Seizures are more common in the newborn (neonatal) period than at any other stage of life. In older children / adults, it is usually possible to recognise a seizure, because the person will show a particular pattern of movement (e.g. convulsion) or behaviour (e.g. absence). In newborns, however, seizures often show no physical signs and can easily go undetected. This is a problem because untreated seizures can cause damage to the brain.

Researchers at Mass General Hospital for Children, in Boston, US, have been trying to find out why neonatal seizures are often 'invisible' clinically.

GABA is the major inhibitory neurotransmitter in the brain, and is released into synapses at the end of inhibitory neurons. When GABA binds to its receptor on the post-synaptic membrane, chloride and potassium ion channels are opened and (usually) chloride ions and potassium ions flow into and out of the neuron respectively. This activity prevents the neuron from firing.

In the developing brain, the 'arrangement' of chloride channels (i.e. the number and types of chloride channel present) at synapses is different to that in the adult brain. When a baby is born, there is a period of transition whereby the adult formation is adopted. Previous studies have shown that this aspect of development happens more quickly in deeper structures of the brain, such as the thalamus, than the brain surface (cortex) where seizures usually start.

In the current study, the team wanted to find out if this difference in chloride channels between brain regions is the reason that newborns often have seizures without convulsions.

The researchers first studied the brains of newborn mice and found that chloride levels in the cells of the thalamus and other deep structures were lower than the chloride levels in the cells of the cortex. This confirmed the different chloride channel activity in the brain layers.

Interestingly, when they applied GABA to the two areas, they saw completely opposite results - neurons in the deep structures were inhibited, but those in the cortex were activated.

The group then induced seizures in the models and then treated them with the anti-epileptic drug phenobarbital (which enhances the effect of GABA). As would be expected from the previous findings, seizure activity was stopped in the deeper brain structures (where GABA inhibits neurons), but not in the cortex (where GABA excites neurons).

This might be important in explaining the suppression of convulsions in newborn seizures, because convulsions require the passage of seizure signals from the cortex through the thalamus (and other deep structures) and out to the muscles. A seizure might be caused by GABA in the newborn cortex (where it still excites neurons), but the seizure won't necessarily spread to deeper structures, because here GABA might already act to inhibit neurons.

The drug bumetanide (a diuretic) acts by blocking the chloride pump responsible for immature neurons' excitatory response to GABA. The team induced seizures in animal models again, but this time they treated them with both bumetanide and phenobarbital. They found that seizure activity both the cortex and deeper brain structures stopped.

Dr Kevin Stayley, Senior Author on the project commented:

"Our study provides a logical mechanism for the clinical invisibility of many neonatal seizures, information that may help determine the best way to monitor newborns with brain injuries for seizures and select the best strategies for anticonvulsant treatment. For example, by blocking the protein responsible for immature brain cells' excitatory response to GABA, bumetanide essentially converts that immature response to a mature response and allows anti-seizure medicines to work properly. We are excited to be participating in a trial of bumetanide as an adjunctive treatment of neonatal seizures currently being carried out in collaboration with colleagues at Children's Hospital Boston and Brigham and Women's Hospital."

Read more here and here