Quantification of ST-Segment Changes During Coronary Angioplasty in Patients with Left Bundle Branch Block Karen S. Stark, MD, Mitchell W. Krucoff, MD, Beverly Schryver, MS, and Kenneth M. Kent MD, PhD
Electrocardiographic manifestations of ischemia are difficult to interpret in the presence of left bundle branch block (LBBB). Recently developed techniques allow continuous computerized digital analysis of ST segments that can be zeroed to the patient’s own baseline electrocardiogram even if that baseline is abnormal conduction. With use of this technology, ST-segment changes during balloon coronary occlusion were compared in 10 patients with LBBB versus an age-, sex-, and coronary anatomy-matched population of 20 control subjects with normal baseline conduction. ST-segment deviation of I1 mm from baseline was present in 80% of patients with LBBB and in 75% of control patients (difference not significant). There was no significant difference between patients with LBBB versus control patients in maximal ST-segment deviation (2.6 f 1.7 vs 2.0 f 1.0 mm) or in ST-segment deviation measured after 60 seconds of occlusion (2.4 f 1.3 vs 1.8 f 1.1 mm). ST-segment deviation reached 1 mm more quickly in patients with LBBB (33 f 11 seconds) than in control subjects (60 f 36 seconds) (p = 0.003). It is concluded that ST-segment analysis is feasible in patients with LBBB using digital self-referenced ST analysis and may provide important clinical information regarding the presence of myocardial ischemia. (Am J Cardiol 1991;67:1219-1222)
From Georgetown University Hospital, 3800 Reservoir Road, NW, Washington, D.C. Manuscript received May 11, 1990; revised manuscript received January 17, 1991, and accepted January 21. Address for reprints: Karen S. Stark, MD, Washington Cardiology Center, PC, 110 Irving Street, NW 4B-14, Washington, D.C. 20010.
lectrocardiographic ST-segment changes are well-described dynamic markers for myocardial ischemia and infarction. During percutaneous transluminal coronary angioplasty, ST-segment monitoring is used to assessthe degree of ischemia induced by balloon coronary occlusion. This type of monitoring has been useful in evaluating the effects of interventions designed to decrease ischemia and in predicting complications. ST-segment analysis, however, has been largely ignored in the presence of left bundle branch block (LBBB) where the abnormal pattern of repolarization makes absolute ST levels and relative changes in ST level difficult to interpret. Recent advances in monitoring technology allow continuous computerized digital analysis of ST segments in multiple leads simultaneously. Digital electrocardiographic information can be selfreferenced or zeroed to the patient’s own baseline electrocardiogram. This allows precise analysis of dynamic ST-segment shifts despite baseline electrocardiographic abnormalities. This study compares the ST deviation from a self-referenced baseline during coronary occlusion in patients with LBBB to the ST changes in matched control patients with normal baseline conduction.
E
METHODS Patient
population: Nine hundred patients undergoing elective coronary angioplasty between March 5, 1985, and April 26, 1988, underwent ST-segment monitoring. Retrospective review of this database identified 11 patients with LBBB. The criteria for LBBB were: ( 1) conduction originating above the atrioventricular node; (2) QRS duration of 20.14 second; (3) predominantly upright complexes with broad slurred R waves in leads I, V5 and Vg; and (4) QS or rS pattern in V1 with a normal intrinsicoid deflection. Control subjects with no conduction abnormality were then matched in a 2:l ratio to patients with LBBB by age f 5 years, sex, angioplasty artery, and presence or absence of intercoronary collateral vessels. ST-segment monitoring: Over the course of this study, 2 devices were used for ST-segment monitoring.
LEFT
BUNDLE
BRANCH
BLOCK
1219
Initially, a 3-channel precalibrated ST/Halter AM recorder (Stole-Alta 3) was used. This system has been previously validated for low-frequency response and phase shift through the entire record/playback loop when paired with the Marquette 8000 playback computer.’ This playback computer was then interfaced to a DEC PDP 1 l/34 mainframe for playback. Bipolar leads were used to address true anterior (VZ), lateral (Vs) and inferior (aVF) vectors. The second monitoring device used was a 1Zlead real-time portable programmable microprocessor-driven device, the Mortara ELI-ST. This device acquires a standard 12-lead electrocardiogram every 15 to 20 se0 onds. The analog wave forms are digitized and compared with the baseline in real time. Further analysis of the digital information is accomplished using graphic and superimposition software developed on an IBM personal computer. For the purposes of this study only data obtained from leads Vz, Vs and aVF were analyzed. ST-segment trends were generated by averaging the ST-segment level at J plus 60 ms of all normally conducted beats in 20-second intervals (Figure 1). ST-segment analysis: ST baseline was defined as the patient’s ST level before angioplasty set as zero. Significant ischemia was defined as I1 mm ST-segment shift from the baseline in any of 3 leads. Data were recorded for each patient from only the most active lead when at least 1 lead showed changes. The most active lead was defined as the lead with the TCA
Baseline
TCA
greatest absolute peak ST change, with elevation taking priority over depression when these changes coexisted. Inflations lasting from 60 to 120 seconds were analyzed. Peak heart rate was recorded for each inflation analyzed. Three parameters were analyzed. The first, referred to as 1 mm time, is the time after inflation in seconds to reach 1 mm of ST-segment deviation from the baseline. The second, referred to as ST 60, is the absolute STsegment deviation from baseline at 60 seconds after inflation. The third, the peak ST, is the maximal absolute ST-segment deviation from baseline in millimeters. RESULTS The mean age was 64 f 8 years for study patients and 63 f 8 years for control patients; 24 of 33 were men (73%) and 9 of 33 were women (27%). The angioplasty coronary artery was the left anterior descending in 9 of 33 (27%), the circumflex in 12 of 33 (36%) and the right in 12 of 33 (36%). Intercoronary collateral vessels were identified on the preangioplasty angiograms in 18 of 33 patients (55%). Significant ischemia during balloon inflation was documented in 9 of 11 patients with LBBB and in 16 of 22 control patients. In 6 of the 9 patients with LBBB with ischemia, both of the matched control subjects also had ischemia in the same most active lead. In the other 3 LBBB patients with ischemia, only 1 of the 2 control subjects also had ischemia. Mean peak heart rate during balloon inflation for
Recovery
Baseline
TCA
TCA
Recovery
6
H
H
H
t -4
A
-61 ’
3.0
6.0
9.0 12.0 15.0 Time (minutes)
16.0
t -6 1
21.0 24.0
B
’
I 3.0
I 6.0
I I I 9.0 12.0 15.0 Time (minutes)
I 16.0
I 21.0
I 24.0
FIGURE 1. ST-segment Horironfa/ bars indicate aVF (Channel 2) during Mock. TCA = transluminal
ST analysis gives a mean ST level every 15 seconds. trend rewrdad during angioplasty. High-resolution balloon inflations. A similar pattern of ST-segment elevation with balloon inftations is evklent in lead right coronary ocdusion in a patient with (A) normal basetins conduction and (6) left bundle branch coronary angioplasty.
1220
JOURNAL
THE AMERICAN
OF CARDIOLOGY
VOLUME
67
JUNE
1, 1991
the 2 groups was 77 f 19 beats/min for the study patients and 71 f 11 beats/min for control subjects (difference not significant). Mean change in heart rate during inflation was 4 f 3 beats/min for both groups of patients. The peak ST-segment deviation from baseline during balloon inflations was 2.6 f 1.6 mm for the group with LBBB and 2.0 f 1.0 mm for the control population (difference not significant). The time taken to reach 1.0 mm of ST-segment deviation was signficantly less for the LBBB group (36 f 13 seconds) than for the control group (55 f 31 seconds) (p
Electrocardiographic manifestations of ischemia are difficult to interpret in the presence of left bundle branch block (LBBB). Recently developed techniques allow continuous computerized digital analysis of ST segments that can be zeroed to the patient’s own baseline electrocardiogram even if that baseline is abnormal conduction. With use of this technology, ST-segment changes during balloon coronary occlusion were compared in 10 patients with LBBB versus an age-, sex-, and coronary anatomy-matched population of 20 control subjects with normal baseline conduction. ST-segment deviation of I1 mm from baseline was present in 80% of patients with LBBB and in 75% of control patients (difference not significant). There was no significant difference between patients with LBBB versus control patients in maximal ST-segment deviation (2.6 f 1.7 vs 2.0 f 1.0 mm) or in ST-segment deviation measured after 60 seconds of occlusion (2.4 f 1.3 vs 1.8 f 1.1 mm). ST-segment deviation reached 1 mm more quickly in patients with LBBB (33 f 11 seconds) than in control subjects (60 f 36 seconds) (p = 0.003). It is concluded that ST-segment analysis is feasible in patients with LBBB using digital self-referenced ST analysis and may provide important clinical information regarding the presence of myocardial ischemia. (Am J Cardiol 1991;67:1219-1222)
From Georgetown University Hospital, 3800 Reservoir Road, NW, Washington, D.C. Manuscript received May 11, 1990; revised manuscript received January 17, 1991, and accepted January 21. Address for reprints: Karen S. Stark, MD, Washington Cardiology Center, PC, 110 Irving Street, NW 4B-14, Washington, D.C. 20010.
lectrocardiographic ST-segment changes are well-described dynamic markers for myocardial ischemia and infarction. During percutaneous transluminal coronary angioplasty, ST-segment monitoring is used to assessthe degree of ischemia induced by balloon coronary occlusion. This type of monitoring has been useful in evaluating the effects of interventions designed to decrease ischemia and in predicting complications. ST-segment analysis, however, has been largely ignored in the presence of left bundle branch block (LBBB) where the abnormal pattern of repolarization makes absolute ST levels and relative changes in ST level difficult to interpret. Recent advances in monitoring technology allow continuous computerized digital analysis of ST segments in multiple leads simultaneously. Digital electrocardiographic information can be selfreferenced or zeroed to the patient’s own baseline electrocardiogram. This allows precise analysis of dynamic ST-segment shifts despite baseline electrocardiographic abnormalities. This study compares the ST deviation from a self-referenced baseline during coronary occlusion in patients with LBBB to the ST changes in matched control patients with normal baseline conduction.
E
METHODS Patient
population: Nine hundred patients undergoing elective coronary angioplasty between March 5, 1985, and April 26, 1988, underwent ST-segment monitoring. Retrospective review of this database identified 11 patients with LBBB. The criteria for LBBB were: ( 1) conduction originating above the atrioventricular node; (2) QRS duration of 20.14 second; (3) predominantly upright complexes with broad slurred R waves in leads I, V5 and Vg; and (4) QS or rS pattern in V1 with a normal intrinsicoid deflection. Control subjects with no conduction abnormality were then matched in a 2:l ratio to patients with LBBB by age f 5 years, sex, angioplasty artery, and presence or absence of intercoronary collateral vessels. ST-segment monitoring: Over the course of this study, 2 devices were used for ST-segment monitoring.
LEFT
BUNDLE
BRANCH
BLOCK
1219
Initially, a 3-channel precalibrated ST/Halter AM recorder (Stole-Alta 3) was used. This system has been previously validated for low-frequency response and phase shift through the entire record/playback loop when paired with the Marquette 8000 playback computer.’ This playback computer was then interfaced to a DEC PDP 1 l/34 mainframe for playback. Bipolar leads were used to address true anterior (VZ), lateral (Vs) and inferior (aVF) vectors. The second monitoring device used was a 1Zlead real-time portable programmable microprocessor-driven device, the Mortara ELI-ST. This device acquires a standard 12-lead electrocardiogram every 15 to 20 se0 onds. The analog wave forms are digitized and compared with the baseline in real time. Further analysis of the digital information is accomplished using graphic and superimposition software developed on an IBM personal computer. For the purposes of this study only data obtained from leads Vz, Vs and aVF were analyzed. ST-segment trends were generated by averaging the ST-segment level at J plus 60 ms of all normally conducted beats in 20-second intervals (Figure 1). ST-segment analysis: ST baseline was defined as the patient’s ST level before angioplasty set as zero. Significant ischemia was defined as I1 mm ST-segment shift from the baseline in any of 3 leads. Data were recorded for each patient from only the most active lead when at least 1 lead showed changes. The most active lead was defined as the lead with the TCA
Baseline
TCA
greatest absolute peak ST change, with elevation taking priority over depression when these changes coexisted. Inflations lasting from 60 to 120 seconds were analyzed. Peak heart rate was recorded for each inflation analyzed. Three parameters were analyzed. The first, referred to as 1 mm time, is the time after inflation in seconds to reach 1 mm of ST-segment deviation from the baseline. The second, referred to as ST 60, is the absolute STsegment deviation from baseline at 60 seconds after inflation. The third, the peak ST, is the maximal absolute ST-segment deviation from baseline in millimeters. RESULTS The mean age was 64 f 8 years for study patients and 63 f 8 years for control patients; 24 of 33 were men (73%) and 9 of 33 were women (27%). The angioplasty coronary artery was the left anterior descending in 9 of 33 (27%), the circumflex in 12 of 33 (36%) and the right in 12 of 33 (36%). Intercoronary collateral vessels were identified on the preangioplasty angiograms in 18 of 33 patients (55%). Significant ischemia during balloon inflation was documented in 9 of 11 patients with LBBB and in 16 of 22 control patients. In 6 of the 9 patients with LBBB with ischemia, both of the matched control subjects also had ischemia in the same most active lead. In the other 3 LBBB patients with ischemia, only 1 of the 2 control subjects also had ischemia. Mean peak heart rate during balloon inflation for
Recovery
Baseline
TCA
TCA
Recovery
6
H
H
H
t -4
A
-61 ’
3.0
6.0
9.0 12.0 15.0 Time (minutes)
16.0
t -6 1
21.0 24.0
B
’
I 3.0
I 6.0
I I I 9.0 12.0 15.0 Time (minutes)
I 16.0
I 21.0
I 24.0
FIGURE 1. ST-segment Horironfa/ bars indicate aVF (Channel 2) during Mock. TCA = transluminal
ST analysis gives a mean ST level every 15 seconds. trend rewrdad during angioplasty. High-resolution balloon inflations. A similar pattern of ST-segment elevation with balloon inftations is evklent in lead right coronary ocdusion in a patient with (A) normal basetins conduction and (6) left bundle branch coronary angioplasty.
1220
JOURNAL
THE AMERICAN
OF CARDIOLOGY
VOLUME
67
JUNE
1, 1991
the 2 groups was 77 f 19 beats/min for the study patients and 71 f 11 beats/min for control subjects (difference not significant). Mean change in heart rate during inflation was 4 f 3 beats/min for both groups of patients. The peak ST-segment deviation from baseline during balloon inflations was 2.6 f 1.6 mm for the group with LBBB and 2.0 f 1.0 mm for the control population (difference not significant). The time taken to reach 1.0 mm of ST-segment deviation was signficantly less for the LBBB group (36 f 13 seconds) than for the control group (55 f 31 seconds) (p
