GRADUAL RECOVERY OF IMPAIRED CARDIAC AUTONOMIC BALANCE
WITHIN FIRST SIX MONTHS AFTER ISCHEMIC CEREBRAL STROKE
Nenad Lakusic, M.D., MSc1, Darija Mahovic, M.D., MSc2, Tomislav Babic, M.D., Ph.D.,
Department of Cardiology, Hospital for Medical Rehabilitation, Krapinske Toplice,
Department of Neurology, University Hospital Center Zagreb, Zagreb, Croatia
Running Title: CARDIAC AUTONOMIC BALANCE AFTER STROKE
Correspondence to Nenad Lakusic, M.D., MSc, Department of Cardiology, Hospital for
Medical Rehabilitation, Gajeva 2, HR – 49 217 Krapinske Toplice, Croatia
Fax: + 385 49 23 21 40
Tel: + 385 49 23 21 22
e-mail: [email protected]
Abstract (225 words)
Background: The level of autonomic dysbalance in the first months after acute
ischemic cerebral stroke has not been thoroughly investigated, and the available data
are uncomplete. The aim of this research is to establish the degree and dynamics of
impaired cardiac autonomic balance recovery within the first six months following the
acute ischemic cerebral stroke.
Methods: This prospective study included 78 patients who had suffered the first
ischemic cerebral stroke and 78 sex and age-matched healthy subjects. We have
analyzed heart rate variability (HRV) from a 24-hour Holter ECG. In the group of
patients with ischemic cerebral stroke, HRV was measured after two and six months
following the acute phase, respectively.
Results: Two and six months after the acute ischemic cerebral stroke, all HRV
variables, except low to high frequency ratio (LF/HF), were significantly lower in the
group of stroke patients when compared to the control group. Furthermore, we found
a significant increase in the overall HRV between months 2 and 6 after the acute
phase of cerebral stroke; p=0.03 for Standard deviation of all normal R-R intervals
(SDNN) and p=0.01 for Total power.
Conclusions: The results point to the gradual recovery of impaired cardiac autonomic
balance in the patients with ischemic cerebral stroke within the first months following
the acute phase. Nevertheless, HRV remains significantly lower even six months after
the acute phase in comparison to healthy subjects.
Key Words: heart rate; autonomic nervous system; cerebral stroke; ischemic
Like an acute myocardial infarction (MI) , acute ischemic cerebral stroke leads to
autonomic dysbalance and lowered heart rate variability (HRV) . The mechanism
of HRV decrease in the acute ischemic cerebral stroke is the damage of brain
structures located in the insular cortex that are modulating the autonomic heart
activity [4,16,18]. The level of autonomic dysbalance in the first months following
ischemic cerebral stroke has not been thoroughly investigated, and available literature
data are uncomplete .
The aim of this research is to establish the degree and dynamics of impaired cardiac
autonomic balance recovery within the first six months following the acute ischemic
Patients and methods
This prospective study included 78 consecutive patients who had suffered the first
ischemic cerebral stroke (53 male and 25 female, mean age 59 ± 11 year) within 10
weeks after the acute phase during rehabilitation treatment, and 78 age and sexmatched healthy subjects.
Inclusion criteria were: age under 70 years, ischemic hemispheric lesions verified by a
CT scan (52% with the left and 48% with the right hemispheric infarction), and ECG
Exclusion criteria were: hemorrhagic stroke, atrial fibrillation, sick sinus syndrome,
AV block of II or III degree, previous MI, percutaneous coronary intervention,
coronary artery bypass grafting (CABG), diabetes mellitus, heart failure and betaadrenergic blockers or antiarrhythmic drugs in therapy.
Cardiac autonomic balance was evaluated by analysis of HRV. All HRV variables
were measured through the 23.2-hour period (ranged 21 to 24 hours). Ambulatory
ECG recordings were made by 3-channel Medilog Digital Holter recorders FD3,
Oxford, with 1024 Hz resolution. HRV was analyzed by computer and over-read
manually. A commercial system (Oxford Instruments, with software Medilog Holter
Management System Excel 2, Version 7.1) was used. Algorithms for arrhythmia
analysis gave a label to each QRS complex. An operator cleaned all recordings from
artefacts, reviewed beats and modified them if needed, under the cardiologist
supervision. Only recordings with less than 15% of ectopic beats were used. Periods
with the highest and lowest average R-R intervals, detected from R-R interval
histograms, were always validated. The corrected data were processed and HRV was
computed. Raw tachogram was used for time domain analysis. The power spectral
analysis was computed using fast Fourier transformation. R-R intervals that included
ectopic beats were excluded and extrapolated by linear interpolation for the spectral
analysis. Details have been published elsewhere .
In the group of patients with ischemic stroke, HRV was measured two (58 ± 6 days)
and six months (178 ± 12 days) after the acute phase and once in healthy subjects.
Most of the variables proposed by the Task Force on the HRV  were analyzed.
Time domain analysis included: Mean RR - mean of R-R intervals for normal beats,
SDNN - standard deviation of all normal R-R intervals, SDANN-i - standard
deviation of the 5-minute means of R-R intervals, SDNN-i - mean of the 5-minute
standard deviations of RR intervals, rMSSD - square root of the mean of the squared
successive differences in R-R intervals and pNN50 - percentage of R-R intervals that
are at least 50 ms different from the previous interval. Frequency domain analysis
covered: TP - Total power (0.0-0.5 Hz), VLF - very low (0.003-0.04 Hz), LF - low
(0.04-0.15 Hz) and HF - high (0.15-0.4 Hz) frequency components, with LF/HF low to high frequency ratio. LF and HF variables are expressed in ms2, as well as in
normalised units (n. u.), which is calculated using the formula: LF or HF norm / (TP –
VLF) x 100 .
Statistical package Microsoft SPSS for Windows, Version 10.0 was used. The results
are expressed by mean values and standard deviations. The normality of distribution
of the variables was tested using the Kolmogorov – Smirnov test. Differences in HRV
were tested by analysis of variance (ANOVA) with post hoc comparison using the
Tukey test. The value p < 0.05 is considered statistically significant.
HRV was analyzed in 74 (95%) out of total of 78 included patients six months after
the acute phase. This was due to the fact that in the follow-up period one patient had
died, one patient had a new stroke and two patients had been excluded due to atrial
After two months from the acute ischemic cerebral stroke, all HRV variables, except
LF/HF ratio, were significantly lower in the group of stroke patients compared to the
control group: p=0.02 for Mean RR; p<0.001 for SDNN, SDANN-i, SDNN-i, pNN50,
TP and LF norm; p=0.006 for rMSSD; p=0.01 for VLF; p=0.001 for LF and HF
norm; and p=0.008 for HF.
Furthermore, in stroke patients, all HRV variables, except LF/HF ratio, remained
significantly lower six months after the acute phase in comparison to healthy control
subjects: p=0.03 for Mean RR, VLF and HF norm; p=0.004 for SDNN, pNN50;
p=0.002 for SDANN-i and TP; p=0.001 for SDNN-i; p=0.01 for rMSSD and LF
norm; p=0.008 for LF and p=0.02 for HF.
Moreover, an increase in all HRV parameters was found between the second and sixth
month after the acute ischemic cerebral stroke, and significant differences were found
in the following variables: p=0.03 for SDNN, SDNN-i and LF norm; p=0.01 for TP;
p=0.02 for SDANN-i and LF (Table 1.).
The obtained results point to the gradual recovery of impaired cardiac autonomic
balance within the first months after ischemic cerebral stroke, but HRV remains
significantly lower even six months after the acute phase when compared to healthy
controls. It is interesting that the LF/HF ratio didn’t show any change whereas TP was
reduced. In addition, the time domain measures of parasympathetic regulation were
reduced compared to controls. Although is LF/HF ratio generally accepted as measure
of sympatho-vagal balance , LF/HF ratio might fail to describe sympatho-vagal
balance in a clinical practice, particulary when long term Holter recodings are analysed
. It seems that LF/HF ratio correctly present sympatho-vagal balance in healthy
subjects, while LF/HF ratio is useless in patients with seriously decreased overall HRV
and sympathetic overactivity [11,19]. By a decrease in overall HRV, TP lose signal in
LF and HF and most of the residual energy is distributed in lower frequencies, explicitly
in the VLF band . Furthermore, the possible explanation for limited benefit from
LF/HF ratio in the analysis of sympatho-vagal balance in cardiac, as well as in stroke
patients, is that disease progression shifts HRV spectra leftward . These findings
explain why we didn’t found significant changes between stroke patients and healthy
controls when we analyzed LF/HF ratio. Regarding to that, we conclude that within first
six months after the acute phase of disease, in patients who had suffered ischemic
cerebral stroke there exist lower parasympathetic modulation activities and a higher
sympathetic tone when compared to healthy persons of the same age.
Previously, a similar investigation in a markedly smaller sample was performed by
Korpelainen JT et al. . Although Korpelainen JT et al. reported that abnormal HRV
persisted up to six months after hemispheric infarction, their results were contradictory
to the results of our investigation. Bearing in mind that Korpelainen et al. measured
HRV from a 24-hour Holter ECG, it is difficult to explain the low values of SDNN even
six months after stroke, the values of which where within the borders of pathological
HRV. Besides the mean SDNN value of 46 ms in the acute phase, or 51 ms six months
after stroke, the measured mean values of TP were 1936 ms2, or 2336 ms2, while Mean
RR was 896 ms, or 935 ms six months after stroke , what is very doubtful according
to our experience and to results of this study. Such low SDNN values cannot be related
to normal TP values, what on the other hand correlates well with SDNN values in the
time domain . Furthermore, the mean SDNN values (68 ms) measured in the control
group of healthy subjects are difficult to accept in comparison to the high values of TP
(4772 ms2). In our previous studies, we have analyzed HRV on a sample of more than
2500 patients and we have offered standards for HRV for different groups of cardiac
patients . In this study, SDNN value considered normal for the “general
cardiological population” was 93 ms, the cut point for pathologically decreased SDNN
was 59 ms, and the cut point for normal TP was 1312 ms2. If we follow the analogy
from the HRV analysis in patients after MI published by Bigger JT. et al. , it is to be
expected that there would be at least partial recovery from diminished levels of HRV in
the first months after the acute phase of cerebral stroke and that trend we found in our
The cognition that cerebral infarction causes autonomic dysbalance is not only of
academic and scientific character, but it also has clinical implications. Therefore, each
patient should be monitored during the acute phase of stroke and it would be ideal if
the degree of the autonomic dysbalance could be quantified by the HRV analysis. By
that screening, we believe that could be possible to identify patients with high risk of
malignant arrhythmias and sudden death. Keeping in mind that some groups of
antihypertensives and antiarrhythmic drugs have a favourable effect on the HRV
[7,9,12], by selection of medications in those patients we could try to affect on the
improvement of autonomic cardiac activity or lowering the autonomic dysbalance,
respectively. This finally could reduce the early and late mortality rate from stroke.
A possible limitation of this research is the fact that HRV had not been
measured in the first days of the disease. However, the aim of this investigation was
to analyze HRV after the acute phase. The severe exclusion criteria from the study
were been necessary to eliminated the impact of other disease and condition like MI
, CABG , diabetes mellitus , etc., on HRV. Further investigations should be
designed to follow up patients with ischemic stroke from the acute phase of the
disease. HRV should be measured in the first days, and also more times up to a year
after the acute phase in order to establish in what degree and in which period recovery
of impaired cardiac autonomic balance occurs, and also whether it comes to its
1. Bernardi L, Ricordi L, Lazzari P, Solda P, Calciati A. et al. Impaired circadian
modulation of sympathovagal activity in diabetes: A possible explanation for
altered temporal onset of cardiovascular disease. Circulation 1992; 86: 1443 - 52.
2. Bigger JT, Fleiss JL, Steinman RC, Rolnitzky LM, Kleiger RE. et al. Frequency
domain measures of heart period variability and mortality after myocardial
infarction. Circulation 1992; 85: 164 - 71.
3. Bigger JT Jr, Fleiss JL, Rolnitzky LM, Steinman RC, Schneider WJ. Time course
of recovery of heart rate variability after myocardial infarction. J Am Coll Cardiol
1991; 18: 1643 - 9.
4. Cheung RT, Hachinski VC. The insula and cerebrogenic sudden death. Arch
Neurol 2000; 57: 1685 – 88.
5. Demirel S, Tukek T, Akkaya V, Atilgan D, Ozcan M. et al. Heart rate variability
after coronary artery bypass grafting. Am J Cardiol 1999; 84: 496 - 7.
6. Kleiger RE, Miller JP, Bigger JT, Moss AJ, and the Multicenter Post-Infarction
Research Group: Decreased heart rate variability and its association with
increased mortality after acute myocardial infarction. Am J Cardiol 1987; 59: 256
7. Kontopoulos AG, Athyros VG, Papageorgiou AA, Skeberis VM, Basayiannis EC.
et al. Effect of angiotensin-converting enzyme inhibitors on the power spectrum of
heart rate variability in post-myocardial infarction patients. Coron Artery Dis
1997; 8: 517 – 24.
8. Korpelainen JT, Sotaniemi KA, Huikuri HV, Vilho VV. Abnormal heart rate
variability as a manifestation of autonomic dysfunction in hemispheric brain
infarction. Stroke 1996; 27: 2059 - 63.
9. Lampert R, Ickovics JR, Viscoli CJ, Horwitz RI, Lee FA. Effects of propranolol
on recovery of heart rate variability following acute myocardial infarction and
relation to outcome in the Beta-Blocker Heart Attack Trial. Am J Cardiol 2003;
91: 137 – 42.
10. Lombardi F, Sandrone G, Mortara A, Torzillo D, La Rovere MT. et al. Linear and
nonlinear dynamics of heart rate variability after acute myocardial infarction with
normal and reduced left ventricular ejection fraction. Am J Cardiol 1996; 77: 1283
11. Lombardi F. Chaos theory, heart rate variability and arrhythmic mortality.
Circulation 2000: 101: 8 - 10.
12. Malik M, Camm AJ, Janse MJ, Julian DG, Frangin GA. et al. Depressed heart
rate variability identifies postinfarction patients who might benefit from
prophylactic treatment with amiodarone: a substudy of EMIAT (The European
Myocardial Infarct Amiodarone Trial). J Am Coll Cardiol 2000; 35: 1263 – 75.
13. Milicevic G, Lakusic N, Szirovicza L, Cerovec D, Majsec M. Different cut-points
of decreased heart rate variability for different groups of cardiac patients. J
Cardiovasc Risk 2001; 8: 93 - 102.
14. Milicevic G, Lakusic N, Majsec M. Disease progression shifts heart rate
variability spectra leftward. Europace 2002; 3 (Suppl. A): A202.
15. Orlandi G, Fanucchi S, Strata G, Pataleo L, Landucci Pellegrini L. et al. Transient
autonomic nervous system dysfunction during hyperacute stroke. Acta Neurol
Scand 2000; 102: 317 – 21.
16. Sander D, Klingelhofer J. Stroke-associated pathological sympathetic activation
related to size of infarction and extent of insular demage. Cerebrovasc Dis 1995;
5: 381 – 5.
17. Task Force of the European Society of Cardiology and The North American
Society of Pacing and Electrophysiology: Heart rate variability. Standards of
measurement, physiological interpretation, and clinical use. Eur Heart J 1996; 17:
354 - 81.
18. Tokgozoglu SL, Batur MK, Topcuoglu MA, Saribas O, Kes S. et al. Effects of
stroke localisation on cardiac autonomic balance and sudden death. Stroke 1999;
30: 1307 – 11.
19. Van de Borne P, Montano N, Pagani M, Oren R, Somers VK. Absence of low –
frequency variability of sympathetic nerve activity in severe heart failure.
Circulation 1997: 95: 1449 - 54.
Table 1. Values of heart rate variability (HRV) in the patients who suffered ischemic
cerebral stroke two and six months after acute phase of disease and control group of
Two months after
Six months after
867 ± 102
875 ± 103
892 ± 117
96 ± 21
107 ± 24
136 ± 31
84 ± 22
92 ± 26
122 ± 20
42 ± 12
47 ± 13
55 ± 16
23 ± 16
27 ± 18
31 ± 18
4.5 ± 3.9
6.0 ± 4.8
8.0 ± 6.4
1989 ± 1305
2638 ± 1617
3968 ± 2857
1568 ± 642
1786 ± 745
2134 ± 979
362 ± 291
576 ± 316
792 ± 485
LF norm (n.u.)
48.2 ± 19.7
59.3 ± 24.5
68.1 ± 28.4
154 ± 126
212 ± 107
309 ± 234
HF norm (n.u.)
20.1 ± 8.6
22.8 ± 10.1
26.2 ± 11.7
2.4 ± 1.4
2.6 ± 1.3
2.6 ± 1.6
Mean RR (ms)
Mean RR - mean of R-R intervals for normal beats, SDNN - standard deviation of all
normal R-R intervals, SDANN-i - standard deviation of the 5-minute means of R-R
intervals, SDNN-i - mean of the 5-minute standard deviations of RR intervals, rMSSD square root of the mean of the squared successive differences in R-R intervals and pNN50 percentage of R-R intervals that are at least 50 ms different from the previous interval, TP total power (0.0-0.5 Hz), VLF - very low (0.003-0.04 Hz), LF - low (0.04-0.15 Hz), HF high (0.15-0.4 Hz) frequency components, LF/HF - low to high frequency ratio