Electrocardiographic criteria which have the best prognostic significance in hypertensive patients with echocardiographic hypertrophy of left ventricle: 15‐year prospective study

Abstract Background Electrocardiography is the first‐choice technique for detecting left ventricular hypertrophy in patients with arterial hypertension. It is necessary to know the probable outcome for every patient during the treatment, with the aim of improving cardiovascular event prevention. Hypothesis Certain electrocardiographic criteria for left ventricular hypertrophy may predict outcomes of patients with left ventricular hypertrophy during a 15‐year follow‐up. Methods Fifteen‐year prospective study of 83 consecutive patients (53 male and 30 female; mean age 55.3 ± 8.1) with echocardiographic left ventricular hypertrophy (left ventricular mass index 170.3 ± 31.6 g/m2). Electrocardiographic left ventricular hypertrophy was determined by means of Gubner‐Ungerleider voltage, Lewis voltage, voltage of R wave in aVL lead, Lyon‐Sokolow voltage, Cornell voltage and Cornell product, voltage RV6 and RV5 ratio, Romhilt‐Estes score, Framingham criterion and Perugia criterion. Results One or more composite events were registered in 32 (38.5%) patients during 15‐year follow‐up. Positive Lyon‐Sokolow score (17.6% vs. 47.3%; P < 0.05), Lewis voltage (9.8% vs. 21.9%; P < 0.05), Cornell voltage (15.7% vs. 37.5%; P < 0.05), and Cornell product (9.8% vs. 34.4%; P < 0.01) were more frequent in a group of patients with composite events. Odd ratio for Cornell product was 4.819 (95% CI 1.486‐15.627). Conclusion Patients with echocardiographic left ventricular hypertrophy who had positive Lewis voltage, Lyon‐Sokolow voltage, Cornell voltage, and Cornell product showed worse 15‐year outcome. The strongest predictor of cardiovascular events was positive result of Cornell product.


| INTRODUCTION
Left ventricular mass is measured by accurate techniques, such as echocardiography, computerized tomography, magnetic resonance, and three-dimensional echocardiography. Electrocardiography (ECG) is the first-choice technique for detecting left ventricular hypertrophy (LVH) in patients with hypertension due to its wide-scale availability, low cost, repeatability, and established value prognostic. 1-3 Current guidelines for diagnosis and treatment of hypertension strongly recommend ECG as the only examination to be performed in all hypertensive subjects for detection of LVH. 4 However, it is well known that majority of LVH are left undetected. The presence of electrocardiographic LVH predicts a several-fold increase cardiovascular morbidity and mortality in patients with essential hypertension. [5][6][7] The most effective strategy for cardiovascular event prevention implies extensive knowledge of the probable outcome for each patient during the treatment.
The aim of the paper is to investigate clinical and prognostic significance of electrocardiographic criteria of hypertensive left ventricular hypertrophy during 15-year follow-up period.

| Population study
This study represented an extension of our previously published study which had analyzed the above stated criteria during 5-year follow-up period. 8 The study was conducted at the Institute for Treatment and Rehabilitation "Niska Banja." The current study is part of the project "Prognostic significance of non-invasive parameters at patients with hypertension and left ventricular hypertrophy." The continuation of a 5-year research was approved on the session of Ethical Committee of Institute "Niska Banja" held on January 26, 2010 and RV 5 ratio (positive >1), Romhilt-Estes score, Framingham criterion, and Perugia criterion (electrocardiograph EKG-300, EI Nis). Additionally, Gubner-Ungerleider voltage as RD 1 + SD 3 ≥ 2.5 mV, and Lewis voltage as (RD 1 + SD 3 ) -(SD 1 + RD 3 ) ≥ 1.7 mV were used for confirming LVH. Positive Lyon-Sokolow voltage was defined as SV 1 + RV 5 or V 6 ≥ 3.5 mV, or in accordance with European Society of Cardiology Guidelines Committee as ≥3.8 mV. 9 In line with Cornell voltage criteria, the presence of LVH was defined as SV 3 + RaVL > 2.0 for women and >2.8 for men. Positive Cornell product was defined as SV 3 + RaVL × QRS duration ≥244 mV × ms. Left ventricular strain was defined as ST-segment depression ≥0.1 mV plus T-wave asymmetric inversion in V 2 to V 6 and in peripheral leads (lateral or inferior).
Perugia criterion was positive if SV 3 + RaVL >2.4 mV (men), >2.0 mV (women), and/or LV strain, and/or Romhilt-Estes score ≥5. Positive Romhilt-Estes score was defined as ≥5 points and calculated from six ECG features with a specific value of points for each feature: R or S wave in any limb lead ≥2.0 mV or S wave in V 1 or V 2 ≥ 3.0 mV or R wave in V 5 or V 6 ≥ 3.0 mV (three points); P-terminal force defined as terminal negativity of P wave in V 1 ≥ 0.10 mV in depth and ≥0.04 ms in duration (three points); LV strain defined as ST segment and T wave in opposite direction to QRS in V 5 or V 6 , without digitalis (three points); left axis deviation defined as QRS axis less than or equal to −30 (two points); QRS duration ≥0.09 ms (one point); and intrinsicoid deflection in V 5 or V 6 ≥ 0.05 ms (one point). 10

| Detection of left ventricular hypertrophy
Detection of echocardiographic left ventricular hypertrophy was done by means Acuson Sequoia C250 with 3.5 MHz, using M-mode technique. 8 Measurements were carried out in accordance with the rules of Penn convention, after which left ventricular mass was calculated. 11,12 Left ventricular mass was indexed by body surface area, while cut off values for left ventricular mass index were defined as ≥110 g/m 2 for women and ≥134 g/m 2 for men. 11 All details of echocardiographic examination were explained in our previously published study. 8

| Coronary artery disease detection
The first test for detecting coronary artery disease implied treadmill Bruce protocol. Ergometric testing was performed every second year or at more frequent intervals, if necessary (clinically suspected coronary artery disease). If the test data were not sufficient, stress echocardiography was carried out. Patients with positive exercise test (ST depression of ≥1 mm) were subjected to coronary angiography.
More details about exercise testing protocol could be seen in our previously published study. 8

| Blood pressure measurement
In addition to continuous blood pressure monitoring in medical office, each patient was subjected to 24-hour ambulatory blood pressure monitoring (Del Mar Avionics, Irvine, CA equipment, model P-VA and P6). 5 Extreme values of blood pressure recorded during 24-hour Baseline characteristic of examined population and therapy at the end of the study

| RESULTS
Statistically, both groups, that is, group with adverse events and group without adverse events, had the same basic characteristics and received the same therapy at the end of the study (Table 1).
Parameters obtained by 24-hour ambulatory blood pressure monitoring and the ones obtained by echocardiography were shown in ECG had low sensitivity and high specificity. 19 Nowadays, numerous experts attempt to redefine the existing LVH criteria by using magnetic resonance as gold standard. Rodrigues et al. 19 noticed that obese individuals had less frequent LHV positive results in terms of voltage criterion and suggested their correction.
The prognostic significance of this should be tested on large number of patients. Each positive step in terms of increasing sensitivity and preserving specificity had a significant clinical importance for this inexpensive and widely available diagnostic method. 20  Recently published study indicated that visit-to-visit blood pressure variability was a predictor of cardiovascular risk category in general population. 23 Hansen et al. 24 showed that correlation between 24-hour ambulatory blood pressure monitoring and LVH was better than the correlation between office-measured blood pressure and LVH. The values of blood pressure were not good predictors of morbidity and mortality in patients with LVH, as it was shown in our 5-year study. 4,25 After completing ergometric test, coronary heart disease was excluded from this study. Patients included in the study showed no difference in stress level or double product.