Abstract
Background
Mitral annular calcification (MAC) is a common finding in the elderly that is associated with worse outcomes. It is postulated that MAC is a different form of atherosclerosis. Epicardial fat tissue (EFT) is in close contact with different heart structures and is especially pronounced in the atrioventricular grooves and the area surrounding the atrioventricular valve apparatus. The amount of EFT is associated with the extent of coronary artery disease, including plaque burden and coronary calcification. The aim of this study was to investigate whether EFT is also associated with the extent of MAC.
Methods
In this retrospective analysis of n = 543 (53.6% female) consecutive patients with high-grade aortic stenosis, EFT volume and MAC were assessed in the pre-transcatheter aortic valve implantation computed tomography scans. Patients with no/mild MAC and moderate/severe MAC were compared in terms of EFT volume and baseline, procedural, and postprocedural characteristics. Univariate and multivariable regression analyses were performed with MAC as dependent variable.
Results
Over one-quarter (n = 154; 28.4%) of the patients had moderate or severe MAC. The EFT volume between those with moderate/severe MAC and those with little or no MAC did not differ significantly: 130 (interquartile range 94-164) cm3 vs. 133 (interquartile range 95-176) cm3; p = 0.704. Accordingly, EFT volume did not predict increased MAC. Female sex, atrial fibrillation, and prior pacemaker implantation were independent predictors of moderate/severe MAC.
Conclusions
Increased EFT is not a predictor of MAC, and thus its unfavorable proinflammatory properties do not seem to play a significant role in the development of MAC.
Keywords: Aortic stenosis, Body mass index, Epicardial fat, Mitral annular calcification, Transcatheter aortic valve implantation
Highlights
-
•
Mitral annular calcification (MAC) is highly prevalent in patients with severe aortic stenosis undergoing transcatheter aortic valve implantation.
-
•
Patients with moderate/severe MAC did not have more epicardial fat tissue (EFT) compared with patients with no or mild MAC.
-
•
Female sex, atrial fibrillation, and preexisting permanent pacemaker, but not increased EFT, were independent predictors of MAC.
Introduction
Mitral annular calcification (MAC) is a chronic, degenerative process of the fibrous support structure of the mitral valve,1 affecting approximately 8% to 15% of the general population.2 Its prevalence is higher in the elderly population, in patients with chronic renal disease or aortic stenosis (AS), and in women.2, 3, 4, 5 MAC is a predictor of increased cardiovascular morbidity and mortality.6 Severe MAC has also been identified as an independent predictor of all-cause mortality and cardiovascular mortality in AS patients undergoing transcatheter aortic valve implantation (TAVI).2 The pathogenesis of MAC remains incompletely understood, but based on the strong association between cardiovascular risk factors and MAC, it is suggested that MAC is a different form of atherosclerosis.7
Epicardial fat tissue (EFT) is located between the myocardium and the visceral layer of the pericardium, directly surrounding the coronary vessels and the adjacent heart structures without any intervening fascia or boundaries.8 Among other structures, it is especially concentrated in the area of the atrioventricular grooves9 and therefore in close proximity to the mitral annulus. EFT has several protective mechanical, thermogenic, and anti-inflammatory functions but also has adverse effects by releasing proinflammatory cytokines and molecules that can potentially be delivered to adjacent heart structures by vasocrine or endocrine mechanisms.10,11 Since a number of studies have shown a clear relationship between the amount of EFT and the extent of coronary artery disease and coronary calcification,12, 13, 14 and based on the assumption that MAC is a different form of atherosclerosis, we hypothesized that greater amounts of MAC are associated with increased EFT. Given that MAC has a higher prevalence in patients with AS and that all AS patients planned for TAVI undergo a routine pre-TAVI computed tomography (CT) scan, we aimed to investigate the relationship between MAC and EFT and correlate the amount of these 2 entities in this special patient subset. Furthermore, patients with no or low amounts of MAC and those with moderate/severe MAC were compared in terms of baseline characteristics and procedural and postprocedural findings, and independent predictors of higher amounts of MAC were identified.
Methods
Patient Cohort
The study cohort was previously described.15 Briefly, between January 2016 and August 2017, 560 consecutive patients with symptomatic high-grade aortic stenosis scheduled for TAVI were included. The diagnosis of severe aortic stenosis was made based upon current guideline recommendations16: the presence of a mean transvalvular gradient of ≥40 mmHg measured over the aortic valve and/or an aortic valve area of ≤1.0 cm2 as calculated via the continuity equation or measured via planimetry on transesophageal echocardiography scans. Routine pre-TAVI multidetector computed tomography (MDCT) examinations were assessed for the amount of epicardial fat volume with an established, fully automated deep-learning algorithm (described below). For the present analysis, 17 patients were excluded due to prior mitral valve surgery, either mitral valve repair (n = 2) or mitral valve replacement (n = 4), or the use of different CT software (n = 11). Two patients had prior mitral valvuloplasty. In 5 cases, semiquantitative but not quantitative grading of MAC was possible due to artifacts. The final study cohort comprised of 543 patients.
The study was conducted in adherence to the Declaration of Helsinki. The local Ethics Committee of the University of Giessen, Germany, approved the study, and patient consent was waived due to the retrospective design of the study.
MDCT Analysis
Electrocardiogram-gated MDCT examinations were performed with a 128-slice or a 384-slice dual-source scanner (SOMATOM Definition or Force; Siemens Healthineers, Forchheim, Germany) as previously described.17 Reconstructions were carried out using a cardiac-gated B26f or I26f algorithm with a slice thickness of 0.6 mm in systole at 35% and in diastole at 70% of the cardiac cycle. Pre-TAVI routine measurements of the aortic root and the vascular access site were performed offline on a dedicated workstation (3mensio Structural Heart, Pie Medical Imaging, Maastricht, the Netherlands).
Quantification of MAC
For the assessment of MAC, as a first step, multiplanar MDCT views by rotation of the crosshairs were obtained offline via the aforementioned workstation (Syn.govia, Siemens Healthineers). For semiquantitative measurement of MAC, the grading system by Amat-Santos18 consisting of 4 categories of MAC severity was used: based on the involvement of the mitral annular ring, no MAC (grade 0), mild MAC defined as circumferential involvement of less than one-third of the annular ring (grade 1), moderate MAC with involvement of one-third to one-half (grade 2), and severe MAC with calcification of more than one-half of the ring (grade 3) was determined (Figure 1). Then, in a second step that quantitatively assessed MAC, the Agatston score (measured in Agatston units [AU] of the mitral annular ring) was derived using axial views of the noncontrast-enhanced scans by carefully tagging the calcified areas. Areas with a lesion density of at least 130 Hounsfield units were considered as calcified. Special attention was paid to the adjacent areas of calcification close to the left ventricular outflow tract or calcification of the adjacent circumflex artery in order to properly distinguish the anatomic involvement.
Figure 1.
Example of a severe MAC in a patient with high-grade aortic stenosis. In an MDCT scan, the calcification extends over more than half of the mitral valve annulus and thus is an example of grade 3 (moderate/severe) mitral annular ring calcification.
Abbreviations: bpm. beats per minute; CAUD, caudal; LAO, left anterior oblique; MAC, mitral annular calcification; MDCT, multidetector computed tomography; MPR, multiplanar reconstruction; ms, milliseconds.
Measurement of EFT Volume
EFT volume was quantified from CT scans in cm3 by using an established, fully automated deep-learning algorithm that is a component of research software (QFAT V2.0, Cedars-Sinai Medical Center, Los Angeles, California) and has been validated and tested in a large, multicenter study.19 For EFT measurements, the bifurcation of the pulmonary trunk was used as the upper boundary, and the most inferior slice with any portion of the heart was used as the lower limit, as recommended by Dey et al.20 When tracings of the pericardial border were insufficient due to limited image quality, the tracings were manually adjusted. Figure 2 shows an example of an EFT measurement in an MDCT slice.
Figure 2.
Example of EFT volume measurement. CT-derived slice of the heart in the axial view with tracing of the pericardial border. The areas marked in purple represent EFT.
Abbreviations: CT, computed tomography; EFT, epicardial fat tissue.
Statistical Analysis
Continuous variables are provided as median with interquartile range (IQR) or as mean with SD, as appropriate. Testing for a normal distribution pattern was carried out with the Shapiro-Wilk test. Two groups were designated: patients with no MAC or mild MAC were allocated to one group, and patients with moderate or severe MAC were assigned to a second group. Comparisons of parameters between the 2 groups were made using the Mann-Whitney U test or the unpaired Student’s t-test, as appropriate. Categorical parameters were compared via the chi2 test. Univariate and multivariable logistic regression analyses were carried out to test for potential predictors of moderate/severe MAC. For this, variables that were significantly different between the 2 groups (Table 1, all variables with a p-value < 0.05) and EFT volume were further tested in the univariate logistic analysis, and subsequently, parameters with a p-value ≤ 0.1 in the univariate testing were included in the age-adjusted multivariable logistic analysis. The odds ratio and the corresponding 95% CI were determined. EFT volume was compared both as a dichotomized variable (according to its median value) and as a continuous variable between the 2 groups. Correlation analysis between EFT and MAC Agatston score was conducted via measurement of the Pearson correlation coefficient.
Table 1.
Baseline characteristics∗
| Total cohort (n = 543) | No/mild MAC (n = 389) | Moderate/severe MAC (n = 154) | p-value† | |
|---|---|---|---|---|
| Age, y | 82 [79-86] | 82 [79-85] | 82 [80-86] | 0.119 |
| Sex, female | 291 (53.6) | 183 (47) | 108 (70.1) | <0.001 |
| Body mass index, kg/m2 | 26.4 [23.7-29.8] | 26.1 [23.7-29] | 27.6 [23.6-31.3] | 0.023 |
| Mitral annular calcification, AU | 1201 [332-3582] | 434 [180-970] | 4631 [2759-7741] | <0.001 |
| NYHA class 3 or 4 | 423 (77.9) | 296 (76.1) | 127 (82.5) | 0.107 |
| Syncope | 92 (16.9) | 65 (116.7) | 25 (17.5) | 0.818 |
| Cardiac decompensation | 157 (28.9) | 108 (27.8) | 49 (31.8) | 0.348 |
| Hypertension | 509 (93.7) | 363 (93.3) | 146 (97.8) | 0.519 |
| Diabetes mellitus | 166 (30.6) | 115 (29.6) | 51 (33.1) | 0.418 |
| Hyperlipidemia | 164 (30.2) | 119 (30.6) | 45 (29.2) | 0.754 |
| STS score, % | 4.1 [2.8-5.9] | 3.9 [2.8-5.7] | 4.3 [2.9-6.4] | 0.093 |
| EuroScore II, % | 4.9 [3-8.2] | 4.5 [2.8-8.1] | 5.4 [3.3-8.5] | 0.068 |
| Coronary artery disease | 331 (61) | 233 (59.9) | 98 (63.6) | 0.421 |
| Previous MI | 77 (14.2) | 59 (15.2) | 18 (11.7) | 0.295 |
| Peripheral artery disease | 87 (16) | 66 (17) | 21 (13.6) | 0.34 |
| CKD (eGFR <60 mL/min/m2) | 312 (58.2) | 234 (60.9) | 78 (51.3) | 0.042 |
| Hemodialysis | 11 (2) | 8 (1.5) | 3 (0.6) | 0.936 |
| COPD | 97 (17.9) | 73 (18.8) | 24 (25.6) | 0.383 |
| Prior stroke | 71 (13.1) | 48 (12.3) | 23 (14.9) | 0.419 |
| Atrial fibrillation | 228 (42) | 153 (39.3) | 75 (48.7) | 0.046 |
| AV block I° | 95 (17.5) | 72 (18.5) | 23 (14.9) | 0.323 |
| LBBB | 57 (10.5) | 42 (10.8) | 15 (9.7) | 0.717 |
| RBBB | 57 (10.5) | 44 (11.3) | 13 (8.4) | 0.325 |
| Prior PPI | 70 (12.9) | 39 (10) | 31 (20.1) | 0.002 |
| Indications: | ||||
| High degree AV block | 32 (45.7) | 15 (38.5) | 17 (54.8) | |
| Sick sinus node disease | 17 (24.3) | 12 (30.8) | 5 (16.1) | |
| Bradyarrhythmia | 8 (11.4) | 3 (7.7) | 5 (16.1) | |
| Other indications | 13 (18.6) | 9 (23.1) | 41 (2.9) | |
| CABG | 67 (12.3) | 53 (13.6) | 14 (9.1) | 0.148 |
| Ejection fraction, % | 65 [55-65] | 65 [55-65] | 65 [60-65] | 0.08 |
| Pmean, mmHg | 42 [32-53] | 41 [31-51] | 43 [35-54] | 0.048 |
| Aortic valve area, cm2 | 0.7 [0.6-0.8] | 0.7 [0.6-0.8] | 0.6 [0.5-0.8] | 0.012 |
| Mitral valve insufficiency (any) | 451 (83.1) | 312 (80.2) | 139 (90.3) | 0.024 |
| Grade II/III | 65 (12.0) | 43 (11.1) | 22 (14.3) | 0.296 |
| Mitral valve stenosis (any) | 39 (7.2) | 8 (2.1) | 31 (20.1) | <0.001 |
| Grade II/III | 17 (3.1) | 3 (0.8) | 14 (9.1) | <0.001 |
| Epicardial fat tissue, cm3 | 131 [94-171] | 133 [95-176] | 130 [94-164] | 0.704 |
| Epicardial fat tissue/BSA, cm3/m2 | 69 [53-90] | 70 [53-91] | 67 [54-86] | 0.792 |
| Epicardial fat tissue/BMI, cm3/kg/m2 | 4.8 [3.6-6.3] | 4.8 [3.7-6.4] | 4.6 [3.6-6.2] | 0.231 |
Abbreviations: AV, atrioventricular; AU, Agatston units; BMI, body mass index; BSA, body surface area; CABG, coronary artery bypass graft; CKD, chronic kidney disease; COPD, chronic obstructive pulmonary disease; eGFR, estimated glomerular filtration rate; LBBB, left bundle branch block; MAC, mitral annular calcification; MI, myocardial infarction; NYHA, New York Heart Association; Pmean, mean transvalvular aortic pressure; PPI, permanent pacemaker implantation; RBBB, right bundle branch block; STS, Society of Thoracic Surgeons.
Values presented as number (%) or median [interquartile range].
Bold and italics denote a p-value below 0.05.
In 20 randomly selected patients, an interobserver and intraobserver intraclass correlation coefficient was determined for quantitative MAC measurements. The interobserver intraclass correlation coefficient was 0.987 (p < 0.001), and the intraobserver intraclass correlation coefficient was 0.982 (p < 0.001). According to Koo et al., the reliability of these results is classified as “excellent”.21
A two-sided level of p < 0.05 was considered significant. All statistical measurements were carried out with SPSS version 22.0 (IBM, Armonk, New York).
Results
Patient Characteristics
Table 1 shows the baseline characteristics of the entire cohort and compares the subgroups no/mild MAC and moderate/severe MAC. A total of 543 patients (53.6% females) were included in the analysis. The median age was 82 [79-86] years. Three hundred eighty-nine patients (71.6%) had no or mild MAC, and 154 patients (28.4%) had moderate or severe MAC. The prevalence of no, mild, moderate, and severe MAC was 27.4, 44.2, 13.1, and 15.3%, respectively. The median Agatston score for MAC of the overall cohort was 1201 [332-3582] AU, for the no/mild MAC group was 434 [180-970] AU, and for the moderate/severe MAC group was 4631 [2759-7741] AU (p < 0.001). The median age of patients with no/mild MAC and of those with moderate/severe MAC was not different: 82 [79-85] years vs. 82 [80-86] years; (p = 0.119). In the no/mild MAC group, the sexes were almost evenly distributed (47% female), whereas in the moderate/severe MAC group, the majority of patients were females (70.1%, p < 0.001). Chronic kidney disease (CKD) was more prevalent in the moderate/severe MAC group compared with the no/mild MAC cohort (33 vs. 25%, p = 0.042), and the body mass index (BMI) was higher (28 [24-31] kg/m2 vs. 26 [24-29] kg/m2; p = 0.023). The median aortic transvalvular gradient was higher in the subgroup of patients with moderate/severe MAC (43 [35-55] vs. 41 [31-51] mmHg; p = 0.048), and the aortic valve area was lower in this group: 0.6 [0.5-0.8] cm2 vs. 0.7 [0.6-0.8] cm2, p = 0.012). The prevalence of atrial fibrillation was greater in the moderate/severe MAC group than in the no/mild MAC group (32.9 vs. 25.1%, p = 0.046), and these patients had higher rates of preexisting permanent pacemaker implantation (PPI): 44.3% vs. 26% (p = 0.002). The presence of any degree of baseline mitral valve insufficiency was similar between the groups.
Procedural and Postprocedural Characteristics
Table 2 shows the procedural/postprocedural characteristics of the entire cohort and of the 2 groups. Self-expanding and balloon-expandable valves were used to a similar extent in the 2 groups. There were no relevant differences in terms of the prevalence of major bleeding, major vascular events, or disabling strokes between the groups. The rates of moderate or severe paravalvular leakage were similar.
Table 2.
Procedural and postprocedural characteristics∗
| Total cohort (n = 543) | No/mild MAC (n = 389) | Moderate/severe MAC (n = 154) | p-value | |
|---|---|---|---|---|
| Length of hospital stay, d | 8 [7-11] | 8 [7-11] | 8 [7-12] | 0.284 |
| Prothesis type | 0.197 | |||
| Balloon-expandable | 253 (46.6) | 188 (48.3) | 65 (42.2) | |
| Self-expanding | 290 (53.4) | 201 (51.7) | 89 (57.8) | |
| Procedural and postprocedural complications | ||||
| Major/life-threatening bleeding | 51 (9.4) | 37 (9.5) | 14 (9.1) | 0.88 |
| Major vascular complications | 42 (7.7) | 29 (7.5) | 13 (8.4) | 0.698 |
| Valve embolization | 9 (1.7) | 9 (2.3) | 0 (0) | 0.057 |
| Left ventricular injury | 1 (0.2) | 1 (0.3) | 0 (0) | 0.529 |
| Aortic root injury | 2 (0.4) | 1 (0.3) | 1 (0.6) | 0.496 |
| Valve-in-valve | 12 (2.2) | 11 (2.8) | 1 (0.6) | 0.12 |
| Device success | 491 (90.4) | 348 (89.5) | 143 (92.9) | 0.225 |
| Disabling stroke | 10 (1.8) | 9 (2.3) | 1 (0.7) | 0.195 |
| AKI stage 2 or 3 | 23 (4.2) | 14 (3.6) | 9 (5.8) | 0.242 |
| New PPI | 63 (11.6) | 51 (13.1) | 12 (7.8) | 0.081 |
| New atrial fibrillation | 19 (3.5) | 13 (3.3) | 6 (3.9) | 0.751 |
| Postprocedural echocardiographic findings | ||||
| Pmean, mmHg | 10 [7-13] | 9 [7-12] | 10 [7-14] | 0.144 |
| Aortic valve area, cm2 | 1.6 [1.3-1.8] | 1.6 [1.4-1.8] | 1.6 [1.3-1.8] | 0.429 |
| Paravalvular leakage ≥ moderate | 13 (2.4) | 7 (1.8) | 6 (4) | 0.146 |
| Follow-up | ||||
| 30-d all-cause mortality | 18 (3.3) | 13 (3.3) | 5 (3.2) | 0.955 |
| 1-y all-cause mortality | 82 (15.2) | 57 (14.7) | 25 (16.4) | 0.601 |
Abbreviations: AKI, acute kidney injury; MAC, mitral annular calcification; Pmean, mean transvalvular aortic pressure; PPI, permanent pacemaker implantation.
Values presented as number (%) or median [interquartile range].
EFT and MAC
The median EFT volume of the entire cohort was 131 [IQR 94-171] cm3 (Table 1). There was no significant difference in EFT volume between the 2 groups: 133 [IQR 95-176] cm3 for the no/mild MAC group vs. 130 [IQR 94-164] cm3 for the moderate/severe MAC group (p = 0.704) (Figure 3). In addition, when indexed for BMI and body surface area, EFT values were similar between the no/mild and moderate/severe MAC groups (Table 1). When dichotomized according to the median value in “low EFT volume” and “high EFT volume,” the values for the 2 groups did not differ (high EFT volume present in 50.9 vs. 49.4% for the no/low MAC and moderate/high MAC groups, respectively; p = 0.745). Accordingly, EFT volume and MAC (according to Agatston score) did not correlate: the Pearson correlation coefficient was −0.07 (p = 0.171).
Figure 3.
Boxplots of EFT volume of no/low MAC and moderate/severe MAC cohorts. No/low MAC and moderate/severe MAC groups do not significantly differ in terms of EFT volume.
Abbreviations: EFT, epicardial fat tissue; MAC, mitral annular calcification.
Predictors of Moderate/Severe MAC
In univariate logistic regression analysis, female sex, age, higher BMI, CKD, atrial fibrillation, prior PPI, mean transvalvular gradient, and lower aortic valve area were associated with the presence of moderate/severe MAC. More EFT did not predict higher MAC. In multivariable regression analysis, only female sex, atrial fibrillation, and prior PPI independently predicted moderate/severe MAC (Table 3).
Table 3.
Identification of potential independent predictors of moderate/severe MAC
| Univariate OR [95% CI] | p-value | Multivariable OR [95% CI] | p-value∗ | |
|---|---|---|---|---|
| Age, y | 1.03 (0.99-1.07) | 0.067 | 1.03 (0.99-1.07) | 0.211 |
| Sex, female | 2.64 (1.77-3.94) | <0.001 | 2.15 (1.38-3.37) | 0.001 |
| Body mass index, kg/m2 | 1.04 (1.00-1.07) | 0.048 | 1.04 (0.99-1.08) | 0.066 |
| Chronic renal insufficiency, eGFR <60 mL/min/1.73 m2 | 1.04 (1.0-1.07) | 0.048 | 1.46 (0.96-2.22) | 0.08 |
| Atrial fibrillation | 1.46 (1.01-2.13) | 0.047 | 1.58 (1.03-2.43) | 0.035 |
| Prior PPI | 2.26 (1.35-3.78) | 0.002 | 2.69 (1.5-4.81) | 0.001 |
| Pmean, mmHg | 1.01 (0.99-1.02) | 0.069 | 1.01 (0.99-1.02) | 0.431 |
| Aortic valve area, mm2 | 0.24 (0.08-0.70) | 0.009 | 0.38 (0.09-1.59) | 0.186 |
| Epicardial fat tissue, cm3 | 1.0 (0.99-1.0) | 0.943 | - | - |
| Epicardial fat tissue low/high† | 0.94 (0.64-1.37) | 0.745 | - | - |
Abbreviations: CI, confidence interval; eGFR, estimated glomerular filtration rate; MAC, mitral annular calcification; OR, odds ratio; Pmean, mean transvalvular aortic pressure; PPI, permanent pacemaker implantation.
Bold and italics denote a p-value below 0.05.
Above/below the median of 131 cm3.
Discussion
The findings of the present study showed that 1) MAC of any degree was present in almost 3-quarters of patients with severe AS undergoing TAVI, and severe MAC was found in 15.3% of patients; 2) patients with moderate/severe MAC did not have more EFT compared with patients with no or mild MAC as measured via the Agatston score; 3) patients with increased MAC were predominantly female, were more obese, suffered more often from CKD and atrial fibrillation, and had higher rates of preexisting PPI; 4) female sex, atrial fibrillation, and preexisting PPI, but not increased EFT, were independent predictors of MAC.
Studies investigating an association between MAC and EFT are scarce. In fact, to our knowledge, only 3 studies have analyzed this relationship, but the patient cohorts and the methods that were used vary extensively. In the study by Analbesi et al.,22 EFT thickness was measured at 4 different locations on noncontrast CT scans from a total of 297 patients, and its correlation with MAC, aortic valve calcium, and the sum of the 2 was investigated. The authors found that, in particular, at the site of the superior interventricular groove and the left atrioventricular groove, EFT and MAC, quantified via the Agatston score, were significantly associated, even after adjustment, albeit only weakly (Spearman correlation coefficient, p = 0.125 [superior interventricular groove] and 0.226 [left atrioventricular groove]).22 A direct comparison of these findings with ours is not possible since the authors used the thickness instead of the volume of EFT, which is potentially less precise as it reflects only a localized and not a global aspect of EFT.
Although a positive association between EFT amount and the presence and extent of coronary artery disease (CAD) has been confirmed in many trials,23, 24, 25 studies yield conflicting results concerning the relationship between EFT and coronary artery calcium,26, 27, 28, 29 showing either a positive association or no association. However, it is worth mentioning that the quantification methods of EFT measurement were different across existing studies (e.g., measurement of total EFT volume or pericoronary EFT) as were patient subsets (e.g., either healthy subjects, diabetics, or patients with suspicion of CAD).
As local proinflammatory and proatherogenic factors of the EFT released by paracrine and/or vasocrine mechanisms into the adjacent tissue are considered to be the main pathway for the association of EFT and coronary atherosclerosis and calcification,30,31 the hypothesis that EFT has similar effects on the mitral valve annulus is conceivable. Two older studies investigated the relationship between MAC and the presence of coronary atherosclerosis and found MAC to be an independent predictor of CAD.7,32 Already in 1986, William Roberts postulated that MAC and aortic valve calcium are in fact a “form of atherosclerosis.”7 However, the role of inflammation as the potential basis for the development of MAC is still not clear, since studies investigating an association between different biomarkers (among them also inflammatory markers like interleukin-6, C-reactive proteine, and soluble cluster of differentiation-14) and MAC failed to show an elevation in affected patients.33,34 Obviously, the fact that we found no association between EFT and MAC does not exclude the existence of an association between MAC and coronary atherosclerosis per se. However, our findings indicate that EFT might not play the same role in the development of MAC as it does for the evolution of CAD. It should be mentioned that in the latter case, despite a large number of studies that found a positive association between EFT and the extent of CAD and CAD progression, a causal role of EFT has yet to be proven.
Our results contrast with the findings of Argan et al.35 and Guler et al.36 The former group found EFT to be an independent predictor of severe MAC in a cohort of 102 patients with MAC and 107 patients without MAC.35 However, the groups were not matched in terms of demographic and cardiovascular risk comorbidities. Additionally, the methods that the authors used were different from ours and less objective: EFT thickness and MAC were assessed via transthoracic echocardiography, a method that is clearly more patient-dependent since the validity of its parameters depends on the patient’s body constitution. Guler et al. also found that EFT was significantly higher in 78 patients with MAC compared with 47 matched controls (5.7 ± 0.9 vs. 4.4 ± 0.9 mm; p < 0.001). However, as in the study by Argan et al., they used echocardiography as a method for EFT and MAC assessment with the limitations mentioned above.36 In both studies, the sample size was rather small compared to ours.
Our finding that patients with severe aortic stenosis and moderate/severe MAC were more often female and had more prevalent CKD concurs with results of other studies.2,37,38 Abramowitz et al.2 investigated the prevalence, characteristics, and outcomes of patients with concomitant MAC and high-grade aortic stenosis undergoing TAVI. The study group also found that patients with MAC had higher mean transvalvular gradients and higher EFT than patients without MAC. Similar to our study, no difference in the presence of mitral regurgitation (any grade) was found between the 2 groups. However, they found a lower prevalence of any degree of MAC in their aortic stenosis cohort (49.3% compared with 72.6% in ours). Nevertheless, the rates of severe calcification of the mitral annulus were comparable between the studies (9.5 vs. 15.3%, respectively), and the main difference was the rate of mild MAC (30.4 vs. 44.2%). One reason for this might be that in our study, patients with very small areas of MAC were also assigned to the “mild MAC” group and not to the “no MAC” group. We speculate that these patients would have been labeled as “no MAC” in the other studies.
Severe MAC was shown to be an independent predictor of new PPI in the Abramowitz et al. study, in contrast to our findings. We found preexisting PPI to be an independent predictor of moderate/severe MAC, but the need for new PPI was not associated with increased MAC. As no information was given on the rates of preexisting right bundle branch block, atrioventricular block, and/or implantation depth of the valve prosthesis in that study, it is difficult to discuss the reasons for this difference. Nevertheless, in both cases—preexisting or new PPI—patients with elevated MAC seem to develop conduction abnormalities more often than patients with lower amounts of MAC. In a subanalysis of the indications for prior PPI in our study cohort, far more patients of the moderate/severe MAC group had high-degree atrioventricular block as an indication for PPI compared with the no/mild MAC group (54.8 vs. 38.5%). Conversely, other indications for prior PPI, such as a sick sinus node, were more frequent in the no/mild MAC group (30.8 vs. 16.1%). These findings support the suggestion that increased MAC is associated with an impairment of the cardiac conduction system. It remains to be elucidated, however, whether calcification at the site of the mitral valve apparatus affects the conduction system per se. We also found atrial fibrillation to be an independent predictor of MAC. This finding is in line with an analysis of an over 1100-patient cohort of the Framingham Heart study39: after multivariable adjustment, the authors found MAC to be strongly associated with atrial fibrillation (hazard ratio: 1.6; 95% CI: 1.1-2.2).39 Another analysis by O’Neal et al.40 of >6000 patients of diverse ethnicity also confirmed this finding in a multivariable model that showed a strong association between MAC and atrial fibrillation (hazard ratio: 1.9; 95% CI: 1.5-2.5). The reasons for this association are most likely multifactorial, but 1 major reason might be the enlargement of the left atrium typically found in patients with MAC, which constitutes a strong trigger for the development of atrial fibrillation.39 Consistent with the findings of several other studies,4,41,42 higher MAC was more frequent in female patients than in males in our analysis. The reasons for this finding are not clear, but 1 suggestion is that due to the increased bone loss in postmenopausal women, more ectopic calcium deposits accumulate at the site of the mitral annulus, leading to higher amounts of MAC.43
Limitations
This is a retrospective analysis associated with the usual limitations and risk of selection bias. As it is a single-center study, the external validity of our results is low. Additionally, differences in MAC quantification cannot be ruled out when other CT protocols are used, as differences in reconstruction parameters might have an impact on the visibility of MAC. As our analysis was only performed in patients with high-grade aortic stenosis, the results cannot be generalized to different subsets of patients. The amount of calcification in the coronary arteries as measured via the Agatston score was not available, since this is not a routine assessment in the analysis of pre-TAVI CT scans. However, the association between calcification of the coronary arteries and MAC and EFT was not the focus of this study.
Conclusion
Increased EFT was not associated with elevated MAC in patients with high-grade aortic stenosis undergoing TAVI. Hence, the unfavorable inflammatory properties of EFT do not seem to play a role in MAC development. Further large-scale studies are needed to confirm these results.
CRediT Authorship Contributions
Maren Weferling, Julia Treiber, Christoph Liebetrau, Andreas Rolf, Yeong-Hoon Choi, Efstratios I. Charitos, Damini Dey, Samuel Sossalla, Won-Keun Kim: Substantial contributions to the conception or design of the work; or the acquisition, analysis, or interpretation of data for the work; drafting the work or revising it critically for important intellectual content; final approval of the version to be published; and agreement to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved.
Ethics Statement
The study was conducted in adherence to the Declaration of Helsinki and was approved by the Ethics Committee of the Justus-Liebig University of Giessen, Giessen, Germany. The need for informed patients’ consent was waived due to the retrospective design of the study.
Funding
No funding was received for conducting this study.
Disclosure Statement
M. Weferling is a speaker honoraria from Boston Scientific, Shockwave Medical, Sahajanand Medical Technologies Pvt Ltd, and Bristol Myers Squibb. S. Sossalla received scientific funding from Novartis and speaker/consulting honoraria from AstraZeneca, Novartis, Berlin-Chemie, Bristol-Myers-Squibb, and Boehringer Ingelheim. W.-K. Kim is a speaker and consulting/advisory honoraria from Abbott, Boston Scientific, and Meril LifeSciences, and speaker honoraria from Edwards Lifesciences and JenaValve.
The other authors had no conflicts to declare.
Acknowledgments
We thank Elizabeth Martinson, PhD, of the Kerckhoff Heart Research Institute (KHFI) for her excellent editorial assistance.
References
- 1.Nestico P.F., Depace N.L., Morganroth J., Kotler M.N., Ross J. Mitral annular calcification: clinical, pathophysiology, and echocardiographic review. Am Heart J. 1984;107(5 Pt 1):989–996. doi: 10.1016/0002-8703(84)90840-8. [DOI] [PubMed] [Google Scholar]
- 2.Abramowitz Y., Kazuno Y., Chakravarty T., et al. Concomitant mitral annular calcification and severe aortic stenosis: prevalence, characteristics and outcome following transcatheter aortic valve replacement. Eur Heart J. 2017;38(16):1194–1203. doi: 10.1093/eurheartj/ehw594. [DOI] [PubMed] [Google Scholar]
- 3.Abramowitz Y.J.H., Chakravarty T., Mack M.J., Makkar R.R. Mitral annulus calcification. J Am Coll Cardiol. 2015;66:1934–1941. doi: 10.1016/j.jacc.2015.08.872. [DOI] [PubMed] [Google Scholar]
- 4.Kanjanauthai S., Nasir K., Katz R., et al. Relationships of mitral annular calcification to cardiovascular risk factors: the Multi-Ethnic Study of Atherosclerosis (MESA) Atherosclerosis. 2010;213(2):558–562. doi: 10.1016/j.atherosclerosis.2010.08.072. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Asselbergs F.W., Mozaffarian D., Katz R., et al. Association of renal function with cardiac calcifications in older adults: the cardiovascular health study. Nephrol Dial Transplant. 2009;24(3):834–840. doi: 10.1093/ndt/gfn544. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Fox C.S., Vasan R.S., Parise H., et al. Mitral annular calcification predicts cardiovascular morbidity and mortality: the Framingham Heart Study. Circulation. 2003;107(11):1492–1496. doi: 10.1161/01.cir.0000058168.26163.bc. [DOI] [PubMed] [Google Scholar]
- 7.Roberts W.C. The senile cardiac calcification syndrome. Am J Cardiol. 1986;58(6):572–574. doi: 10.1016/0002-9149(86)90045-7. [DOI] [PubMed] [Google Scholar]
- 8.Iacobellis G., Willens H.J. Echocardiographic epicardial fat: a review of research and clinical applications. J Am Soc Echocardiogr. 2009;22(12):1311–1319. doi: 10.1016/j.echo.2009.10.013. [DOI] [PubMed] [Google Scholar]
- 9.Rabkin S.W. Epicardial fat: properties, function and relationship to obesity. Obes Rev. 2007;8(3):253–261. doi: 10.1111/j.1467-789X.2006.00293.x. [DOI] [PubMed] [Google Scholar]
- 10.Mazurek T., Zhang L., Zalewski A., et al. Human epicardial adipose tissue is a source of inflammatory mediators. Circulation. 2003;108(20):2460–2466. doi: 10.1161/01.CIR.0000099542.57313.C5. [DOI] [PubMed] [Google Scholar]
- 11.Villasante Fricke A.C., Iacobellis G. Epicardial adipose tissue: clinical biomarker of Cardio-Metabolic risk. Int J Mol Sci. 2019;20(23):5989. doi: 10.3390/ijms20235989. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Rosito G.A., Massaro J.M., Hoffmann U., et al. Pericardial fat, visceral abdominal fat, cardiovascular disease risk factors, and vascular calcification in a community-based sample: the Framingham Heart Study. Circulation. 2008;117(5):605–613. doi: 10.1161/CIRCULATIONAHA.107.743062. [DOI] [PubMed] [Google Scholar]
- 13.Kim B.J., Kang J.G., Lee S.H., et al. Relationship of echocardiographic epicardial fat thickness and epicardial fat volume by computed tomography with coronary artery calcification: data from the CAESAR study. Arch Med Res. 2017;48(4):352–359. doi: 10.1016/j.arcmed.2017.06.010. [DOI] [PubMed] [Google Scholar]
- 14.Abe M., Morimoto T., Akao M., et al. Relation of contrast-induced nephropathy to long-term mortality after percutaneous coronary intervention. Am J Cardiol. 2014;114(3):362–368. doi: 10.1016/j.amjcard.2014.05.009. [DOI] [PubMed] [Google Scholar]
- 15.Weferling M., Rolf A., Fischer-Rasokat U., et al. Epicardial fat volume is associated with preexisting atrioventricular conduction abnormalities and increased pacemaker implantation rate in patients undergoing transcatheter aortic valve implantation. Int J Cardiovasc Imaging. 2022;38(6):1399–1406. doi: 10.1007/s10554-021-02502-x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Joint Task Force on the Management of Valvular Heart Disease of the European Society of Cardiology (ESC), European Association for Cardio-Thoracic Surgery (EACTS), Vahanian A., et al. Guidelines on the management of valvular heart disease (version 2012) Eur Heart J. 2012;33(19):2451–2496. doi: 10.1093/eurheartj/ehs109. [DOI] [PubMed] [Google Scholar]
- 17.Kim W.K., Renker M., Rolf A., et al. Annular versus supra-annular sizing for TAVI in bicuspid aortic valve stenosis. EuroIntervention. 2019;15(3):e231–e238. doi: 10.4244/EIJ-D-19-00236. [DOI] [PubMed] [Google Scholar]
- 18.Amat-Santos I.J., Revilla A., Lopez J., et al. Value of CT in patients undergoing self-expandable TAVR to assess outcomes of concomitant mitral regurgitation. JACC Cardiovasc Imaging. 2015;8(2):226–227. doi: 10.1016/j.jcmg.2014.05.019. [DOI] [PubMed] [Google Scholar]
- 19.Commandeur F., Goeller M., Razipour A., et al. Fully automated CT quantification of epicardial adipose tissue by deep learning: a multicenter study. Radiol Artif Intell. 2019;1(6) doi: 10.1148/ryai.2019190045. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Dey D., Wong N.D., Tamarappoo B., et al. Computer-aided non-contrast CT-based quantification of pericardial and thoracic fat and their associations with coronary calcium and Metabolic Syndrome. Atherosclerosis. 2010;209(1):136–141. doi: 10.1016/j.atherosclerosis.2009.08.032. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Koo T.K., Li M.Y. A guideline of selecting and reporting intraclass correlation coefficients for reliability research. J Chiropr Med. 2016;15(2):155–163. doi: 10.1016/j.jcm.2016.02.012. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Alnabelsi T.S., Alhamshari Y., Mulki R.H., et al. Relation between epicardial adipose and aortic valve and mitral annular calcium determined by computed tomography in subjects aged >/=65 years. Am J Cardiol. 2016;118(7):1088–1093. doi: 10.1016/j.amjcard.2016.07.012. [DOI] [PubMed] [Google Scholar]
- 23.Mancio J., Azevedo D., Saraiva F., et al. Epicardial adipose tissue volume assessed by computed tomography and coronary artery disease: a systematic review and meta-analysis. Eur Heart J Cardiovasc Imaging. 2018;19(5):490–497. doi: 10.1093/ehjci/jex314. [DOI] [PubMed] [Google Scholar]
- 24.Silaghi A., Piercecchi-Marti M.D., Grino M., et al. Epicardial adipose tissue extent: relationship with age, body fat distribution, and coronaropathy. Obesity (Silver Spring) 2008;16(11):2424–2430. doi: 10.1038/oby.2008.379. [DOI] [PubMed] [Google Scholar]
- 25.Wu F.Z., Chou K.J., Huang Y.L., Wu M.T. The relation of location-specific epicardial adipose tissue thickness and obstructive coronary artery disease: systemic review and meta-analysis of observational studies. BMC Cardiovasc Disord. 2014;14:62. doi: 10.1186/1471-2261-14-62. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Ahmadi N., Nabavi V., Yang E., et al. Increased epicardial, pericardial, and subcutaneous adipose tissue is associated with the presence and severity of coronary artery calcium. Acad Radiol. 2010;17(12):1518–1524. doi: 10.1016/j.acra.2010.08.017. [DOI] [PubMed] [Google Scholar]
- 27.McClain J., Hsu F., Brown E., et al. Pericardial adipose tissue and coronary artery calcification in the Multi-ethnic Study of Atherosclerosis (MESA) Obesity (Silver Spring) 2013;21(5):1056–1063. doi: 10.1002/oby.20090. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Gorter P.M., de Vos A.M., van der Graaf Y., et al. Relation of epicardial and pericoronary fat to coronary atherosclerosis and coronary artery calcium in patients undergoing coronary angiography. Am J Cardiol. 2008;102(4):380–385. doi: 10.1016/j.amjcard.2008.04.002. [DOI] [PubMed] [Google Scholar]
- 29.Cosson E., Nguyen M.T., Rezgani I., et al. Epicardial adipose tissue volume and coronary calcification among people living with diabetes: a cross-sectional study. Cardiovasc Diabetol. 2021;20(1):35. doi: 10.1186/s12933-021-01225-6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Alexopoulos N.K.D., Raggi P. Visceral adipose tissue as a source of inflammation and promoter of atherosclerosis. Atherosclerosis. 2014;233:104–112. doi: 10.1016/j.atherosclerosis.2013.12.023. [DOI] [PubMed] [Google Scholar]
- 31.Lee H.Y., Després J.P., Koh K.K. Perivascular adipose tissue in the pathogenesis of cardiovascular disease. Atherosclerosis. 2013;230:177–184. doi: 10.1016/j.atherosclerosis.2013.07.037. [DOI] [PubMed] [Google Scholar]
- 32.Atar S., Jeon D.S., Luo H., Siegel R.J. Mitral annular calcification: a marker of severe coronary artery disease in patients under 65 years old. Heart. 2003;89(2):161–164. doi: 10.1136/heart.89.2.161. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Bortnick A.E., Bartz T.M., Ix J.H., et al. Association of inflammatory, lipid and mineral markers with cardiac calcification in older adults. Heart. 2016;102(22):1826–1834. doi: 10.1136/heartjnl-2016-309404. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Rezaeian P., Miller P.E., Haberlen S.A., et al. Extra-coronary calcification (aortic valve calcification, mitral annular calcification, aortic valve ring calcification and thoracic aortic calcification) in HIV seropositive and seronegative men: multicenter AIDS Cohort Study. J Cardiovasc Comput Tomogr. 2016;10(3):229–236. doi: 10.1016/j.jcct.2016.02.002. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Argan O., Avci E., Safak O., Yildirim T. Epicardial adipose tissue predicts severe mitral annular calcification in patients aged >/=60 years. Med Sci Monit. 2020;26 doi: 10.12659/MSM.921553. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.Guler S., Varol E. The relation between echocardiographic epicardial fat thickness and mitral annular calcification. Afr Health Sci. 2019;19(1):1657–1664. doi: 10.4314/ahs.v19i1.41. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37.Marwick T.H., Amann K., Bangalore S., et al. Chronic kidney disease and valvular heart disease: conclusions from a kidney disease: improving global outcomes (KDIGO) controversies conference. Kidney Int. 2019;96(4):836–849. doi: 10.1016/j.kint.2019.06.025. [DOI] [PubMed] [Google Scholar]
- 38.Deslandes M., Paquin A., Guzzetti E., et al. Sex-specific correlates of valvular and arterial calcification burden in patients with moderate aortic stenosis. Open Heart. 2022;9(2) doi: 10.1136/openhrt-2022-002139. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39.Fox C.S., Parise H., Vasan R.S., et al. Mitral annular calcification is a predictor for incident atrial fibrillation. Atherosclerosis. 2004;173(2):291–294. doi: 10.1016/j.atherosclerosis.2003.12.018. [DOI] [PubMed] [Google Scholar]
- 40.O'Neal W.T., Efird J.T., Nazarian S., Alonso A., Heckbert S.R., Soliman E.Z. Mitral annular calcification and incident atrial fibrillation in the Multi-Ethnic Study of Atherosclerosis. Europace. 2015;17(3):358–363. doi: 10.1093/europace/euu265. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 41.Boon A., Cheriex E., Lodder J., Kessels F. Cardiac valve calcification: characteristics of patients with calcification of the mitral annulus or aortic valve. Heart. 1997;78(5):472–474. doi: 10.1136/hrt.78.5.472. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 42.Elmariah S., Budoff M.J., Delaney J.A., et al. Risk factors associated with the incidence and progression of mitral annulus calcification: the multi-ethnic study of atherosclerosis. Am Heart J. 2013;166(5):904–912. doi: 10.1016/j.ahj.2013.08.015. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 43.Roberts W.C. Morphologic features of the normal and abnormal mitral valve. Am J Cardiol. 1983;51(6):1005–1028. doi: 10.1016/s0002-9149(83)80181-7. [DOI] [PubMed] [Google Scholar]