About one-third of people with epilepsy continue to have seizures despite taking medication. For many of them, deep-brain stimulation (DBS) offers a way out, electrodes implanted deep in the brain deliver small electrical pulses that quiet the neural storm. But the most common surgical target, a structure called the anterior thalamic nucleus (ANT), comes with an ironic trade-off: it helps control seizures during the day, but it can wreck the sleep that the brain needs to recover at night.
A preliminary study published July 23 in the Journal of Neurology suggests there may be a better option. Researchers in France and Italy compared sleep outcomes in patients receiving DBS targeting the pulvinar (PuM), a large nucleus at the back of the thalamus, either alone or in combination with ANT stimulation. The early signal is striking: the pulvinar appears to leave sleep architecture intact, while ANT stimulation, even when delivered simultaneously, fragments it.
What they found
The team led by Pietro Mattioli at APHM Timone Hospital in Marseille analyzed five nights of polysomnography from three patients with drug-resistant epilepsy, each implanted with bilateral DBS electrodes targeting the pulvinar, the anterior thalamic nucleus, or both.
The most instructive case was Patient 1, who was recorded across three consecutive nights under different DBS configurations: ANT and pulvinar both active (ANT+/PuM+), ANT active with pulvinar turned off (ANT+/PuM-), and then both active again. This within-subject design let the researchers isolate the effect of each target on sleep.
When the pulvinar stimulator was switched off, leaving only ANT stimulation, the patient’s sleep architecture degraded noticeably. The disruption was most pronounced in the latter part of the night, during the final NREM sleep cycle, where normalized delta power, a marker of deep, restorative sleep, dropped significantly compared with the night when both targets were active.
The two patients with pulvinar-only DBS (one receiving cyclic stimulation, the other continuous) told a different story. Their sleep structure was preserved. Sleep efficiency remained normal, and delta power was higher across the night compared with periods of ANT stimulation. In short, pulvinar stimulation did not appear to interfere with the brain’s natural sleep machinery at all.
Why it matters
Sleep disruption is one of the most common and most undertreated complaints in epilepsy. Patients with drug-resistant epilepsy already contend with seizure-related waking, the sedating effects of antiseizure medications, and the psychological burden of a chronic neurological condition. Adding iatrogenic sleep disruption from a treatment meant to help them compounds the problem.
The anterior thalamic nucleus became the dominant DBS target for epilepsy after the landmark SANTE trial, published in 2010, showed that bilateral ANT stimulation reduced seizure frequency by roughly 40 percent at one year and nearly 70 percent at five years. But the SANTE trial also documented what clinicians had long suspected: ANT-DBS can fragment sleep, likely because the nucleus sits at a crossroads of the brain’s sleep-wake circuitry. The anterior thalamus has dense reciprocal connections with the hippocampus, the cingulate cortex, and the default mode network, all of which oscillate between sleep and wake states.
The pulvinar, by contrast, is primarily a visual and associative relay. It routes information between cortical regions rather than governing arousal. If the early results hold, it may offer a way to uncouple seizure control from sleep disruption, a separation that has proved difficult with ANT stimulation alone.
“The patients with PuM-DBS alone had preserved sleep structure and normal sleep efficiency, with higher normalized delta power compared to ANT stimulation,” the authors note. For a patient population in which restorative sleep is already fragile, preserving delta power is not a minor detail. It is the difference between waking rested and waking depleted.
Limits
The study is small, three patients, five nights, and the authors are upfront about its preliminary nature. The within-subject comparison in Patient 1 is suggestive but cannot rule out night-to-night variability unrelated to stimulation settings. The sample includes only patients with drug-resistant epilepsy, so the findings may not generalize to other indications for DBS. Longer recordings and larger cohorts will be needed to confirm whether pulvinar stimulation reliably spares sleep architecture over months and years of treatment.
Bottom line
For epilepsy patients whose seizures are not controlled by medication, DBS is a proven option, but the choice of target matters beyond seizure reduction. This early evidence suggests that the pulvinar may deliver the therapeutic benefit of thalamic stimulation without the sleep-wrecking side effect that has shadowed ANT-DBS for more than a decade. If replicated in larger studies, the finding could shift surgical practice toward a target that treats seizures and lets patients sleep through the night.
Source
Mattioli P, Akyol Gurses A, Simula S, Carron R, Bartolomei F, Lambert I, Pizzo F. Effects of pulvinar and combined pulvinar/anterior thalamic deep-brain stimulation on sleep: preliminary results. J Neurol. 2026 Jul 23;273(8):484. doi: 10.1007/s00415-026-14012-8. PMID: 42493672.