Author + information
- David S. Celermajer, MB, BS, PhD, FRACP1 ()
- ↵1Reprint requests and correspondence:
Dr. David S. Celermajer, Department of Cardiology, Royal Prince Alfred Hospital, Missenden Road, Camperdown NSW 2050, Australia.
We should be able to detect atherosclerosis early, before clinical events occur. After all, it is a disease process that begins in childhood (1), takes many decades to progress in most subjects, and affects an organ (the vasculature) for which we have multiple diagnostic modalities available.
Nevertheless, we have not yet established a reliable methodology for seeing clearly through this long presymptomatic window of atherosclerotic disease. This failure is probably because the pathogenesis of the disease is very complex, involving a plethora of contributing moieties (lipoproteins, a variety of cells, extracellular matrix, oxidants, cytokines and thrombomodulators, among others) and because much of the disease processes occur inside the vessel wall, rather than adjacent to or within the lumen. Furthermore, there is only a loose association between the burden of plaque and the risk of an event, with factors such as vascular function, plaque composition, plaque geometry and remodeling, inflammation, and collateralization all confounding the relationships among plaque size, luminal narrowing, and clinical events.
Preventive strategies for atherosclerosis often are divided rather simplistically into “primary” and “secondary” prevention, on the basis of the presence or absence of a prior clinical event. This dichotomy, however, does not emphasize identification of those asymptomatic individuals with particularly high risk nor target such subjects for intensive preventive therapy, that is, those who require “primary and a half” prevention. In this regard, novel blood tests and imaging modalities have been developed for these high-risk asymptomatic subjects that now might improve our diagnostic capabilities and therapeutic targeting strategies.
With such developments, preventive cardiology efforts may now move beyond the identification of epiphenomenal “risk factors,” to the measurement of processes more directly related to the pathology within the arterial wall (i.e., how the risk factors have impacted on atherogenesis in a particular individual).
A number of large prospective population-based studies have established the potential importance of measuring serum inflammatory markers as a means of identifying high-risk asymptomatic subjects. Of these, the high-sensitivity C-reactive protein (hsCRP) measurement (2) is the best studied to date. However, not all researchers agree on the magnitude of risk conferred by an elevated hsCRP level (3), and further data are required concerning day-to-day variability in individual subjects as well as defining the clinical benefits that might accrue from treatment and risk stratification according to hsCRP results. Nevertheless, this is an exciting area in preventive medicine. Encouraging work also is now appearing about inflammatory markers that might be more specifically related to atherogenic processes, such as lipoprotein-associated phospholipase A2, a lipoprotein-associated enzyme implicated in the vascular inflammatory pathway leading to plaque formation (4).
Coronary calcium scanning
Vascular imaging is an even more appealing concept for identification for high-risk subjects because it might provide information about specific location of vulnerable plaque(s) in addition to identifying vulnerable patients. The detection of coronary artery calcium with fluoroscopy has long been recognized as valuable in predicting the presence of obstructive coronary artery disease, and it has been more than a decade since computed tomography scanning for quantification of coronary calcium was popularized (5). This promising technique does appear to add extra diagnostic information over and above the measurement of traditional risk factors (6), and the absence of coronary calcium appears to have a high negative predictive value for subsequent events. Nevertheless, some questions remain about lower specificity than sensitivity, particularly in low-risk subjects, reproducibility of measurements with serial studies, and aspects related to radiation and cost-effectiveness.
By contrast, ultrasonography is a very appealing technique for studying atherosclerosis because it is noninvasive, relatively inexpensive, widely available, simple, and generally reproducible. Although the coronary arteries cannot be well visualized by noninvasive ultrasonography, early changes in peripheral arteries have been widely studied in the hope and expectation that this might provide useful insights into presymptomatic atherosclerosis, given the systemic nature of this disease. Furthermore, it became apparent more than a decade ago that ultrasonography was useful for the detection of both early structural changes in the artery walls (such as thickening or minor plaque), as well as early functional changes (such as loss of endothelium-dependent dilatation, which precedes plaque formation in high-risk subjects).
Thus, in 1992, we first described the technique of ultrasonography-based measurement of flow-mediated dilatation (FMD) in children and young adults at risk of atherosclerosis (7), which was later shown to be mediated predominantly by endothelial nitric oxide release in response to increased shear stress. This technique has been used by many research groups, who have documented a relationship between FMD and most cardiovascular risk factors in both low-risk and high-risk subjects. Recently, several centers have demonstrated a relationship between decreased brachial FMD and a higher incidence of cardiovascular events during follow-up (e.g., reference 8). Nevertheless, long-term prognostic data in a large population of asymptomatic subjects remain lacking.
At around the same time, Salonen et al. (9) reviewed the data concerning the measurement of intima-media thickness (IMT) of the common carotid artery as an accurate and reproducible measure of atherosclerosis. Carotid IMT has been shown to correlate well with traditional risk factors and burden of atheroma elsewhere in the body, and it also appears to be predictive of subsequent cardiovascular events in large population studies. As with FMD, there is some variation in the protocols used for measurement carotid IMT, but most investigators have shown a significant relationship between ultrasonography-based measurement of this parameter and subsequent risk of events. Carotid IMT has now been used extensively as a surrogate endpoint in disease reversibility trials.
FMD, IMT, and population screening
Three recent studies have investigated the relationship between ultrasound-based measurements for vascular health using FMD and/or IMT and subsequent risk, two in this issue of the Journal(10,11) and one in a recent issue of Circulation(12).
In the larger of the two population-based ultrasound studies, Juonala et al. (12) studied the interrelations between brachial endothelial function and carotid IMT in 2,109 healthy adults ages 24 to 39 years, documenting a very strong inverse association between FMD and IMT (p < 0.001), in a multivariate model adjusted for traditional risk factors. Furthermore, this inverse relationship was strongest in subjects with intermediate or impaired FMD, suggesting that very good FMD might protect the vasculature from the propensity of risk factors to lead to structural wall thickening. This study encourages the use of both FMD and IMT, in appropriately expert hands, for the identification of high-risk asymptomatic individuals.
In sharp contrast, the publication by the Firefighters And Their Endothelium (FATE) investigators in this issue of the Journal(10) found no significant relationship at all between carotid IMT and brachial artery FMD in 1,578 middle-aged men without known cardiovascular disease. Although possible interpretations of these data include inaccuracies in the measurement techniques used or a lack of validity of carotid IMT and/or brachial FMD as good measures of early vascular disease, the authors argue convincingly that carotid IMT and brachial artery FMD probably provide complementary pathophysiologic insights into early atherosclerosis. Nevertheless, it is worrisome for the potential applicability of both techniques that two large population-based studies by highly experienced investigators have produced such strikingly discordant results.
In the third of these studies, Witte et al. (11) postulate, on the basis of meta-analysis, that the association between FMD and cardiovascular risk may be limited to low-risk populations, in whom they document a significant association between these parameters. However, it should be noted that the median sample size of the meta-analyzed studies was only 20 subjects, that many of the smaller studies recruited highly selected individuals rather than population-based samples, and that the resulting analysis will necessarily reflect the findings of the largest studies included, as well as being potentially confounded by publication bias and the combination of results from studies which used very different methodologies for the measurement of FMD. Thus, such findings could best be regarded as hypothesis-generating, rather as a definitive indication that FMD is best suited to risk stratification in low-risk populations.
There is little doubt that the measurement of both FMD and IMT has provided an extremely valuable insight into the relationship between risk factors and early arterial disease, as well as important data concerning the effects of treatment on early atherogenic processes. Large population studies support the predictive value of IMT, as do the only prospective studies of FMD and event rates that have been published to date, although as yet these have only been very small; thus, the final results of the FATE investigation concerning the predictive value of FMD for clinical events are anxiously awaited. However, at the moment, the measurement of FMD must be regarded as a valuable clinical research tool but certainly not ready for “prime time,” for population screening, or clinical decision-making.
Therefore, the quest for “the Holy Grail” of identification of high-risk asymptomatic subjects continues. One feels, however, that the chalice is not far away, with blood tests and ultrasonography the most likely modalities to permit cost-effective identification of vulnerable patients and advanced imaging modalities more likely to allow localization of vulnerable plaques.
↵* Editorials published in the Journal of the American College of Cardiologyreflect the views of the authors and do not necessarily represent the views of JACCor the American College of Cardiology.
- American College of Cardiology Foundation
- Ridker P.M.
- Agatston A.S.,
- Janowitz W.R.,
- Hildner F.J.,
- et al.
- Gokce N.,
- Keaney J.F.,
- Hunter L.M.,
- et al.
- Yan R.T.,
- Anderson T.J.,
- Charbonneau F.,
- Title L.,
- Verma S.,
- Lonn E.
- Witte D.R.,
- Westerink J.,
- de Koning E.J.,
- van der Graaf Y.,
- Grabbee D.E.,
- Bots M.L.
- Juonala M.,
- Vikari J.S.A.,
- Laitinen T.,
- et al.