increased only when compared to the anterolateral segment. In the non-ComVE group, median ECV% in the inferolateral segment was not significantly increased compared to the other myocardial segments (Figure 3C).
Table 2 shows ECG and CMR characteristics according to ComVE and sex. Overall, there were no significant differences between groups, with the exception of higher global native T1 (1302±49 vs 1249±50 ms, p=0.008) and inferolateral native T1 (1335±73 vs 1263±78 ms, p=0.01) in women with ComVE compared to those without ComVE. In men, global native T1 (1255±55 vs 1245±39 ms, p=0.52) and inferolateral native T1 (1292±38 vs 1277±60 ms, p=0.48) were similar between ComVE and non-ComVE. MVPs with ComVE were more likely to have bileaflet involvement and lower LVEF in both women and men (Table 2). The proportion of MVPs with MAD and/or LGE did not differ between groups among women and men (Table 2).
Predictors of Complex Ventricular Ectopy
In an exploratory analysis, we assessed the association of ECV% with the risk of ComVE after adjusting for the Charlson comorbidities index, presence of LGE, bileaflet MVP, inferolateral MAD, and LVEF. In univariable analysis, ComVE risk was only predicted by ECV% (OR: 1.18, 95% CI 1.04 - 1.35; p=0.01). Of note, when stratified based on sex, ECV% was significantly associated with the risk of ComVE in women (OR: 1.44, 95% CI 1.07 - 1.94; p=0.01), but not in men (OR: 1.08, 95% CI 0.93 - 1.26; p=0.31). After multiple multivariable models, ECV% remained independently associated with an increased risk of ComVE in all multivariable models (Figure 4).
DISCUSSION
In this cross-sectional analysis of consecutive and unselected MVPs with predominantly no or mild MR undergoing CMR, we investigated the relationship of interstitial myocardial fibrosis with ComVE, including cases with VT, VF arrest and SCD. The main findings were as follows: 1) Diffuse interstitial fibrosis as quantified by global ECV%/T1 mapping was significantly greater in MVPs with ComVE, regardless of presence of LGE; 2) In the arrhythmic MVP group, higher segmental ECV% was not limited to the inferolateral/inferior LV wall, but was also demonstrated in segments not typically subject to myocardial traction, suggesting a diffuse myopathic process in MVP beyond abnormal valvular-myocardial mechanics; 3) The association between interstitial fibrosis and ComVE was more prominent in women compared to men with MVP (Graphical Abstract).
Myocardial Fibrosis and Arrhythmic MVP
Seminal studies support the role of myocardial fibrosis, resulting from increased mechanical traction of prolapsing valve, as a substrate for ComVE and SCD in MVP.5, 8, 9, 11 Basso et al.9 previously reported that replacement fibrosis is a common feature of young SCD victims (≤40 years of age) with arrhythmic MVP. However, replacement fibrosis is not consistently found in SCD victims with MVP,10 nor in those with ComVE alone (prevalence of LGE 36-45%).20, 32 These findings suggest that other myocardial substrates are contributing to arrhythmic outcomes in MVP.
Non-invasive measurement of diffuse interstitial fibrosis can be achieved by measuring ECV% by T1 mapping on CMR, a technique that has been validated against histology.33, 34 Prior studies have reported abnormal myocardial T1 and ECV%, suggestive of greater diffuse interstitial fibrosis in arrhythmic MVP.21, 28, 35 However, such investigations have included predominantly MVP cases with severe MR or MAD, or have linked abnormal ECV% to a few cases of sudden cardiac arrest rather than the wider spectrum of ComVE.20, 21 ComVE is frequently detected (>80%) in patients with MVP before lethal events and has been linked to mortality excess.5, 6 Therefore, the identification of imaging parameters able to predict ComVE are essential for detection of intermediate categories of risk and overall improved arrhythmic risk stratification. Selection of specific categories of MVP in prior studies (severe MR or MVP with MAD and abnormal valve-myocardial mechanics) have limited our understanding of the distribution and pattern of interstitial fibrosis in MVP beyond a simple response to chronic volume load or a focal fibrotic response to myocardial traction. Another study has assessed ECV% in mild or moderate MR, but only in 12 MVP patients and without investigating segmental contributions to global ECV%.36 In the present study, we demonstrate for the first time in a larger sample of CMR studies unselected for severe MR or MAD, that ComVE in MVP is strongly linked to greater ECV%, even in the subset without LGE or abnormal valvular-myocardial mechanics. Our imaging findings are in line with our prior post-mortem data demonstrating that multisite interstitial fibrosis was more prevalent than replacement fibrosis (100% vs 30%) on autopsy in consecutive and unselected MVPs with sudden arrhythmic death adjudication.10 Therefore, diffuse interstitial fibrosis may represent an alternative substrate for ComVE and SCD in MVP with mild MR, even in the absence of LGE. As such, MVP patients undergoing CMR for assessment of more traditional parameters of risk such as MAD or LGE, may greatly benefit from routine T1 mapping/ECV% calculation as part of their clinical care.
To date, the exact mechanisms for ventricular arrhythmia and SCD in MVP without severe MR are not well understood. It has been hypothesized that the arrhythmogenic risk associated with bileaflet MVP with mild MR could be attributed to stretching forces exerted on the papillary muscles and adjacent LV wall by the prolapsing leaflets.11, 12 These mechanical stresses could trigger myocardial inflammation and contribute to fibrosis as a substrate for ventricular arrhythmia.36 Moreover, MAD, which is more common in bileaflet MVP,37 can augment such abnormal valve-myocardial mechanics. In the present study, we confirmed the regionalized expansion of myocardial fibrosis in arrhythmic MVP. Among MVP ComVE, the inferolateral segment had the highest ECV% within the LV mid-section. Further, the absolute ECV% in the LV inferolateral segment was significantly greater in MVP with ComVE compared to non-ComVE. However, among MVP patients with mild MR, we also highlight a global, multi-segment pattern of interstitial fibrosis which may underline a diffuse myopathic process initially proposed in genetic studies highlighting mutations in cardiomyopathy genes (FLNC, LMNA, ALPK3A).38, 39 A primary myopathy may represent the “substrate” for ventricular arrhythmia in the presence of bileaflet MVP, MAD and myocardial stretch (the “trigger”). Alternatively, such diffuse myopathy may act alone, when bileaflet MVP and associated imaging arrhythmic features are absent. Indeed, in our study sample only 50% of MVP patients with mild MR had the traditional phenotype of bileaflet MVP with MAD and curling. Longitudinal studies are needed to determine the prognostic role of CMR-based diffuse interstitial myocardial fibrosis in arrhythmic MVP either as an imaging surrogate of a primary myopathy or, when regionalized to the inferolateral LV, as a precursor of replacement fibrosis.
Sex Differences, Expansion of Myocardial Fibrosis, and Risk of Arrhythmias
There are conflicting findings regarding the potential effect of sex on MVP arrhythmogenicity. Prior reports of MVP-related SCD/SCA cases have showed a predominance of young women.5, 9
In our study, there was a predominance of women with ComVE (60%), which is consistent with prior observations.40 Importantly, we observed a stronger association between global ECV% and ComVE in women compared to men. In addition, women with MVP and ComVE had significantly greater ECV% in the LV inferolateral segment but also in other segments typically not subject to myocardial traction. Of note, both women and men had similar distributions of replacement fibrosis as quantified by LGE. To our knowledge, this is the first study to investigate and report sex-related differences in the expansion of myocardial fibrosis as measured by CMR in MVP.
Several factors may contribute to sex-specific mechanisms of interstitial fibrosis in arrhythmic MVP including neurohormonal differences, cardiac structural and functional variations, and genetic predisposition.41, 42 Sex hormones, particularly estrogen, are well-known for their impact on the myocardium.43 In this setting, hormonal fluctuations may contribute to ventricular fibrosis and arrhythmias.41, 44 Greater global/diffuse interstitial fibrosis in women with MVP, mild MR, and ComVE may suggest a primary/genetic diffuse myopathy with variable, sex-related penetrance similar to that observed for valvular features, i.e. bileaflet MVP and MAD, which are more prevalent in women.45 Indeed, in addition to greater global ECV%, women with arrhythmic MVP in our study exhibited greater “regionalized” interstitial fibrosis, specifically higher ECV% in the mid inferolateral LV, which was not observed in men. Altogether, these findings support the role of sex-specific factors in the development of myocardial fibrosis and subsequent risk of ventricular arrhythmia in MVP. However, further studies are needed to provide a clearer understanding of the effect of sex in arrhythmic MVP.
Clinical Implications
The identification of those at highest risk for ventricular arrhythmia and SCD within the larger population of benign MVP cases remains a major and unmet challenge.1 Current guidelines focus on management of significant MR, with less emphasis on routine monitoring of ComVE in MVP without MR.46, 47 In the absence of LGE, arrhythmic risk stratification in MVP is even less clear to the practicing cardiologist. In our study, we demonstrate greater interstitial fibrosis in arrhythmic MVP, regardless of MR or LGE, and more so in women. Hence, ECV% quantified by T1 mapping represents an alternative surrogate marker for myocardial fibrosis that is expected to improve risk stratification of “malignant” MVP. Individuals with MVP and abnormal ECV%, particularly women without significant MR and/or LGE, may benefit from closer clinical and imaging monitoring. Furthermore, our findings support the need for better understanding of the underlying sex-specific mechanisms involved in the pathogenesis of arrhythmic MVP.
Limitations
The retrospective design and small sample size of the present study require additional studies to confirm our findings. Nonetheless, this study is the first to our knowledge to characterize the association between interstitial fibrosis and ComVE, in a large series of MVPs undergoing CMR studies unselected for severe MR or ventricular arrhythmia. Importantly, our findings suggest for the first-time sex differences in patterns of fibrosis in MVP which may explain the higher prevalence of arrhythmic complications in women. Our findings need to be validated in further experimental and longitudinal observational studies.
CONCLUSION
In the first large series of consecutive MVP patients with mild MR, ComVE is associated with greater interstitial fibrosis as quantified by ECV% using CMR/T1 mapping, even in the absence of LGE. Measurement of ECV% could help enhance arrhythmic risk stratification particularly in female arrhythmic MVP patients, who exhibit greater interstitial fibrosis compared to men. Further studies are needed to determine if sex-specific mechanisms are involved in the pathogenesis of arrhythmic MVP.
Data Availability
All data produced in the present study are available upon reasonable request to the authors.
Sources of Funding
This work was supported by the National Institute of Health NHLBI R01HL153447.
Disclosures
Nothing to disclose.
ABBREVIATIONS
- MVP
- mitral valve prolapse
- MR
- mitral regurgitation
- SCD
- sudden cardiac death
- LV
- left ventricular/ventricle
- ComVE
- complex ventricular ectopy
- LGE
- late gadolinium enhancement
- CMR
- cardiac magnetic resonance
- ECV
- extracellular volume fraction