Int J Cardiovasc Imaging DOI 10.1007/s10554-014-0547-0
ORIGINAL PAPER
Normal mitral annulus dynamics and its relationships with left ventricular and left atrial function Sorina Mihaila • Denisa Muraru • Marcelo Haertel Miglioranza Eleonora Piasentini • Diletta Peluso • Umberto Cucchini • Sabino Iliceto • Dragos Vinereanu • Luigi P. Badano
•
Received: 18 August 2014 / Accepted: 3 October 2014 Ó Springer Science+Business Media Dordrecht 2014
Abstract Mitral annulus (MA) geometry and dynamics are crucial for preserving normal mitral valve (MV) function. Static reference values for MA parameters have been reported, but the normal MA dynamics during the entire cardiac cycle remains controversial. MV full-volume datasets were obtained by three-dimensional transthoracic echocardiography from 50 healthy volunteers (18–74 years; 31 men) to assess MA changes in size and shape during entire cardiac cycle. Using simultaneous multiplanar review, projected MA area (MAA) and circumference (MAC), antero-posterior (AP) and anterolateral-posteromedial (ALPM) diameters, and sphericity index (SphI) were obtained at: mitral valve closure (MVC), mid- and end-systole (ES), early- (EDF) and late-diastolic filling, and end-diastole. MAA and AP diameter were the most ‘‘active’’ parameters, changing in all reference frames (p \ 0.001). MAA and AP diameter started to contract before MVC (during the left atrial contraction), reaching their minimum at MVC. Maximum MAA occurred at ES,
S. Mihaila (&) D. Muraru M. H. Miglioranza E. Piasentini D. Peluso U. Cucchini S. Iliceto L. P. Badano Department of Cardiac, Thoracic and Vascular Sciences, University of Padua, Via Giustiniani 2, 35128 Padua, Italy e-mail: [email protected] S. Mihaila D. Vinereanu University of Medicine and Pharmacy ‘‘Carol Davila’’, Bucharest, Romania M. H. Miglioranza Cardiology Institute of Rio Grande Do Sul, Porto Alegre, Brazil
while maximum AP diameter and SphI occurred at EDF. MAA fractional shortening was 35 ± 10 %. AP diameter change was 25 ± 10 %. MAC, ALPM and SphI showed similar patterns during left ventricular (LV) systole, and remained unchanged during diastole. Fractional change was 35 ± 10 % for MAC, and 13 ± 8 % for ALPM diameter. Our study provides the normal dynamics of the MA during the entire cardiac cycle. It reveals ‘‘pre-systolic’’ contraction of the MA, related to left atrial (LA) contraction, and minimal MAA during early LV systole. Therefore, the normal MA dynamics relates to a ‘‘physiologic LA-LV coupling’’, and a complete MA contraction requires both and properly timed LA and LV systole. Keywords Three-dimensional echocardiography Mitral annulus Mitral valve Normal dynamics Abbreviations 2D Two-dimensional 3D Three-dimensional ALPM Anterolateral-posteromedial AP Anterior-posterior ED End-diastole EDF Early diastolic filling ES End-systole LA Left atrium LDF Late diastolic filling LV Left ventricle MA Mitral annulus MR Mitral regurgitation MS Mid-systole MV Mitral valve MVC Mitral valve closure SphI Sphericity index TTE Transthoracic echocardiography
123
Int J Cardiovasc Imaging
Introduction Mitral regurgitation (MR) is highly prevalent [1] and, usually, it is associated with mitral annulus (MA) dysfunction [2–5]. Accurate characterization of the geometry and function of the MA plays an important role for better understanding the pathophysiology of the regurgitant mitral valve (MV) and for designing annular prosthesis [6, 7]. Although the MA dynamics have been characterized in pathological MV [2, 3, 5], the normal human MA dynamics during the entire cardiac cycle remains to be clarified. The translation function of the MA and the dynamic of its ‘‘saddle shape’’ have been thoroughly analyzed in normal individuals [8, 9]. Conversely, the changes in MA diameters and area are still a matter of debate, and the minimal MA size has been reported to occur at either early [3, 10], mid [4] or late systole [2]. Consequently, it is still controversial if the MA enlarges or decreases in size during the LV systole [2, 10, 11]. Moreover, MA area reduction has been mainly related to LV systolic function [12]. However, animal studies suggested that the area reduction of the normal MA starts before the onset of LV systole [13], suggesting a left atrial (LA) influence on it. In addition, experimental studies revealed that LA contraction contributes significantly to the effective MV closure (MVC) [14]. In those studies, the ‘‘pre-systolic’’ area reduction of the MA was absent in the animals in which the LA systole was abolished [13] and diminished during rapid LA pacing [15]. MA ‘‘pre-systolic’’ area reduction has been suggested in humans, too [11], using two-dimensional echocardiography (2DE). Even though the current guidelines recommend 2DE as the first imaging technique before MV surgery and for follow-up of the patients [16], 2DE has important limitations for MA assessment. It depends on correct alignment of anatomical landmarks [17] and underestimates the true dimensions of the MA [18]. Conversely, three-dimensional echocardiography (3DE) has the ability to provide anatomically sound images of the MV apparatus and quantitative analysis of the geometry and dynamics of the MA without geometrical assumptions. 3DE provided more accurate dimensions than 2DE for both normal and pathological MA, similar to those obtained by cardiac magnetic resonance [19]. However, even though the dynamics of the MA has been assessed by 3DE in different MV pathologies [3, 5, 20] or on very small number of normal subjects [10] using dedicated software that tracks the MA during the LV systole only, a thorough analysis of normal MA dynamics regarding its changes in diameters, area and shape during the entire cardiac cycle has not been reported yet. Accordingly, our study aim was to perform a quantitative assessment of the dynamics of the MA in healthy
123
volunteers during the entire cardiac cycle and to evaluate its relation to LA and LV systole.
Methods Study population 50 healthy Caucasian volunteers were prospectively recruited between January 2012 and September 2013 in a single tertiary center among hospital employees, fellows in training, their relatives and people who underwent medical visits for driving or working license. The inclusion criteria were: age [17 years, no history or symptoms of cardiovascular or lung disease, no cardiovascular risk factors (i.e. systemic arterial hypertension, smoking, diabetes, and hypercholesterolemia), normal electrocardiogram and physical examination, no cardio- or vasoactive treatment. Exclusion criteria were: trained athletes, pregnancy, body mass index [30 kg/m2, and poor apical acoustic window. Blood pressure, height and weight were measured in all subjects immediately before the echocardiographic examination. Body surface area was calculated according to the formulas by DuBois and DuBois [21]. An ECG was performed to confirm sinus rhythm and to exclude electrical abnormalities. University of Padua Ethics Committee approved the study (Protocol 2380 P) and all volunteers provided informed consent before entering the study. Echocardiography Using a standardized acquisition protocol, all examinations were performed using commercially available Vivid E9 system (GE Vingmed Ultrasound, Norway) equipped with M5S and 4 V probes. First, study subjects underwent standard transthoracic echocardiography to exclude subclinical heart diseases. Then, in order to obtain high temporal resolution of the datasets, separate 3D full-volume multi-beat acquisitions of the MV, the LA and the LV were obtained, by combining six consecutive ECG-triggered sub-volumes, during breath-hold and avoidance of patient or probe movement. Care was taken to encompass the MV in the full-volume throughout the cardiac cycle, and the entire LA and LV cavities in the dataset. By using multislice display, the absence of stitching artifacts was carefully checked. Image analysis 3DE LA full-volume datasets were converted to DICOM format and analyzed using dedicated software designed for volumetric analysis of the LA and recently validated
Int J Cardiovasc Imaging Fig. 1 Multiplanar review of the mitral annulus at mitral valve closure (MVC, upper panels) and early diastolic filling (EDF, lower panels). Red and green arrows point the frames of the cardiac cycle used for analysis. Flexi slice tool displays three 2D slices of the dataset, and one 3D volumerendered image (a, b) at each time frame. The transversal 2D slice (green border) shows a cut plane taken at the level of the rendered image (green dotted line on longitudinal planes). Cut-planes are manipulated to obtain a section at the level of the mitral annulus (MA). The aortic valve (Ao) and the tricuspid annulus (TA) were identified in both volumerendered image and derived 2D slices as anatomical landmarks
against cardiac magnetic resonance (LA analysis 2.3, TomTec Imaging Systems, Unterschleissheim, D) [17]. As previously described [17], the maximum (LAVmax), minimum (LAVmin) and pre-atrial contraction (LAVpreA) volumes of the LA were measured. 3DE datasets of the LV and MV were stored digitally in raw-data format for offline analysis using commercially available software package (EchoPAC BT 12, GE Vingmed Ultrasound, Norway). Quantification of 3D LV volumes and ejection fraction was performed using 4D Auto LVQ option in EchoPAC, as previously described [22, 23]. The quality of MV datasets was judged subjectively, considering the signal-to-noise ratio, the degree of bloodtissue contrast and the visualization of the MA. They were
classified as excellent, fair or poor, and only dataset with excellent visual quality and tracking of the MA were included in the study. Using frame-by-frame motion, 3D datasets of the MV were analyzed at six reference frames during the cardiac cycle: the mitral valve closure (MVC); mid-systole (MS); end-systole (ES), as the first frame after the aortic valve closure; early diastolic filling (EDF), as the frame with the MV wide opened during early LV filling; late diastolic filling (LDF), as the frame with the MV wide opened during late LV filling (after the P wave); and end-diastole (ED), as the frame just before the onset of the R wave. Using the ‘‘flexi-slice’’ option, the MV dataset was displayed in a multi-planar review mode. The MV full-
123
Int J Cardiovasc Imaging
between their maximum and minimum value, divided by the maximum values and expressed in percentages. Statistical analysis
Fig. 2 Layout with the correspondent 2D slice of the rendered 3D image of the mitral annulus. View from the atrium, during early diastolic filling. On this layout were performed the measures for the MA area and circumference, the AP and ALPM diameters, and sphericity index (left panel). ALPM anterolateral-posteromedial, AP anterior-posterior, Ao aorta, MA mitral annulus, TA tricuspid annulus
volume was automatically sectioned in two longitudinal and one transverse cut planes, displaying one 3D volumerendered image and three 2D slices of the dataset (Fig. 1). The 2D slice with the solid borders displayed the same view as the 3D volume-rendered image. The user had the flexibility to adjust both 2D slices by manipulating the position of the cut-planes and their intersection in any direction, to obtain a view of interest in both the 3D volume-rendered image and the correspondent 2D slice. The transverse cut-plane was set at the level of the MA and rotated by 180°, to visualize the MA from the LA side. Then, the correspondent 2D slice was rotated to display the MA with the aorta at 12 o’clock [24] (Fig. 1). Finally, the longitudinal cut-planes were adjusted to section the MA in its longest and shortest diameters. The resulting longitudinal 2D slices of the data set provided the anatomical landmarks for correct assessment of the MA (MV leaflets, LV and LA basal segments, aortic valve hinge). The 2D and 3D gain settings were manually adjusted to obtain the best visual delineation of the MA borders (Fig. 2) and remained unchanged throughout the entire cardiac cycle, to avoid consequent variability of the MA measurements. On the correspondent 2D slice, the following parameters of the MA were measured at each reference frame: projected MA area (MAA); projected MA circumference (MAC); antero-posterior (AP) diameter, as the shortest size of the MA; anterolateral-posteromedial (ALPM) diameter, as the longest diameter of the MA; and sphericity index (SphI) as the ratio between AP and ALPM diameter (Figs. 2, 3). Fractional shortening of the MAA and MAC, and of AP and ALPM diameters were obtained by the difference
123
Normal distribution of variables was verified by Kolmogorov–Smirnov test. Continuous variables were summarized as mean ± SD. MA measurements obtained at different reference frames during the cardiac cycle were compared using ANOVA for repeated measurements. Pearson’s correlation was used to analyze the relationships between different MA parameters and LV volumes, and between the fractional changes of MA parameters. Interobserver variability for MA assessment using 3DE was analyzed in 15 random subjects by two independent observers (M.M. and S.M.). S.M., who repeated the measurements 7 days later, assessed the intra-observer variability. Reproducibility was reported as intra-class correlation coefficient (ICC). All analyses were carried out using SPSS version 20.0 (SPSS, Inc., Chicago, IL). Differences among variables were considered significant at p \ 0.05.
Results We recruited 50 healthy volunteers between 18 and 74 years (31 men). Subjects’ demographics, physical and 3DE findings are summarized in Table 1. Temporal resolution of the 3D datasets was 34 ± 1 volumes/s. The average analysis time for each dataset was 30 ± 4 min. Mean values for MA parameters at the six selected reference frames of the cardiac cycle are summarized in Table 2, and schematic curves for MA parameters dynamic throughout the cardiac cycle in Fig. 4. MAA and AP diameter were the most ‘‘active’’ parameters, showing significant changes in all the reference frames during the cardiac cycle (all p \ 0.001). Minimal MAA, AP diameter and SphI occurred at the MVC and increased afterwards, during the LV systole. Maximum MAA was reached at ES, and maximum AP diameter and SphI during EDF. MAA started to decrease abruptly before the R wave, mainly in its AP diameter, reaching its minimum at the MVC (Fig. 2; 4). Of the total MA area reduction calculated from the maximal ES value to the minimal MVC value, 61 ± 22 % occurred before the R wave, and 43 ± 37 % after the R wave (i.e. between R wave and MVC). MAC and ALPM diameter showed significant changes from the onset of the R wave until EDF, with a pattern similar to the one documented for MAA and AP diameter. Conversely, both MAC and ALPM diameter remained rather constant during diastole (between EDF and LDF).
Int J Cardiovasc Imaging
Fig. 3 The analysis of the mitral annulus shape and dimensions at six reference frames of the cardiac cycle. View from the atrial side. a 3D rendered images of the mitral annulus; b 2D correspondent layouts,
Table 1 Demographic, physical and three-dimensional echocardiography findings of the healthy volunteers Clinical features
N = 50
Males no (%)
31 (62 %)
Age (years)
46 ± 14
Weight (kg)
70 ± 10
Height (cm)
172 ± 8
Body mass Index (kg/m2)
23.2 ± 2.6
Body surface area (m2)
1.8 ± 0.2
Systolic blood pressure (mmHg)
121 ± 10
Diastolic blood pressure (mmHg)
73 ± 8
Heart rate (bpm)
68 ± 8
Echocardiographic features LV end-diastolic volume (ml)
114 ± 21
Indexed LV end-diastolic volume (ml/m2)
62 ± 9
LV end-systolic volume (ml)
43 ± 10
Indexed LV end-systolic volume (ml/m2)
23 ± 4
LV ejection fraction (%)
62 ± 4
LV sphericity index
0.36 ± 0.07
for performing measurements. Ao aorta, ED end-diastole, EDF early diastolic filling, ES end-systole, LDF late diastolic filling, MA mitral annulus, MS mid-systole, MVC mitral valve closure
ALPM diameter shortening (Fig. 5). There was no relationship between the fractional shortening of the MA parameters and ageing of the normal individuals. LV end-diastolic and end-systolic volumes correlated with both MAA and MAC measured at every reference frame of the cardiac cycle (r [ 0.5, p \ 0.001). As expected, the strongest correlations were found between the LV end-diastolic volume and MAA and MAC measured at ED, and between LV end-systolic volume and MAA and MAC measured at ES (Fig. 6). Conversely, there was no relationship between MAA fractional change and LV ejection fraction. LAVmin correlated with the MAA and MAC measured at all reference frames of the cardiac cylce (r [ 0.6, all p \ 0.001). The strongest correlations were found between the LAVmin and MAA measured at LDF and ED (after the LA contraction), and between the LAVmin and MAC measured at LDF and MVC (after the LA contraction, and at early LV systole respectively; Fig. 7). The LAVmax did not correlate with either MAA or MAC. Reproducibility
MAA and AP diameter fractional changes throughout the cardiac cycle were 35 ± 10 %, and 25 ± 10 % respectively. The mean fractional change of MAC was 35 ± 10 %, whereas the one of ALPM diameter was 13 ± 8 %, only. The fractional changes of the projected MAA and MAC were related more to the AP than to
MA parameters showed good intraobserver and interobserver reproducibility in all analyzed reference frames (Table 3). The best intraobserver reproducibility for all MA parameters was observed at MVC and MS, while for the interobserver variability only at MS frame.
123
Int J Cardiovasc Imaging Table 2 Mitral annulus geometry parameters during phases of the cardiac cycle Time-points during cardiac cycle Mitral valve closure
Circumference (cm)
Area (cm2)
AP diameter (cm)
ALPM diameter (cm)
SphI
9.6 ± 1*
6.0 ± 1.3*
2.1 ± 0.3*
3.4 ± 0.3*
0.60 ± 0.08*
Mid-systole
10.9 ± 1.2*
8.1 ± 1.8*
2.5 ± 0.3*
3.8 ± 0.4*
0.64 ± 0.07*
End-systole
11.6 ± 1.1*
9.3 ± 1.9*
2.7 ± 0.3*
3.9 ± 0.4*
0.70 ± 0.08*
Early diastolic filling
10.9 ± 1.2
8.7 ± 1.8*
2.8 ± 0.3*
3.7 ± 0.5
0.76 ± 0.08
Late diastolic filling
10.6 ± 1.1
8.2 ± 1.7*
2.7 ± 0.7*
3.7 ± 0.5
0.72 ± 0.08
7.1 ± 1.7*
2.4 ± 0.3*
3.6 ± 0.5*
0.67 ± 0.07*
End-diastole
9.9 ± 1.4*
Data are summarized as mean ± SD AP anterior-posterior, ALPM anterior-lateral posterior-medial, SphI sphericity index * Statistical significance with p \ 0.005 for all marked cases, by Anova repeated measures, followed by paired T test
Fig. 4 The magnitude of changes of mitral annulus parameters during the cardiac cycle. Circumference (cm); area (cm2); anterior-posterior (AP) diameter (cm); anterolateralposteromedial (ALPM) diameter; sphericity index (SphI)
123
Int J Cardiovasc Imaging
Fig. 5 Relationships between the contraction of the mitral annulus area (upper panels) and circumference (lower panels) with the fractional changing of mitral annulus diameters. Data are expressed in percentages (%). AP antero-posterior, ALPM anterolateral-posteromedial
Discussion The present study provides a quantitative analysis of the dynamic changes in size and shape of the MA during the entire cardiac cycle in healthy volunteers, using 3DE. The main findings of the study were: (1) MA area reduction started before the onset of LV systole (in parallel to LA contraction) and reached its maximum at the MVC; (2) MAA and AP size presented significant dynamic changes during all phases of the cardiac cycle, while MAC, ALPM size and MA sphericity remained unchanged during diastole; (3) MA reached its minimal size and sphericity at the MVC and increased afterwards during LV systole; (4) The largest MAA was measured at ES, while maximum MA sphericity occurred at EDF. The MA shape, size, and dynamics influence the leaflets’ geometry, which is crucial for complete coaptation of MV leaflets and avoidance of MR. Consequently, changes of the normal MA dynamics are closely related to the development of MR [2, 3, 5], and mitral annuloplasty is the most common
surgical procedure to repair a regurgitant MV [25, 26]. Therefore, understanding normal dynamics of the MA is particularly important not only to characterize changes in mitral disease states but also to guide further refinements in MV repair towards normal MA geometry and function. We used transthoracic 3DE and multiplanar review analysis, which uses multiple cut-planes of the MV apparatus and allows a better delineation of the MA, to analyze the dynamics of the MA throughout the entire cardiac cycle in healthy volunteers with a wide age range and excellent quality of the 3D MV datasets. The parameters for MA dimensions obtained by 3DE were reproducible in all six frames of the cardiac cycle. In addition, the assessment of the MA by 3DE was already proved more reliable than 2DE, with results similar to those obtained by CMR [18, 19]. Previous human studies have used 3DE to reveal MA dysfunction in pathological MV [5, 20, 27, 28] and in the follow-up after the repair of the MV [6, 7] or the surrounding structures [6, 29–31]. 3DE assessment of the MA dynamics also served to design more physiological MA
123
Int J Cardiovasc Imaging
Fig. 6 The relationship between mitral annulus area (left panels) and circumference (right panels) at end-diastole and end-systole and left ventricular end-diastolic and end-systolic volumes, respectively. ED
end-diastole, ES end-systole, LVED left ventricle end-diastolic, LVES left ventricle end-systolic
prosthesis [32]. However, the normal dynamics of the MA provided by these studies remained controversial [2–5], especially in terms of MA area reduction. MA ‘‘saddle shape’’ and translation function were analyzed both in normal and pathological settings, therefore they were not the aim of our study [5, 8, 9, 20, 33]. Conversely, the extent and moreover, the timing of the maximal MA area reduction in normal individuals was the focus of our study. From our data, overall reduction of MA area is similar to that reported in experimental animals studies by Davis and Kinmonth [34], the first investigators that implanted radiopaque markers around canine MA and showed a MA area change of 30 % from diastole to systole. Tsakiris et al. [35] also observed a 19–34 % area reduction in dogs, and reported that MA area reduction starts in the pre-systolic period. MA ‘‘contraction’’ function, as it was called by Silbiger et al. [36], is of clinical importance. It was found decreased
in patients with functional MR [3] in relation with MR severity [20], and also in patients with organic MR due to MV prolapse or Barlow disease [5]. Conversely, Little et al. reported normal MA ‘‘contraction’’ in patients with organic MR, by comparing the MA area change of patients with organic MR with the MA area change of only 26 ± 8 % found in 15 healthy controls throughout the entire cardiac cycle. Therefore, a standardization of the overall MA area reduction would be beneficial when comparing normal with pathological MA. Our study found maximum MA area reduction during the early LV systole, with a consequent minimum size of the MA occurring at the MVC, similar to those reported by Kwan [10] and Topilsky [3] in a limited number of healthy subjects. Conversely, our results were different from data found by Kaplan [4] or Little et al. [2] that reported a minimum MA area at MS and at ES, respectively. Similar to Kwan [10] findings, we showed that, after the minimal
123
Int J Cardiovasc Imaging
Fig. 7 The relationship between the mitral annulus area (left panels) and circumference (right panels) measured at either late diastolic filling, end-diastole, or mitral valve closure and minimal left atrium
volume, measured after the left atrial contraction. ED end-diastole, LAVmin the minimal left atrial volume, LDF late diastolic filling, MVC mitral valve closure
MA size occurring at early systole, the MA size increases further during the LV systole. Indeed, the 3DE studies mentioned above used dedicated MV software that can perform a thorough analysis of the MA geometry, including MA height and nonplanarity. However, they were able to perform a dynamic analysis of the MA during the cardiac systole, only. Therefore, to be able to analyze the dynamics of the MA during the entire cardiac cycle, we performed our study without dedicated software for the 3DE dataset. We found that the normal MA area reduction starts before the LV systole. Thus, we confirmed the data suggested by Glasson et al. [13] in animal studies or by Ormiston et al. [11] in humans, using 2DE. Timek et al. also showed in canine models that the MA area reduction occurs in parallel with the atrial contraction [15], which proved important for MV leaflet closure [14]. Our results strengthen the idea of an important role of the LA for the MA ‘‘contraction’’ function and for an effective MVC. The amount of ‘‘pre-systolic’’ area reduction of the
MA found by us in normal humans (61 ± 22 %), was relatively lower than 89 % reported by Glasson [13] in experimental ovine study. Both LA dysfunction and decreased MA ‘‘contraction’’ have been reported in patients with functional or organic MR [5, 20, 37, 38]. Considering that normal MA area reduction starts in parallel with the LA contraction, it would be of interest to further assess the relation between LA and MA dysfunction in MV disease with consequent MR, and to analyze if these events might have a prognostic impact in these clinical settings. MAA and AP diameter showed significant changes from each reference frame of the cardiac cycle to the other, during both LV systolic and diastolic period, as previously reported [4, 10]. These findings suggest an ‘‘active’’ MA during the entire cardiac cycle, contradicting the static MA behavior during LV diastole reported by Grewal et al. [5]. As expected, MAA change was closely related to MA circumference reduction, and we found that the AP
123
Int J Cardiovasc Imaging Table 3 Intra-observer and inter-observer reproducibility for the mitral annulus area, circumference and diameters measured with multi-planar review
MA parameter
MA area
MA circumference
AP diameter
AP antero-posterior, ALPM anterolateral-posteromedial, CI confidence interval, ED enddiastole, EDF end-diastolic filling, ES end-systole, ICC intra-class coefficient, LD late diastole, LDF late diastolic filling, MA mitral annulus, MS mid-systole, MVC mitral valve closure * Statistical significance, for all ICC p \ 0.01
ALPM diameter
Reference frame
Inter-observer
ICC*
ICC*
95 % CI interval
95 % CI interval
MVC
0.98
0.94–0.99
0.87
0.33–0.97
MS
0.97
0.89–0.99
0.95
0.75–0.99
ES
0.95
0.83–0.99
0.89
0.47–0.98
EDF
0.92
0.65–0.98
0.92
0.55–0.98
LDF
0.95
0.78–0.98
0.87
0.34–0.97
ED
0.95
0.81–0.99
0.88
0.44–0.98
MVC
0.98
0.94–0.99
0.87
0.33–0.97
MS
0.97
0.89–0.99
0.91
0.56–0.98
ES
0.96
0.83–0.99
0.82
0.25–0.96
EDF
0.92
0.65–0.98
0.84
0.20–0.97
LDF ED
0.94 0.95
0.79–0.98 0.78–0.98
0.81 0.70
0.21–0.95 0.45–0.98
MVC
0.83
0.31–0.96
0.71
0.56–0.98
MS
0.85
0.30–0.97
0.84
0.25–0.97
ES
0.75
0.40–0.94
0.72
0.32–0.81
EDF
0.82
0.60–0.86
0.80
0.43–0.98
LDF
0.85
0.30–0.97
0.86
0.29–0.97
ED
0.85
0.16–0.94
0.84
0.23–0.96
MVC
0.94
0.76–0.98
0.85
0.24–0.96
MS
0.97
0.90–0.99
0.93
0.66–0.98
ES
0.97
0.88–0.99
0.85
0.27–0.97
EDF
0.84
0.30–0.96
0.84
0.22–0.97
LDF
0.95
0.79–0.98
0.90
0.45–0.98
ED
0.92
0.69–0.98
0.88
0.45–0.97
diameter change contributed more than the ALPM diameter change to MA area reduction [4, 5]. The unequal changes of the MA diameters lead to a minimum MA sphericity at the MVC, which is probably due to MA folding, and a maximum sphericity during EDF, when the MV is wide opened. MA from our healthy volunteers presented maximum area at the ES, similar to data reported by Khabbaz [28] and Grewal et al. [5] in 15 ‘‘control individuals’’. The timing of the maximum MAA was however controversial among different studies, and has been also reported either at early [9] or late diastole [2]. Even though Ormiston [11] reported a maximum MAA before the P wave, using 2DE, the study also specify an increase of the MA size during ES and isovolumic relaxation time. We found a significant correlation between MA size and LV volumes. However, the fractional changes of the MA parameters showed no correlation with the LV ejection fraction. This might be due to the significant, presystolic contribution of the LA systole to the MA area reduction, with consequent shape changes during the pre-systolic phase of the cardiac cycle. Therefore, the minimal LA volume measured after the LA contraction correlated with
123
Intra-observer
the MA size measured at all reference frames, and especially at the frames occurring immediately after the LA contraction (LDF, ED and MVC). Conversely, the maximum LA volume reached at ES did not correlated with either MAA or MAC. Limitations of the study The limited number of reference frames may be considered a limitation of the study, as now there is a commercially available dedicated MV software package for 3DE that allows the tracking and analysis of the MA frame-byframe. However, this software package analysis capability is limited only to the systolic period, and the purpose of our study was to analyze the changes of the MA during the entire cardiac cycle, and to identify its most important events. We provided data about the shape, size and fractional changes of MA diameters and projected area and circumference during the entire cardiac cycle, but not about the MA ‘‘saddle shape’’ and translation function. However, the latter have been previously described in normal subjects in other studies [8, 9].
Int J Cardiovasc Imaging
‘‘End-systole (ES)’’ was variably defined in the different studies, either before the aortic valve closure [3], the frame before MV starts to open [19], or in between [10], similar to our study. These differences in the choice of the ‘‘reference frames’’ may be partially responsible for the inconsistent MA dynamics reported in the literature, as the MA can present significant changes in size at just a few frames distance.
Conclusions Our study provides the normal dynamics of the MA during the entire cardiac cycle and documents the ‘‘pre-systolic’’ area reduction of the MA, which is related to the atrial systole, and maximum MA area reduction during early LV systole. We emphasize the importance of a physiologic coupling of the LA-LV systole for a normal ‘‘contraction’’ function of the MA and an efficient MVC. The normal dynamic of the MA provided by our study can be used as a reference for 3DE studies of the dynamics of annular prosthesis or of native MA in patients with pathological MV and after the interventions performed on the surrounding structures. Acknowledgments Dr. Mihaila is an EACVI Research Grant winner on 2011 and an ESC Training Grant winner on 2014. Dr Mihaila is also a winner of a Grant for PhD research of the University of Medicine and Pharmacy ‘‘Carol Davila’’ Bucharest, POSDRU/159/ 1.5/S/141531. Conflict of interest Doctors Badano and Muraru are consultants and received equipment grants from GE Healthcare and TomTec Imaging Systems.
References 1. Iung B, Baron G, Butchart EG, Delahaye F, Gohlke-Barwolf C, Levang OW et al (2003) A prospective survey of patients with valvular heart disease in Europe: the euro heart survey on valvular heart disease. Eur Heart J 24(13):1231–1243 2. Little SH, Ben Zekry S, Lawrie GM, Zoghbi WA (2010) Dynamic annular geometry and function in patients with mitral regurgitation: insight from three-dimensional annular tracking. J Am Soc Echocardiogr 23(8):872–879 3. Topilsky Y, Vaturi O, Watanabe N, Bichara V, Nkomo VT, Michelena H et al (2013) Real-time 3-dimensional dynamics of functional mitral regurgitation: a prospective quantitative and mechanistic study. J Am Heart Assoc 2(3):e000039 4. Kaplan SR, Bashein G, Sheehan FH, Legget ME, Munt B, Li XN et al (2000) Three-dimensional echocardiographic assessment of annular shape changes in the normal and regurgitant mitral valve. Am Heart J 139(3):378–387 5. Grewal J, Suri R, Mankad S, Tanaka A, Mahoney DW, Schaff HV et al (2010) Mitral annular dynamics in myxomatous valve disease: new insights with real-time 3-dimensional echocardiography. Circulation 121(12):1423–1431
6. Caiani EG, Fusini L, Veronesi F, Tamborini G, Maffessanti F, Gripari P et al (2011) Quantification of mitral annulus dynamic morphology in patients with mitral valve prolapse undergoing repair and annuloplasty during a 6-month follow-up. Eur J Echocardiogr 12(5):375–383 7. Mahmood F, Subramaniam B, Gorman JH III, Levine RM, Gorman RC, Maslow A et al (2009) Three-dimensional echocardiographic assessment of changes in mitral valve geometry after valve repair. Ann Thorac Surg 88(6):1838–1844 8. Ryan LP, Jackson BM, Enomoto Y, Parish L, Plappert TJ, St John-Sutton MG et al (2007) Description of regional mitral annular nonplanarity in healthy human subjects: a novel methodology. J Thorac Cardiovasc Surg 134(3):644–648 9. Veronesi F, Corsi C, Sugeng L, Caiani EG, Weinert L, Mor-Avi V et al (2008) Quantification of mitral apparatus dynamics in functional and ischemic mitral regurgitation using real-time 3-dimensional echocardiography. J Am Soc Echocardiogr 21(4):347–354 10. Kwan J, Jeon MJ, Kim DH, Park KS, Lee WH (2009) Does the mitral annulus shrink or enlarge during systole? A real-time 3D echocardiography study. J Korean Med Sci 24(2):203–208 11. Ormiston JA, Shah PM, Tei C, Wong M (1981) Size and motion of the mitral valve annulus in man. I. A two-dimensional echocardiographic method and findings in normal subjects. Circulation 64(1):113–120 12. Kwan J, Qin JX, Popovic ZB, Agler DA, Thomas JD, Shiota T (2004) Geometric changes of mitral annulus assessed by real-time 3-dimensional echocardiography: becoming enlarged and less nonplanar in the anteroposterior direction during systole in proportion to global left ventricular systolic function. J Am Soc Echocardiogr 17(11):1179–1184 13. Glasson JR, Komeda M, Daughters GT, Foppiano LE, Bolger AF, Tye TL et al (1997) Most ovine mitral annular three-dimensional size reduction occurs before ventricular systole and is abolished with ventricular pacing. Circulation 96(9 Suppl):II-115–II-22 discussion II-23 14. Timek T, Dagum P, Lai DT, Green GR, Glasson JR, Daughters GT et al (2001) The role of atrial contraction in mitral valve closure. J Heart Valve Dis 10(3):312–319 15. Timek TA, Lai DT, Dagum P, Green GR, Glasson JR, Daughters GT et al (2000) Mitral annular dynamics during rapid atrial pacing. Surgery 128(2):361–367 16. Lancellotti P, Tribouilloy C, Hagendorff A, Popescu BA, Edvardsen T, Pierard LA et al (2013) Recommendations for the echocardiographic assessment of native valvular regurgitation: an executive summary from the European association of cardiovascular imaging. Eur Heart J Cardiovasc Imaging 14(7):611–644 17. Foster GP, Dunn AK, Abraham S, Ahmadi N, Sarraf G (2009) Accurate measurement of mitral annular dimensions by echocardiography: importance of correctly aligned imaging planes and anatomic landmarks. J Am Soc Echocardiogr 22(5):458–463 18. Anwar AM, Soliman OI, ten Cate FJ, Nemes A, McGhie JS, Krenning BJ et al (2007) True mitral annulus diameter is underestimated by two-dimensional echocardiography as evidenced by real-time three-dimensional echocardiography and magnetic resonance imaging. Int J Cardiovasc Imaging 23(5):541–547 19. Anwar AM, Soliman OI, Nemes A, Germans T, Krenning BJ, Geleijnse ML et al (2007) Assessment of mitral annulus size and function by real-time 3-dimensional echocardiography in cardiomyopathy: comparison with magnetic resonance imaging. J Am Soc Echocardiogr 20(8):941–948 20. Daimon M, Saracino G, Fukuda S, Koyama Y, Kwan J, Song JM et al (2010) Dynamic change of mitral annular geometry and motion in ischemic mitral regurgitation assessed by a computerized 3D echo method. Echocardiography 27(9):1069–1077
123
Int J Cardiovasc Imaging 21. Dubois D, Dubois EF (1916) Clinical calorimetry. A formula to estimate the approximate surface area if height and weight be known. Arch Intern Med 17:863–871 22. Muraru D, Badano LP, Piccoli G, Gianfagna P, Del Mestre L, Ermacora D et al (2010) Validation of a novel automated borderdetection algorithm for rapid and accurate quantitation of left ventricular volumes based on three-dimensional echocardiography. Eur J Echocardiogr 11(4):359–368 23. Muraru D, Badano LP, Peluso D, Dal Bianco L, Casablanca S, Kocabay G et al (2013) Comprehensive analysis of left ventricular geometry and function by three-dimensional echocardiography in healthy adults. J Am Soc Echocardiogr 26(6):618–628 24. Lang RM, Badano LP, Tsang W, Adams DH, Agricola E, Buck T et al (2012) EAE/ASE recommendations for image acquisition and display using three-dimensional echocardiography. J Am Soc Echocardiogr 25(1):3–46 25. Ling LH, Enriquez-Sarano M, Seward JB, Orszulak TA, Schaff HV, Bailey KR et al (1997) Early surgery in patients with mitral regurgitation due to flail leaflets: a long-term outcome study. Circulation 96(6):1819–1825 26. Enriquez-Sarano M, Avierinos JF, Messika-Zeitoun D, Detaint D, Capps M, Nkomo V et al (2005) Quantitative determinants of the outcome of asymptomatic mitral regurgitation. N Engl J Med 352(9):875–883 27. De Simone R, Wolf I, Mottl-Link S, Hoda R, Mikhail B, Sack FU et al (2006) A clinical study of annular geometry and dynamics in patients with ischemic mitral regurgitation: new insights into asymmetrical ring annuloplasty. Eur J Cardiothorac Surg 29(3):355–361 28. Khabbaz KR, Mahmood F, Shakil O, Warraich HJ, Gorman JH III, Gorman RC et al (2013) Dynamic 3-dimensional echocardiographic assessment of mitral annular geometry in patients with functional mitral regurgitation. Ann Thorac Surg 95(1):105–110 29. Warraich HJ, Matyal R, Bergman R, Hess PE, Khabbaz K, Manning WJ et al (2013) Impact of aortic valve replacement for
123
30.
31.
32.
33.
34. 35.
36. 37.
38.
aortic stenosis on dynamic mitral annular motion and geometry. Am J Cardiol 112(9):1445–1449 Tsang W, Meineri M, Hahn RT, Veronesi F, Shah AP, Osten M et al (2013) A three-dimensional echocardiographic study on aortic-mitral coupling in transcatheter aortic valve replacement. Eur Heart J Cardiovasc Imaging 14(10):950–956 Tsang W, Veronesi F, Sugeng L, Weinert L, Takeuchi M, Jeevanandam V et al (2013) Mitral valve dynamics in severe aortic stenosis before and after aortic valve replacement. J Am Soc Echocardiogr 26(6):606–614 Mahmood F, Gorman JH III, Subramaniam B, Gorman RC, Panzica PJ, Hagberg RC et al (2010) Changes in mitral valve annular geometry after repair: saddle-shaped versus flat annuloplasty rings. Ann Thorac Surg 90(4):1212–1220 Jensen MO, Hagege AA, Otsuji Y, Levine RA, Leducq Transatlantic MN (2013) The unsaddled annulus: biomechanical culprit in mitral valve prolapse? Circulation 127(7):766–768 Davis PK, Kinmonth JB (1963) The movements of the annulus of the mitral valve. J Cardiovasc Surg (Torino) 4:427–431 Tsakiris AG, Von Bernuth G, Rastelli GC, Bourgeois MJ, Titus JL, Wood EH (1971) Size and motion of the mitral valve annulus in anesthetized intact dogs. J Appl Physiol 30(5):611–618 Silbiger JJ (2012) Anatomy, mechanics, and pathophysiology of the mitral annulus. Am Heart J 164(2):163–176 Gertz ZM, Raina A, Saghy L, Zado ES, Callans DJ, Marchlinski FE et al (2011) Evidence of atrial functional mitral regurgitation due to atrial fibrillation: reversal with arrhythmia control. J Am Coll Cardiol 58(14):1474–1481 Cameli M, Lisi M, Righini FM, Massoni A, Natali BM, Focardi M et al (2013) Usefulness of atrial deformation analysis to predict left atrial fibrosis and endocardial thickness in patients undergoing mitral valve operations for severe mitral regurgitation secondary to mitral valve prolapse. Am J Cardiol 111(4):595–601
ORIGINAL PAPER
Normal mitral annulus dynamics and its relationships with left ventricular and left atrial function Sorina Mihaila • Denisa Muraru • Marcelo Haertel Miglioranza Eleonora Piasentini • Diletta Peluso • Umberto Cucchini • Sabino Iliceto • Dragos Vinereanu • Luigi P. Badano
•
Received: 18 August 2014 / Accepted: 3 October 2014 Ó Springer Science+Business Media Dordrecht 2014
Abstract Mitral annulus (MA) geometry and dynamics are crucial for preserving normal mitral valve (MV) function. Static reference values for MA parameters have been reported, but the normal MA dynamics during the entire cardiac cycle remains controversial. MV full-volume datasets were obtained by three-dimensional transthoracic echocardiography from 50 healthy volunteers (18–74 years; 31 men) to assess MA changes in size and shape during entire cardiac cycle. Using simultaneous multiplanar review, projected MA area (MAA) and circumference (MAC), antero-posterior (AP) and anterolateral-posteromedial (ALPM) diameters, and sphericity index (SphI) were obtained at: mitral valve closure (MVC), mid- and end-systole (ES), early- (EDF) and late-diastolic filling, and end-diastole. MAA and AP diameter were the most ‘‘active’’ parameters, changing in all reference frames (p \ 0.001). MAA and AP diameter started to contract before MVC (during the left atrial contraction), reaching their minimum at MVC. Maximum MAA occurred at ES,
S. Mihaila (&) D. Muraru M. H. Miglioranza E. Piasentini D. Peluso U. Cucchini S. Iliceto L. P. Badano Department of Cardiac, Thoracic and Vascular Sciences, University of Padua, Via Giustiniani 2, 35128 Padua, Italy e-mail: [email protected] S. Mihaila D. Vinereanu University of Medicine and Pharmacy ‘‘Carol Davila’’, Bucharest, Romania M. H. Miglioranza Cardiology Institute of Rio Grande Do Sul, Porto Alegre, Brazil
while maximum AP diameter and SphI occurred at EDF. MAA fractional shortening was 35 ± 10 %. AP diameter change was 25 ± 10 %. MAC, ALPM and SphI showed similar patterns during left ventricular (LV) systole, and remained unchanged during diastole. Fractional change was 35 ± 10 % for MAC, and 13 ± 8 % for ALPM diameter. Our study provides the normal dynamics of the MA during the entire cardiac cycle. It reveals ‘‘pre-systolic’’ contraction of the MA, related to left atrial (LA) contraction, and minimal MAA during early LV systole. Therefore, the normal MA dynamics relates to a ‘‘physiologic LA-LV coupling’’, and a complete MA contraction requires both and properly timed LA and LV systole. Keywords Three-dimensional echocardiography Mitral annulus Mitral valve Normal dynamics Abbreviations 2D Two-dimensional 3D Three-dimensional ALPM Anterolateral-posteromedial AP Anterior-posterior ED End-diastole EDF Early diastolic filling ES End-systole LA Left atrium LDF Late diastolic filling LV Left ventricle MA Mitral annulus MR Mitral regurgitation MS Mid-systole MV Mitral valve MVC Mitral valve closure SphI Sphericity index TTE Transthoracic echocardiography
123
Int J Cardiovasc Imaging
Introduction Mitral regurgitation (MR) is highly prevalent [1] and, usually, it is associated with mitral annulus (MA) dysfunction [2–5]. Accurate characterization of the geometry and function of the MA plays an important role for better understanding the pathophysiology of the regurgitant mitral valve (MV) and for designing annular prosthesis [6, 7]. Although the MA dynamics have been characterized in pathological MV [2, 3, 5], the normal human MA dynamics during the entire cardiac cycle remains to be clarified. The translation function of the MA and the dynamic of its ‘‘saddle shape’’ have been thoroughly analyzed in normal individuals [8, 9]. Conversely, the changes in MA diameters and area are still a matter of debate, and the minimal MA size has been reported to occur at either early [3, 10], mid [4] or late systole [2]. Consequently, it is still controversial if the MA enlarges or decreases in size during the LV systole [2, 10, 11]. Moreover, MA area reduction has been mainly related to LV systolic function [12]. However, animal studies suggested that the area reduction of the normal MA starts before the onset of LV systole [13], suggesting a left atrial (LA) influence on it. In addition, experimental studies revealed that LA contraction contributes significantly to the effective MV closure (MVC) [14]. In those studies, the ‘‘pre-systolic’’ area reduction of the MA was absent in the animals in which the LA systole was abolished [13] and diminished during rapid LA pacing [15]. MA ‘‘pre-systolic’’ area reduction has been suggested in humans, too [11], using two-dimensional echocardiography (2DE). Even though the current guidelines recommend 2DE as the first imaging technique before MV surgery and for follow-up of the patients [16], 2DE has important limitations for MA assessment. It depends on correct alignment of anatomical landmarks [17] and underestimates the true dimensions of the MA [18]. Conversely, three-dimensional echocardiography (3DE) has the ability to provide anatomically sound images of the MV apparatus and quantitative analysis of the geometry and dynamics of the MA without geometrical assumptions. 3DE provided more accurate dimensions than 2DE for both normal and pathological MA, similar to those obtained by cardiac magnetic resonance [19]. However, even though the dynamics of the MA has been assessed by 3DE in different MV pathologies [3, 5, 20] or on very small number of normal subjects [10] using dedicated software that tracks the MA during the LV systole only, a thorough analysis of normal MA dynamics regarding its changes in diameters, area and shape during the entire cardiac cycle has not been reported yet. Accordingly, our study aim was to perform a quantitative assessment of the dynamics of the MA in healthy
123
volunteers during the entire cardiac cycle and to evaluate its relation to LA and LV systole.
Methods Study population 50 healthy Caucasian volunteers were prospectively recruited between January 2012 and September 2013 in a single tertiary center among hospital employees, fellows in training, their relatives and people who underwent medical visits for driving or working license. The inclusion criteria were: age [17 years, no history or symptoms of cardiovascular or lung disease, no cardiovascular risk factors (i.e. systemic arterial hypertension, smoking, diabetes, and hypercholesterolemia), normal electrocardiogram and physical examination, no cardio- or vasoactive treatment. Exclusion criteria were: trained athletes, pregnancy, body mass index [30 kg/m2, and poor apical acoustic window. Blood pressure, height and weight were measured in all subjects immediately before the echocardiographic examination. Body surface area was calculated according to the formulas by DuBois and DuBois [21]. An ECG was performed to confirm sinus rhythm and to exclude electrical abnormalities. University of Padua Ethics Committee approved the study (Protocol 2380 P) and all volunteers provided informed consent before entering the study. Echocardiography Using a standardized acquisition protocol, all examinations were performed using commercially available Vivid E9 system (GE Vingmed Ultrasound, Norway) equipped with M5S and 4 V probes. First, study subjects underwent standard transthoracic echocardiography to exclude subclinical heart diseases. Then, in order to obtain high temporal resolution of the datasets, separate 3D full-volume multi-beat acquisitions of the MV, the LA and the LV were obtained, by combining six consecutive ECG-triggered sub-volumes, during breath-hold and avoidance of patient or probe movement. Care was taken to encompass the MV in the full-volume throughout the cardiac cycle, and the entire LA and LV cavities in the dataset. By using multislice display, the absence of stitching artifacts was carefully checked. Image analysis 3DE LA full-volume datasets were converted to DICOM format and analyzed using dedicated software designed for volumetric analysis of the LA and recently validated
Int J Cardiovasc Imaging Fig. 1 Multiplanar review of the mitral annulus at mitral valve closure (MVC, upper panels) and early diastolic filling (EDF, lower panels). Red and green arrows point the frames of the cardiac cycle used for analysis. Flexi slice tool displays three 2D slices of the dataset, and one 3D volumerendered image (a, b) at each time frame. The transversal 2D slice (green border) shows a cut plane taken at the level of the rendered image (green dotted line on longitudinal planes). Cut-planes are manipulated to obtain a section at the level of the mitral annulus (MA). The aortic valve (Ao) and the tricuspid annulus (TA) were identified in both volumerendered image and derived 2D slices as anatomical landmarks
against cardiac magnetic resonance (LA analysis 2.3, TomTec Imaging Systems, Unterschleissheim, D) [17]. As previously described [17], the maximum (LAVmax), minimum (LAVmin) and pre-atrial contraction (LAVpreA) volumes of the LA were measured. 3DE datasets of the LV and MV were stored digitally in raw-data format for offline analysis using commercially available software package (EchoPAC BT 12, GE Vingmed Ultrasound, Norway). Quantification of 3D LV volumes and ejection fraction was performed using 4D Auto LVQ option in EchoPAC, as previously described [22, 23]. The quality of MV datasets was judged subjectively, considering the signal-to-noise ratio, the degree of bloodtissue contrast and the visualization of the MA. They were
classified as excellent, fair or poor, and only dataset with excellent visual quality and tracking of the MA were included in the study. Using frame-by-frame motion, 3D datasets of the MV were analyzed at six reference frames during the cardiac cycle: the mitral valve closure (MVC); mid-systole (MS); end-systole (ES), as the first frame after the aortic valve closure; early diastolic filling (EDF), as the frame with the MV wide opened during early LV filling; late diastolic filling (LDF), as the frame with the MV wide opened during late LV filling (after the P wave); and end-diastole (ED), as the frame just before the onset of the R wave. Using the ‘‘flexi-slice’’ option, the MV dataset was displayed in a multi-planar review mode. The MV full-
123
Int J Cardiovasc Imaging
between their maximum and minimum value, divided by the maximum values and expressed in percentages. Statistical analysis
Fig. 2 Layout with the correspondent 2D slice of the rendered 3D image of the mitral annulus. View from the atrium, during early diastolic filling. On this layout were performed the measures for the MA area and circumference, the AP and ALPM diameters, and sphericity index (left panel). ALPM anterolateral-posteromedial, AP anterior-posterior, Ao aorta, MA mitral annulus, TA tricuspid annulus
volume was automatically sectioned in two longitudinal and one transverse cut planes, displaying one 3D volumerendered image and three 2D slices of the dataset (Fig. 1). The 2D slice with the solid borders displayed the same view as the 3D volume-rendered image. The user had the flexibility to adjust both 2D slices by manipulating the position of the cut-planes and their intersection in any direction, to obtain a view of interest in both the 3D volume-rendered image and the correspondent 2D slice. The transverse cut-plane was set at the level of the MA and rotated by 180°, to visualize the MA from the LA side. Then, the correspondent 2D slice was rotated to display the MA with the aorta at 12 o’clock [24] (Fig. 1). Finally, the longitudinal cut-planes were adjusted to section the MA in its longest and shortest diameters. The resulting longitudinal 2D slices of the data set provided the anatomical landmarks for correct assessment of the MA (MV leaflets, LV and LA basal segments, aortic valve hinge). The 2D and 3D gain settings were manually adjusted to obtain the best visual delineation of the MA borders (Fig. 2) and remained unchanged throughout the entire cardiac cycle, to avoid consequent variability of the MA measurements. On the correspondent 2D slice, the following parameters of the MA were measured at each reference frame: projected MA area (MAA); projected MA circumference (MAC); antero-posterior (AP) diameter, as the shortest size of the MA; anterolateral-posteromedial (ALPM) diameter, as the longest diameter of the MA; and sphericity index (SphI) as the ratio between AP and ALPM diameter (Figs. 2, 3). Fractional shortening of the MAA and MAC, and of AP and ALPM diameters were obtained by the difference
123
Normal distribution of variables was verified by Kolmogorov–Smirnov test. Continuous variables were summarized as mean ± SD. MA measurements obtained at different reference frames during the cardiac cycle were compared using ANOVA for repeated measurements. Pearson’s correlation was used to analyze the relationships between different MA parameters and LV volumes, and between the fractional changes of MA parameters. Interobserver variability for MA assessment using 3DE was analyzed in 15 random subjects by two independent observers (M.M. and S.M.). S.M., who repeated the measurements 7 days later, assessed the intra-observer variability. Reproducibility was reported as intra-class correlation coefficient (ICC). All analyses were carried out using SPSS version 20.0 (SPSS, Inc., Chicago, IL). Differences among variables were considered significant at p \ 0.05.
Results We recruited 50 healthy volunteers between 18 and 74 years (31 men). Subjects’ demographics, physical and 3DE findings are summarized in Table 1. Temporal resolution of the 3D datasets was 34 ± 1 volumes/s. The average analysis time for each dataset was 30 ± 4 min. Mean values for MA parameters at the six selected reference frames of the cardiac cycle are summarized in Table 2, and schematic curves for MA parameters dynamic throughout the cardiac cycle in Fig. 4. MAA and AP diameter were the most ‘‘active’’ parameters, showing significant changes in all the reference frames during the cardiac cycle (all p \ 0.001). Minimal MAA, AP diameter and SphI occurred at the MVC and increased afterwards, during the LV systole. Maximum MAA was reached at ES, and maximum AP diameter and SphI during EDF. MAA started to decrease abruptly before the R wave, mainly in its AP diameter, reaching its minimum at the MVC (Fig. 2; 4). Of the total MA area reduction calculated from the maximal ES value to the minimal MVC value, 61 ± 22 % occurred before the R wave, and 43 ± 37 % after the R wave (i.e. between R wave and MVC). MAC and ALPM diameter showed significant changes from the onset of the R wave until EDF, with a pattern similar to the one documented for MAA and AP diameter. Conversely, both MAC and ALPM diameter remained rather constant during diastole (between EDF and LDF).
Int J Cardiovasc Imaging
Fig. 3 The analysis of the mitral annulus shape and dimensions at six reference frames of the cardiac cycle. View from the atrial side. a 3D rendered images of the mitral annulus; b 2D correspondent layouts,
Table 1 Demographic, physical and three-dimensional echocardiography findings of the healthy volunteers Clinical features
N = 50
Males no (%)
31 (62 %)
Age (years)
46 ± 14
Weight (kg)
70 ± 10
Height (cm)
172 ± 8
Body mass Index (kg/m2)
23.2 ± 2.6
Body surface area (m2)
1.8 ± 0.2
Systolic blood pressure (mmHg)
121 ± 10
Diastolic blood pressure (mmHg)
73 ± 8
Heart rate (bpm)
68 ± 8
Echocardiographic features LV end-diastolic volume (ml)
114 ± 21
Indexed LV end-diastolic volume (ml/m2)
62 ± 9
LV end-systolic volume (ml)
43 ± 10
Indexed LV end-systolic volume (ml/m2)
23 ± 4
LV ejection fraction (%)
62 ± 4
LV sphericity index
0.36 ± 0.07
for performing measurements. Ao aorta, ED end-diastole, EDF early diastolic filling, ES end-systole, LDF late diastolic filling, MA mitral annulus, MS mid-systole, MVC mitral valve closure
ALPM diameter shortening (Fig. 5). There was no relationship between the fractional shortening of the MA parameters and ageing of the normal individuals. LV end-diastolic and end-systolic volumes correlated with both MAA and MAC measured at every reference frame of the cardiac cycle (r [ 0.5, p \ 0.001). As expected, the strongest correlations were found between the LV end-diastolic volume and MAA and MAC measured at ED, and between LV end-systolic volume and MAA and MAC measured at ES (Fig. 6). Conversely, there was no relationship between MAA fractional change and LV ejection fraction. LAVmin correlated with the MAA and MAC measured at all reference frames of the cardiac cylce (r [ 0.6, all p \ 0.001). The strongest correlations were found between the LAVmin and MAA measured at LDF and ED (after the LA contraction), and between the LAVmin and MAC measured at LDF and MVC (after the LA contraction, and at early LV systole respectively; Fig. 7). The LAVmax did not correlate with either MAA or MAC. Reproducibility
MAA and AP diameter fractional changes throughout the cardiac cycle were 35 ± 10 %, and 25 ± 10 % respectively. The mean fractional change of MAC was 35 ± 10 %, whereas the one of ALPM diameter was 13 ± 8 %, only. The fractional changes of the projected MAA and MAC were related more to the AP than to
MA parameters showed good intraobserver and interobserver reproducibility in all analyzed reference frames (Table 3). The best intraobserver reproducibility for all MA parameters was observed at MVC and MS, while for the interobserver variability only at MS frame.
123
Int J Cardiovasc Imaging Table 2 Mitral annulus geometry parameters during phases of the cardiac cycle Time-points during cardiac cycle Mitral valve closure
Circumference (cm)
Area (cm2)
AP diameter (cm)
ALPM diameter (cm)
SphI
9.6 ± 1*
6.0 ± 1.3*
2.1 ± 0.3*
3.4 ± 0.3*
0.60 ± 0.08*
Mid-systole
10.9 ± 1.2*
8.1 ± 1.8*
2.5 ± 0.3*
3.8 ± 0.4*
0.64 ± 0.07*
End-systole
11.6 ± 1.1*
9.3 ± 1.9*
2.7 ± 0.3*
3.9 ± 0.4*
0.70 ± 0.08*
Early diastolic filling
10.9 ± 1.2
8.7 ± 1.8*
2.8 ± 0.3*
3.7 ± 0.5
0.76 ± 0.08
Late diastolic filling
10.6 ± 1.1
8.2 ± 1.7*
2.7 ± 0.7*
3.7 ± 0.5
0.72 ± 0.08
7.1 ± 1.7*
2.4 ± 0.3*
3.6 ± 0.5*
0.67 ± 0.07*
End-diastole
9.9 ± 1.4*
Data are summarized as mean ± SD AP anterior-posterior, ALPM anterior-lateral posterior-medial, SphI sphericity index * Statistical significance with p \ 0.005 for all marked cases, by Anova repeated measures, followed by paired T test
Fig. 4 The magnitude of changes of mitral annulus parameters during the cardiac cycle. Circumference (cm); area (cm2); anterior-posterior (AP) diameter (cm); anterolateralposteromedial (ALPM) diameter; sphericity index (SphI)
123
Int J Cardiovasc Imaging
Fig. 5 Relationships between the contraction of the mitral annulus area (upper panels) and circumference (lower panels) with the fractional changing of mitral annulus diameters. Data are expressed in percentages (%). AP antero-posterior, ALPM anterolateral-posteromedial
Discussion The present study provides a quantitative analysis of the dynamic changes in size and shape of the MA during the entire cardiac cycle in healthy volunteers, using 3DE. The main findings of the study were: (1) MA area reduction started before the onset of LV systole (in parallel to LA contraction) and reached its maximum at the MVC; (2) MAA and AP size presented significant dynamic changes during all phases of the cardiac cycle, while MAC, ALPM size and MA sphericity remained unchanged during diastole; (3) MA reached its minimal size and sphericity at the MVC and increased afterwards during LV systole; (4) The largest MAA was measured at ES, while maximum MA sphericity occurred at EDF. The MA shape, size, and dynamics influence the leaflets’ geometry, which is crucial for complete coaptation of MV leaflets and avoidance of MR. Consequently, changes of the normal MA dynamics are closely related to the development of MR [2, 3, 5], and mitral annuloplasty is the most common
surgical procedure to repair a regurgitant MV [25, 26]. Therefore, understanding normal dynamics of the MA is particularly important not only to characterize changes in mitral disease states but also to guide further refinements in MV repair towards normal MA geometry and function. We used transthoracic 3DE and multiplanar review analysis, which uses multiple cut-planes of the MV apparatus and allows a better delineation of the MA, to analyze the dynamics of the MA throughout the entire cardiac cycle in healthy volunteers with a wide age range and excellent quality of the 3D MV datasets. The parameters for MA dimensions obtained by 3DE were reproducible in all six frames of the cardiac cycle. In addition, the assessment of the MA by 3DE was already proved more reliable than 2DE, with results similar to those obtained by CMR [18, 19]. Previous human studies have used 3DE to reveal MA dysfunction in pathological MV [5, 20, 27, 28] and in the follow-up after the repair of the MV [6, 7] or the surrounding structures [6, 29–31]. 3DE assessment of the MA dynamics also served to design more physiological MA
123
Int J Cardiovasc Imaging
Fig. 6 The relationship between mitral annulus area (left panels) and circumference (right panels) at end-diastole and end-systole and left ventricular end-diastolic and end-systolic volumes, respectively. ED
end-diastole, ES end-systole, LVED left ventricle end-diastolic, LVES left ventricle end-systolic
prosthesis [32]. However, the normal dynamics of the MA provided by these studies remained controversial [2–5], especially in terms of MA area reduction. MA ‘‘saddle shape’’ and translation function were analyzed both in normal and pathological settings, therefore they were not the aim of our study [5, 8, 9, 20, 33]. Conversely, the extent and moreover, the timing of the maximal MA area reduction in normal individuals was the focus of our study. From our data, overall reduction of MA area is similar to that reported in experimental animals studies by Davis and Kinmonth [34], the first investigators that implanted radiopaque markers around canine MA and showed a MA area change of 30 % from diastole to systole. Tsakiris et al. [35] also observed a 19–34 % area reduction in dogs, and reported that MA area reduction starts in the pre-systolic period. MA ‘‘contraction’’ function, as it was called by Silbiger et al. [36], is of clinical importance. It was found decreased
in patients with functional MR [3] in relation with MR severity [20], and also in patients with organic MR due to MV prolapse or Barlow disease [5]. Conversely, Little et al. reported normal MA ‘‘contraction’’ in patients with organic MR, by comparing the MA area change of patients with organic MR with the MA area change of only 26 ± 8 % found in 15 healthy controls throughout the entire cardiac cycle. Therefore, a standardization of the overall MA area reduction would be beneficial when comparing normal with pathological MA. Our study found maximum MA area reduction during the early LV systole, with a consequent minimum size of the MA occurring at the MVC, similar to those reported by Kwan [10] and Topilsky [3] in a limited number of healthy subjects. Conversely, our results were different from data found by Kaplan [4] or Little et al. [2] that reported a minimum MA area at MS and at ES, respectively. Similar to Kwan [10] findings, we showed that, after the minimal
123
Int J Cardiovasc Imaging
Fig. 7 The relationship between the mitral annulus area (left panels) and circumference (right panels) measured at either late diastolic filling, end-diastole, or mitral valve closure and minimal left atrium
volume, measured after the left atrial contraction. ED end-diastole, LAVmin the minimal left atrial volume, LDF late diastolic filling, MVC mitral valve closure
MA size occurring at early systole, the MA size increases further during the LV systole. Indeed, the 3DE studies mentioned above used dedicated MV software that can perform a thorough analysis of the MA geometry, including MA height and nonplanarity. However, they were able to perform a dynamic analysis of the MA during the cardiac systole, only. Therefore, to be able to analyze the dynamics of the MA during the entire cardiac cycle, we performed our study without dedicated software for the 3DE dataset. We found that the normal MA area reduction starts before the LV systole. Thus, we confirmed the data suggested by Glasson et al. [13] in animal studies or by Ormiston et al. [11] in humans, using 2DE. Timek et al. also showed in canine models that the MA area reduction occurs in parallel with the atrial contraction [15], which proved important for MV leaflet closure [14]. Our results strengthen the idea of an important role of the LA for the MA ‘‘contraction’’ function and for an effective MVC. The amount of ‘‘pre-systolic’’ area reduction of the
MA found by us in normal humans (61 ± 22 %), was relatively lower than 89 % reported by Glasson [13] in experimental ovine study. Both LA dysfunction and decreased MA ‘‘contraction’’ have been reported in patients with functional or organic MR [5, 20, 37, 38]. Considering that normal MA area reduction starts in parallel with the LA contraction, it would be of interest to further assess the relation between LA and MA dysfunction in MV disease with consequent MR, and to analyze if these events might have a prognostic impact in these clinical settings. MAA and AP diameter showed significant changes from each reference frame of the cardiac cycle to the other, during both LV systolic and diastolic period, as previously reported [4, 10]. These findings suggest an ‘‘active’’ MA during the entire cardiac cycle, contradicting the static MA behavior during LV diastole reported by Grewal et al. [5]. As expected, MAA change was closely related to MA circumference reduction, and we found that the AP
123
Int J Cardiovasc Imaging Table 3 Intra-observer and inter-observer reproducibility for the mitral annulus area, circumference and diameters measured with multi-planar review
MA parameter
MA area
MA circumference
AP diameter
AP antero-posterior, ALPM anterolateral-posteromedial, CI confidence interval, ED enddiastole, EDF end-diastolic filling, ES end-systole, ICC intra-class coefficient, LD late diastole, LDF late diastolic filling, MA mitral annulus, MS mid-systole, MVC mitral valve closure * Statistical significance, for all ICC p \ 0.01
ALPM diameter
Reference frame
Inter-observer
ICC*
ICC*
95 % CI interval
95 % CI interval
MVC
0.98
0.94–0.99
0.87
0.33–0.97
MS
0.97
0.89–0.99
0.95
0.75–0.99
ES
0.95
0.83–0.99
0.89
0.47–0.98
EDF
0.92
0.65–0.98
0.92
0.55–0.98
LDF
0.95
0.78–0.98
0.87
0.34–0.97
ED
0.95
0.81–0.99
0.88
0.44–0.98
MVC
0.98
0.94–0.99
0.87
0.33–0.97
MS
0.97
0.89–0.99
0.91
0.56–0.98
ES
0.96
0.83–0.99
0.82
0.25–0.96
EDF
0.92
0.65–0.98
0.84
0.20–0.97
LDF ED
0.94 0.95
0.79–0.98 0.78–0.98
0.81 0.70
0.21–0.95 0.45–0.98
MVC
0.83
0.31–0.96
0.71
0.56–0.98
MS
0.85
0.30–0.97
0.84
0.25–0.97
ES
0.75
0.40–0.94
0.72
0.32–0.81
EDF
0.82
0.60–0.86
0.80
0.43–0.98
LDF
0.85
0.30–0.97
0.86
0.29–0.97
ED
0.85
0.16–0.94
0.84
0.23–0.96
MVC
0.94
0.76–0.98
0.85
0.24–0.96
MS
0.97
0.90–0.99
0.93
0.66–0.98
ES
0.97
0.88–0.99
0.85
0.27–0.97
EDF
0.84
0.30–0.96
0.84
0.22–0.97
LDF
0.95
0.79–0.98
0.90
0.45–0.98
ED
0.92
0.69–0.98
0.88
0.45–0.97
diameter change contributed more than the ALPM diameter change to MA area reduction [4, 5]. The unequal changes of the MA diameters lead to a minimum MA sphericity at the MVC, which is probably due to MA folding, and a maximum sphericity during EDF, when the MV is wide opened. MA from our healthy volunteers presented maximum area at the ES, similar to data reported by Khabbaz [28] and Grewal et al. [5] in 15 ‘‘control individuals’’. The timing of the maximum MAA was however controversial among different studies, and has been also reported either at early [9] or late diastole [2]. Even though Ormiston [11] reported a maximum MAA before the P wave, using 2DE, the study also specify an increase of the MA size during ES and isovolumic relaxation time. We found a significant correlation between MA size and LV volumes. However, the fractional changes of the MA parameters showed no correlation with the LV ejection fraction. This might be due to the significant, presystolic contribution of the LA systole to the MA area reduction, with consequent shape changes during the pre-systolic phase of the cardiac cycle. Therefore, the minimal LA volume measured after the LA contraction correlated with
123
Intra-observer
the MA size measured at all reference frames, and especially at the frames occurring immediately after the LA contraction (LDF, ED and MVC). Conversely, the maximum LA volume reached at ES did not correlated with either MAA or MAC. Limitations of the study The limited number of reference frames may be considered a limitation of the study, as now there is a commercially available dedicated MV software package for 3DE that allows the tracking and analysis of the MA frame-byframe. However, this software package analysis capability is limited only to the systolic period, and the purpose of our study was to analyze the changes of the MA during the entire cardiac cycle, and to identify its most important events. We provided data about the shape, size and fractional changes of MA diameters and projected area and circumference during the entire cardiac cycle, but not about the MA ‘‘saddle shape’’ and translation function. However, the latter have been previously described in normal subjects in other studies [8, 9].
Int J Cardiovasc Imaging
‘‘End-systole (ES)’’ was variably defined in the different studies, either before the aortic valve closure [3], the frame before MV starts to open [19], or in between [10], similar to our study. These differences in the choice of the ‘‘reference frames’’ may be partially responsible for the inconsistent MA dynamics reported in the literature, as the MA can present significant changes in size at just a few frames distance.
Conclusions Our study provides the normal dynamics of the MA during the entire cardiac cycle and documents the ‘‘pre-systolic’’ area reduction of the MA, which is related to the atrial systole, and maximum MA area reduction during early LV systole. We emphasize the importance of a physiologic coupling of the LA-LV systole for a normal ‘‘contraction’’ function of the MA and an efficient MVC. The normal dynamic of the MA provided by our study can be used as a reference for 3DE studies of the dynamics of annular prosthesis or of native MA in patients with pathological MV and after the interventions performed on the surrounding structures. Acknowledgments Dr. Mihaila is an EACVI Research Grant winner on 2011 and an ESC Training Grant winner on 2014. Dr Mihaila is also a winner of a Grant for PhD research of the University of Medicine and Pharmacy ‘‘Carol Davila’’ Bucharest, POSDRU/159/ 1.5/S/141531. Conflict of interest Doctors Badano and Muraru are consultants and received equipment grants from GE Healthcare and TomTec Imaging Systems.
References 1. Iung B, Baron G, Butchart EG, Delahaye F, Gohlke-Barwolf C, Levang OW et al (2003) A prospective survey of patients with valvular heart disease in Europe: the euro heart survey on valvular heart disease. Eur Heart J 24(13):1231–1243 2. Little SH, Ben Zekry S, Lawrie GM, Zoghbi WA (2010) Dynamic annular geometry and function in patients with mitral regurgitation: insight from three-dimensional annular tracking. J Am Soc Echocardiogr 23(8):872–879 3. Topilsky Y, Vaturi O, Watanabe N, Bichara V, Nkomo VT, Michelena H et al (2013) Real-time 3-dimensional dynamics of functional mitral regurgitation: a prospective quantitative and mechanistic study. J Am Heart Assoc 2(3):e000039 4. Kaplan SR, Bashein G, Sheehan FH, Legget ME, Munt B, Li XN et al (2000) Three-dimensional echocardiographic assessment of annular shape changes in the normal and regurgitant mitral valve. Am Heart J 139(3):378–387 5. Grewal J, Suri R, Mankad S, Tanaka A, Mahoney DW, Schaff HV et al (2010) Mitral annular dynamics in myxomatous valve disease: new insights with real-time 3-dimensional echocardiography. Circulation 121(12):1423–1431
6. Caiani EG, Fusini L, Veronesi F, Tamborini G, Maffessanti F, Gripari P et al (2011) Quantification of mitral annulus dynamic morphology in patients with mitral valve prolapse undergoing repair and annuloplasty during a 6-month follow-up. Eur J Echocardiogr 12(5):375–383 7. Mahmood F, Subramaniam B, Gorman JH III, Levine RM, Gorman RC, Maslow A et al (2009) Three-dimensional echocardiographic assessment of changes in mitral valve geometry after valve repair. Ann Thorac Surg 88(6):1838–1844 8. Ryan LP, Jackson BM, Enomoto Y, Parish L, Plappert TJ, St John-Sutton MG et al (2007) Description of regional mitral annular nonplanarity in healthy human subjects: a novel methodology. J Thorac Cardiovasc Surg 134(3):644–648 9. Veronesi F, Corsi C, Sugeng L, Caiani EG, Weinert L, Mor-Avi V et al (2008) Quantification of mitral apparatus dynamics in functional and ischemic mitral regurgitation using real-time 3-dimensional echocardiography. J Am Soc Echocardiogr 21(4):347–354 10. Kwan J, Jeon MJ, Kim DH, Park KS, Lee WH (2009) Does the mitral annulus shrink or enlarge during systole? A real-time 3D echocardiography study. J Korean Med Sci 24(2):203–208 11. Ormiston JA, Shah PM, Tei C, Wong M (1981) Size and motion of the mitral valve annulus in man. I. A two-dimensional echocardiographic method and findings in normal subjects. Circulation 64(1):113–120 12. Kwan J, Qin JX, Popovic ZB, Agler DA, Thomas JD, Shiota T (2004) Geometric changes of mitral annulus assessed by real-time 3-dimensional echocardiography: becoming enlarged and less nonplanar in the anteroposterior direction during systole in proportion to global left ventricular systolic function. J Am Soc Echocardiogr 17(11):1179–1184 13. Glasson JR, Komeda M, Daughters GT, Foppiano LE, Bolger AF, Tye TL et al (1997) Most ovine mitral annular three-dimensional size reduction occurs before ventricular systole and is abolished with ventricular pacing. Circulation 96(9 Suppl):II-115–II-22 discussion II-23 14. Timek T, Dagum P, Lai DT, Green GR, Glasson JR, Daughters GT et al (2001) The role of atrial contraction in mitral valve closure. J Heart Valve Dis 10(3):312–319 15. Timek TA, Lai DT, Dagum P, Green GR, Glasson JR, Daughters GT et al (2000) Mitral annular dynamics during rapid atrial pacing. Surgery 128(2):361–367 16. Lancellotti P, Tribouilloy C, Hagendorff A, Popescu BA, Edvardsen T, Pierard LA et al (2013) Recommendations for the echocardiographic assessment of native valvular regurgitation: an executive summary from the European association of cardiovascular imaging. Eur Heart J Cardiovasc Imaging 14(7):611–644 17. Foster GP, Dunn AK, Abraham S, Ahmadi N, Sarraf G (2009) Accurate measurement of mitral annular dimensions by echocardiography: importance of correctly aligned imaging planes and anatomic landmarks. J Am Soc Echocardiogr 22(5):458–463 18. Anwar AM, Soliman OI, ten Cate FJ, Nemes A, McGhie JS, Krenning BJ et al (2007) True mitral annulus diameter is underestimated by two-dimensional echocardiography as evidenced by real-time three-dimensional echocardiography and magnetic resonance imaging. Int J Cardiovasc Imaging 23(5):541–547 19. Anwar AM, Soliman OI, Nemes A, Germans T, Krenning BJ, Geleijnse ML et al (2007) Assessment of mitral annulus size and function by real-time 3-dimensional echocardiography in cardiomyopathy: comparison with magnetic resonance imaging. J Am Soc Echocardiogr 20(8):941–948 20. Daimon M, Saracino G, Fukuda S, Koyama Y, Kwan J, Song JM et al (2010) Dynamic change of mitral annular geometry and motion in ischemic mitral regurgitation assessed by a computerized 3D echo method. Echocardiography 27(9):1069–1077
123
Int J Cardiovasc Imaging 21. Dubois D, Dubois EF (1916) Clinical calorimetry. A formula to estimate the approximate surface area if height and weight be known. Arch Intern Med 17:863–871 22. Muraru D, Badano LP, Piccoli G, Gianfagna P, Del Mestre L, Ermacora D et al (2010) Validation of a novel automated borderdetection algorithm for rapid and accurate quantitation of left ventricular volumes based on three-dimensional echocardiography. Eur J Echocardiogr 11(4):359–368 23. Muraru D, Badano LP, Peluso D, Dal Bianco L, Casablanca S, Kocabay G et al (2013) Comprehensive analysis of left ventricular geometry and function by three-dimensional echocardiography in healthy adults. J Am Soc Echocardiogr 26(6):618–628 24. Lang RM, Badano LP, Tsang W, Adams DH, Agricola E, Buck T et al (2012) EAE/ASE recommendations for image acquisition and display using three-dimensional echocardiography. J Am Soc Echocardiogr 25(1):3–46 25. Ling LH, Enriquez-Sarano M, Seward JB, Orszulak TA, Schaff HV, Bailey KR et al (1997) Early surgery in patients with mitral regurgitation due to flail leaflets: a long-term outcome study. Circulation 96(6):1819–1825 26. Enriquez-Sarano M, Avierinos JF, Messika-Zeitoun D, Detaint D, Capps M, Nkomo V et al (2005) Quantitative determinants of the outcome of asymptomatic mitral regurgitation. N Engl J Med 352(9):875–883 27. De Simone R, Wolf I, Mottl-Link S, Hoda R, Mikhail B, Sack FU et al (2006) A clinical study of annular geometry and dynamics in patients with ischemic mitral regurgitation: new insights into asymmetrical ring annuloplasty. Eur J Cardiothorac Surg 29(3):355–361 28. Khabbaz KR, Mahmood F, Shakil O, Warraich HJ, Gorman JH III, Gorman RC et al (2013) Dynamic 3-dimensional echocardiographic assessment of mitral annular geometry in patients with functional mitral regurgitation. Ann Thorac Surg 95(1):105–110 29. Warraich HJ, Matyal R, Bergman R, Hess PE, Khabbaz K, Manning WJ et al (2013) Impact of aortic valve replacement for
123
30.
31.
32.
33.
34. 35.
36. 37.
38.
aortic stenosis on dynamic mitral annular motion and geometry. Am J Cardiol 112(9):1445–1449 Tsang W, Meineri M, Hahn RT, Veronesi F, Shah AP, Osten M et al (2013) A three-dimensional echocardiographic study on aortic-mitral coupling in transcatheter aortic valve replacement. Eur Heart J Cardiovasc Imaging 14(10):950–956 Tsang W, Veronesi F, Sugeng L, Weinert L, Takeuchi M, Jeevanandam V et al (2013) Mitral valve dynamics in severe aortic stenosis before and after aortic valve replacement. J Am Soc Echocardiogr 26(6):606–614 Mahmood F, Gorman JH III, Subramaniam B, Gorman RC, Panzica PJ, Hagberg RC et al (2010) Changes in mitral valve annular geometry after repair: saddle-shaped versus flat annuloplasty rings. Ann Thorac Surg 90(4):1212–1220 Jensen MO, Hagege AA, Otsuji Y, Levine RA, Leducq Transatlantic MN (2013) The unsaddled annulus: biomechanical culprit in mitral valve prolapse? Circulation 127(7):766–768 Davis PK, Kinmonth JB (1963) The movements of the annulus of the mitral valve. J Cardiovasc Surg (Torino) 4:427–431 Tsakiris AG, Von Bernuth G, Rastelli GC, Bourgeois MJ, Titus JL, Wood EH (1971) Size and motion of the mitral valve annulus in anesthetized intact dogs. J Appl Physiol 30(5):611–618 Silbiger JJ (2012) Anatomy, mechanics, and pathophysiology of the mitral annulus. Am Heart J 164(2):163–176 Gertz ZM, Raina A, Saghy L, Zado ES, Callans DJ, Marchlinski FE et al (2011) Evidence of atrial functional mitral regurgitation due to atrial fibrillation: reversal with arrhythmia control. J Am Coll Cardiol 58(14):1474–1481 Cameli M, Lisi M, Righini FM, Massoni A, Natali BM, Focardi M et al (2013) Usefulness of atrial deformation analysis to predict left atrial fibrosis and endocardial thickness in patients undergoing mitral valve operations for severe mitral regurgitation secondary to mitral valve prolapse. Am J Cardiol 111(4):595–601
