Signal averaged ECGs
124
Signal averaged ECGs
SIGNAL AVERAGED ECGS
7 4 5 Quantitative analysis of high-frequency components of
the p-wave in patients with paroxysmal atrial fibrillation
1
M. Villani, G. lannucci, S. lannuzzl, L MacheUa, F. Caruso , G. Baaarello.
Arrhythmia Control Unit, La Saplenza University;1 CanSomecBc S.p a. Rome,
Italy
NX
N
P
NY
NZ
26.0 24.9
±6.4 ±6.9 ±8.3
17.4 2 2 9 24.2
±6.3 ±7 4 ±9.0
0.256 0.227 0.811
FAP 20.0
NT
ABX
ABY
70.9
±16.9
64.6
±18.6
0.309
974 9
±425 8
674.0
±270.6
0.022
1338 1 1285 0
±441 9 ±489 6
889.1 926.5
±319.7 ±370.2
0002 0.024
ABZ
ABT
Pd
RMS20
3431.7
±1399 0
2483.4
±810.1
0 024
112.6
±16.4
88.2
±12.0
00000
2.1
±0.9
32
±1 1
0.02
Our preliminary results demonstrate the presence of LAP and of a great
number of high frequency components in pts wtth PAF. The clinical and
electrophysiological implications of those high frequency signals in the entire
P-wave have to be better Interpreted.
7 4 6 The pre-operatlve signal-averaged P-wave compared to
clinical, standard electrocardlographic or
echocardiographlc variables for prediction of atrial
fibrillation after coronary bypass grafting
P.J. Stafford, S. Kotvekar', J. Cooper, T.J. Spyt', C.J. Garratt Cardiac
Department;1 Cardiothoractc Surgery Department, Glenfleld Hospital,
Leicester, UK
The preoperative signal averaged P wave (SAPW) may be a useful Indicator
of atrial fibrillation (AF) after coronary bypass grafting (CBG), but standard
etectrocardiography (ECG) or echocardiography may be equally predictive. We
therefore prospectivety analysed the predictive value of SAPW compared to
clinical variables, ECG and echocardlogram for AF after CBG.
137 unsetected patients (pts) undergoing first elective CBG were recruited
over 6 months. Age, sex, vessels grafted, cardJothoracIc ratio, and cardloactive
drugs were noted. Pts had P wave specific SAPW, standard ECG and left atrial
diameter from the M mode echocardlogram (LA) within 24 hrs before surgery.
From the SAPW, filtered P wave duration (SAPWD), spatial velocity, and energy
were measured. From the standard ECG, lead II P wave duration, P terminal
force in lead V, (Morris), total P wave duration and isoetectric interval (Buxton)
were calculated. Postoperatively pts had 48 hrs Hotter monitoring and dairy
ECGs. 2 (1.5%) pts died; of the remaining 135, 38(28%) had AF lasting > 1 hr
a mean of 2 (0.5-7) days after CBG. These pts had longer SAPWD (149 ( ± 2)
vs 141 (± 2) ms, p = 0.002) and more mates (95% vs 70%, p = 0.01) than those
remaining ki sinus rhythm. Stepwise logistic regression identified onry SAPWD
as predictive of postoperative AF. A SAPWD of > 143 ms predfcted post CBG
AF with a positive predictive accuracy (+PA) of 4 1 % and -PA of 87%.
In these pts SAPWD was a bettor predictor of AF after CBG man clinical,
standard ECG or LA. This technique may predict groups of pts at risk of AF
after CBG, but its value in individual pts Is likely to be low.
P. Steinbigler, R Haberi, J. Vogel, A. Spiegl, I. SchmOcking, K. Nespithal,
T. BrOggemann1, D. Andresen 1 , G. Steinbeck.1 University of Berlin and
University of Munich, Germany
Post-infarction risk stratification with late potential analysis is still limited by a low
positive predictive value (ppV). We therefore investigated, whether functional
changes of ventricular late potentials (LP) in the Hotter-ECG during periods of
decreased heart rate variability (HRV|) and increased heart rate (HRf) might
improve the results.
Holter-ECG-tapes, recorded In the 2. week after myocardial infarction In 250
patients (P), were divided In 12 segments with 120 min each. The transient
appearance of LP wtth spectral analysis during periods of HRVJ. (SDNN < 20
ms) and during periods of HR-f (A > 20% compared to the following segment
and the segment before) was determined In each segment. All patients were
followed over a period of 12 months. Follow-up revealed 8 P wtth sudden
cardtec death (SCO), 13 patients wtth cardiac, non-sudden death (CD), 8
patients wtth resuscitation from ventricular fibrillation (VF) and 9 patients with
sustained ventricular tachycardia (VT). LP during HRV| were found in 76% of
P with CD and in 88% of P with VT, LP during H R | In 86% of P with SCD (p
< 0.05 compared to asymptomatic patients). Transient LP during HRV| had a
ppV of 53% for CD, 58% for VT and the appearance of LP during HRt 48%
for SCD. Conventional LP-analysIs In short-term recordings at rest revealed LP
with spectral analysis in 7 1 % of P with SCD (ppV only 18%) and in 77% P with
VT (ppV 10%), with the Simson-method in 69% of P with CD (ppV 19%) (p <
0.05 compared to asymptomatic patients)
Thus, functional LP-anatysis In the Holter-ECG compared to conventional
LP-analysi8 in short-term recordings Improves post-infarction risk-stratification.
748 Correlation of QT dispersion and late potentials and
inducibiltty at the electrophyslology laboratory
A.H. Madrid, C. Mora, A. Chadli, E. Garcia, P. Catalan, L Alonso.. Arrhythmia
Unit, Ramdn y Cajal Hospital, AicalA de Henares University, Madrid, Spain
QT dispersion has been proposed as a measure of repolartzatJon heterogeneity
and may be a marker of an Increased risk of ventricular arrhythmias. We tested
the hypothesis that Increased repdarlzatton dispersion might be a useful
prognostic parameter in Identifying patients with presence of late potentials and
induclbility of ventricular tachycardia at the EP laboratory.
Clinical evaluation, angtographic studies, PES, Hotter recordings, 12 lead
ECG and signal averaged ECG were performed In 60 consecutive patients,
with clinical history of AVM (VT or VF). Patients were in sinus rhythm with no
bundle branch block. They were not on drugs that Influence the QT Interval.
QT dispersion, defined as the difference between maximal and minimal QT
measured in the 12 leads of the surface ECG, was assessed by two physicians
blinded to the results.
Results: thirty-nine of the 60 patients were Inducible (65%). SAECG was +
In 44 / 60 patients (75%).
Or. I SMVT (39 pt)
Or. II NSVT (9 pt)
Gr. IIINotnd. (12pt)
75
SAECG+
75±26
66 ±18*
80 ±14-
35/39(89%)
6/9 (66%)
3/12(25%)
*p < 0 05 vs group I
Conclusions: The patients with VT had significantly longer QT dispersion. VT
inducfbSity was predicted by abnormal SAECG (sensitivity 89%) and Increased
QT dispersion In 12 lead ECG > 70 mseg, (sensitivity 74%). Combination of
abnormal SAECG and increased QT dispersion gave an overall accuracy of
96% and relative odds of 9:1 in predkflng the VT Inducibility. SAECG combined
wtth QT dispersion gives the best accuracy In predicting the inducibility to VT.
In previous papers we have demonstrated, by means of trie time domain
analysis (TD) with the signal averaging electrocardiogram (SAE), the presence
of high frequency late atrial potentials (LAP), probably substrate of a reentrant
circuit, in patients (pts) with paroxysmal atrial fibrillation (PAF).
The purpose of the present study was to analyse the number of components
of high frequency signa (...truncated)
