Re-evaluation of the hypoxia theory as the mechanism of hyperventilation-induced EEG slowing

Hyperventilation-induced high-amplitude rhythmic slowing: A mimicker of absence seizures in children

PURPOSE

Hyperventilation (HV) in children can lead to HV-induced high-amplitude rhythmic slowing (HIHARS) on the EEG (electroencephalogram) which is sometimes associated with altered awareness (AA) and concomitant semiological features. Our aims were to determine the frequency of HIHARS in children, to assess if the associated semiological features were temporally related to HV, and to evaluate if specific semiological features can differentiate HIHARS with AA from absence seizures.

METHODS

Consecutive children with suspected new onset seizure(s) underwent HV and awareness testing during video-EEG acquisition. Hyperventilation-induced high-amplitude rhythmic slowing was defined as 2.5- to 5-Hz generalized rhythmic slowing with amplitude ≥100 μv lasting for ≥3 s. The associated semiological features were compared between the group of children with HIHARS and AA, an age- and gender-matched control group without HIHARS, and in children who experienced absence seizures during HV.

RESULTS

One hundred sixteen children with a mean age of 9.8 years were included. Hyperventilation-induced high-amplitude rhythmic slowing occurred in 39 children (33.6%) with AA documented in 30 (76.9%). The probability of developing AA during HIHARS was significantly and positively correlated with the HIHARS duration. The frequencies of HIHARS were not significantly different between children diagnosed with seizure(s) and those with nonepileptic spells. Hyperventilation cessation and staring did not occur in any child of the control group. Fidgeting and yawning were significantly more common in the group with HIHARS with AA while staring and blinking were significantly more frequent in the group of children with absence seizures.

CONCLUSIONS

We ascertained that HIHARS with AA is a relatively common occurrence in children and most likely represents an age-related nonepileptic phenomenon. When associated with fidgeting or yawning, it can help differentiate this phenomenon from absence seizures. However, recording the concomitant presence of generalized spike wave discharges on the EEG remains essential to confirm the diagnosis of absence seizures.

Electroencephalography of completely locked-in state patients with amyotrophic lateral sclerosis

Patients in completely locked-in state (CLIS) due to amyotrophic lateral sclerosis (ALS) lose the control of each and every muscle of their body rendering them motionless and without any means of communication. Though some studies have attempted to develop brain-computer interface (BCI)-based communication methods with CLIS patients, little information is available of the neuroelectric brain activity of CLIS patients. However, because of the difficulties with and often loss of communication, the neuroelectric signature may provide some indications of the state of consciousness in these patients. We recorded electroencephalography (EEG) signals from 10 CLIS patients during resting state and compared their power spectral densities with those of healthy participants in fronto-central, central, and centro-parietal channels. The results showed significant power reduction in the high alpha, beta, and gamma bands in CLIS patients, indicating the dominance of slower EEG frequencies in their oscillatory activity. This is the first study showing group-level EEG change of CLIS patients, though the reason for the observed EEG change cannot be concluded without any reliable communication methods with this population.

Change in the electroencephalogram delta wave in the frontal cranial region of rats with the hyperventilation

Hyperventilation is one way to cause activation on the electroencephalogram (EEG) to diagnose brain disorders. The hyperventilation is also known to affect on the delta power in EEG. This study divided the total delta wave into low, middle, and high bands corresponding to the wave frequency. The power in these three delta wave bands was examined in the frontal cranial region of adult male Sprague-Dawley rats hyperventilated with ventilation (VE) of 360, 540, and 720 ml/min for 5 min. The control group was ventilated normally with a volume of 160 ml/min. The results show that the relative power of the low delta band in the rats hyperventilated at 360 ml/min VE was significantly increased compared with powers of pre-hyperventilation (p<0.05). The relative power of the middle delta band was not significantly affected by hyperventilation at any VE, and in the high delta band, all of the relative powers were decreased significantly in all hyperventilated rats compared with powers of pre-hyperventilation (p<0.05). We concluded that hyperventilation affects the frontal cranial region, by increasing the low delta band and decreasing the high delta band.

Relationship between Hyperventilation-Induced Electroencephalographic Changes and PCO2 Level

Background and Purpose:

We conducted this study to define the relationship between the hyperventilation-induced EEG changes (HV-EEG changes) and PCO₂

Methods:

In consecutive EEG recordings of 190 patients, we gathered data on PCO₂ during the hyperventilation procedure. The data included baseline PCO₂ (B-PCO₂), PCO₂ after 5 min of hyperventilation (5 min-PCO₂), the mean value of the PCO₂ (M-PCO₂), and the difference between B-PCO₂ and 5 min-PCO₂ (ΔPCO₂). We divided the enrolled patients into two groups by hyperventilation response (response group and no-response group), presence of epilepsy (epileptic group and non-epileptic group) and age (child-adolescent group and adult group) repeatedly. We compared the four variables between the two groups in each pair.

Results:

ΔPCO₂ was 14.2±5.0 mmHg (mean±SD) in response group (n=48) and 12.4±5.0 in no-response group (n=142; p=0.033) for all the patients. For adult patients only, 5 min-PCO₂ was 24.3±3.4 in response group (n=30) and 26.2±4.6 in no-response group (n=115; p=0.048), and ΔPCO₂ was 15.8±4.0 and 12.9±5.0, respectively (p=0.004). In non-epileptic patients, 5min-PCO₂ was 23.4±2.2 in response group (n=7) and 26.3±3.8 in no-response group (n=44; p=0.026), and ΔPCO₂ was 15.9±4.3 and 12.7±3.9, respectively (p=0.053).

Conclusions:

In adults and non-epileptic patients, ΔPCO₂ and 5 min-PCO₂ may be crucial to the induction of EEG changes by hyperventilation. PCO₂ could be a crucial factor for provoking HV-EEG changes in a limited group of patients.

The difficult-to-treat electroconvulsive therapy patient - Strategies for augmenting outcomes

BACKGROUND

Several treatment strategies for augmenting outcomes with ECT (concurrent antidepressant treatment, frequency of ECT treatments, hyperventilation and use of remifentanil) are discussed in the context of a difficult clinical case, accompanied by a review of the relevant existing literature.

METHODS

Literature on the above aspects of ECT technique was identified via a PubMed search and was critically reviewed.

RESULTS

There is preliminary evidence that concurrent administration of some antidepressant medications may be useful in the highly treatment resistant patient, though due attention should be given to potential risks in combining these with ECT; reduction of the treatment frequency to twice a week; hyperventilation prior to each treatment; and the use of remifentanil to minimise the dosage of induction anaesthetics with anticonvulsant properties, may be useful strategies to enhance seizure production in cases where a high or rapidly rising seizure threshold is a major impediment to ECT treatment.

LIMITATIONS

It should be noted that empirical evidence for the effectiveness of each of the above strategies in producing better outcomes with ECT is not definitive, pointing to the need for further research in these areas.

CONCLUSIONS

The above strategies may be useful in clinical ECT practice, particularly in patients who are apparently treatment resistant, but the practitioner should be aware that the level of evidence underpinning these approaches is at present, preliminary.

Changes in visual-evoked potential habituation induced by hyperventilation in migraine

Hyperventilation is often associated with stress, an established trigger factor for migraine. Between attacks, migraine is associated with a deficit in habituation to visual-evoked potentials (VEP) that worsens just before the attack. Hyperventilation slows electroencephalographic (EEG) activity and decreases the functional response in the occipital cortex during visual stimulation. The neural mechanisms underlying deficient-evoked potential habituation in migraineurs remain unclear. To find out whether hyperventilation alters VEP habituation, we recorded VEPs before and after experimentally induced hyperventilation lasting 3 min in 18 healthy subjects and 18 migraine patients between attacks. We measured VEP P100 amplitudes in six sequential blocks of 100 sweeps and habituation as the change in amplitude over the six blocks. In healthy subjects, hyperventilation decreased VEP amplitude in block 1 and abolished the normal VEP habituation. In migraine patients, hyperventilation further decreased the already low block 1 amplitude and worsened the interictal habituation deficit. Hyperventilation worsens the habituation deficit in migraineurs possibly by increasing dysrhythmia in the brainstem-thalamo-cortical network.

Effect of carbon dioxide on background cerebral electrical activity and fractional oxygen extraction in very low birth weight infants just after birth

Decreased arterial carbon dioxide tension (PaCO2) results in decreased cerebral blood flow, which is associated with diminished cerebral electrical activity. In such a situation, cerebral fractional oxygen extraction (CFOE) would be expected to increase to preserve cerebral oxygen delivery. This study aimed to determine whether changes in blood gases in infants less than 30 wk' gestation were associated with changes in background electroencephalograms (EEG) and CFOE. Thirty-two very low birth weight infants were studied daily for the first three days after birth. Digital EEG recordings were performed for 75 min each day. CFOE, mean blood pressure and arterial blood gases were measured midway through each recording. EEG was analysed by (a) spectral analysis and (b) manual calculation of interburst interval. Blood pressure, pH and PaCO2 did not have any effect on the EEG. On day one, only PaCO2 showed a relationship with the relative power of the delta frequency band (0.5-3.5 Hz) and the interburst interval. The relative power of the delta band remained within normal limits when PaCO2 was between 24 and 55 mmHg on day one. There was a negative association between PaCO2 and CFOE. The associations between PaCO2 and EEG measurements were strongest on day one, weaker on day two, and absent on day three. The slowing of EEG and increased CFOE at lower levels of PaCO2 are likely to be due to decreased cerebral oxygen delivery induced by hypocarbia. When PaCO2 was higher, there was suppression of the EEG.

BOLD-contrast functional MRI signal changes related to intermittent rhythmic delta activity in EEG during voluntary hyperventilation-simultaneous EEG and fMRI study

Differences in the blood oxygen level dependent (BOLD) signal changes were studied during voluntary hyperventilation (HV) between young healthy volunteer groups, (1) with intermittent rhythmic delta activity (IRDA) (N = 4) and (2) controls (N = 4) with only diffuse arrhythmic slowing in EEG (normal response). Subjects hyperventilated (3 min) during an 8-min functional MRI in a 1.5-T scanner, with simultaneous recording of EEG (successful with N = 3 in both groups) and physiological parameters. IRDA power and average BOLD signal intensities (of selected brain regions) were calculated. Hypocapnia showed a tendency to be slightly lighter in the controls than in the IRDA group. IRDA power increased during the last minute of HV and ended 10-15 s after HV. The BOLD signal decreased in white and gray matter after the onset of HV and returned to the baseline within 2 min after HV. The BOLD signal in gray matter decreased approximately 30% more in subjects with IRDA than in controls, during the first 2 min of HV. This difference disappeared (in three subjects out of four) during IRDA in EEG. BOLD signal changes seem to depict changes, which precede IRDA. IRDA due to HV in healthy volunteers represent a model with a clearly defined EEG pattern and an observable BOLD signal change.