Results
The study included 103 patients with MVP and a control group of 40 people and 34 of these MVP patients (33%) had MAD. Patients were divided into groups as MAD (+) (n=34), MAD (-) (n=69) and control (n=40). The clinical and demographic characteristics of the study groups are shown in Table-1. There was no statistically significant difference between the groups in terms of gender, age, BMI, beta-blocker use and smoking habits. When symptom types such as palpitation, dyspnea, fatigue, chest pain and syncope were compared, no significant difference was found between MAD (+) and MAD (-) groups. However, number of symptoms on admission was not similar. Number of patients with 2 or more symptoms was higher in the MAD (+) group than in the MAD (-) group (4.3% vs 44.1%, p<0.001). T negativity in the inferior leads on electrocardiography was more frequent in the MAD (+) group than in the MAD (-) patients (4.3% vs 20.6%, p=0.014).
When the groups were compared in terms of echocardiographic and MRI findings (table 2), left ventricular ejection fraction (LVEF), posterior wall thickness (PW), interventricular septum to posterior wall ratio (IVS/PW), left ventricular end-diastolic diameter (LVEDD) and end-systolic diameter (LVESD), apical longitudinal strain (ALS) levels were similar. Mitral regurgitation degree (p=0.001), frequency of Pickelhaube sign (17.6% vs. 1.4%, p=0.005) and LGE (35.3% vs. 10.6%, p=0.002) were higher in MAD (+) patients compared to MAD (-) patients. When the left atrial volume index, TDI lateral Sa, global longitudinal strain (GLS), basal longitudinal strain (BLS), mid-ventricular longitudinal strain(MVLS) levels were compared, the difference between the 3 groups was statistically significant. GLS of the MAD (+) group was -23.1±2.1 and this was -23.5±2.3 for the MAD (-), -24.5±2.2 for the control group (p<0.001). BLS of the MAD (+), MAD (-) and control groups were -19.6±1.5, -20.5±1.9 -23.3±1.2, respectively, with a P value <0.001. For the BLS of MAD (+), MAD (-) and control groups, these values were -22.2±1.7, -23.2±2.2 and -24.5±1.8; (p<0.001) and for MVLS, -22.2±1.7, -23.2±2.2 and -24.5±1.8 (P<0.001), respectively. The mean MAD distance was 7.8±3.2 mm in MAD (+) patients.
The relationship between GLS level and MAD distance in MAD (+) patients was evaluated by pearson correlation analysis (figure 1). It was observed that as the GLS level decreased, the MAD distance increased statistically significantly (r:0.694, p<0.001). Then, the relationship between BLS level and MAD distance in MAD (+) patients was evaluated by pearson correlation analysis (figure 2). It was observed that as the BLS level decreased, the MAD distance increased statistically significantly (r:0.715, p<0.001).
Discussion
The main finding of our study is the numeric decrease in longitudinal strain in MVP patients with MAD when compared to both MAD (-) MVP and control groups. This difference was detected for both global and segmental longitudinal strain. We interpreted this result as a reflection of partial loss of basal support during systole for the ventricular myocardium in significant MAD. The normal mitral annulus has roots located within the ventricular myocardium and these roots provide tight anchoring of the annulus to the ventricular myocardium (7). MAD occurs when these roots of the mitral annulus separate from the ventricular myocardium or fail to adhere sufficiently. However, the data about the functional implications of MAD and related ventricular dysfunction in the literature is scarce. First study in Medline about this subject was reported by Lee et al which investigated the 3D structure of MAD and its functional importance. In their study 42% of the 101 MVP patients were having MAD and mean dislodgement distance was 8.9 cm. Both results were compatible with ours. However, they found that although MVP had a reductive effect on left ventricular global longitudinal strain (LV-GLS) when compared to the control group with normal mitral valve, MAD positivity didn’t have any additional effect on LV-GLS. This result was opposing our main finding. One of the main differences between our and their study population was age. Mean age of our and their MAD+ MVP groups were 35.7±14.2 and 56±13 respectively. A systematic review defining the features of MAD has revealed that average age of these patients was 62 years in the literature. Therefore, our study population is younger, mostly composed of Barlow’s disease, suggesting that functional effects of MAD may show different characteristics according to the age period in which it is detected, and this may be due to etiological differences (8).
A case-control study by Wang et al examined the strain parameters of 21 MAD+ and 21MAD- MVP patients. MAD was causing a significant decrease in basal longitudinal strain by TTE, basal inferolateral (BIL), basal circumferential and basal radial strain by cardiac magnetic resonance imaging (CMR). Moreover, they concluded that global circumferential strain by CMR was independently associated with the diagnosis of MAD. They claimed the difference between TTE and CMR as the TTE strain analysis couldn’t catch the significant difference in strain parameters of MAD+ and MAD- MVP patients however CMR could. These results may be attributable to some confounding factors as the relatively small sample size and study design. In our study, all the strain differences were found by TTE evaluation (5). More recently, Sanoglioni et al analyzed 93 MVP patients, 34.4% of which were MAD+ and showed significantly impaired LV-GLS (-17.2 ± 1.4 vs -19.4 ± 3.0%, p < 0.001) and LV-GCS (-16.3 ± 4.1 vs -20.4 ± 4.9, p < 0.001) in MAD+ MVP patients compared to MAD- MVP cases, despite similar LV ejection fraction (9).
In our study 103 MVP patients and 40 control subjects were included and MAD was detected in 34(33%) of the MVP patients, compatible with most other studies in the literature. MAD is detected in 20 to 58 percent of the patients diagnosed with MVP and the prevalence is even higher in myxomatous MVP (1). Nonetheless, 21% of MAD patients don’t have MVP (4). In our study, all MAD patients were having MVP.
Our results provided that longitudinal strain is significantly affected in MAD+ patients when compared to MAD- MVP patients. Longitudinal strain and especially the GLS provides information about the functioning of longitudinal fibres and LV remodelling. It is known to be a predictor of cardiovascular outcome (10). Therefore our results are important in terms of detecting early myocardial dysfunction related to MAD especially in the younger patients. Moreover, in our study MVP patients with MAD had more severe mitral insufficiency when compared to the MAD-group. This result may be explained by the changes observed in annulus dynamics in patients with MAD. The mitral annulus enlarge more at the end of the systole (paradoxical systolic dilatation) and flattening of the annulus rather than physiologic saddling (paradoxical annular unsaddling) is a feature of MAD. It has been shown that MAD is associated with deformity of mitral leaflets and chordae tendienae. Furthermore, extend of disjunction and degree of mitral regurgitation are shown to be correlated (8). Likewise, Dumont et al have proved that annular dilatation due to MAD+ Barlow’s mitral valve disease causes more severe mitral regurgitation (11).
All the above mentioned features of MAD positivity in patients with MAD also may explain the increased number of symptoms in this group, another finding of our study, when compared to MAD- group, owing to more severe functional deformation.
Malignant MVP is characterized with increased sudden cardiac death risk, bileaflet prolapse, ECG repolarization abnormalities, papillary muscle and inferobasal wall fibrosis and its main echo finding is MAD. This entity also includes another term arrhythmic MVP (AMVP) which usually presents with frequent premature ventricular contractions and T-wave inversion in the inferolateral leads (12). Although we didn’t define a subgroup as malignant MVP of AMVP, our results were compatible with the literature such that T wave inversion in the inferolateral leads was detected in 20.6% of MAD+ group which was significantly more common than the MAD (-) group (4.3%) (P=0.014). LV fibrosis especially at the papillary muscles and inferobasal wall of the left ventricle can be detected as late gadolinium enhancement(LGE) in patients with MAD and previous reports showed that LGE can be seen as high as 40-47% of these patients (5,13). Likewise, in our study 35.3% of MAD+ MVP patients had LGE in CMR whereas this percentage was only 10% for MAD- MVP patients. Actually these areas of LGE in MAD+ MVP patients were found to be related with unipolar low voltage areas which are accepted as a potential substrate for ventricular arrhythmias(VA) and sudden cardiac death(SCD) (14). This relationship between LGE and VA and SCD in patients with MVP was also supported by other studies (3,15).
Another important finding of our study was the higher frequency of Pickelhaube sign in MAD+ MVP patients. This sign is a high velocity mid-systolic spike gathered from the lateral mitral annulus by tissue Doppler imaging, due to the stretching effect of prolapsing leaflet on posteromedial papillary muscle. As LGE, this sign was found to be a noninvasive imaging marker associated with SCD in MVP in previous studies (16).
We detected this sign in 17.6% of MAD+ MVP patients whereas only %1.4 of the MAD- MVP group had this sign on TTE.
In our study, mean disjunction distance was 7.8±3.2 mm There is no limit to define the movement of annulus as disjunction. In fact, detachment of the mitral annulus from myocardium may range between a few millimeters to more than 1 centimeter in the literature (17). Increased distance also increases the risk of SCD and it is reported that disjunction > 8.5 mm was predictive of VA (18).