Short Communication

Rate-Dependent Repolarization Dynamics:

Correlation between Electrocardiographic T Wave and U Wave

Shioto Yasuda,, MD.1,2, Chiharu Yoshida, MD.1, Satoko Ideishi, MD.1,

Kei Irie, MD.1, Shohei Moriyama, MD.1, Mitsuhiro Fukata, MD.1,

Taku Yokoyama, MD.1, Takeshi Arita, MD.1, Keita Odashiro, MD.1,

Yoichiro Hiramoto, MD.2, Toru Maruyama, MD.1,2, and Koichi Akashi, MD.1

1 Heart Center, Kyushu University Hospital, Fukuoka 812-8582, Japan.

2 Department of Medicine, Koga Central Hospital, Koga 811-3101, Japan.

Running Head: Rate-dependent T-U waves

Word count of the main title: 11 words and 97 characters including spaces.

Word count of the running head: 3 words and 24 characters including spaces.

Word count for the main text: 2063 words, Number of Tables: 0, Number of Figures: 3

Address correspondence to: Toru Maruyama, MD., PhD.

Heart Center, Kyushu University Hospital, Fukuoka 812-8582, Japan

Tel: +81-92-583-7863, Fax: +81-92-592-2866, e-mail address: maruyama@artsci.kyushu-

u.ac.jp

Funding: The authors have declared no financial support.

Conflict of Interest: The authors have declared no competing interest.

Authorship: All authors declare that each author listed has contributed to the creation of this

manuscript to a degree worthy of authorship. All authors had access to the data and a role in

writing this manuscript.

Abstract

1

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22

23

24

12

U wave is the last ventricular repolarization component in electrocardiogram

(ECG). Rate-dependent amplitude adaptation of U wave was less known relative to that

of T wave. We experienced a patient presenting palpitation and showing ECG

associated with clear physiologic upward U wave. Echocardiogram demonstrated no

structural heart diseases and normal left ventricular (LV) function. T and U wave

amplitudes were simultaneously plotted as a function of the preceding RR intervals

using ambulatory ECG. U wave amplitude was augmented monotonously according to

an increase of the preceding RR interval, whereas T wave amplitude was increased to

the maximum level in the RR interval ranging from 1,000 to 1,500 msec and decreased

to a steady-state level as RR interval was further prolonged. These indicate that rate-

dependent repolarization gradient in T wave differs from that in physiologic U wave,

which is mediated by mechano-electrical coupling under pause-dependent LV pressure

and/or volume augmentation.

(149 words)

Key words: T wave, U wave, Rate-dependency, wave amplitude

Introduction

T wave amplitude is dependent greatly on the heart rate. This amplitude variability

2

25

26

27

28

29

30

31

32

33

34

35

36

37

38

39

40

41

34

is useful for risk assessment of the occurrence and recurrence of ventricular

tachyarrhythmia in patients with and without structural heart diseases [1-3], whereas U

wave amplitude dynamicity is not fully investigated. U wave is a small wave component

followed by T wave in the electrocardiogram (ECG). Polarity of the U wave determines

its abnormalities, i.e., positive U wave is generally physiologic, whereas negative U

wave is pathologic [4]. The amplitude of the physiologic U wave is less than one fourth

of the positive T wave amplitude [5]. However, rate-dependent U wave dynamics

remains unclear in comparison with T wave dynamics. We experienced an old female

patient showing clear physiologic U wave and analyzed rate-dependent dynamics of

both T and U waves simultaneously.

Case Description

A 80-year-old woman complained of palpitation and visited our hospital in 2012.

On her first visit to our hospital, the results of her serum chemistry were unremarkable,

i.e., serum potassium concentration was 4.0 mEq/l, and calcium concentration was 9.0

mg/dl. ECG showed respectable positive U wave associated with normal sinus rhythm

(heart rate of 50 beats per minute) and normal QRS axis. Automatic measurement of

3

42

43

44

45

46

47

48

49

50

51

52

53

54

55

56

57

56

absolute and rate-corrected QT interval indicated 445 msec and 410 msec, respectively

(Figure 1). QT interval rate correction was based on Bazett’s formula (QTc =

QT/RR1/2). Transthoracic echocardiogram demonstrated normal left ventricular (LV)

function without any structural heart diseases or false tendons. Papillary muscles size

was intermediate. Two-channel ambulatory ECG (Digital Walk, FM-160, Fukuda

Denshi, Tokyo, Japan) demonstrated 69,980 total heart beats per day, and average heart

rate was 49 beats per minute associated with nocturnal sino-atrial conduction block and

repetitive premature supraventricular contractions (PSVCs). Based on symptom diary,

her palpitation was considered to be due to the short runs of PSVCs, and antiarrhythmic

agents (diltiazem 60 mg per day or bisoprolol 1.25 mg per day) were prescribed. U

wave polarity was persistently positive. She complained no symptoms except for

palpitation and demonstrated no signs susceptible for heart failure and coronary heart

disease. Ambulatory ECG showed no diurnal ST-T changes. These indicate that her U

wave is not pathological.

Ambulatory Data Analyses

Based on the ambulatory monitoring applied to this case, U wave amplitude as well

4

58

59

60

61

62

63

64

65

66

67

68

69

70

71

72

73

78

as T wave amplitude was plotted as a function of preceding RR interval. Ambulatory

lead showing greater T and U wave complex (CM5) was adopted. On this analysis,

ECG data satisfying the following three criteria were acceptable; 1) ambulatory data

showing clear positive T wave and consecutive physiologic U wave without baseline

drift, and 2) data demonstrating that incisura between the T wave and U wave was

above the baseline, 3) whether sinus beats or PSVCs did not matter if the beats satisfied

the above two conditions.

The amplitudes of repolarization waves and the preceding RR intervals were

measured manually by two independent cardiologists, who were unaware of this

patient’s baseline characteristics. The amplitudes of the T and U wave complex in CM5

were persistently greater than those in CC5 and measured from isoelectric line including

the consecutive P wave onsets to the highest levels of the T and U wave, respectively.

The intra-observer and inter-observer coefficient of variance for manual measurements

of preliminary ECG was 1.4% and 2.5%, respectively. Representative ambulatory

monitors containing long RR intervals were demonstrated (Figure 2). Repetitive PSVCs

yielded compensatory pause, and pause-dependent behavior of T wave amplitude

5

74

75

76

77

78

79

80

81

82

83

84

85

86

87

88

89

910

differed from that of U wave. The amplitude of T wave preceded by long pause was

smaller than that of the preceding T wave, whereas the amplitude of the corresponding

U wave was greater than that of the preceding U wave (Figure 2A). These phenomena

were observed also immediately after the sino-atrial block (Figure 2B).

The amplitudes of T and U waves were plotted simultaneously as a function of the

preceding RR intervals. To avoid data plots deviation, equal numbers of data points (n =

10) were plotted within a same RR segment of 500 msec (i.e., 500 to 1,000 msec, 1,500

to 2,000 msec, 2,000 to 2,500 msec and 2,500 to 3,000 msec segments) except for RR

segment ranging from 1,000 to 1,500 msec, which corresponds to sinus beats. Sinus

beats data were selected randomly as one datum sampling of every hour segment (n =

20), if aforementioned inclusion criteria were met. This trial of random sampling was

repeated, and the number of this trial was 10. Inter-trial variance of sampled data (n =

60 in one trial) showed 8.2%, which was considered to satisfy the multiple random

sampling validation. Least square method was applied to fit the scattered plots to

polynomial regression curves by using Microsoft Excel 2010 (Redmond, WA, USA),

and regression curve showing the greatest correlation of coefficient (R) was accepted.

6

90

91

92

93

94

95

96

97

98

99

100

101

102

103

104

105

1112

The amplitudes of both T and U waves were abbreviated according to the shortening of

the preceding RR interval less than 1,000 msec as observed in the occurrence of PSVCs.

However, the behaviors of repolarization dynamics under the preceding RR interval

ranging from 1,000 to 1,500 msec in T wave were quite different from those of

physiologic U wave. The amplitude of T wave in this RR interval range was scattered

but was the maximum, whereas the amplitude of U wave was augmented as the RR

interval was lengthened. According to the further prolongation of the preceding RR

interval, T wave amplitude was diminished and saturated to the steady-state amplitude

around 0.6 mV. On the other hand, physiologic U wave amplitude was further

augmented and reached to the maximum level of approximately 0.25 mV (Figure 3).

Discussion

Physiologic U wave is sometimes clear in healthy individuals [4]. In this study

using ambulatory data in a female patient showing physiologic U wave, polynomial

7

106

107

108

109

110

111

112

113

114

115

116

117

118

119

120

121

122

1314

curve fitting was attempted (Figure 3), which yielded curvilinear augmentation of U

wave amplitude according to an increase of the preceding RR intervals (R2 = 0.51) and

the maximum hump of the T wave amplitude around the RR interval of 1,000 to 1,500

msec (R2 = 0.55).

Ventricular repolarization is greatly influenced by the preceding RR interval and

additionally modified by autonomic nervous system, electrolytes, gender and many

other factors. Rate-dependent adaptation of repolarization amplitude is useful for risk

stratification of ventricular tachyarrhythmia in patients with and without structural heart

diseases [1-3]. Among repolarization waves, genesis of T wave completely differs from

that of physiologic U wave [5], i.e., T wave represents overall ventricular repolarization,

whereas physiologic U wave reflects local repolarization of Purkinje fibers, M cell layer

or papillary muscles [4]. Electrophysiologic features of rate-dependency between the

whole ventricle and the local components underlie the differences between the T wave

and U wave dynamics.

Amplitude of the T wave is reduced according to an increase of the constant atrial

pacing rate, whereas that of the U wave remains unknown [6]. However, such rate-

8

123

124

125

126

127

128

129

130

131

132

133

134

135

136

137

138

1516

adaptation of repolarization wave amplitudes in steady state differs from dynamic

relationship between the wave amplitudes and the preceding RR intervals. As a matter

of fact, pause-dependent augmentation of repolarization wave amplitude was clear in

physiologic U wave but not in positive T wave (Figure 3). T wave amplitude in CM5 is

influenced by LV pressure and/or volume [7], i.e., T wave amplitude is augmented by

increased LV filling pressure and/or volume after the long RR intervals in healthy

individuals [8]. Temporary changes of LV loading condition modulates transmural

repolarization gradient and hence repolarization wave morphologies, which is mediated

by mechano-electrical coupling [9]. However, mechano-electrical coupling depends

heterogeneously on the period when and the extent to which LV preloading affects, i.e.,

preload is reduced during PSVCs and increased after the following compensatory pause.

These complexities lead to complicated T wave amplitude change as a function of RR

interval, whereas U wave reflecting local component repolarization may have been less

influences by such complexities.

Our results should be interpreted carefully due to some limitations. First, this

analysis is based on a single patient showing clear physiologic U wave. Second,

9

139

140

141

142

143

144

145

146

147

148

149

150

151

152

153

154

1718

alteration of T and U wave amplitude as a function of the preceding RR interval was

mainly obtained at nighttime due to analytical inclusion criteria. Reportedly, nocturnal

rate-dependent repolarization dynamics are different from daytime ones [10]. Third

limitation is manual measurement of repolarization waves using analog ambulatory

data. In spite of such limitations, repolarization dynamics of physiologic U wave is

concluded to be different from that of T wave in structurally normal heart. Rate-

dependent U wave amplitude behavior in heart diseases is a matter of future study with

perspective of arhythmogenicity.

Acknowledgement:

The authors thank Ms. Kanae Nagao for secretarial assistance.

Funding:

None

10

155

156

157

158

159

160

161

162

163

164

165

166

167

168

169

170

171

172

1920

Conflicts of interest:

None

References

1. Sobue Y, Watanabe E, Yamamoto M, Sano K, Harigaya H, Okuda K, Ozaki Y. Beat-

to-beat variability of T-wave amplitude for the risk assessment of ventricular

tachyarrhythmia in patients without structural heart disease. Europace 2011; 13(11):

1612-1618

2. Baumert M, Lambert E, Vaddadi G, Sari CI, Esler M, Lambert G, Sanders P,

11

173

174

175

176

177

178

179

180

181

182

183

184

185

186

187

188

189

190

2122

Nalivaiko E. Cardiac repolarization variability in patients with postural tachycardia

syndrome during graded head-up tilt. Clin Neurophysiol 2011; 122(2): 405-409.

3. Couderc JP, Zareba W, McNitt S, Maison-Blanche P, Moss AJ. Repolarization

variability in the risk stratification of MADIT II patients. Europace 2007; 9(9): 717-

723.

4. Pérez Riera AR, Ferreira C, Filho CF, Ferreira M, Meneghini A, Uchida AH,

Schapachnik E, Dubner S, Zhang L. The enigmatic sixth wave of the

electrocardiogram: the U wave (review). Cardiol J 2008; 15(5): 408-421.

5. Hlaing T, DiMino T, Kowey PR, Yan GX. ECG repolarization waves: their genesis

and clinical implications. Ann Noninvas Electrocardiol 2005; 10(2): 211-223.

6. Vahedi F, Haney MF, Jensen SM, Näslund U, Bergfeldt L. Effect of heart rate on

ventricular repolarization in healthy individuals applying vectorcardiographic T

vector and T vector loop analysis. Ann Noninvasive Electrocardiol 2011; 16(3): 287-

294.

7. Aeschbacher S, Metin F, Bossard M, Schoen T, von Rotz M, Mettler H, Abächerli R,

Risch M, Risch L, Conen D. Relationships of electrocardiographic parameters with

12

191

192

193

194

195

196

197

198

199

200

201

202

203

204

205

206

207

2324

ambulatory hypertension in young and healthy adults. Int J Cardiol 2016; 202(1):

300-304.

8. Nakagawa M, Ooie T, Hara M, Ichinose M, Nobe S, Yonemochi H, Saikawa T.

Dynamics of T-U wave in patients with idiopathic ventricular tachycardia

originating from the right ventricular outflow tract. Pacing Clin Electrophysiol

2004; 27(2): 148-155.

9. Meijborg VMF, Belterman CNW, de Bakker JMT, Coronel R, Conrath CE.

Mechano-electric coupling, heterogeneity in repolarization and the

electrocardiographic T-wave. Prog Biophys Mol Biol 2017; 130: 356-364.

10. Extramiana F, Leenhardt A, Maison-Blanche P. ECG evaluation of ventricular

properties: the importance of cardiac cycle length. Ann Noninvasive Electrocardiol

2009; 14 (Suppl 1): S54-S59.

13

208

209

210

211

212

213

214

215

216

217

218

219

220

221

222

223

224

2526

Legends for Figures

Figure 1: Twelve-lead electrocardiogram (ECG) in a 80-year-old female patient

showing clear physiologic U-wave. Absolute QT interval was 445 msec, whereas rate-

corrected QT interval was 410 msec, respectively.

Figure 2: Ambulatory monitor of this patient demonstrating long RR intervals caused

by sino-atrial block at night (A) and preceding repetitive premature supraventricular

beats observed in the evening (B). Upper lead is CM5 and lower lead is CC5. Note the

pause-dependent increase in the U wave (open triangles) amplitude and decrease in the

T wave (closed triangles) amplitude observed in CM5 lead of both (A) and (B).

Figure 3: Scattered plots of the amplitudes of T wave and U wave as a function of the

14

225

226

227

228

229

230

231

232

233

234

235

236

237

238

239

240

241

242

2728

preceding RR intervals. Regression curves showed R2 of 0.55 for the T waves (closed

squares) and 0.51 for the U waves (closed circles), respectively.

(Total word count of 2063 words)

15

243

244

245

246

247

2930

Figure 1

16

248

249

3132

Figure 2 A

17

▼ ▼ ▽▽

250

251

252

253

254

3334

Figure 2 B

18

▽▽▼▼

255

256

257

258

259

260

3536

Figure 3

0 500 1000 1500 2000 2500 30000

0.2

0.4

0.6

0.8

1

1.2

R² = 0.512322062759414

R² = 0.545172752899131

RR interval (msec)

Wav

e Am

plitu

des

(

mV)

19

261

262

3738