| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
(Circulation. 2003;107:1492.)
© 2003 American Heart Association, Inc.
Clinical Investigation and Reports |
From the National Heart, Lung, and Blood Institutes Framingham Heart Study (all authors), Framingham, Mass; the Cardiology Division, Department of Medicine, Massachusetts General Hospital, Boston, Mass (C.J.O.); the Cardiology Section, Department of Medicine and Preventive Medicine Section, Boston University School of Medicine, Boston, Mass (R.S.V., D.L., E.J.B.); the Department of Mathematics, Boston University, Boston, Mass (H.P., R.D.); and the National Heart, Lung, and Blood Institute, National Institutes of Health, Bethesda, Md (C.S.F., D.L., C.J.O.).
Correspondence to Emelia J. Benjamin, MD, ScM, Framingham Heart Study, 73 Mt Wayte Ave, Suite #2, Framingham, MA 01702-5827. E-mail emelia{at}fram.nhlbi.nih.gov
| Abstract |
|---|
|
|
|---|
Methods and Results We examined the association between MAC assessed by M-mode echocardiography and the incidence of CVD, CVD death, and all-cause death over 16 years of follow-up in the Framingham Heart Study subjects who attended a routine examination between 1979 and 1981. Cox proportional hazards models were used to estimate hazard ratios (HRs) associated with the presence of MAC for each outcome. Of 1197(445 male, 752 female) subjects who had adequate echocardiographic assessment, 14% had MAC. There were 307 incident CVD events and 621 deaths. In multivariable adjusted analyses, MAC was associated with an increased risk of incident CVD (HR, 1.5; 95% CI, 1.1, 2.0), CVD death (HR, 1.6; 95% CI, 1.1, 2.3), and all-cause death (HR, 1.3; 95% CI, 1.04, 1.6). For each 1-mm increase in MAC, the risk of incident CVD, CVD death, and all-cause death increased by
10%.
Conclusions The independent association of MAC with incident CVD and CVD death underscores that cardiac calcification is a marker of increased CVD risk.
Key Words: calcium echocardiography cardiovascular diseases mortality
| Introduction |
|---|
|
|
|---|
Subclinical cardiovascular disease (CVD) measurements have been demonstrated to have the same risk factors as clinically overt CVD14 and to be predictive of incident CVD events.1518 Other valvular calcification measurements, including aortic sclerosis and aortic stenosis, have been shown to be manifestations of subclinical CVD by virtue of their association with CVD risk factors19 and with an increased risk of CVD events, including coronary artery disease,8 stroke,20 and CVD death.21
Because risk factors for MAC are similar to risk factors for CVD, including age, hypertension, hyperlipidemia, diabetes, and obesity,2225 we hypothesized that MAC also may be a manifestation of subclinical atherosclerotic CVD. Thus, we sought to examine the association of MAC with incident CVD events in the community-based Framingham Heart Study sample. We hypothesized that the risk of incident CVD events would be higher in subjects with MAC.
| Methods |
|---|
|
|
|---|
Echocardiography was performed with 2-dimensional guidance through a parasternal window. M-mode recordings were made with a Hoffrel 201 ultrasound machine with an Aerotech 2.25-MHz transducer and a Jason thermographic printer. Echocardiographers reinterpreted M-mode echocardiograms, blinded to all clinical information about the subjects, to determine the presence and severity of MAC as defined previously.13 Briefly, the disorder was considered present if an echodense band was visualized throughout systole and diastole, was distinguishable from the posterior mitral valve leaflet, and was located anterior and parallel to the posterior left ventricular wall. MAC was measured in millimeters by a cardiologist from the leading anterior to the trailing posterior edge at its greatest width during the cycle on at least 3 cardiac cycles.
Risk factors were characterized at the index Framingham Heart Study clinic examination. Details about the methods of risk factor measurement and laboratory analysis have been described.28 Diabetes was defined as a fasting glucose level
140 mg/dL, a random nonfasting glucose level
200 mg/dL, or use of insulin or oral hypoglycemic agents. Because only M-mode echocardiography was available at the index examination, baseline valvular disease was defined as any diastolic murmur, or a systolic murmur >2/6 on the physical examination performed by the Heart Study physician. Covariates were obtained at the index examination except for smoking, HDL cholesterol, and serum creatinine, which were ascertained at the prior examination.
Subjects were followed for up to 16 years for CVD events. A panel of 3 experienced investigators reviewed all CVD events, blinded to MAC status.28 Incident CVD events included myocardial infarction, coronary insufficiency, congestive heart failure, and nonhemorrhagic stroke; fatal CVD additionally included sudden cardiac death defined by previously reported criteria.
Statistical Analyses
For principal analyses, MAC was dichotomized (present/absent); in secondary analyses, MAC was examined as a continuous variable. Differences in baseline risk factors in subjects with and without MAC were tested by ANOVA or logistic regression after adjustment for age and sex. All incidence rates were age and sex adjusted and presented as events per 10 000 person-years of follow-up. Cox proportional hazards regression models were used to estimate the association of MAC with the risks of myocardial infarction, congestive heart failure, incident CVD events, CVD death, and all-cause death. Because the interaction between MAC and sex was not significant, all analyses were performed for pooled sexes, with adjustment for sex. All multivariable models were adjusted for age, sex, systolic blood pressure, hypertension treatment, diabetes, total/HDL cholesterol, body mass index (BMI), and ECG left ventricular hypertrophy with strain. Additional covariates were outcome specific according to prior literature and were as follows:
| Results |
|---|
|
|
|---|
|
Overall, there were 307 cases of incident CVD, with an age- and sex-adjusted incidence rate of 554 per 10 000 person-years in subjects with MAC compared with 268 in 10 000 in subjects without MAC (Table 2). The Figure shows the age- and sex-adjusted cumulative incidence of CVD; the corresponding hazard ratio (HR) was 1.7 (Table 3). After multivariable adjustment, subjects with MAC were 50% more likely to have CVD in follow-up (HR, 1.5; 95% CI, 1.1, 2.0).
|
|
|
Overall, there were 621 deaths; 213 were attributed to CVD. The CVD mortality rate was 428 per 10 000 compared with 162 in 10 000 in subjects with and without MAC, respectively. The cumulative incidence of CVD death is displayed in the Figure. The age- and sex-adjusted HR for CVD death was 2.0, which remained significant after multivariable adjustment for baseline risk factors (HR, 1.6) and further adjustment for interim myocardial infarction and congestive heart failure (HR, 1.5; 95% CI, 1.1, 2.2).
The all-cause mortality rate among subjects with MAC was 847 per 10 000 person-years, compared with 443 per 10 000 person-years in subjects without MAC; the Cox proportional hazards data are presented in the Figure. In age- and- sex-adjusted analyses, the HR was 1.5. After multivariable adjustment, the increased risk of all-cause death remained statistically significant (HR, 1.3). In an exploratory analysis adjusting for interim myocardial infarction or congestive heart failure, these results were essentially unchanged (HR, 1.3; 95% CI, 1.03, 1.6; Table 3).
In secondary analyses, MAC was analyzed as a continuous variable. For each 1-mm increase in MAC, the risk of incident CVD, CVD death, and all-cause death adjusted for relevant baseline risk factors increased by 9% (P<0.007), 12% (P<0.004) and 9% (P<0.001), respectively.
There were 138 cases of myocardial infarction and 191 cases of congestive heart failure in follow-up. In age- and sex-adjusted models, MAC was significantly related to incident myocardial infarction (HR, 1.7; P<0.01) and heart failure (HR, 1.9; P<0.001). After further adjustment for baseline covariates, MAC was not an independent risk factor for myocardial infarction (HR, 1.4; 95% CI, 0.8, 2.2) or heart failure (HR, 1.3; 95% CI, 0.9, 1.8). Given the small number of cases for these outcomes, we only had power of 30% and 12%, respectively, to detect HRs of 1.4 and 1.3 at an
level of 0.05 in fully adjusted dichotomous models. Of interest, when we analyzed MAC as a continuous variable in fully adjusted models, MAC was not associated with myocardial infarction but was associated with a significantly increased risk of heart failure. For each 1-mm increase in MAC, the HR for heart failure was 1.1 (95% CI, 1.03, 1.2; P<0.01). Thus, our lack of significant association between the presence or absence of MAC and myocardial infarction and congestive heart failure should be interpreted with caution.
| Discussion |
|---|
|
|
|---|
Possible Mechanisms
Increasingly, it is recognized that vascular calcification, assessed by multiple modalities, including electron beam CT,29,30 echocardiography,21,25 and aortic arch calcification,32 is associated with CVD risk factors. Recent data also suggest an association between vascular calcification and incident CVD events.3234 The pathophysiology of cardiac calcification is not known but may stem from metabolic causes. For example, it is known that coronary artery calcification is common in patients with end-stage kidney disease undergoing dialysis,35 and it has also been observed that MAC is common in patients with end-stage kidney disease.36 Thus, derangements in calcium-phosphorous metabolism may contribute to MAC.
The association between vascular calcification, specifically MAC, and increased cardiovascular risk may also be due to the burden of shared risk factors, including age, hypertension, hyperlipidemia, diabetes, and obesity.2225 Furthermore, MAC may function as a bioassay for longitudinal exposure to risk factors. For example, MAC may more accurately "measure" the impact of longitudinal blood pressure elevation than a blood pressure reading at one examination. MAC may also be a marker for atherosclerotic disease burden; the association between MAC and aortic plaque37,38 and coronary calcification6 has been previously noted. It is also possible to speculate that other unmeasured factors such as metabolic, inflammatory, and hemostatic risk factors might be associated and may account for relations of MAC with CVD end points.
Study Limitations and Strengths
Some limitations of our study warrant attention. The Framingham Heart Study sample is neither ethnically diverse nor nationally representative. However, the coronary heart disease risk factor relations from Framingham have recently been validated in 6 ethnically and geographically diverse cohorts and were found to be applicable in other populations.39 Second, our sample was elderly, and our findings may not be generalizable to younger individuals. In addition, to have sufficient numbers of events longitudinally to provide adequate statistical power for analyses, we relied on the echocardiographic technology available at the baseline examination, M-mode, which is less sensitive for the detection of MAC. Finally, as noted above, our study lacked sufficient power to adequately examine the relation of MAC to many clinically important CVD outcomes such as endocarditis and heart failure.
There are several strengths of our study. Many prior studies of cardiovascular outcomes with MAC have been limited by sample sizes as small as 38 subjects,7,1012 retrospective12 or cross-sectional designs,6,7,25 nonuniform ascertainment of echocardiography,5,7,40,41 lack of recruitment of an appropriate control group,1012 and lack of routine ascertainment of covariate data and multivariable analyses.8,25,40
The present study is among the first large-scale community-based cohort studies to examine the prospective association between MAC and incident cardiovascular events and death. Our study is the first to show a graded relation between MAC severity and clinical outcome. To our knowledge, the only prior longitudinal cohort study to examine the relation of MAC to CVD outcomes was the Cardiovascular Health Study,9 which examined multiple echocardiographic variables including left ventricular mass, dimensions, and wall thickness and left atrial dimensions, in addition to MAC (dichotomously analyzed). The investigators observed that the presence of MAC was associated with an unadjusted increased hazard for incident stroke, heart failure, coronary heart disease, and all-cause death. However, after adjustment for CVD risk factors, MAC only remained predictive of incident coronary heart disease. Possible reasons for the disparate findings between the two studies may include the older mean age (73 years), the shorter duration of follow-up (7 years), and the utilization of 2-dimensional MAC assessment in the Cardiovascular Health Study. Conceivably, our reliance on the less-sensitive M-mode technology may have led to the detection of more severe degrees of MAC, with a greater risk of incident events.
Implications
In conclusion, MAC was independently associated with incident CVD, cardiovascular death, and all-cause death after adjusting for traditional CVD risk factors, suggesting that MAC is a marker of increased cardiovascular risk. We have shown that this increased risk occurs in a graded fashion by MAC severity. The mechanisms of increased risk are undoubtedly multifactorial and include mechanical characteristics of MAC, shared risk factors between MAC and cardiovascular outcomes, and valvular calcification as an overall marker of atherosclerotic burden. The association of calcification of the mitral annulus with increased CVD risk parallels the previously reported elevated cardiovascular risk observed with the presence of echocardiographic aortic sclerosis. The prognostic importance of the calcification phenotype in multiple cardiac locations underscores the importance of understanding the genetic and environmental determinants of cardiac calcification.
| Acknowledgments |
|---|
Received October 4, 2002; revision received December 19, 2002; accepted January 2, 2003.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
M. R. Akram, T. Chan, S. McAuliffe, and A. Chenzbraun Non-rheumatic annular mitral stenosis: prevalence and characteristics Eur J Echocardiogr, June 9, 2008; (2008) jen179v1. [Abstract] [Full Text] [PDF] |
||||
![]() |
N E MANGHAT, V RACHAPALLI, R V. LINGEN, A M VEITCH, C A ROOBOTTOM, and G J MORGAN-HUGHES Imaging the heart valves using ECG-gated 64-detector row cardiac CT Br. J. Radiol., April 1, 2008; 81(964): 275 - 290. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. N. Bella, W. Tang, A. Kraja, D. C. Rao, S. C. Hunt, M. B. Miller, V. Palmieri, M. J. Roman, D. W. Kitzman, A. Oberman, et al. Genome-Wide Linkage Mapping for Valve Calcification Susceptibility Loci in Hypertensive Sibships: The Hypertension Genetic Epidemiology Network Study Hypertension, March 1, 2007; 49(3): 453 - 460. [Abstract] [Full Text] [PDF] |
||||
![]() |
R. C. Johnson, J. A. Leopold, and J. Loscalzo Vascular Calcification: Pathobiological Mechanisms and Clinical Implications Circ. Res., November 10, 2006; 99(10): 1044 - 1059. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. A. Allison, P. Cheung, M. H. Criqui, R. D. Langer, and C. M. Wright Mitral and Aortic Annular Calcification Are Highly Associated With Systemic Calcified Atherosclerosis Circulation, February 14, 2006; 113(6): 861 - 866. [Abstract] [Full Text] [PDF] |
||||
![]() |
C. S. Fox, M. G. Larson, R. S. Vasan, C.-Y. Guo, H. Parise, D. Levy, E. P. Leip, C. J. O'Donnell, R. B. D'Agostino Sr., and E. J. Benjamin Cross-Sectional Association of Kidney Function with Valvular and Annular Calcification: The Framingham Heart Study J. Am. Soc. Nephrol., February 1, 2006; 17(2): 521 - 527. [Abstract] [Full Text] [PDF] |
||||
![]() |
E.A. de Vrey, A.J.H.A. Scholte, X.H. Krauss, R.A. Dion, D. Poldermans, E.E. van der Wall, and J.J. Bax Intracardiac pseudotumor caused by mitral annular calcification Eur J Echocardiogr, January 1, 2006; 7(1): 62 - 66. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. R. Kizer, D. O. Wiebers, J. P. Whisnant, J. M. Galloway, T. K. Welty, E. T. Lee, L. G. Best, H. E. Resnick, M. J. Roman, and R. B. Devereux Mitral Annular Calcification, Aortic Valve Sclerosis, and Incident Stroke in Adults Free of Clinical Cardiovascular Disease: The Strong Heart Study Stroke, December 1, 2005; 36(12): 2533 - 2537. [Abstract] [Full Text] [PDF] |
||||
![]() |
A K Nightingale and J D Horowitz Aortic sclerosis: not an innocent murmur but a marker of increased cardiovascular risk Heart, November 1, 2005; 91(11): 1389 - 1393. [Abstract] [Full Text] [PDF] |
||||
![]() |
R. Sharma, D. Pellerin, D. C. Gaze, H. Gregson, C. P. Streather, P. O. Collinson, and S. J. D. Brecker Dobutamine stress echocardiography and the resting but not exercise electrocardiograph predict severe coronary artery disease in renal transplant candidates Nephrol. Dial. Transplant., October 1, 2005; 20(10): 2207 - 2214. [Abstract] [Full Text] [PDF] |
||||
![]() |
B. D. Horne, N. J. Camp, J. B. Muhlestein, and L. A. Cannon-Albright Evidence for a Heritable Component in Death Resulting From Aortic and Mitral Valve Diseases Circulation, November 9, 2004; 110(19): 3143 - 3148. [Abstract] [Full Text] [PDF] |
||||
![]() |
C. M. Otto Why is aortic sclerosis associated with adverse clinical outcomes? J. Am. Coll. Cardiol., January 21, 2004; 43(2): 176 - 178. [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
Circulation Home | Subscriptions | Archives | Feedback | Authors | Help | AHA Journals Home | Search Copyright © 2003 American Heart Association, Inc. All rights reserved. Unauthorized use prohibited. |