Increased QT dispersion in patients with Prinzmetal s variant angina and cardiac arrest

Transcription

1 Cardiovasular Researh 50 (2001) loate/ ardiores loate/ ardiores Inreased QT dispersion in patients with Prinzmetal s variant angina and ardia arrest Abstrat Nikhil Parhure, Velislav Bathvarov, Marek Malik, A. John Camm, Juan Carlos Kaski* Coronary Artery Disease Researh Unit, Department of Cardiologial Sienes, St George s Hospital Medial Shool, Cranmer Terrae, London SW17 0RE, UK Reeived 13 September 2000; aepted 7 November 2000 Objetives: We sought to ompare QT dispersion in patients presenting with Prinzmetal s variant angina ompliated by ardia arrest or synope and patients with unompliated variant angina. Bakground: Despite the usually benign ourse of treated Prinzmetal s variant angina, a proportion of vasospasti angina patients develop ventriular arrhythmias and sudden death in assoiation with oronary spasm. Inreased QT dispersion has been suggested to inrease suseptibility to ventriular arrhythmias in patients with oronary artery spasm. Methods: We studied 25 onseutive patients (mean age 58 years; 14 men) with lassial Prinzmetal s variant angina and doumented oronary artery spasm. None of the patients had oronary artery stenoses #40%. Five patients had suffered a doumented ardia arrest, two had reurrent synope and 18 had no arrhythmi events or synopal episodes. In all patients QT dispersion (QT maximum2qt minimum in every ECG lead) was measured on the baseline 12-lead eletroardiogram at study entry using a digitising board. Results: Mean (6S.D.) QT dispersion of study patients was ms. QT dispersion in patients with ardia arrest and synope ( ms) was signifiantly higher ompared to patients with no suh events ( ms), (95% CI , P50.005). No signifiant linial, biohemial or angiographi differenes were found between patients with and those without ardia arrest or synope. Conlusion: QT dispersion is inreased in patients with Prinzmetal s variant angina ompliated by ardia arrest and synope ompared to patients without suh events. Inreased QT dispersion may be both a substrate for sudden ardia death and a marker of risk in patients with Prinzmetal s variant angina Elsevier Siene B.V. All rights reserved. Keywords: Coronary disease; QT dispersion; Sudden death; Ventriular arrhythmias 1. Introdution patients have abnormalities of ventriular repolarization, even during asymptomati phases of their disease [7]. It Treated patients with Prinzmetal s variant angina usually has also been suggested that in the presene of a redued have good long-term prognosis [1 3]. However, myoar- threshold, or a substrate for ventriular arrhythmias, ordial infartion, sudden ardia death and synope have onary artery spasm may be a triggering mehanism for been shown to our in a sizeable proportion of patients life-threatening rhythm disturbanes in some vitims of with variant angina [4 6]. Life threatening ventriular ardia arrest [11 13]. Lengthening of the QT interval arrhythmias have been doumented during spontaneous may represent suh a substrate. oronary spasm in 5 15% of patients with Prinzmetal s Prolongation of the QT interval is reognised as a variant angina [1,4 10]. The fators responsible for in- marker of inreased risk of malignant ventriular arrhythreased risk of malignant ventriular arrhythmias and/ or mias and/ or sudden ardia death in a variety of linial sudden death in patients with oronary artery spasm onditions [14 16]. Moreover, QT interval has been remain to be eluidated. Studies have suggested that these shown to exhibit a ertain degree of interlead spatial variability (dispersion) whih may be onsidered to represent *Corresponding author. Tel.: ; fax: an approximate measure of ventriular repolarization address: jkaski@sghms.a.uk (J. Carlos Kaski). Time for primary review 26 days / 01/ $ see front matter 2001 Elsevier Siene B.V. All rights reserved. PII: S (00)00290-X

2 380 N. Parhure et al. / Cardiovasular Researh 50 (2001) abnormalities [17,18]. Based on the evidene that abnor- elevation developed at rest in the presene of ompletely malities of ventriular repolarization provide a substrate normal oronary arteries. for malignant ventriular arrhythmias, the variation in QT Of the 25 patients, seven had a history of ardia arrest interval duration has been advoated as a marker of or synope (Tables 1 and 2) [Group I: mean age 55 years arrhythmogeniity [19]. Interestingly, a reent study (range 44 71; 4 men)]. The remaining 18 patients (Tables showed that variant angina patients with inreased QT 1 and 3) [Group II: mean age 59 years (range 39 82; 10 dispersion have an inreased suseptibility to ventriular men)] had no history of synope or ardia arrest. arrhythmias during aetylholine indued oronary artery Patients in Group I met the riteria desribed by spasm [7]. However, the relationship between ardia Myerburg et al. [4]: (1) ardia arrest with doumented death, QT dispersion and spontaneous oronary spasm in ventriular fibrillation or sustained rapid ventriular tahyvariant angina patients has not been assessed systematial- ardia; (2) the absene of previous history of angina ly. petoris or aute myoardial infartion; (3) normal angiog- In the present study, we investigated the assoiation raphi left ventriular ejetion fration and no wall motion among oronary artery spasm, inreased QT dispersion and abnormalities; (4) the absene of signifiant oronary ardia arrest or synope in patients with Prinzmetal s artery narrowing (50%) or any strutural ardia abnorvariant angina petoris. malities; (5) the absene of identifiable or orretable auses of ventriular arrhythmia, suh as pre-exitation syndrome, long QT interval, metaboli or eletrolyte 2. Methods disturbanes or drug toxiity. At the time of the study, patients in both groups were reeiving treatment with 2.1. Patients alium antagonists, usually diltiazem ( mg daily) and isosorbide mononitrate (40 80 mg daily). Clinial We studied 25 onseutive patients (14 men and 11 harateristis and biohemial variables in the two patient women; mean age 58 years, range years) referred to groups are presented in Table 1. our unit for the assessment and management of oronary artery spasm from 1992 to All patients had a history 2.2. Measurement of QT dispersion lassial of Prinzmetal s variant angina with ST-segment elevation assoiated with typial anginal hest pain at rest, Manual eletroardiographi measurement was perusually ourring in the early morning hours or at night formed independently by two experiened observers time. All patients had a preserved exerise apaity and blinded to the linial and angiographi profile of the normal exerise stress test responses. In nine patients who patients. Measurement was performed on standard paperunderwent ergonovine provoative testing of oronary based 12-lead ECGs reorded 3 months after the index artery spasm, the typial symptoms and ECG hanges of event at a paper speed of 25 mm/s and standard gain of 1 variant angina were reprodued with the administration of mv/ m. All 12 leads were aquired simultaneously. The intravenous or intraoronary ergonovine. In the remainder, reordings were omposed of four triplets of simultaneousspontaneous oronary artery spasm ourred during or- ly aquired ECG leads (i.e. leads I, II, III, leads avr, avl, onary angiography (six patients) or transient ST segment avf, leads V1, V2, V3, and leads V4, V5, V6). A high Table 1 Clinial harateristis and oronary artery disease risk fators in 25 patients with Prinzmetal s variant angina Group I Group II P value Cardia arrest/ synope Unompliated (n57) (n518) 1. Age Sex M/F 4/3 10/8 3. Body mass index Heart rate, bpm Systoli BP, mm Hg Diastoli BP, mm Hg Smoking, Current 0 1 Ex 6 12 Never Plasma gluose, mmol/ l Total holesterol, mmol/ l LDL-holesterol, mmol/ l HDL-holesterol, mmol/ l Triglyerides, mmol/ l

3 N. Parhure et al. / Cardiovasular Researh 50 (2001) Table 2 Angiographi harateristis of patients in Group I [variant angina patients with ardia arrest (n55) and synope (n52)] Patient Age/ Sex Previous MI % Diameter stenosis EF % Ergonovine Site of Site of ST segment Cardia arrest No. test spasm elevation LAD LCx RCA Rhythm Ativity 1 46 F None LAD Anterior VF Resting 2 71 M None LAD Anterior VF Walking 3 53 M None RCA lateral a * Resting 4 44 M None ND ND Anterior VF Driving 5 59 F None Cx Ant1Inf b ** Sitting 6 62 F None ND ND Inferior VF Resting 7 52 M None RCA Lateral VF Resting a *5Synope preeded by fast palpitations. b **5Synope preeded by hest pain and irregular fast palpitations. ND5Not done. resolution (650 mm in both horizontal and vertial wave [20]. In ase of omplete fusion of the T and U wave diretions) digitising board (Cherry, Harpender, UK) no determination of the T wave end was attempted and the driven by a purpose-built in-house interative software was orresponding lead was exluded from measurement. As used for measurement. In eah lead of eah eletroar- suggested in other studies [21,22], leads in whih the T diographi traing, the RR, QRS and QT interval was wave amplitude was very low (i.e.,0.05 mv (,0.05 mm)) measured in at least two onseutive ardia yles of were exluded from measurement. T wave inversion did sinus rhythm and the results were subsequently averaged. not interfere in QT dispersion measurements. QT disper- Before the atual measurement of eah ECG, the sion was defined as the differene between the maximum observer was required by the software programme to and minimum QT interval in all measured leads. lassify the repolarisation pattern in eah lead into one of seven ategories, as desribed in previous studies from this 2.3. Statistial analysis department [37]. The main distintion was the T wave of very low amplitude, T U pattern onsisting of distint T Data are expressed as the mean value6s.d., mean and U wave separated by an isoeletri segment, partial differene6s.d., and 95% onfidene interval of mean fusion of the T and U wave with learly visible transition differene as appropriate. The unpaired Student s t-test between both waves (nadir or sudden hange in the slope was used to ompare QT dispersion in Group I and Group of the T wave), and omplete fusion of the T and U wave. II. Differenes in the linial and angiographi harateris- In ase of partial T U fusion the T wave end was tis of patients in Group I and II were ompared using determined to the nadir between the T wave and the U either a hi-square test or unpaired Student s t-test as Table 3 Angiographi harateristis of patients in group II (variant angina patients without arrhythmi ompliations) Patient Age/ Sex Previous MI % Diameter stenosis EF % ECG leads showing ST No. segment elevation LAD LCx RCA 1 53 F None Inferior 2 64 F None Lateral 3 60 M None Anterior 4 71 F None Anterior 5 71 F None Inferior 6 52 M Inferior Anterior 7 45 F None Lateral 8 63 M None Anterior 9 39 M None Anterior1Inferior F Inferior Anterior F None Inferior1lateral M None Anterior M Inferior Lateral M Inferior Anterior1Inferior M None Anterior M None Anterior1Inferior M None Anterior F Ant non Q Anterior

4 382 N. Parhure et al. / Cardiovasular Researh 50 (2001) appropriate. A probability (P) value,0.05 was onsidered statistially signifiant. The variability of measurement of QT dispersion by the two observers was assessed by relative errors alulated aording to the formula ((A2B)/ ((A1B)/2))3100 [%], where A and B were the values of QT dispersion measured by the two observers, respetively [23]. 3. Results The linial harateristis and oronary artery disease risk fators of the patients inluded in the study are presented in Table 1. Exept for the higher systoli blood pressure in Group I there were no differenes in linial, biohemial, hemodynami and angiographi variables between the two groups. All patients had a normal physial examination and no Fig. 1. QT dispersion was signifiantly higher in patients with ardia previous history of ardia disease. Six patients had an arrest (group 1) ompared to those without events (group 2). abnormal baseline ECG i.e. one patient had right bundle branh blok and five had T wave inversion (two in groups. The mean QT dispersion in Prinzmetal s variant inferior and three in anterior leads). Two patients in Group angina patients with ardia arrest and/ or synope (Group I and five in Group II were found to be hypertensive. I) was (range ) as ompared to Seven of the 25 patients (three in group I and four in (range ) in patients without these group II) had angiographially demonstrable oronary events (Group II) (Fig. 1). The differene was statistially artery stenoses #40% luminal diameter (Tables 2 and 3). signifiant (95% CI: , ombined standard error Five patients (in group II) had previous history of myoar- 7.56; P50.005). dial infartion and their average left ventriular ejetion QT Dispersion Measurement were reproduible and the fration was 68%61.57 (range 66% 70%). Bland Altman plot [24] of the differenes between the All patients had normal resting 2-dimensional ehoar- measurements of the two observers was %. diograms. No patient was reeiving antiarrhythmi agents or other therapy that ould affet the QT interval but all patients were treated with alium antagonists (diltiazem) 4. Disussion and nitrates (isosorbide mononitrate). Our study showed for the first time that QT dispersion is 3.1. QT dispersion on surfae ECG signifiantly higher in patients with Prinzmetal s variant angina ompliated by ardia arrest and synope than in Table 4 summarises the ECG variables assessed in patients without suh events. This finding suggests that patients with Prinzmetal s variant angina. The QTmax in patients with variant angina ompliated by ardia arrest Group I was signifiantly higher (P5,0.001) than Group may have abnormalities of ventriular repolarization, that II but there was no differene in QTmin between the two may inrease the risk of malignant ventriular arrhythmias. Table 4 QT measurements in patients with prinzmetal s variant angina Group I Group II P value Patients with ardia arrest Patients without ardia arrest and synope and/ or synope QT MAX (ms) , QT MIN (ms) Maximum QT dispersion , Minimum QT dispersion , Mean QT dispersion , Standard Deviation Combined Standard Error % Confidene Interval Two tailed P 0.005

5 N. Parhure et al. / Cardiovasular Researh 50 (2001) Our results also indiate that evaluation of QT dispersion patients with Prinzmetal s variant angina and found to be on the surfae ECG may provide useful information about higher ompared to patients with atypial hest pain. the suseptibility of ertain variant angina patients to suffer However, the present investigation is the first study to ardia arrest. ompare QT dispersion in Prinzmetal s variant angina patients who presented with ardia arrest versus those 4.1. Ventriular arrhythmias in variant angina with an unompliated linial ourse. The findings in the present study support previous reports [31 34] demon- The ourrene of ventriular arrhythmias in patients strating that inreased QT dispersion represents a marker with variant angina is well established. Prinzmetal s origi- of inreased risk of sudden ardia death in different nal desription of vasospasti angina inluded one patient patient subsets i.e. oronary artery disease, hypertrophi who had runs of ventriular tahyardia during episodes of ardiomyopathy and heart failure. Although ontroversy pain [25]. Sine then, several studies have shown that exists as to what exatly QT dispersion represents in serious ventriular arrhythmias may our in this syndrome eletrophysiologial terms, a reent editorial artile [35] [5,7,9,10,6] and sudden ardia death has also been has suggested that QT dispersion is an approximate doumented [5,6] often in assoiation with ventriular expression of ventriular repolarization abnormalities. arrhythmias. Studies have reported that myoardial infar- Suzuki et al. [7] showed both that baseline QT dispertion and sudden ardia death our in 10 20% of patients sion was signifiantly greater in patients with vasospasti with variant angina over several years of follow-up [26 angina than in patients with atypial hest pain and that 28]. Whether silent or symptomati, transient myoardial QT dispersion was signifiantly greater in patients who ishemia may be the triggering mehanism for fatal developed ventriular arrhythmias during provoation tests ventriular arrhythmias in vitims of ardia arrest [11 for spasm. However there were no patients with dou- 13]. mented ardia arrest in Suzuki s study. Contrary to the Previous studies of survivors of arrhythmi ardia findings of Suzuki et al. [7], Ashikaga et al. [36] observed arrest due to isolated oronary artery spasm have been no signifiant differenes in QT dispersion in 35 patients sare and limited in size. Kerin et al. [10] demonstrated with vasospasti angina ompared to 30 patients with that most of their study patients had high-grade obstrutive atypial hest pain and normal oronary arteries (2768 vs. lesions (signifiant oronary artery disease.50% stenosis) 2867 ms, P5NS). However, in the Ashikaga study [37] and that the angiographi site of spasm orrelated well QT dispersion was signifiantly inreased by dipyridamole with eletroardiographi findings. Furthermore, arrhyth- administration in patients with vasospasti angina (53614 mias in this study were more ommon in the group with ms, P,0.0001) ompared to patients with atypial hest oronary obstrutive disease. Conversely, Nakamura et al. pain. [29] demonstrated that the risk of serious arrhythmias or sudden death did not depend on the presene of severe Clinial and angiographi features fixed stenoses. This is onsistent with our findings in the These have been shown to be of limited value in present study and with data reported by Maseri et al. [30], reognising patients with variant angina who are at risk of who found no orrelation between the development of fatal developing life threatening ventriular arrhythmias. Therearrhythmia and the patients oronary anatomy. In our fore, our observations in the present study, that assessment study most of the patients had ompletely normal oronary of QT dispersion even when obtained in the asymptomati arteriograms and the seven patients with oronary stenoses state, may represent a useful, noninvasive preditive (#40% diameter redution) were evenly distributed be- variable have important linial impliations. tween the two groups. Interestingly, Myerburg et al. [4] showed that even silent 4.3. Study limitations myoardial ishemia due to oronary artery spasm an initiate potentially fatal arrhythmias in patients without Manual measurement of QT dispersion may have limitaflow-limiting oronary artery lesions. Unfortunately how- tions, partiularly low interobserver reproduibility as ever, the relationship between QT intervals and ardia reported by other authors [37 39]. arrest was not assessed in Myerburg s study. It is oneiv- The reliability of manual measurement of the QT able that as suggested by our findings, Prinzmetal s variant dispersion is influened by many fators, suh as the angina patients with an arrhythmi substrate, of greater QT amplitude of the T wave, the noise level, and the number dispersion, may be more liable to develop potentially fatal of measured leads. Moreover, it has been suggested that arrhythmias in assoiation with episodes of oronary the use of a ruler instead of a alliper and even the spasm. respiratory phase of the patient during whih the reording is made [40] may influene QT measurements QT dispersion and variant angina To minimise these limitations we used a high resolution (650 mm in both horizontal and vertial diretions) QT dispersion has been previously studied [7] in digitising board (Cherry, Harpender, UK) driven by a

6 384 N. Parhure et al. / Cardiovasular Researh 50 (2001) purpose-built in-house interative software whih was [5] Miller DD, Waters DD, Szlahi J, Theroux P. Clinial harateris- demonstrated reently to provide aurate measurement of tis assoiated with sudden death in patients with variant angina. Cirulation 1982;66: QT dispersion [41]. [6] Freedman SB, Rihmond DR, Kelly DT. Clinial studies of patients Another potential limitation of this study is the relatively with oronary spasm. Am J Cardiol 1983;52:67A 71A. small number of patients. However, Prinzmetal s variant [7] Suzuki M, Nishizaki M, Arita M, Ashikaga T, Ymawake N, Kakuta angina is not a ommon ondition and variant angina T, Numano F, Hiraoka M. Inreased QT dispersion in patients with patients presenting with ardia arrest are even more rare. vasospasti angina. Cirulation 1998;98: [8] Salerno JA, Previtali M, Chimiente M, Klersey C, Bobba P. Vasospasm and ventriular arrhythmias. Ann NY Aad Si 1984;427: Conlusions [9] Previtali K, Klersey C, Salerno JA, Chimienti M, Paniroli C, Marangoni E, Spehia G, Comolli M, Bobba P. Ventriular tahyarrhythmias in Prinzmetal s variant angina: linial signifiane QT dispersion is signifiantly higher in patients with and relation to the degree and time ourse of ST segment elevation. Prinzmetal s variant angina ompliated by ardia arrest Am J Cardiol 1983;52: and/ or synope ompared to unompliated patients. In- [10] Kerin NZ, Rubenfire M, Naini M, Wajszzuk WJ, Pamatmat A, reased QT dispersion may represent both a substrate for Casade PN. Relationship of arrhythmias to ST segment elevation arrhythmi sudden ardia death and a marker of inreased and R-wave hanges. Cirulation 1979;60: mortality risk in variant angina patients. [11] Myerburg RJ, Kessler KM, Castellanos A. Pathophysiology of sudden ardia death. PACE, Paing Clin Eletrophysiol 1991;14: [12] Morady F, DiCarlo LA, Winston S, Davis JC, Sheinman MM. 6. Condensed abstrat Clinial features and prognosis of patients with out of hospital ardia arrest and a normal eletrophysiologial study. J Am Coll In this study we ompared QT dispersion in patients Cardiol 1984;4: [13] Morady F, DiCarlo Jr. LA, Krol RB et al. Role of myoardial with Prinzmetal s variant angina ompliated by ardia ishaemia during programmed stimulation in survivors of ardia arrest or synope and patients with unompliated variant arrest with oronary artery disease. J Am Coll Cardiol 1987;9:1004 angina. We observed that QT dispersion in patients with ardia arrest and synope ( ms) was signifi- [14] Surawiz B, Knoebel SB. Long QT: Good, bad or indifferent? J Am antly higher ompared to patients with no suh events Coll Cardiol 1984;4: [15] Ward DE. Prolongation of the QT interval as an indiator of risk of ( ms), (95% CI , P50.005). No signifia ardia event. Eur Heart J 1988;9(Supp 1G): ant linial, biohemial or angiographi differenes were [16] Shwartz PJ, Wolf S. QT interval prolongation as preditor of found between patients with, and those without ardia sudden death in patients with myoardial infartion. Cirulation arrest or synope. Inreased QT dispersion may represent 1978;57: both a substrate for arrhythmi sudden ardia death/ [17] Mirvis DM. Spatial variation of QT interval in normal persons and patients with aute myoardial infartion. J Am Coll Cardiol synope and a marker of inreased risk in variant angina 1985;5: petoris. [18] Sylven JC, Horaek BM, Spener CA et al. QT interval variability on the body surfae. J Eletroardiol 1984;17: [19] Surawiz B. Will QT dispersion play a role in linial deision Aknowledgements making? J Cardiovas Eletrophysiol 1996;7: [20] Lepeshkin E, Surawiz B. The measurement of the Q T interval of the eletroardiogram. Cirulation 1952;VI: N. Parhure is a British Heart Foundation Junior Re- [21] Murray A, MLaughlin NB, Bourke JP, Doig JC, Furniss SS, searh Fellow. Campbell RWF. Errors in manual measurement of QT intervals. Br Heart J 1994;71: [22] Savelieva I, Yi G, Guo H-H, Hnatkova K, Malik M. Agreement and Referenes Reproduibility of automati versus manual measurement of QT interval and QT dispersion. Am J Cardiol 1998;81: [23] Kautzner J, Ti G, Camm AJ, Malik M. Short and long-term [1] Yasue H, Takizawa A, Nagao M, Nishida S, Horie M, Kubota J, reproduibility of QT, QT, and QT dispersion measurement in Omote S, Takaoka K, Okumora K. Long-term prognosis of patients healthy subjets. PACE 1994;17: with variant angina and influential fators. Cirulation 1988;78:1 9. [24] Bland JM, Altman DG. Statistial methods for assessing agreement [2] Bott-Silverman C, Heupler FA. Natural history of pure oronary between two methods of linial measurement. Lanet 1986;1:307 spasm in patients treated medially. J Am Coll Cardiol 1982;2: [25] Prinzmetal M, Kennamar R, Merliss R, Wada T, Bor N. A variant [3] Shroeder JS, Lamb IH, Bristow MR, Ginsburg R, Hung J, form of angina petoris. Am J Med 1959;27: MAuley BJ. Prevention of ardiovasular events in variant angina [26] Waters DD, Szlahi J, Miller DD, Theroux P. Clinial haraterisby long-term diltiazem therapy. J Am Coll Cardiol 1983;1:1507 tis of patients with variant angina ompliated by myoardial infartion or death within one month. Am J Cardiol 1982;49:658 [4] Myerburg R, Kessler KM, Mallon SM, Cox MM, DeMarhena E, 664. Interian A, Castellanos A. Life threatening ventriular arrhythmias [27] Waters DD, Miller DD, Szlhi J, Bouhard A, Methe M, Kreeft J, in patients with silent myoardial ishemia due to oronary artery Theroux P. Fators influening the long term prognosis of treated spasm. N Engl J Med 1992;326: patients with variant angina. Cirulation 1983;68:

7 N. Parhure et al. / Cardiovasular Researh 50 (2001) [28] Mark DB, Cliff RM, Morris KG, Harrell FE, Pryor DB, Hlatky MA, [35] Malik M. QT dispersion: Time for an obituary? Eur Heart J Lee KL, Rosati RA. Clinial harateristis and long term survival 2000;21(12): of patients with variant angina. Cirulation 1984;69: [36] Ashikaga T, Nishizaki M, Arita M, Yamawake N, Suzuki M, [29] Nakamura M, Takeshita A, Nose Y. Clinial harateristis assoi- Hashimoto Y, Kishi Y, Numano F, Hiraoka M. Effet of ated with myoardial infartion, arrhythmias and sudden death in dipyridamole on QT dispersion in vasospasti angina petoris. Am J patients with vasospasti angina. Cirulation 1987;75: Cardiol 1999;84: [30] Maseri A, Severi S, De Nes M, L Abbate A, Chierhia M, Marzilli [37] Kautzner J, Gang Yi, Kishore R, Copie X, Janota T, Nagayoshi H, M, Ballestra A, Parodi O, Biagini A, Distante A. Variant angina: one Camm AJ, Malik M. Interobserver reproduibility of QT interval aspet of a ontinuous spetrum of vasospasti myoardial is- measurement and QT dispersion in patients after aute myoardial haemia. Pathogeneti mehanisms, estimated inidene and linial infartion. ANE 1996;1: and oronary angiographi findings in 138 patients. Am J Cardiol [38] Bhullar HK, Chia P, Ong K, Ng WL. Assessment of inter-observer 1978;42: and intra-observer variability in the measurement of qt dispersion. [31] Bogun F, Chan K, Harvey M, Goyal R, Castellani M, Niebauer M, Comput Cardiol 1996;: Daoud E, Man KC, Strkberger SA, Morady F. QT dispersion in [39] Van de Loo A, Arendts W, Hohnloser SH. Variability of QT nonsustained ventriular tahyardia and oronary artery disease. dispersion measurements in the surfae eletroardiogram in patients Am J Cardiol 1996;77: with aute myoardial infartion and in normal subjets. Am J [32] Day CP, M Comb JM, Cambell RWF. QT dispersion: an indiation Cardiol 1994;74: of arrhythmia risk in patients with long QT intervals. Br Heart J [40] Krupieniz A, Czarneki R, Adamus J. QT dispersion magnitude is 1990;63: related to the respiratory phase in healthy subjets. Am J Cardiol [33] Buja G, Miorelli M, Turrini P, Nava A. A omparison of QT 1997;80: dispersion in hypertrophi ardiomyopathy between patients with [41] Malik M, Bradford A. Human preision of operating a digitizing and without ventriular arrhythmias and sudden death. Am J Cardiol board: impliations for eletroardiogram measurement. PACE 1993;72: ;21: [34] Barr CS, Naas A, Freeman M, Lang CC, Struthers AD. QT dispersion and sudden unexpeted death in hroni heart failure. Lanet 1994;343: