Search

Objectives: Hyperventilation is effective in inducing generalized spike-wave discharges in patients with absence seizures and improves visual function and normalizes visual function in patients with multiple sclerosis. Hyperventilation increases the excitability of cutaneous and motor axons. In experimental animals, hyperventilation increases excitability of hippocampal neurons. There is however no direct evidence of a hyperventilation-induced increase in neuronal excitability within the central nervous system in humans. In this study we determined the effects of hyperventilation on the human corticospinal system., Methods: We studied the effects of hyperventilation on (1) motor evoked potentials (MEPs) induced by transcranial magnetic pulse stimulation of the motor cortex and (2) F-wave responses. Six subjects were studied., Results: Hyperventilation resulting in an end-tidal pCO2 of 15 mm Hg or less enhanced the amplitude of the MEP and resulted in a shortened onset latency. F-wave amplitudes were enhanced without any change in onset latency., Conclusions: These findings indicate that hyperventilation increases the excitability of the human corticospinal system. A hyperventilation-induced increase in excitability within the central nervous system may account for clinical phenomena such as facilitation of spike-wave discharges.

Shifts in scalp-recorded direct current (DC) potential were studied in relation to the changes in end-tidal partial CO2 (PACO2) or O2 (PAO2) of the expired gas either during hyperventilation (HV), hypoventilation (HYPO) or inhalation of high CO2 content air during HV or HYPO in 10 healthy subjects. The DC potential was obtained through a chopper stabilized type of DC amplifier from Cz referred to linked earlobes. Each session was comprised of 3 min control, 3 min experimental and 5 min recovery periods. HV induced a negative shift of the DC potential of 765.5 +/- 203.0 microV (mean +/- SEM). Inhalation of 6% CO2 air during HYPO induced a positive shift of the DC potential of 280.6 +/- 62.8 microV (mean +/- SEM). The magnitude of the DC potential shifts was linearly dependent on the changes in the end-tidal PACO2 (r = 0.78, P < 0.0001). There was no change in the cephalic inter-electrodes impedance during each experimental session. The results suggest that the scalp-recorded DC potentials reflect the changes in cortical excitability associated with the PACO2 level.

Clinical (paresthesiae and Trousseau's sign), EMG, blood biochemical (lactic acid and total Ca2+) and gas-analytic (pH, PCO2, PO2, HCO3) changes were studied during and after 5 min voluntary hyperventilation (HV) in 15 normal subjects. Paresthesiae and spontaneous motor activity were common manifestations (87% and 67% respectively) in our test. They were significantly related only to changes in pH, PCO2 and PO2, with paresthesiae arising earlier and at milder gas-analytic changes than motor activity. Paresthesiae and motor activity induced by HV cannot be used as reliable indicators of pathology. Their presence simply reflects the degree of the blood gas-analytic changes (pH, PCO2 and PO2) which therefore should be carefully monitored before drawing any conclusion from an HV test.

We studied the effects of hyperventilation under control of the end-tidal PCO2, on the electromyographic silent period evoked by transcranial magnetic brain stimulation and by peripheral nerve stimulation. We also studied the effects of hyperventilation on the threshold, latency and amplitude of motor potentials. Hyperventilation significantly reduced the duration of the cortical silent period, but did not affect the length of the peripheral silent period. Neither did it alter the latency, amplitude or threshold of the motor potentials. These findings suggest that hyperventilation selectively depresses motor cortical inhibition in humans.

Hyperventilation leads to an increase in slow EEG activity as well as to a decrease in alpha activity. These effects may be considered a result of reduction in cerebral blood flow due to vasoconstriction, but metabolic factors, such as alkalosis and the increased formation of cerebral lactate, may also have to be taken into account. As indomethacin decreases cerebral blood flow it is possible to study cerebral vasoconstriction, without concomitant metabolic alkalosis or cerebral lactate formation. Two parallel groups of 12 healthy male subjects (age 20-25) were studied with quantitative EEG (qEEG) and cerebral blood flow velocity as parameters. In the first group the effect of 100 mg indomethacin was studied. In the parallel group a standardized hyperventilation procedure was performed. In the indomethacin group the blood flow velocity decreased to 60% of the initial value; the qEEG showed a 0.5 Hz slowing of the alpha peak frequency (P less than 0.01) and a decrease in the power of the alpha band without any change in the delta or theta band. In the hyperventilation group the blood flow velocity decreased to 63% of the initial value and the qEEG showed a marked increase in delta and theta activity (P less than 0.01), but a non-significant change in alpha peak frequency. Indomethacin and hyperventilation caused similar degrees of vasoconstriction; however, the increase in qEEG slow wave activity, which was observed only in the hyperventilation group, is apparently related to metabolic rather than haemodynamic factors.

To investigate the role of cerebral hypoxia as a causative factor in the alteration of the qEEG during hyperventilation, qEEG changes caused by progressive hypocapnia were compared with qEEG changes due to progressive normobaric hypoxia in two parallel groups of 12 and 10 healthy male subjects (age 20-27 years), respectively. In the first group, qEEG records were obtained before and during hyperventilation to pCO2 levels of 4.0, 3.0 and 2.0 kPa. In the second group, the qEEG samples were taken before and during hypoxia with hemoglobin oxygen saturations of 80, 70 and 60%. In both groups, blood flow velocity in the middle cerebral artery was also recorded. Hyperventilation caused an exponential increase in slow activity and a decrease in alpha power. No shift in the alpha mean frequency and alpha peak frequency was observed, except with the pCO2 level of 4.0 kPa, which caused an increase in both variables. Hypoxia with a hemoglobin oxygen saturation of 60% caused a much less pronounced increase in slow activity. No change in total power in the alpha band was found, but both the alpha peak frequency and alpha mean frequency decreased. Lesser degrees of hypoxia caused only minimal EEG changes. Blood flow velocity was decreased by hyperventilation but increased by hypoxia. It is concluded that the EEG changes observed during hyperventilation must mainly or totally be attributed to factors other than cerebral hypoxia.

The effects on the EEG rhythms recorded from the sensorimotor cortex (post-sigmoid gyrus) of anaesthetized cats were studied under 4 conditions of artificial mechanical hyperventilation (HV) before and after cervical bilateral vagotomy. In animals with intact vagus nerves, using visual examination, EEG changes were only observed within the 2nd min during HV produced by increased stroke volume (delta V) with associated hypocapnia. Quantitative EEG (qEEG) showed that, for the same increase in minute ventilation and the same degree of hypocapnia, delta V induced a greater and earlier relative decrease (2nd min) in the power density of delta, theta and alpha bands, than increased pump frequency (delta F). The delta F tests produced a fall only in the theta band and within the 3rd min. With constant paCO2, transient modifications occurred only with delta V and were limited to the first 30 sec. In bivagotomized cats, moderate EEG responses to delta V plus associated hypocapnia persisted partly in the alpha band. Finally, no changes appeared with delta V or delta F when the vagus nerves were cut and paCO2 was maintained constant. The present data suggest strongly that, in anaesthetized cats, peripheral vagal afferents from the respiratory system play a major role in the EEG changes caused by artificial hyperventilation.

EEG development through advancing clinical stages was studied in 30 cases of Rett syndrome, a progressive encephalopathy seen so far only in girls. Besides slowing of background activity and epileptic patterns, which are found in many types of 'degenerative' brain diseases, we observed in advanced stages of Rett syndrome unusual EEG patterns resembling those seen in slow virus encephalitis. On the other hand, the EEG was normal in one case of advanced Rett syndrome. The typical steps of EEG development were: initially normal EEG; as the first abnormality rolandic spikes, often followed by other epileptic patterns; and, in more advanced stages, a pseudoperiodic delta pattern and occasionally generalized periodic spike activity. The results motivate a discussion concerning other possible origins of Rett syndrome in addition to the supposed genetic one. The question is raised whether Rett syndrome might be etiologically heterogeneous.

The hyperventilation syndrome (HVS) is a functional disorder with repeated involuntary hyperventilation attacks together with symptoms of respiratory alkalosis. We have studied the EEG and end-tibial pCO2 in the resting state and during hyperventilation activation in 12 HVS patients in order to find out whether there is a greater susceptibility to cerebral vasoconstriction in HVS patients than in controls, as indicated by slowing of the EEG. A surprisingly high proportion (58%) of abnormal resting EEGs was found in HVS patients, although the patients were neurologically normal. More theta and beta background activity was usually revealed in a quantitative computer analysis, especially frontally. Although the hyperventilation activation caused the same degree of hypocapnia in HVS patients and in controls, peripheral symptoms like tingling and numbness of fingers, as well as carpopedal spasms, occurred much more often in HVS patients. However, the EEG changes due to hyperventilation were similar in both HVS patients and normal controls, and it thus seems that the reason for cerebral symptoms in HVS patients is not a greater susceptibility to cerebral vasoconstriction.

A standardized hyperventilation (HV) procedure has been developed in which the end-tidal pCO2 was decreased to 2 kpa. In 24 young male subjects blood flow velocity and qEEG were studied before, during and after HV. This standardized hyperventilation procedure gave rise to a decrease in blood flow velocity to 40% of baseline value and highly significant qEEG changes in 3 derivations. Both relative and absolute band power estimates showed an increase in slow activity and a decrease in alpha and beta activity. The use of subtraction spectra led to a more precise and detailed presentation of these changes than the use of classical qEEG parameters. These changes were reproducible after 1 week. The effects found in the presented model of HV-induced ischaemia appeared to be twice as large as those found in a model of hypobaric hypoxia. The present model might be used to test the efficacy of anti-ischaemic drugs in young human subjects.

An unusual transient electroencephalographic pattern consisting of repetitive sharp-contoured wave forms in the range of 4--7 c/sec was seen in recordings from 65 patients who were between 42 and 80 years of age. This pattern was usually predominant over the parietal and temporal regions. No clinical alteration was associated with the pattern. Although the pattern resembled a subclinical EEG seizure discharge, it did not correlate with clinical seizures. The pattern occurred in patients with diverse clinical complaints, and although its mechanism of origin remains uncertain, it appears to represent a benign EEG phenomenon that has little diagnostic significance.

Two cases of Moyamoya disease in children are presented. The EEG of each child demonstrated minimal abnormalities at rest. However, during 3 min of hyperventilation, the record developed rhythmic delta activity which persisted in excess of 5 min after termination of over breathing. This activity did not respond to glucose administration and was maximal in the areas correlating with the clinical symptoms. It appears to represent a physiologic correlate of an abnormal vascular response to hypocapnia. Although admittedly a non-specific response, the persistence of delta activity easily provoked by hyperventilation, the topographic correspondence to the ultimately demonstrated vascular anomaly, and the lack of response to glucose ingestion represent a combination of factors which, in the appropriate clinical setting, are distinctive and of diagnostic value. Currently, invasive arteriography is required for definitive diagnosis of Moyamoya. Improved selection of children for invasive studies may be accomplished by EEG screening.

Acute hypocapnia decreases CBF, increases hemoglobin affinity for oxygen and causes cerebral tissue hypoxia. This tissue hypoxia is reversed with inhalation of 100% O2 in dogs. EEG slowing produced by hyperventilation is considered a manifestation of cerebral hypoxia due to decreased CBF and is thought to be reversed with hyperoxia. This study evaluated the effects of 3 gas mixtures (16% O2, 21% O2, 100% O2) on posterior frequencies of the resting and hyperventilatory EEG in normal subjects aged 23-37. Hypocapnia was maintained to an end-tidal pCO2 of 21 mm Hg for 3 min. Respiratory measures, heart rate, saO2, minute ventilation and side effects were recorded. EEG was analyzed by visual inspection and by spectral analysis. Spectral analysis evaluated total amplitude, percentile frequencies, and peak frequencies. There were significant changes from eucapnia to hypocapnia for the group in all physiologic parameters, total amplitude by spectral analysis, and posterior frequencies by visual analysis. There were no significant differences among the gases. We conclude that the EEG changes of hyperventilation are independent of the concentration of inspired oxygen over the range studied in our subjects. Symptoms of hyperventilation are likewise independent of the inspired oxygen concentration for the range studied.