M. Fioranelli, P. Azzolini, G.M.
Mileto, F. Sgreccia, G. Speciale, M.P. Risa, C. Peraldo Neja, R. Ricci, A. Puglisi.
Division of Cardiology, Fatebenefratelli Hospital, Tiberina Island, Rome, Italy
|
|
Background. Mortality after
myocardial infarction may be due to arrhythmic events, reinfarction, progressive heart
failure and non-cardiac cause.
Various experimental and clinical observations suggest changes in sympathetic and
vagal neural regulatory mechanisms play a critical role in altering cardiac electrical
properties and favor the occurrence of arrhythmic events.
Aim of this study was to investigate the role of heart rate variability, ejection
fraction and late potentials in risk stratification after a MI.
Methods. We analyzed the outcome of 53 patients after a miocardial
infarction (MI); 34 pts (64%) had an anterior MI and in 35 pts (66%) was performed
thrombolysis. We recorded a predischarge Ecg Holter monitoring, an echocardiographic
ejection fraction (EF), and a signal averaging ecg (SAECG). We grouped the patients as
follows: Group 1 those free from cardiovascular events at the end of follow-up; Group 2
with a cardiovascular event such as angor, by-pass, ptca; Group 3 patients with sudden
death.
For statistical analysis we used analysis of variance for numeric variables,
chi-squared analysis for non numeric variables, and t test for analysis between two
groups.
Results. At the end of follow-up (mean 26 ± 8 months), 26% of
patients were free from cardiovascular events (Group 1), 64% had a cardiovascular event
such as angor, by-pass, ptca (Group 2), 5 patients died (9%) (Group 3), 4 from sudden
death, 1 of cardiac death.
EF in group 3 (with a death in follow-up) was lower than that in group 2 (with a
cardiovascular event) and than that in group 1 (without a cardiovascular event) (F 2.50 =
5.39, p = 0.008).
Late potentials (positive for three criteria) were present in 10 pts (18%); 87% with
negative LP were alive at the end of follow-up, whereas 80% of the patients that died had
positive late potential: the ratio alive/death has a chi-squared of 13.15 with a p of
0.001.
Analysis of variance of total power, a HRV frequency domain parameter, demonstrated a
clear difference of values in free-event patients against the event group and the death
group (F 2.50 = 1.93, p = 0.13); when we compare alive and death group of patients, the t
test gives a very strong statistical significance (t = 2.63; p = 0.01).
Standard deviation, a time domain parameter of HRV, shows a clear statistical
difference between alive and death group of patients (t = 2.26, p = 0.024).
We performed analysis of sensitivity, specificity and positive predictive accuracy for
death of three parameters, HRV expressed such TP, presence of LP and EF < 40%, alone
and in combination each other; with a depressed HRV and presence of LP we have the best
sensitivity (60%), the best specificity (93%), and the best positive predictive accuracy
for death (50%).
Conclusion. A depressed HRV is a potent and indipendent predictor of
cardiovascular events after myocardial infarction; the prognostic power of HRV is
increased when associated to positive late potentials and depressed ejection fraction.
Signal-averaging is only able to detect an arrhythmogenic substrate that is
permanently present and may express a propensity to sustained ventricular tachycardia.
However the mechanism underlying the occurrence of sudden cardiac death is multifactorial
and may often only be transiently present like a neurovegetative modulation.
|
|
Signal averaged ECG
acute myocardial infarction, heart rate variability, left ventricular ejection fraction,
risk stratification, OA
Heart rate variability
acute myocardial infarction, late potentials, left ventricular ejection fraction, risk
stratification, OA
|
