This case report highlights an intricate, dual neurobiological observation in an adolescent patient undergoing maintenance therapy with LAI aripiprazole for complex BD and cannabis use disorder. While the therapy achieved its primary clinical targets—yielding psychiatric remission and the apparent structural stabilization of the cerebellum and vermis—it simultaneously unmasked a neuroelectrophysiological vulnerability, culminating in cortical hyperexcitability, generalized EEG alterations, and clinical myoclonus. This juxtaposition of macrostructural volume preservation and microstructural electrical destabilization offers valuable insights into the functional topography of fronto-cerebellar networks.

The relationship between structural cerebellar pathology and major psychoses is historically grounded. Heath et al. (1982) [18] demonstrated that approximately 50% of patients with functional psychoses exhibited prominent cerebellar vermal atrophy upon visual inspection of computed tomography (CT) scans, compared to a mere 0.5% to 3.7% in control groups. Contemporary neuroscience has established that the cerebellum plays a critical role in cognitive processing and emotional regulation, a concept heavily implicated in the “cognitive dysmetria” hypothesis [11]. Crucially, the integrity of cerebellar-cortical networks dictates clinical trajectories. Utilizing baseline resting-state functional MRI (rs-fMRI), Velioglu et al. (2025) [12] revealed that stronger baseline functional connectivity between cerebellar cognitive systems and midline cerebral structures predicts superior treatment outcomes across both aripiprazole and risperidone cohorts.

The lack of detectable progression in volumetric changes over 9 months points toward an apparent absence of detectable macrostructural regression during aripiprazole maintenance. Given the substantial methodological limitations and manual segmentation variability acknowledged herein—particularly the calculated +24.77% change in vermis volume—this trajectory is more appropriately described as an absence of detectable change on repeat CT rather than true biological tissue stabilization [24]. Cumulative affective episodes, high-stress states, and chronic cannabis use trigger neurotoxic cascades, including pro-inflammatory cytokine upregulation and accelerated oxidative stress, which may lead to structural gray matter loss [13]. While our clinical data cannot directly measure molecular changes, existing preclinical literature offers hypothetical frameworks that may explain these findings. For instance, the literature suggests aripiprazole may modulate brain-derived neurotrophic factor (BDNF) and attenuate neuroinflammatory pathways in animal models [14,15]; however, these mechanisms remain entirely presumptive in our patient and cannot be verified without fluid biomarkers or advanced functional imaging. Furthermore, it must be emphasized that because no prior childhood neuroimaging was available for comparison, we cannot definitively characterize this baseline cerebellar and vermis atrophy as an active, neuroprogressive degenerative process. It is highly plausible that this structural deficit represents a static, congenital neurodevelopmental variant or a stable hypoplastic vulnerability that primed the patient’s early-onset behavioral phenotype and lowered her baseline threshold for subsequent pharmacological side effects.

The sudden emergence of generalized myoclonus and the highly pathological awake EEG findings—characterized by bursts of generalized 3–4 Hz spike-wave complexes—represent a critical counter-mechanism that warrants rigorous scientific deconstruction. This phenomenon generates the hypothesis of a potential drug-induced lowering of the seizure threshold, which could theoretically be mediated by a complex neuroanatomical mismatch [25,26]. Physiologically, the cerebellum exerts a continuous, modulating inhibitory tone over cerebral cortical structures via the cerebello-thalamo-cortical inhibition (CTCI) pathway. We hypothesize that the patient’s pre-existing, baseline cerebellar and vermis atrophy could theoretically have compromised this homeostatic inhibitory loop, potentially rendering the cerebral cortex electrophysiologically vulnerable. When a high-dose, long-acting atypical antipsychotic was introduced, its complex modulation of serotonin and dopamine receptors—which influence cortical GABAergic interneurons—likely altered cortical polarity and synchronization. In a structurally intact brain, this modification might remain subclinical; however, in this patient’s pre-compromised neuroanatomical substrate, the theorized loss of cerebello-thalamo-cortical inhibition combined with the drug’s electrophysiological footprint unmasked diffuse cortical hyperexcitability.

Crucially, the specific neuroelectrophysiological contribution of LAI aripiprazole cannot be isolated from the patient’s concurrent psychopharmacological regimen. Lithium carbonate possesses well-documented pro-convulsant properties at therapeutic or sub-toxic ranges, frequently inducing intentional tremors, myoclonic jerks, and generalized EEG slowing or paroxysmal discharges when co-administered with atypical antipsychotics. Furthermore, although oral methylphenidate was maintained at a conservative dose (10 mg/day), psychostimulants intrinsically enhance pro-excitatory catecholaminergic neurotransmission. This combination likely created an adverse synergistic effect, where lithium-induced cortical irritability and psychostimulant-mediated excitation converged on a structurally compromised, atrophic cerebellar substrate to precipitate the clinical myoclonus.

A critical differential diagnosis that must be rigorously debated given the patient’s demographic profile is Juvenile Myoclonic Epilepsy (JME). JME typically manifests during mid-to-late adolescence and is characterized pathognomonically by brief myoclonic jerks of the upper extremities, frequently accompanied by generalized 3–4 Hz spike-wave complexes on an awake EEG—identical to the neurophysiological footprint captured in our patient. Given the lack of a baseline EEG, we cannot exclude an underlying, genetically primed JME phenotype. However, the absence of prior childhood absence seizures and the strict temporal alignment with high-dose maintenance therapy strongly favor a drug-induced or polypharmacy-precipitated unmasking of cortical hyperexcitability. This phenotypic intersection highlights the emerging role of the cerebellum in global epilepsy networks. Far from being a purely motor or emotional structure, recent data confirm that the cerebellum acts as a powerful endogenous seizure modulator. Specifically, the cerebello-thalamo-cortical pathway can exert an inhibitory ‘braking’ action on cortical synchronization; we hypothesize that our patient’s baseline cerebellar vermis atrophy structurally compromised this protective brake, allowing the drug-associated electrophysiological footprint to easily precipitate paroxysmal activity. The rapid, total resolution of myoclonus following the addition of levetiracetam (500 mg BID)—a broad-spectrum agent highly effective in both JME and secondary cortical hyperexcitability—further supports this multi-hit network hypothesis.