International Journal of Cardiology, 36 (1992) 151-162 0 1992 Elsevier Science Publishers B.V. All rights reserved 0167-5273/92/$05.00
151
CARD10 01490
Dynamic changes of the QRS complex in unstable angina pectoris Mikael Dellborg,
Gunnar
Gustafsson,
Martin
Riha and Karl Swedberg
Department of Medicine, Unirlersity of Giiteborg, &tra Hospital 416 85 GGteborg, Sweden
(Received 25 November 1991; revision accepted 26 February 1992)
Dellborg M, Gustafsson G, Riha M, Swedberg K. Dynamic changes of the QRS complex in unstable angina pectoris. Int J Cardiol 1992;36:151-162. Despite intensive medical treatment to control chest pain, about one-third of patients with unstable angina have an unfavourable outcome within a period of 1 to 2 months. Holter monitoring can identify patients with silent myocardial ischaemia that are at a high risk of sustaining a major cardiovascular event. The present paper describes the use of dynamic, continuous, computerized on-line vectorcardiography for real-time monitoring of QRS-complex and ST-segment changes in patients with unstable coronary disease. In many patients a pattern of frequent repetitive episodes of QRS change was observed, with or without concomitant ST change. Whereas no patient had episodes of ST-vector change without also having episodes of significant QRS change, 15 patients had several episodes of QRS changes without any episode of significant ST change. The number of episodes of significant increase of the QRS vector difference correlated weakly but significantly with the number of episodes of significant ST-vector magnitude change (r = 0.34, p < 0.05). The present study suggests that myocardial ischaemia will influence the QRS complex as well as the ST segment. The mechanism behind the QRS changes observed is not clear but episodes of QRS change without ST change or chest pain, may reflect sudden depressions of left ventricular function, as has been reported by others to occur in patients with coronary artery disease. Dynamic vectorcardiography offers the opportunity to monitor all parts of the QRST complex in real time. Key words: Unstable angina; ECG monitoring; Vector cardiography
Correspondence
to: M. Dellborg M.D., Dept. of Medicine,
&tra Hospital, 416 85 Gteborg, Sweden. Tel. 46-31-374000. Fax 46-31-258933. Supported by a grant from AFA, the Labour Market Insurance Company, Stockholm, Sweden, TryggHansa Insurante Company, Stockholm, Sweden, The Heart-Lung Foundation, Stockholm, Sweden and AB Hassle Cardiovascular Research Laboratories, Molndal, Sweden.
Introduction Unstable angina is a condition caused by myocardial ischaemia, characterized by symptoms of recent onset, a crescendo pattern or the occurrence of symptoms at rest. Despite intensive medical treatment to control chest pain, about one-
152
third of patients with unstable angina have an unfavourable outcome within a period of 1 to 2 months [l-3]. Continuous 2-channel Holter monitoring can identify patients with silent myocardial ischaemia that are at a high risk of sustaining a major cardiovascular event and thus have an impaired short-term prognosis [4,51. We have previously described the use of dynamic, continuous, computerized on-line vectorcardiography to monitor patients with acute myocardial infarction as well as during coronary angioplasty [6-81. This system allows on-line monitoring of QRS-complex and ST-segment changes with analysis and presentation in real time, potentially identifying high-risk patients and allowing monitoring of acute interventions in unstable coronary disease. The present report describes the system and summarizes our experience with real-time, dynamic vectorcardiography in patients with unstable angina. Methods Normal subjects Twenty healthy volunteers, 10 men and 10 women, aged under 30, with normal standard 1Zlead eIectrocardiogram were monitored from 6 p.m. to 6 a.m. while resting in bed. They were all without on-going medication and experienced no symptoms during the 12-h monitoring period. Patient series Patients with unstable angina pectoris as defined below, included in a randomized trial of intravenous vs buccal nitroglycerin (n = 28) or included in a comparative multi-centre study of vectorcardiography vs Holter-monitoring (n = 151, were monitored with continuous dynamic vectorcardiography for 24 h after admission to the coronary care unit. Patients were treated with beta-blockers, aspirin and if necessary calcium antagonists in addition to high-dose nitrates. Changes in the resting 12-lead scalar electrocardiogram were not necessary for inclusion. All patients were closely followed throughout the hospital stay and were, if not stabilized medically,
subject to coronary angiography and if required, revascularization was performed. Continuous vectorcardiography The monitoring system (MIDA, Myocardial Infarction Diagnosis and Analysis, Ortivus Medical AB, Taby, Sweden) consists of a microprocessor controlled data-acquisition module, an IBM compatible personal computer and a graphic printer. Electrocardiographic signals are continuously collected from conventional body surface electrodes applied to the patient’s chest according to the Frank lead system [9]. The sensitivity of the system is 1 PV and the sampling rate of each lead is 500 samples/s. Electrocardiographic complexes are detected, collected and placed according to their shape into 1 of 5 classes. The most dominant beat type is automatically determined during the first 10 s of recording and termed the zero class. The 3 orthogonal vectorcardiographic leads X, Y and Z are computed from sampled unipolar Frank leads and continuously displayed on the colour monitor and averaged to form mean vectorcardiographic complexes. Averaging is performed for consecutive periods, each period being 10 s to 4 min long. For this study, averaging periods of 1 min were used. Only beats classified as class zero were used for subsequent analysis. For each period the resulting mean vectorcardiographic complex was analyzed and compared a reference complex collected during the second averaging period. Any of the trend parameters may be presented as a trend curve updated continuously during the whole recording period. The screen of a colour monitor attached to the personal computer may be divided in several windows with different information, such as running vectorcardiographic signal, any mean vectorcardiographic complex collected or trend curves, presented in each window. The data collection and calculations are all done on-line. An alarm may be set to identify a change in any of the trend parameters to alert the attending staff to significant changes in the QRS complex or the ST segment. The computer will automatically store all mean complexes and trend curves on the hard disk and keep them available for on-line review-
153
ing at any time during the recording. The recording may then also be reviewed and summarized in off-line mode. The data may be transferred to a floppy disk and stored. Coronary care unit nurses were trained to connect the patient, start monitoring and recording and to supervise signal quality. Initial training took approximately 3 h. Monitored vectorcardiographic
variables
Based on previous results and observations [6-81 3 trend parameters were closely studied: the QRS-vector difference, which summarizes all changes within the QRS complex, the ST-vector magnitude, which is the deflection of the ST segment from the isoelectric level 20 ms after the J point and the ST change vector magnitude which is the length of the difference vector between the initial ST vector and the current ST vector. These parameters are described in Fig. 1. In Fig. 2 trend curves from 3 patients are shown. Fig. 2a shows the simultaneous changes occurring with and without chest pain in ST-vector magnitude, ST-change vector magnitude and QRS-vector difference in a patient with unstable angina. Fig. 2b shows a patient who develops an acute myocardial infarction illustrated by the continuous and persistent increase in QRS vector difference. This patient was not part of the present study but is shown for comparative reasons only. A patient with episodes of significant QRSvector difference change but without significant increase in either the ST-vector magnitude or the ST-change vector magnitude is illustrated in Fig. 2c. Monitoring
of chest pain
Pain was evaluated hourly during vectorcardiographic monitoring by the attending nurse, asking the patient to recall the time, number and severity of episodes of chest pain. Definitions
chest pain occurring at rest or at a very low level of exercise. The pain should give strong suspicion of ischaemic heart disease but not myocardial infarction and last for at least 5 but not more than 30 min and should have been present at least twice during the last 24 h. Significant vectorcardiographic change. Based on findings from previous studies [7,8,10] we postulated that a significant change in the QRS-vector difference indicating ischaemia would consist of a reversible increase of more than 15 ~VS, lasting > 1 min and with > 1 min between episodes. Such episodes will be referred to as “QRS-episodes” in the text. For ST-vector magnitude and ST-change vector magnitude an increase to above 0.1 mV (corresponding to an ST-segment change of > 1 mm in either 1 of the 3 orthogonal leads) was considered to indicate ischaemia. These changes will be referred to as “ST episodes” or “ST-change episodes”. Vectorcardiographic changes are expressed as number of episodes and/or total duration per patient per 24-h period. Intervention. Coronary angiography, percutaneous transluminal coronary angioplasty or coronary artery bypass grafting during the hospitai stay due to failure to achieve medical stabiiization of angina were considered interventions. Myocardial infarction. Myocardial infarction was defined with the usual criteria, i.e. 2 out of 3 of: typical chest pain, elevation of cardiac enzymes, electrocardiographic changes. Vectorcardiographic changes were not used for diagnosis. There were 4 patients in whom a definite acute myocardial infarction was documented by release of cardiac enzymes during the initial 24 h of monitoring. These patients were considered to have had an ongoing myocardial infarct already when included; they were therefore not included in the analysis. One of these patients is illustrated in Fig. 2b for comparative reasons. Statistics
Unstable angina. Recent onset angina or a sudden change in a previously stable pattern, with
All values are expressed as mean + SEM. Groups were compared using Mann-Whitney U-
154 ORS
- Vector
Difference
statistics. Distribution of observations was compared with chi-square analysis with continuity correction or when appropriate with 4-fold table test. A p value of < 0.05 was considered significant. Results Normal subjects Ax
Ay. AZ i
QRS-VD =
ST - Vecior
Ax2
Magnitude
in,
ih_ir\, Current
complex
-
ST-VM=\/Xf+
STC
t Ay2 + AZ’
- Vector
Y:
+
The healthy control group reported no episodes of chest pain during the monitoring period. Compared to patients with unstable angina the control group had fewer ST and QRS episodes (see Table 1). Most notably, no normal subject had any ST-change episode. The effect of changing body position is shown in Fig. 3.
2:
Magnitude
Fig. 1. The upper panel shows a trend curve with time on the horizontal axis and QRS-vector difference and ST-vector magnitude on the vertical axis. An example of an episode of reversible change of both parameters is shown. The reference complex is the second average recorded for an individual patient. The ischaemic complex is taken from the point of maximum change on the trend curve. The computer will in real time calculate the area of difference between the reference complex and any subsequent complexes. The figure shows the comparison between the reference complex and the ischaemic complex. The QRS-vector difference is the hatched area of difference between the 2 average complexes. The ST-vector magnitude is the deviation of the ST segment from the baseline, measured at 20 ms after the J point. The ST-deviation is quadratically added for all three leads, producing the total ST-vector magnitude. The ST-change (STC) vector magnitude is the length of the vector going from the tip of the ST vector of the reference complex to the tip of the ST vector of the ischaemic complex. The ST-change vector magnitude is thus the ST-vector magnitude corrected for the degree of ST deviation observed on the reference complex. However, the ST-change vector magnitude also takes into consideration the angular changes of the ST vector. STC = ST change; ST-VM: ST-vector magnitude; QRS-VD = QRSvector difference.
Unstable angina group There were 43 patients with unstable angina pectoris monitored for 24 h. Baseline characteristics of these patients are shown in Table 2. Clinical data regarding chest pain, myocardial infarction, death or the need for acute intervention are given in Table 3 for patients with unstable angina. QRS episodes were seen in 88% of the patients whereas ST or ST-change episodes were seen in 52 and 43%, respectively (Table 1). Correlation between QRS and ST episodes. In Fig. 4, for every patient the number of ST or ST-change episodes is related to the number of QRS episodes. The number of QRS episodes correlated weakly but significantly with the number of ST episodes (r = 0.34, p < 0.05). Only 1 patient had ST or ST-change episodes without also having QRS episodes. However, 15 patients had several QRS episodes without any ST or ST-change episodes. Of these 15 patients, 12 had chest pain at rest associated with QRS episodes. Medical therapy failed to stabilize 1 of these patients and acute coronary angiography revealed 2-vessel disease with a highly significant stenosis of the left anterior descending artery (patient recording shown in Fig. 2~).
155
Silent myocardial ischaemia. The occurrence of ST-change episodes without concomitant chest pain was analyzed as an indication of silent my-
ocardial ischaemia. The unstable angina group was divided according to the presence (n = 17) or absence (n = 25) of asymptomatic ST-change
(b)
Fig. 2. a. Part of the trend curve from a patient with unstable angina. The QRS-vector difference, the ST-vector magnitude and the ST-change vector magnitude are shown. Simultaneous changes occur in all 3 parameters, in some cases associated with chest pain (arrow). b. The same parameters recorded from a patient with unstable angina who develops an acute myocardial infarction. Note the sustained increase of the QRS-vector difference. This patient was not part of the present report but is included for comparative reasons. c. Patient with repeated QRS episodes. The patient has no ST or ST-change episodes. Angiography during the hospital stay revealed a high-grade stenosis of the left anterior descending artery; an uncomplicated angioplasty was performed. Arrows indicate chest pain. ~VS = microvoltseconds, unit for QRS-vector difference; mV = millivolt, unit for ST and ST-change vector magnitude; STC-VM = ST-change vector magnitude; ST-VM = ST-vector magnitude; QRS-VD = QRS-vector difference.
156
TABLE
1
Comparison
of vectorcardiographic
parameters Normal
between
normal
subjects
NS = not significant.
ST change:
and patients Unstable
No. Patients ST episodes no. of episodes/patient duration/patient (mitt) patients with episodes ST-change episodes no. of episodes/patient duration/patient (min) patients with episodes QRS episodes no. of episodes/patient duration/patient (mitt) patients with episodes Episodes of chest pain
subjects
9%
angina. P
angina
No.
20
%
43
4*2 144 * 78 5
5*1 140 f 56 23
25
0 0 0
2+1 27* 15 19
6+2 118+38 14 0
11+1 180 f 24 38 3*1
ST-change
with unstable
70
vector
magnitude.
episodes (see Table 4). Overall 62 of 78 ST-change episodes (79%) were asymptomatic. Only 1 patient had symptomatic ST-change episodes without having asymptomatic ST-change episodes. Znteruention. Patients with unstable angina were divided according to the need for interven-
Values
52
NS NS < 0.05
43
< 0.001 < 0.001 < 0.001 < 0.05 < 0.05 NS < 0.001
88
are given as mean f SEM.
tion. In the 12 patients that failed medical apy, ST and ST-change episodes occurred frequently and were of longer duration as pared to patients who responded to medical apy and thus were not subject to intervention Table 5).
thermore comther(see
Myocardial infarction. Myocardial infarction after the initial 24-h monitoring but before hospiTABLE
2
Baseline
characteristics
of 43 patients
Age, yr (range) Previous infarction Men/women Previous CABG/PTCA Medication on admission beta-blockers aspirin long-acting nitrates calcium antagonists Electrocardiogram on admission ST elevation ST depression T abnormalities normal CABG = coronary neous transluminal
with unstable
No.
%
59+F3 16 29/14 0
(39-75) 37 67/33
32 17 12 7
74 40 28 16
3 13 14 13
7 30 33 30
artery bypass grafting; coronary angioplasty.
angina.
PTCA = percuta-
TABLE
3
Clinical
data,
patients
with unstable
angina.
No. During initial 24 h Patients with chest pain at rest l-3 episodes > 3 episodes Patients without chest pain at rest Death Total From 24 h to hospital discharge Coronary angiography/ CABG/PTCA Myocardial infarction Death CABG = coronary neous transluminal
%
29
67 12 17
14 0 43
12 2 0
artery bypass grafting; coronary angioplasty.
28 39 33 100
28 7
PTCA = percuta-
157
tal discharge developed (5%).
in 2 of the 43 patients
Discussion The present study investigated the use of computerized on-line vectorcardiographic monitoring of patients with unstable angina. The definition of unstable angina is often difficult and therefore it is difficult to compare different studies. In the present study 28% needed acute intervention as they could not be stabilized medically. Observations from these patients were compared to observations made in a group of normal subjects.
ing, we cannot completely rule out that such information influenced the attending physician in the making of this decision. Therefore, our identification of high-risk patients by monitoring the ST-change vector magnitude must be cautiously interpreted and subjected to further study. QRS-vector changes in myocardial ischaemia Although QRS episodes were frequently seen among normal subjects they were observed in a
(I.3 1 STC-VM
ST-vector changes in myocardial ischaemia 0.1s
The ST-vector magnitude with the presently used limit for defining myocardial ischaemia (STVM > 0.1 mV) seems to have a low specificity since normal subjects had a similar number and duration of ST episodes as the unstable patients. The healthy volunteers were all young and many had a high baseline ST-vector magnitude at rest due to vagotonic ST-segment elevation. A possible solution to this problem would be to measure the ST-vector magnitude as a percentage of change from baseline in the individual patient. The ST-change vector magnitude is a measurement of change in relation to the initial ST vector in an individual patient. In the present study it was superior to the ST-vector magnitude and the QRS-vector difference in separating normal subjects from patients with coronary disease. The ST-change vector magnitude also identified a group of patients with electrocardiographic changes without symptoms, tentatively defined as having silent myocardial ischaemia. With conventional ST monitoring, patients with unstable angina and frequent asymptomatic ST changes are at high risk of sustaining a coronary event within the following 1 to 2 months after hospitalization for unstable angina 14,111. In the present study, patients that could not be stabilized medically had a significantly higher number and a longer total duration of ST-change episodes. Although the decision to intervene was not based on the results of the vectorcardiographic monitor-
0.5
ST-VM I
QRS-VD
SUPiTlC
left side
supine
qht
sldc
sup,nc
sIllins
sup,nc
Fig. 3. Mean f standard deviation of the QRS-vector difference, the ST-vector magnitude and the ST-change-vector magnitude of 20 healthy volunteers. Body position was changed every 5 min in a non-random way as indicated in the figure. For every position and person, the maximum value was used. STC-VM = ST-change vector magnitude; ST-VM = STvector magnitude: QRS-VD = QRS-vector difference.
158
significantly higher frequency and were of longer duration in patients with unstable angina. As shown in Figs. 2c and 4, many QRS episodes occurred without ST or ST-change episodes. Actually, in many patients a pattern of frequent repetitive QRS episodes with or without ST or ST-change episodes was detected. This pattern was not seen among controls. In previous reports we have found that myocardial infarction will result in a persistent increase of the QRS-vector difference after 24 h of recording [6,8]. The QRS-vector difference thus seems to be reversibly influenced by myocardial ischaemia but the development of necrosis will cause sustained changes.
Mechanism myocardial
behind the QRS changes ischaemia
during
The precise mechanism behind QRS changes during myocardial ischaemia is not known. Several mechanisms have been proposed, including the Brodie effect 1121. In normal subjects positional changes will cause relatively large QRS changes (see Fig. 3). The distance from the electrode to the left ventricular wall is of critical importance. This distance has been shown to decrease significantly when the subject is in the left supine position [13]. Simultaneous changes in the QRS complex occur and the QRS-vector loop will then increase its area, “dilate”. With the
ST episodes 35 30
I 0
25
ORS
rpisodes
STC episodes 20 18 16 14 12 10
a 6 4 2 0
0
5
10
15
20
25
30
35 ORS
Fig. 4. Number
of ST (a) or ST-change episodes Overlaps are indicated
episodes
(b) per 24 h is plotted against the number of QRS episodes by increased size of the symbols used. STC = ST change.
for individual
patients.
present technique, this will appear as a simuItaneous increase in the QRS-vector difference, i.e. a QRS episode. We have previously reported that changes in the QRS loops and of the QRS-vector difference occur during myocardial ischaemia in all patients during percutaneous coronary angioplasty [71. Reversible changes of the QRS complex occur when a coronary artery is temporarily occluded by the angioplasty balloon. This ischaemic reaction of the QRS complex is believed to be caused by delayed activation of the ischaemic myocardium. Since non-ischaemic myocardium will be depolarized earlier, the delayed activation of the ischaemic part of the myocardium will be unopposed electrically and produce a larger vector loop [14]. Myocardial ischaemia will also induce increased end-systolic volume [15,16], which will shorten the distance from the myocardium to the electrode. Tamaki and co-workers monitored ejection fraction by an ambulatory radionucleide recorder in patients with coronary artery disease [17]. They reported a
TABLE
high incidence of simultaneously occurring chest pain, ST-segment depression and a decrease in ejection fraction. Among 39 patients they also observed 20 episodes in 14 patients of transiently depressed ejection fraction without concomitant ST-segment changes. They suggested that ischaemia may not always produce neither pain nor ST depression but its sole manifestation may be depressed left ventricular function. We speculate that the QRS episodes that we have observed in patients with unstable angina may reflect sudden depressions of left ventricular function, with or without ST-segment changes and chest pain. Similar episodes of changes in left ventricular function in the absence of ECG changes or chest pain have also been reported following coronary angioplasty [ 181. The mechanism behind a change in the QRSvector difference is different between normal subjects and patients with myocardial ischaemia. The appearance of the vector loop is, however, similar in both instances. Therefore, the present tech-
4
Vectorcardiographic
and clinical
data for patients
with and without
With asymptomatic episodes No. Patients Chest pain at rest,
17
no. of episodes/patient patients with episodes ST episodes
3+ 13
no. of episodes/patient duration/patient (min) patients with episodes ST-change episodes
asymptomatic
ST-change
episodes
No.
ischaemia”). P
No asymptomatic episodes %
(=“silent
%
2.5 1
2 6+ 182 k 124 13
76
4 15
71
5 118 10
+
55 66~ 14 13*
2
10
duration/patient (min) patients with episodes AMI or intervention
202+ 17 9
31
173 21 4
100
100 53
NS NS
40
NS NS = 0.05
f 2 f 53
0.04 t 0.08 k
1 34
no. of episodes/patient duration/patient (min) patients with episodes QRS episodes no. of episodes/patient
60
1
0.04 0.08
1
4 f? * 35 84 16
< 0.00 1 < 0.001 < 0.001 NS NS NS = 0.05
AMI = acute myocardial infarction; intervention = coronary angiography, angioplasty or coronary artery bypass surgery during hospital stay. NS = not significant. All values are given as mean + SEM. One patient had no information about chest pain.
160 TABLE
5
Vectorcardiographic
and clinical
data for patients
with acute intervention
Intervention
Stabilized
No. Patients Mean age (yr) Male/female Chest pain at rest, no. of episodes/patient patients with episodes
2.5 + 11
ST episodes no. of episodes/patient duration/patient (min) patients with episodes ST-change episodes no. of episodes/patient duration/patient (min) patients with episodes ST-change episodes, silent no. of episodes duration/patient (min) patients with episodes QRS episodes no. of episodes/patient duration/patient (min) patients with episodes Intervention = coronary angiography, values are given as mean f SEM.
1
8* 3 334 f 177 10 4.8 f 76 + 8
2 49
4* 65 + 8
1 46
12 191 12
+ +
3.5 * 18
83
4+1 63 + 28 13
67
0.7 f 7+ 11
67
0.5 + 0.1 5 +13 10
angioplasty
100 or coronary
nique cannot differentiate and pathogenetically separate individual episodes of QRS-vector difference change. Because of a low number of patients and controls the statistical power of the present study is limited. Our findings are preliminary and the relative value of the different vectorcardiographic parameters for detection and quantification of myocardial ischaemia are at present not clear. Other monitoring
10 176 26
systems
Holter recording and off-line analysis is today a frequently used technique to monitor myocardial ischaemia. The merit of ischaemia monitoring as a guide to therapy and intervention in unstable angina pectoris has not been proven in controlled clinical trials. However, it is of substantial practical value if the results of ischaemia monitoring in patients with unstable angina can
artery
bypass
medically.
I-J
7
92
2 42
stabilized
%
31 59 f 21/10
9
and for patients
medically
No.
%
12 60 F 8/4
(=“intervention”l
1 58
NS NS
42
< 0.05 < 0.05 = 0.05
35
< 0.001 < 0.05 NS
32
< 0.001 < 0.01 NS
84
NS NS NS
0.2 4
+ 2 + 30
surgery
during
hospital
stay. NS = not significant.
All
be made available to the responsible clinician in real time. The “Monitor One” is a solid-state Holter recorder for on-line analysis and its use in patients with stable angina and during exercise testing has been reported 119,201. It has been pointed out that multi-lead recordings provide improved detection of ischaemic episodes and may also provide important prognostic information when used for evaluation of the total ischaemic burden [21]. At present there are only 2 systems that will allow multiple-lead, on-line monitoring of electrocardiographic changes. The Mortara system, which uses continuous real-time monitoring of the conventional 12-lead electrocardiogram, has been extensively investigated by Krucoff and coworkers in patients with acute myocardial ischaemia [22] and has also been applied to patients with unstable angina [lo]. Von Essen and co-workers [23] described a multiple-lead system
I61
for real-time monitoring of patients with myocardial infarction and unstable angina. Their system would also allow monitoring of QRS changes in addition to ST changes. The relative sensitivity and specificity for detecting myocardial ischaemia for the vectorcardiographic system we have used in relation to other monitoring systems cannot be assessed at present. In conclusion, continuous on-line vectorcardiography is feasible for monitoring of patients with coronary artery disease. Myocardial ischaemia will influence the QRS complex as well as the ST segment and this technique offers the opportunity to monitor all parts of the QRST complex in real time. Vectorcardiography has the advantage over other multiple lead systems to summarize all changes in the QRS complex in one easily monitored parameter, the QRS-vector difference. In addition it offers several ST-vector parameters. The present study suggests that continuous vectorcardiography may be helpful in identifying patients with unstable angina at high risk and separate the patients with acute myocardia1 infarction. Acknowledgements We are grateful to Kerstin Korssell R.N. and Ann-Marie Svensson R.N. for technical assistance and to the nursing staff at the coronary care unit of &tra Hospital. We are also indebted to Johan Ubby M.S., Ortivus Medical for software support and development. References 1 Russel RO Jr, Moraski RE, Kouchoukos N et al. Unstable angina pectoris: national cooperative study group to compare medical and surgical therapy. II. In-hospital experience and initial follow-up results in patients with one, two and three vessel disease. Am J Cardiol 1978;42:839-848. 2 Gerstenblith G. Ouyang P, Achuff SC et al. Nifedipine in unstable angina: a double-blind, randomized trial. N Engl J Med 1982;306:885-889. 3 Gottlieb SO, Weisfeldt ML, Ouyang P et al. Effect of the addition of propranolol to therapy with nifedipine for unstable angina pectoris: a randomized, double-blind placebo-controlled trial. Circulation 1986:73:331-337.
4 Gottlieb SO, Weisfeldt ML, Ouyang P, Mellits ED, Gerstenblith G. Silent ischemia as a marker for early unfavourable outcomes in patients with unstable angina. N Engl J Med 1986:314:1214-1219. 5 Johnson SM, Mauritson DR. Winniford MD et al. Continuous electrocardiographic monitoring in patients with unstable angina pectoris: identification of high-risk subgroup with severe coronary disease, variant angina and/or impaired early prognosis. Am Heart J 1982;103:4-12. 6 Dellborg M, Riha M, Swedberg K. Dynamic QRS- and ST-segment changes in myocardial infarction monitored by continuous on-line vectorcardiography. J Electrocardiol 1990;23(suppl): l-9. 7 Dellborg M, Emanuelsson H, Riha M, Swedberg K. Dynamic QRS-complex and ST-segment monitoring by continuous vectorcardiography during coronary angioplasty. Coronary Artery Dis 1991;2:43-53. 8 Dellborg M, Riha M, Swedberg K, for the TEAHAT study-group. Dynamic QRS-complex and ST-segment monitoring in acute myocardial infarction during recombinant tissue plasminogen activator therapy. Am J Cardiol 1991;67:343-349x. 9 Frank E. Accurate, clinically practical system for spatial vectorcardiography. Circulation 1956:13:737-744. 10 von Arnim TH, Reuschel-Janetschek E. Continuous bedside monitoring of the ECG for detection of silent myocardial ischemia. Eur Heart J 1988:9(suppl N):89-92. 11 Nademanee K, lntarachot V, Josephson M et al. Prognostic significance of silent myocardial ischemia in patients with unstable angina. J Am Coll Cardiol 1987:10:1-9. 12 Brody D. A theoretical analysis of intracavitary blood mass influence on the heart-lead relationship. Circ Res 1956;4:731-738. 13 Feldman T, Borow K. Neumann A, Lang R, Childers R. Relation of electrocardiographic R-wave amplitude to changes in left ventricular chamber size and position in normal subjects. Am J Cardiol 1985:55:1168-l 174. 14 Selvester R, Wagner N, Wagner G. Ventricular excitation during percutaneous transluminal angioplasty of the left anterior descending coronary artery. Am J Cardiol 19X8:62:1116-1121, 15 Visser C, David G, Kan G et al. Two-dimensional echocardiography during percutaneous transluminal coronary angioplasty. Am Heart J 1986;lll: 1035-1041. I6 Bertrand M, Lablanche J, Fourrier J. Traisnel G. Mirsky I. Left ventricular systolic and diastolic function during acute coronary artery balloon occlusion in humans. J Am Cob Cardiol 1988: 12:341-347. 17 Tamaki N. Yasuda T, Moore R et al. Continuous monitoring of left ventricular function by an ambulatory radionucleide detector in patients with coronary artery disease. J Am Coil Cardiol 1988;12:669-679. 18 Breisblatt W, Schulman D, Follansbee W. Continuous on-line monitoring of left ventricular function with a new nonimaging detector: validation and clinical use in the evaluation of patients post angioplasty. Am Heart J 1991:121:1609-1617.
162 19 Barry J, Cambell S, Nabel EG, Mead K, Selwyn AP. Ambulatory monitoring of the digitized electrocardiogram for detection and early warning of transient myocardial ischemia in angina pectoris. Am J Cardiol 1987;60:483-488. 20 Jamal SM, Mitra-Duncan L, Kelly DT, Freedman SB. Validation of a real-time electrocardiographic monitor for detection of myocardial ischemia secondary to coronary artery disease. Am J Cardiol 1987;60:525-527. 21 Krucoff M. Identification of high-risk patients with silent myocardial ischemia after percutaneous transluminal coronary angioplasty by multi-lead monitoring. Am J Cardiol 1988:61:29F_34F.
22 Krucoff M, Wagner N, Pope J et al. The portable programmable microprocessor-driven real-time 1Zlead electrocardiographic monitor: a preliminary report of a new device for the noninvasive detection of successful reperfusion or silent coronary reocclusion. Am J Cardiol 1990;65:143-148. 23 Von Essen R, Hinsen R, Louis R et al. On-line monitoring of multiple precordial leads in high risk patients with coronary artery disease - a pilot study. Eur Heart J 1984;5:203-210.
151
CARD10 01490
Dynamic changes of the QRS complex in unstable angina pectoris Mikael Dellborg,
Gunnar
Gustafsson,
Martin
Riha and Karl Swedberg
Department of Medicine, Unirlersity of Giiteborg, &tra Hospital 416 85 GGteborg, Sweden
(Received 25 November 1991; revision accepted 26 February 1992)
Dellborg M, Gustafsson G, Riha M, Swedberg K. Dynamic changes of the QRS complex in unstable angina pectoris. Int J Cardiol 1992;36:151-162. Despite intensive medical treatment to control chest pain, about one-third of patients with unstable angina have an unfavourable outcome within a period of 1 to 2 months. Holter monitoring can identify patients with silent myocardial ischaemia that are at a high risk of sustaining a major cardiovascular event. The present paper describes the use of dynamic, continuous, computerized on-line vectorcardiography for real-time monitoring of QRS-complex and ST-segment changes in patients with unstable coronary disease. In many patients a pattern of frequent repetitive episodes of QRS change was observed, with or without concomitant ST change. Whereas no patient had episodes of ST-vector change without also having episodes of significant QRS change, 15 patients had several episodes of QRS changes without any episode of significant ST change. The number of episodes of significant increase of the QRS vector difference correlated weakly but significantly with the number of episodes of significant ST-vector magnitude change (r = 0.34, p < 0.05). The present study suggests that myocardial ischaemia will influence the QRS complex as well as the ST segment. The mechanism behind the QRS changes observed is not clear but episodes of QRS change without ST change or chest pain, may reflect sudden depressions of left ventricular function, as has been reported by others to occur in patients with coronary artery disease. Dynamic vectorcardiography offers the opportunity to monitor all parts of the QRST complex in real time. Key words: Unstable angina; ECG monitoring; Vector cardiography
Correspondence
to: M. Dellborg M.D., Dept. of Medicine,
&tra Hospital, 416 85 Gteborg, Sweden. Tel. 46-31-374000. Fax 46-31-258933. Supported by a grant from AFA, the Labour Market Insurance Company, Stockholm, Sweden, TryggHansa Insurante Company, Stockholm, Sweden, The Heart-Lung Foundation, Stockholm, Sweden and AB Hassle Cardiovascular Research Laboratories, Molndal, Sweden.
Introduction Unstable angina is a condition caused by myocardial ischaemia, characterized by symptoms of recent onset, a crescendo pattern or the occurrence of symptoms at rest. Despite intensive medical treatment to control chest pain, about one-
152
third of patients with unstable angina have an unfavourable outcome within a period of 1 to 2 months [l-3]. Continuous 2-channel Holter monitoring can identify patients with silent myocardial ischaemia that are at a high risk of sustaining a major cardiovascular event and thus have an impaired short-term prognosis [4,51. We have previously described the use of dynamic, continuous, computerized on-line vectorcardiography to monitor patients with acute myocardial infarction as well as during coronary angioplasty [6-81. This system allows on-line monitoring of QRS-complex and ST-segment changes with analysis and presentation in real time, potentially identifying high-risk patients and allowing monitoring of acute interventions in unstable coronary disease. The present report describes the system and summarizes our experience with real-time, dynamic vectorcardiography in patients with unstable angina. Methods Normal subjects Twenty healthy volunteers, 10 men and 10 women, aged under 30, with normal standard 1Zlead eIectrocardiogram were monitored from 6 p.m. to 6 a.m. while resting in bed. They were all without on-going medication and experienced no symptoms during the 12-h monitoring period. Patient series Patients with unstable angina pectoris as defined below, included in a randomized trial of intravenous vs buccal nitroglycerin (n = 28) or included in a comparative multi-centre study of vectorcardiography vs Holter-monitoring (n = 151, were monitored with continuous dynamic vectorcardiography for 24 h after admission to the coronary care unit. Patients were treated with beta-blockers, aspirin and if necessary calcium antagonists in addition to high-dose nitrates. Changes in the resting 12-lead scalar electrocardiogram were not necessary for inclusion. All patients were closely followed throughout the hospital stay and were, if not stabilized medically,
subject to coronary angiography and if required, revascularization was performed. Continuous vectorcardiography The monitoring system (MIDA, Myocardial Infarction Diagnosis and Analysis, Ortivus Medical AB, Taby, Sweden) consists of a microprocessor controlled data-acquisition module, an IBM compatible personal computer and a graphic printer. Electrocardiographic signals are continuously collected from conventional body surface electrodes applied to the patient’s chest according to the Frank lead system [9]. The sensitivity of the system is 1 PV and the sampling rate of each lead is 500 samples/s. Electrocardiographic complexes are detected, collected and placed according to their shape into 1 of 5 classes. The most dominant beat type is automatically determined during the first 10 s of recording and termed the zero class. The 3 orthogonal vectorcardiographic leads X, Y and Z are computed from sampled unipolar Frank leads and continuously displayed on the colour monitor and averaged to form mean vectorcardiographic complexes. Averaging is performed for consecutive periods, each period being 10 s to 4 min long. For this study, averaging periods of 1 min were used. Only beats classified as class zero were used for subsequent analysis. For each period the resulting mean vectorcardiographic complex was analyzed and compared a reference complex collected during the second averaging period. Any of the trend parameters may be presented as a trend curve updated continuously during the whole recording period. The screen of a colour monitor attached to the personal computer may be divided in several windows with different information, such as running vectorcardiographic signal, any mean vectorcardiographic complex collected or trend curves, presented in each window. The data collection and calculations are all done on-line. An alarm may be set to identify a change in any of the trend parameters to alert the attending staff to significant changes in the QRS complex or the ST segment. The computer will automatically store all mean complexes and trend curves on the hard disk and keep them available for on-line review-
153
ing at any time during the recording. The recording may then also be reviewed and summarized in off-line mode. The data may be transferred to a floppy disk and stored. Coronary care unit nurses were trained to connect the patient, start monitoring and recording and to supervise signal quality. Initial training took approximately 3 h. Monitored vectorcardiographic
variables
Based on previous results and observations [6-81 3 trend parameters were closely studied: the QRS-vector difference, which summarizes all changes within the QRS complex, the ST-vector magnitude, which is the deflection of the ST segment from the isoelectric level 20 ms after the J point and the ST change vector magnitude which is the length of the difference vector between the initial ST vector and the current ST vector. These parameters are described in Fig. 1. In Fig. 2 trend curves from 3 patients are shown. Fig. 2a shows the simultaneous changes occurring with and without chest pain in ST-vector magnitude, ST-change vector magnitude and QRS-vector difference in a patient with unstable angina. Fig. 2b shows a patient who develops an acute myocardial infarction illustrated by the continuous and persistent increase in QRS vector difference. This patient was not part of the present study but is shown for comparative reasons only. A patient with episodes of significant QRSvector difference change but without significant increase in either the ST-vector magnitude or the ST-change vector magnitude is illustrated in Fig. 2c. Monitoring
of chest pain
Pain was evaluated hourly during vectorcardiographic monitoring by the attending nurse, asking the patient to recall the time, number and severity of episodes of chest pain. Definitions
chest pain occurring at rest or at a very low level of exercise. The pain should give strong suspicion of ischaemic heart disease but not myocardial infarction and last for at least 5 but not more than 30 min and should have been present at least twice during the last 24 h. Significant vectorcardiographic change. Based on findings from previous studies [7,8,10] we postulated that a significant change in the QRS-vector difference indicating ischaemia would consist of a reversible increase of more than 15 ~VS, lasting > 1 min and with > 1 min between episodes. Such episodes will be referred to as “QRS-episodes” in the text. For ST-vector magnitude and ST-change vector magnitude an increase to above 0.1 mV (corresponding to an ST-segment change of > 1 mm in either 1 of the 3 orthogonal leads) was considered to indicate ischaemia. These changes will be referred to as “ST episodes” or “ST-change episodes”. Vectorcardiographic changes are expressed as number of episodes and/or total duration per patient per 24-h period. Intervention. Coronary angiography, percutaneous transluminal coronary angioplasty or coronary artery bypass grafting during the hospitai stay due to failure to achieve medical stabiiization of angina were considered interventions. Myocardial infarction. Myocardial infarction was defined with the usual criteria, i.e. 2 out of 3 of: typical chest pain, elevation of cardiac enzymes, electrocardiographic changes. Vectorcardiographic changes were not used for diagnosis. There were 4 patients in whom a definite acute myocardial infarction was documented by release of cardiac enzymes during the initial 24 h of monitoring. These patients were considered to have had an ongoing myocardial infarct already when included; they were therefore not included in the analysis. One of these patients is illustrated in Fig. 2b for comparative reasons. Statistics
Unstable angina. Recent onset angina or a sudden change in a previously stable pattern, with
All values are expressed as mean + SEM. Groups were compared using Mann-Whitney U-
154 ORS
- Vector
Difference
statistics. Distribution of observations was compared with chi-square analysis with continuity correction or when appropriate with 4-fold table test. A p value of < 0.05 was considered significant. Results Normal subjects Ax
Ay. AZ i
QRS-VD =
ST - Vecior
Ax2
Magnitude
in,
ih_ir\, Current
complex
-
ST-VM=\/Xf+
STC
t Ay2 + AZ’
- Vector
Y:
+
The healthy control group reported no episodes of chest pain during the monitoring period. Compared to patients with unstable angina the control group had fewer ST and QRS episodes (see Table 1). Most notably, no normal subject had any ST-change episode. The effect of changing body position is shown in Fig. 3.
2:
Magnitude
Fig. 1. The upper panel shows a trend curve with time on the horizontal axis and QRS-vector difference and ST-vector magnitude on the vertical axis. An example of an episode of reversible change of both parameters is shown. The reference complex is the second average recorded for an individual patient. The ischaemic complex is taken from the point of maximum change on the trend curve. The computer will in real time calculate the area of difference between the reference complex and any subsequent complexes. The figure shows the comparison between the reference complex and the ischaemic complex. The QRS-vector difference is the hatched area of difference between the 2 average complexes. The ST-vector magnitude is the deviation of the ST segment from the baseline, measured at 20 ms after the J point. The ST-deviation is quadratically added for all three leads, producing the total ST-vector magnitude. The ST-change (STC) vector magnitude is the length of the vector going from the tip of the ST vector of the reference complex to the tip of the ST vector of the ischaemic complex. The ST-change vector magnitude is thus the ST-vector magnitude corrected for the degree of ST deviation observed on the reference complex. However, the ST-change vector magnitude also takes into consideration the angular changes of the ST vector. STC = ST change; ST-VM: ST-vector magnitude; QRS-VD = QRSvector difference.
Unstable angina group There were 43 patients with unstable angina pectoris monitored for 24 h. Baseline characteristics of these patients are shown in Table 2. Clinical data regarding chest pain, myocardial infarction, death or the need for acute intervention are given in Table 3 for patients with unstable angina. QRS episodes were seen in 88% of the patients whereas ST or ST-change episodes were seen in 52 and 43%, respectively (Table 1). Correlation between QRS and ST episodes. In Fig. 4, for every patient the number of ST or ST-change episodes is related to the number of QRS episodes. The number of QRS episodes correlated weakly but significantly with the number of ST episodes (r = 0.34, p < 0.05). Only 1 patient had ST or ST-change episodes without also having QRS episodes. However, 15 patients had several QRS episodes without any ST or ST-change episodes. Of these 15 patients, 12 had chest pain at rest associated with QRS episodes. Medical therapy failed to stabilize 1 of these patients and acute coronary angiography revealed 2-vessel disease with a highly significant stenosis of the left anterior descending artery (patient recording shown in Fig. 2~).
155
Silent myocardial ischaemia. The occurrence of ST-change episodes without concomitant chest pain was analyzed as an indication of silent my-
ocardial ischaemia. The unstable angina group was divided according to the presence (n = 17) or absence (n = 25) of asymptomatic ST-change
(b)
Fig. 2. a. Part of the trend curve from a patient with unstable angina. The QRS-vector difference, the ST-vector magnitude and the ST-change vector magnitude are shown. Simultaneous changes occur in all 3 parameters, in some cases associated with chest pain (arrow). b. The same parameters recorded from a patient with unstable angina who develops an acute myocardial infarction. Note the sustained increase of the QRS-vector difference. This patient was not part of the present report but is included for comparative reasons. c. Patient with repeated QRS episodes. The patient has no ST or ST-change episodes. Angiography during the hospital stay revealed a high-grade stenosis of the left anterior descending artery; an uncomplicated angioplasty was performed. Arrows indicate chest pain. ~VS = microvoltseconds, unit for QRS-vector difference; mV = millivolt, unit for ST and ST-change vector magnitude; STC-VM = ST-change vector magnitude; ST-VM = ST-vector magnitude; QRS-VD = QRS-vector difference.
156
TABLE
1
Comparison
of vectorcardiographic
parameters Normal
between
normal
subjects
NS = not significant.
ST change:
and patients Unstable
No. Patients ST episodes no. of episodes/patient duration/patient (mitt) patients with episodes ST-change episodes no. of episodes/patient duration/patient (min) patients with episodes QRS episodes no. of episodes/patient duration/patient (mitt) patients with episodes Episodes of chest pain
subjects
9%
angina. P
angina
No.
20
%
43
4*2 144 * 78 5
5*1 140 f 56 23
25
0 0 0
2+1 27* 15 19
6+2 118+38 14 0
11+1 180 f 24 38 3*1
ST-change
with unstable
70
vector
magnitude.
episodes (see Table 4). Overall 62 of 78 ST-change episodes (79%) were asymptomatic. Only 1 patient had symptomatic ST-change episodes without having asymptomatic ST-change episodes. Znteruention. Patients with unstable angina were divided according to the need for interven-
Values
52
NS NS < 0.05
43
< 0.001 < 0.001 < 0.001 < 0.05 < 0.05 NS < 0.001
88
are given as mean f SEM.
tion. In the 12 patients that failed medical apy, ST and ST-change episodes occurred frequently and were of longer duration as pared to patients who responded to medical apy and thus were not subject to intervention Table 5).
thermore comther(see
Myocardial infarction. Myocardial infarction after the initial 24-h monitoring but before hospiTABLE
2
Baseline
characteristics
of 43 patients
Age, yr (range) Previous infarction Men/women Previous CABG/PTCA Medication on admission beta-blockers aspirin long-acting nitrates calcium antagonists Electrocardiogram on admission ST elevation ST depression T abnormalities normal CABG = coronary neous transluminal
with unstable
No.
%
59+F3 16 29/14 0
(39-75) 37 67/33
32 17 12 7
74 40 28 16
3 13 14 13
7 30 33 30
artery bypass grafting; coronary angioplasty.
angina.
PTCA = percuta-
TABLE
3
Clinical
data,
patients
with unstable
angina.
No. During initial 24 h Patients with chest pain at rest l-3 episodes > 3 episodes Patients without chest pain at rest Death Total From 24 h to hospital discharge Coronary angiography/ CABG/PTCA Myocardial infarction Death CABG = coronary neous transluminal
%
29
67 12 17
14 0 43
12 2 0
artery bypass grafting; coronary angioplasty.
28 39 33 100
28 7
PTCA = percuta-
157
tal discharge developed (5%).
in 2 of the 43 patients
Discussion The present study investigated the use of computerized on-line vectorcardiographic monitoring of patients with unstable angina. The definition of unstable angina is often difficult and therefore it is difficult to compare different studies. In the present study 28% needed acute intervention as they could not be stabilized medically. Observations from these patients were compared to observations made in a group of normal subjects.
ing, we cannot completely rule out that such information influenced the attending physician in the making of this decision. Therefore, our identification of high-risk patients by monitoring the ST-change vector magnitude must be cautiously interpreted and subjected to further study. QRS-vector changes in myocardial ischaemia Although QRS episodes were frequently seen among normal subjects they were observed in a
(I.3 1 STC-VM
ST-vector changes in myocardial ischaemia 0.1s
The ST-vector magnitude with the presently used limit for defining myocardial ischaemia (STVM > 0.1 mV) seems to have a low specificity since normal subjects had a similar number and duration of ST episodes as the unstable patients. The healthy volunteers were all young and many had a high baseline ST-vector magnitude at rest due to vagotonic ST-segment elevation. A possible solution to this problem would be to measure the ST-vector magnitude as a percentage of change from baseline in the individual patient. The ST-change vector magnitude is a measurement of change in relation to the initial ST vector in an individual patient. In the present study it was superior to the ST-vector magnitude and the QRS-vector difference in separating normal subjects from patients with coronary disease. The ST-change vector magnitude also identified a group of patients with electrocardiographic changes without symptoms, tentatively defined as having silent myocardial ischaemia. With conventional ST monitoring, patients with unstable angina and frequent asymptomatic ST changes are at high risk of sustaining a coronary event within the following 1 to 2 months after hospitalization for unstable angina 14,111. In the present study, patients that could not be stabilized medically had a significantly higher number and a longer total duration of ST-change episodes. Although the decision to intervene was not based on the results of the vectorcardiographic monitor-
0.5
ST-VM I
QRS-VD
SUPiTlC
left side
supine
qht
sldc
sup,nc
sIllins
sup,nc
Fig. 3. Mean f standard deviation of the QRS-vector difference, the ST-vector magnitude and the ST-change-vector magnitude of 20 healthy volunteers. Body position was changed every 5 min in a non-random way as indicated in the figure. For every position and person, the maximum value was used. STC-VM = ST-change vector magnitude; ST-VM = STvector magnitude: QRS-VD = QRS-vector difference.
158
significantly higher frequency and were of longer duration in patients with unstable angina. As shown in Figs. 2c and 4, many QRS episodes occurred without ST or ST-change episodes. Actually, in many patients a pattern of frequent repetitive QRS episodes with or without ST or ST-change episodes was detected. This pattern was not seen among controls. In previous reports we have found that myocardial infarction will result in a persistent increase of the QRS-vector difference after 24 h of recording [6,8]. The QRS-vector difference thus seems to be reversibly influenced by myocardial ischaemia but the development of necrosis will cause sustained changes.
Mechanism myocardial
behind the QRS changes ischaemia
during
The precise mechanism behind QRS changes during myocardial ischaemia is not known. Several mechanisms have been proposed, including the Brodie effect 1121. In normal subjects positional changes will cause relatively large QRS changes (see Fig. 3). The distance from the electrode to the left ventricular wall is of critical importance. This distance has been shown to decrease significantly when the subject is in the left supine position [13]. Simultaneous changes in the QRS complex occur and the QRS-vector loop will then increase its area, “dilate”. With the
ST episodes 35 30
I 0
25
ORS
rpisodes
STC episodes 20 18 16 14 12 10
a 6 4 2 0
0
5
10
15
20
25
30
35 ORS
Fig. 4. Number
of ST (a) or ST-change episodes Overlaps are indicated
episodes
(b) per 24 h is plotted against the number of QRS episodes by increased size of the symbols used. STC = ST change.
for individual
patients.
present technique, this will appear as a simuItaneous increase in the QRS-vector difference, i.e. a QRS episode. We have previously reported that changes in the QRS loops and of the QRS-vector difference occur during myocardial ischaemia in all patients during percutaneous coronary angioplasty [71. Reversible changes of the QRS complex occur when a coronary artery is temporarily occluded by the angioplasty balloon. This ischaemic reaction of the QRS complex is believed to be caused by delayed activation of the ischaemic myocardium. Since non-ischaemic myocardium will be depolarized earlier, the delayed activation of the ischaemic part of the myocardium will be unopposed electrically and produce a larger vector loop [14]. Myocardial ischaemia will also induce increased end-systolic volume [15,16], which will shorten the distance from the myocardium to the electrode. Tamaki and co-workers monitored ejection fraction by an ambulatory radionucleide recorder in patients with coronary artery disease [17]. They reported a
TABLE
high incidence of simultaneously occurring chest pain, ST-segment depression and a decrease in ejection fraction. Among 39 patients they also observed 20 episodes in 14 patients of transiently depressed ejection fraction without concomitant ST-segment changes. They suggested that ischaemia may not always produce neither pain nor ST depression but its sole manifestation may be depressed left ventricular function. We speculate that the QRS episodes that we have observed in patients with unstable angina may reflect sudden depressions of left ventricular function, with or without ST-segment changes and chest pain. Similar episodes of changes in left ventricular function in the absence of ECG changes or chest pain have also been reported following coronary angioplasty [ 181. The mechanism behind a change in the QRSvector difference is different between normal subjects and patients with myocardial ischaemia. The appearance of the vector loop is, however, similar in both instances. Therefore, the present tech-
4
Vectorcardiographic
and clinical
data for patients
with and without
With asymptomatic episodes No. Patients Chest pain at rest,
17
no. of episodes/patient patients with episodes ST episodes
3+ 13
no. of episodes/patient duration/patient (min) patients with episodes ST-change episodes
asymptomatic
ST-change
episodes
No.
ischaemia”). P
No asymptomatic episodes %
(=“silent
%
2.5 1
2 6+ 182 k 124 13
76
4 15
71
5 118 10
+
55 66~ 14 13*
2
10
duration/patient (min) patients with episodes AMI or intervention
202+ 17 9
31
173 21 4
100
100 53
NS NS
40
NS NS = 0.05
f 2 f 53
0.04 t 0.08 k
1 34
no. of episodes/patient duration/patient (min) patients with episodes QRS episodes no. of episodes/patient
60
1
0.04 0.08
1
4 f? * 35 84 16
< 0.00 1 < 0.001 < 0.001 NS NS NS = 0.05
AMI = acute myocardial infarction; intervention = coronary angiography, angioplasty or coronary artery bypass surgery during hospital stay. NS = not significant. All values are given as mean + SEM. One patient had no information about chest pain.
160 TABLE
5
Vectorcardiographic
and clinical
data for patients
with acute intervention
Intervention
Stabilized
No. Patients Mean age (yr) Male/female Chest pain at rest, no. of episodes/patient patients with episodes
2.5 + 11
ST episodes no. of episodes/patient duration/patient (min) patients with episodes ST-change episodes no. of episodes/patient duration/patient (min) patients with episodes ST-change episodes, silent no. of episodes duration/patient (min) patients with episodes QRS episodes no. of episodes/patient duration/patient (min) patients with episodes Intervention = coronary angiography, values are given as mean f SEM.
1
8* 3 334 f 177 10 4.8 f 76 + 8
2 49
4* 65 + 8
1 46
12 191 12
+ +
3.5 * 18
83
4+1 63 + 28 13
67
0.7 f 7+ 11
67
0.5 + 0.1 5 +13 10
angioplasty
100 or coronary
nique cannot differentiate and pathogenetically separate individual episodes of QRS-vector difference change. Because of a low number of patients and controls the statistical power of the present study is limited. Our findings are preliminary and the relative value of the different vectorcardiographic parameters for detection and quantification of myocardial ischaemia are at present not clear. Other monitoring
10 176 26
systems
Holter recording and off-line analysis is today a frequently used technique to monitor myocardial ischaemia. The merit of ischaemia monitoring as a guide to therapy and intervention in unstable angina pectoris has not been proven in controlled clinical trials. However, it is of substantial practical value if the results of ischaemia monitoring in patients with unstable angina can
artery
bypass
medically.
I-J
7
92
2 42
stabilized
%
31 59 f 21/10
9
and for patients
medically
No.
%
12 60 F 8/4
(=“intervention”l
1 58
NS NS
42
< 0.05 < 0.05 = 0.05
35
< 0.001 < 0.05 NS
32
< 0.001 < 0.01 NS
84
NS NS NS
0.2 4
+ 2 + 30
surgery
during
hospital
stay. NS = not significant.
All
be made available to the responsible clinician in real time. The “Monitor One” is a solid-state Holter recorder for on-line analysis and its use in patients with stable angina and during exercise testing has been reported 119,201. It has been pointed out that multi-lead recordings provide improved detection of ischaemic episodes and may also provide important prognostic information when used for evaluation of the total ischaemic burden [21]. At present there are only 2 systems that will allow multiple-lead, on-line monitoring of electrocardiographic changes. The Mortara system, which uses continuous real-time monitoring of the conventional 12-lead electrocardiogram, has been extensively investigated by Krucoff and coworkers in patients with acute myocardial ischaemia [22] and has also been applied to patients with unstable angina [lo]. Von Essen and co-workers [23] described a multiple-lead system
I61
for real-time monitoring of patients with myocardial infarction and unstable angina. Their system would also allow monitoring of QRS changes in addition to ST changes. The relative sensitivity and specificity for detecting myocardial ischaemia for the vectorcardiographic system we have used in relation to other monitoring systems cannot be assessed at present. In conclusion, continuous on-line vectorcardiography is feasible for monitoring of patients with coronary artery disease. Myocardial ischaemia will influence the QRS complex as well as the ST segment and this technique offers the opportunity to monitor all parts of the QRST complex in real time. Vectorcardiography has the advantage over other multiple lead systems to summarize all changes in the QRS complex in one easily monitored parameter, the QRS-vector difference. In addition it offers several ST-vector parameters. The present study suggests that continuous vectorcardiography may be helpful in identifying patients with unstable angina at high risk and separate the patients with acute myocardia1 infarction. Acknowledgements We are grateful to Kerstin Korssell R.N. and Ann-Marie Svensson R.N. for technical assistance and to the nursing staff at the coronary care unit of &tra Hospital. We are also indebted to Johan Ubby M.S., Ortivus Medical for software support and development. References 1 Russel RO Jr, Moraski RE, Kouchoukos N et al. Unstable angina pectoris: national cooperative study group to compare medical and surgical therapy. II. In-hospital experience and initial follow-up results in patients with one, two and three vessel disease. Am J Cardiol 1978;42:839-848. 2 Gerstenblith G. Ouyang P, Achuff SC et al. Nifedipine in unstable angina: a double-blind, randomized trial. N Engl J Med 1982;306:885-889. 3 Gottlieb SO, Weisfeldt ML, Ouyang P et al. Effect of the addition of propranolol to therapy with nifedipine for unstable angina pectoris: a randomized, double-blind placebo-controlled trial. Circulation 1986:73:331-337.
4 Gottlieb SO, Weisfeldt ML, Ouyang P, Mellits ED, Gerstenblith G. Silent ischemia as a marker for early unfavourable outcomes in patients with unstable angina. N Engl J Med 1986:314:1214-1219. 5 Johnson SM, Mauritson DR. Winniford MD et al. Continuous electrocardiographic monitoring in patients with unstable angina pectoris: identification of high-risk subgroup with severe coronary disease, variant angina and/or impaired early prognosis. Am Heart J 1982;103:4-12. 6 Dellborg M, Riha M, Swedberg K. Dynamic QRS- and ST-segment changes in myocardial infarction monitored by continuous on-line vectorcardiography. J Electrocardiol 1990;23(suppl): l-9. 7 Dellborg M, Emanuelsson H, Riha M, Swedberg K. Dynamic QRS-complex and ST-segment monitoring by continuous vectorcardiography during coronary angioplasty. Coronary Artery Dis 1991;2:43-53. 8 Dellborg M, Riha M, Swedberg K, for the TEAHAT study-group. Dynamic QRS-complex and ST-segment monitoring in acute myocardial infarction during recombinant tissue plasminogen activator therapy. Am J Cardiol 1991;67:343-349x. 9 Frank E. Accurate, clinically practical system for spatial vectorcardiography. Circulation 1956:13:737-744. 10 von Arnim TH, Reuschel-Janetschek E. Continuous bedside monitoring of the ECG for detection of silent myocardial ischemia. Eur Heart J 1988:9(suppl N):89-92. 11 Nademanee K, lntarachot V, Josephson M et al. Prognostic significance of silent myocardial ischemia in patients with unstable angina. J Am Coll Cardiol 1987:10:1-9. 12 Brody D. A theoretical analysis of intracavitary blood mass influence on the heart-lead relationship. Circ Res 1956;4:731-738. 13 Feldman T, Borow K. Neumann A, Lang R, Childers R. Relation of electrocardiographic R-wave amplitude to changes in left ventricular chamber size and position in normal subjects. Am J Cardiol 1985:55:1168-l 174. 14 Selvester R, Wagner N, Wagner G. Ventricular excitation during percutaneous transluminal angioplasty of the left anterior descending coronary artery. Am J Cardiol 19X8:62:1116-1121, 15 Visser C, David G, Kan G et al. Two-dimensional echocardiography during percutaneous transluminal coronary angioplasty. Am Heart J 1986;lll: 1035-1041. I6 Bertrand M, Lablanche J, Fourrier J. Traisnel G. Mirsky I. Left ventricular systolic and diastolic function during acute coronary artery balloon occlusion in humans. J Am Cob Cardiol 1988: 12:341-347. 17 Tamaki N. Yasuda T, Moore R et al. Continuous monitoring of left ventricular function by an ambulatory radionucleide detector in patients with coronary artery disease. J Am Coil Cardiol 1988;12:669-679. 18 Breisblatt W, Schulman D, Follansbee W. Continuous on-line monitoring of left ventricular function with a new nonimaging detector: validation and clinical use in the evaluation of patients post angioplasty. Am Heart J 1991:121:1609-1617.
162 19 Barry J, Cambell S, Nabel EG, Mead K, Selwyn AP. Ambulatory monitoring of the digitized electrocardiogram for detection and early warning of transient myocardial ischemia in angina pectoris. Am J Cardiol 1987;60:483-488. 20 Jamal SM, Mitra-Duncan L, Kelly DT, Freedman SB. Validation of a real-time electrocardiographic monitor for detection of myocardial ischemia secondary to coronary artery disease. Am J Cardiol 1987;60:525-527. 21 Krucoff M. Identification of high-risk patients with silent myocardial ischemia after percutaneous transluminal coronary angioplasty by multi-lead monitoring. Am J Cardiol 1988:61:29F_34F.
22 Krucoff M, Wagner N, Pope J et al. The portable programmable microprocessor-driven real-time 1Zlead electrocardiographic monitor: a preliminary report of a new device for the noninvasive detection of successful reperfusion or silent coronary reocclusion. Am J Cardiol 1990;65:143-148. 23 Von Essen R, Hinsen R, Louis R et al. On-line monitoring of multiple precordial leads in high risk patients with coronary artery disease - a pilot study. Eur Heart J 1984;5:203-210.
