Minor electrocardiogram abnormalities in arterial hypertension

The appropriate use of non-invasive cardiac testing in asymptomatic persons is different from that in patients with symptoms suggestive of heart disease. In symptomatic patients, testing is used to confirm a suspected diagnosis and to estimate near term prognosis and to guide further diagnostic procedures and immediate treatment. By contrast, in asymptomatic patients, the emphasis is on the assessment of long-term risk and primary prevention of future clinical disease [1]. Although several multiple risk factor model have been suggested, their accuracy is considerably below 100%. Sensitivity and specificity of diagnostic preocedures highly depends on the characteristics of the population studied. A widely applicable and standardized screening procedure contributing to the assessment of cardiovascular risk is the electrocardiogram (ECG). Its diagnostic accuracy is improved by simultaneous exercise testing. A number of studies have shown that exercise-induced ischaemia in healthy individuals is associated with an increased risk of future major cardiac events (MACE) [1]. In addition to ST-segment changes, useful data may be derived in this diagnostic setting such as exercise capacity, heart rate recovery, presence or absence of dysrhythmias, and signs of left ventricular hypertrophy (LVH). However, only a few data are available on the accuracy of ECG recordings at rest in healthy individuals or those without known cardiovascular disease. Hypertensive individuals are prone to a variety of cardiac alterations, including development of (a)symmetrical LVH, interstitial and perivascular fibrosis with diastolic dysfunction, left ventricular enlargement with systolic dysfunction and potential dysrhythmias, endothelial dysfunction, stenoses of epicardial arteries and/or coronary microangiopathy. All these mechanisms may reduce an adequate ratio of coronary blood flow to oxygen demand of myocardium, causing angina, dyspnoea and major cardiac events. The increase in coronary vascular resistance may be attributed to functional and morphological alterations in the resistive arteries [2]. In addition, it is widely accepted that LVH and remodelling is an important independent risk factor for cardiovascular morbidity and mortality in hypertensive patients [3]. Antihypertensive therapy aimed at reducing blood pressure produces regression of ECG LVH, and regression of LVH appears to be associated with improved prognosis [4]. ECG may detect impairment of coronary conductance due to stenosis in epicardial arteries and/or diminished dilatory capacity of coronary microvasculature. Furthermore, LVH, an independent predictor of cardiovascular mortality, may be detected in ECG recordings. However, its sensitivity is far less compared to echocardiography, which additionally may provide significant information on the geometry of LVH. On the other hand, ECG is less costly and time consuming and can be derived by almost all physicians, supporting its usefulness as a screening tool in patients at increased cardiovascular risk. In addition to its lack of appropriate sensitivity its specificity might be questioned. In this issue of the journal, Schillaci et al. [5] address the problem of specificity of minor ECG abnormalities in patients with arterial hypertension with respect to its potential prognostic significance. The authors address a common clinical problem that faces many physicians in their daily clinical routine: do minor alterations in ST segment or t-wave in ECG matter in hypertensive patients? It is clearly demonstrated that minor repolarization changes in hypertensive patients with a previous unknown history of cardiac disease are associated with an increased likelihood of MACE during a follow-up period of 10 years. By contrast, the rate of cerebrovascular events was not affected. This prospective study exhibits a remarkable follow-up rate of > 99%, excluding major statistical errors. To increase the accuracy of ECG in detecting LVH not only the classical Sokolow–Lyon voltage index was used, but also the Cornell criteria were applied with sex-specific adjustment. Although the degree of gender-related adjustment is still a matter of debate [4], the Cornell criteria appear more sensitive than the Sokolow–Lyon index. Currently, it is widely accepted that hypertensive patients develop diastolic dysfunction and remodelling of left ventricular geometry with or without LVH very early in the time course of hypertensive heart disease; both of which have been shown to affect coronary conductance, and therefore might also affect ECG. In other words, a significant proportion of the patients in this study may have suffered from coronary conductance abnormalities not only on the level of epicardial arteries (the primary endpoint of the present study), but also because of dysfunction of the coronary microvasculature. Moreover, recently, it has been shown that impaired coronary flow reserve at the level of coronary resistive arteries may impair endothelial function of upstream arteries, which has prognostic significance [6]. In clinical practice, it should be remembered that assessing global cardiovascular risk of hypertensive patients into low, medium, high and very high classes is the primary goal. Global risk stratification follows the diagnostic routine procedures suggested by the 1999 WHO–International Society of Hypertension guidelines: medical history, physical examination and clinical blood pressure measurement; routine blood chemistry and urine analysis; and electrocardiogram. However, in a large cohort of hypertensive patients Cuspidi et al. [7] showed that this routine work-up may be too insensitive to detect patients belonging to the high global risk category. Thus, routine cardiac [8,9] and vascular ultrasonography [7] appear to be necessary to avoid this error [7]. Risk stratification may be complemented by measurement of microalbuminuria with redefined thresholds [10,11] and peripheral artery endothelial function, as recently shown to reflect the severity of disturbed circadian patterns of blood pressure [12]. However, the implementation of sonographic measurements [13] of carotid intima–media thickness [7], peripheral vascular compliance and endothelial dysfunction depends on more reliable and effective methods that make the accuracy and reproducibility more applicable for clinical routine [14,15]. The recent guidelines of the European Societies of Hypertension and Cardiology suggest the frequent inclusion of cardiac and vascular ultrasonography and microalbuminuria among the recommended examinations to search for target organ damage and to stratify total cardiovascular risk in hypertensive patients [16]. Apart from its effects on stroke, renal failure and peripheral artery disease, systemic arterial hypertension is a major risk factor for cardiac complications (e.g. coronary artery disease, heart failure and sudden cardiac death). Although trivial at first sight, the management of hypertension appears a major goal because hypertension is principally a treatable risk factor. However, as previously summarized, the percentage of hypertensives in the general population who are identified and receiving treatment, and whose arterial pressure is under control, remains unacceptable. Hypertension management thus becomes an important goal to reduce the incidence of hypertension-related cardiovascular complications and outcomes [17]. What are the consequences for clinical routine and daily practice? The important message of the study from the Perugian group is that the finding of minor repolarization changes in patients with so far uncomplicated arterial hypertension and LVH, and who are at increased cardiovascular risk, thus deserves further meticulous diagnostic procedures such as stress testing to unmask silent disturbances of coronary conductance, either at the level of epicardial arteries or in the coronary microvasculature due to intravascular and extravascular alterations.