Content uploaded by Juan Carlos Kaski
Author content
All content in this area was uploaded by Juan Carlos Kaski
Content may be subject to copyright.
European Heart Journal (1997)
18,
1269-1277
Effect
of
magnesium sulphate
in
patients with
unstable angina
A double blind, randomized, placebo-controlled study
S.
R.
Redwood,
Y.
Bashir,
J.
Huang,
E. W.
Leatham, J.-C. Kaski and
A. J.
Camm
Department
of
Cardiological Sciences,
St
George's Hospital Medical School, London,
U.K.
Aims Administration
of
intravenous magnesium sulphate
has been shown
to be
protective during acute myocardial
ischaemia
and it may
therefore have beneficial effects
in
unstable angina.
The
purpose
of
this study was
to
assess
the
effects
of a 24-h
infusion
of
magnesium
in
patients with
unstable angina.
Methods
and
results Patients
who
presented with
unstable angina with electrocardiographic changes were
randomized
to
receive
a 24-h
intravenous infusion
of
mag-
nesium
or
placebo within 12
h of
admission.
The
primary
endpoint
was
myocardial ischaemia,
as
assessed
by 48 h
Holter monitoring. Resting 12-lead ECGs, creatine kinase-
MB release
and
urinary catecholamines were also assessed.
Patients were followed
for 1
month. Thirty-one patients
received magnesium sulphate
and 31
placebo. Baseline
characteristics
and
extent
of
coronary disease were similar
in both groups.
On 48 h
Holter monitoring,
14
patients
(50%)
had
transient
ST
segment shifts
in the
magnesium
group
vs 12
patients (46%)
in the
placebo group. However,
there were fewer ischaemic episodes
in the
magnesium
group (51
vs
101, /)<0001)
and
there
was a
trend towards
an increase
in the
total duration
of
ischaemia
in the
placebo
group compared
to the
magnesium group
in the
second
24 h
(2176 min
vs
719 min respectively,
P-008).
Regression
of
T wave changes
on the
24
h
ECG occurred more frequently
in patients
who
received magnesium compared
to
those
treated with placebo (11 patients
vs 0
patients respectively,
/><0005). Creatine kinase-MB release
was
significantly
less
at 6 and
24
h in
patients
who
received magnesium
compared
to
those treated with placebo. Catecholamine
excretion
was
lower
in
patients treated with magnesium
than
in
those treated with placebo (adrenaline: 1 05
±016
vs
1-61 ±
0-32
ng .
mmol
~ '
creatinine; noradrenaline:
9-99 ±1-82
vs
18-48
±
2-41 ng.mmol"' creatinine
respectively
in the
first 12
h
sample, P<005).
Conclusions Intravenous magnesium reduces ischaemic
ECG changes, creatine kinase-MB release
and
urinary
catecholamine excretion
in the
acute phase
of
unstable
angina. Thus, magnesium
may be a
beneficial additional
therapy
for
these patients. Further studies
are
required
to
confirm these findings.
(Eur Heart
J
1997;
18:
1269-1277)
Key Words: Magnesium, unstable angina, ischaemia,
catecholamines.
Introduction
There are several lines of evidence supporting a poten-
tially beneficial role of magnesium during acute myocar-
dial ischaemia'1'. Several studies have suggested that
magnesium may reduce mortality and serious arrhyth-
mias post acute myocardial infarction'2^*1. However, this
was not supported by the recently published results of
ISIS-4'51,
which showed that a 24-h infusion of mag-
nesium has no beneficial effect in those receiving throm-
bolysis for acute myocardial infarction. In this study,
Submitted
8
October 1996,
and
accepted 30 October
1996.
Correspondence:
Dr
Simon Redwood, Department
of
Cardiology,
The Rayne Institute,
St
Thomas' Hospital, London, SE1
7EH.
patients were randomized after the thrombolytic agent
had been administered, a mean of 8 h after the onset of
pain, compared to a median of
3
h in the LIMIT-2 trial,
which showed a 25% reduction in mortality'21. Animal
studies suggest that the effect of magnesium is greatest if
given before spontaneous or induced reperfusion'61, and
this may account, in part, for the lack of benefit seen
in ISIS-4.
The pathophysiology of unstable angina involves
a combination of an ulcerated atherosclerotic plaque
with superimposed dynamic coronary obstruction,
mediated by coronary vasoconstriction and/or platelet
aggregation and thrombosis'7'8'. In patients who have
ST segment and/or T wave changes on resting 12-lead
electrocardiograms (ECGs), there is a higher risk of
failure of medical therapy, progression to myocardial
0195-668X/97/081269+09 S18.00/01997 The European Society
of
Cardiology
by guest on July 21, 2011eurheartj.oxfordjournals.orgDownloaded from
1270 S. R. Redwood et al.
infarction and an increased mortality'9'101. Furthermore,
in patients undergoing continuous ECG monitoring,
the presence of myocardial ischaemia predicts adverse
prognosis, with an increased frequency of major cardiac
events111-121.
Magnesium has several effects which may be of
benefit in unstable angina. For example, it has been
shown to reduce intracellular calcium overload'13'141,
thought to be a central mechanism in ischaemic myo-
cardial death'151; it has been shown to dilate coronary
arteries'16'171 and to have favourable haemodynamic
effects, reducing total peripheral resistance'17'181. In
addition, it inhibits platelet function'19'20', possibly by an
effect on prostacyclin secretion'2'1. Furthermore, mag-
nesium deficiency has been associated with enhanced
sympathetic activity'22'231 and magnesium has been
shown to attenuate catecholamine release'241 and to
reduce the size of catecholamine induced myocardial
necrosis'251.
The use of magnesium in unstable angina has not
been previously assessed; however, one study in sus-
pected acute myocardial infarction suggested a reduced
incidence of confirmed infarction in those given magne-
sium'261, but this has not been confirmed in other studies.
The aims of this study were therefore to assess the effects
of a 24-h infusion of magnesium sulphate on myocardial
ischaemia, creatine kinase MB release, and catechol-
amine excretion in patients with unstable angina.
Methods
All patients with a diagnosis of unstable angina with
ECG changes who satisfied entry criteria were eligible.
Patients were recruited between March 1993 and April
1994 from the coronary care unit of St George's Hospi-
tal,
with a population typical of a district general
hospital. During this period, 380 patients were admitted
with chest pain as their main presenting symptom. Of
these, 169 patients (45%) had a diagnosis of myocardial
infarction, 179 patients (47%) had a primary diagnosis
of unstable angina and 32 patients (8%) had non-cardiac
chest pain. Of the 169 patients with unstable angina, 85
(47%) had associated ECG changes. Eight patients (4%)
initially diagnosed as unstable angina were subsequently
shown to have sustained a myocardial infarction, based
on cardiac enzyme and ECG results. The study protocol
was approved by the Regional Ethics Committee and all
patients gave fully informed written consent.
Entry criteria were unstable angina, defined as
angina at rest of at least 10 min duration within the last
24 h, with associated transient ST segment depression
(planar or downsloping ST segment shift >01 mV
measured at 008 s after the J point) and/or transient T
wave inversion, and absence of exclusion criteria. Exclu-
sion criteria included serum creatinine >300 umol. 1~ ',
complete heart block, baseline ECG changes making ST
segment interpretation difficult (left bundle branch
block, paced rhythm or left ventricular hypertrophy)
and patient inability/refusal to give informed consent.
Women of childbearing potential were not included in
the study.
Study protocol
Magnesium infusion
Patients were randomized to receive either magnesium
or placebo in a double-blind manner. Identical treat-
ment packs for magnesium and placebo, identified by
number only, were manufactured according to a rand-
omization schedule kept only by the St George's Hospi-
tal pharmacy. Packs contained either magnesium
sulphate (8 mmol bolus diluted to 100 ml in normal
saline to be given intravenously over 20 min and
72 mmol in 36 ml administered at a rate of
3
mmol. h ~'
for 24 h) or an identical volume of normal saline.
Unless contraindicated, all patients received oral
aspirin 150mg.day~', oral beta-adrenergic blocking
agents, intravenous nitrates at the minimum dose to
render them pain free and intravenous heparin (to keep
the activated partial thromboplastin time between 1-5
and 2-5 x baseline). The activated partial thromboplas-
tin time was measured at 6 h from study entry and daily
thereafter.
Data collected included medical history, present-
ing symptoms, antianginal medication, laboratory data,
12-lead ECGs at study entry and at 24 and 48 h and
events during admission and follow-up. ECGs taken at
24 and 48 h after admission were compared with the
admission ECG. ST segment and T wave changes were
classified as inferior, anterior, septal or lateral according
to ECG leads involved. Resolution of ECG changes was
defined as abolition of changes or a reduction in the
number of territories involved. Progression was defined
as the development of new changes or the appearance of
changes in a new ECG territory.
24 h Holier monitoring
All patients underwent 48 h continuous ECG monitor-
ing using a Marquette two-channel recorder. Recordings
were made from two bipolar leads (II and CM5) as
previously described'271. Tapes were analysed by an
independent cardiologist without knowledge of patient
identity. An ischaemic episode was defined as >01 mV
ST segment depression at 008 s after the J point lasting
for >
1
min. Return of the ST segment to baseline for
>
1
min was required between episodes. The number
and duration of ischaemic episodes were assessed during
the period of monitoring. The duration of each episode
was determined from the lead that showed the more
prolonged ECG changes.
Blood and urine tests
Venous blood samples were obtained, centrifuged and
frozen at - 20 °C at 0, 6, 24 and 48 h for measurement
of creatine kinase-MB and its isoforms'281. Four iso-
forms were measured: creatine kinase-MB (MB2, tissue
isoform and MB1, plasma-modified isoform) and
Eur Heart J, Vol. 18, August 1997
by guest on July 21, 2011eurheartj.oxfordjournals.orgDownloaded from
Magnesium sulphate in unstable angina 1271
Table 1 Clinical characteristics of the patients in the magnesium (n=31) and
placebo (n=31) groups
Age
(mean
± SEM)
Male
Onset
of
pain
(h)
Duration
of
pain
(h)
Last pain
to
study entry
Smoking:
Previous infarct
Previous angina
Previous revascularization
Family history
of IHD
Hypertension
Diabetes
Never
Ex
Current
Inferior
Anterior
Duration (years)
PTCA
CABG
Magnesium
63-6 ± 2-4
20 (65%)
9-8 ± 1-4
2-4 ± 0-4
7-4 ± 1-4
10 (32%)
13 (42%)
8 (26%)
9 (29%)
7 (23%)
2 (7%)
23 (74%)
5-9 ±1-5
9 (29%)
1 (3%)
8 (29%)
8 (26%)
12 (39%)
1 (3%)
Placebo
620 ±2-5
19 (61%)
12-3 ± 1-4
2-6 ±0-5
9-7 ±1-3
7 (23%)
9 (29%)
15 (48%)
13 (42%)
10 (32%)
3 (10%)
19 (61%)
l-4± 1-8*
8 (26%)
5 (16%)
3 (10%)
9 (29%)
9 (29%)
0
*P<005
PTCA = percutaneous transluminal coronary
angioplasty;
CABG = coronary artery bypass
surgery;
lHD=ischaemic heart
disease.
creatine kinase-MM (MM3, tissue isoform and MM1,
plasma-modified isoform). In addition, serum mag-
nesium was measured in the first 32 patients. Four
consecutive 12 h samples of urine were collected from
study entry in 30 patients. All samples were acidified,
measured and 20 ml aliquots retained
and frozen at
—
70 °C for later analysis using high
performance liquid chromatography.
Outcome variables
The primary endpoint of this study was to assess the
effects of magnesium on ischaemic ST segment shifts. In
addition, creatine kinase-MB and isoform release and
catecholamine excretion were studied. The occurrence of
death, non-fatal myocardial infarction and need for
myocardial revascularization (percutaneous translumi-
nal coronary angioplasty or coronary artery bypass
surgery) were also recorded. Myocardial infarction was
defined as the development of new pathological Q waves
with at least a twofold increase in cardiac enzyme levels.
Statistical analysis
Sample size calculation was based on an expected inci-
dence of transient ST segment change of 65% in this
population of patients'29'. Assuming that magnesium
resulted in a 30% reduction in transient myocardial
ischaemia, it was calculated that 30 patients in each
group would have an 80% chance of detecting this
reduction. Discrete data were compared by contingency
table analysis with the chi-square test. Differences in
continuous variables were analysed using the non-paired
t-test or non-parametric statistics (Kruskal-Wallis) as
appropriate. Data expressed as mean ± SEM. A P value
<005 was considered significant.
Results
Patients
Sixty-two patients satisfied all entry criteria and were
therefore randomized to receive magnesium (31 patients;
age 636±2-4 years) or placebo (31 patients; age
620 ± 2-5 years). Baseline clinical characteristics (Table
1),
full blood count and urea and electrolytes were
similar between the two treatment groups.
The magnesium infusion was well tolerated and
was completed in all but two patients (3%) due to
flushing. No patients developed significant hypotension
or conduction disturbances. Serum magnesium rose
from 0-84±003 to
1-69±005
mmol.
1
at 24h in the
magnesium group (Fig. 1) and returned towards baseline
values at 48 h, with no change in the placebo group.
Following hospital admission (Table 2), a similar
proportion of patients in the magnesium and placebo
groups received oral aspirin, beta-blockers and calcium
antagonists, intravenous nitrates and intravenous
heparin before study entry. Activated partial thrombo-
plastin times were similar in the two groups at 6, 24
and 48 h.
Twenty-seven patients (87%) in the magnesium
group and 25 (81%) in the placebo group underwent
pre-discharge left ventriculography and selective cor-
onary arteriography. There was no significant difference
Eur
Heart
J, Vol. 18,
August
1997
by guest on July 21, 2011eurheartj.oxfordjournals.orgDownloaded from
1272
S. R.
Redwood
et al.
1.80
1.40 -
1.00 -i
0.60
12 18
24
30
Time
(h)
36
42 48
Figure
1 The
effect
of the
study infusion (magnesium
sulphate
8
mmol bolus followed
by 3-0
mmol. h
~ ' for
24
h, or
placebo)
on
serum magnesium concentrations
in
the magnesium
(•,
n
=
16)
and
placebo
(D,
n
=
16) groups.
01
in
the
number
of
diseased vessels,
the
distribution
of
disease
or
left ventricular function.
Evolution
of
resting
ECG
changes
and
continuous
48 h ECG
monitoring
All patients
had ST
segment and/or
T
wave changes
on
entry
to the
study,
and
these were similarly distributed
between
the two
groups.
A
similar proportion
had
pre-existing
Q
waves
in a
similar distribution.
At
24
h,
there
was
resolution
of T
wave changes
in 11
patients
(36%)
in the
magnesium group
and in no
patients
in the
placebo group (Z'<0005), with progression
in
seven
patients
(23%) in the
magnesium group
and in 10
patients
(32%) in the
placebo group.
Continuous
ECG
monitoring (Table
3) was ana-
lysed
in 54
patients
(28
patients
in the
magnesium group
and
26 in the
placebo group).
In the
remaining eight
patients, recordings were
of
poor quality
and
analysis
was
not
possible.
A
total
of
2387
h of ST
segment
monitoring
was
obtained (1282
h in the
magnesium
Table
2
Additional drug therapy following admission
to
hospital
MagnesiumPlacebo
Aspirin
Beta-blocker
Calcium antagonist
iv nitrate
Duration (days, mean ± SEM)
iv heparin
Duration (days, mean ± SEM)
27 (87%)
20 (65%)
17 (55%)
26 (84%)
3-1 ±0-5
28 (90%)
3-5 ±0-6
28 (90%)
18 (58%)
12(39%)
27 (87%)
2-6
±0-5
30 (97%)
3-6 ±0-6
iv=intravenous.
Table
3
Continuous monitoring of the
ST
segment
in the
first
and
second
24
h
in the
magnesium
and
placebo groups
Patients (n)
First 24 h
No.
(%) with >
1
episode
ST change
Total episodes (n)
Mean ± SEM
Total duration of ST change (min)
Mean ± SEM
Of those with ST change (mean ± SEM)
Second 24 h
No.
(%) with >
1
episode of ST change
Total episodes (n)
Mean ± SEM
Total duration of ST change (min)
Mean ± SEM
Of those with ST change (mean ± SEM)
Magnesium
28
11 (39%)
37
1-3 ±0-5
917-3
32-8
±141
83-4
± 181
5 (18%)
14f
0-5 ±0-3
7191
25-7 ±17-0
143-8
±310
Placebo
26
10 (39%)
54*
21 ± 10
884-8
340 ± 151
88-5
± 19-3
10 (39%)
47*
1-8 ±0-7
2175-9§
83-7 ±400
217-6 ±55-3
*/3<0001 compared to the magnesium group, t/><005 compared to first 24 h.
§/><005
compared
to the first 24 h.
Eur Heart J. Vol. 18. August 1997
by guest on July 21, 2011eurheartj.oxfordjournals.orgDownloaded from
Magnesium sulphate in unstable angina 1273
Table 4
Patient
No.
Details
Age
(years)
Magnesium Group
1
2
3
4
5
6
7
8
9
10
11
12
13
14
72
72
76
39
40
61
77
57
61
71
80
56
59
64
63-2|
Placebo Group
1
2
3
4
5
6
7
8
9
10
11
12
44
83
61
60
57
78
60
62
73
69
57
58
63-5f
of
the
patients with one
Sex "
M
F
M
F
M
F
M
F
M
F
M
F
M
M
M
M
M
M
F
M
M
F
F
F
M
F
Oh
55
92
59
80
92
60
71
83
83
100
86
100
66
82
79-2f
86
92
50
92
60
75
75
75
90
65
78
74
76-Of
Heart rate
24 h
56
86
50
86
63
59
55
71
55
63
73
72
59
64
65 It
59
92
46
86
55
100
50
75
85
72
82
82
73-7f
or more
48 h
52
92
52
75
80
60
60
75
51
60
67
67
64
68
65-9f
55
75
63
75
53
86
60
60
80
68
69
59
66-9f
episodes
iof ST
segment
ST changes*
1st 24 h
Y
Y
N
Y
Y
Y
N
Y
Y
Y
Y
Y
Y
N
11 patients
Y
Y
Y
Y
Y
Y
Y
Y
N
Y
Y
N
10 patients
' change
On admission
2nd 24 h Beta-blocker|
N
Y
Y
N
N
N
Y
N
Y
N
N
N
N
Y
5 patients 4
Y
Y
N
Y
Y
Y
N
Y
Y
Y
Y
Y
10 patients 3
A, 50
P,
160
A, 50
M, 75
patients
A, 50
P,
30
A, 50
patients
Calcium
antagonist§
D,
180
N,
40
D,
60
D,
180
A, 10
5 patients
D,
360
D,
120
D,
180
D,
180
N,
40
5 patients
After admission
Beta-blockerJ:
A, 50
A, 50
P,
160
A, 50
M, 75
A, 50
6 patients
M, 100
A, 50
M, 75
A, 50
A, 50
5 patients
Calcium
antagonist§
D,
180
N,
40
D,
240
D,
180
D,
180
A, 10
6 patients
D,
360
D,
120
D,
180
N,
40
4 patients
*Y
=
yes,
N =
no;
f
=
mean;
JA = atenolol, P=propranolol, M
=
metoprolol. Dose given in mg . day ' (e.g A, 50=atenolol, 50 mg . day ');
§D= diltiazem, N
=
nifedipine, A
=
amlodipine.
group and 1105 h in the placebo group). There were one
or more episodes of reversible ST segment shift in 14
patients (50%) of the magnesium group and 12 patients
(46%) of the placebo group, giving a total of
51
episodes
in the magnesium group and 101 episodes in the placebo
group (/)<0001). This difference in the number of
episodes was evident in both the first and second 24 h of
recording. In addition, there was a reduction in the
number of episodes that occurred in the magnesium
group in the second 24 h compared with the first 24 h
(14 vs 37 episodes respectively, /><005), with no signifi-
cant change in the placebo group (47 vs 54 episodes
respectively).
There was a trend towards a reduction in the
total duration of ischaemia in the magnesium group
compared to the placebo group, but this did not reach
statistical significance (1636 vs 3061 min respectively).
Most of this difference in total duration of ischaemia
was during the second 24 h of recording (after the study
infusion had been completed); 719 min in the mag-
nesium group and 2176 min in the placebo group
(/)=0-08). However, there was a significant increase in
the total duration of ischaemia in the placebo group in
the second 24 h compared to the first 24 h (2176 min vs
885 min, /)<0-05), which was not seen in the magnesium
group. In addition, during the second 24 h of recording,
five patients (18%) in the magnesium group and 10
(39%) in the placebo group had one or more episodes of
ST segment shift.
Details of the 26 patients who had >
1
episode of
ST segment shift are given in Table 4. There were no
significant differences in the mean age, heart rates at 0,
24 or 48 h, or in the proportion of patients on beta-
blockers or calcium antagonists either before or after
admission between the two groups.
Cardiac enzymes
There was no difference in total creatine kinase, aspar-
tate transaminase or lactate dehydrogenase either on
admission, or at 24 or 48 h. Creatine kinase-MB release
was less in the magnesium group (Fig. 2) at 6 and 24 h
(6-6 ±
1 -2
vs 28-3 ± 7-4 ng
.
ml" ' at
6
h, and 3-3 ± 0-7 vs
Eur Heart J, Vol. 18, August 1997
by guest on July 21, 2011eurheartj.oxfordjournals.orgDownloaded from
1274
S. R
Redwood
et al.
2448
Hours
Figure
2 The
serum levels
of
creatine kinase-MB
in
the magnesium
(D) and
placebo
(•)
groups. *P<0-05
vs
magnesium.
14-5
±
7-0
ng.
ml"1
at 24 h,
P<005). However, there
were
no
differences
in the
isoform fractions
on
admis-
sion
or at 6, 24 or 48 h
between
the two
groups.
Urinary catecholamine excretion
In
30
patients, four 12
h
urine samples were collected
for estimation
of
urinary adrenaline
and
noradrenaline,
two during
and two
after
the
infusion
(Fig. 3).
There
was
a
reduction
in
adrenaline excretion
in the mag-
nesium group
for the
first (0-12
h:
105
±016 vs
161
±
0-32
mg
. mmol~
'
creatinine respectively)
and
fourth (36-48
h: 069 ±
0-14
vs
1-38
± 018 mg
. mmol"
'
creatinine respectively) samples (/><005); this increase
in urinary adrenaline
in the
36-48
h
time period
in
the placebo group
was not
limited
to
those patients
who
had
ischaemic
ST
segment changes
on
Holter
monitoring. Noradrenaline excretion
was
lower
in
the first (0-12
h)
sample
in the
magnesium group
(9-95
±
1 -82
vs
18-48
±
2-41
mg
.
mmol"
'
creatinine
respectively, P<0-05),
but
similar
in the
other three
collections.
Adverse events
The occurrence
of
adverse events within
one
month
are
shown
in
Table
5.
There
was no
significant difference
in
the occurrence
of
myocardial infarction, revasculariz-
ation
or
death between
the two
groups.
Discussion
This
is the
first randomized, double-blind, placebo-
controlled study investigating
the
effects
of
magnesium
in unstable angina.
In
this study,
we
have observed that
magnesium reduces creatine kinase-MB release,
ST seg-
ment changes
and
catecholamine excretion
and
thus
2
1.8
1.6
CD
|l.4
|
12
1*0.6
6
0.4
0.2
0
30
25
CD
C
•|
20
"8
2
7°
15
"o
fi
10
bi
e5
0
(a) *
T
:
—
-
-
1
1
1
1.
1
BBi 1 1
0-12
(b)
-*
T
•L
•
•
•
1
•
i
0-12
T
T
1
I
1
|
, _ ,
1
•.
••
LJ ""i i
12-24
24-36
Hours
T
1
•
T
T|
r-U-r
r-m
nT
•
•
•
1
m
i
12-24
Hours
•
•
•
•
m
i
24-36
*
T
I
1
|
—MM—
36-48
T
X
\T
L
•
•
1
•
36-48
Figure
3
Urinary adrenaline excretion
(a) and
urinary
noradrenaline excretion
(b) in the
magnesium
(D) and
placebo
(•)
groups. Four
12 h
collections were obtained
in
15
patients
in
each group. Data corrected
for
urinary
creatinine
and
expressed
as
mg/mmol creatinine.
*/><0>05
vs
magnesium.
may have
a
beneficial role
in
addition
to
standard
therapy
in the
acute phase
of
unstable angina.
The prognostic implications
of
transient
myo-
cardial ischaemia
in
unstable angina
are
well estab-
lished'"121.
In
this study, magnesium resulted
in
more
resolution
of T
wave changes with
a
trend towards
a
reduction
in ST
segment changes
on
resting 12-lead
ECGs.
On
continuous
ECG
monitoring,
a
similar
pro-
portion
of
patients
in
each treatment group
had
evidence
of myocardial ischaemia
in the
first 24
h of
recording.
However,
the
number
of
episodes
was
significantly less
in the magnesium group
and the
number
of
patients with
transient myocardial ischaemia
in
this group fell from
11 patients
(39%) to
five
(18%) in the
second 24
h
of recording with
no
change
in the
placebo group
(10 patients
[39%] in
both
the
first
and
second 24 h).
Furthermore, there
was a
non-significant reduction
in
the total duration
of
ischaemia during
the
second 24
h
recording period
in the
magnesium group, with
an
increase
in the
placebo group.
Eur Heart
J,
Vol. 18, August
1997
by guest on July 21, 2011eurheartj.oxfordjournals.orgDownloaded from
Magnesium sulphate in unstable angina 1275
Table 5 Adverse events that occurred in the two patient groups during 1 month
follow-up
Q wave MI
Revascularization
Death in hospital
Death within 1 month
PTCA
CABG
Magnesium
1 (3%)
15(48%)
5 (16%)
10 (32%)
4(13%)
5 (16%)
Placebo
2 (7%)
16 (52%)*
9 (29%)
9 (29%)
3 (10%)
3 (10%)
Two patients required CABG for unsatisfactory results post PTCA.
MI
=
myocardial infarction; PTCA
=
percutaneous transluminal angioplasty: CABG
=
coronary
artery bypass grafting.
Creatine kinase-MB and its isoforms have been
shown to be sensitive markers of myocardial cell in-
jury'281 and are predictors of an adverse prognosis in
patients with unstable angina'30'. In this study, we
showed a significant reduction in creatine kinase-MB
release in the magnesium group at 6 and 24 h; however,
no change in the isoforms of creatine kinase-MB. In
ischaemic heart disease, studies of creatine kinase-MM
isoforms demonstrate a transient increase in the relative
proportions of the tissue isoform MM3 (in comparison
with the plasma-modified isoform MM1) several hours
before the total creatine kinase or creatine kinase-MB
activities rise'31'. In our study, patients were randomized
at a mean of 111 h after the onset of chest pain. It seems
likely, therefore, that if there was an increase in these
isoforms indicating myocardial damage, that it would
have occurred early and may have been missed with
blood sampling at 0, 6, 24 and 48 h.
Intracellular calcium overload is thought to be a
central mechanism in ischaemic myocardial death'15'.
Magnesium has been shown to inhibit cellular calcium
influx'13', reduce mitochondrial calcium overload'14' and
conserve intracellular ATP as Mg2+-ATP'32', and raising
extracellular magnesium has been shown to be protec-
tive during ischaemia'33'. The use of coronary vasodila-
tors in patients with unstable angina has been shown to
be beneficial'34'. Magnesium inhibits the contractility of
coronary arteries'35', increases coronary blood flow and
decreases coronary vascular resistance without increas-
ing myocardial oxygen consumption1'7'. In patients with
variant angina it has been shown to suppress coronary
spasm'36'.
The potentially beneficial effect of magnesium on
myocardial ischaemia could be mediated either by a
reduction in coronary spasm (directly or indirectly via
a reduction in catecholamines), a reduction in platelet
aggregability or a direct cellular effect of magnesium.
Increased catecholamine levels are known to
have an important role in the pathogenesis and prog-
nosis of unstable angina and are thought to play an
important part in extension of an infarct'37'. In our
study, magnesium was shown to reduce adrenaline ex-
cretion in the first (0-12 h) and fourth (36-48 h) samples
and to reduce noradrenaline excretion in the first (0-
12 h) sample; a similar proportion of patients in each
group were treated with beta-blockers. Magnesium de-
ficiency has been associated with enhanced sympathetic
activity'2223' and an increased pressor and arrhythmo-
genic response to adrenaline'38'. Magnesium has been
shown to inhibit catecholamine release from the adrenal
medulla and to reduce the sensitivity of a-adrenergic
receptors to catecholamines'39'. In addition, it has been
shown to reduce both the likelihood of exogenous
epinephrine to precipitate arrhythmias'40' and the size of
catecholamine induced myocardial damage'41'.
Tracheal intubation is associated with release
of catecholamines in large amounts and James et
al.142^
used this as a model to show that infusions of mag-
nesium significantly reduce both adrenaline and nor-
adrenaline secretion. In this study, we have shown that
magnesium may reduce neuroendocrine activation in
patients with unstable angina and this may have
favourable prognostic implications.
Study limitations
In this study, unstable angina was defined as chest pain
with ECG changes; this selects a subgroup with a worse
prognosis, and the results of this study may not be
applicable to other groups of patients. However, it was
carried out in a population typical of a district general
hospital on unselected patients and included 73% of
those admitted with unstable angina and ECG changes
during the study period. Patients who subsequently were
found to have elevated cardiac enzymes were not ex-
cluded as these patients would be initially treated as
unstable angina and would not be candidates for throm-
bolytic therapy based on ECG criteria. This study was
designed to detect a reduction in myocardial ischaemia
in patients with unstable angina, based on an expected
incidence of 65%, but its power was not adequate to
detect an effect on mortality.
Conclusion
Magnesium has been shown to be safe and well tolerated
and to have potentially beneficial effects, with a re-
duction in transient myocardial ischaemia, creatine
Eur Heart J, Vol. 18, August 1997
by guest on July 21, 2011eurheartj.oxfordjournals.orgDownloaded from
1276 S. R. Redwood et al.
kinase-MB release and catecholamine excretion. The
implications of these findings may be clinically import-
ant and deserve further investigation.
During the conduct of this study, Dr Redwood was a British
Heart Foundation Junior Research Fellow. This study was funded
by The British Heart Foundation and South West Thames
Regional Health Authority. We are grateful to the staff of the
Coronary Care Unit and Pharmacy at St George's Hospital. In
addition, we thank the Analytical Unit of our department for
performing the magnesium, creatine kinase-MB and catecholamine
assays.
References
[1] Woods KL. Possible pharmacological actions of magnesium
in acute myocardial infarction. Br J Clin Pharm 1991; 32:
3-10.
[2] Woods K, Fletcher S, Roffe C, Haider Y. Intravenous mag-
nesium sulphate in suspected acute myocardial infarction:
results of the second Leicester intravenous magnesium inter-
vention trial. (LIMIT-2). Lancet 1992; 339: 1553-8.
[3] Teo KK, Yusuf S, Collins R, Held PH, Peto R. Effects of
intravenous magnesium in suspected acute myocardial infarc-
tion: overview of randomised trials. BMJ
1991;
303: 1499-503.
[4] Homer S. Efficacy of intravenous magnesium in acute myo-
cardial infarction in reducing arrhythmias and mortality.
Circulation 1992; 86: 774-9.
[5] ISIS-4 Collaborative Group. ISIS-4. a randomised factorial
trial assessing early oral captopril, oral mononitrate, and
intravenous magnesium sulphate in 58,050 patients with sus-
pected acute myocardial infarction. Lancet 1995; 345: 669-85.
[6] du Toit EF, Opie L. Modulation of severity of reperfusion
stunning in the isolated rat heart by agents altering calcium
flux at onset of reperfusion. Circ Res 1992; 70: 960-7.
[7] Epstein SE, Palmeri ST. Mechanisms contributing to precipi-
tation of unstable angina and acute myocardial infarction:
Implications regarding therapy. Am J Cardiol 1984; 54. 1245—
52.
[8] Fuster V, Badimon L, Cohen M, Ambrose JA, Badimon JJ,
Chesebro JH. Insights into the pathogenesis of acute ischemic
syndromes. Circulation 1988; 77: 1213-20.
[9] de Servi S, Ghio S, Ferrario
M
et
al.
Clinical and angiographic
findings in angina at rest. Am Heart J 1986; 111: 6-11.
[10] Sclarovsky S, Davidson E, Strasberg B, Lewin RF, Arditti A,
Agmon J. Unstable angina pectoris evolving to acute myocar-
dial infarction: significance of ECG changes during chest pain.
Am Heart J 1986; 112: 459-62.
[11] Gottlieb S, Weisfeldt M, Ouyang P, Mellits D, Gerstenblith
G. Silent ischemia as a marker for early unfavorable outcomes
in patients with unstable angina. N Engl J Med 1986; 314:
1214-9.
[12] Nademanee K. Intarachot V, Josephson AM, Rieders D,
Vaghaiwalla F, Singh B. Prognostic significance of silent
myocardial ischemia in patients with unstable angina. J Am
Coll Cardiol 1987; 10: 1-9.
[13] White RE, Hartzell HC. Effects of intracellular free mag-
nesium on calcium current in isolated cardiac myocytes.
Science 1988; 239: 778-80.
[14] Ferrari R, Albertini A, Curello S et al. Myocardial recovery
during post-ischaemic reperfusion: effects of nifedipine, cal-
cium and magnesium. J Moll Cell Cardiol 1986; 18: 487-98.
[15] Tani M, Neely J. Role of intracellular Na+ in Ca2+ overload
and depressed recovery of ventricular function of reperfused
ischemic rat hearts. Circ Res 1989; 65: 1045-56.
[16] Kimura T, Yasue H. Sakaino N. Rokutanda M, Jougasaki M,
Araki H. Effects of magnesium on the tone of isolated human
coronary arteries. Circulation 1989: 79: 1118-24.
[17] Vigorito C. Giordano A. Ferraro P et al. Hemodynamic
effects of magnesium sulfate on the normal human heart. Am
J Cardiol 1991; 67: 1435-7.
[18] Rasmussen H, Larsen O, Meier K, Larsen J. Hemodynamic
effects of intravenously administered magnesium on patients
with ischemic heart disease. Clin Cardiol 1988; 11. 824-8.
[19] Baudouin-Legros M, Dard B, Guicheney P. Hyperreactivity
of platelets from spontaneously hypertensive rats. Role of
external magnesium. Hypertension 1986; 8: 694-9.
[20] Adams JH, Mitchell JRA. The effects of agents which modify
platelet behaviour and of magnesium ions on thrombus for-
mation in vivo. Thrombos Haemostas 1979; 42: 603-10.
[21] Nadler J, Goodson S, Rude R. Evidence that prostacyclin
mediates the vascular action of magnesium in humans. Hyper-
tension 1987; 9: 379-83.
[22] Ising H. Interaction of noise-induced stress and Mg decrease.
Artery 1981; 9:
205-11.
[23] Tackett RL. Enhanced sympathetic activity as a mechanism
for cardiac glycoside toxicity in hypomagnesemia. Pharma-
cology 1986; 32: 141-6.
[24] James M, Cork R, Harlen G, White J. Interactions of adrena-
line and magnesium on the cardiovascular system of the
baboon. Magnesium 1988; 7: 37^3.
[25] Vormann J, Fischer G, Classen H, Thoni H. Influence of
decreased and increased magnesium supply on the cardiotoxic
effects of epinephrine in rats. Drug Res 1983; 33: 205-10.
[26] Rasmussen H, Gronbaek M, Cintin C, Balslov S, Norregard
P,
McNair P. One-year death rate in 270 patients with
suspected acute myocardial infarction, initially treated with
intravenous magnesium or placebo. Clin Cardiol 1988; 11:
377-81.
[27] Quyyumi A, Mockus L, Wright C, Fox K. Morphology of
ambulatory ST segment changes in patients with varying
severity of coronary artery disease. Br Heart J 1985; 53:
186-93.
[28] Hossein-Nia M, Kallis P, Brown PA et
al.
Creatine kinase MB
isoforms: sensitive markers of ischemic myocardial damage.
Clin Chem 1994; 40:
1265-71.
[29] Gottlieb SO, Weisfeldt M, Ouyang P et al. Effect of the
addition of propranolol to therapy with nifedipine for un-
stable angina pectoris: A randomized, double-blind, placebo-
controlled trial. Circulation 1986; 73: 331-7.
[30] Botker HE, Ravkilde J, Sogaard P, Jorgensen PJ, Horder M,
Thygesen K. Gradation of unstable angina based on a sensi-
tive immunoassay for serum creatine kinase MB. Br Heart J
1991;
65: 72-6.
[31] Puleo P, Guadagno P, Roberts R et al. Early diagnosis of
acute myocardial infarction based on assay for subforms of
creatine kinase-MB. Circulation 1990; 82: 759-64.
[32] Hearse D. Reperfusion of the ischemic myocardium. J Mol
Cell Cardiol 1977; 9: 605-16.
[33] Bersohn M, Shine K, Sterman W. Effect of increased mag-
nesium on recovery from ischemia in rat and rabbit hearts.
Am Physiol Soc 1982; 242: H89-93.
[34] Hugenholtz PG, Michels HR, Serruys PW, Brower RW.
Nifedipine in the treatment of unstable angina coronary
spasm and myocardial ischemia. Am J Cardiol 1981; 47:
163-73.
[35] Altura BM, Altura BT, Carella A, Gebrewold A, Murakawa
T, Nishio A. Mg2+ - Ca2+ interaction in contractility of
vascular smooth muscle: Mg2+ versus organic calcium channel
blockers on myogenic tone and agonist-induced responsive-
ness of blood vessels. Can J Physiol Pharmacol 1987; 65:
729-45.
[36] Kugiyama K, Yasue H, Okumura K et al. Suppression of
exercise-induced angina by magnesium sulfate in patients with
variant angina. J Am Coll Cardiol 1988; 12: 1177-83.
[37] Penny WJ. The deleterious effects of myocardial catechol-
amines on cellular electrophysiology and arrhythmias during
ischaemia and reperfusion. Eur Heart J 1984; 5: 960-73.
[38] Bean B, Varghese P. Role of dietary magnesium deficiency in
the pressor and arrhythmogenic response to epinephrine in the
intact dog. Am Heart J 1994; 127: 96-102.
[39] Von Euler U, Lishajko F. Effects of Mg2* and Ca2+ on
noradrenaline release and uptake in adrenergic nerve granules
in different media. Acta Med Scand 1973; 89: 415-22.
Eur Heart J, Vol. 18, August 1997
by guest on July 21, 2011eurheartj.oxfordjournals.orgDownloaded from
Magnesium sulphate in unstable angina 1277
[40] Mayer D, Miletich D, Feld J. Albrecht R. The effects of structural and cytochemical study. Cardioscience 1992; 3:
magnesium salts on the duration of epinephrine induced 197-203.
ventricular tachyarrhythmias in anesthetized rats. Anesthesi- [42] James MFM, Beer RE, Esser JD. Intravenous magnesium
ology 1989;
71:
923-8. sulfate inhibits catecholamine release associated with tracheal
[41] Herbaczynska Cedro K, Gajkowska B. Effect of magnesium intubation. Anesth Analg 1989; 68: 772-6.
on myocardial damage induced by epinephrine. Ultra-
Eur Heart J, Vol. 18, August 1997
by guest on July 21, 2011eurheartj.oxfordjournals.orgDownloaded from
