Spontaneous Oscillations in Cerebral Blood Flow Velocities in Middle Cerebral Arteries in Control Subjects and Patients With Epilepsy
Background and Purpose Cardiac arrhythmias mediated by the sympathetic nervous system have been implicated in sudden, unexplained deaths in patients with epilepsy. Cerebral blood flow velocities (CBFV) as measured by transcranial Doppler are characterized by slow spontaneous oscillations in part attributed to changes in sympathetic activity (M waves, 3 to 9 cycles per minute) and to discharges of monoaminergic neurons in the brain stem (B waves, 0.5 to 2 cycles per minute). This study was designed to compare spontaneous fluctuations of CBFV in patients with epilepsy with those in normal control subjects .
Methods Simultaneous registrations of scalp electroencephalograms, with electrodes placed according to the 10–20 System, and transcranial Doppler recordings of both middle cerebral arteries were performed in 27 patients (9 with primary generalized epilepsy, 18 with focal epilepsy). Data analysis of CBFV was based on the envelope curves of the Doppler spectrum. A fast Fourier transformation over the 20-minute CBFV curve was performed, and the amplitudes of B and M waves were calculated and compared with those in 20 normal, age-matched control subjects.
Results While the amplitudes of the B waves in both groups were similar, patients with epilepsy showed significantly increased M waves. Patients with focal epilepsy did not present asymmetries between the normal hemisphere and the side of the epileptic focus with respect to both M and B waves.
Conclusions Enhanced M waves in epileptic patients may reflect increased sympathetic activity even in the absence of seizures. This study provides further evidence for an autonomic dysfunction as a possible mechanism for sudden unexplained death in patients with epilepsy.
Sudden unexplained death is a significant cause of mortality in patients with epilepsy. Estimated rates vary from 1.5 to 9 in 1000 patient-years.1 2 The etiologic factor discussed is ictal cardiac arrhythmia suggestive of autonomic dysfunction. During partial seizures, tachycardias have been observed. Epstein et al3 studied the neuroanatomical correlates of ictal tachycardias in patients with epilepsy; the amount of change in cardiac frequency depended on the volume of the cerebral structures involved in epileptogenic discharges. By contrast, others4 5 have observed bradycardias during seizures.
Blood pressure, heart rate, and cerebral blood flow velocities (CBFV) are never entirely constant over time but show slow oscillations occurring at various frequencies. M waves are oscillations of a frequency of 3 to 9/min. Simultaneous recordings of sympathetic nerve activity and arterial blood pressure have shown that the M waves of the systemic arterial blood pressure correlate with discharges of sympathetic neurons, which in turn account for the 3 to 9/min CBFV variations in the cerebral vasculature.6 B waves were first described by Lundberg7 in 1960 as rhythmic spontaneous 0.5 to 2/min oscillations of intracranial pressure in patients with acute brain lesions. Changes in the simultaneously measured CBFV were in the same frequency range.8 These phenomena are also seen in healthy subjects.9 10 An autonomic brain stem rhythm is postulated as the direct pacemaker of these oscillations in intracranial pressure and cardiovascular oscillations.11
Because of the close correlation of autonomic nervous function and changes in CBFV evaluated by transcranial Doppler sonography, we investigated whether spontaneous oscillations in CBFV differed in patients with epilepsy from those in normal control subjects. If so, this would be further evidence for autonomic nervous dysfunction in epileptic patients.
Subjects and Methods
Twenty-seven patients with epilepsy (18 women, 9 men; age, 29±8.4 years) were included in the study. Nine had primary generalized epilepsy, and 18 had focal epilepsy of temporal lobe semiology. The mean duration of the epilepsy was 14.6±11.9 years, and the mean seizure frequency at the time of the study was 5.8±6.2 seizures per month. All patients had normal brain MRI studies. Five of the patients with primary generalized epilepsy were on valproic acid, and 2 were on combinations of valproic acid and primidone or carbamazepine, respectively. Two were on no medication at the time of the study. Nine of the patients with focal epilepsy were on monotherapy with carbamazepine, 7 were on various combinations including phenytoin, lamotrigine, gabapentin, and topiramate, and 2 were on no medication. All drug levels were within the therapeutic range. The patients were on no other drugs except for the anticonvulsants. All patients had been seizure free for at least 48 hours. The control population consisted of 20 age-matched healthy volunteers (10 women and 10 men). Patients and control subjects gave informed consent before inclusion in the study.
The recordings took place in a quiet, dark room with the patients resting comfortably in a chair in a reclined position with closed eyes. All recordings were made during evening hours.
Registrations of the scalp electroencephalograms with electrodes placed according to the 10–20 System and transcranial Doppler recordings of both middle cerebral arteries were performed simultaneously by use of a 2-MHz transcranial Doppler device (TCD, DWL). After identification of the MCA signal at an insonation depth of 50 mm,12 the probes were adjusted and mechanically fixed with a specially developed probe holder that does not impede electrode placement. In 4 of the 27 patients, only unilateral registration of the Doppler signal was possible because of an insufficient bone window. Data analysis of CBFV was based on the envelope curves of the Doppler spectrum, which were subsequently transformed into percentage units. A fast Fourier transformation over the 20-minute CBFV curve was performed, showing the time sequence of CBFV deviations around its mean. The amplitudes of oscillations in CBFV were indicated as percent change relative to mean CBFV, the so-called coefficient of variation. Amplitudes of B and M waves in epileptic patients were calculated and compared with those in the 20 control subjects.
To assess whether there was any left/right desynchronization of the various waves, the coherence was also calculated by cross-spectral analysis.
The Mann-Whitney U test for paired data was used to assess statistical significance of differences in one group. The Mann-Whitney U test for unpaired samples was used to compare different groups.
Although the amplitudes of the B waves were similar in patients with epilepsy and in control subjects, patients showed significantly increased M waves compared with normal subjects (P<.01, Figure⇓). Table 1⇓ shows the covariance and coherence results of M and B waves in both groups.
In all patients with focal epilepsy, no hemispheric asymmetries were detected. At the time of the recordings, 14 of the 18 patients with focal epilepsy had a clear lateralizing epileptogenic focus: 9 foci were located on the left, 5 on the right. Even in these patients, no asymmetries between the side of the epileptogenic focus and the normal hemisphere were found with respect to M or B waves, respectively.
All of the 9 patients with primary generalized epilepsy had paroxysmal interictal generalized epileptic activity during the recordings. No hemispheric asymmetry was found. There were no significant differences with respect to M or B waves between patients with primary generalized and focal epilepsy (Table 2⇓).
CBFV are known to be influenced by autonomic activity, thus causing slow oscillations in the M- and B-wave frequencies.
An autonomic brain stem rhythm has been postulated as the direct pacemaker of intracranial pressure and cardiovascular oscillations of the B-wave frequency.11 This theory was further supported by animal experiments showing that neuronal discharges of the locus caeruleus and the raphe nuclei parallel changes in intracranial pressure.13 Given that monoaminergic neurons were involved in the generation of B waves, we hypothesized B-wave abnormalities in epileptic patients as an indicator of autonomic dysfunction. Our results did not confirm this hypothesis. Significant differences in B-wave frequencies were not found.
There was, however, a considerable increase in M waves of CBFV in patients with epilepsy, thus supporting the theory of an autonomic dysfunction in these patients due to an increased sympathetic activity. The fact that none of the patients had seizures at the time of the recording strongly indicates that these changes in sympathetic activity reflect a characteristic feature also present interictally. Interestingly, the activity of the epileptogenic focus did not cause any significant difference with respect to M and B waves.
Several authors have also described an abnormal interictal heart rate variability reflecting autonomic nervous dysfunction. Frysinger et al14 thought that the fluctuations in heart rate in patients with epilepsy reflected an exaggerated Mayer wave frequency.
Devinsky et al15 demonstrated normal interictal heart rate and blood pressure with significantly greater variance, however, during orthostasis and two other autonomic tests. To differentiate the effects of medication and disease, these authors also tested normal control subjects after they had received carbamazepine. Only part of the greater variability could be attributed to carbamazepine. Structural lesions, effects of chronic epilepsy such as hippocampal neuronal loss, and increased inhibition were also considered as the cause of autonomic dysfunction. In our study most of the patients had long-standing epilepsy and received antiepileptic drugs. The number of patients without medication was too small for a statistically viable comparison. Future studies should address this issue.
Severe or intractable epilepsy has been considered a risk factor for sudden unexplained death.16 Another interesting aspect requiring further investigation is whether the duration and severity of the epilepsy also have an impact on the aforementioned interictal autonomic abnormalities.
Some controversy exists about whether there is a hemispheric difference regarding autonomic changes. The types of cardiac arrhythmias observed during focal seizures were not dependent on hemispheric lateralization.14 In the intracarotid amobarbital procedure (ie, Wada’s test), the heart rate increased after left hemisphere inactivation but decreased after right hemisphere inactivation.17 Experiments in both animals and humans suggest that the insular cortex is the most important brain area for the control of sympathetically and parasympathetically mediated cardiovascular regulation. Stimulation of the human right insula increases sympathetic cardiovascular tone, whereas left insular stimulation increases parasympathetic activity.18 Several studies in stroke patients showed abnormal heart rate variability without any hemispheric difference.19 20 This can be the result of damage to cortical or subcortical structures known to regulate the cardiovascular autonomic tone. Disturbance of autonomic function, either by stimulation or inactivation, can be elicited from both hemispheres and from different cortical and subcortical levels. In our study neither patients with focal seizures nor those with an active epileptic focus showed asymmetries with respect to B or M waves. This suggests a direct brain stem involvement in the generation of these abnormalities.
In conclusion, our data are in favor of an autonomic dysfunction characterized by an increased sympathetic activity in patients with primary generalized as well as focal epilepsy. Further insight into the autonomic changes in patients with epilepsy might be expected by cross-spectral analysis of continuous blood pressure and heart rate curves in addition to bilateral TCD measures.
This study was supported by a grant from Glaxo Wellcome, Hamburg, Germany.
- Received May 23, 1997.
- Revision received September 11, 1997.
- Accepted September 12, 1997.
- Copyright © 1997 by American Heart Association
Epstein MA, Sperling MR, O’Connor MJ. Cardiac rhythm during temporal lobe seizures. Neurology. 1992;42:50–53.
Lundberg N. Continuous recordings and control of ventricular fluid pressure in neurosurgical practice. Acta Psychiatr Neurol Scand. 1960;149(suppl):1–193.
Einhäupl KM, Garner C, Dirnagel U, Schmieder G, Schmiedek P, Kufer G, Rieder J. Oscillations of ICP related to cardiovascular parameters. In: Miller JD, Taesdale GM, Rowan JE, Galbraith SL, Mendelow AD, eds. Intracranial Pressure VI. Berlin, Germany: Springer; 1986:290–297.
Maeda M, Takahashi K, Miyazaki M, Ishii S. The role of central monoamine system and the cholinoceptive pontine area on theoscillation of ICP ’pressure waves.’ In: Miller JD, Taesdale GM, Rowan JO, Galbraith SL, Mendelow AD, eds. Intracranial Pressure VI. Berlin, Germany: Springer; 1986:151–155.
Frysinger RC, Engel J, Harper RM. Interictal heart rate patterns in partial seizure disorders. Neurology. 1993;43:2136–2139.
Zamrini EY, Meador KJ, Loring DW, Nichols FT, Lee GP, Figueroa RE, Thompson WO. Unilateral cerebral inactivation produces differential left/right heart rate responses. Neurology. 1990;40:1408–1411.
Oppenheimer SM, Gelb A, Girvin JP, Hachinski VC. Cardiovascular effects of human insular cortex stimulation. Neurology. 1992;42:1727–1732.
Barron SA, Rogovski Z, Hemli J. Autonomic consequences of cerebral hemisphere infarction. Stroke. 1994;25:113–116.
Kopelainen JT, Sotaniemi KA, Huikuri HV, Myllylä VV. Abnormal heart rate variability as a manifestation of autonomic dysfunction in hemispheric brain infarction. Stroke. 1996;27:2059–2063