Effect of n-3 fatty acids from fish on electrocardiographic characteristics
in patients with frequent premature ventricular complexes
Anouk Geelen1*, Peter L. Zock1, Ingeborg A. Brouwer1, Martijn B. Katan1, Jan A. Kors2,
Henk J. Ritsema van Eck2and Evert G. Schouten1
1Wageningen Centre for Food Sciences (WCFS) and Division of Human Nutrition, Wageningen University, Wageningen, The Netherlands
2Department of Medical Informatics, Faculty of Medicine and Health Sciences, Erasmus University Rotterdam, Rotterdam,
(Received 1 September 2004 – Revised 23 December 2004 – Accepted 15 January 2005)
n-3 Fatty acids may protect against heart disease mortality by preventing fatal arrhythmias. Underlying effects on cardiac electrophysiology may be demon-
strable in the standard electrocardiogram (ECG) and provide insight into the mechanism. Therefore, we investigated the effect of dietary n-3 fatty acids on
heart-rate-corrected QT interval, T-loop width, spatial QRS-T angle and spatial U-wave amplitude in patients with frequent premature ventricular com-
plexes. Seventy-four patients received either capsules providing 1·5g n-3 fatty acids daily or placebo for approximately 14 weeks. ECG were recorded
before and after intervention. None of the ECG characteristics was significantly affected by treatment. The present results do not provide additional support
for the hypothesis that n-3 fatty acids prevent cardiac arrhythmia through generic electrophysiologic effects on heart cell membranes. However, we cannot
exclude effects of n-3 fatty acids on clinical relevant endpoints that are not easily detected by prior changes in the ECG.
n-3 Fatty acids: Electrocardiogram: Heart-rate-corrected QT interval: Electrophysiology
Evidence from human observational studies and clinical trials
indicates that n-3 fatty acids can protect against fatal heart disease
by preventing cardiac arrhythmias (Burr et al. 1989; Albert et al.
2002; Marchioli et al. 2002). In addition, experimental studies
show that n-3 fatty acids prevent and terminate arrhythmias in
in vitro and animal models. n-3 Fatty acids stabilize the electrical
activity of cardiomyocytes by elevating the action potential
threshold and prolonging the relative refractory time. These
effects may result from an action of n-3 fatty acids on ion trans-
port through heart cell membranes, which is essential for heart
rhythm (Kang & Leaf, 2000). Effects on the electrophysiology
of the whole heart may be demonstrable in a surface electrocar-
diogram (ECG) in man. We previously reported no effect of
n-3 fatty acids on several ECG characteristics in healthy subjects
(Geelen et al. 2002). However, effects of n-3 fatty acids may be
detected only in abnormal ECG of more susceptible subjects. We
therefore performed a study in patients with frequent premature
ventricular complexes, a common form of arrhythmia that may
trigger more life-threatening arrhythmias.
The heart-rate-corrected QT interval (QTc) on the ECG is a rel-
evant measure for arrhythmia risk. In the general population, sub-
jects with a longer QTc have an increased mortality risk
(Schouten et al. 1991; Dekker et al. 1994; de-Bruyne et al. 1999).
Thus, a decrease in QTc by n-3 fatty acids would support a protec-
tive effect of n-3 fatty acids on heart disease. An earlier study
showed that n-3 fatty acids decreased the duration of QTc in dogs
(Billman et al. 1997). n-3 Fatty acids may also affect other more
exploratory ECG characteristics. For instance, T-loop width (Kors
et al. 1999) and the spatial QRS-T angle (Kardys et al. 2003) have
been proposed as markers for heterogeneity of ventricular repolar-
ization, which provides the condition for the genesis of ventricular
arrhythmias.ThespatialQRS-Tangle hasbeenrecognized asarisk
predictor of cardiac mortality in the elderly (Kardys et al. 2003).
Furthermore, U-wave changes may predict the occurrence of
arrhythmias (Trusz-Gluza et al. 2002), although little is known
lier study on the effects of n-3 fatty acids on ECG characteristics
(Geelen et al. 2002), we found a tendency for n-3 fatty acids to
decrease the U-wave (S Quak, A Geelen, IA Brouwer, PL Zock,
HJRitsemavan Eck,JAKors,MB KatanandEG Schouten,unpub-
lished results). Effects of n-3 fatty acids on specific ECG character-
istics could provide insight into the mechanism of a possible
antiarrhythmic effect. It could also suggest new biomarkers for
the study of the antiarrhythmic potential of drugs and food ingredi-
ents in man. Therefore, we investigated the effect of intake of very
long-chain n-3 fatty acids from fish on QTc, T-loop width, spatial
QRS-T angle and spatial U-wave amplitude in subjects with fre-
quent premature ventricular complexes.
Subjects and methods
The Medical Ethical Committee of Wageningen University
approved the study protocol. Patients gave their written informed
*Corresponding author: Dr Anouk Geelen, fax þ31 317 483342, email firstname.lastname@example.org
Abbreviations: ECG, electrocardiogram; QTc, heart-rate-corrected QT interval.
British Journal of Nutrition (2005), 93, 787–790
q The Authors 2005
consent after the study protocol had been explained to them.
Cardiologists from three Dutch hospitals recruited and enrolled
patients aged 18 years or older with at least 1440 premature
ventricular complexes per 24h in a previous Holter recording
made less than 6 months before the study. We included patients
with and without a history of myocardial infarction, but patients
who used antiarrhythmic drugs other than beta-blockers were
excluded. Ninety-two patients who met the inclusion criteria were
The primary purpose of the present double-blind, placebo-
controlled study with parallel design was to investigate effects
of n-3 fatty acids on the occurrence of premature ventricular com-
plexes (results reported elsewhere). Patients were randomized per
centre, stratified for history of myocardial infarction, to receive
either a daily dose of 3·5g fish oil or placebo oil (high oleic
sunflower oil; Loders Croklaan, Wormerveer, The Netherlands)
during the intervention period of 14 (SD 1) weeks. The oils
were administered in seven soft gelatin capsules daily (Banner
Pharmacaps Europe BV, Tilburg, The Netherlands). Fish oil
capsules provided approximately 700mg EPA (C20:5n-3),
560mg DHA (C22:6n-3) and 260mg other n-3 fatty acids per
d. The placebo capsules contained mainly oleic acid (C18:1n-9).
Compliance of the patients was checked by analysis of n-3 fatty
acids in serum cholesteryl esters from non-fasting blood samples
taken at the beginning and end of the intervention period (Zock
et al. 1997). Intakes of energy, fatty acids, cholesterol and alcohol
were estimated by a telephone-administered 24h dietary recall. In
addition, fish intake was assessed twice by interviewing patients
using a questionnaire on the frequency of fish consumption.
Complete ECG measurements at both baseline and the end of
the intervention period were available for seventy-four patients.
Standard twelve-lead ECG were recorded for 1min with a
Cardio Perfect Portable recorder and digitally stored on a
Cardio Control workstation (Cardio Control NV, Delft, The
Netherlands). We processed ECG recordings without knowledge
of treatment type or other subject variables, using the Modular
ECG Analysis System (MEANS; van-Bemmel et al. 1990).
Bazett’s formula (QTc¼QT/pRR) was used to correct QT
duration for heart rate. T-loop width and QRS-T angle were
determined as described previously (Kors et al. 1999; Kardys
et al. 2003). The amplitude of the spatial U-wave gradient
vector was calculated by taking the integral of the x, y and z
component of the U-wave using the interactive computer program
‘Intraval’ (Ritsema van Eck, 2004).
Differences in response between the fish oil and placebo groups
were analysed by Student’s t test. A subgroup analysis for patients
with and without prior myocardial infarction was planned before-
hand and included in the protocol.
There were no substantial differences in baseline characteristics
between the fish oil and the placebo group (Table 1). EPA in
serum cholesteryl esters (g/100g total fatty acids) confirmed
compliance; it changed during intervention from 1·30 (SD 0·78)
to 3·66 (SD 1·68) in the fish oil group and remained constant in
the placebo group (1·03 (SD 0·70) to 0·98 (SD 0·62)). QTc was
not significantly affected by intake of n-3 fatty acids; it decreased
by 0·3ms (95% CI 10·7, 10·1ms) or 0·1% in the fish oil group
compared with placebo (Fig. 1, Table 2). Intake of n-3 fatty
acids also did not significantly affect T-loop width, spatial
QRS-T angle and spatial U-wave amplitude (Table 2). Subgroup
analyses in patients with and without prior myocardial infarction
also revealed no significant effects (Table 2).
Background dietary intake was similar in the two treatment
groups. The fish oil group consumed 33% energy as fat, 0·6%
energy (1·5g/d) as total n-3 fatty acids (mostly a-linolenic
acid), 4% energy as alcohol and 23mg cholesterol/MJ. In the pla-
cebo group the corresponding figures were 33%, 0·5% (1·1g),
4% and 28mg/MJ. Fish intake was similar in the treatment
groups and did not change during the intervention.
We observed no effect of a daily intake of 3·5g fish oil for
approximately 14 weeks v. placebo on QTc, QRS-T angle,
T-loop width or spatial U-wave amplitude in these patients with
frequent premature ventricular complexes. It is unlikely that this
is caused by a lack of power, because the 95% CI of the differ-
ence in response of QTc was narrow. A true effect of n-3 fatty
acids on QTc of more than 2·5% is not expected. In a previous
study, we also observed no effect of n-3 fatty acids on several
ECG characteristics in eighty-four healthy middle-aged men and
women (Geelen et al. 2002). The effects of n-3 fatty acids on
the electrophysiology of the human heart may be too small to
cause appreciable ECG changes or there may be multiple effects
at the cellular level that cancel each other out in the ECG. A stan-
dard ECG of short duration (1min) may not be useful to demon-
strate possible effects of n-3 fatty acids on the heart.
Protective effects of n-3 fatty acids on hard endpoints in
clinical trials have been found exclusively for post-myocardial
QTc response (ms)
Fig. 1. Response of the heart-rate-corrected QT interval (QTc) of men and
women who consumed 3·5g fish oil/d (1·5g n-3 fatty acids/d; n 38) or pla-
cebo (n 36) daily for approximately 14 weeks.
Table 1. Characteristics of the forty-four men and thirty women for whom
complete electrocardiographic measurements were available
(Values are means and standard deviations unless indicated otherwise)
Fish oil (n 38)Placebo (n 36)
Male sex, n (%)
Background n-3 fatty acid intake
from fish* (g/month)
History of myocardial infarction,
11 (28·9)8 (22·2)
*Sum of a-linoleic acid plus EPA plus DHA.
A. Geelen et al.788
infarction patients (Burr et al. 1989; GISSI-Prevenzione Investi-
gators, 1999) and not in patients with angina (Burr et al. 2003).
It could be hypothesized that the cardioprotective effect of n-3
fatty acids is restricted to patients with earlier myocardial infarc-
tion. n-3 Fatty acids may interact with structural abnormalities in
cardiac tissue due to previous infarctions and in that way prevent
fatal arrhythmias. If so, n-3 fatty acids would affect the ECG only
in post-infarction patients. We did not find effects of n-3 fatty
acids on the ECG in our subgroup of such patients; however,
this group was small (n 19).
The dose of 3·5g fish oil in the present study provided approxi-
mately 1·5g n-3 fatty acids daily. Several observational studies
suggest that the cardioprotective effect is already present at low
doses of about 200mg n-3 fatty acids/d. Most importantly, in the
GISSI-Prevenzione trial, 1g n-3 fatty acids/d was enough to lower
the rate of death, non-fatal myocardial infarction and stroke
that the dose of n-3 fatty acids in the present study was too low to
detect relevant effects on the ECG, if any exist.
A definitive answer as to whether n-3 fatty acids can reduce
risk of ventricular arrhythmia will have to come from long-term
trials on n-3 fatty acids and arrhythmia incidence in high-risk
patients with an implantable cardioverter defibrillator (Brouwer
et al. 2003). We are aware of three such trials; the first in 200
such patients recently reported a trend towards increased rather
than decreased recurrence of ventricular arrhythmias in patients
who received n-3 fatty acids over a follow-up of 2 years (Cleland
et al. 2004). Two other trials are running and will report within
Despite the body of evidence that n-3 fatty acids reduce risk of
fatal heart disease by preventing ventricular arrhythmia, we could
not demonstrate effects on cardiac electrophysiology as measured
in body surface ECG. The present results do not provide
additional support for the hypothesis that n-3 fatty acids prevent
cardiac arrhythmia through generic electrophysiologic effects on
heart cell membranes. However, we cannot exclude effects of
n-3 fatty acids on clinical relevant endpoints that are not easily
detected by prior changes in the ECG. Mechanistic studies at the
organ level and clinical trials on endpoints will have to provide
This work was funded by the Wageningen Centre for Food
Sciences, an alliance of major Dutch food industries, Maastricht
University, TNO Nutrition and Food Research, and Wageningen
University and Research Centre, with financial support by the
Albert CM, Campos H, Stampfer MJ, Ridker PM, Manson JE, Willett WC
& Ma J (2002) Blood levels of long-chain n-3 fatty acids and the risk of
sudden death. N Engl J Med 346, 1113–1118.
Billman GE, Kang JX & Leaf A (1997) Prevention of ischemia-induced
cardiac sudden death by n-3 polyunsaturated fatty acids in dogs.
Lipids 32, 1161–1168.
Brouwer IA, Zock PL, Wever EF, Hauer RN, Camm AJ, Bocker D,
Otto-Terlouw P, Katan MB & Schouten EG (2003) Rationale and
design of a randomised controlled clinical trial on supplemental
intake of n-3 fatty acids and incidence of cardiac arrhythmia: SOFA.
Eur J Clin Nutr 57, 1323–1330.
Burr ML, Ashfield WP, Dunstan FDJ, Fehily AM, Breay P, Ashton
T, Zotos PC, Haboubi NAA & Elwood PC (2003) Lack of benefit of
dietary advice to men with angina: results of a controlled trial. Eur
J Clin Nutr 57, 193–200.
Burr ML, Fehily AM, Gilbert JF, Rogers S, Holliday RM, Sweetnam PM,
Elwood PC & Deadman NM (1989) Effects of changes in fat, fish, and
fibre intakes on death and myocardial reinfarction: diet and reinfarction
trial (DART). Lancet 2, 757–761.
Cleland JG, Freemantle N, Kaye G, Nasir M, Velavan P, Lalukota K,
Mudawi T, Shelton R, Clark AL & Coletta AP (2004) Clinical trials
update from the American Heart Association meeting: Omega-3 fatty
acids and arrhythmia risk in patients with an implantable defibrillator,
ACTIV in CHF, VALIANT, the Hanover autologous bone marrow
Table 2. Electrocardiographic characteristics at the start and end of the intervention period and the difference in response between the fish oil
and placebo group, for all patients together and for the patients with and without prior myocardial infarction (MI) separately
(Values are means and standard deviations or mean and 95% CI)
Difference in response
Mean (95% CI)
Spatial QRS-T angle (8)
Width of T-loop (8)
Spatial U-wave amplitude (mV)
Patients with MI
Spatial QRS-T angle (8)
Width of T-loop (8)
Spatial U-wave amplitude (mV)
Patients without MI
Spatial QRS-T angle (8)
Width of T-loop (8)
Spatial U-wave amplitude (mV)
(n 38)(n 36)
20·3 (210·7, 10·1)
0·4 (28·0, 8·7)
3·0 (24·4, 10·5)
21·4 (215·8, 13·0)
(n 11)(n 8)
2·5 (226·3, 31·3)
24·1 (227·5, 19·2)
21·2 (215·2, 12·9)
4·4 (240·3, 49·1)
(n 27)(n 28)
21·5 (212·4, 9·4)
1·6 (27·0, 10·2)
4·7 (24·4, 13·7)
22·6 (216·1, 10·9)
QTc, heart-rate-corrected QT duration.
n-3 Fatty acids and electrocardiographic characteristics789
transplantation study, SPORTIF V, ORBIT and PAD and DEFINITE.
Eur J Heart Fail 6, 109–115.
de-Bruyne MC, Hoes AW, Kors JA, Hofman A, van-Bemmel JH &
Grobbee DE (1999) Prolonged QT interval predicts cardiac and
all-cause mortality in the elderly. The Rotterdam Study. Eur Heart
J 20, 278–284.
Dekker JM, Schouten EG, Klootwijk P, Pool J & Kromhout D (1994)
Association between QT interval and coronary heart disease in
middle-aged and elderly men. The Zutphen Study. Circulation 90,
Geelen A, Brouwer IA, Zock PL, Kors JA, Swenne CA, Katan MB &
Schouten EG (2002) (N-3) fatty acids do not affect electrocardiographic
characteristics of healthy men and women. J Nutr 132, 3051–3054.
GISSI-Prevenzione Investigators (1999) Dietary supplementation with n-3
polyunsaturated fatty acids and vitamin E after myocardial infarction:
results of the GISSI-Prevenzione trial. Gruppo Italiano per lo Studio
della Sopravvivenza nell’Infarto miocardico. Lancet 354, 447–455.
Kang JX & Leaf A (2000) Prevention of fatal cardiac arrhythmias by
polyunsaturated fatty acids. Am J Clin Nutr 71, 202S–207S.
Kardys I, Kors JA, van der Meer I, Hofman A, van der Kuip DA &
Witteman JC (2003) Spatial QRS-T angle predicts cardiac death in a
general population. Eur Heart J 24, 1357–1364.
Kors JA, van-Herpen G & van-Bemmel JH (1999) QT dispersion as an
attribute of T-loop morphology. Circulation 99, 1458–1463.
Marchioli R, Barzi F, Bomba E, et al. (2002) Early protection against
sudden death by n-3 polyunsaturated fatty acids after myocardial infarc-
tion – time-course analysis of the results of the Gruppo Italiano per lo
Studio della Sopravvivenza nell’Infarto Miocardico (GISSI)-Preven-
zione. Circulation 105, 1897–1903.
Ritsema van Eck HJ (2004) Fiducial segment averaging to improve
cardiac time interval estimates. J Electrocardiol 35, Suppl., 89–93.
Schouten EG, Dekker JM, Meppelink P, Kok FJ, Vandenbroucke JP
& Pool J (1991) QT interval prolongation predicts cardiovascular
mortality in an apparently healthy population. Circulation 84,
Trusz-Gluza M, Szydlo K, Kukla P, Wozniak-Skowerska I, Filipecki
A, Peszek E & Wojcik E (2002) Postextrasystolic repolarization
abnormalities in ST-U segment in patients with ventricular arrhythmias.
Ann Noninvasive Electrocardiol 7, 17–21.
van-Bemmel JH, Kors JA & van-Herpen G (1990) Methodology of the
modular ECG analysis system MEANS. Methods Inf Med 29, 346–353.
Zock PL, Mensink RP, Harryvan J, de-Vries JH & Katan MB (1997) Fatty
acids in serum cholesteryl esters as quantitative biomarkers of dietary
intake in humans. Am J Epidemiol 145, 1114–1122.
A. Geelen et al.790