JACC Vol. 26, No. 6 150] November 15, 1995:1501-7 ELECTROPHYSIOLOGY Dynamic On-Line Vectorcardiography Improves and Simplifies In-Hospital Ischemia Monitoring of Patients With Unstable Angina MIKAEL DELLBORG, MD, KLAS MALMBERG, MD,* LARS RYDI~N, MD, FACC,* ANN-MARIE SVENSSON, RN, KARL SWEDBERG, MD, FACC GOteborg and Stockholm, Sweden Objectives. This study sought to validate computerized vector- of recording with both channels technically adequate for analysis cardiography against the established technique of Holter electro- compared with 23.7 -+ 1.77 h of vectorcardiographic recording cardiographic (ECG) monitoring and to compare the feasibility of that could be analyzed (p < 0.01). Of the 15 symptomatic episodes the two methods for monitoring patients with unstable angina detected by Holter ECG monitoring, 13 were also detected with pectoris. dynamic vectorcardiography. In contrast, eight patients had 18 Background. Detection of myocardial ischemic episodes is an episodes of chest pain, with simultaneous ST segment changes important objective in patients admitted to the hospital for detected by dynamic vectorcardiography; only 9 of these episodes unstable angina pectoris. Standard ECG monitoring may be were also detected by Holter ECG monitoring. sufficient for detection of symptomatic episodes but will often Conclusions. Monitoring of myocardial ischemia with dynamic overlook silent ischemia. Holter ECG monitoring has a higher vectorcardiography seems to be more efficient than Holter moni- likelihood of discovering such episodes, but analysis is time- toring and may have a higher sensitivity. Computerized, continu- consuming, and the results are not available on-line. ous vectoreardiography has a complete real-time capacity, allow- Methods. We simultaneously monitored 53 consecutive patients ing monitoring over prolonged periods of time, and the results are with unstable angina, 46 of whom had technically adequate 24-h immediately available without time-consuming analysis. Holter ECGs and computerized vectorcardiograms. (JAm CoU Cardiol 1995;26:1501-7) Results. The Holter tapes had a mean (_+SD) of 15.3 + 10.3 h Unstable angina is a condition caused by myocardial ischemia, coronary angioplasty and unstable angina. With this technique, characterized by symptoms of recent onset, a crescendo pat- the dynamic ECG changes seen during myocardial ischemia tern or the occurrence of symptoms at rest. Despite intensive and infarction can be monitored in real time. Our previous medical treatment to control chest pain, approximately one- experience with this technique indicated that dynamic vector- third of patients with unstable angina have an unfavorable cardiography could improve detection of ischemia during outcome within a period of 2 months (1-3). During the acute monitoring of patients with unstable angina. hospital period, ordinary electrocardiographic (ECG) moni- The vectorcardiographic ST segment criteria used by us toring in the coronary care unit may detect symptomatic initially were shown to yield a significant incidence of apparent episodes of myocardial ischemia. However, silent episodes may ischemic episodes in 20 healthy volunteers, that is, the speci- pass undetected. Continuous Holter ECG monitoring identi- ficity of our criteria was poor. Reanalysis of our earlier data fies silent episodes of myocardial ischemia in patients with provided criteria with improved specificityfor the original data unstable angina, thereby increasing the probability of obtaining set. The present study was performed to validate our new more complete information on the pattern of disease and vectorcardiographic criteria against the established technique improved prognostic information (4,5). of Holter ECG monitoring to compare the feasibility of We previously described (6-8) the use of dynamic contin- dynamic vectorcardiography and Holter ECG monitoring in uous computerized on-line vectorcardiography to monitor patients with unstable angina pectoris and, in addition, to patients with acute myocardial infarction as well as during evaluate the prognostic information obtained by the two monitoring systems. From the Department of Medicine,Ostra Hospital, Universityof GOteborg, Grteborg and *Department of Cardiology, Karolinska Hospital, Karolinska Methods Institute, Stockholm, Sweden. This study was supported by a grant from AFA, the Labor Market InsuranceCompany, Stockholm and by the SwedishHeart and Patients. Fifty-three consecutive patients with unstable an- Lung Foundation, Stockholm and Ortivus Medical, T/iby, Sweden. gina admitted to the coronary care units of our hospitals Manuscript received August 22, 1994; revised manuscript received July 3, underwent simultaneous Holter ECG monitoring and dynamic 1995, accepted July 10, 1995. Address for correspondence:Dr. Mikael Dellborg, Department of Medicine, vectorcardiography. Informed oral consent was obtained from Universityof GOteborg 0stra Hospital, S-416 85 Grteborg, Sweden. all patients before the procedure, and the protocol was ap- ©1995 by the American College of Cardiology 0735-1097/95/$9.50 0735-1097(95)00361-7 1502 DELLBORG ET AL. JACC Vol. 26, No. 6 VECTORCARDIOGRAPHY IN UNSTABLE ANGINA November 15, 1995:1501-7 proved by the ethics committee of the University of G~3teborg. ST segment depression >1 mm or elevation >2 mm continu- In 7 patients either the Halter (5 patients) or the vectorcar- ously present for at least 60 s. Ischemic changes had to be diographic data (1 patient), or both (1 patient), could not be separated by at least 60 s to be counted as two discrete analyzed for technical reasons, leaving 46 patients for inclusion episodes. All episodes of ST segment depression or elevation in the final analysis (34 men, 12 women; mean [_+SD] age 64 +_ were printed out for manual check by a technician with no 10 years, range 43 to 82). Unstable angina was defined as knowledge of any clinical or vectorcardiographic data from the follows: 1) anginal pain commencing no longer than 4 weeks patient. before admission or sudden deterioration of a previously stable Dynamic veetoreardiography. Continuous on-line comput- angina pectoris within the same time frame; 2) at least one erized vectorcardiographic monitoring was performed using a episode of anginal chest pain within the past 24 h, each episode MIDA 1000 system (Ortivus Medical AB, T~iby, Sweden). The lasting <30 rain and strongly indicating ischemic heart disease method has been described in detail elsewhere (9). Briefly, but not acute myocardial infarction. Both criteria had to be eight electrodes (seven plus one ground electrode) were fulfilled for inclusion. Furthermore, patients with bundle positioned according to Frank (10), and three orthogonal leads branch block (QRS width >0.12) and suspicion of unstable (X, Y and Z) were continuously monitored and analyzed. After angina secondary to anemia, arrhythmias or heart failure were the operator has accepted the template beat as the normal beat not included. for that patient, the computer automatically discards all beats Clinical course. Of the 46 patients included, 4 were dis- with a grossly different configuration. Accepted beats are charged with a diagnosis of acute myocardial infarction, and 2 averaged every 30 s to form a mean beat for that period. The with noncardiac chest pain; the remaining 40 had unstable first two 30-s periods are used as the reference for all further angina. Medical treatment was given at the discretion of the comparisons. All calculations are done in real time. The result responsible physician. Beta-adrenergic blocking agents were is presented on a computer screen, and changes are related to given to 40 of the 46 patients (87%), aspirin to 35 (76%), the initial reference mean beat for each patient to be presented heparin to 15 (33%), long-acting nitrates to 24 (52%), intra- over time as trend curves. The following vectorcardiographic venous nitrates to 17 (37%) and calcium-channel blocking variables were studied: QRS vector difference, ST segment agents to 13 (28%). Coronary angiography was performed in vector magnitude 20 ms after the J point (STVM20), ST 20 patients (43%) whose angina could not be stabilized by segment vector magnitude 60 ms after the J point (STVM60) medical treatment within 2 to 3 days. Three of these patients and ST segment change vector magnitude. The QRS vector had one-vessel disease (~50% narrowing), and eight each had difference, ST segment vector magnitude and ST segment two- and three-vessel disease. The final patient did not reveal change vector magnitude are shown in Figure 1. any significant coronary stenosis. Revascularization was per- Definitions. According to previous results (8) in a different formed during the initial hospital phase in 14 of these 20 group of (angina) patients and 20 normal control subjects, with patients (2 had angioplasty; 12 had coronary artery bypass some modifications, myocardial ischemia as detected by dy- grafting). The result of ischemia monitoring was not used for namic vectorcardiography was defined as follows: 1) QRS the decision to perform revascularization. Nineteen of the 26 vector difference = reversible increase to >15/xV for >1 min patients whose angina was stabilized by medical treatment (i.e., lasting for two averaged periods); 2) STI/M20, STVM60 = underwent a submaximal exercise test before hospital dis- reversible increase >50 ~V from the individual baseline for charge. The test results were suggestive of coronary artery >1 min; 3) ST segment change vector magnitude = reversible