EDITORIAL COMMENT
The Multi-Ethnic Study of Atherosclerosis (MESA) and Myocardial Function
Where Is the Boundary Between Risk Factor and Disease?*
Nathaniel Reichek, MD, FACC, FAHA*
Research and Education Department, St. Francis Hospital-The Heart Center, and the Division of Cardiology and Department of Biomedical Engineering, Stony Brook University, Stony Brook, New York
* Reprint requests and correspondence: Dr. Nathaniel Reichek, DeMatteis Center, Research & Education Foundation, St. Francis Hospital, 100 Port Washington Boulevard, Roslyn, New York 11568. (Email: Nathaniel.Reichek{at}chsli.org).
In this issue of the Journal, Rosen et al. (1) report the relationship between segmental myocardial function and cardiovascular risk factors in a multiethnic population with no symptoms or history of cardiovascular disease, drawn from the larger population in the Multi-Ethnic Study of Atherosclerosis (MESA). The results demonstrate apparent regional myocardial dysfunction in subjects with the cardiovascular risk factors of diastolic hypertension and/or smoking despite the absence of clinically manifest disease. These novel observations challenge the very structure of conventional clinical reasoning about the distinction between risk factor and disease. There are no precedents for such findings and, at first look, no established mechanisms that explain them.
The authors used cardiac magnetic resonance imaging tissue tagging (spatial modulation of magnetization [SPAMM] tagging with harmonic phase analysis) to examine segmental function in over 1,100 MESA subjects. While little known in the clinical community, magnetic resonance imaging tissue tagging, originally developed by Zerhouni et al. (2), and the SPAMM version thereof, developed by Axel et al. (3), remain our most effective methods for assessment of segmental myocardial contraction or, in engineering mechanics terms, myocardial deformation and rigid body motion. The harmonic phase technique has recently made analysis of the resultant tagged images automated, fast, and efficient, so that the method can be applied much more widely than in the past (4). These MESA results are the first fruits of this advance in data analysis and the largest study by far performed with tissue tagging. The principal variable evaluated was circumferential strain, the shortening of the myocardium in the circumferential direction, which is the orientation of most left ventricular myocardial fibers, particularly in the midwall. Circumferential strain was evaluated by vascular territory using short-axis images of myocardial segments attributed to the left anterior descending, circumflex, and right coronary perfusion territories. The authors found overall that circumferential strain was reduced in subjects with diastolic hypertension relative to normotensive subjects after adjustment for age, gender, calcium score, and other variables, but not in those with borderline systolic or isolated systolic hypertension. Similarly, current cigarette smoking was associated with reduced strain, and there was a dose–response relationship to pack-years smoked. Diabetes and lipid abnormalities had no demonstrable effects. Within the overall data, some ethnic differences were observed between non-Hispanic and Hispanic whites and African-American and Chinese subjects.
The authors are eager to relate their findings to coronary artery disease but are appropriately cautious in discussing possible explanations for these results, given the lack of mechanistic data. However, the reader also must ask what known mechanisms could account for these findings. Briefly, there are two possibilities: 1) an admixture of clinically unrecognized overt coronary disease in the population. In such subjects, segmental myocardial function might be reduced by the usual pathways of stunning, hibernation or infarction; and 2) direct effects of known early cardiovascular adaptations to hypertension and smoking. Clearly, an admixture in the population of a small number of subjects with clinically unrecognized coronary disease with resultant segmental dysfunction could have produced the results reported in pooled data. The authors note that only <2% of subjects had abnormal Q waves on electrocardiogram, and believe these are too few to account for the findings. In addition, they found no correlation between calcium score and circumferential strain reduction. However, abnormal Q waves are an insensitive marker for coronary stenoses, and calcium score reflects atherosclerosis burden, not stenosis per se. Thus, the possibility of an admixture of subjects with clinically significant myocardial disease cannot be excluded. A critical piece of evidence in this regard would be the distribution of circumferential strain values in the population by segment. If an admixture of unrecognized but overtly diseased myocardial segments explains the results, there should be more than one peak in the prevalence distribution across the range of circumferential strain values. Unfortunately, this information is not provided.
What about the alternative of direct risk factor effects on myocardial function? It has long been known from experimental models that adaptive myocardial changes to myocardial pressure overload begin early and result in changes in myocyte metabolism, calcium handling, and contractile protein performance, all of which lead to reduced fiber shortening. These alterations begin before the development of overt hypertrophy (5). Overt reduction of circumferential shortening in mild hypertensive left ventricular hypertrophy (by imaging, not electrocardiography) in human subjects without clinical cardiovascular disease who have normal ejection fraction has also been shown previously (6). Presumably mild hypertrophy offsets reduced intramyocardial function and keeps endocardial excursion and chamber ejection fraction normal in this setting. In the present study, adjustment for left ventricular mass reduced the association between blood pressure and circumferential strain, but we are not told how many subjects actually had left ventricular hypertrophy or concentric remodeling. The fact that diastolic hypertension rather than systolic was the closest correlate of reduced strain may simply reflect the fact that those with diastolic hypertension also had higher or more consistently elevated systolic pressures. Thus there is strong reason to suspect that early myocardial adaptation to pressure overload in hypertensives, with or without overt hypertrophy, accounts for this association between diastolic hypertension and reduced strain. It should also be noted that the population limits of normal left ventricular mass index are sufficiently wide so that early hypertrophy in an individual subject, with an increase in left ventricular mass in that individual, could occur with left ventricular mass remaining within overall normal population limits.
The findings with regard to smoking may also relate to known effects of smoking on endothelial and microvascular function even in the absence of overt clinical effects of atherosclerosis (7–10). These could result in mildly reduced myocardial perfusion and perfusion reserve. Evidence exists that even in normal myocardium at rest local myocardial function is closely matched to variations in perfusion (10). Thus, despite the absence of overt ischemia, mildly reduced regional function could result from reductions in perfusion that do not cause overt ischemia.
Well, which is the answer? Occult disease or risk factor effects? The ball is in the MESA investigators’ court. Hopefully, further examination of the data in this study, reporting of the myocardial stress perfusion magnetic resonance imaging studies also done in a MESA subset, and additional longitudinal follow-up data will provide the answer in due course.
 |
Footnotes
|
|---|
* Editorials published in the Journal of the American College of Cardiology reflect the views of the authors and do not necessarily represent the views of JACC or the American College of Cardiology 
|
References
|
|---|
1. Rosen BD, Saad MF, Shea S, et al. Hypertension and smoking are associated with reduced regional left ventricular function in asymptomatic individualsthe Multi-Ethnic Study of Atherosclerosis. J Am Coll Cardiol 2006;47:1150-1158.[Abstract/Free Full Text]2. Zerhouni EA, Parish DM, Rogers WJ, Yang A, Shapiro EP. Human hearttagging with MR imaging—a method for noninvasive assessment of myocardial motion. Radiology 1988;169:59-63.[Abstract/Free Full Text] 3. Axel L, Dougherty L. Heart wall motionimproved method of spatial modulation of magnetization for MR imaging. Radiology 1989;172:349-350.[Abstract/Free Full Text] 4. Osman NF, Kerwin WS, McVeigh ER, Prince JL. Cardiac motion tracking using CINE harmonic phase (HARP) magnetic resonance imaging Magn Reson Med 1999;42:1048-1060.[CrossRef][Web of Science][Medline] 5. Yamazaki T, Komuro I, Yazaki Y. Signalling pathways for cardiac hypertrophy Cell Signal 1998;10:693-698.[CrossRef][Web of Science][Medline] 6. Palmon LC, Reichek N, Yeon SB, et al. Intramural myocardial shortening in hypertensive left ventricular hypertrophy with normal pump function Circulation 1994;89:122-131.[Abstract/Free Full Text] 7. Wiesmann F, Petersen SE, Leeson PM, et al. Global impairment of brachial, carotid, and aortic vascular function in young smokersdirect quantification by high-resolution magnetic resonance imaging. J Am Coll Cardiol 2004;44:2056-2064.[Abstract/Free Full Text] 8. Meeder JG, Blanksma PK, Van der Wall EE, et al. Long-term cigarette smoking is associated with increased myocardial perfusion heterogeneity assessed by positron emission tomography Eur J Nucl Med 1996;23:1442-1447.[CrossRef][Web of Science][Medline] 9. Deanfield JE, Shea MJ, Wilson RA, Horlick P, deLandsheere CM, Selwyn AP. Direct effects of smoking on the heartsilent ischemic disturbances of coronary flow. Am J Cardiol 1986;57:1005-1009.[CrossRef][Web of Science][Medline] 10. Bassingthwaighte JB, Li Z. Heterogeneities in myocardial flow and metabolismexacerbation with abnormal excitation. Am J Cardiol 1999;83:7H-12H.[Web of Science][Medline]
|
Top
References