Which anti-seizure medications (ASMs) are optimal alongside ketogenic diet therapy (KDT)? Although commonly considered by clinicians, evidence has been limited. Sharma et al. advance the field by demonstrating in 127 children that five ASMs are associated with KDT efficacy and none are linked to worse outcomes. ^{Reference Sharma, Aschner, Kabir and Donner1}

Notably, vigabatrin showed a strong association with KDT response (OR 5.58), although the authors appropriately flag the likely confound: vigabatrin is predominantly used for epileptic spasms, which respond well to KDT. ^{Reference Sharma, Aschner, Kabir and Donner1} However, beyond this confounder, vigabatrin and KDT independently modulate the mechanistic target of rapamycin (mTOR) pathway, which is a key regulator of neuronal excitability and epileptogenesis, suggesting possible synergistic effects. ^{Reference Schubert-Bast and Strzelczyk2} This is especially relevant in tuberous sclerosis complex and cortical malformations, which are associated with mTOR hyperactivation. Prospective studies combining KDT with mTOR inhibitors are needed.

The topiramate association (OR 3.19) ^{Reference Sharma, Aschner, Kabir and Donner1} may reflect shared mechanisms, as they both affect carbonic anhydrase and GABA signaling, with conceivable additive effects. ^{Reference Morrison, Pyzik and Hamdy3} However, bicarbonate should be monitored for risk of metabolic acidosis. The lamotrigine finding ^{Reference Sharma, Aschner, Kabir and Donner1} (OR 5.00) is also reassuring, given that prior literature suggests KDT may reduce lamotrigine levels through UGT1A6 induction and lamotrigine can potentially impair ketosis. ^{Reference Morrison, Pyzik and Hamdy3}

Valproate and carbamazepine showed no association, ^{Reference Sharma, Aschner, Kabir and Donner1} although they require monitoring for hepatic and lipid effects and pharmacokinetic interactions with KDT. ^{Reference Morrison, Pyzik and Hamdy3} Phenobarbital’s neutral effect contrasts with earlier reports of reduced KDT efficacy; however, drug-level monitoring remains prudent. ^{Reference Armeno and Kossoff4} The discordance among studies likely reflects syndrome-level confounding and variable dosing and drug levels, underscoring the difficulty of isolating individual ASM effects in a polytherapy population.

Absent from this study are newer agents including cannabidiol and fenfluramine. The cannabidiol–KDT combination is of particular emerging interest. Pharmacokinetically, a fivefold increase in C-max with high-fat meals has been documented, suggesting KDT may meaningfully enhance cannabidiol bioavailability. ^{Reference Armeno and Kossoff4} In the BC Children’s Hospital experience of cannabidiol, 4/4 children on cannabidiol-KDT had >50% seizure reduction, compared to 22% overall. ^{Reference Kang, Grewal and Boelman5} Although small numbers, this aligns with pharmacokinetic rationale and supports further study. Beyond ASMs, therapies like vagal nerve stimulation with KDT may provide additional benefit, with possible synergistic seizure reduction through shared neural modulation. ^{Reference Kossoff, Pyzik and Rubenstein6}

Ultimately, this study supports current clinical insight: ASM selection alongside KDT should be driven by electroclinical phenotype and tolerability. The absence of any negatively associated ASM is itself clinically valuable. However, the field urgently needs prospective, mechanistically informed studies, particularly as newer agents enter routine practice, to move beyond association toward actionable, individualized guidance.