Kidney International, Vol. 57 (2000), pp.
2117–2122
QTc dispersion increases during hemodialysis with
low-calcium dialysate
SATU E. NÄPPI, VESA K. VIRTANEN, HEIKKI H.T. SAHA, JUKKA T. MUSTONEN,
and AMOS I. PASTERNACK
University of Tampere, Medical School, Tampere University Hospital, Tampere, Finland
QTc dispersion increases during hemodialysis with low-calcium
dialysate.
Background. The risk of ventricular arrhythmias is known
to increase during hemodialysis (HD) treatment, but the cause
of this phenomenon has remained unidentified. QT dispersion
(⫽ QTmax ⫺ QTmin) reflects heterogeneity of cardiac repolariza-
tion, and increased dispersion is known to predispose the heart
to ventricular arrhythmias and sudden cardiac death.
Methods. We studied the effect of dialysate calcium concentra-
tion on cardiac electrical stability during HD treatment in 23 end-
stage renal disease patients. Three HD treatments were applied
with dialysate Ca⫹⫹ concentrations of 1.25 mmol/L (dCa⫹⫹1.25),
1.5 mmol/L (dCa⫹⫹1.5), and 1.75 mmol/L (dCa⫹⫹1.75). The
QTc interval and QTc dispersion were measured before and
after the three sessions.
Results. With the dCa⫹⫹1.5 and dCa⫹⫹1.75 dialyses, serum
Ca⫹⫹ increased and the QTc interval remained stable (dCa⫹⫹1.5)
or decreased (dCa⫹⫹1.75), but no significant change was noted
in QTc dispersion. With dCa⫹⫹1.25 HD, serum Ca⫹⫹ decreased
(1.24 ⫾ 0.11 vs. 1.20 ⫾ 0.09 mmol/L, P ⬍ 0.05), and both the
QTc interval (403 ⫾ 27 vs. 419 ⫾ 33 ms, P ⬍ 0.05) and QTc
dispersion increased (38 ⫾ 19 vs. 49 ⫾ 18 ms, P ⬍ 0.05). The
change in the QTc interval correlated inversely with the change
in serum Ca⫹⫹ (r ⫽ ⫺0.68, P ⬍ 0.0001). Except for serum
Ca⫹⫹ and plasma intact parathyroid hormone, predialysis and
postdialysis values in other blood chemistry, blood pressure,
heart rate, body weight, and total ultrafiltration were equal in
the three dialysis sessions.
Conclusion. This study is the first, to our knowledge, to
demonstrate that HD increases QTc dispersion if a low-calcium
(dCa⫹⫹1.25) dialysate is used. This indicates that the use of low-
calcium dialysate may predispose HD patients to ventricular
arrhythmias and that perhaps it should be avoided, at least
when treating patients with pre-existing cardiac disease.
There is a high prevalence of cardiac arrhythmias in
patients with end-stage renal disease (ESRD) [1, 2], and
Key words: dialysis, calcium, chronic renal failure, QT interval, arrhyth-
mia, sudden cardiac death, ventricular arrhythmia.
Received for publication April 19, 1999
and in revised form November 16, 1999
Accepted for publication December 1, 1999
 2000 by the International Society of Nephrology
2117
the incidence of ventricular arrhythmias is further increased
during hemodialysis (HD) [1–4]. This arrhythmogenic
effect has been thought to result from pre-existing coro-
nary artery disease [5], cardiac systolic dysfunction [2],
changes in plasma potassium during HD [6], and abnor-
malities in serum calcium [2, 4] and parathyroid hormone
(PTH) [3, 4] metabolism. The ultimate cause, however,
remains obscure.
During HD, serum-ionized calcium varies directly with
the dialysate calcium concentration [7]. Changes in se-
rum calcium, in turn, strongly affect the length of the QT
interval [8]. Consequently, HD-induced acute changes in
serum calcium may contribute to the arrhythmogenic
effect of HD treatment.
Any regional differences in cardiac electrical recovery
predispose the heart to ventricular re-entry arrhythmias
[9]. Such disparity in repolarization can be assessed by
measuring the QT dispersion, that is, the variation in QT
interval length, on a standard 12-lead electrocardiogram
(ECG; QT dispersion ⫽ QTmax ⫺ QTmin). This procedure
has been shown to be a useful and reliable means of
identifying patients at risk of ventricular arrhythmias
[10–13]. High values for QT dispersion have been re-
ported in patients with left ventricular (LV) hypertrophy
[14, 15], myocardial infarction [11, 13], and ESRD [16].
The present study was designed to evaluate the effect
of dialysate calcium concentration on cardiac electrical
stability during HD treatment by measuring the QT in-
terval and QT dispersion before and after three HD
sessions, each with a different concentration of dialysate
calcium.
METHODS
Subjects
The subjects involved comprised 23 (20 males and 3
females) ESRD patients undergoing regular HD three
times a week, four to five hours per session. The age of
the patients ranged from 24 to 84 (median 51) years, and
the duration of HD therapy ranged from 1 to 108 (median
2118 Näppi et al: QTc dispersion and dialysate calcium
Fig. 1. Method of using a tangent drawn to the steepest point on the
down slope of the T wave to measure the QT interval.
12.5) months. The causes of ESRD were chronic glomer-
ulonephritis (N ⫽ 8), diabetic nephropathy (N ⫽ 5), poly-
cystic kidney disease (N ⫽ 4), hypertensive nephroscle-
rosis (N ⫽ 3), secondary amyloidosis (N ⫽ 2), and
hypoplastic kidney disease (N ⫽ 1). Nineteen patients
were hypertensive. Three had congestive heart disease,
and three had a myocardial infarction in their medical
history. Sixteen were receiving ␤-blockers, 12 calcium
antagonists, 5 angiotensin-converting enzyme inhibitors,
4 angiotensin-receptor antagonists, 5 nitrate agents, and
6 diuretics. All patients were taking vitamin B and vita-
min C supplements. Eighteen used calcium-containing
phosphate binders, and 9 used a vitamin D analogue.
Sixteen were on erythropoietin. No patient received
medication known to affect the QT interval, and all medi-
cations were continued unchanged throughout the study.
All patients gave informed consent, and the study was
approved by the Ethics Committee of Tampere Univer-
sity Hospital.
Protocol
All patients underwent three study HD sessions with
the dialysate Ca⫹⫹ concentrations of 1.25 mmol/L
(dCa⫹⫹1.25), 1.50 mmol/L (dCa⫹⫹1.50), and 1.75 mmol/L
(dCa⫹⫹1.75). Except for the concentration of Ca⫹⫹, the
compositions of the three dialysates were comparable.
At the beginning and the end of the three study HD
sessions, standard 12-lead ECG was performed at a pa-
per speed of 50 mm/s (Sicard 440; Siemens, Hoffmann
Estates, IL, USA). QT intervals were measured manu-
ally by a single observer (S.E.N), who was unaware of
the concentration of dialysate calcium used during the
session and whether the recordings were made before
or after HD treatment. Three consecutive complexes
were analyzed for each lead. The end of the T wave
was defined as the intersection of the tangent to the
descending limb of the T wave with the isoelectric line
(Fig. 1) [17]. If the T wave was flat or could not be
properly determined, the lead in question was excluded
from analysis. In no case were there fewer than seven
measurable leads. QT intervals were corrected for heart
rate with Bazett’s formula: QTc ⫽ QT/(RR)1/2. QT dis-
persion was defined as the difference between the maxi-
mum and minimum QT interval, and QTc dispersion was
computed.
To evaluate intraobserver variability in QT interval
and QT interval dispersion measurement, the ECGs of
15 patients were analyzed by a single observer (S.E.N)
on two different occasions. There were no significant
differences in QTc interval (404 ⫾ 26 vs. 404 ⫾ 28 ms,
P ⫽ NS) or QTc dispersion (42 ⫾ 21 vs. 45 ⫾ 18 ms,
P ⫽ NS) between the two recordings. According to the
method of Bland and Altman, the mean of the differ-
ences was 0.0 ⫾ 7 ms for the QTc interval and ⫺3.0 ⫾
27 ms for QTc dispersion, the corresponding 95% confi-
dence intervals ranging from ⫺3.9 to 3.9 ms and from
⫺18.0 to 12.0 ms [18].
Before and after each of the three study HD sessions,
blood pressure and heart rate were measured, and blood
samples were collected for the measurement of ionized
calcium, intact PTH, phosphate, sodium, potassium,
magnesium, creatinine, urea, hemoglobin, pH, and bicar-
bonate. The concentration of serum-ionized calcium was
measured with a Radiometer ICA-1 Analyzer (Radiom-
eter A/S, Copenhagen, Denmark) and that of plasma
intact PTH by two-site immunoradiometry (N-tact威 PTH
IRMA; Incstar Corp., Stillwater, MN, USA). Other labo-
ratory analyses were made by routine automatic methods.
Statistical analyses
The means and standard deviations of all variables
were calculated. Student’s t-test for paired samples and
simple-regression and multiple-regression analyses was
used to determine statistical significance, and P values ⬍
0.05 were considered significant. The Statgraphics威 (ver-
sion 7.0) statistical package was used.
RESULTS
dCa⫹⫹1.25 HD induced a decrease in serum-ionized cal-
cium (from 1.24 ⫾ 0.11 to 1.20 ⫾ 0.09 mmol/L, P ⬍ 0.05)
and a slight, although not significant, increase in plasma
intact PTH (Table 1). With dCa⫹⫹1.5 and dCa⫹⫹1.75
dialyses, serum ionized calcium increased, and plasma
intact PTH decreased (Table 1). The predialysis values
for serum-ionized calcium and plasma intact PTH were
comparable between the treatments. Both the predialysis
and the postdialysis values (except for plasma phos-
phate) and the dialysis-induced changes in other blood
chemistry, systolic, diastolic and mean arterial blood
pressure values, heart rate, body weight, and total ultra-
filtration were equal in the three sessions (Tables 1 and 2).
With dCa⫹⫹1.25 HD, the QTc interval was prolonged
(from 403 ⫾ 27 to 419 ⫾ 33 ms, P ⬍ 0.05), whereas with
dCa⫹⫹1.5 HD, it remained stable, and with dCa⫹⫹1.75
HD it was curtailed (Table 2 and Fig. 2). QTc dispersion
 Näppi et al: QTc dispersion and dialysate calcium 2119

Table 1. Laboratory values in 23 hemodialysis (HD) patients measured before (pre) and (post) after three dialysis treatments
with alternative dialysate
Dialysate Ca⫹⫹ 1.25 mmol/L Dialysate Ca⫹⫹ 1.5 mmol/L Dialysate Ca⫹⫹ 1.75 mmol/L
Laboratory values pre post pre post pre post
Serum ionized calcium
mmol/L 1.24 ⫾ 0.11 1.20 ⫾ 0.09a 1.25 ⫾ 0.10 1.35 ⫾ 0.06a,b 1.23 ⫾ 0.10 1.48 ⫾ 0.08a,b,c
Plasma phosphate mmol/L 1.89 ⫾ 0.53 0.94 ⫾ 0.27a 1.98 ⫾ 0.65 1.02 ⫾ 0.27a,b 1.91 ⫾ 0.60 1.04 ⫾ 0.29a,b
Plasma intact PTH pmol/L 5.3 (median) 9.0 (median) 11.2 (median) 6.4 (median)a,b 10.9 (median) 2.9 (median)a,b,c
1.2–103.0 (range) 1.4–102.0 (range) 1.9–124.0 (range) 0.8–100.0 (range) 1.2–138.0 (range) 0.8–76.5 (range)
Plasma magnesium mmol/L 0.93 ⫾ 0.16 0.84 ⫾ 0.07a 0.91 ⫾ 0.14 0.82 ⫾ 0.07a 0.95 ⫾ 0.15 0.83 ⫾ 0.07a
Blood hemoglobin g/L 113 ⫾ 9 120 ⫾ 13a 111 ⫾10 118 ⫾ 13a 116 ⫾ 13 121 ⫾ 14a
Plasma sodium mmol/L 141 ⫾ 2 142 ⫾ 2 141 ⫾ 3 141 ⫾ 2 140 ⫾ 2 141 ⫾ 2
Plasma potassium mmol/L 5.1 ⫾ 0.7 3.7 ⫾ 0.4a 5.0 ⫾ 0.8 3.8 ⫾ 0.3a 5.0 ⫾ 0.7 3.8 ⫾ 0.31a
Plasma chloride mmol/L 103 ⫾ 4 101 ⫾ 3a 103 ⫾ 4 102 ⫾ 2a 102 ⫾ 4 101 ⫾ 2a
Plasma creatinine lmol/L 847 ⫾ 260 346 ⫾ 127a 853 ⫾ 270 348 ⫾ 123a 821 ⫾ 251 346 ⫾ 117a
Serum urea mmol/L 20.3 ⫾ 4.8 6.6 ⫾ 2.7a 20.6 ⫾ 4.4 6.6 ⫾ 2.2a 19.9 ⫾ 3.2 6.3 ⫾ 1.7a
pH 7.38 ⫾ 0.04 7.46 ⫾ 0.05a 7.38 ⫾ 0.04 7.46 ⫾ 0.04a 7.38 ⫾ 0.04 7.45 ⫾ 0.05a
HCO3 mmol/L 23.3 ⫾ 2.4 28.5 ⫾ 1.2a 23.3 ⫾ 2.5 28.9 ⫾ 1.2a 23.1 ⫾ 2.3 28.1 ⫾ 1.2a
Reference ranges are: ionized calcium 1.20–1.35 mmol/L, phosphate 0.80–1.40 mmol/L, intact PTH 1.0–6.8 pmol/L, magnesium 0.75–1.00 mmol/L, hemoglobin
130–180 g/L, sodium 135–146 mmol/L, potassium 3.3–4.8 mmol/L, chloride 96–111 mmol/L, creatinine ⬍115 ␮mol/L, urea 3.0–8.5 mmol/L, pH 7.35–7.44, HCO3
22.0–27.0 mmol/L.
a
Predialysis vs. postdialysis, P ⬍ 0.05
b
Postdialysis value significantly (P ⬍ 0.05) different from that in dCa⫹⫹ 1.25 HD
c
Postdialysis value significantly (P ⬍ 0.05) different from that in dCa⫹⫹ 1.5 HD

Table 2. QT and QTc intervals, QT and QTc dispersions (⫽ QTmax ⫺ QTmin) and clinical data on the 23 hemodialysis (HD) patients
before (pre) and after (post) dialysis treatments with alternative dialysate Ca⫹⫹ concentrations
Dialysate Ca⫹⫹ 1.25 mmol/L Dialysate Ca⫹⫹ 1.5 mmol/L Dialysate Ca⫹⫹ 1.75 mmol/L
pre post pre post pre post
QT interval ms 375 ⫾ 42 378 ⫾ 48 383 ⫾ 49 373 ⫾ 42 a,b
377 ⫾ 45 352 ⫾ 44a,b,c
QTc interval ms 403 ⫾ 27 419 ⫾ 33a 403 ⫾ 40 401 ⫾ 35b 407 ⫾ 34 389 ⫾ 32a,b,c
QT dispersion ms 32 ⫾ 16 35 ⫾ 16 32 ⫾ 13 31 ⫾ 17 29 ⫾ 11 34 ⫾ 13
QTc dispersion ms 38 ⫾ 19 49 ⫾ 18a 36 ⫾ 15 41 ⫾ 21 36 ⫾ 14 40 ⫾ 15
Systolic blood pressure mm Hg 156 ⫾ 24 149 ⫾ 26 157 ⫾ 24 163 ⫾ 25 152 ⫾ 23 161 ⫾ 30
Diastolic blood pressure mm Hg 89 ⫾ 10 88 ⫾ 14 90 ⫾ 13 92 ⫾ 10 89 ⫾ 14 88 ⫾ 14
Mean arterial blood pressure mm Hg 122 ⫾ 16 119 ⫾ 18 124 ⫾ 17 128 ⫾ 15 121 ⫾ 16 125 ⫾ 19
Heart rate beats/min 69 ⫾ 13 74 ⫾ 16a 69 ⫾ 11 74 ⫾ 15a 70 ⫾ 14 77 ⫾ 17a
Body weight kg 79.9 ⫾ 19.2 77.9 ⫾ 18.3a 80.4 ⫾ 19.1 78.4 ⫾ 18.2a 79.5 ⫾ 18.8 77.9 ⫾ 18.7a
Total ultrafiltration ml 2525 ⫾ 1340 2156 ⫾ 1944 2204 ⫾ 1741
a
Before vs. after hemodialysis, P ⬍ 0.05
b
Postdialysis value significantly (P ⬍ 0.05) different from that in dCa⫹⫹ 1.25 HD
c
Postdialysis value significantly (P ⬍ 0.05) different from that in dCa⫹⫹ 1.5 HD

tended to increase with the dCa⫹⫹1.5 and dCa⫹⫹1.75
dialyses, but the dCa⫹⫹1.25 HD was the only procedure
that induced a significant increase in QTc dispersion
(from 38 ⫾ 19 to 49 ⫾ 18 ms, P ⬍ 0.05; Table 2 and Fig. 2).
The minimum QT interval changed from 354 to 356
ms (dCa⫹⫹1.25 HD, P ⫽ NS), from 367 to 354 ms
(dCa⫹⫹1.5 HD, P ⬍ 0.05) and from 360 to 333 ms
(dCa⫹⫹1.75 HD, P ⬍ 0.05) and the maximum QT interval
from 386 to 391 ms (dCa⫹⫹1.25 HD, P ⫽ NS), from 399
to 385 ms (dCa⫹⫹1.5 HD, P ⬍ 0.05), and from 389 to
367 ms (dCa⫹⫹1.75 HD, P ⬍ 0.05). The minimum QTc
interval changed from 380 to 385 ms (dCa⫹⫹1.25 HD,
P ⫽ NS), from 383 to 379 ms (dCa⫹⫹1.5 HD, P ⫽ NS)
and from 387 to 369 ms (dCa⫹⫹1.75 HD, P ⬍ 0.05), and
the maximum QTc interval from 418 to 434 ms
(dCa⫹⫹1.25 HD, P ⬍ 0.05), from 420 to 419 ms
(dCa⫹⫹1.5 HD, P ⫽ NS), and from 422 to 409 ms
(dCa⫹⫹1.75 HD, P ⬍ 0.05).
The change in QTc interval induced by the three study
HD sessions correlated inversely with the change in se-
rum ionized calcium (r ⫽ ⫺0.68, r 2 ⫽ 0.46, P ⬍ 0.0001;
Fig. 3). In multiple regression analyses, where the
changes in ionized calcium, intact PTH, phosphate, mag-
nesium, sodium, potassium, chloride, hemoglobin, creati-
nine, urea, bicarbonate, pH, mean body weight, total
ultrafiltration (UF), age, sex, duration of HD, systolic,
and diastolic and mean arterial blood pressure were inde-
pendent variables, the change in serum ionized calcium
was the only variable correlating independently with
change in the QTc interval. No correlations could be
found between any of these variables and change in QTc
dispersion (simple- and multiple-regression analyses).
2120 Näppi et al: QTc dispersion and dialysate calcium
Fig. 2. QTc interval (A) and QTc dispersion (⫽ QTmax ⫺ QTmin) (B) in
the 23 HD patients, measured before (䊐) and after (䊏) the three HD
treatments with different concentrations of dialysate Caⴙⴙ.
DISCUSSION
QT dispersion comprises the difference between the
longest and the shortest QT interval on a standard 12-
lead ECG, and reflects differences in ventricular recov-
ery times. A nonuniform electrical conduction in the myo-
cardium exposes the heart to ventricular arrhythmias [9].
There is convincing evidence of a relationship between
increased QTc dispersion and the occurrence of ventricu-
lar arrhythmias [11, 13, 19] and sudden cardiac death
[20] in patients with prolonged QTc interval [19], chronic
heart failure [20], or myocardial infarction [11, 13].
Cardiac structural and functional changes (for exam-
Fig. 3. Relationship between the changes in serum Caⴙⴙ and QTc inter-
val in the 23 HD patients during the three HD treatments with different
concentrations of dialysate Caⴙⴙ (r ⫽ 0.68; P ⬍ 0.0001).
ple, LV hypertrophy, chronic congestive heart disease,
and myocardial infarction) increase the QT dispersion
[11, 13, 14, 21, 22] and are at least partly responsible
for the high overall incidence of cardiac arrhythmias in
patients with ESRD. However, although the arrhythmo-
genic effect of HD treatment is well known [1–4], the
exact cause of HD-induced arrhythmias has remained
unidentified. These arrhythmias are most probably mul-
tifactorial in origin, and changes in serum calcium and
PTH metabolism have been proposed to be involved
[2–4]. This is well conceivable in that changes in serum
calcium are known to affect the electrical conduction
times of the myocardium [8]. In the present study, we
sought to establish whether the dialysate calcium concen-
tration affects the QT interval, especially QT dispersion.
The changes in the QTc interval here were, as expected,
inversely related to the changes in serum-ionized calcium
induced by the three different study HD sessions. The
QTc interval remained stable with the dCa⫹⫹1.5 HD
and decreased with the dCa⫹⫹1.75 HD. QTc dispersion
tended to increase with both dCa⫹⫹1.5 and dCa⫹⫹1.75
dialyses, although the changes were not significant. How-
ever, the decrease in serum-ionized calcium and prolon-
gation of QTc interval with the dCa⫹⫹1.25 HD were
accompanied by a significant increase in QTc dispersion.
Except for ionized calcium and intact PTH, the dialysis-
induced changes in other blood chemistry, blood pres-
sure, heart rate, body weight, and total UF were compa-
rable between the three procedures. Accordingly, with
respect to these parameters, the significant increase in
 Näppi et al: QTc dispersion and dialysate calcium
QTc dispersion with the dCa⫹⫹1.25 HD can well be ex-
plained by the decrease in serum ionized calcium.
To our knowledge, there are only two reports on the
effects of HD on QT dispersion [23, 24]. In the study by
Cupisti et al, in contrast to our results, the QTc dispersion
was seen to increase significantly during HD with a high-
calcium (2.0 mmol/L) dialysate [23]. However, the in-
crease in serum total calcium was surprisingly accompa-
nied by an increase in QTc interval, which would suggest
that other confounding factors in their study contributed
to the HD-induced changes in QTc interval and QTc
dispersion. Compatible with our results, Morris et al have
recently demonstrated that HD treatment with reduced
serum calcium increased both the QTc interval and QTc
dispersion [24].
The overall QTc intervals in our patients were rela-
tively short, but this arises from the use of a tangent
method [13, 15, 17, 25] in defining the end of the T wave.
Correspondingly short QTc intervals in HD patients have
been reported by other authors who used the same
method [26]. In the present study where the QT intervals
were measured in each patient before and after dialysis,
the patients thus serving as their own controls, the tan-
gent method offered a reliable means of observing the
HD-induced changes in QT interval.
No correlation was found between the HD-induced
changes in QTc dispersion and changes in serum-ionized
calcium or other electrolytes. Furthermore, QTc disper-
sion tended to also increase during the dCa⫹⫹1.5 and
dCa⫹⫹1.75 dialyses. It has been demonstrated that a
decreased intracellular or extracellular potassium con-
centration [27, 28] or a high removal rate of potassium
during HD increases the incidence of arrhythmias [6, 29].
Additionally, changes in the autonomic nervous system
may also enhance the arrhythmogenic effect of HD treat-
ment [16, 29]. This implies that the cause of the increase
in QTc dispersion during HD is multifactorial. However,
in the present study, the increase in QTc dispersion only
during the dCa⫹⫹1.25 HD cannot be explained by
changes in plasma potassium concentration, since the
decrease in plasma potassium was equal during all three
procedures. It is also possible that changes in serum
calcium, and hence those in the QTc interval, in either
direction increase the QTc dispersion, but that only the
prolongation of the QTc interval has a significant effect
on QTc dispersion. The tendency of QTc dispersion to
increase along with the QTc interval has been demon-
strated in several studies [21–24, 30], and prolongation
of the QTc interval, per se, has been documented as a
risk factor for ventricular arrhythmias in HD patients
[2, 31]. Thus, an increase in both QT interval and QT
dispersion represents abnormal ventricular repolariza-
tion and hence increases the risk of ventricular arrhyth-
mias. This is in keeping with recent study results showing
the prolongation of the QTc interval and QTc dispersion
2121
to be independent risk factors underlying increased car-
diovascular mortality in patients with pre-existing cardio-
vascular diseases [30].
Our present study is, to our knowledge, the first to
show the effect of dialysate Ca⫹⫹ concentration on QT
interval and QT dispersion, and the findings may have
clinical implications. In our previous studies, we have
shown that an acute induction of hypercalcemia by cal-
cium infusion [32] or by HD with a high-calcium dialysate
[33] impairs LV relaxation. The present results, in turn,
show that the use of a low-calcium dialysate increases
the QT interval and dispersion and hence may predis-
pose HD patients to ventricular arrhythmias during
treatment. This suggests that the dialysate calcium con-
centration affects both the relaxation properties and the
electrical stability of the myocardium; thus, the use of a
dialysate Ca⫹⫹ concentration which raises serum Ca⫹⫹
near to the upper limit of the normal range might be
advocated, at least when treating patients with pre-exist-
ing cardiac diseases.
ACKNOWLEDGMENTS
This study was supported by grants from the Finnish Kidney Founda-
tion, the Pirkanmaa Kidney Society, and the Medical Research Fund
of Tampere University Hospital.
Reprint requests to Heikki Saha, M.D., University of Tampere, Medi-
cal School, Box 607, FIN-33101 Tampere, Finland.
E-mail: llhesa@uta.fi
REFERENCES
1. Gruppo Emodialisi e Patologie Cardiovascolari: Multicenter,
cross-sectional study of ventricular arrhythmias in chronically
haemodialysed patients. Lancet 6:305–309, 1988
2. Kimura K, Tabei K, Asano Y, Hosoda S: Cardiac arrhythmias in
hemodialysis patients. Nephron 53:201–207, 1989
3. Ramirez G, Brueggemeyer CD, Newton JL: Cardiac arrhythmias
on hemodialysis in chronic renal failure patients. Nephron 36:212–
218, 1984
4. Strata G, Limbruno U, Leoncini GP, Di Vincenzo A, Palla R,
Mariani M: Physiopathology of cardiac arrhythmias in end-stage
renal disease. Contrib Nephrol 106:30–35, 1994
5. Wizemann V, Kramer W, Schutterle G: Dialysis-induced cardiac
arrhythmias: Fact or fiction? Nephron 39:356–360, 1985
6. Redaelli B, Locatelli F, Limido D, Andrulli S, Signorini MG,
Sforzini S, Bonoldi L, Vincenti A, Cerutti S, Orlandini G:
Effect of new model of hemodialysis potassium removal on the
control of ventricular arrhythmias. Kidney Int 50:609–617, 1996
7. Saha H, Harmoinen A, Pietilä K, Mörsky P, Pasternack A:
Measurement of serum ionized versus total levels of magnesium
and calcium in hemodialysis patients. Clin Nephrol 46:326–331,
1996
8. Braunwald E, Sonnenblick E, Ross J: The ECG and electrolyte
abnormalities, in Heart Disease (4th ed), edited by Braunwald
E, Philadelphia, Saunders, 1992, pp 149–151
9. Merx W, Yoon MS, Han J: The role of local disparity in conduction
and recovery time on ventricular vulnerability to fibrillation. Am
Heart J 94:603–610, 1977
10. Day CP, McComb JM, Matthews J, Campbell RWF: Reduction
in QT dispersion by sotalol following myocardial infarction. Eur
Heart J 12:423–427, 1991
11. Van de Loo A, Arendts W, Hohnloser SH: Variability of QT
dispersion measurements in the surface electrocardiogram in pa-
2122 Näppi et al: QTc dispersion and dialysate calcium
tients with acute myocardial infarction and in normal subjects. Am
J Cardiol 74:1113–1118, 1994
12. Ashikaga T, Nishizaki M, Arita M, Yamawake N, Kishi Y, Nu-
mano F, Hiraoka M: Increased QTc dispersion predicts lethal
ventricular arrhythmias complicating coronary angioplasty. Am J
Cardiol 82:814–816, 1998
13. Perkiömäki JS, Koistinen MJ, Yli-Mäyry S, Huikuri HV: Disper-
sion of QT interval in patients with and without susceptibility to
ventricular tachyarrhythmias after previous myocardial infarction.
J Am Coll Cardiol 26:174–179, 1995
14. Ichkhan K, Molnar J, Somberg J: Relation of left ventricular
mass and QT dispersion in patients with systematic hypertension.
Am J Cardiol 79:508–511, 1997
15. Perkiömäki JS, Ikäheimo MJ, Pikkujämsä SM, Rantala A, Lilja
M, Kesäniemi A, Huikuri HV: Dispersion of the QT interval and
autonomic modulation of heart rate in hypertensive men with and
without left ventricular hypertrophy. Hypertension 28:16–21, 1996
16. Kirvelä M, Yli-Hankala A, Lindgren L: QT dispersion and
autonomic function in diabetic and non-diabetic patients with renal
failure. Br Heart J 73:801–804, 1994
17. Browne KF, Prystowsky E, Heger JJ, Zipes DP: Modulation of
the QT interval by the autonomic nervous system. Pacing Clin
Electrophysiol 6:1050–1056, 1983
18. Bland MJ, Altman DG: Statistical methods for assessing agree-
ment between two methods of clinical measurement. Lancet 8:307–
310, 1986
19. Day CP, McComb JM, Campbell RW: QT dispersion: An indica-
tion of arrhythmia risk in patients with long QT intervals. Br Heart
J 63:342–344, 1990
20. Barr CS, Naas A, Freeman M, Lang CC, Struthers AD: QT
dispersion and sudden unexpected death in chronic heart failure.
Lancet 343:327–329, 1994
21. Mayet J, Shahi M, McGrath K, Poulter NR, Sever PS, Foale
RA, Thom SA: Left ventricular hypertrophy and QT dispersion
in hypertension. Hypertension 28:791–796, 1996
22. Gonzales-Juanatey JR, Garcia-Acuna JM, Pose A, Varela A,
Calvo C, Cabezas-Cerrato J, De la Pena MG: Reduction of
QT and QTc dispersion during long-term treatment of systemic
hypertension with enalapril. Am J Cardiol 81:170–174, 1998
23. Cupisti A, Galetta F, Morelli E, Tintori G, Sibilia G, Meola
M, Barsotti G: Effect of hemodialysis on the dispersion of the
QTc interval. Nephron 78:429–432, 1998
24. Morris STW, Galiatsou E, Stewart GA, Rodger RSC, Jardine
AG: QT dispersion before and after hemodialysis. J Am Soc
Nephrol 10:160–163, 1999
25. Fei L, Statters DJ, Camm JA: QT-interval dispersion on 12-lead
electrocardiogram in normal subjects: Its reproducibility and rela-
tion to the T wave. Am Heart J 127:1654–1655, 1994
26. Erem C, Kulan K, Göldeli Ö, Tuncer C, Komsuoglu B, Mocan
M: Impact of hemodialysis on QT interval. Acta Cardiol 3:177–185,
1995
27. Rombola G, Colussi G, Elisabetta De Ferrari M, Frontini A,
Minetti L: Cardiac arrhythmias and electrolyte changes during
haemodialysis. Nephrol Dial Transplant 7:318–322, 1992
28. Choy AM, Lang CC, Chomsky DM, Rayos GH, Wilson JR,
Roden DM: Normalization of acquired QT prolongation in hu-
mans by intravenous potassium. Circulation 96:2149–2154, 1997
29. Abe S, Yoshizawa M, Nakanishi N, Yazawa T, Yokota K, Honda
M, Sloman G: Electrocardiographic abnormalities in patients re-
ceiving hemodialysis. Am Heart J 131:1137–1144, 1996
30. Elming H, Holm E, Jun L, Torp-Pedersen C, Kober L, Kirschoff
M, Malik M, Camm J: The prognostic value of the QT interval
and QT interval dispersion in all-cause and cardiac mortality and
morbidity in a population of Danish citizens. Eur Heart J 19:1391–
1400, 1998
31. Suzuki R, Tsumura K, Inoue T, Kishimoto H, Morii H: QT
interval prolongation in the patients receiving maintenance hemo-
dialysis. Clin Nephrol 49:240–244, 1998
32. Virtanen VK, Saha HHT, Groundstroem KWE, Seppälä ES,
Pasternack AI: Calcium infusion and left ventricular diastolic
function in patients with chronic renal failure. Nephrol Dial Trans-
plant 13:384–388, 1998
33. Näppi SE, Saha HHT, Virtanen VK, Mustonen JT, Pasternack
AI: Hemodialysis with high-calcium dialysate impairs cardiac relax-
ation. Kidney Int 55:1091–1096, 1999