Journal of Human Hypertension (2004) 18, 33–40 & 2004 Nature Publishing Group All rights reserved 0950-9240/04 $25.00 www.nature.com/jhh ORIGINAL ARTICLE Electrocardiographic assessment of left with time–voltage QRS and QRST-wave areas

L Oikarinen1,2, M Karvonen2,3, M Viitasalo1, P Takala2,3, M Kaartinen1, J Rossinen1, I Tierala1,2,HHa¨nninen1,2, T Katila2,3, MS Nieminen1 and L Toivonen1 1Division of Cardiology, Helsinki University Central Hospital, Helsinki, Finland; 2BioMag Laboratory, Helsinki University, Central Hospital, Helsinki, Finland; 3Laboratory of Biomedical Engineering, Helsinki University of Technology, Helsinki, Finland

The sum of time–voltage QRS areas in the 12-lead for LVH detection. The 12-lead QRS area sum performed electrocardiogram (ECG) has outperformed other 12- similarly. Taking T-wave areas into account did not lead ECG indices for detection of left ventricular improve the results. QRS area sum from two most hypertrophy (LVH). We assessed indices of time–voltage informative leads (located in the upper and lower right QRS and T-wave (QRST) areas from body surface precordium) also separated the LVH group from con- potential mapping (BSPM) for detection of and quantita- trols (61.1 7 23.5 vs 27.8 7 6.5 lV s, respectively; tion of the degree of LVH. We studied 42 patients with Po0.00001). This 2-lead QRS area sum showed 90% echocardiographic LVH (LVH group) and 11 healthy sensitivity with 100% specificity for LVH detection and controls (controls). QRST area sums were calculated maintained high correlation to indexed left ventricular from 123-lead BSPM and from the 12-lead ECG for mass (r ¼ 0.732; Po0.001). In conclusion, the BSPM QRS comparison. Leadwise discriminant indices and correla- area sum compared to 12-lead QRS area sum does not tion coefficients were used to identify optimal recording substantially improve LVH assessment. The 2-lead QRS locations for QRST area-based LVH assessment. BSPM area sum may improve ECG QRS area-based LVH QRS area sum was greater in the LVH group than in assessment. controls (3752 7 1259 vs 2278 7 627 lV s, respectively; Journal of Human Hypertension (2004) 18, 33–40. Po0.001) and at 91% specificity showed 74% sensitivity doi:10.1038/sj.jhh.1001631

Keywords: body surface potential mapping; electrocardiogram; left ventricular hypertrophy

Introduction voltage-based LVH indices fail to identify LVH patients with sufficient accuracy.8 In the orthogo- Both identification of patients with left ventricular nal-lead signal-averaged ECG, the QRS time–voltage (LV) hypertrophy (LVH) and assessment of the area in the horizontal plane significantly improves degree of LVH are clinically important, because ECG detection of LVH.9,10 Even better performance detection of LVH independently predicts cardiovas- for eccentric LVH detection can be achieved by cular morbidity as well as mortality and the relative summing all the QRS time–voltage areas in the 12- risk of these events increases with increasing LV 11 1–3 lead ECG. However, the electrode positioning used mass. Echocardiography and magnetic resonance in the 12-lead ECG may not be optimal for QRS area- imaging, in particular, provide accurate estimates of based LVH detection.12 In addition, the T-wave may anatomic LV mass, but neither of these methods is 4–6 contain information of LV mass not revealed by applicable to wide-scale screening of LVH. restricting the measurements to the QRS com- According to the solid-angle theory, LVH aug- plex.12–15 ments QRS voltages on the body surface electro- 7 In the present study, our first aim was to assess the cardiogram (ECG), but for clinical purposes QRS capability of body surface potential mapping (BSPM) QRS area analysis for detection and quanti- fication of pressure overload-induced LVH com- Correspondence: Dr Lasse Oikarinen, Division of Cardiology, pared to the 12-lead ECG QRS area sum as well as to Department of Medicine, Helsinki University Central Hospital, the Sokolow–Lyon index,16 and to study if calculat- Haartmaninkatu 4, 00290 Helsinki, Finland. E-mail: [email protected] ing QRST areas improves the BSPM methodology. Received 25 February 2003; revised 9 May 2003; accepted 3 Our second aim was to identify the optimal August 2003 electrode positioning for QRS area-based LVH ECG detection of left ventricular hypertrophy L Oikarinen et al 34 assessment and thereby derive a simple, but in- LV mass) from M-mode measurements made accord- formative ECG LVH index. ing to the Penn convention.4 Calculated LV mass and also LV mass indexed to body surface area (to account for the effect of physiological variation in Subjects and methods LV mass) were both used in the analyses. The patients included in the LVH group showed indexed Study population LV mass 4116 g/m2 in men and 4104 g/m2 in 18 The study population (n ¼ 53) consisted of two women. In patients with echocardiographic LVH, groups. The LVH group (n ¼ 42) included patients the LV geometric pattern was considered as con- 19 with haemodynamically significant centric if the relative wall thickness exceeded 0.43; stenosis (n ¼ 27) and patients with essential arterial otherwise, it was considered to be eccentric. hypertension (n ¼ 15), all of whom showed echo- cardiographic LVH by gender-specific criteria (see below). The patients with aortic valve stenosis were BSPM being invasively evaluated in the Helsinki Univer- sity Central Hospital. None of these patients showed Measurement protocol: The BSPM recording device X70% stenosis of luminal diameter of epicardial and procedure have been described in detail else- 20–22 coronary arteries in routine coronary angiography. where. In brief, the BSPM device records No patient with essential arterial hypertension had a unipolar potentials with 120 electrodes placed on history of effort . No patient in the LVH group the subject’s thorax (Figure 1), and three standard showed pathological Q-waves in the ECG or regional limb potentials. The recorded signals were digitized wall-motion abnormalities in LV cineangiography or with a sampling frequency of 1 kHz, and selectively two-dimensional echocardiography suggesting a averaged off-line. previous . Patients with a right LVH measures: From the signal-averaged BSPM or left were not included. data, the time instants of QRS onset, QRS offset, and The control subjects (n ¼ 11; controls) were T-wave offset in each lead were determined as 23,24 healthy middle-aged volunteers who were recruited described previously. The following BSPM in- with a newspaper advertisement. Consecutive sub- dices were calculated in each lead: (1) QRS area by jects who had no major cardiovascular risk factors integrating the absolute value of the BSPM signal 11 and who had no history or signs of cardiovascular over the QRS complex. (2) T-wave area by disease were further examined. If they showed a integrating the value of the BSPM signal from the normal routine echocardiogram and a symptom- QRS offset to the T-wave offset. To relate this index limited bicycle exercise test without angina or ST- segment changes, they were included as controls. All patients in both study groups were in sinus rhythm and none was under medication known to influence QRS or T-wave morphology. The study complied with the Declaration of Helsinki. Informed consent was obtained from all participants and the study protocol was approved by the ethical review committee of the Helsinki University Central Hos- pital.

Echocardiography A skilled cardiologist (MaK) performed a standard M-mode and two-dimensional echocardiography to all study subjects. Two-dimensionally guided M- mode tracings were recorded from the parasternal long-axis view with particular care taken to ensure proper orientation of the ultrasonic beam. End- systolic and end-diastolic LV wall thicknesses and dimensions were then measured on a digitizing table by the cardiologist who was blinded to any Figure 1 The electrode layout of the 123-lead body surface clinical data according to the recommendations of potential mapping system. The electrodes are mounted on 18 the American Society of Echocardiography.17 An strips with a vertical interelectrode distance of 5 cm. The average of up to five measurements from consecu- horizontal spacing of the strips is defined by the individual thoracic dimensions. The vertical line indicates the level of the tive cardiac cycles for each measure was used in the fourth intercostal space. The squares show the locations of the

final analyses. LV mass was calculated with an conventional 12-lead ECG precordial leads V1–V6. The three limb anatomically validated formula (r ¼ 0.90 vs autopsy leads (leads 1–3) are not shown.

Journal of Human Hypertension ECG detection of left ventricular hypertrophy L Oikarinen et al 35 value to QRS-T discordance (opposite polarity of both good discriminative power between LVH group main QRS deflection and T-wave), areas of the T- and controls as well as with a good correlation wave in leads with discordant QRS and T waves between QRS area and LV mass index for further were considered negative (Figure 2). (3) QRS-T area analysis. by subtracting the T-wave area parameter from the QRS area parameter. The rationale of calculating QRS and T-wave areas differently was based on a 12-lead ECG previous study showing that the ECG response to LVH is flattening of the T-wave or polarity reversal of A 12-lead digital ECG was recorded before the BSPM the T-wave in relation to the main QRS deflection.13 measurement. However, because in some of the QRS area sum, T-wave area sum, and QRS-T area ECGs there were a few missing leads, we used a sum were derived by calculating the sum of derived 12-lead ECG from the BSPM measurement respective values from all leads. in all study subjects. Precordial leads V1–V2 and V4– Optimal recording locations for QRS area:To V6, as well as bipolar limb leads are part of the identify the recording locations with the best BSPM layout (Figure 1). The signal in precordial lead V3 was interpolated from four neighbouring discriminative power for LVH detection, we used E the discriminant index (DI).12 In each lead and for leads (distances 2 cm) and the signals in unipolar augmented limb leads were calculated according to each index, DIs were calculated by subtracting the 27 control group mean value from the LVH group mean standard equations. value. The obtained difference was then divided by From the averaged 12-lead ECG signal, we measured the Sokolow–Lyon index (SV1+RV5/V6 the pooled standard deviation of the study groups to 16 derive leadwise DIs.12,25 The pooled standard devia- (whichever is taller)). In each lead, the time– tion (s) was derived from pooled variance (s2) voltage QRS area was calculated similarly as in the calculated by the equation s2 ¼ [(n À1)s2+(n À1)s2]/ BSPM leads, and the areas from the 12 leads were 1 1 2 2 summed (12-lead QRS area sum).11 In patients with [n1+n2À2], where s1 and s2 are the standard devia- 12,25,26 complete measurements, the QRS area sum from the tions of the two groups of sizes n1 and n2. A large DI value indicates good performance of that BSPM-derived 12-lead ECG correlated extremely lead in separating the LVH patients from the closely with that obtained from the original 12-lead controls. We then constructed discriminant maps12 ECG (r ¼ 0.956). displaying DI value distributions over the thorax. As the DI uses groups based on dichotomized values, it does not necessarily identify informative Statistical methods leads for quantification of a continuous parameter, Statistical data analysis was performed with SPSS such as LV mass. Therefore, we also constructed version 10.1 software (SPSS Inc, Chicago, IL, USA). correlation coefficient maps displaying the correla- All continuous data are presented as mean 7 S.D. tion coefficient between QRS area and LV mass The significance of difference in continuous vari- index in each lead. In an attempt to derive a simple ables between the study groups was determined index for LVH detection and quantification, we with the Mann–Whitney U-test. Correlation between searched for particularly informative leads with continuous variables was calculated using Pearson’s correlation coefficients and using partial correlation coefficients when adjusting for covariates. All categorical variables were tested with the w2 test or with Fisher’s exact test when appropriate. Two- tailed P-values o0.05 were considered statistically significant.

Results Clinical and echocardiographic characteristics The clinical and echocardiographic characteristics of the study groups are shown in Table 1. Of the 42 patients in the LVH group, 33 (78.6%) had con- centric and nine (21.4%) eccentric LVH.

Figure 2 Calculation of QRS, T-wave and QRS- areas in each BSPM lead. QRS area is the sum of absolute values of Q-, R- and S-wave areas. In measurement of the T-wave area, negative BSPM and 12-lead ECG indices of LVH in the study deflections were subtracted from the positive ones (a). If the main groups QRS deflection and the T-wave were discordant, the T-wave area was multiplied by À1 (b). To obtain the QRS-T area, the T-wave In BSPM, the QRS area sum separated the LVH area was subtracted from the QRS area. See text for details. group from controls better than T-wave area sum or

Journal of Human Hypertension ECG detection of left ventricular hypertrophy L Oikarinen et al 36 Table 1 Clinical and echocardiographic characteristics of the study groups

Controls (n=11) LVH group (n=42)P-value

Gender (male/female) 8/3 25/17 0.503 Age (years) 55 7 7637 12 0.008 Body mass index (kg/m2) 24.4 7 2.8 26.3 7 3.6 0.119 Body surface area (m2) 1.89 7 0.14 1.89 7 0.17 0.878 IVS thickness, diastole (cm) 0.85 7 0.21 1.44 7 0.32 o0.001 PW thickness, diastole (cm) 0.93 7 0.19 1.45 7 0.31 o0.001 LV internal diameter, diastole (cm) 4.59 7 0.67 5.01 7 0.71 0.177 LV mass (g) 148 7 68 344 7 93 o0.001 LV mass index (g/m2)797 37 181 7 45 o0.001 Fractional shortening (%) 38 7 6367 8 0.525

Data are given as number of individuals or mean 7 S.D. IVS: interventricular septal; LV: left ventricular; LVH: left ventricular hypertrophy; PW: posterior wall.

Table 2 Body surface potential mapping and 12-lead ECG indices of left ventricular hypertrophy in the study groups

Controls (n=11) LVH group (n=42) P-value

BSPM index QRS area sum (mV s) 2278.2 7 626.5 3751.7 7 1259.1 o0.001 T wave area sum (mV s) 412.9 7 1649.2 À1223.4 7 1953.5 0.033 QRS-T area sum (mV s) 1865.3 7 1703.8 4975.2 7 2988.2 0.002 2-lead QRS area sum (mV s) 27.8 7 6.5 61.1 7 23.5 o0.00001

12-lead ECG index Sokolow–Lyon voltage (mV) 2.32 7 0.62 3.96 7 1.37 o0.001 QRS area sum (mV s) 344.0 7 85.0 537.4 7 178.5 o0.001

Data are given as mean 7 S.D. BSPM: body surface potential mapping; ECG: electrocardiographic; LVH: left ventricular hypertrophy.

QRS-T area sum (Table 2). In the 12-lead ECG, the Sokolow–Lyon voltage as well as the QRS area sum performed equally well as the BSPM QRS area sum for identifying LVH patients (Table 2). Using the conventional cutoff value of 3.5 mV, the sensitivity of Sokolow–Lyon voltage for detection of LVH was 50% with 91% specificity. With matched specificity of 91% (one false positive in the controls) and optimizing cutoff points in the present study population, the sensitivity of the 12-lead QRS area sum (cutoff value 440 mV s) was 69% and that of BSPM QRS area sum (cutoff value 3000 mV s) was 74%. Figure 3 The group mean (GM) and DI maps for QRS area (QRS), T-wave area (T) and QRS-T area (QRS-T) are displayed over the thorax. In the GM maps, the solid lines represent positive, dotted Optimal BSPM leads for QRS area assessment in LVH lines negative, and dashed line the zero value of integrals. In the detection and quantitation DI maps, the respective lines represent positive, negative, and zero values of DI. Contour steps are 20 mV s for GM maps and By use of discriminant maps and correlation maps, 0.5 units for DI maps. The best discriminating areas for these we identified several locations with both a high DI indices can be found on the right and lower parts of the anterior value for QRS area (Figure 3) and a high correlation thorax and on the left flank. coefficient value between QRS area and LV mass index. These were leads 12 and 13 on the lower right precordium, leads 10 and 16 in the upper right group from controls (Table 2), and the sensitivity for precordium, and lead 82 on the left lower flank area LVH detection (cutoff value 39 mV s) was 90% with (see Figure 1). The sum of QRS areas in leads 13 and 100% specificity (Figure 4). Other additional sum 16 (2-lead QRS area sum) also separated the LVH combinations from these five leads did not improve

Journal of Human Hypertension ECG detection of left ventricular hypertrophy L Oikarinen et al 37 Table 3 Correlation coefficients of left ventricular mass and left ventricular mass index with body surface potential mapping and 12-lead ECG LVH indices

All subjects (n=53) LVH group (n=42)

LV mass LVMI LV mass LVMI

BSPM index QRS area sum 0.716* 0.725* 0.633* 0.659* T wave area sum À0.603* À0.591* À0.666* À0.666* QRS-T area sum 0.703* 0.699* 0.702* 0.713* 2-lead QRS area sum 0.724* 0.732* 0.608* 0.626* 12-lead ECG index Sokolow–Lyon 0.539* 0.582* 0.342z 0.413w QRS area sum 0.669* 0.687* 0.582* 0.623*

*Po0.001, wPo0.01, zPo0.05. BSA: body surface area; BSPM: body surface potential mapping; ECG: electrocardiographic; LVH: left ventricular hypertrophy; LVMI: left ventricular mass indexed to body surface area.

area sum (Table 3). In fact, the BSPM and 12-lead QRS area sums were almost linearly correlated (r ¼ 0.967; Po0.001), suggesting that they provide practically identical information. Scatterplots show- ing the actual values for BSPM 2-lead QRS area sum and 12-lead QRS area sum vs LV mass index are shown Figure 4. In the overall study population, BSPM QRS area sum and 2-lead QRS area sum correlated strongly with echocardiographic inter- ventricular septal wall thickness (r ¼ 0.694 and 0.654, respectively; Po0.001 in both) as did also Sokolow–Lyon voltage and 12-lead QRS area sum (r ¼ 0.542 and 0.685, respectively; Po0.001 in both). None of these indices correlated with LV internal diastolic diameter (r values betweeen À0.085 and 0.103; P40.4 in all). All of the above correlations remained significant and substantially in the same order when using partial correlation coefficients adjusting for the possible effect of age and/or gender Figure 4 Scatterplots showing the correlation of 12-lead QRS area on these parameters. sum (a) and 2-lead QRS area sum (b) to left ventricular mass index in the overall study population (different markers used in controls and left ventricular hypertrophy (LVH) patients). The r values are from the overall study population (n ¼ 53). the performance. Furthermore, summing up other Discussion additional combinations of QRS area, T-wave area Main findings and/or QRS-T area values from informative leads did not perform substantially better (data not shown). Our results show that BSPM QRS area sum can separate the group of LVH patients without evidence of from controls, but for Correlation of BSPM and 12-lead ECG QRS area identifying patients with LVH it does not perform measures to LV mass and structure substantially better than the 12-lead QRS area sum. The sum of QRS areas from the two most informa- Both in the overall study population and within the tive leads identified by leadwise BSPM analysis LVH group, the BSPM QRS and QRS-T area sums improved detection of LVH compared to 12-lead showed a stronger correlation to LV mass and LV ECG analyses and retained the ability to quantitate mass index than the Sokolow–Lyon voltage, whereas the degree of LVH. Both these leads were located the BSPM index correlations were only slightly outside the conventional 12-lead ECG electrode stronger than or equal to those of the 12-lead QRS positions.

Journal of Human Hypertension ECG detection of left ventricular hypertrophy L Oikarinen et al 38 BSPM and 12-lead ECG detection and quantitation of BSPM technique have improved the identification LVH and even quantification of anatomic LVH,12,29–34 suggesting that accurate assessment of anatomic Detection of patients with LVH is clinically impor- LVH from electrical signals is indeed possible. tant, because the risk of cardiovascular morbidity Based on the above background, we hypothesized and death is 2–4-fold higher in subjects with LV that BSPM mapping of QRS and QRST areas may mass above normal limits than in those with normal improve the detection and quantitation of LVH. In LV mass.1–3,6,18 In addition, quantification of the addition, we determined whether the information degree of LVH is equally important, since the risk of redundancy inherent in BSPM might be compressed these events increases as LV mass increases.2,3 Direct to yield a simple index that could be recorded by a visualization of the LV wall thicknesses and dimen- simple modification of lead placement with the 12- sions by echocardiography enables calculation of LV lead ECG equipment. Our results show that BSPM mass based on certain geometrical assumptions,4,6 QRS area sum can separate the group of patients but the accuracy with this methodology is obviously with pressure overload-induced LVH with mostly expertise-dependent. Magnetic resonance imaging concentric LVH from controls and can quantify the offers accurate assessment of anatomic LV mass, but degree of LVH, but the 12-lead QRS area sum is, in the upper normal limits with this methodology have both respects, a very accurate estimate of BSPM QRS only recently been established28 and the method area sum. Furthermore, with the approach we used atpresent is unsuitable for screening purposes.5 taking T-wave areas into consideration does not Therefore, an attempt to develop a widely applic- improve performance. However, our results suggest able screening method for LVH assessment is still that a simple sum of QRS areas from the upper and warranted. lower right precordial areas improves performance Pressure-overload of the LV induces myocardial compared to the 12-lead QRS area sum. Although cellular hypertrophy and, thus, an increase in this 2-lead QRS area sum was not an a priori index, electrically active LV mass. According to both the previous BSPM mapping studies have shown computer simulations and the solid-angle theory, that for QRS voltage measurements the right pre- an increase in LV mass is expected to strengthen cordial area is particularly informative for LVH QRS voltage on the body surface ECG.7,13 However, assessment.12,30 This suggests that our results may clinically QRS voltage-based LVH indices in the 12- be more generalizable and prompts the prospective lead ECG fail to identify individual LVH patients testing of our approach. with sufficient accuracy. This may in part be explained by the effect of several extracardiac factors, such as age, gender, chest diameter, and Study limitations body build, on ECG QRS voltages.8 Attempts to take these factors into account alone does not suffice for An obvious limitation to our study is the relatively critical improvement of the methodology. Measuring small study population and the use of echocardio- only maximum QRS deflection amplitudes over- graphy as a reference method,5 since even with looks the more powerful discriminating and quanti- meticulous measurement technique, the assessment tation capabilities of information in other segments of LV mass with M-mode echocardiography is not within the QRS complex.12,29,30 Accordingly, the very accurate.35 However, at present screening of time–voltage integrated QRS area in the horizontal patients for a study like ours with magnetic plane from the orthogonal signal-averaged ECG resonance imaging is impractical. The numerical improves ECG detection of LVH,9,10 and the sum of values, including the values of upper limit of QRS areas in the 12-lead ECG results in further normal, in the article by Okin et al11 should be improvement of performance for LVH detection.11 treated with caution and their approach was not Importantly, the 12-lead QRS area sum may be tested in an independent test set. Thus, because of relatively independent of extracardiac factors and the lack of values for upper normal limits, we gender.11 Even though the QRS area scans the compared our results with matched specificities. temporal aspects of the depolarization wavefront, The upper limits of normal should be derived from a it is possible that the positioning of the electrodes in larger sample of unselected subjects and include the 12-lead ECG is not spatially optimal for QRS investigation of potential gender differences, area-based LVH detection. This is suggested by the although Okin et al11 found their criteria not gender distribution of most informative leads in BSPM dependent. Deriving the 2-lead QRS area sum index outside the conventional 12-lead ECG electrode from a training set and not testing it subsequently in positioning12,29,30 and by the fact that orthogonal an independent test set is likely to optimize the ECG does not capture all the available information results. Therefore, to test if the 2-lead QRS area for the assessment of the activation wave- index truly improves clinical LVH assessment, the fronts.9,10,12,29,30 In addition, the T-wave may contain sensitivity and specificity of our approach should be complementary information of LV mass,12–15,29,30 but assessed in a separate population and also including QRST areas have not been assessed for this purpose. more patients with borderline LVH. Although with Methods developed for LVH detection utilizing the our approach we found no additional information in

Journal of Human Hypertension ECG detection of left ventricular hypertrophy L Oikarinen et al 39 the T-wave area sums, other methods incorporating 8 Surawicz B. Electrocardiographic diagnosis of chamber separately QRS and T-wave area information or enlargment. J Am Coll Cardiol 1986; 8: 711–724. using, for example, the concept of QRST gradients 9 Okin PM et al. Electrocardiographic diagnosis of left may offer additional information from repolariza- ventricular hypertrophy by the time–voltage integral of tion for LVH assessment.36 Indexing LV mass to body the QRS complex. J Am Coll Cardiol 1994; 23: 133–140. 10 Okin PM, Roman MJ, Devereux RB, Kligfield P. Time– surface area is an attempt to account for the voltage area of the QRS for the identification of left variation in physiological determinants of LV mass ventricular hypertrophy. Hypertension 1996; 27: and to enable assessment of correlation to patholo- 251–258. gically increased LV mass, but has certain limita- 11 Okin PM et al. Time–voltage QRS area of the 12-lead tions.37 electrocardiogram. Detection of left ventricular hyper- trophy. Hypertension 1998; 31: 937–942. 12 Kornreich F et al. Identification of best electrocardio- graphic leads for diagnosing left ventricular hypertro- Study implications phy by statistical analysis of body surface potential maps. Am J Cardiol 1988; 62: 1285–1291. Our present results show that the 12-lead QRS area 13 Thiry PS, Rosenberg RM, Abbott JA. A mechanism for sum captures the same information for LVH detec- the electrocardiogram response to left ventricular tion and quantitation as the BSPM QRS area sum. hypertrophy and acute . Circ Res 1975; 36: 92–104. The 12-lead QRS area sum performs relatively well 14 Devereux RB, Reichek N. abnormalities in a study population of patients with mainly of left ventricular hypertrophy. Clinical, echocardio- concentric LVH, which has not been previously graphic and hemodynamic correlates. J Electrocardiol shown.11 A simple 2-lead QRS area sum from the 1982; 15: 47–54. right precordium may be a way to improve ECG 15 Okin PM et al, for The LIFE Study Investigators. detection of LVH with 12-lead ECG equipment. If Relationship of the electrocardiographic strain pattern prospectively validated, this index may be useful in to left ventricular structure and function in hyperten- screening patients with LVH and in preliminary sive patients: The LIFE Study. J Am Coll Cardiol 2001; assessment of the degree of LVH as well as in 38: 514–520. interventional studies for assessment of regression 16 Sokolow M, Lyon TP. The ventricular complex in left ventricular hypertrophy as obtained by unipolar of LVH. precordial and limb leads. Am Heart J 1949; 37: 161–168. 17 Sahn DJ, DeMaria A, Kisslo J, Weyman A. Recommen- Acknowledgements dations regarding quantitation in M-mode echocardio- graphy: results of a survey of echocardiographic This work was supported by grants from the Finnish measurements. Circulation 1978; 58: 1072–1083. Foundation for Cardiovascular Research and the 18 Liu JE, Devereux RB. Clinical assessment of cardiac Aarne Koskelo Foundation. hypertrophy. In: Sheridan DJ (ed). Left Ventricular Hypertrophy. Churchill-Livingstone: London, 1998, pp 11–16. 19 Reichek N, Devereux RB. Reliable estimation of peak References left ventricular systolic pressure by M-mode echo- graphic-determined end-diastolic relative wall thick- 1 Casale PN et al. Value of echocardiographic measure- ness: identification of severe valvular in ment of left ventricular mass in predicting cardiovas- adult patients. Am Heart J 1982; 103: 202–209. cular morbid events in hypertensive men. Ann Intern 20 Simelius K et al. A body surface potential mapping Med 1986; 105: 173–178. system in clinical use. Med Biol Eng Comput 1996; 34: 2 Levy D et al. Prognostic implications of echocardio- 107–108. graphically determined left ventricular mass in the 21 Takala P et al. Magnetocardiographic and electrocar- Framingham Heart Study. N Engl J Med 1990; 322: diographic exercise mapping in healthy subjects. Ann 1561–1566. Biomed Eng 2001; 29: 501–509. 3 Schillaci G et al. Continuous relation between left 22 Ha¨nninen H et al. ST-segment level and slope in ventricular mass and cardiovascular risk in essential exercise-induced myocardial ischemia evaluated with hypertension. Hypertension 2000; 35: 580–586. body surface potential mapping. Am J Cardiol 2001; 4 Devereux RB et al. Echocardiographic assessment of 88: 1152–1156. left ventricular hypertrophy: comparison to necropsy 23 Simson MB. Use of signals in the terminal QRS findings. Am J Cardiol 1986; 57: 450–458. complex to identify patients with ventricular tachy- 5 Myerson SG, Bellenger NG, Pennell DJ. Assessment of cardia after myocardial infarction. Circulation 1981; left ventricular mass by cardiovascular magnetic 64: 235–242. resonance. Hypertension 2002; 39: 750–755. 24 Oikarinen L et al. Magnetocardiographic QT interval 6 Devereux RB, Pini R, Aurigemma GP, Roman MJ. dispersion in postmyocardial infarction patients with Measurement of left ventricular mass: methodology sustained ventricular : validation of auto- and expertise. J Hypertens 1997; 15: 801–809. mated QT measurements. Pacing Clin Electrophysiol 7 Holland RP, Arnsdorf MF. Solid angle theory and the 1998; 21: 1934–1942. electrocardiogram: physiologic and quantitative inter- 25 Kornreich et al. Identification of best electrocardio- pretations. Prog Cardiovasc Dis 1977; 19: 431–457. graphic leads for diagnosing anterior and inferior

Journal of Human Hypertension ECG detection of left ventricular hypertrophy L Oikarinen et al 40 myocardial infarction by statistical analysis of body 32 Holt JH, Barnard ACL, Kramer JO. Multiple dipole surface potential maps. Am J Cardiol 1986; 58: . A comparison of electrically and 863–871. angiographically determined left ventricular masses. 26 Altman DG. Comparing groups F continuous data. In: Circulation 1978; 57: 1129–1133. Altman DG (ed). Practical Statistics For Medical 33 Dunn RA et al. Performance of conventional orthogo- Research. Chapman & Hall: London, 1995, pp 179–228. nal and multiple-dipole electrocardiograms in estimat- 27 Macfarlane PW. Lead systems. In: Macfarlane PW, ing left ventricular muscle mass. Circulation 1979; 60: Lawrie TDV (eds). Comprehensive Electrocardiology. 1350–1353. Pergamon Press: Oxford, 1989, pp 315–352. 34 Yamaki M et al. Improved diagnostic performance on 28 Salton CJ et al. Gender differences and normal left the severity of left ventricular hypertrophy with body ventricular anatomy in an adult population free of surface potential mapping. Circulation 1989; 79: 312– hypertension. A cardiovascular magnetic resonance 323. study of the Framingham Heart Study offspring cohort. 35 Palmieri V et al. Reliability of echocardiographic J Am Coll Cardiol 2002; 39: 1055–1060. assessment of left ventricular structure and function. 29 Kornreich F et al. Diagnostic body surface potential The PRESERVE Study. J Am Coll Cardiol 1999; 34: map patterns in left ventricular hypertrophy during 1625–1632. PQRST. Am J Cardiol 1989; 63: 610–617. 36 Montague TJ et al. Isointegral analysis of body surface 30 Kornreich F et al. Improved prediction of left ven- maps: surface distribution and temporal variability in tricular mass by regression analysis of body surface normal subjects. Circulation 1981; 63: 1166–1172. potential maps. Am J Cardiol 1990; 66: 485–492. 37 Rautaharju PM, Zhou SH, Park LP. Improved ECG 31 Holt JH, Barnard ACL, Lynn MS. A study of the human models for left ventricular mass adjusted for body size, heart as a multiple dipole electrical source. II. with specific algorithms for normal conduction, Diagnosis and quantitation of left ventricular hyper- bundle branch blocks, and old myocardial infarction. trophy. Circulation 1969; 40: 697–710. J Electrocardiol 1996; 29(Suppl): 261–269.

Journal of Human Hypertension