Discussion
DBS of the bilateral anterior nuclei of the thalamus (ANT) has emerged as an effective device therapy used in patients where pharmacotherapy fails to adequately manage seizures. However, given the connections between the anterior thalamus with limbic and prefrontal circuits, ANT-DBS has been demonstrated to have potential side effects on behavior and mood. It is well recognized that epilepsy and psychiatric disorders are bidirectionally associated, with up to one-third of people with epilepsy experiencing a lifetime psychiatric illness (Mula et al., 2021).
The introduction of neurostimulation further increases the risk of serious symptoms, particularly in patients with pre-existing neuropsychiatric disorders. Clinical trials have documented notable psychiatric side effects in a subset of DBS patients. In the SANTE trial of ANT-DBS, 14-15% of patients reported new or worsened depression during the blinded stimulation phase, compared to only ~2% in the sham group (Imbach et al., 2023). Over long-term follow-up, the prevalence of depressive mood symptoms rose to 37% of patients (Salanova et al., 2015), and about 12% reported suicidal ideation, although most of those who reported suicidal ideation had prior histories of depression. Anxiety symptoms and psychotic symptoms have additionally been observed in some patients during chronic stimulation.
Fortunately, these psychiatric effects are usually reversible with adjustments to stimulation parameters, such as reducing stimulation current or duty cycle duration. For example, one report stated that two DBS recipients with prior histories of depression developed acute depressive episodes immediately after certain programming settings, and two others with no prior history of psychosis gradually developed paranoid anxiety (Järvenpää et al., 2018); in all cases, changing the DBS settings led to resolution of the psychiatric symptoms. This reversibility suggests that DBS itself, likely the intensity and/or location of stimulation, may contribute to the psychiatric irregularities.
In the context of our patient, the worsening behavioral disturbances after DBS despite a partial reduction in seizures is consistent with these reported phenomena. It is possible that her history of psychosis and intellectual disability may have predisposed her to such complications. It is noteworthy that improving seizure control can sometimes unmask or exacerbate psychiatric symptoms, a phenomenon known as forced normalization: the emergence of psychosis or behavioral disorder when epileptic activity is suppressed in a previously uncontrolled patient. This has been described with various therapies including pharmacological treatments, VNS, and DBS. It is thought to reflect neurochemical shifts when ictal control is suddenly established. One case report describes a man with mental retardation and medically intractable epilepsy who became delusional after his seizures abruptly ceased post-VNS implantation (Loganathan et al., 2015).
In our patient, a similar mechanism could explain why reducing her seizure burden via DBS coincided with heightened psychotic and behavioral issues, in addition to the inherent risks of neuropsychiatric side effects as a result of thalamic stimulation. Awareness of this possibility is critical in determining adequate treatment plans and calls for close psychiatric monitoring of DBS patients and a careful titration of stimulation to maintain a delicate balance that controls seizures without inducing intolerable mood or behavioral changes.
Treatment-resistant epilepsy (TRE) in pediatric patients is increasingly understood as a manifestation of underlying genetic and neurodevelopmental disorders, especially when conventional imaging fails to reveal a focal structural lesion. Literature has linked early-onset, intractable epilepsy to chromosomal abnormalities, single-gene mutations, and disorders of neuronal development, particularly those affecting GABAergic interneurons and the excitatory/inhibitory (E/I) balance in the maturing brain.
Balanced chromosomal translocations involving the X chromosome and autosomes, particularly the t(14;X), have been associated with intellectual disability, developmental delay, and various epilepsy syndromes. Although such reciprocally balanced translocations do not result in any gross loss or duplication of genetic material, their pathogenicity could arise from gene disruption at breakpoints or aberrant gene regulation due to altered chromatin structure or skewed X-inactivation. Disruptions involving FGF13 and PHF8, two X-linked genes critical for neurodevelopment, have been implicated in prior t(14;X) cases with phenotypes including generalized epilepsy with febrile seizures plus (GEFS+) and intellectual disability. Notably, FGF13 plays a role in neuronal excitability and the function of voltage-gated sodium channels, while PHF8 is a histone demethylase involved in X-linked mental retardation and cleft lip/palate (Loenarz et al., 2010).
A particularly relevant study (Puranam et al., 2015) described a family with a t(X;14)(q27;q21) translocation disrupting FGF13, in which affected males exhibited febrile seizures in childhood progressing to refractory temporal lobe epilepsy in adolescence and adulthood. Animal models of Fgf13 haploinsufficiency showed decreased inhibitory synaptic input and heightened neuronal excitability, supporting the theory that deficits in GABAergic signaling underlie the clinical phenotype (Puranam et al., 2015). This mechanism aligns with the broader hypothesis that impaired interneuron development or migration during critical periods of brain maturation can lead to a lasting E/I imbalance, predisposing individuals to seizures that become increasingly resistant to treatment over time.
Although the precise impact of a 14;X translocation varies with the genes affected, the clinical trajectory observed in many of these cases, including early febrile seizures followed by medically refractory focal epilepsy, supports a model in which subtle widespread disruption of inhibitory networks contributes to epileptogenesis. This hypothesized mechanism provides strong implications for diagnosis and management, as traditional neuroimaging and EEG findings may underestimate the extent of diffuse network dysfunction. In such cases, genome sequencing can be instrumental in identifying causal mutations and informing prognosis, potential targeted therapies, or eligibility for clinical trials.
Given these findings, it is reasonable to hypothesize that the epilepsy observed in the present case may stem from a similar genetic disruption affecting early inhibitory circuit formation. While confirmatory genomic data is not yet available for this patient, the clinical features and resistance to therapy are consistent with previously described translocation-linked developmental epilepsies.