A mother and daughter with agenesis of the corpus callosum

Interhemispheric coherence of EEG rhythms in children: Maturation and differentiation in corpus callosum dysgenesis

OBJECTIVES

To evaluate the evolution of interhemispheric coherences (ICo) in background and spindle frequency bands during childhood and use it to identify individuals with corpus callosum dysgenesis (CCd).

METHODS

A monocentric cohort of children aged from 0.25 to 15 years old, consisting of 13 children with CCd and 164 without, was analyzed. The ICo of background activity (ICOBckgrdA), sleep spindles (ICOspindles), and their sum (sICO) were calculated. The impact of age, gender, and CC status on the ICo was evaluated, and the sICO was used to discriminate children with or without CCd.

RESULTS

ICOBckgrdA, ICOspindles and sICO increased significantly with age without any effect of gender (p < 10-4), in both groups. The regression equations of the different ICo were stronger, with adjusted R2 values of 0.54, 0.35, and 0.57, respectively. The ICo was lower in children with CCd compared to those without CCd (p < 10-4 for all comparisons). The area under the precision recall curves for predicting CCd using sICO was 0.992 with 98.9 % sensitivity and 87.5 % specificity.

DISCUSSION

ICo of spindles and background activity evolve in parallel to brain maturation and depends on the integrity of the corpus callosum. sICO could be an effective diagnostic biomarker for screening children with interhemispheric dysfunction.

Role of corpus callosum in sleep spindle synchronization and coupling with slow waves

Sleep spindles of non-REM sleep are transient, waxing-and-waning 10–16 Hz EEG oscillations, whose cortical synchronization depends on the engagement of thalamo-cortical loops. However, previous studies in animal models lacking the corpus callosum due to agenesis or total callosotomy and in humans with agenesis of the corpus callosum suggested that cortico-cortical connections may also have a relevant role in cortical (inter-hemispheric) spindle synchronization. Yet, most of these works did not provide direct quantitative analyses to support their observations. By studying a rare sample of callosotomized, split-brain patients, we recently demonstrated that the total resection of the corpus callosum is associated with a significant reduction in the inter-hemispheric propagation of non-REM slow waves. Interestingly, sleep spindles are often temporally and spatially grouped around slow waves (0.5–4 Hz), and this coordination is thought to have an important role in sleep-dependent learning and memory consolidation. Given these premises, here we set out to investigate whether total callosotomy may affect the generation and spreading of sleep spindles, as well as their coupling with sleep slow waves. To this aim, we analysed overnight high-density EEG recordings (256 electrodes) collected in five patients who underwent total callosotomy due to drug-resistant epilepsy (age 40–53, two females), three non-callosotomized neurological patients (age 44–66, two females), and in a sample of 24 healthy adult control subjects (age 20–47, 13 females). Individual sleep spindles were automatically detected using a validated algorithm and their properties and topographic distributions were computed. All analyses were performed with and without a regression-based adjustment accounting for inter-subject age differences. The comparison between callosotomized patients and healthy subjects did not reveal systematic variations in spindle density, amplitude or frequency. However, callosotomized patients were characterized by a reduced spindle duration, which could represent the result of a faster desynchronization of spindle activity across cortical areas of the two hemispheres. In contrast with our previous findings regarding sleep slow waves, we failed to detect in callosotomized patients any clear, systematic change in the inter-hemispheric synchronization of sleep spindles. In line with this, callosotomized patients were characterized by a reduced extension of the spatial association between temporally coupled spindles and slow waves. Our findings are consistent with a dependence of spindles on thalamo-cortical rather than cortico-cortical connections in humans, but also revealed that, despite their temporal association, slow waves and spindles are independently regulated in terms of topographic expression.

Role of corpus callosum in interhemispheric coherent activity during sleep

OBJECTIVE

To investigate to what extent the increase in interhemispheric coherent activity observed from wakefulness to sleep depends on the integrity of the corpus callosum (CC).

METHODS

Interhemispheric coherent activity was analyzed in two epileptic patients selected for callosotomy because of multifocal refractory epilepsy, before and 4 months after callosotomy. One patient underwent complete callosotomy and another was subjected to callosotomy of the anterior 2/3, which offered the possibility of comparing the role of the CC in the coherent activity increase from wakefulness to sleep, between anterior regions with interrupted CC communication (in the two patients) and posterior regions with intact communication (in one of them). Results were compared with a group of normal subjects.

RESULTS

Both patients showed increased coherent activity from wakefulness to sleep after surgery.

CONCLUSIONS

Results demonstrate that interhemispheric coherent activity, despite an attenuation after surgery, is higher during SWS than during wakefulness after sectioning the CC; however, they have to be taken with caution because they come from two patients only.

SIGNIFICANCE

Present results show that the increase in coherent activity during sleep does not depend exclusively on callosal integrity but also on state-dependent influences from sleep-promoting mechanisms, probably spread throughout the thalamo-cortical network.