Umeå University Medical Dissertations, New Series 1583

Long QT Syndrome - studies of diagnostic methods

Ulla-Britt Diamant

Department of Public Health and Clinical Medicine,

Medicine.

Umeå 2013

Responsible publisher under Swedish law: the Dean of the Medical Faculty

This work is protected by the Swedish Copyright Legislation (Act 1960:729)

ISBN: 978-91-7459-693-9

ISSN: 0346-6612

Cover: Wooden relief in basswood made by surolle ℅ Jögge Sundqvist. Part of commisson work at Umeå University Hospital 2013. Photographer: Jögge Sundqvist Elektronic version available at http://umu.diva-portal.org/

Printed by: Print & Media

Umeå, Sweden 2013

i

To the study participants and their relatives

The Indian heart, as seen by the Ojibwa Indians, is full of strength

and beats powerfully in space.

The cover: Relief “I live by the clear space”

Artist Jögge Sundqvist.

ii

Table of Contents

Table of Contents ii Abstract iv Abbreviations v Svensk sammanfattning vii List of papers viii Introduction 1 Background 2

Historical background 2 LQTS a genetic cardiac ion channelopathy - Genotype 4

Functional effects 5 Heredity in LQTS 6 Prevalence of LQTS 7 Genetic testing for LQTS 7 Founder mutations and founder populations 7

Clinical presentation of LQTS - Phenotype 8 Arrhythmias in the LQTS 8 Other ECG changes associated with LQTS 9 The “Schwartz score” 10 Risk stratification in the LQTS 11 Prophylactic treatment of symptoms in Long QT syndrome 11

Physiologic basis of the ventricular repolarization 12 Action potential of the ventricular myocytes in the normal heart 12 Prolonged action potential 13 Ventricular gradient 13 Heterogeneities of ventricular repolarization 14

Manual methods to define the end of the T-wave 15 “Normal” QT interval 16

The QT interval corrected for heart rate 18 The corrected QT interval and bundle branch block 18

Other clinical conditions associated with prolonged QT interval 19 Electrocardiographic recordings 19

Vectorcardiography 19 VCG loop characteristics 23 Other VCG parameters 25

Aims 27 Materials 28

Overview of the LQTS study population in paper I-IV 28 Subjects 28 Ethical considerations 30

iii

Methods 31 12-lead ECG 31

Automatic measuring and interpretation of the QT interval 31 Manual measurement of the QT interval 32

VCG according Frank lead system 32 Automatic measuring of the QT interval 32 Off-line analysis of the VCG 33

Genetic testing 33 Genealogy, geography and haplotype analysis 34

Genealogical and geographic analysis 34 Haplotype analysis 35 Mutation age and prevalence of the Y111C mutation in the KCNQ1 gene 36

Statistic methods 36 Limitations 37

Summary of Results 39 Paper I and II 39 Paper III 39 Paper IV 40

Discussion 43 Manual measurements 43 Automatic methods 44 QTc cut-off values 45 The “ideal” QT measurement 47 Electrophysiological phenotype 47 Ventricular repolarization 49 Risk markers 49 Founder population Y111C 50

Conclusion 51 Acknowledgements 52 References 54

Appendices 1 I - Instructions in manual measurement of the QT interval (translated) 2 II - Long QT syndrome questionnaire (translated) 3

iv

Abstract

Background: The Long QT Syndrome (LQTS) is a hereditary heart

disease with risk of malignant ventricular arrhythmia and sudden

cardiac death. Despite our increased knowledge about genotype and

phenotype correlation we still rely on the 12-lead ECG for assessment of

the QT interval and the T-wave morphology for diagnosis and risk

stratification. Intra- and -inter individual variability in manually QT

measurement and, e.g., difficulties in defining the end of the T-wave may

impair the diagnosis of LQTS. Increased heterogeneity in ventricular

repolarization (VR) may be an important factor in the arrhythmogenicity

in cases of LQTS. In a LQTS founder population the same mutation is

carried by numerous individuals in many families which provide a

unique opportunity to study diagnostic methods, risk assessment, VR

and the correlation between genotype and phenotype.

Methods: Resting 12-lead ECG and vectorcardiogram (VCG) were

recorded in 134 LQTS mutation carriers and 121 healthy controls, to

investigate the capability and precision in measuring the QT interval.

For assessment of the VR, VCG was compared in individuals with

mutations in the KCNQ1 and KCNH2 gene. Genealogical and geographic

studies were performed in 37 index cases and their relatives to

determine if Swedish carriers of the Y111C mutation in the KCNQ1 gene

constitute a founder population. To confirm kinship, haplotype analysis

was performed in 26 of the 37 index cases. The age and prevalence of the

Y111C mutation were calculated in families sharing a common haplotype

Results: VCG by automatic measurement of the QT interval provided

the best combination of sensitivity (90%) and specificity (89%) in the

diagnosis of LQTS. VCG showed no consistent pattern of increased VR

heterogeneity among KCNQ1 and KCNH2 mutation carriers. Living

carriers of the Y111C mutation shared a common genetic (haplotype),

genealogic and geographic origin. The age of the Y111C mutation was

approximately 600 years. The prevalence of living carriers of the Y111C

mutation in the mid-northern Sweden was estimated to 1:1,500-3,000.

Conclusion: We have shown that VCG provides a valuable contribution

to the diagnosis and risk assessment of LQTS in adults and children. No

consistent pattern of increased VR heterogeneity was found among the

LQTS mutation carriers. The identified Swedish LQTS founder

population will be a valuable source to future LQTS research and may

contribute to increase our understanding of LQTS and the correlation of

phenotype, genotype and modifying factors.

v

Abbreviations

AP Action potential from ventricular myocytes

APD Action potential duration

Bpm Beats per minute

EAD Early after depolarization

ECG Electrocardiogram

LQTS Long QT Syndrome

µV Microvolt

µVs Microvolt seconds

Ms Millisecond

QRSamplitude Amplitude of the maximum QRS vector in

space

QRSarea The spatial area under the curve formed by the

moving heart vector during the QJ interval

QRS-T angle The angle between the maximum QRS and T

vectors

Tamplitude Amplitude of the maximum T vector in space

Tarea The spatial area under the curve formed by the

moving heart vector during the JT interval

Tavplan The mean distance between the periphery of

the T loop and both sides of the preferential

plane

Tazimuth The angle of the maximum T vector in the

transverse plane (0◦ left, +90◦ front, -90◦ back

and 180◦ right)

TdP Torsade de Pointes

Teigenvalue The squared quotient between the two largest

perpendicular axes (eigenvalues) of the T loop

in the preferential plane [(d1/d2)2 where d1 ≥

d2]

Televation The angle of the maximum T vector in the

cranio-caudal direction by us defined from 0◦

(caudal direction) to 180◦ (cranial direction)

Tp-e T peak to T end, the last part of the QT interval

and final repolarization

VCG Vectorcardiogram

VG Ventricular gradient or QRST area is the

spatial area under the curve formed by the

moving heart vector during the QT interval;

also the sum of the QRSarea vector and Tarea

vector, taking into account the angle between

vi

them; it describes the dispersion of action

potential morphology throughout the

ventricles.

VR Ventricular repolarization

The intervals and waves in an ECG complex

vii

Svensk sammanfattning

Bakgrund och syfte med avhandlingen: Långt QT syndrom (LQTS) är

en ärftlig hjärtsjukdom med risk för plötslig död p.g.a. livshotande

hjärtrytmrubbning (kammartakykardi). EKG är ett viktigt instrument vid

diagnos och riskbedömning av LQTS. Avhandlingens syfte är att förbättra

diagnos och riskbedömning av LQTS genom att undersöka olika EKG

metoder, samt finna ut om svenska bärare av Y111C mutationen

(sjukdomsanlag) i KCNQ1 genen har samma ursprung och utgör en

founderpopulation. En founderpopulation utgör en genetisk homogen grupp

och är en viktig källa för framtida forskning av bl.a. diagnostiska metoder.

Metod och material: Registrering av 12 avlednings EKG och

vektorkardiogram (VKG) utfördes på 134 LQTS mutationsbärare- och 121

friska kontroller. Metodernas diagnostiska precision samt förmåga till

riskbedömning avseende LQTS värderades. Genealogiska och geografiska

studier utfördes på 37 indexpersoner och deras släktingar. Haplotypanalys

(släktskaps analys) utfördes på 26 av de 37 indexpersonerna.

Resultat: VKG erbjöd den bästa kombinationen av sensitivitet (0,90),

specificitet (0,89) för diagnos av LQTS. Genealogiska studier visade att 26

bärare av Y111C tillhörde samman släktträd med en anfader/moder som

levde för ca 400 år sedan. Alla nu levande bärare av Y111C visade sig ha ett

gemensamt geografiskt och genetiskt ursprung.

Slutsatser: Vi har visat att VKG är en effektiv metod med hög precision

som utgör ett värdefullt tillskott vid LQTS diagnostik. Avseende

riskmarkörer i LQTS populationen, fann vi vid VKG mätning förlängt QT

intervall men för övrigt normala fynd i repolarisationsprocessen. Vi har

identifierat den första svenska LQTS founderpopulationen som utgör en

viktig källa för framtida forskning.

Betydelse: För enskilda individer med risk för potentiellt livshotande

hjärtrytmrubbningar är tidig diagnostik, riskvärdering och profylaktisk

behandling av stor vikt. Vi har påvisat vektorkardiografins fördelar vilket

skulle kunna ge metoden en ökad betydelse i kliniken. Dock krävs ytterligare

studier för att befästa vektorkardiografins position vid diagnostik och

riskvärdering vid LQTS. Att vi identifierat en founderpopulation öppnar

unika möjligheter för internationell betydelsefull forskning inom ett snabbt

expanderande område. På sikt kommer detta att kunna bidra till minskad

dödlighet bland patientgrupper med LQTS.

viii

List of papers

This thesis is based on the following papers, which are referred to in the text

by their Roman numerals:

I Two automatic QT algorithms compared with manual

measurement in identification of long QT syndrome

Diamant UB, Winbo A, Stattin EL, Rydberg A, Kesek M,

Jensen SM

J Electrocardiol. 2010 Jan-Feb;43(1):25-30

Reproduced with kind permission from Elsevier.

II Vectorcardiographic Recordings of the Q-T Interval in a

Pediatric Long Q-T Syndrome Population

Diamant UB, Jensen SM, Winbo A, Stattin EL, Rydberg A

Pediatr Cardiol. 2013 Feb;34(2):245-9

Reproduced with kind permission from Springer Science and

Business Media

III Electrophysiological Phenotype in the LQTS Mutations Y111C

and R518X in the KCNQ1 Gene

Diamant UB*, Farzad Vahedi*, Winbo A, Rydberg A, Stattin

EL, Jensen SM, Bergfeldt L

*both authors contributed equally

Manuscript

IV Origin of the Swedish long QT syndrome Y111C/KCNQ1

founder mutation

Winbo A, Diamant UB, Rydberg A, Persson J, Jensen SM,

Stattin EL

Heart Rhythm. 2011 Apr;8(4):541-7.

Reproduced with kind permission from Elsevier

1

Introduction

The Long QT Syndrome (LQTS) is a hereditary heart disease that has been

further clarified in the era of molecular genetics. The electrocardiogram

(ECG) is still one of the most important elements in the description of the

clinical phenotype in LQTS, in spite of the increased knowledge about

genotype and phenotype correlation. In the daily work of diagnosis and risk

stratification, we still rely on the 12-lead ECG for the evaluation of the

duration of the QT interval and the T-wave morphology. However, because

of pitfalls in the manual measurement of the QT interval, the diagnosis of

LQTS may be misclassified due to, e.g., difficulties in defining the end of the

T-wave. There are also intra- and inter- individual differences in

measurements of the QT interval (1), which further worsens the accuracy of

manual measurements. The ECG is fundamental in determining the

diagnosis and in the risk stratification of LQTS. Nonetheless, there is no

consensus on how to measure the QT interval or how to correct the QT

interval for heart rate.

The aim of this thesis was to improve the capability of ECG as a diagnostic

and prognostic tool in LQTS. Paper I and II compare two automatic methods

(12-lead ECG and Frank vectorcardiogram VCG) and manual measurement

of the QT interval in healthy individuals and a population of LQTS subjects.

In paper I we also state intra- and inter- variability of manual QT

measurement for four observers. Promising results in the automatic

measurement of the QT interval by VCG are included in paper I and II, which

deals with the time-consuming and error-prone manual measurement of the

QT interval. Increased ventricular repolarization (VR) heterogeneity may be

an important factor in the arrhythmogenicity in LQTS. This has, to the best

of our knowledge, not been evaluated in humans in any previous study. In

paper III, VR was studied in two populations with two different LQTS

mutations and a group of healthy individuals. Genetic homogeneous

populations are of great value to develop and improve diagnostic methods

and risk assessment tools. A so-called founder population - comprising

individuals with a common geographic and genealogic origin and carrying a

common mutation - is ideal for studies of correlation between genotype and

phenotype including ventricular repolarization. During our work in the

LQTS clinic we noticed a high number of families carrying the same

mutation Y111C in the KCNQ1 gene. After systematic genealogic and genetic

studies, presented in paper IV, we proved that these families formed a LQTS

founder population.

2

Background

Historical background

The first ECG in a human was recorded by Augustus Waller in 1887 (2).

Later William Einthoven invented the string galvanometer and named the

wave deflections P, Q, R, S and T (3), instead of ABCD that had been used

with the capillary electrometer. He avoided N and O, which were already in

use with mathematic/geometric matters and started with P to label the first

deflection. In 1912, he also defined the current standard ECG leads I, II and

III also known as “Einthoven’s triangle” (4). Einthoven and colleagues

described the first electrical heart vector and the angle that was between a

horizontal reference line and the heart vector. It was time consuming to

calculate the loop, since each lead (I,II and III) had to be aligned manually

for the calculation of the loop in the frontal plane (5). The study of

orthogonal leads became easier, and was further developed with the

introduction of the cathode-ray oscillograf in the 1950´s. More than 30

corrected lead systems was developed but the Frank lead system became the

most frequently used because it was a compromise between accuracy and

practical manageability (6). With today’s computer technology, it is easy to

record a VCG with presentation of the loops in real-time. During the 1930´s

the augmented limb leads (aVR, aVF and aVL) and the precordial leads (V1-

V6) came into use and in the 1950´s the use of the standard 12 lead ECG was

widely spread. In 1930´s the American Heart Association and the Cardiac

Society of Great Britain defined the standard positions and wiring of the 12-

lead ECG, which today forms the basis of manual and automatic

interpretation of ECG - one of the most important diagnostic tools for

cardiac diseases (7, 8).

Congenital LQTS is a cardiac arrhythmogenic disorder associated with

prolongation of the QT interval on the ECG, syncope and sudden cardiac

death in young individuals with no structural heart disease. It was first

described by Jervell and Lange-Nielsen 1957 in a family with congenital

deafness, prolonged QT interval, syncope and sudden death transmitted in a

autosomal recessive pattern (9). Romano reported in 1963 (10) and Ward in

1964 (11) a similar syndrome but without deafness and transmitted in an

autosomal dominant way. In 1991 came the first report that presented LQTS

as a genetic disorder, a finding that was achieved after gene linkage studies.

The study was performed in a pedigree with both healthy and LQTS affected

relatives, and uncovered a strong linkage between LQTS and a DNA marker

at the Harvey ras-1 locus on chromosome 11 (12). In 1995, the connection

between gene mutations and cardiac potassium and sodium channels was

3

discovered – since then LQTS has been considered as a cardiac ion

channelopathy (13, 14). The early use of comprehensive national population

records in Sweden provides good opportunities to track the ancestry of living

persons back in time - this registration is a golden source for mapping of

monogenetic diseases (15). Following a church law of 1686, the priests were

enjoined since the late 17th and early 18th century to register in church

archives every dweller in their parish. As a result of this we now, in Sweden,

possess registers of all individual habitants of the Swedish parishes in the

catechetical examination records. These were replaced in 1895-1900 by the

“congregation book” which, in its turn, was abolished in 1991. The church

registered all migration, births, deaths and marriages, and from 1860 every

tenth year census were held and archived in the Swedish Central Bureau of

Statistics. Tracking the geographical origin of the population carrying the

Y111C mutation in the KCNQ1 gene brought us to the area along Ångerman

River valley and its tributaries. As the other historical provinces of Northern

Sweden, the landscape Ångermanland was first populated in the coastal

areas and the river valleys (16). Occasional Swedish settlements were

founded during the middle ages in the area where our founder couple settled

in the mid-17th century. During the 16th, century there was also some

colonization in the area by settlers coming from Finland. The settlers were

principally farmers, but they also provided for themselves by hunting and

fishing. There was a Sami indigenous population in the area that was

predominantly nomads, living by keeping reindeers, hunting and fishing.

The state encouraged the colonization of the interior of northern Sweden

through the so called “Lappmarksplakatet” of 1673, and this was the starting

point for the era of colonization of the inlands of northern Sweden that

lasted for over 200 years. No great influx from other parts of the country

occurred during the first half of the 18th century. In 1749, the Crown

supplemented the bill with the Lapplands regulations settling a distribution

of trades between the colonizers and the Sami natives. Settlers should

foremost pursue farming, wherewith the trades of hunting and fishing would

essentially be protected for the Sami. The main object of the regulations was

to stimulate colonization and the settlers were granted 15-25 years free of

taxation. The regulations didn’t accomplish any significant results towards

an increase in settlings until a few decades later. During the 19th century,

there was an increase of immigration from other parts of the country, and

the population of the inland of northern Sweden grew. In the latter part of

the 19th century there was a surge in the commercial interests in forestry

affording new sources of income. The first half of the 20th century saw a

certain increase of movement from the river valleys to the coastal cities, but

not until after the 1950´s has there been a net decrease of population in the

studied area.

4

LQTS a genetic cardiac ion channelopathy - Genotype

LQTS is the most well-known monogenetic heart disease and several

hundreds of mutations in 13 genes have been described (Table 1). Most of

the mutations can be found in 3 genes, the KCNQ1 (40-55%), KCNH2 (30-

45%) and SCN5A gene (5-10%) (14, 17, 18). LQTS is mostly a heterogenetic

disease, meaning that each family carries a mutation unique for their family.

Table 1. LQTS genes.

Gene Syndrome Frequency

%

Locus Functional effect

Phenotype (in

vitro)

KCNQ1 (LQT1) RWS, JLNS 40-55 11p15.5 Loss-of-function

KCNH2 (LQT2) RWS 30-45 7q35-36 Loss-of-function

SCN5A (LQT3) RWS 5-10 3p21-p24 Gain-of-function

ANKB (LQT4) RWS <1 4q25-q27 Loss-of-function

KCNE1 (LQT5) RWS, JLNS <1 21q22.1 Loss-of-function

KCNE2 (LQT6) RWS <1 21q22.1 Loss-of-function

KCNJ2 (LQT7) AS <1 17q23 Loss-of-function

CACNA1C (LQT8) TS <1 12p13.3 Gain-of-function

CAV3 (LQT9) RWS <1 3p25 Loss-of-function

SVN4B (LQT10) RWS <1 11q23.3 Loss-of-function

AKAP9 (LQT11) RWS <1 7q21-q22 Loss-of-function

SNTA1 (LQT12) RWS <1 20q11.2 Loss-of-function

KCNJ5 (LQT13) RWS <1 11q24 Loss-of-function

LQTS Long QT Syndrome, RWS Romano Ward Syndrome, JLNS Jervell and

Lange-Nielsen Syndrome, AS Andersen Syndrome, TS Timothy Syndrome.

Modified from (19).

5

Functional effects

Gene mutations cause changes in the proteins of the cardiac ion channels.

This creates a disturbance in the functional effect of the ion current through

the cell membrane (19) (Figure 1). The mutations can be divided according

to their phenotypical effects into different categories. A loss-of-function

mutation is a mutation that does not create any functional gene product. A

dominant loss-off-function mutation creates a defect gene product that

disturbs the normal functional effect of the wild-type gene. A gain-of-

function mutation results in a gene product that has a new feature, most of

these mutations are dominant. A gain-of-function mutation can be due to a

new amino acid sequence which alters protein function. Moreover, it can be

a mutation in the regulatory regions, which expresses as a new tissue or in a

new stage where it has not previously been active (20).

The KCNQ1 gene (LQT1) encodes the α-subunit of the K+ channel that

generates the slow potassium current (IKs) (Figure 1) (21). During

sympathetic activation with increased heart rate the repolarization duration

is hastened by an increase of the IKs (19). When a KCNQ1 mutation causes

defective IKs the repolarization duration does not shorten appropriately to

sympathic stimulation. This can be seen in an ECG as a prolonged QT

interval that does not adapt properly to increased heart rate.

The KCNH2 gene (LQT2) encodes the α-subunit of the K+ channel and

thereby for the rapid potassium current (IKr). A mutation in this region

causes an effect similar to a mutation in the KCNQ1 gene with reduction of

outwardly IKr and thus a prolongation of the duration of the repolarization.

Both mutations in the KCNQ1 and KCNH2 gene reduces components of the

outwardly potassium current (IK) through the cardiac ion channels.

The third common gene causing LQTS is the SCN5A gene (LQT3) that

encodes the α-subunit of the cardiac sodium channel conducting the inward

sodium current under the depolarization. The result of a mutation in the

SCN5A gene produces an increase in the delayed Na+ inward current and the

consequence is a prolongation of the action potential (AP) of the cardiac

myocytes.

The result of all of these mutations in the different genes thus creates an

arrhythmogenic cardiac condition.

6

Figure 1: Illustration of the effect of altered ion-channel currents on the

ventricular action potential (AP) duration in LQTS. The direction of ion

currents: inward = below the line; outward = above the line. Hatched

rectangles = time location of the effect of mutations in LQT1, LQT2 and

LQT3 on sodium and potassium ion-channel currents. A prolonged AP is

seen when (horizontal arrow) there is an inappropriate gain of function

(GOF) in late sodium current (INa) or loss of function (LOF) in slowly (IKs) or

rapidly (IKr) acting repolarization potassium currents. Reprinted from (21).

Reproduced with kind permission from Elsevier.

Heredity in LQTS

Two hereditary variants of LQTS are known, the most common being

Romano-Ward syndrome (RWS). If one parent carries a mutation the risk is

50 % for each child to develop RWS – this hereditary pattern is known as

autosomal dominant. The other variant is Jervell and Lange-Nielsen

syndrome (JLNS) which is transmitted in an autosomal recessive way. When

both parents carry a mutation, the risk of each child inheriting the mutation

is 50% (RWS) and 25% to inherit both mutations and get the disorder JLNS

7

syndrome. JLNS is a serious disease and is in addition to cardiac

arrhythmias associated with congenital deafness (22).

Prevalence of LQTS

The prevalence of LQTS has gradually increased due to the growing use of

molecular genetic diagnostics which has consequently increased the number

of diagnosed asymptomatic carriers (23). The estimated prevalence has

increased from 1:10 000 in 2000 (24), to 1:5 000 in 2008 (25) and 1:2000 in

2009 (26). In paper IV, we present prevalence data from the area of mid-

northern Sweden (counties of Jämtland, Västernorrland and Västerbotten)

of the first Swedish LQTS founder mutation Y111C in the KCNQ1 gene.

Genetic testing for LQTS

Genetic testing for the LQTS has entered routine clinical practice and is a

valuable instrument in the diagnosis of LQTS. An individual (index case)

with the clinical diagnosis LQTS can be offered genetic testing to identify the

mutation causing LQTS. Genetic testing identifies mutations in up to 75-80%

of clinical affected individuals (27). When a mutation is identified in a

family, this gives the first-degree relatives the opportunity to undergo

genotyping thus providing important information in guiding the

management of individuals.

Founder mutations and founder populations

A founder mutation can be described as a mutation that appears in a limited

gene pool in a population living in the same geographic area and thus being

enriched (15). Environmental factors, socio-ethnical constructions and

characteristics of the mutation itself affect the expression of the disease.

LQTS founder mutations have been found in South Africa (28) and Finland

(29). A population growth during the 17th century and onward made the

population in northern Sweden a breeding ground for the growth of

monogenetic diseases (15). The river valleys of northern Sweden running

from northwest to northeast has separated the populations living within the

river valleys from other populations. This has contributed to sub-isolates of

people living in the same geographic area within the river valleys. Moreover,

Sweden has a unique opportunity of genealogic studies via the

comprehensive population records which allows genealogical mapping of

monogenetic diseases in the population. A LQTS founder mutation with

large numbers of individuals provides an opportunity to study the genotype-

phenotype correlations and impact of modifying factors on the phenotype.

8

Clinical presentation of LQTS - Phenotype

The LQTS phenotype is diverse and may include syncope, sudden cardiac

death and an ECG with or without QT interval prolongation. Specific factors

have been showed to trigger syncopal episodes, where LQT1 patients are

more likely to develop cardiac events during exercise and swimming (30).

Emotions and sudden noise give rise to cardiac events in individuals affected

with mutations in the KCNH2 gene (LQT2), and LQT3 carriers can

experience events on awakening and during sleep. The family history of

syncope and sudden death among first-degree and second-degree relatives is

important in the initial evaluation of a patient with suspected LQTS.

Notably, about 30-50 % of the carriers of a LQTS mutation are asymptomatic

throughout life.

Arrhythmias in the LQTS

The ventricular arrhythmia associated with the cardiac events in LQTS is

Torsade de Pointes (TdP) (Figure 2). TdP is a polymorphic ventricular

tachycardia characterized by a gradual change in the amplitude and twisting

of the QRS complexes around the isoelectric line. The onset of the TdP is

often preceded by a short-long-short sequence in R-R intervals, with the last

sequence interrupting the T-wave (R on T phenomenon). In most of the

cases the TdP is self-terminating and causes palpitations and/or syncope,

but in the worst case the TdP degenerates into ventricular fibrillation (VF)

and causes cardiac arrest or sudden cardiac death (31).

Figure 2. ECG from a loop-recorder implanted in an 18 year old female with

unknown syncope episodes, showing the typical pattern seen in Torsade de

Pointes.

Experimental studies have shown that the abnormal prolongation of the AP

creates triggered activity, e.g., early after depolarizations (EAD), and the

abnormal dispersion of repolarization is the substrate that initiates and

9

perpetuates TdP (32-34). Why TdP stops or why it continues is not known. A

study in a LQTS population of 50 patients including 151 episodes of TdP,

showed that the majority of these episodes (56 %) and QT-related extra

systoles (70%) originated from the outflow tract (35). The outflow tract was

defined as the area inferior to the pulmonary and aortic valve and other

essentially contiguous structures, e.g., the right and left ventricular outflow

tracts, and areas superior to the mitral and tricuspid annulus.

Other ECG changes associated with LQTS

A prolonged QT interval has been shown to be associated with a high-grade

AV-block especially in children (36). T-wave alternans is a macroscopic

every-other-beat variation in T-waves and is a sign of a major electrical

instability, and identifies particular patients with high risk of malignant

arrhythmias (37) (Figure 3). In infants, subtle notches in the T-wave can be

normal. However, in adults these notches are not normal, and subtle

notched T-waves in lead II or V4-V6 should give suspicion of LQTS (38). It

has been recognized that a typical T-wave morphology may be associated

with specific genetic types. However, because there are overlaps in the T-

wave morphology between the affected genes, the T-wave morphology is

difficult to use in prediction of the affected gene in individuals with LQTS

(39).

Figure 3. Twenty-four hour Holter recording from an 11 year old boy with

Long QT syndrome showing T-wave macroscopic alternans.

10

The “Schwartz score”

The “Schwartz score” was presented in 1993 (updated 2011) and proposed as

a quantitative approach in the diagnosis of LQTS (Table 2) (19). The scoring

system evaluates not only the length on the QT interval but includes other

ECG findings together with clinical history and family history.

Table 2. LQTS Diagnostic Criteria Schwartz 2011

Points

Electrocardiographic

findings*

A QTc **, ms

≥480 3

460-479 2

450-459 (men) 1

B QTc** 4th minute of recovery from

exercise stress test ≥480

1

C Torsade-de-Pointes*** 2

D T-wave alternans 1

E Notched T-wave in 3 leads 1

F Low heart rate for age **** 0.5

Clinical history

A Syncope ***

With stress 2

Without stress 1

B Congenital deafness 0.5

Family history

A Family member with definite LQTS

*****

1

B Unexplained SCD ‹30 year of age

among immediate family members

*****

0.5

LQTS long QT syndrome; *absence of medications or disorders known to

affect these electrocardiographic features; **QTc calculated by Bazett

formula QTc=QT√RR (RR in seconds); ***mutually exclusive; ****resting

heart rate below the second percentile for age;*****the same family

member cannot be counted in A and B.

Score: ≤1 point: low probability of LQTS; 1.5-3 points: intermediate

probability of LQTS; ≥3.5 points: high probability. Modified from (19).

11

Risk stratification in the LQTS

In the general LQTS population, and in particular in the LQT1 population

there is a higher risk of cardiac events until puberty in males, whereas

females has a higher risk during adulthood (40). Based on the probability of

a first cardiac event (syncope, cardiac arrest or sudden death) before the age

of 40, and without therapy it has been proposed a protocol for risk

stratification among patients with LQTS according to genotype and sex. For

instance, there is a high risk (≥ 50 %) if the QTc is ≥ 500 ms in a carrier of a

mutation in the KCNQ1 or KCNH2 gene or in a male carrier with a mutation

in the SCN5A gene (41). LQTS patients have variability in the QTc between

follow-up ECGs. The maximum QTc interval contain prognostic information

in addition to the baseline ECG, therefore it has been suggested in the risk

stratification of LQTS to always record follow-up ECGs (42).

Prophylactic treatment of symptoms in Long QT syndrome

With prophylactic medication and minor lifestyle changes it is possible to

effectively reduce mortality and morbidity in LQTS (43). It is thus important

to identify even asymptomatic individuals and for this we need accurate

diagnostic tools. ß-adrenergic blocking agents are the first choice in

prophylactic therapy of LQTS patients (43). This medication is most effective

in individuals with mutations in the KCNQ1 gene, but somewhat less

effective in KCNH2 gene (44). Many of the cardiac events during

prophylactic treatment are thought to be caused by non-compliance or use of

QT-prolonging drugs (45).

Implantable cardioverter defibrillator (ICD) is recommended in patients

with cardiac arrest (43, 46, 47). Left cardiac sympathetic denervation (LCSD)

may be used when prophylactic beta-blocking therapy has failed and/or were

there has been “storms” of repeated defibrillation because of TdP in subjects

with ICD (47).

12

Physiologic basis of the ventricular repolarization

Action potential of the ventricular myocytes in the normal heart

The cardiac action potential (AP) is divided in four different phases (48). The

cardiac AP of the cardiac myocytes is illustrated in the upper part of Figure

1. Phase 0; the rapid depolarization in the ventricular tissue, caused by the

rapid movement of Na+ (INa) through ion channels (when the membrane

potential is approximal -65 mV) into the intracellular spaces of the myocytes.

After just a few milliseconds the ion channels are inactivated and the

channels are closed. Phase 1; a slight repolarization, the upstroke of phase

0 activates the channels which causes the transient efflux of K+ from the

myocytes (“transient outward current” Ito). This brief repolarization phase is

rapidly over and is seen as a notch in the AP. Phase 2; the plateau phase,

this is the major determinant of the duration of the AP and is caused by the

inwardly depolarization of calcium currents (ICa) and the outwardly

repolarization of potassium currents (IK). The ion channels (L-type calcium

channels and potassium channels) have a slow activation rate and are

opened in connection to phase 0, the balance of inward and outward

currents between the competing ion channels determine the duration of the

plateau phase. Phase 3; inactivation of the depolarizing current of calcium

(ICa) is coupled to an increasing net outward potassium current (IK)

consisting of rapid (IKr) and slow (IKs) components of the delayed potassium

channels. Phase 4; during phase 4 the resting membrane potential is

restored (-90 mV) and the baseline potential is maintained by the inward-

rectifier potassium current (IK1).

There are slight differences between the AP in the myocardium and the AP in

the Purkinje fibers in the ventricular conducting system. There is also a

subtle difference between the myocardial layers APs where the mid-

myocardium (M cells) having the longest action potential duration (APD)

(49). The duration of ventricular repolarization is longer in the endocardium

than the epicardium, and shorter in de basal regions compared with the apex

(50). This heterogeneity of repolarization is seen in healthy adult hearts.

The interval between the beginning of Q and end of T in a surface ECG is an

indirect measure of the duration of the ventricular action potentials (51)

(Figure 4). Measured on a surface ECG, the QT interval consists of two

parts, the QRS interval and the JT interval, which reflects the depolarization

(in the His-Purkinje system and ventricles) and the duration of the

repolarization, respectively. The depolarization is spread through the

13

Purkinje fibers from endocardium to epicardium and the repolarization is

synchronized in opposite direction from epicardium to endocardium.

Figure 4. The QT interval corresponds to the interval between the Q wave

and the end of the T-wave. When the QT interval is rate corrected (Bazett’s

formula) the preceding R-R interval (seconds) is used QTc=QT/(√RR).

Prolonged action potential

To maintain the correct balance between internal and external ion

concentrations, it is essential that the ion currents can flow through the cell

membrane of the myocytes. If there is an increase of the inward current or

decrease of the outward current during the repolarization, a prolonged AP

occurs (50). This will be reflected in the ECG and appears as a prolonged QT

interval with alterations of the ST segment and T-wave morphologies.

Abnormalities in the ST segment and T-wave of the QT interval can be

classified as primary or secondary repolarization abnormalities (52). In

absence of changes in the depolarization, the primary changes depend on

changes in the shape and/or the duration of the repolarization phase 3 of the

AP. Such changes may be caused by, e.g., ischemia, electrolyte abnormalities

(Ca++ and K+), myocarditis, ion channelopathies (e.g., LQTS) but also abrupt

changes in heart rate or body position. A change in the QRS that gives rise to

abnormalities in the ST segment and T-wave is called secondary

repolarization abnormalities. These may occur because of voltage gradients

and become manifest due to changes in the depolarization that alter the

repolarization. Secondary ST and T-wave abnormalities occur with bundle-

branch blocks, ventricular pre-excitation and ectopic ventricular beats and

paced ventricular complexes.

Ventricular gradient

Wilson et al. introduced the concept ventricular gradient (VG) that deals

with primary versus secondary repolarization abnormalities (53, 54). The

QRST area in a single ECG lead reflects the summarized local electrical

activity in the myocardium, viewed from the angle of the particular ECG-

lead. Commonly the orthogonal leads X, Y and Z in a VCG are used, since the

14

QRST-areas in the X, Y and Z-leads, together with the angle between the

separate QRS and T areas, are used in the mathematical expression of the

VG. An abnormal T-wave axis with a normal QRS axis indicates primary

repolarization abnormalities. Secondary repolarization abnormalities, e.g., in

left bundle-branch block the ST and T-wave vectors changes in the opposite

direction of the mean QRS vector (52).

Heterogeneities of ventricular repolarization

The differences in the time-course of the repolarization between cells in

different myocardial layers (endocardium, mid-myocardium and

epicardium) contribute to the T-wave of the ECG (55). Voltage gradients,

developing as a result of the different time course of repolarization of phase

2 and 3 in cells of the three myocardial layers, give rise to opposing voltage

gradients on either side of the M region (mid-myocardium), which are in

large part responsible for the inscription of the T-wave (56). In an upright T-

wave, the earliest repolarization is seen in the epicardial cells and the latest

is in the AP of the M cells. Full repolarization of the epicardial cells´ AP and

the M cells´ AP, coincides with the peak of the T-wave and end of the T-

wave, respectively. Therefore, the duration of the M cells´ AP, determines

the QT interval, whereas, the duration of the epicardial AP determine the

QTpeak interval.

Heterogeneities of VR have long been associated with arrhythmogenesis.

Increase of the spatial heterogeneity (hereafter called dispersion) of

repolarization has been identified as the substrate of arrhythmias both in

LQTS and acquired LQTS (55). Triggered activity, which is both substrate

and trigger for TdP in LQTS, may be induced by EADs which in turn may be

induced by accentuated spatial dispersion secondary to an increase of the

transmural, trans-septal or apico-basal dispersion of repolarization (55).

Experimental models of LQT1, LQT2 and LQT3 have been developed from

canine arterially perfused left ventricular wedge preparations (57, 58). Such

studies suggests that the prolongation of the APD in the M cells leads to a

prolongation of the QT interval, as well as an increased transmural

dispersion of repolarization, both of which contributes to development of

TdP (59-61).

The Tpeak-Tend interval (T peak to T end, the last part of the QT interval and

final repolarization) has been proposed as an index of transmural dispersion

of repolarization (62). It has been shown that this interval is increased in

patients with LQTS (63). Tpeak-Tend have been suggested to have a potential

value in predicting the risk of developing TdP (30, 64-66). However, further

15

studies are needed to evaluate these indices of electrical dispersion and the

prognostic value in the assignment of arrhythmic risk.

Manual methods to define the end of the T-wave

It has been shown in several studies that the manual measurement of the QT

interval is bound to have errors because of the difficulties to define the end of

the T-wave. This leads to both intra observer variability and variability

between different observers.

The difficulty to identify the end of the T-wave may be amplified by a fusion

of the U-wave and the T-wave, by a T-wave that coincide with the following

P-wave (high heart rates), by a biphasic T-wave, and by low amplitudes of

the T-wave. There is also a problem to choose in which lead the QT interval

should be determined since no consensus has emerged among different

researchers. In paper I and II we used the method to define the end of the T-

wave that was described by Goldenberg (67) (Figure 4). This method was

proposed to be used in the day-to-day practice in the diagnosis of LQTS and

other repolarization disorders. The longest QT value from the mean of 3-5

ECG complexes in lead II and V5 or V6 should be used. As with other similar

methods, this may lead to elements of subjectivity, in particular when

biphasic T-waves and U-waves interrupt the return of the T-wave to the

baseline.

Another way to measure the QT interval, is by using the intersection of a

tangent to the steepest slope (“peak-slope”) of the last limb of the T-wave

and the baseline in lead II or V5 (Figure 5). The proponents of the tangent

method believe that it gives a greater consistency in the measurement of the

QT interval (68). When comparing the tangent method with the “threshold

method” (T-wave offset when the T-wave reaches the isoelectric baseline) in

healthy individuals with a normal T-waves, the tangent method gives shorter

QT interval by up to 10 ms (69). The tangent method provides longer QT

intervals than the threshold method when used in subjects with changed T-

wave morphology, flat T-waves and U-waves as seen in drug induced QT

prolongation (70).

16

Figure 4. A) Normal T-wave: T end is when the descending limb returns to

TP baseline. B) Separated T and U wave: T end is when the descending limb

of the T-wave returns to TP baseline before the onset of the U wave. C)

Biphasic T-wave: T1 and T2 with similar amplitude, T end is when T2 returns

to TP baseline. D) When a second low-amplitude interrupts the end of the

larger T-wave: T end can both be at the nadir (1) of the two waves and at the

final return to TP baseline (2) U wave? TP baseline corresponds to the

baseline between the end of the T-wave and the start of next P wave.

Modified from (67).

Figure 5. Tangent method: With a tangent drawn to the steepest slope in

the end of the T-wave in lead II or V5. The intersection between baseline and

the tangent is the end of the T-wave. The R to R interval is measured from

the preceding R-R interval.

“Normal” QT interval

Is the QTc normal? This is the question that we face every time we measure

and evaluate the QT interval. In two large population-based studies,

including 12,500 and 40,000 healthy individuals, the following normal

limits of QTc were suggested according to the 97.5th percentiles for the lower

17

and upper limits in healthy subjects: Adult males 350-450 milliseconds (ms)

and, for adult females 360-460 ms (71, 72). In the guidelines from

AHA/ACC/HRS QTc values ≥ 450 ms for men and ≥ 460 for women have

been stated as prolonged QT (52). For the diagnosis of LQTS, Sami Viskin

has proposed an upper limit for prolonged QTc ≥ 470 ms for males and

QTc ≥ 480 ms for females, even if the individual is asymptomatic and have a

negative family history (23). It is unusual that the LQTS diagnosis is based

only on a prolonged QTc - usually more evidence is required (73). The most

accepted QTc limits among both adults and children have been suggested by

Goldenberg et al. (67, 74) (Table 3). Goldenberg’s study further revealed a

difference in QTc depending on age and sex among adults. The QTc

difference depending on sex seems to disappear with old age (75).

Table 3. Bazett corrected QTc values for diagnosing QT prolongation

Children 1-15 y

(ms)

Adult male

(ms)

Adult female

(ms)

Normal QTc <440 <430 <450

Borderline QTc 440-460 430-450 450-470

Prolonged QTc >460 >450 >470

Modified from (67).

Other sources used to define limits for the QT interval are studies of

populations with genetically confirmed carriers and non-carriers of LQTS

mutations (76). Such studies have documented carriers with normal QTc as

well as non-carriers with prolonged QTc. It has been found that the

diagnostic criteria QTc ≥ 450 ms fails to identify 10% of the carriers and

incorrectly diagnose 10% of the non-carriers. It is known that the mutation

carriers with normal QTc account for about 36% of the LQT1 carriers, 19% of

the LQT2 carriers and 10% of the LQT3 carriers (41).

There is a pronounced diurnal variation of the QT interval in the normally

innervated heart. This follows changes in neurally mediated autonomic tone,

mainly parasympathetic, during sleep and on the circadian variation in

circulating catecholamines (77). It has been demonstrated in healthy

individuals that autonomic conditions, independent of the heart rate,

directly affects the ventricular myocardium and causes changes in the QT

interval, and this can complicate the clinical assessment (78).

18

The QT interval corrected for heart rate

In healthy individuals, the QT interval shortens with increasing R-R interval.

Therefore, it has become standard practice to use a correction formula, such

as the Bazett formula (79), to normalize the QT interval to a heart rate of 60

beats per minute (bpm), yielding the rate-corrected QT or QTc = QT/(√RR),

where RR is the preceding R-R interval in seconds. When using the Bazett

formula for heart rates below 60 bpm the QTc is somewhat overcorrected

resulting in too short QTc intervals (false-negative values). The opposite

takes place when correcting heart rates above 90 bpm, where Bazett´s

formula gives too long QT intervals (false-positive values) (67). In borderline

cases, it would be advisable to repeat the measurements during heart rate

> 60 bpm or < 90 bpm. The Fridericia formula (80) QTc=3√R-R has mostly

been used in children because it reflects a more accurate QTc in higher heart

rates, but it has the same limitations at slow heart rates as Bazett´s formula

(67). Numerous other formulas have been suggested that may give a more

uniform correction over a wide range of heart rates (48). However, the

Bazett and Fridericia formulas are the most frequently used because they

provide nearly equivalent results for the diagnosis of QT prolongation in

subjects with normal heart rate, i.e., between 60-90.

The relationship between the QT interval and the heart rate has a substantial

inter-subject variability and a strong intra-subject stability (81). It is also

known that the QT interval duration does not instantly adapts to heart rate

changes, since there is a lag time before the QT is stabilized phenomenon

known as QT-RR hysteresis (82, 83). The lag time has been shown to be

individual and is approximately 2 minutes (84), but may be prolonged in

cardiac disease and may be altered by autonomic perturbations (85).

The corrected QT interval and bundle branch block

In the left and right bundle branch-block (BBB) the excitation time is

prolonged and QT interval prolongation is induced. The easiest correction

would be to subtract 70 ms from the calculated QTc in case of left bundle-

branch block (LBBB), or to subtract 40 ms in right bundle-branch block

(RBBB) (86). To detect prolonged repolarization in BBB, the JT interval or

Bazett’s QTc-QRS has been proposed (87). A study evaluating the previous

proposal was performed in 11,739 individuals with normal conduction and

1,251 individuals with BBB (88), but still there was a residual correlation

between the heart rate and the QT interval (r=0.54) in BBB, whereas the

residual correlation was r=0.32 in the group with normal conduction. When

comparing six different formulas for QT or JT correction in BBB in two

different groups at different heart rates, the Hodges formula [QTcH = QT +

19

1.75 (heart rate – 60)] was found to produce the smallest heart rate

dependency (89).

Other clinical conditions associated with prolonged QT interval

There are other conditions than LQTS that are associated with a

prolongation of the QT interval and may cause TdP. Drugs commonly cause

the “acquired long QT syndrome”, e.g., certain antibiotics, some

antidepressants, antihistamines, diuretics, anti-arrhythmic drugs,

cholesterol-lowering drugs, diabetes medications and some antifungal and

antipsychotic (www.azcert.org). Also cardiac disorders such as chronic heart

failure, cardiomyopathies and bradycardia due to sinus dysfunction or as

mentioned, conduction block has been shown to prolong the QT interval. It

is not uncommon to find a prolonged QT interval in, e.g., people suffering of

anorexia-nervosa (90, 91), or subarachnoidal heamorrhage. Electrolyte

imbalance especially hypokalemia, hypo magnesaemia and hypocalcaemia, is

also a common cause of prolonged QT interval as well as intracoronary

contrast injection and resuscitation (32).

Electrocardiographic recordings

The electromotive forces of the heart can be recorded by two different

systems, the scalar and the vectorial system. The scalar system measures the

difference in the APs of the heart between two leads, a negative and a

positive on the body surface, or a positive on the body surface and a

constructed zero potential. The vectorial system presents the projection of

the electromotive forces of the heart viewed in three mutually perpendicular

directions. In addition to the differences in potential and speed, the

vectorcardiogram adds the direction of the electromotive forces.

Vectorcardiography

The clinical usefulness of vectorcardiography (VCG) is well documented. It

has advantages over the 12-lead ECG, in particular in describing the

increased dispersion in ventricular repolarization (VR) and the association to

T loop morphology (92-95). With a distinct stop, the end of the T-wave loop

is easy to detect with VCG compared with ECG. VCG also gives information

about the spatial orientation of the repolarization and also a more

20

anatomically reliable T vector loop, as compared to a 12-lead ECG. VCG

describes the variation in electrical activity in the heart as a single dipole. In

each moment of a heartbeat, a spatial vector with an arrowhead is depicted

representing the orientation and the magnitude (length of the vector) of the

dipole. When each instantaneous single vector is plotted consecutively in a

heartbeat, continuous vector loops are formed in space for each heartbeat. If

the configuration of all three loops in all three planes are analyzed

simultaneously, it is possible to calculate the vector magnitude by using the

direction and magnitude of the spatial vector (Pythagorean formula);

vector magnitude = √X2+Y2+Z2 (Figure 6).

Figure 6. The vector magnitude constructed from Frank leads X, Y and Z.

The interval between the letters Q and F correspond to the QT interval.

Reprinted from (96). Reproduced with permission from Elsevier.

21

From the three orthogonal leads X, Y and Z, the spatial VCG can be recorded

and projected. Ernest Frank developed a system to compensate for the non-

spherical human torso and the eccentric origin of the heart’s electrical

activity. The three Frank orthogonal leads X, Y and Z are calculated from the

electrodes using the equation shown in (Figure 7) (6, 97, 98).

Figure 7. Frank lead system, with the 3 orthogonal leads X, Y, and Z

recorded by 8 electrodes. Reprinted from (96). Reproduced with permission

from Elsevier.

X = (0.610 × A) + (0.171 × C) − (0.781 × I)

Y = (0.655 × F) + (0.345 × M) − (1.000 × H)

Z = (0.133 × A) + (0.736 × M) − (0.264 × I) − (0.374 × E) − (0.231 × C).

22

If two of the three leads X, Y and Z are depicted in a coordinate system a

separate loop can be constructed for each ECG component P, QRS and T-

loops and presented in three planes (Figure 8).

X; Y Frontal plane

X; Z Transversal plane (or horizontal)

Y; Z Sagittal plane

Figure 8. QRS and T loops calculated from leads X, Y, and Z. The upper

loops showing the QRS and T loop together, the lower only the T loop.

Reprinted from (96). Reproduced with permission from Elsevier.

23

VCG loop characteristics

The depolarization and repolarization of the heart can be described by

analyzing the direction, the morphology and the interrelationship between

the characteristics of each loop.

The T vector orientation with its maximum vector in space under the

repolarization is described in following parameters:

Tazimuth (◦) – is the angle in the transverse plane (X-Z). The angle is

0◦ when the vector is pointing to the left. Forward direction is

defined as 0-180◦ and backward direction as 0-(-180) (Figure 9) .

Televation (◦) –is the angle between the vector and the Y axis. It is 0◦

when the vector is pointing downwards and 180◦ when the vector is

pointing in the cranial direction (Figure 9).

Figure 9. Dotted line: T vector loop, dashed line arrow: maximum T vector

in space. A) Tazimuth expresses the angle in the transverse plane (X;Z)

vector pointing to the left 0˚, forward direction 0˚ to 180˚, backward

direction 0˚to -180˚. B) Televation expresses the angle between the T vector

and the Y-axis perpendicular to the transverse plane (0˚= vector pointing

downward, 180˚= vector pointing in the cranial direction). Reprinted from

(99). Reproduced with permission from Elsevier.

24

The QRS and T vector orientation and their interrelationship can be

described by the orientation of the maximum vector in space.

QRS – T vector angle (◦) – Maximal angle between the QRS and T

vector loop.

The morphology of the T vector loop can be described by Tavplan, Teigenv and

Tarea (93);

Tavplan (µV) -is the mean distance between the sample values of the T

loop and its preferential plane, a measure of the T loops bulginess.

Increased dispersion of the repolarization is reflected as a high value

of the Tavplan (100) (Figure 10).

Figure 10. Tavplan (µV) expresses the bulginess of the T loop in relation to a

preferential plane; Tavplan can also be defined as the mean distance of the T

loop from the preferential plane. Reprinted from (101). Reproduced with

permission from Elsevier.

Teigenv (unitless) expresses the form and symmetry of the T loop by

the quotient between the two highest eigenvalues (≈ diameters; d1

and d2) of the matrix of inertia. Teigenv = d1/d2 where d1>d2,

Teigenv = 1 corresponds to a circular T loop (sign of increased

dispersion of the repolarization) and a higher value correspond to a

normal more elongated T loop. Teigenv can be described as the longest

axis (d1) of the T loop that can be rotated most easily and d2 is the

perpendicular axis to d1 that allows the easiest rotation. There is a

third perpendicular axis d3, but it has a value close to 0 and is

therefore negligible (Figure 11).

25

Figure 11. Teigenv (unitless) expresses the form and symmetry of the T loop

and is the quotient between the highest d1 and d2 (≈ diameters) of the T

loop. Reprinted from (101). Reproduced with permission from Elsevier.

Tarea (µVs) is the spatial area under the curve formed by the moving

heart vector during the J-Tend interval in X, Y and Z leads: (Tx2 + Ty2

+ Tz2)½ .

QRSarea (µVs) is the spatial area under the curve formed by the

moving heart vector during the Q-J interval in X, Y and Z leads:

(QRSx2 + QRSy2 + QRSz2)½.

Other VCG parameters

Tp-e (ms) - Tpeak to Tend, the last part of the QT interval and final

repolarization in the QRST complex.

Ventricular Gradient (VG µVs) describes the dispersion of action

potential morphology throughout the ventricles, (Figure 12).

26

Figure 12. Ventricular Gradient (VG µVs). The spatial ventricular

gradient (VG) sometimes referred to as the QRSTarea is the vectorial sum of

the QRSarea and Tarea vectors, taking into account the angle between them.

VG = (QRSarea2 + Tarea

2 + 2*QRSarea*Tarea*cosine)1/2.

is the angle between the QRSarea vector and the Tarea vector (0° to 180°),

QRSarea and Tarea are the spatial areas between the baseline and the curve

formed by the moving vector during QJ and JT intervals respectively.

Reprinted from (102). Reproduced with permission from Elsevier.

27

Aims

The aim of this thesis was to improve the diagnostics and risk assessment in

the LQTS by

I: Examining electrocardiographic methods for measurement of the QT

interval

II: Determining the precision of QT measurements in a pediatric population

III: Comparing vectorcardiographic parameters between individuals with

mutations in the KCNQ1 or KCNH2 gene.

IV: Investigating if Swedish carriers of the Y111C mutation compose a

founder population which could be an important source for future studies.

28

Materials

Overview of the LQTS study population in paper I-IV

Subjects

The index cases and family members in paper I-III were recruited from the

LQTS Family Clinic at the Centre for Cardiovascular Genetics at Umeå

University Hospital. The healthy volunteers were recruited from hospital

staff at Umeå University Hospital and their relatives. The healthy volunteers

had to fulfill the following criteria: absence of known heart or lung disease;

no medication affecting the cardiac repolarization; and no history of

unexplained syncope or SCD in any relative younger than 40 years of age.

Genetic testing was performed in index cases and family members, but not in

the healthy volunteers.

The LQTS population (by genetic testing confirmed carrier of LQTS

mutation) was matched for age and sex with a control population (healthy

volunteers and healthy family members’, non-carriers of LQTS mutation) 1:1

(paper I-II).

29

Seventy-five LQT1 individuals (≥ 17 year of age) were matched (age and sex)

1:1; and 3 individuals were matched 1:2 with 78 healthy controls. Five LQT1

children (≤ 16 year of age) were matched 1:1; and 19 were matched 1:2 with

43 healthy children (paper III).

In paper IV, index cases and family members were included from the LQTS

Family Clinic at the Centre for Cardiovascular Genetics at Umeå University

Hospital and also through national referrals to the Department of Clinical

Genetics at Umeå University Hospital.

Subjects in paper I, included from 2005 until 2008

The LQTS population consisted of 94 carriers: 84 LQT1 and 10 LQT2, 16

were index cases and 78 family members. There were 25 adult males: 12

carriers of Y111C, 5 R518X, 6 with other LQT1 mutations and 2 with LQT2

mutations. There were 45 adult females: 16 carriers of Y111C, 18 R518X, 6

with other LQT1 mutations and 5 with LQT2 mutations. There were 24

children of whom 10 were boys: 2 carrying Y111C, 6 R518X, one other LQT1

mutation and one LQT2 mutation; 14 were girls: 5 carrying Y111C, 7 R518X

and 2 with LQT2 mutations.

The control population consisted of 28 family members that were genetically

confirmed non-carriers and 66 healthy volunteers. There were 25 adult

males: 11 non-carriers and 14 healthy volunteers. There were 45 adult

females: 13 non-carriers and 32 healthy volunteers. There were 24 children:

10 boys of which 2 were non-carriers and 8 healthy volunteers; 14 were girls

of which 2 were non-carriers and 12 were healthy volunteers.

Subjects paper in II, included from 2005 until 2009

The pediatric LQTS population consisted of 35 mutation carriers: 29 LQT1

carriers (13 Y111C, 14 R518X and 2 other LQT1 mutations) and 6 LQT2

carriers.

The pediatric control population consisted of 10 family members that were

genetically confirmed non-carriers and 25 healthy volunteers.

Subjects in paper III, included from 2005 until 2011

The LQTS population consisted of 118 carriers: 99 LQT1 and 19 LQT2. There

were 36 adult males: 21 carriers of Y111C, 7 R518X and 8 with LQT2

mutations. There were 58 adult females: 26 carriers of Y111C, 21 R518X and

30

11 with LQT2 mutations. There were 24 children: 10 carriers of Y111C and 14

carriers of R518X.

The control population consisted of 121 individuals. There were 30 adult

males: 16 family members that were genetically confirmed non-carriers and

14 healthy volunteers. There were 48 adult females: 16 family members that

were genetically confirmed non-carriers and 32 healthy volunteers. There

were 43 children: 15 family members that were genetically confirmed non-

carriers and 28 healthy volunteers.

Subjects in paper IV, included from 2005 until 2010

The LQTS population consisted of 37 index cases and 21 family members all

carriers of the Y111C mutation in the KCNQ1 gene.

The control population consisted of 84 healthy military recruits, matched for

origin (northern Sweden), corresponding to 168 control chromosomes.

Ethical considerations

All studies were conducted in accordance with the Helsinki Declaration for

Ethical principles for Medical Research Involving Human Subjects (2000).

All studies were approved by the Regional Ethical Review board at Umeå

University (Umeå Sweden) Dnr: 05-127M.

All participants in the study gave written consent including the legal

guardians of the participating children. Information concerning genetic

testing and execution of haplotype analysis was provided to all.

31

Methods

Paper I-III: In all individuals a 12-lead ECG was recorded during rest in a

supine position, and approximately 2 minutes later this was followed by the

recording of a VCG. For VCG recordings, five electrodes were applied around

the chest at the level where the 4th intercostal space meets the sternum, one

in the neck, one on the left hip, and one zero point on the right hip

(according the Frank system) (103, 104). All the recordings were performed

by the same person.

12-lead ECG

Automatic measuring and interpretation of the QT interval

In paper I and II the 12-lead ECGs were recorded with a Mac 5000 version

008B using the equipments 12SL algorithm for automatic calculation and

interpretation of the QT interval (GE Medical System, Information

Technologies, Milwaukee, WI, USA) (105). The paper speed was 50 mm/s,

the amplitude gain 10 mm/mV, and the sampling rate 500 samples per

second. The signal was amplified in a band between 0.31 and 150 Hertz (Hz).

QRS complexes of the same morphology were aligned in time during 10

seconds recording without noises and a representative complex was

generated from the medians obtained at the successive sampling points. The

end of the T-wave was defined on the vector magnitude, which was

calculated from the sum of the absolute value of I, II and V1-V6 and its first-

order difference. The T-wave end was defined as when a new point

contributes to the T area and when that point in the accumulated T area is

below a threshold value (1%). The QT interval was then calculated from the

earliest Q wave in any lead to the T-wave end. According to the equipments

12SL algorithm the QT interval was categorized as normal, borderline, or

prolonged according to the specific criteria of age, sex and heart rate (105)

(Figure 13).

32

Figure 13. Presentation of values and interpretation in 12-lead ECG from

the equipment Mac 5000 version 008B. Calculated parameters: Heart rate

56 BPM, PQ-interval 130 ms, QRS-duration 94 ms, QT/QTc 524/505 ms

and PR-T axis 56, 52 and 39. Interpretation: Slightly slow sinus rhythm

with sinus arrhythmia, prolonged QT-interval, pathological ECG.

Manual measurement of the QT interval

All 12-lead ECGs were coded and, no automatic values or interpretations

were visible for the observers (paper I-II). Four experienced observers

repeated manual measurement of the QT interval in 42 12-lead ECGs on two

occasions with a one week interval between measurements (paper I). All

observers had verbal and written instructions on how to perform the

measurements (Appendices I): the T-wave end was defined according to the

proposal by Goldenberg I et al.(67). The QT interval was corrected for heart

rate according Bazett´s formula. Each observer made a statement based on

cut-off QTc values according to age and gender for diagnosing LQTS (Table

3); normal, borderline (suspected LQTS) or prolonged QT interval (LQTS).

The observer with the lowest intra-observer variability performed the

measurements in the subsequent 146 ECG recordings in paper I and also the

70 ECGs in paper II.

VCG according Frank lead system

Automatic measuring of the QT interval

The CoroNet II system (Ortivus AB, Danderyd, Sweden) was used to

recording all VCGs in paper I, II and III. In paper I the older MIDA1000 was

also used, but there are no relevant differences between the systems. The

orthogonal leads X, Y and Z were recorded in 3-4 minutes, where the signals

were sampled at 500 Hz (amplified band width 0.03-170 Hz). The recording

was performed during period 60 seconds (s) for determination of the

parameters describing the spatial QRS and T vector, T vector loop and QRST

intervals.

33

The average vector magnitude was computed from the X, Y and Z leads

(vector magnitude = √X2+Y2+Z2), the detections points were located in

relation to the Q-point which had the value 0 and the isoelectric line was

located 30 ms before the start of Q. The detection point Q to F corresponded

to the QT interval (Figure 6), and the mean heart rate for one minute was

used in the automatic calculation of the QTc.

The CoroNet II algorithm defines the beginning of QRS and the end of the T-

wave from the computed derivative of the vector magnitude. In order to find

out the detections points for T end, the algorithm starts from the T peak and

seeks forward until the magnitude decreases with 1/3 of the value of T peak.

The algorithm seeks further forward to a predefined smaller value, and when

the slope of the magnitude crosses that point the algorithm defines this as

the T end. In a similar way the detection point of the QRS start is defined,

from the highest derivative of the QRS complex and backwards until the

magnitude has been decreased to 1/10 or a predefined value. When the

detection point of T end is defined and a U wave is an early part of the T-

wave, the U-wave was included in the QT interval. If the U-wave comes after

the T-wave, it is not included in the QT interval.

Off-line analysis of the VCG

Paper III: All off-line analyses were performed by one observer. Detection

points were automatically positioned and presented in the vector magnitude

(Figure 6). The last minute of the recording was chosen for analysis; if there

were extrasystolic beats the previous minute were chosen. The end of the T-

wave was manually assessed by the tangent method (106). Roughly 90 % of

the T-wave end detection points needed manual adjustment. After all

recordings had been reviewed, the data was processed and the VCG

parameters were determined.

Genetic testing

Genomic DNA was prepared from whole blood using the standard salting out

method. Evaluation was performed in order to confirm the presence of

mutation in KCNQ1 and KCNQ2 genes. Amplified Polymerase Chain

Reaction (PCR) fragments were screened for sequence variants by

denaturing high-performance liquid chromatography. In case of variant

elution peaks, further analyses were performed with DNA sequencing

(CEQ8000 sequencer Beckman Coulter, Fullerton, CA). When a mutation

was identified in an index case, the family members subsequently underwent

34

direct analysis using MGB-probes by TaqMan 7000 (Applied Biosystem,

Carlsbad, CA). Based on the molecular diagnosis each subject was classified

as either genotype positive or genotype negative for LQTS (Paper I-IV).

Genealogy, geography and haplotype analysis

Paper IV investigated if the high number of index cases and family members

carrying the Y111C mutation in the KCNQ1 gene constitutes a population

sharing a common haplotype, together with common geographic and

genealogic origin. The age of the mutation and prevalence were also

calculated.

Genealogical and geographic analysis

The genealogic searches were performed by tracing all maternal and paternal

ancestors of 37 index cases, and then connecting as many index cases as

possible in as few generations as possible with a common couple (founder

couple). The maternal or paternal ancestors connecting to the index cases

and the common founder couple were defined as obligate carriers of the

family mutation. All obligate carriers including siblings and spouses were

traced: sex, place of birth and death, were noted if possible. Spouses were

traced to investigate the marriage pattern within or outside the river valley.

The birth places were traced to establish a possible clustering of ancestors

within a geographic area.

According to the population records initiated by the Crown in Sweden in the

year of 1686, the priesthood was obliged to execute annual catechetical

examinations in their local parishes and register the inhabitants by including

births, deaths, marriage and migration. The documentation before 1686 was

often based on e.g., estate inventory proceedings and court proceedings.

These registries give us a unique opportunity to investigate our ancestors.

Genealogical data were collected from church registries, parish books, and

census archives at the Research Archive at Umeå University. Data were also

collected from the Swedish archive information homepage (www.svar.ra.se)

and researcher’s private genealogic database.

All genealogic data were entered and stored in DISGEN computer software

(Lakewood, CO, USA). Open Source software (Inkscape vector graphic

editor) and GIMP (GNU image manipulation program) were used to

construct pedigrees and geographic maps.

35

Below is an example from a church registry of people who died in the parish,

with names and causes of death documented. The individuals in the example

registry have no connection with the LQTS families.

Haplotype analysis

To determine if the index cases and their family members were related and

had the same genetically origin due to founder effect, we examined if they

shared a common haplotype. The haplotype represents alleles inherited

together in related individuals. The closer the microsatellite markers are to

each other, the smaller is the probability of recombination.

We performed haplotype analysis in each index case whenever possible,

including two generations from each family. These analyses were compared

with 84 healthy controls of north Swedish origin.

The point mutation c.332A>G p.Y111C is located in the KCNQ1 gene on the

short arm (p) of chromosome 11 at position 15, where the base A have been

changed to G causing the amino acid to exchange from Tyrosin to Cystein.

We investigated 5-6 microsatellite markers upstream and 9 downstream

from the point mutation. These microsatellites are located over a physical

distance of approximately 8 x 106 base pairs of the chromosome. To work out

the pattern of shared alleles, we reconstructed the most likely ancestral

haplotype. To compare the frequency of disease associated alleles in healthy

non-carriers, analysis of the same microsatellite markers were performed in

168 control chromosomes in 84 healthy controls. We used the National

Centre for Biotechnology Information Entrez Gene database, and the

deCODE, Généthon and Marshfield genetic maps, to choose markers for the

haplotype analysis. The markers with the highest heterogeneity according to

the CEPH Genotype database were selected. For each separate marker

forward and reverse primers flanking the microsatellite region were used

(Sigma-Aldrich Inc.). Fragment analysis was performed according to

manufactory instructions, and the PCR product was analyzed using an

automated capillary electrophoresis–based DNA Sequencer (Wave® 3500

HT, Transgenomic Inc, Omaha, Nebraska). GE Healthcare (United

36

Kingdom) and Applied Biosystems (Foster City, California) supplied

solutions and material for the PCR mix. GeneMapper software version 3.7

(Applied Biosystems Inc, Foster City, California) was used to analyze the

microsatellite data.

Mutation age and prevalence of the Y111C mutation in the KCNQ1 gene

The prevalence and age of a mutation can be investigated by analyzing the

haplotypes in mutation carriers from a common descent. To possibly reduce

the uncertainty in these calculations two different computer programs were

used. The extent of shared alleles in index cases, the frequency of disease

associated alleles in healthy controls, and the recombination frequencies for

the 15 microsatellite markers was computed in the ESTIAGE software (paper

IV). The software estimated the age and determined the 95% confidence

interval (CI). The mutation age was defined as the age of the most recent

common ancestor of the index cases. Recombination frequencies were

estimated using the distance from the Y111C mutation and the markers, and

the standard correspondence 1 cM = 106 base pairs.

The DMLE software (www.dmle.org) was used to estimate the prevalence of

the mutation. Data were entered including the haplotypes of index cases and

controls. The average population growth rate per generation was estimated

based on regional demographic data from the Jämtland and Västernorrland

Counties from the period 1570-1950 in Statistic Sweden (www.scb.se). The

“true value” of the mutation age was assumed when the confidence interval

(CI) of the DMLE and ESTIAGE software overlapped.

Statistic methods

Data were analyzed using SPSS version 16.0 and 18.0 (SPSS Inc. Chicago, IL,

USA) in paper I-III. All values were calculated as mean, range, standard

deviation (SD) and min-max. To assess and identify the efficacy in

diagnosing LQTS for the different electrocardiographic methods (paper I and

II), sensitivity, specificity and positive and negative predictive value (PPV

and NPV, respectively) were computed.

In paper I-II, we compared the electrocardiographic methods by computing

receiver operating curves (ROC), and the area under the curve (AUC) (SAS

version 9.1.3; SAS Institute, Cary, NC). For nonparametric comparison of

AUC, the SAS-macro %ROC version 1.6 was used.

37

Relative error (paper I) and ĸ coefficient were used to assess inter- and intra-

individual observer variability and the agreement between the QT

measurements. Relative error was calculated as; absolute error/mean error

of the pair A – B/[(A + B)/2)]. Absolute error was calculated as the error

from two pairs (A, B) from 2 occasions of measuring as the absolute values of

their difference A – B.

Molecular genetic diagnoses were used as gold standard when comparing the

measured QTc between the different methods in paper I and II.

A value of the ĸ coefficient > 0.75 was considered as excellent agreement,

and values < 0.40 was considered poor agreement. p ≤ 0.05 was considered

statistically significant.

Regression analysis was used to investigate the influence of sex, age, KCNQ1

mutation on the VR (paper III). As we found a strong relationship between

age and most VR parameters, we divided the LQTS population into two age

groups: ≤ 16 y (children), and ≥ 17 y (adults), respectively.

We analyzed if healthy volunteers and non-carriers of the family mutation

could be merged to one control group. Linear regression was used to

compare controls and KCNQ1 mutation carriers.

Limitations

Paper I – III: DNA analysis was not performed in healthy volunteers in the

control group. However, the risk for an individual to be carrier of LQTS in

the healthy group is only 1/2000 (0.005%). The generalizability in paper I-

III may be limited because the LQTS populations were composed preferably

of KCNQ1 gene mutations and a low number of KCNH2 mutations, but no

SCN5A gene mutations.

The electrocardiographic recordings were performed without consideration

of the diurnal variation or day-to-day variation of the QT interval. More

stringent registration times may possibly have given somewhat different

results. Paper I, II and III: No consideration was taken to beta-blocking

therapy in the analyses of the LQTS population. However, in study III we

showed that in our LQTS population there was only a small change in the QT

interval (16 ms) caused by beta blocking therapy and no changes in the QTc.

38

Paper IV: DNA was available for haplotype analysis in 26 out of 37 index

cases, however, we found that index cases with or without genealogic

connection to the common couple shared alleles to the same extent. This fact

makes it likely that the analyzed index cases sample is representative.

Twenty-six out of 37 index-cases were connected in a pedigree to a common

couple. Some of the reasons why not all could be connected in the pedigree

could be that a child had other parent/s then as stated in the church books,

or because of destroyed church books, or that incorrect data were entered by

the authorities (e.g., the vicar).

39

Summary of Results

Paper I and II

We found a considerable intraindividual and interindividual

variability in the manual measurement of the QT interval in 12-lead

ECG among 4 observers.

The automatic interpretation of the QTc from 12-lead ECG showed

the lowest sensitivity (0.40). It provided 38 out of 94 mutation

carriers of LQTS with a correct diagnosis.

Automatic QTc measurement using VCG showed the highest

sensitivity (0.90) and a positive predictive value (PPV) 0.89 and an

AUC 0.948. It provided 85 out of 94 mutation carriers of LQTS with

a correct diagnosis based on an age and sex limited QTc value.

Also in a pediatric population VCG had the highest accuracy for a

correct diagnosis, 30 of 35 (86%) mutation carriers were correctly

diagnosed as LQTS and 28 of 35 (80 %) non-carriers received the

correct diagnosis.

Automatic measuring with VCG, 12-lead ECG and manually

measuring in 12-lead ECG showed roughly the same specificity 0.87-

0.89. In the pediatric study specificity was nearly equal in the four

methods.

Paper III

In adult men and women ≥17 years of age:

A significant longer QT interval (p=0.037) was found in the

symptomatic group when comparing 23 symptomatic and 52

asymptomatic KCNQ1 mutation carriers.

There was a significant difference (p=0.02) in QTc between the

populations carrying Y111C and R518X mutation in the KCNQ1 gene,

where the Y111C population had a longer QTc.

There were a significant difference between the LQT1 population

and the control population were the LQT1 population had longer

QTc (p=0.000) and longer Tp-e interval (p=0.000).

The QTc and Tp-e/QT was longer in LQT2 carriers and the Tarea and

Tamplitude were greater in LQT1 carriers, but all within a normal range.

40

In children 1-16 years of age:

There were a significant difference between the LQT1 population

and the control population, were the LQT1 population had longer QT

interval (p=0.009) and longer QTc (p=0.000), while the Tp-e/QT

ratio was larger in the control population (p=0.000).

Paper IV

The Y111C population being a founder mutation is supported by:

Genealogic studies connected 26 of 37 living carriers of the Y111c

mutation in the KCNQ1 gene through a line of obligate mutation

carriers to a common founder couple born 1605 resp. 1614 (Figure

15).

Geographic clustering back to a common geographic origin in the

region of the Ångerman river valley and their tributaries were traced

in ancestors to all 37 index cases (Figure 16).

Haplotype analysis traced an original ancestral haploblock in the

DNA in 26 index cases and 21 family members.

The age of the mutation was calculated to 600 years (CI 450; 850)

and 575 years (95% CI 400; 900) with the ESTIAGE and DMLE

software, respectively.

A prevalence of 1:1,500 – 3,000 carriers was calculated for the Y111C

mutation in the KCNQ1 gene in the counties of Jämtland,

Västernorrland and Västerbotten.

41

Figure 15 Twenty-six index cases carrying the Y111C mutation in the

KCNQ1 gene connected through obligate carriers of the mutation to a

common couple born in 1605 and 1614 spanning 13 generations. =Index

cases; H=Index cases with haplotype data. Reprinted from (107).

Reproduced with permission from Elsevier.

42

Figure 16. Birthplaces of the 37 index cases and their ancestors, all carriers

of the Y111C mutation in the KCNQ1 gene. Birthplaces in the 19th century are

depicted in a map illustrating the geographic clustering of the mutation (grey

shaded area). Depicted lines illustrate the rivers and tributaries that flow

from northwest to southeast and the dots indicate the birthplace of the

ancestors. The grey shaded geographic area of interest is magnified in the

upper left corner. Reprinted from (107). Reproduced with permission from

Elsevier.

43

Discussion

Manual measurements

Long QT is a widespread and potentially lethal disease because of the risk of

the feared arrhythmia - known as TdP. The warning sign could be a noticed

long QT interval, but it has been shown that many physicians, including

many cardiologists, do not recognize a prolonged QT interval (108). This

may be due to the difficulty in defining the end of the T-wave and/or a lack

of experience in performing the measurements, which produces a significant

intra- and inter-variability in the measurements of the QT interval. It is of

utter importance to accurately measure and identify a long QT interval - both

for the individual that is at risk of arrhythmia and for those without risk. In

children, manual measurements of the QT interval are even more difficult

due to high and variable heart rate (sinus arrhythmia).

In accordance with other studies (1, 108), we found (paper I) considerable

problems with the manual method for measuring QT intervals. In a study by

Postema et al., 151 students with no experience in manual measurement of

the QT interval were taught to measure the QT interval by the tangent

method (Figure 5). They measured and calculated QTc and interpreted four

ECG (two normal and two LQTS) on two occasions and their results were

compared with those of arrhythmia experts, cardiologists and non-

cardiologists (109). The students achieved a correct classification rate of 71%

and 77% on the first and second occasion, compared with 62% for the

arrhythmia experts and less than 25% for the cardiologists and non-

cardiologists. As a consequence the authors proposed that the tangent

method should be used to achieve a better stratification for individuals that

are at risk for sudden cardiac death (SCD).

The tangent method is reliable and easy to learn and has acceptable

variability. However, the tangent method does not account for the entire QT

interval and is problematic in T-waves with low amplitude (110), which may

have implications for diagnosis and risk stratification. In paper I and II we

used the method, as proposed by Goldenberg et al., since it includes the

whole end of the T-wave and thereby the entire QT interval. The tangent

method was used in the VCG off-line analysis (paper III), which makes study

III comparable with other studies performed with the same VCG equipment

(93, 102, 111-114).

In VCG recordings, the automatically measured QTc were compared with

QTc manually measured, using the tangent method (unpublished results).

44

The study population consisted of both adults and children including 105

individuals with DNA verified LQTS and 118 healthy controls. In accordance

with other studies we found that the mean QTc in all individuals was

significant shorter, with the tangent method (427±35 ms) compared with the

VCG method, (445 ±35 ms, p<0.0001) (Table 4) (69).

Table 4. QTc measured automatically with the VCG equipment and

measured manually using the tangent method in the same recordings

(unpublished results) Mean± SD;CI

All

n=223

Mean± SD;CI

Control

population n=118

Mean± SD;CI

LQTS

population

n=105

VCG QTc ms,

automatically

445±35

440;450

421±19

418;425

472±31

465; 476

VCG QTc ms,

manually tangent

427±35

423;432

404±21

400;407

454±29

447; 458

QTc: Heart rate corrected QT interval (Bazett`s formula); SD: Standard

deviation; CI: Confidence interval; Control population: DNA confirmed non-

carriers of a LQTS mutation and healthy volunteers; LQTS population: DNA

confirmed carriers of a LQTS mutation; ms: milliseconds; VCG

vectorcardiogram

Automatic methods

To overcome the intra– and inter-individual variability in manual

measurements of the QT interval, a reliable automatic method would be

valuable. But in paper I and II, we showed that automatic ECG interpretation

may lead to an erroneous diagnosis. A study of 97 046 ECGs analyzed by the

Marquette 12 SL ECG Analysis Program (GE Healthcare), a system that also

was used in this thesis, showed a manifest underreporting of prolonged QT.

Out of 16 235 (16.7%) ECGs with prolonged QTc only 7709 (47.5%) were

stated as “prolonged QT” (115).

We showed in paper I a higher accuracy in automatic QTc measurement by

VCG. The VCG recordings with electrodes applied according to the Frank

lead system (Figure 7) and the algorithm that performed the automatic

measurements of the QT interval provided the best combination of

sensitivity and specificity (AUC 0.948).

45

The higher accuracy in automatic QTc measurement by VCG may be

explained by that the vector magnitude is calculated from three orthogonal

leads and that there are electrodes that cover the posterior parts of the heart

(Z lead).

When investigating a pediatric population of LQTS mutation carriers and

healthy controls we found that, based on QTc and a QTc limit of 440 ms,

VCG identified 30/35 (86%) of the carriers. VCG incorrectly diagnosed five

children as having LQTS. The automatic interpretation of the 12-lead ECG

identified only 17/35 (49%) of the carriers. One reason to VCGs superiority

and high precision in measurements of QT interval in a pediatric population

could be that the VCG recording is longer compared to a standard ECG

recording. The sampling time for the heart rate was 60 seconds for the VCG

and 10 seconds for 12-lead ECG. This may have influenced the QTc,

especially in a pediatric population with varying heart rate. When comparing

the calculated mean heart rate between VCG and 12-lead ECG in 70 children,

the mean heart rate was nearly the same. But, when analyzing each

individual’s heart rate, calculated with both methods, we found that in 14 out

of 70 children (20 %) the heart rate differed ≥10 bpm (max 22 bpm) between

VCG and 12-lead ECG (unpublished data).

QTc cut-off values

There are carriers of LQTS mutations with normal QTc, and there are

healthy individuals with prolonged QTc interval, as has been shown in

earlier studies (73, 76). It is thus not possible to define a QTc value adjusted

for age and sex that can sort out all carriers of a LQTS mutation from healthy

individuals.

The distribution and different cut-off values of QTc were examined in 81

adult carriers (50 women) and 82 non-carriers (52 women) (unpublished

results). Applying QTc cut-offs 430, 440 or 450 ms in the control population

would state 24%, 12% and 2%, as having prolonged QT interval. If the same

cut-off limits were used in the LQTS population, 96%, 90% and 80% would

have prolonged QT interval. Table 5 show the percentages calculated for

both populations. In Figure 14, individual QTc values are plotted against

age for controls and carriers of a LQTS mutation, where lines mark the

different investigated cut-off limits for prolonged QTc.

Over time there is a significant variability in the duration of the QTc. At

baseline 25 % of 375 LQTS children had QTc ≥500 ms while 40 % during the

46

following 10 years at some point had QTc ≥500 ms (42). This highlights the

importance of performing repeated ECG measurements of the QTc for the

diagnosis of LQTS, and to investigate the clinical and family history before

deciding to perform genetic testing (73).

Table 5. The effect of different QTc cutoff values on the statement -

prolonged QT.

QTc cut-off

Statement

prolonged

QT

Control

Population

LQTS

Population

All

n=82

Female

n=52

Male

n=30

All

n=81

Female

n=50

Male

n=31

430 ms 20

(24%)

3

(10%)

78

(96%)

28

(90%)

440 ms 10

(12%)

9

(17%)

1

(3%)

73

(90%)

47

(94%)

26

(84%)

450 ms 2

(2%)

2

(4%)

65

(80%)

45

(90%)

QTc: Corrected QT interval (Bazett formula); Control population: Healthy family members and healthy volunteers; LQTS population: Genetically confirmed carriers of a LQTS mutation.

Figure 14. QTc automatically measured with VCG against age for controls

(left) and the LQTS study population (right). Lines show different QTc limits.

47

QTc cut-off levels for LQTS diagnosis were set to 430 ms, 440 and 450 ms

for males, children and females respectively (paper I and II). These values

correspond to the lower limit of borderline QTc, as proposed by Goldenberg

et al. (67) (Table 3). These QTc limits should not be considered as absolute

limits for a LQTS diagnosis, but as a wake up bell raising the question - could

this be a LQTS patient?

The “ideal” QT measurement

The ideal ECG equipment would give us an accurate QT measurement

without the manual procedure's problems. It has previously been shown that

automated QT measurement provides considerable less variability (116).

Modern ECG equipment provides digital on-line recordings with automatic

measurements and interpretation of the ECG. It is however still important

that the ECG is examined by an experienced person.

A standardized algorithm and a uniform definition of recording

circumstances would improve the diagnostic of diseases that alter the QT

interval. It would also improve the possibilities to compare values from

different studies. This is difficult today where several different methods are

in use for QT measuring and often, also in high-impact LQTS studies,

without accounting of how the QT intervals were measured (method, lead,

etc.). ECG recordings are performed in various settings by staff with varying

educations. To maintain the quality and the clinical value it is important to

have well educated paramedical staff that correctly perform the ECG

recordings, knowing the importance of: correctly placed electrodes,

especially the precordial leads, in relation to body surface anatomical

landmarks, connecting the correct electrode cable to the correct electrode,

and assuring that there is minimal interference from skeletal muscles and

artifacts from skin-electrodes or implanted or external electrical or electronic

devices that otherwise hampers the use of the recordings.

Electrophysiological phenotype

The clinical risk stratification in LQTS is based on LQTS type, age, sex,

symptoms, family history and QTc duration (41). Within a family and

between families affected by the same LQTS type and even the same

mutation QTc can vary considerably. We have shown in paper I and II that

VCG accurately estimates QT interval as compared with 12-lead ECG (96,

48

117). VCG may be more sensitive to changes in VR and provide more

information than just the duration of the QT interval (101). Therefore, VCG

may be a valuable contribution in the assessment of increased VR dispersion.

The Y111C and R518X mutations in the KCNQ1 gene have different effects on

the IKs ion channel function (118). R518X is a nonsense mutation causing a

≤50% reduction of the channel function (haplo-insufficiency). The Y111C

mutation is a point mutation with a dominant negative effect with >50%

reduction of channel function (>75% has been shown in vitro) (119).

The electrophysiological phenotypes (e.g., VG, Tarea, and Tazimut) of the two

mutations were examined with VCG. The findings were compared with a

healthy control population and a population of mutation carriers in the

KCNH2 gene. Data analyses showed that age, sex and carriership of a LQTS

mutation affected most VR parameters; consequently, the population was

divided into adult men, adult women (≥17 years of age), and children 1-16

years of age. All recordings were performed during rest with the individuals

in supine position, which might explain why we did not find signs of

increased dispersion in the ventricular repolarization except for a longer QTc

in the LQTS populations. We have earlier found that carriers of the Y111C

mutation in Sweden has a low incidence of life threatening cardiac events

despite the severe functional defects shown in in vitro studies (120).

Is this a sign of modifying factors? Could unknown mechanisms compensate

the impaired channel function by physiological redundancy expressed as

retained repolarization reserve?

A more prolonged QTc and a higher number of cardiac events have been

found in LQTS populations with dominant negative effect compared with

haplo-insufficiency (119). The clinical phenotype is linked to the mutation

and its biophysical effects on the cell membrane. It has been shown that the

localization of the mutation has importance, a first cardiac event were seen

more often in missense mutations and mutations located in the trans-

membrane part of the protein rather than C-terminally located mutations

(119).

49

Ventricular repolarization

In healthy individuals it has been shown that the VR duration shortens and

the QRSarea, Tarea and Tamplitude decreases with increased heart rate (113). The

conclusion was that an increased heart rate in healthy individuals decreases

the ventricular depolarization instant, AP morphology and thereby the global

VR. We found no significant difference in heart rate and no significant

changes in VG, Tarea and QRSarea in paper III.

Increased autonomous nervous activity can also be critical in LQTS if the

repolarization duration does not shorten appropriately during sympathetic

stimulation. Pharmacologic perturbations of autonomic nervous system

(ANS) and the response of VR parameters recorded by VCG have been

studied in healthy individuals (102). β-adrenergic stimulation after

muscarinic blockade (atropine) gave a transient prolonged QT interval and

delayed heart rate adaption of Tarea and VG, thus mimicking pathological

conditions with increased risk of arrhythmia such as LQTS. In Y111 carriers

syncope was most commonly seen in association to physical activity or

emotional stress (120), as also has been shown to be a trigger in other

KCNQ1 mutations.

Risk markers

LQTS is a genetic heterogenic inherited disease with variable penetrance

(30-50% of the carriers never have symptoms) and expressivity. More

knowledge of the association link between genotype and phenotype is

needed. The degree of QT interval prolongation is a known risk marker for

syncope and/or cardiac arrest in LQTS. Despite that, syncope and/or cardiac

arrest happens in 5% of LQTS mutation carriers with a normal QT interval

(121).

Other diagnostic parameters and risk markers than QTc have been evaluated

in different LQTS populations. Increased short-term variability of the QT

interval in drug-induced LQTS and AP instability were useful in the

predicting arrhythmia (122, 123), and may be used as an additive non-

invasive diagnostic marker in LQTS (124). We previously performed a beat to

beat analysis of the QT interval and found that most VR parameters during

resting conditions showed a significant larger instability (by a factor of 2) in

carriers of a KCNQ1 or KCNH2 mutation compared with a control

population (114). The beat to beat instability of the QT interval increased

with increasing QT interval, also seen as increased VR dispersion measures -

50

Tarea (global dispersion) and VG (action potential morphology dispersion).

This may be a marker of VR variability with valuable clinical implications,

but this has to be confirmed in further studies with additional clinical

(phenotype) and genetically (genotype) data (125). It would be valuable to

perform such studies in populations with a high number of individuals with

the same mutation.

Founder population Y111C

Founder populations are of special interest because they give, in a genetic

homogenous population, a possibility to investigate the relation between

genotype, phenotype and modifying factors.

With our genealogical, geographical, epidemiological, and haplotype data we

have identified a founder population in the area around the Ångerman river

valley consisting of carriers of an Y111c mutation in the KCNQ1 gene. The

mutation was estimated to date back to the 15th century. The immigrants to

the inland of the northern river valleys lived relatively isolated from other

river valleys, the geographic isolation created condition for the occurrences

of founding of genetic sub-isolates. We investigated the marriage-pattern of

the obligate carriers in the pedigree and found that the spouses under the

17th and 18th centuries were born in the Ångerman river valley. Other LQTS

founder mutations have been described, e.g., the A341V founder mutation in

the KCNQ1 gene in South Africa (126), for which in vitro studies found it to

be functional mild although the clinical phenotype is severe. These findings

have given valuable contributions to the understanding of risk modifiers in

LQTS.

When we started the LQTS Clinic at the Centre for Cardiovascular Genetics

at Umeå University Hospital we found a high number of families (unaware of

each other) carrying the Y111C mutation in the KCNQ1 gene. We have earlier

described the mutation to have a mild clinical phenotype with a low 0.005%

annual incident of life- threatening events (120). The mild phenotype

enabled the enrichment in the isolated river valley and has contributed to the

high number of today living carriers of the Y111C mutation (> 200 carriers

2013).

These findings do not correspond with the severe in vitro

electrophysiological characteristic of the mutation which has >75%

mutation-associated reduction of Ks channel function (118, 127).

Consequently, there must be other unknown factors affecting the clinical

51

phenotype. This large founder population is a valuable basis for future

studies of phenotype and genotype correlations and modifying factors that

affect the disease.

Conclusion

VCG was found to be an accurate method for the initial diagnosis of LQTS.

QT interval measurements performed with vectorcardiography have high

precision in a pediatric population.

Ventricular repolarization analysis by resting VCG showed a significant

difference in QTc between Y111C and R518X mutations in the KCNQ1 gene.

VCG showed no increased ventricular repolarization dispersion in KCNQ1

and KCNH2 mutation carriers.

We have identified the first Swedish LQTS founder population carrier of the

Y111C mutation in the KCNQ1 gene. This founder population will be an

important source to future LQTS research in diagnostics methods and the

correlation of phenotype, genotype and modifying factors.

52

Acknowledgements

My first thanks and tokens of appreciation must be directed to Steen M

Jensen, my principal supervisor and workmate through many years. I thank

you for the faith and support you have awarded me and the help that has

attended me throughout the process resulting in the completion of this

thesis. You have freely shared your friendship and your cardiological

insights.

Next I want to express my great gratitude to my co-supervisor Annika

Rydberg. You are always there when you are needed with your

encouragement, boosting my self-confidence and sharing your generous

friendship and knowledge.

Thanks to my co-authors and the members of our LQTS research group of

which Steen M Jensen and Annika Rydberg are part and also:

Annika Winbo, You are a whirlwind! Thank you for the inspiring

conversations and the fruitful collaboration.

Eva-Lena Stattin, I thank you for sharing your insights in genetics, always

willing and able to give good advice.

Lennart Bergfeldt, the Sahlgrenska Academy at the university of

Gothenburg, my co-author in paper III, thank you for your amiability, good

advice, and your liberality in sharing your vast knowledge.

Farzad Vahedi, the Sahlgrenska Academy, University of Gothenburg, my

co-author in paper III, thank you for en enriching collaboration and your

hospitality at my visit in Gothenburg.

Milos Kesek, co-author in my first work – thank you for sharing your

knowledge.

Johan Persson, statistician and co-author of paper IV – thank you for the

analysis of the “statistics of the founder effect”.

Further, I would also like to thank:

Marcus Karlsson for all the invaluable help and all the good advice you

have been giving me through the years.

53

Urban Wiklund always willing to help, thank you so much.

Ina, Mona, Barbro and Anita, you are fantastic – thank you for all the

amusing family receptions.

Stefan Escher for introducing me into the world of genealogy.

Mats Danielsson and the staff at the Research Archive at the University of

Umeå, for all the good advice and aid in my genealogical research, and the

interpretations of “fuzzy handwritings” in the church registers.

Rolf Hörnsten for your support and for finding the right ECG–strips.

Urban Hellman, thank you for the inspiring conversations in “genetics”.

Katarina Englund, my colleague and friend at the Center of

Cardiovascular Genetics, for your existence, for always being there during

my absence and splendidly running our daily work.

Ulf Näslund, Stellan Mörner, Kerstin Franzen thank you for

providing time, your support and working conditions.

Kerstin Rosenquist for the administrative service during the making of

this thesis.

The staff at Clinical Physiology, Clinical Genetics and Center of

Cardiovascular Genetics, thank you for all the help and support you’ve

given me towards the completion of this thesis.

My family – my children and their families especial John, and first and

last you, Thomas – thank you for all the happiness, the love and the

support.

There are many people as have stood me by in the making of this thesis, not

mentioned in these acknowledgments. I send you all my thanks and my

appreciation.

This thesis was supported by grants from the Swedish Heart-Lung

foundation and Northern County Councils Cooperation

Committee.

54

References

1. Murray A, McLaughlin NB, Bourke JP, Doig JC, Furniss SS, Campbell RW. Errors in manual measurement of QT intervals. Br Heart J. 1994 Apr;71(4):386-90. PubMed PMID: 8198894.

2. Waller A. A demonstration on man of electromotive changes

accompanying the heart's beat. J Physiol. 1887;8:229-34.

3. Einthoven W. Üeber die Form des menschlichen Electrocardiogramms. Arch f d Ges Physiol. 1895;60:101-23.

4. Einthoven W. The different forms of the human electrocardiogram and their signification. Lancet.1912(1):853-61.

5. Mann H. Method of analyzing the electrocardiogram. Arch Int

Med. 1920;25:283.

6. Frank E. An accurate, clinically practical system for spatial vectorcardiography. Circulation. 1956 May;13(5):737-49. PubMed PMID: 13356432.

7. Wilson NF JF, Macleod AG, Barker PS. . Electrocardiograms that represent the potential variations of a single electrode. Am Heart J. 1934;9:447-58.

8. Barnes AR, Pardee HEB, White PD, et al. Standardization of

precordial leads. Am Heart J 1938;15:235-9.

9. Jervell A, Lange-Nielsen F. Congenital deaf-mutism, functional heart disease with prolongation of the Q-T interval and sudden death. Am Heart J. 1957 Jul;54(1):59-68. PubMed PMID: 13435203.

10. Romano C, Gemme G, Pongiglione R. [Rare Cardiac Arrythmias of the Pediatric Age. Ii. Syncopal Attacks Due to Paroxysmal Ventricular Fibrillation. (Presentation of 1st Case in Italian Pediatric Literature)]. Clin Pediatr (Bologna). 1963 Sep;45:656-83. PubMed PMID: 14158288. Aritmie cardiache rare dell'eta' pediatrica. ii. accessi sincopali per fibrillazione ventricolare parossistica. (presentazione del primo caso della letteratura pediatrica italiana).

11. Ward OC. A New Familial Cardiac Syndrome in Children. J Ir

Med Assoc. 1964 Apr;54:103-6. PubMed PMID: 14136838.

55

12. Keating M, Atkinson D, Dunn C, Timothy K, Vincent GM, Leppert M. Linkage of a cardiac arrhythmia, the long QT syndrome, and the Harvey ras-1 gene. Science. 1991 May 3;252(5006):704-6. PubMed PMID: 1673802.

13. Curran ME, Splawski I, Timothy KW, Vincent GM, Green ED,

Keating MT. A molecular basis for cardiac arrhythmia: HERG mutations cause long QT syndrome. Cell. 1995 Mar 10;80(5):795-803. PubMed PMID: 7889573.

14. Wang Q, Shen J, Splawski I, Atkinson D, Li Z, Robinson J, et al. SCN5A mutations associated with an inherited cardiac arrhythmia, long QT syndrome. Cell. 1995 Mar 10;80(5):805-11. PubMed PMID: 7889574.

15. Einarsdottir E, Egerbladh I, Beckman L, Holmberg D, Escher SA. The genetic population structure of northern Sweden and its implications for mapping genetic diseases. Hereditas. 2007 Nov;144(5):171-80. PubMed PMID: 18031350.

16. Norrlands Universitetsförlag. Norrländsk Uppslagsbok vol 2-

4, p.128-349.

17. Splawski I, Shen J, Timothy KW, Lehmann MH, Priori S, Robinson JL, et al. Spectrum of mutations in long-QT syndrome genes. KVLQT1, HERG, SCN5A, KCNE1, and KCNE2. Circulation. 2000 Sep 5;102(10):1178-85. PubMed PMID: 10973849.

18. Kapplinger JD, Tester DJ, Salisbury BA, Carr JL, Harris-Kerr C, Pollevick GD, et al. Spectrum and prevalence of mutations from the first 2,500 consecutive unrelated patients referred for the FAMILION long QT syndrome genetic test. Heart Rhythm. 2009 Sep;6(9):1297-303. PubMed PMID: 19716085. Pubmed Central PMCID: 3049907.

19. Schwartz PJ, Crotti L, Insolia R. Long-QT syndrome: from

genetics to management. Circulation Arrhythmia and electrophysiology. 2012 Aug 1;5(4):868-77. PubMed PMID: 22895603. Pubmed Central PMCID: 3461497.

20. Escher Stefan A, Saura Anssi. Genetik: Liber AB; 2009 p.153-164.

21. Goldenberg I, Zareba W, Moss AJ. Long QT Syndrome. Current problems in cardiology. 2008 Nov;33(11):629-94. PubMed PMID: 18835466.

56

22. Winbo A, Stattin EL, Diamant UB, Persson J, Jensen SM, Rydberg A. Prevalence, mutation spectrum, and cardiac phenotype of the Jervell and Lange-Nielsen syndrome in Sweden. Europace : European pacing, arrhythmias, and cardiac electrophysiology : journal of the working groups on cardiac pacing, arrhythmias, and cardiac cellular electrophysiology of the European Society of Cardiology. 2012 Dec;14(12):1799-806. PubMed PMID: 22539601.

23. Viskin S. The QT interval: too long, too short or just right.

Heart Rhythm. 2009 May;6(5):711-5. PubMed PMID: 19389656.

24. Chiang CE, Roden DM. The long QT syndromes: genetic basis and clinical implications. J Am Coll Cardiol. 2000 Jul;36(1):1-12. PubMed PMID: 10898405.

25. Goldenberg I, Moss AJ. Long QT syndrome. J Am Coll Cardiol. 2008 Jun 17;51(24):2291-300. PubMed PMID: 18549912.

26. Schwartz PJ, Stramba-Badiale M, Crotti L, Pedrazzini M,

Besana A, Bosi G, et al. Prevalence of the congenital long-QT syndrome. Circulation. 2009 Nov 3;120(18):1761-7. PubMed PMID: 19841298. Pubmed Central PMCID: 2784143.

27. Schwartz PJ, Crotti L. QTc behavior during exercise and genetic testing for the long-QT syndrome. Circulation. 2011 Nov 15;124(20):2181-4. PubMed PMID: 22083145.

28. Crotti L, Spazzolini C, Schwartz PJ, Shimizu W, Denjoy I, Schulze-Bahr E, et al. The common long-QT syndrome mutation KCNQ1/A341V causes unusually severe clinical manifestations in patients with different ethnic backgrounds: toward a mutation-specific risk stratification. Circulation. 2007 Nov 20;116(21):2366-75. PubMed PMID: 17984373.

29. Marjamaa A, Salomaa V, Newton-Cheh C, Porthan K,

Reunanen A, Karanko H, et al. High prevalence of four long QT syndrome founder mutations in the Finnish population. Ann Med. 2009;41(3):234-40. PubMed PMID: 19160088. Pubmed Central PMCID: 2704397.

30. Schwartz PJ, Priori SG, Spazzolini C, Moss AJ, Vincent GM, Napolitano C, et al. Genotype-phenotype correlation in the long-QT syndrome: gene-specific triggers for life-threatening arrhythmias. Circulation. 2001 Jan 2;103(1):89-95. PubMed PMID: 11136691.

31. Viskin S. Long QT syndromes and torsade de pointes. Lancet. 1999 Nov 6;354(9190):1625-33. PubMed PMID: 10560690.

57

32. Surawicz B. Electrophysiologic substrate of torsade de pointes: dispersion of repolarization or early afterdepolarizations? J Am Coll Cardiol. 1989 Jul;14(1):172-84. PubMed PMID: 2661626.

33. Jalife JD, Anumonowo, J, Berenfeld, O, Kalifa, J. Basic

Cardiac Electrophysiology for the Clinician; Inheritable arrythmogenic diseases: Blackwell Publishing; 2009. 276-315 p.

34. Antzelevitch C, Sicouri S. Clinical relevance of cardiac arrhythmias generated by afterdepolarizations. Role of M cells in the generation of U waves, triggered activity and torsade de pointes. J Am Coll Cardiol. 1994 Jan;23(1):259-77. PubMed PMID: 8277090.

35. Birati EY, Belhassen B, Bardai A, Wilde AA, Viskin S. The site of origin of torsade de pointes. Heart. 2011 Oct;97(20):1650-4. PubMed PMID: 21561894.

36. Trippel DL, Parsons MK, Gillette PC. Infants with long-QT

syndrome and 2:1 atrioventricular block. Am Heart J. 1995 Nov;130(5):1130-4. PubMed PMID: 7484750.

37. Schwartz PJ, Malliani A. Electrical alternation of the T-wave: clinical and experimental evidence of its relationship with the sympathetic nervous system and with the long Q-T syndrome. Am Heart J. 1975 Jan;89(1):45-50. PubMed PMID: 1109551.

38. Malfatto G, Zaza A, Vanoli E, Schwartz PJ. Muscarinic effects on action potential duration and its rate dependence in canine Purkinje fibers. Pacing Clin Electrophysiol. 1996 Nov;19(11 Pt 2):2023-6. PubMed PMID: 8945090.

39. Zhang L, Timothy KW, Vincent GM, Lehmann MH, Fox J,

Giuli LC, et al. Spectrum of ST-T-wave patterns and repolarization parameters in congenital long-QT syndrome: ECG findings identify genotypes. Circulation. 2000 Dec 5;102(23):2849-55. PubMed PMID: 11104743.

40. Locati EH, Zareba W, Moss AJ, Schwartz PJ, Vincent GM, Lehmann MH, et al. Age- and sex-related differences in clinical manifestations in patients with congenital long-QT syndrome: findings from the International LQTS Registry. Circulation. 1998 Jun 9;97(22):2237-44. PubMed PMID: 9631873.

41. Priori SG, Schwartz PJ, Napolitano C, Bloise R, Ronchetti E, Grillo M, et al. Risk stratification in the long-QT syndrome. N Engl J Med. 2003 May 8;348(19):1866-74. PubMed PMID: 12736279.

58

42. Goldenberg I, Mathew J, Moss AJ, McNitt S, Peterson DR, Zareba W, et al. Corrected QT variability in serial electrocardiograms in long QT syndrome: the importance of the maximum corrected QT for risk stratification. J Am Coll Cardiol. 2006 Sep 5;48(5):1047-52. PubMed PMID: 16949500.

43. Zipes DP, Camm AJ, Borggrefe M, Buxton AE, Chaitman B,

Fromer M, et al. ACC/AHA/ESC 2006 Guidelines for Management of Patients With Ventricular Arrhythmias and the Prevention of Sudden Cardiac Death: a report of the American College of Cardiology/American Heart Association Task Force and the European Society of Cardiology Committee for Practice Guidelines (writing committee to develop Guidelines for Management of Patients With Ventricular Arrhythmias and the Prevention of Sudden Cardiac Death): developed in collaboration with the European Heart Rhythm Association and the Heart Rhythm Society. Circulation. 2006 Sep 5;114(10):e385-484. PubMed PMID: 16935995.

44. Priori SG, Napolitano C, Schwartz PJ, Grillo M, Bloise R, Ronchetti E, et al. Association of long QT syndrome loci and cardiac events among patients treated with beta-blockers. JAMA : the journal of the American Medical Association. 2004 Sep 15;292(11):1341-4. PubMed PMID: 15367556.

45. Chatrath R, Bell CM, Ackerman MJ. Beta-blocker therapy failures in symptomatic probands with genotyped long-QT syndrome. Pediatr Cardiol. 2004 Sep-Oct;25(5):459-65. PubMed PMID: 15534720.

46. Mönnig G, Kobe J, Loher A, Eckardt L, Wedekind H, Scheld

HH, et al. Implantable cardioverter-defibrillator therapy in patients with congenital long-QT syndrome: a long-term follow-up. Heart Rhythm. 2005 May;2(5):497-504. PubMed PMID: 15840474.

47. Schwartz PJ, Priori SG, Cerrone M, Spazzolini C, Odero A, Napolitano C, et al. Left cardiac sympathetic denervation in the management of high-risk patients affected by the long-QT syndrome. Circulation. 2004 Apr 20;109(15):1826-33. PubMed PMID: 15051644.

48. Bednar MM, Harrigan EP, Anziano RJ, Camm AJ, Ruskin JN. The QT interval. Prog Cardiovasc Dis. 2001 Mar-Apr;43(5 Suppl 1):1-45. PubMed PMID: 11269621.

49. Rosen MR, Anyukhovsky EP, Sosunov EA, Gainullin RZ. The

M cell. J Cardiovasc Electrophysiol. 1999 Sep;10(9):1297-9. PubMed PMID: 10517665.

59

50. Rosen MR, Cohen IS. Molecular/genetic determinants of repolarization and their modification by environmental stress. J Intern Med. 2006 Jan;259(1):7-23. PubMed PMID: 16336510.

51. Josephson ME. Clin Cardiac Electrophysiology; Techniques

and Interpretation. ed2, editor: Lea&Febiger; 1993. p 1-3.

52. Rautaharju PM, Surawicz B, Gettes LS, Bailey JJ, Childers R, Deal BJ, et al. AHA/ACCF/HRS recommendations for the standardization and interpretation of the electrocardiogram: part IV: the ST segment, T and U waves, and the QT interval: a scientific statement from the American Heart Association Electrocardiography and Arrhythmias Committee, Council on Clinical Cardiology; the American College of Cardiology Foundation; and the Heart Rhythm Society: endorsed by the International Society for Computerized Electrocardiology. Circulation. 2009 Mar 17;119(10):e241-50. PubMed PMID: 19228821.

53. Wilson F. The determination and signification of the areas of the ventricular deflection of the electrogram. American Heart J. 1934;10:46-61.

54. Wilson F. The T deflection of the electrocardiogram Trans

Assoc Am Physicians. 1931;46:29-38.

55. Antzelevitch C, Oliva A. Amplification of spatial dispersion of repolarization underlies sudden cardiac death associated with catecholaminergic polymorphic VT, long QT, short QT and Brugada syndromes. J Intern Med. 2006 Jan;259(1):48-58. PubMed PMID: 16336513.

56. Yan GX, Antzelevitch C. Cellular basis for the normal T wave and the electrocardiographic manifestations of the long-QT syndrome. Circulation. 1998 Nov 3;98(18):1928-36. PubMed PMID: 9799215.

57. Belardinelli L, Antzelevitch C, Vos MA. Assessing predictors

of drug-induced torsade de pointes. Trends in pharmacological sciences. 2003 Dec;24(12):619-25. PubMed PMID: 14654302.

58. Antzelevitch C, Shimizu W. Cellular mechanisms underlying the long QT syndrome. Current opinion in cardiology. 2002 Jan;17(1):43-51. PubMed PMID: 11790933.

59. Shimizu W, Antzelevitch C. Cellular basis for the ECG features of the LQT1 form of the long-QT syndrome: effects of beta-adrenergic agonists and antagonists and sodium channel blockers on transmural

60

dispersion of repolarization and torsade de pointes. Circulation. 1998 Nov 24;98(21):2314-22. PubMed PMID: 9826320.

60. Shimizu W, Antzelevitch C. Sodium channel block with

mexiletine is effective in reducing dispersion of repolarization and preventing torsade des pointes in LQT2 and LQT3 models of the long-QT syndrome. Circulation. 1997 Sep 16;96(6):2038-47. PubMed PMID: 9323097.

61. Shimizu W, Antzelevitch C. Differential effects of beta-adrenergic agonists and antagonists in LQT1, LQT2 and LQT3 models of the long QT syndrome. J Am Coll Cardiol. 2000 Mar 1;35(3):778-86. PubMed PMID: 10716483.

62. Antzelevitch C, Shimizu W, Yan GX, Sicouri S, Weissenburger J, Nesterenko VV, et al. The M cell: its contribution to the ECG and to normal and abnormal electrical function of the heart. J Cardiovasc Electrophysiol. 1999 Aug;10(8):1124-52. PubMed PMID: 10466495.

63. Lubinski A, Lewicka-Nowak E, Kempa M, Baczynska AM,

Romanowska I, Swiatecka G. New insight into repolarization abnormalities in patients with congenital long QT syndrome: the increased transmural dispersion of repolarization. Pacing Clin Electrophysiol. 1998 Jan;21(1 Pt 2):172-5. PubMed PMID: 9474667.

64. Tanabe Y, Inagaki M, Kurita T, Nagaya N, Taguchi A, Suyama K, et al. Sympathetic stimulation produces a greater increase in both transmural and spatial dispersion of repolarization in LQT1 than LQT2 forms of congenital long QT syndrome. J Am Coll Cardiol. 2001 Mar 1;37(3):911-9. PubMed PMID: 11693770.

65. Shimizu M, Ino H, Okeie K, Yamaguchi M, Nagata M, Hayashi K, et al. T-peak to T-end interval may be a better predictor of high-risk patients with hypertrophic cardiomyopathy associated with a cardiac troponin I mutation than QT dispersion. Clinical cardiology. 2002 Jul;25(7):335-9. PubMed PMID: 12109867.

66. Takenaka K, Ai T, Shimizu W, Kobori A, Ninomiya T, Otani H,

et al. Exercise stress test amplifies genotype-phenotype correlation in the LQT1 and LQT2 forms of the long-QT syndrome. Circulation. 2003 Feb 18;107(6):838-44. PubMed PMID: 12591753.

67. Goldenberg I, Moss AJ, Zareba W. QT interval: how to measure it and what is "normal". J Cardiovasc Electrophysiol. 2006 Mar;17(3):333-6. PubMed PMID: 16643414.

61

68. Ireland RH, Robinson RTCE , Heller SR, ND. H, editors. QT measurement for TU fused ECG morphology as exhibited during hypoglcaemia. 1998: Proceeding of the Twentieth Annual International Conference of the IEEE Engineering in Medicine and Biology Society.

69. Panicker GK, Karnad DR, Natekar M, Kothari S, Narula D,

Lokhandwala Y. Intra- and interreader variability in QT interval measurement by tangent and threshold methods in a central electrocardiogram laboratory. J Electrocardiol. 2009 Jul-Aug;42(4):348-52. PubMed PMID: 19261293.

70. Panicker GK, Karnad DR, Joshi R, Shetty S, Vyas N, Kothari S, et al. Z-score for benchmarking reader competence in a central ECG laboratory. Ann Noninvasive Electrocardiol. 2009 Jan;14(1):19-25. PubMed PMID: 19149789.

71. Gallagher MM, Magliano G, Yap YG, Padula M, Morgia V, Postorino C, et al. Distribution and prognostic significance of QT intervals in the lowest half centile in 12,012 apparently healthy persons. The American journal of cardiology. 2006 Oct 1;98(7):933-5. PubMed PMID: 16996877.

72. Kobza R, Roos M, Niggli B, Abacherli R, Lupi GA, Frey F, et

al. Prevalence of long and short QT in a young population of 41,767 predominantly male Swiss conscripts. Heart Rhythm. 2009 May;6(5):652-7. PubMed PMID: 19303371.

73. Schwartz PJ, Ackerman MJ. The long QT syndrome: a transatlantic clinical approach to diagnosis and therapy. Eur Heart J. 2013 Mar 18. PubMed PMID: 23509228.

74. Moss AJ. Measurement of the QT interval and the risk associated with QTc interval prolongation: a review. The American journal of cardiology. 1993 Aug 26;72(6):23B-5B. PubMed PMID: 8256751.

75. Rautaharju PM, Zhou SH, Wong S, Calhoun HP, Berenson

GS, Prineas R, et al. Sex differences in the evolution of the electrocardiographic QT interval with age. Can J Cardiol. 1992 Sep;8(7):690-5. PubMed PMID: 1422988.

76. Vincent GM, Timothy KW, Leppert M, Keating M. The spectrum of symptoms and QT intervals in carriers of the gene for the long-QT syndrome. N Engl J Med. 1992 Sep 17;327(12):846-52. PubMed PMID: 1508244.

77. Bexton RS, Vallin HO, Camm AJ. Diurnal variation of the QT interval--influence of the autonomic nervous system. Br Heart J. 1986

62

Mar;55(3):253-8. PubMed PMID: 3513807. Pubmed Central PMCID: 1232162.

78. Magnano AR, Holleran S, Ramakrishnan R, Reiffel JA,

Bloomfield DM. Autonomic nervous system influences on QT interval in normal subjects. J Am Coll Cardiol. 2002 Jun 5;39(11):1820-6. PubMed PMID: 12039498.

79. Bazett H. An analysis of the time relations of electrocardiogram. Heart. 1920;7:355-70.

80. Fridericia L. Die sytolendauer in elektrokardiogramm bei normalen menschen und bei herzkranke. Acta Med Scand. 1920;53:469-86.

81. Batchvarov VN, Ghuran A, Smetana P, Hnatkova K, Harries

M, Dilaveris P, et al. QT-RR relationship in healthy subjects exhibits substantial intersubject variability and high intrasubject stability. American journal of physiology Heart and circulatory physiology. 2002 Jun;282(6):H2356-63. PubMed PMID: 12003846.

82. Franz MR, Swerdlow CD, Liem LB, Schaefer J. Cycle length dependence of human action potential duration in vivo. Effects of single extrastimuli, sudden sustained rate acceleration and deceleration, and different steady-state frequencies. J Clin Invest. 1988 Sep;82(3):972-9. PubMed PMID: 3417875. Pubmed Central PMCID: 303610.

83. Lau CP, Freedman AR, Fleming S, Malik M, Camm AJ, Ward DE. Hysteresis of the ventricular paced QT interval in response to abrupt changes in pacing rate. Cardiovascular research. 1988 Jan;22(1):67-72. PubMed PMID: 3167930.

84. Malik M, Hnatkova K, Novotny T, Schmidt G. Subject-specific

profiles of QT/RR hysteresis. American journal of physiology Heart and circulatory physiology. 2008 Dec;295(6):H2356-63. PubMed PMID: 18849333.

85. Pueyo E, Smetana P, Caminal P, de Luna AB, Malik M, Laguna P. Characterization of QT interval adaptation to RR interval changes and its use as a risk-stratifier of arrhythmic mortality in amiodarone-treated survivors of acute myocardial infarction. IEEE Trans Biomed Eng. 2004 Sep;51(9):1511-20. PubMed PMID: 15376499.

86. Talbot S. QT interval in right and left bundle-branch block. Br Heart J. 1973 Mar;35(3):288-91. PubMed PMID: 4692659. Pubmed Central PMCID: 458603.

63

87. Crow RS, Hannan PJ, Folsom AR. Prognostic significance of corrected QT and corrected JT interval for incident coronary heart disease in a general population sample stratified by presence or absence of wide QRS complex: the ARIC Study with 13 years of follow-up. Circulation. 2003 Oct 21;108(16):1985-9. PubMed PMID: 14517173.

88. Rautaharju PM, Zhang ZM, Prineas R, Heiss G. Assessment of

prolonged QT and JT intervals in ventricular conduction defects. The American journal of cardiology. 2004 Apr 15;93(8):1017-21. PubMed PMID: 15081446.

89. Chiladakis J, Kalogeropoulos A, Koutsogiannis N, Zagkli F, Vlassopoulou N, Chouchoulis K, et al. Optimal QT/JT interval assessment in patients with complete bundle branch block. Ann Noninvasive Electrocardiol. 2012 Jul;17(3):268-76. PubMed PMID: 22816546.

90. Lesinskiene S, Barkus A, Ranceva N, Dembinskas A. A meta-analysis of heart rate and QT interval alteration in anorexia nervosa. The world journal of biological psychiatry : the official journal of the World Federation of Societies of Biological Psychiatry. 2008;9(2):86-91. PubMed PMID: 17853296.

91. Isner JM, Roberts WC, Heymsfield SB, Yager J. Anorexia

nervosa and sudden death. Annals of internal medicine. 1985 Jan;102(1):49-52. PubMed PMID: 3966745.

92. Jensen SM, Johansson G, Osterman G, Reiz S, Naslund U. On-line computerized vectorcardiography monitoring of myocardial ischemia during coronary angioplasty: comparison with 12-lead electrocardiography. Coron Artery Dis. 1994 Jun;5(6):507-14. PubMed PMID: 7952410.

93. Nowinski K, Jensen S, Lundahl G, Bergfeldt L. Changes in ventricular repolarization during percutaneous transluminal coronary angioplasty in humans assessed by QT interval, QT dispersion and T vector loop morphology. J Intern Med. 2000 Aug;248(2):126-36. PubMed PMID: 10947891.

94. Wecke L, Poci D, Schwieler J, Johansson B, Edvardsson N,

Lundahl G, et al. Vectorcardiography shows cardiac memory and repolarization heterogeneity after ablation of accessory pathways not apparent on ECG. Int J Cardiol. 2011 Nov 9:doi 10.1016/j.ijcard.2011.10.16. PubMed PMID: 22078977.

64

95. Zareba W, Badilini F, Moss AJ. Automatic detection of spatial and dynamic heterogeneity of repolarization. J Electrocardiol. 1994;27 Suppl:66-72. PubMed PMID: 7884378.

96. Diamant UB, Winbo A, Stattin EL, Rydberg A, Kesek M,

Jensen SM. Two automatic QT algorithms compared with manual measurement in identification of long QT syndrome. J Electrocardiol. 2010 Jan-Feb;43(1):25-30. PubMed PMID: 20005993.

97. Frank E. Analysis of R, L, F, B systems of spatial vectorcardiography. Am Heart J. 1956 Jan;51(1):34-52. PubMed PMID: 13268309.

98. Frank E. Planar dipole loops. Circ Res. 1956 May;4(3):257-62. PubMed PMID: 13317015.

99. Wecke L, Gadler F, Linde C, Lundahl G, Rosen MR, Bergfeldt

L. Temporal characteristics of cardiac memory in humans: vectorcardiographic quantification in a model of cardiac pacing. Heart Rhythm. 2005 Jan;2(1):28-34. PubMed PMID: 15851261.

100. Fayn J, Robel P, Arnaud P. A new methodology for optimal comparision of serial vectorcardiogram Computer in cardiology. 1983:467-70.

101. Wecke L, Rubulis A, Lundahl G, Rosen MR, Bergfeldt L. Right ventricular pacing-induced electrophysiological remodeling in the human heart and its relationship to cardiac memory. Heart Rhythm. 2007 Dec;4(12):1477-86. PubMed PMID: 17997360.

102. Vahedi F, Odenstedt J, Hartford M, Gilljam T, Bergfeldt L.

Vectorcardiography analysis of the repolarization response to pharmacologically induced autonomic nervous system modulation in healthy subjects. J Appl Physiol. 2012 Aug;113(3):368-76. PubMed PMID: 22582212.

103. Langner PH, Jr., Okada RH, Moore SR, Fies HL. Comparison of four orthogonal systems of vectorcardiography. Circulation. 1958 Jan;17(1):46-54. PubMed PMID: 13511621.

104. Berson AS, Haisty WK, Pipberger HV. Electrode position effects on Frank lead electrocardiograms. Am Heart J. 1978 Apr;95(4):463-73. PubMed PMID: 636984.

105. Healthcare GE. Marquette™ 12SL™ ECG Analysis Program

Physician’s Guide. 1 February 2007;416791-004 Revision D.

65

106. Lepeschkin E, Surawicz B. The measurement of the Q-T interval of the electrocardiogram. Circulation. 1952 Sep;6(3):378-88. PubMed PMID: 14954534.

107. Winbo A, Diamant UB, Rydberg A, Persson J, Jensen SM,

Stattin EL. Origin of the Swedish long QT syndrome Y111C/KCNQ1 founder mutation. Heart Rhythm. 2011 Apr;8(4):541-7. PubMed PMID: 21129503.

108. Viskin S, Rosovski U, Sands AJ, Chen E, Kistler PM, Kalman JM, et al. Inaccurate electrocardiographic interpretation of long QT: the majority of physicians cannot recognize a long QT when they see one. Heart Rhythm. 2005 Jun;2(6):569-74. PubMed PMID: 15922261.

109. Postema PG, De Jong JS, Van der Bilt IA, Wilde AA. Accurate electrocardiographic assessment of the QT interval: teach the tangent. Heart Rhythm. 2008 Jul;5(7):1015-8. PubMed PMID: 18598957.

110. Pater C. Methodological considerations in the design of trials

for safety assessment of new drugs and chemical entities. Current controlled trials in cardiovascular medicine. 2005 Feb 3;6(1):1. PubMed PMID: 15691384. Pubmed Central PMCID: 549209.

111. Rubulis A, Jensen J, Lundahl G, Tapanainen J, Bergfeldt L. Ischemia induces aggravation of baseline repolarization abnormalities in left ventricular hypertrophy: a deleterious interaction. J Appl Physiol. 2006 Jul;101(1):102-10. PubMed PMID: 16565349.

112. Sahlen A, Rubulis A, Winter R, Jacobsen PH, Stahlberg M, Tornvall P, et al. Cardiac fatigue in long-distance runners is associated with ventricular repolarization abnormalities. Heart Rhythm. 2009 Apr;6(4):512-9. PubMed PMID: 19324313.

113. Vahedi F, Haney MF, Jensen SM, Naslund 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 Jul;16(3):287-94. PubMed PMID: 21762257.

114. Vahedi F, Diamant UB, Lundahl G, Bergqvist G, Gransberg L, Jensen SM, et al. Instability of repolarization in LQTS mutation carriers compared to healthy control subjects assessed by vectorcardiography. Heart Rhythm. 2013 May 2. PubMed PMID: 23643511.

115. Garg A, Lehmann MH. Prolonged QT interval diagnosis suppression by a widely used computerized ECG analysis system. Circulation

66

Arrhythmia and electrophysiology. 2013 Feb;6(1):76-83. PubMed PMID: 23275261.

116. Darpö B, Agin M, Kazierad DJ, Layton G, Muirhead G, Gray

P, et al. Man versus machine: is there an optimal method for QT measurements in thorough QT studies? Journal of clinical pharmacology. 2006 Jun;46(6):598-612. PubMed PMID: 16707406.

117. Diamant UB, Jensen SM, Winbo A, Stattin EL, Rydberg A. Vectorcardiographic recordings of the QT interval in a pediatric long QT syndrome population. Pediatr Cardiol. 2012 Feb;34(2):245-9. PubMed PMID: 22806714.

118. Dahimene S, Alcolea S, Naud P, Jourdon P, Escande D, Brasseur R, et al. The N-terminal juxtamembranous domain of KCNQ1 is critical for channel surface expression: implications in the Romano-Ward LQT1 syndrome. Circ Res. 2006 Nov 10;99(10):1076-83. PubMed PMID: 17053194.

119. Moss AJ, Shimizu W, Wilde AA, Towbin JA, Zareba W,

Robinson JL, et al. Clinical aspects of type-1 long-QT syndrome by location, coding type, and biophysical function of mutations involving the KCNQ1 gene. Circulation. 2007 May 15;115(19):2481-9. PubMed PMID: 17470695.

120. Winbo A, Diamant UB, Stattin EL, Jensen SM, Rydberg A. Low incidence of sudden cardiac death in a Swedish Y111C type 1 long-QT syndrome population. Circ Cardiovasc Genet. 2009 Dec;2(6):558-64. PubMed PMID: 20031635.

121. Moss AJ, Schwartz PJ, Crampton RS, Tzivoni D, Locati EH, MacCluer J, et al. The long QT syndrome. Prospective longitudinal study of 328 families. Circulation. 1991 Sep;84(3):1136-44. PubMed PMID: 1884444.

122. Hinterseer M, Thomsen MB, Beckmann BM, Pfeufer A,

Schimpf R, Wichmann HE, et al. Beat-to-beat variability of QT intervals is increased in patients with drug-induced long-QT syndrome: a case control pilot study. Eur Heart J. 2008 Jan;29(2):185-90. PubMed PMID: 18156612.

123. Jacobson I, Carlsson L, Duker G. Beat-by-beat QT interval variability, but not QT prolongation per se, predicts drug-induced torsades de pointes in the anaesthetised methoxamine-sensitized rabbit. Journal of pharmacological and toxicological methods. 2011 Jan-Feb;63(1):40-6. PubMed PMID: 20451633.

124. Hinterseer M, Beckmann BM, Thomsen MB, Pfeufer A, Dalla Pozza R, Loeff M, et al. Relation of increased short-term variability of QT

67

interval to congenital long-QT syndrome. The American journal of cardiology. 2009 May 1;103(9):1244-8. PubMed PMID: 19406266.

125. Goldenberg I, Barsheshet A. Instability of Ventricular

Repolarization in Long QT Syndrome: Is the Corrected QT Interval Sufficient for Risk Assessment? Heart Rhythm. 2013 May 31. PubMed PMID: 23732227.

126. Brink PA, Crotti L, Corfield V, Goosen A, Durrheim G, Hedley P, et al. Phenotypic variability and unusual clinical severity of congenital long-QT syndrome in a founder population. Circulation. 2005 Oct 25;112(17):2602-10. PubMed PMID: 16246960.

127. Jespersen T, Rasmussen HB, Grunnet M, Jensen HS, Angelo K, Dupuis DS, et al. Basolateral localisation of KCNQ1 potassium channels in MDCK cells: molecular identification of an N-terminal targeting motif. Journal of cell science. 2004 Sep 1;117(Pt 19):4517-26. PubMed PMID: 15316073.

1

Appendices

I Instructions in manual measurement of the QT interval, translated.

II Long QT syndrome Questionnaire, translated.

2

I - Instructions in manual measurement of the QT interval (translated)

Manual measurement of the QT interval

Measure 3 QT interval in lead II, V5 and V6, the interval should be specified

in milliseconds (ms). Use a ruler or sender, whatever you normally use.

Measure 3 RR-intervals in lead II, specify the interval in ms.

Mark clearly on the ECG the measuring points and write down the value

under the ECG complex.

Each ECG has a code - 1, 2, 3 and so on - in the attached table. Enter your

measurements from ECG coded 1 into code 1 in the table and so on. Leave

the box or boxes empty if you have no value to enter.

At the end of each row in the table there are 3 options for the assessment of

measured QTc. The 3 options are normal, borderline and prolonged QTc

according to the table below. On each recorded ECG you can read age and

sex on the individual. Cross mark the corresponding box for your

assessment.

Children 1-15

year (ms)

Adult male

(ms)

Adult female

(ms)

Normal <440 <430 <450

Borderline 440-460 430-450 450-470

Prolonged >460 >450 >470

It is well known that the T-wave end is difficult to assess because of low

amplitudes, biphasic T-waves, discerning which is-U wave and which is T-

wave. These are among the issues often mentioned that render the

assessment problematic. You are one of four observers in a group where the

measurements of the QT intervals are compared. Two observers are

classified as habituated to measure the QT interval and two as less habit.

QT interval values, recorded and automatically measured by two different

ECG equipments will be compared with each observer’s QT interval values. If

you have any questions during the course of your work please contact any of

the following. Thank you for participating in this work!

Ulla-Britt Diamant phone: 52729 or Steen Jensen phone: 51760.

3

II - Long QT syndrome questionnaire (translated)

Patient questionnaire

Patient information: name birth year, index relative with the disease

1. Have you had symptoms of the disease Long QT Syndrome that

occurs in your family? (for example dizziness, loss of consciousness,

convulsions) Yes/No

If Yes, describe your symptoms.

If No, go to question 5.

2. In which situations have you experienced the symptoms described?

3. How old were you at symptoms debut?

4. When did you experience you most recent symptom?

5. Have you had contact with the health care system in relation to

syncope, loss of consciousness or convulsions? Yes/No

If Yes, when, and at which hospital?

6. Do you have normal hearing? Yes/No

If No, have you consulted an Auditory Clinic? At which hospital?

7. Do you have any additional diseases for which you attend controls

and/or are receiving treatment for? Yes/No

If Yes, what diseases and/treatments?

8. Have you been diagnosed with Long QT Syndrome?

If Yes, when?

9. Have you been prescribed/recommended medical therapy? Yes/No

If Yes, what medicine and in which dosage?

10. Are you currently taking the medication? Yes/No

4

If Yes, since when are you taking the medication? Are you taking the

prescribed dosage?

If No, why are you not taking the medication?

11. When taking the medication, do you experience any adverse effects?

Yes/No

If Yes, describe the adverse effects.

12. Is there anything you like to add with regards to the disease,

symptoms, controls, medication or other aspects?