Rate Variability After icute Myocardial Infarction With Normal and Reduced Left Ventricular Ejection Fraction Federico Lombardi, MD, Giulia Sandrone, MD, Andrea Mortara, MD, Daniela Torzillo, MD, Maria Teresa La Rovere, MD, Maria Gabriella Signorini, BS, Sergio Cerutti, BS, and Albert0 Malliani, MD We analyzed heart rate variability (HRV) in 2 groups of patients after acute myocardial infarction with normal and reduced ejection fraction (EF) by considering both the power of the 2 major harmonic components at low and high frequency and 2 indexes of nonlinear dynam- ics, namely the 1 /f slope and the correlation dimension Dz. HRV of patients with a reduced EF was characterized by a diminished RR variance as well as a different dis- tribution of the residual power in all frequency ranges, with lower values of the low-frequency component ex- pressed in both absolute and normalized units, and of the low- to high-frequency ratio. In these patients we also observed a steeper slope of the negative regression line between power and frequency in the very low fre- quency range. The presence of a smaller fractal dimen- sion was suggested by a lower Dz. Thus, in patients after acute myocardial infarction with a reduced EF, the re- duction in HRV is associated with a different distribution of the residual power in the entire frequency range, which suggests a diminished responsiveness of sinus node to neural modulatory inputs. (Am J Cardiol 1996;77: 1283- 1288) A reduction in heart rate variability (HRV) has been consistently reported after acute myocar- dial infarction (AMI) and generally interpreted as the result of an alteration of autonomic modulation of sinus node (i.e., increased sympathetic and di- minished vagal activity) .I-3 Despite the negative prognostic value of a reduced HRV after AMI, ‘-’ interpretation of the mechanisms that modify HRV remains elusive, especially in patients with a reduced ejection fraction (EF) . It is generally accepted that sympathetic and parasympathetic influences on sinus node are reflected by 2 major rhythmic oscillations (at relatively low and high frequencies) that char- acterize the spectral profile >0.03 Hz.~-’ When con- sidering long-term recordings, it is evident’ that most of the variance is distributed below 0.03 Hz (i.e., in a frequency range of unsettled interpreta- tion), which requires a different mathematic ap- proach to analyze it appropriately.8-” The present study analyzed HRV in 2 groups of patients after AM1 with normal and reduced EF, by considering both the power of 2 major harmonic components measured on short-term recordings and 2 indexes of nonlinear dynamics8-14 namely, the l/f slope and the correlation dimension D2 computed on long-term recordings corresponding to a 6-hour period. From Centro Ricerche Cardiovoscolarr, CNR, Medicina lnterna II, Ospedale Sacco, Universita dr Mifano; Drvisione dr Cardiologia, Centro Medico Montescano, Fondozione Clinrca Del lavoro Pavia; and Dipartrmento di Bloingegneria, Politecnico di Miiano, Mrlan, Italy Manuscript recerved November 1 3, 1995; revised manuscript received and accepted February 1, 1996. Address for rearints. Federico tombardi. MD. Medicrna Interno. Osp L Sacco, University of Milan, vra G.B. Grossi 74, 2015? Mifano, Italy 0 1996 by Excerpta Medica, Inc. Ail rights reserved. METHODS We studied 35 male patients (mean age 53 +- 3 years, range 40 to 69) who were admitted to the Montescano Cardiac Rehabilitation Centre after a first AMI. Only patients ~70 years of age, with a low frequency of ventricular arrhythmias ( < 10 ven- tricular premature complexes/hour) and no clinical signs of peripheral neuropathies, were considered el- igible for the study. The site of AM1 was anterior in 20 and inferior in 15 cases. In all patients, 2-dimen- sional echocardiography was performed to calculate EF according to the area-length technique. Patients were then dichotomized into 2 groups by using a cutoff value of 40%. Accordingly, normal (54 f- 2%, range 40% to 62%) and reduced (32 2 3%, range 17% to 39% ) EFs were present in 20 and 15 patients, respectively. The 2 groups did not differ in terms of mean age and site of infarction. All patients were in pharmacologic washout (24-hour withdrawal from nitrates and nifedipine, and 5-day withdrawal from /I blockers, diltiazem, and angiotensin-converting enzyme inhibitors) except for 4 patients with low EF, who continued to receive lisinopril, nitrates, and digoxin (2 cases). All subjects gave informed written consent to the study, which was approved by the Committee for the Protection of Human Subjects from the Risk From Research at the Fondazione Clinica de1 Lavoro, Pa- via, Italy. In all patients, 24-hour Holter monitoring was performed 32 i- 2 days after AMI. After connection to a 2-channel Holter recorder (model 465; Del Mar Avionics, Irvine, California), patients were asked to lie comfortably on a bed in a quiet room for 15 minutes and then to adhere strictly to a normal daily routine consisting of meals and waking and noctur- 0002.9 149/96/$15 00 1283 PII SOOO2-9149(96)00193-2