Dynamic QT interval analysis in uraemic patients receiving
chronic haemodialysis
Simonetta Genovesi
a
, Rodolfo Rivera
b
, Paolo Fabbrini
a
, Chiara Dossi
a
,
Giuseppe Bonforte
b
, Luca Mircoli
d
, Alberto U. Ferrari
a
, Andrea Stella
a
and
Marco Stramba-Badiale
c
Objective To analyse the duration of the QT interval and
its relationship with heart rate changes in patients with
uraemia, before and during haemodialysis.
Methods QT and RR intervals were measured
automatically using a dedicated algorithm with 24-h Holter
recordings in 29 patients (15 women) receiving chronic
haemodialysis. QT corrected for heart rate (QTc) and the
slope of QT/RR linear regression were calculated. Arterial
blood pressure (ABP) was measured before and during
haemodialysis. Plasma concentrations of K , Mg
2
and
Ca
2
were assessed before and after haemodialysis.
Results ABP decreased significantly from baseline
(102.7 6 11.0 mmHg) during the first (100.6 6 8.8 mmHg,
P < 0.05), second (95.6 6 10.6 mmHg, P < 0.05), and third
(94.9 6 10.3 mmHg, P < 0.05) hours of haemodialysis. QTc
was longer during haemodialysis than during a 4-h period
of no dialysis (447 6 28 ms compared with 429 6 22 ms,
P < 0.001), and increased progressively during
haemodialysis, with the greatest value during the last hour
of haemodialysis (454 6 32 ms compared with
426 6 22 ms, P < 0.001). QT/RR slopes and correlation
coefficients were lower during haemodialysis than during
the period of no dialysis (0.13 6 0.08 compared with
0.20 6 0.07, P < 0.001 and 0.48 6 0.30 compared with
0.81 6 0.20, respectively; P < 0.001), suggesting a reduced
ability to adapt the QT interval in response to changes in
heart rate. The effects of haemodialysis on QT interval and
the QT/RR relationship were greater in women than in
men. QTc variations during dialysis were not correlated
with changes in ABP, but were inversely related to changes
in Ca
2
concentration (r
2
0.35; P 0.001).
Conclusions In patients with uraemia, the haemodialysis
session induces a progressive increase in QT interval and
modifies its relationship with heart rate. These effects may
predispose some individuals to ventricular arrhythmias at
the end of and immediately after the haemodialysis
session. J Hypertens 21:1921–1926 & 2003 Lippincott
Williams & Wilkins.
Journal of Hypertension 2003, 21:1921–1926
Keywords: 24-h Holter monitoring, sex, haemodialysis, QT interval, QT/RR
relationship, ventricular arrhythmias
a
DIMEP Universita ` degli Studi di Milano Bicocca, Monza,
b
Ospedale di Desio,
Desio,
c
IRCCS Istituto Auxologico Italiano and
d
IRCCS Ospedale Maggiore,
Milan, Italy.
Correspondence and requests for reprints to Dr Simonetta Genovesi, University
of Milano-Bicocca, Dipartimento di Medicina Clinica Prevenzione e Biotecnologie
Sanitarie, Via Cadore 48 20052 Monza, Italy.
Tel: +39 039 2332375; fax: +39 039 2332376;
e-mail: simonetta.genovesi@unimib.it
Received 30 April 2003 Revised 1 July 2003
Accepted 4 July 2003
Introduction
Patients with end-stage renal disease requiring main-
tenance haemodialysis have a high mortality rate that is
primarily attributable to cardiovascular diseases, includ-
ing ventricular arrhythmias and sudden death [1]. The
incidence of arrhythmias increases during and immedi-
ately after the haemodialysis session [2,3]. Previous
studies have shown that haemodialysis induces altera-
tions in the electrocardiogram (prolongation of the QT
interval) and increases spatial QT dispersion [4–7].
These abnormalities of ventricular repolarization may
favour ventricular arrhythmias. Prolongation of the QT
interval is associated with an increased risk of sudden
death in the long QT syndromes [8], in patients with
myocardial infarction [9] and even in healthy adult
individuals [10] or during infancy [11]. Furthermore,
increased QT dispersion is associated with greater risk
of malignant arrhythmias after myocardial infarction
[12] and in congestive heart failure [13].
Data on haemodialysis-related alterations in the ECG
have been obtained by measuring QT interval and
spatial QT dispersion from the standard 12-lead ECG
before and after the haemodialysis session. However,
the QT interval measured from a short ECG tracing
may not reflect the dynamic changes in duration of
repolarization and its relationship with heart rate.
Bazett’s formula is currently used to correct QT inter-
val for heart rate in resting conditions, but its use for
the long-term analysis of ventricular repolarization is
questionable. New developments in the computerized
analysis of Holter recording have made possible the
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Original article 1921
0263-6352 & 2003 Lippincott Williams & Wilkins DOI: 10.1097/01.hjh.0000084770.37215.c0