MAD or MADness?

Abstract

Mitral valve prolapse (MVP) is a common condition affecting approximately 3% of the population, typically with a benign clinical course. However, a small subset of patients (5–10%) may develop severe mitral regurgitation or arrhythmias, which can lead to sudden cardiac death (SCD). Among the morphological features of MVP, mitral annular disjunction (MAD) has emerged as a potential marker of malignant MVP, with some studies suggesting an association with ventricular arrhythmias and SCD. MAD refers to a structural abnormality where there is a separation between the posterior mitral annulus and the ventricular myocardium, particularly during systole. Initially described in the 1980s, MAD has been primarily studied through echocardiography, although its dynamic nature during the cardiac cycle has complicated its diagnosis. The clinical significance of MAD has been debated, as its presence is not exclusive to pathological MVP, being observed in structurally normal mitral valves as well. Recent research, using advanced imaging techniques such as three-dimensional echocardiography, cardiac magnetic resonance and computed tomography, has provided a more refined understanding of MAD. These studies suggest that MAD can be found in normal hearts, particularly in the posterior mitral annulus, and is often considered a benign anatomical variant. However, the occurrence of MAD in patients with MVP, especially those with leaflet redundancy, has been linked to an increased risk of arrhythmias and SCD. The exact role of MAD in arrhythmogenesis remains unclear, but it is hypothesized that MAD may contribute to electrical instability by altering the mechanical properties of the mitral valve, potentially promoting fibrosis in the surrounding myocardium. Despite these associations, the direct causal role of MAD in SCD requires further investigation, and it may ultimately prove to be an innocent bystander rather than the primary cause of fatal arrhythmias.

Graphical Abstract

Schematic illustration of the current definition of MAD and the new hypothesis

Introduction

Mitral valve prolapse (MVP) is a relatively common condition, affecting approximately 3% of the general population, and is typically associated with a benign clinical course. Indeed, over 90% of individuals with MVP are asymptomatic or exhibit clinically insignificant symptoms such as atypical chest pain, exertional dyspnoea, and palpitations [1, 2]. However, a minority of patients (5–10%) may develop severe mitral regurgitation requiring surgical intervention. Of particular concern, ventricular arrhythmias and sudden cardiac death (SCD) occur in 1–3% of younger patients with MVP, even in the absence of severe valvular regurgitation or left ventricular dysfunction. This subset of the condition is referred to as “arrhythmic” or “malignant” MVP [3].

Since the advent of M-mode echocardiography, cardiologists have sought to identify specific morphological features or “stigmata” associated with arrhythmic malignancy, i.e. characteristics of the mitral valve that correlate with complex arrhythmias and SCD. Historically, leaflet redundancy was the only morphological feature consistently linked to these adverse outcomes. However, more recently, the “mitral annular disjunction” (MAD), a structural abnormality at the junction of the left atrium, left ventricle, and posterior mitral leaflet has emerged as a potential hallmark of malignant MVP.

This review aims to reevaluate the role of MAD in the context of malignant MVP based on the latest evidence. Specifically, the review will address a brief history of MAD, challenges and controversies in its echocardiographic diagnosis, the occurrence of MAD in normally structured mitral valves, and its potential role as a trigger for severe arrhythmias or SCD.

Brief history of MAD

MAD was first described by Bharati et al. in a 1981 case report [4]. The authors detailed a case of sudden death, characterizing the posterior mitral annulus as “elongated, with the valve moored on the left atrial side.” Subsequently, in 1986, Hutchins et al. systematically investigated this structural anomaly and introduced the term “mitral annular disjunction.” They defined MAD as an “anomalous attachment of the posterior leaflet directly to the atrial wall.” Their pathological study identified MAD in 23 out of 25 patients with bi-leaflet MVP and hypothesized that this structural anomaly might serve as the “primum movens”, leading to leaflet floppiness and redundancy [5]. However, key questions remained unresolved regarding the extent of posterior leaflet involvement and the nature of the tissue occupying the gap between the hinge line and the ventricular crest. Indeed, whether this space was filled by the atrial myocardium below the leaflet insertion or by fibrotic tissue was not clarified. Angelini et al. [6] investigated this phenomenon further, finding MAD in prolapsed and standard mitral valves. They argued that the high prevalence of MAD in MVP reported by Hutchins et al. could have been overestimated due to artifacts resulting from abnormal leaflet thickness at the hinge line. Angelini’s group concluded that MAD might represent a simple anatomical variation rather than a pathological finding.

Despite these controversies, the cardiology community initially regarded MAD as clinically irrelevant. Over time, however, clinical studies established a link between MAD and ventricular arrhythmias. Perazzolo Marra et al. [7] were among the first to report convincing evidence of this association. Subsequently, a growing body of research has suggested that MAD, particularly when associated with bi-leaflet MVP and redundant leaflets, could be a pivotal contributor to ventricular arrhythmias and sudden or aborted cardiac death (as discussed in subsequent sections).

Challenges and controversial echocardiographic diagnosis of MAD

MAD has historically been described as a systolic phenomenon. In an intriguing editorial titled “Mitral annular disjunction: The forgotten component of myxomatous mitral valve disease”, Maurice-Enrique Sarano proposed the dynamic nature of MAD. He defined MAD as follows: “In diastole, the diagnosis cannot be made because the ventricular myocardium is appropriately situated under the annulus. Over systole, as the posterolateral myocardium contracts, the annulus ‘slides’ and becomes detached from the ventricular myocardium” [8]. This dynamic conceptualization has been widely accepted in the cardiology community. A recent expert consensus on arrhythmogenic [9] MVP described MAD as “a systolic separation between the ventricular myocardium and the mitral annulus supporting the posterior mitral leaflet…” In this context, MAD is absent when the mitral annulus remains attached to the atrial and ventricular myocardium/endocardium throughout the cardiac cycle. The extent of MAD is quantified as the gap between the posterior leaflet’s hinge line and the crest of the left ventricular myocardium, often referred to as the MAD length. Quantitative assessments have demonstrated that this displacement can range from millimeters to 1 cm.

Using three-dimensional transesophageal echocardiography, Lee et al. analyzed a cohort of 42 patients with MAD and found that the disjunction predominantly involved the P1 and P2 scallops of the posterior leaflet. Interestingly, MAD was associated with a functional loss of systolic annular function, with the “disjuncted” annulus exhibiting paradoxical systolic expansion, effectively decoupling from the left ventricular myocardium in mid-to-late systole [10].

However, this dynamic model raises anatomical challenges. The suggestion that the posterior mitral annulus “slides” upward toward the left atrial roof during systole and returns to its original position in diastole lacks anatomical corroboration. To date, cardiac surgeons have not reported such slackness or mobility of the mitral annulus. During mitral valve repair, surgeons consistently observe that the posterior leaflet and annulus are firmly attached, with prosthetic rings being sutured directly onto the anatomical annulus.

Our group [11] proposed an alternative hypothesis, utilizing cardiac magnetic resonance (CMR) imaging in patients diagnosed with bi-leaflet MVP and MAD via echocardiography. We observed that many “systolic” MAD cases might represent a misinterpretation. During systole, the billowing leaflet often presses against the atrial wall, making distinguishing between the leaflet tissue and the adjacent atrial myocardium challenging. This can create a false appearance for MAD. Additionally, the hyperdynamic motion of the basal ventricular myocardium may contribute to this misinterpretation. We termed this phenomenon “pseudo-MAD.”

In this scenario, the paradoxical systolic expansion of the annulus is not unexpected. Measurements taken on the falsely identified neo-annulus, which corresponds to the atrial wall, naturally reflect atrial systolic expansion. This misinterpretation is widespread in transthoracic echocardiographic images obtained from a long-axis parasternal view. Due to this image’s limited spatial and temporal resolution, the prolapsing leaflet’s belly often appears indistinguishable from the atrial wall (Figs. 1 and 2).

Fig. 1

A Magnified transthoracic echocardiographic image of the posterior atrioventricular junction during diastole. The yellow marker indicates the true annular insertion of the posterior mitral leaflet. B Corresponding image during systole. The yellow marker again identifies the true annular insertion, while the red marker indicates the pseudo-annulus. The image demonstrates that the belly of the posterior leaflet is opposed to the atrial wall, creating a false impression of MAD

Fig. 2

A The end-systolic frame shows the red marker, indicating a neo-migrated annulus. B One frame earlier and C two frames earlier, it becomes evident that the so-called neo-migrated annulus does not exist. The yellow marker consistently identifies the true annulus across all frames. This sequence highlights the importance of frame-by-frame in distinguishing pseudo-MAD from true anatomical disjunction. Abbreviations: LV = left ventricle; LA = left atrium

In rare cases, the hinge line of the posterior mitral leaflet appears disjunct from the atrial wall in both diastole and systole. We define this structural anomaly as true MAD. Frame-by-frame analysis of high-resolution imaging techniques, such as modern echocardiographic systems, cardiac magnetic resonance (CMR), or computed tomography (CT), offers a more detailed view.

Specifically, in mid-diastole, during the diastasis phase, i.e. a brief period when the mitral leaflet is neither fully open nor closed, the hinge line of the posterior leaflet becomes visible. In this “suspension phase,” it is possible to differentiate between true-MAD and pseudo-MAD with greater accuracy (Fig. 3).

Fig. 3

Long axis view in mid-diastole with (A) CMR and (B) TEE. In both cases, the true MAD is visible. LV = left ventricle; LA = left atrium; AO = aorta

This observation underscores the importance of advanced imaging modalities (such as the newest echo machines, CMR, or CT scan) for precisely identifying and classifying mitral annular disjunction. Using all three imaging modalities mentioned above, Fiore et al. [12] reviewed 603 consecutive patients with MVP. By following frame-by-frame the exact position of the hinge line of the posterior leaflet, they found that 221 out of 263 (84%) patients previously classified as “MAD” were reclassified as pseudo-MAD. The prevalence of true MAD was only 7%. Importantly, paradoxical systolic expansion was not observed when annular diameters were measured on the true annulus.

MAD in the normally structured mitral valve

Nearly fifty years ago, anatomical studies were the first to report mitral annular disjunction (MAD) in structurally standard mitral valves. In 1976, McAlpine conducted a thorough investigation of standard heart specimens, identifying a collar-shaped fibrous extension measuring between 3 and 8 mm in height that connects the posterior leaflet’s hinge to the crest of the ventricular wall [13]. Subsequently, Angelini et al. illustrated that this subvalvular membrane is discontinuous, showing interruptions at various points along the posterior hinge line [6]. In the modern era, Dejgaard et al. [14], utilizing cardiac magnetic resonance (CMR), confirmed the presence of MAD in structurally normal hearts. Their study found that MAD was “scattered” along the posterior annulus, with areas of disjunction interspersed with regions of regular annular arrangement. More recently, Toh et al. [15], using cardiac computed tomography, found MAD to be a common finding in normal adult hearts, with a prevalence of 96%. MAD was primarily observed in the P1 and P3 scallops, exhibiting a bimodal distribution. Similarly, Zugwitz et al. reported that 76% of 2,607 participants in the U.K. Biobank study had at least one MAD site. These studies highlighted the high prevalence of MAD when lower diagnostic thresholds (1–2 mm separation between the posterior leaflet hinge line and the ventricular crest) were applied (Fig. 4) [16].

Fig. 4

CMR images (A) in long axis view and (B) bi-commissural view in a patient with normal structured mitral valve. While the hinge line of posterior leaflets in the long axis view is normally located (arrow), in the bi-commissural view, the disjunction is visible (arrows). LV = left ventricle; LA = left atrium; AO = aorta

A pathological investigation by Krawczyk-Ozog et al. [17] examined 224 adult heart specimens without evidence of mitral valve disease. MAD was identified in 20% of cases, primarily in the posterior leaflet (P1: 8.9%, P2: 5.4%, P3: 4.5%) and at commissural sites (superolateral commissure: 9.8%, inferoseptal commissure: 5.8%).

While anatomical and histological studies remain the gold standard for diagnosing MAD, discrepancies with imaging-based studies are expected. These differences may arise from the dynamic nature of the cardiac cycle assessed through imaging, which cannot be captured in post-mortem pathological evaluations. Imaging techniques provide a “dynamic” view of the annulus during systole and diastole. In contrast, pathological studies evaluate a static representation, potentially contributing to variability in MAD’s reported prevalence and characteristics.

Does MAD trigger complex arrhythmias and SCD?

Numerous studies have investigated the relationship between mitral annular disjunction (MAD) and complex arrhythmias, including sudden cardiac death (SCD).

One of the most intriguing studies regarding the causal relationship between MAD and arrhythmias was that of Dejgaard et al. [14]. The study included 116 patients with a diagnosis of MAD at standard clinical echocardiogram. In all, 40 (34%) patients had ventricular arrhythmias i.e., severe arrhythmic events or non-sustained ventricular tachycardia. These patients were younger (age 37 ± 13 years vs. 51 ± 14 years; p = 0.001), had lower ejection fraction (51 ± 5% vs. 57 ± 7%; p = 0.002), and frequently had papillary muscle fibrosis (4 [36%] vs. 6 [9%]; p = 0.03). MVP was evident in 90 out of 116, (78%) patients and, intriguingly, was not associated with ventricular arrhythmia. Conversely, they found that 22% of cases of MAD without MVP, had complex arrhythmias arguing that MAD itself is an “arrhythmogenic entity “. However, these conclusions were never confirmed by other authors who, on the contrary, found that a higher risk of arrhythmias was associated with the combination of MVP + MAD.

Essayagh BA et al. [18], in a large cohort study of 595 patients with bileaflet mitral valve prolapse (MVP) and leaflet redundancy, compared the characteristics and long-term survival of patients with MAD (n = 186) versus those without MAD (n = 409). While no significant difference in overall survival was observed between the groups, patients with MAD exhibited a markedly higher risk of arrhythmias compared to those without MAD. This increased risk persisted even after adjusting for demographic and clinical variables.

Van Wijngaarden et al. [19] analyzed 610 patients with MVP and 133 (22%) with echocardiography-proven MAD. Within this subgroup, patients with ventricular arrhythmias (26 of 67; 39%) had a significantly greater MAD height (8 mm vs. 7 mm) than those without arrhythmias. Furthermore, the prevalence of combined MAD and MVP was notably higher in patients with arrhythmias (39%) than in those without arrhythmias (20%).

However, neither study distinguished between pseudo-MAD and true-MAD, leaving ambiguity about whether one subtype carries a higher arrhythmic risk. Fiore et al. [12] addressed this issue and found that both true-MAD and pseudo-MAD were associated with a similar risk of arrhythmias. Notably, cardiac magnetic resonance (CMR) analysis revealed greater late gadolinium enhancement (LGE), indicative of myocardial fibrosis, in patients with both true-MAD and pseudo-MAD compared to those without MAD.

The observation of increased fibrosis in patients with both forms of MAD recalls earlier hypotheses. In 1992, Sanfilippo et al. [20] proposed that abnormal tension at the tips of the papillary muscles due to MVP could cause electrical instability, predisposing to ventricular arrhythmias. This hypothesis was primarily overlooked until 2002 when Han et al. [21] provided confirmatory evidence. In their seminal study, the authors identified focal regions of LGE in the papillary muscle tips, suggestive of fibrosis, in 63% (10 of 16) of MVP patients but in none of the controls. Furthermore, papillary muscle LGE was significantly associated with complex ventricular arrhythmias in MVP patients.

Figliozzi et al. [22] studied 474 patients with MVP without significant mitral regurgitation or left ventricular dysfunction (mean age 47 ± 16 years, 244 women) over a mean follow-up period of 3.3 years. They found that LGE, particularly its extent, was a stronger predictor of adverse events, such as sustained ventricular tachycardia, SCD, and unexplained syncope, than the presence of MAD (P = 0.89).

Fibrosis likely represents the most powerful trigger of arrhythmias. As a wound-healing response, fibrosis enhances the risk of potentially lethal cardiac arrhythmias by creating a vulnerable myocardial substrate. Arrhythmias arise when two factors converge: a trigger, such as a premature ventricular complex, and a vulnerable substrate, characterized by interspersed collagen bundles within myocardial strands. This trigger–substrate combination facilitates the initiation of anatomical or functional re-entry circuits, which may degenerate into arrhythmias and SCD [23, 24].

Conclusions

The following conclusions (though not definitive) can be drawn:

• The majority of patients with MVP (> 80%) had a pseudo-MAD. This entity is formed by the belly of the billowing posterior leaflet juxtaposed with the adjacent left atrial wall, creating the illusion of a disjunction.

• True-MAD is a real abnormal attachment of the leaflet to the atrial wall and is visible in both systole and diastole. It is rarely seen in patients with MVP (7% of cases).

• MAD is present in the majority of normal mitral valves and can be considered a standard variant. It is typically localized near the P1-P3 segments. However, significant differences in prevalence and locations have been found between anatomic and imaging studies.

• Considering MAD as the “culprit lesion” for severe arrhythmias and SCD is still debated. Many CMR studies have shown that the arrhythmogenic MVP often presents fibrosis, which is a well-known trigger of arrhythmias. Thus, although the direct role of MAD in arrhythmogenesis needs further investigation, we favour MAD as an innocent bystander rather than the “guilty” lesion.