Ketogenic Diet for Super-refractory Status Epilepticus: A Case Series and Review of Literature.

Sir,

Super-refractory status epilepticus (SRSE) is a life-threatening condition with high mortality (16–23%).[1] There are no clear guidelines for its treatment. Failure of anesthetic agents has prompted the use of immunotherapy, intravenous magnesium, pyridoxine, hypothermia, resective surgery, and ketogenic diet (KD) in SRSE.

We share our experience and discuss the literature regarding use of KD in SRSE.

CASES

Our protocol for KD in SRSE is given in Figure 1. Details of MRI, EEG, and KD are given in Table 1.

Figure 1

Protocol for the use of ketogenic diet in superrefractory status epilepticus

Table 1

Clinical details, diagnosis, and treatment details of patients treated with ketogenic diet

Parameters	Patient 1 (60 years/F)	Patient 2 (27 years/M)	Patient 3 (24 years/M)	Patient 4 (7 years/M)
Diagnosis	Diabetes mellitus 2, hypertension, right MCA infarct, aspiration pneumonia, sepsis, super refractory status epilepticus	NORSE/Encephalitis of unknown origin	Superrefractory status epilepticus	Past encephaliits, HIE (poststatus epilepticus), refractory status epilepticus
		Superrefractory status epilepticus	In a diagnosed case of Lennox-Gastaut syndrome
GCS at admission	6/15	9/15	9 /15	13/15
Status epilepticus duration before admission (days)	3	5	1	6
Status epilepticus duration before initiation of ketogenic diet (days)	4	13	2	11
Antiepileptics used before initiation of intravenous anesthetics and ketogenic diet	Phenytoin, levetiracetam, valproate, lorazepam, clobazam	Phenytoin, levetiracetam, lacosamide, valproate, lorazepam, midazolam	Phenytoin, levetiracetam, loeazepam	Phenytoin, clobazam, levetiracetam, oxcarbamazepine
Intravenous anesthetic agents started on (day*), agent used	3, Midazolam	8, Thiopentone, midazolam	2, Midazolam	None
Ketogenic diet composition (fat:carbohydrates/proteins)	4:1	4:1	4:1	4:1
Ketogenic diet started on (day*)	4	14	2	12
Total duration of intravenous anesthetic agents (days)	3	20	9	None
Total duration of ketogenic diet	18	32 days	1 month	10 months
Antiepileptic drugs while on ketogenic diet	Phenytoin, clobazam	Valpraote, levetiracetam	Levetiracetam, valproate, phenobarbital	Phenytoin, clobazam, lacosamide, oxcarbamazepine, levetiracetam
Ketosis achieved after initiation of ketogenic diet (days)	2	4	2	5
Cessation of seizures after initiation of ketogenic diet (days)	2	9	5	7
Mechanical ventilation	Required	Required	Required	Not required
MRI brain [Figure 2]	Acute infarct in right middle cerebral artery territory	Multifocal T2/FLAIR white mater hyperintense lesions in both cerebral hemispheres	Normal	Right > left cerebral gyral hyperintensities
EEG	Generalized spike-wave activity (3-3.5 Hz) intermittent generalized delta-slowing and right frontal-temporal focal spikes. Regained background alpha rhythm by fifth day of ketogenic diet	Intermittent generalized sharp-slow wave activity. Regained normal background alpha rhythm by 10 th day of initiating ketogenic diet	Continuous 3.5-4 Hz generalized spike waves	Ictal EEG showed rhythmic theta activity in right frontocentral leads with rapid spread to left side
			Generalized slow spike-wave activity with intermittent multifocal spikes after control of status epilepticus	Theta-delta slowing without any epileptiform activity on first seizurefree day

*Day from the onset of refractory status epilepticus

Case 1

A 60-years-old lady presented with seizures leading to status epilepticus and left-hemiparesis for 3 days. She had left face/arm clonic seizures spreading to other side every few minutes. She had received lorazepam, loading dose of phenytoin sodium, levetiracetam, sodium valproate, and clobazam. In view of lower lobe pneumonia and sepsis, she was intubated and mechanically ventilated. She did not respond to midazolam (0.3 mg/kg/h). After 2 days of initiating classic KD, ketosis and seizure cessation were achieved. The KD was gradually weaned over 18 days. At 1 month, she was seizure free on phenytoin sodium and clobazam.

Case 2

A 27-year-old man presented with headache, vomiting, and altered behavior for 5 days. On fifth day, he developed generalized tonic–clonic seizures that evolved to status epilepticus. He received diazepam, lorazepam, loading dose of phenytoin, levetiracetam, and lacosamide over next 5 days. He was referred to us on midazolam infusion (0.3 mg/kg/h). On examination, his consciousness was 8/16 (FOUR score), no focal weakness, and extensor planters. His seizure continued despite thiopentone sodium (5 mg/kg bolus followed by 5 mg/kg/h infusion). In view of new-onset refractory status epilepticus and white-mater hyperintensities on MRI, he was given methylprednisolone (1 g/day) for 3 days with no response. On day 14, he was initiated on KD that improved his seizures. Patient remained seizure free on levetiracetam and clobazam. He was successfully weaned from KD over 18 days. Autoantibody panel and repeat cranial MRI were normal.

Case 3

A 24-year-old male, a known case of Lennox–Gastaut syndrome, developed generalized status epilepticus for 1 day after missing his antiepileptic drugs. Midazolam infusion (0.3 mg/kg/h) produced only transient seizure freedom. From day 2 of status epilepticus, patient was given KD (4:1). On day 5, patient had complete cessation of seizures. He is seizure free for 1 month on KD, sodium valproate, levetiracetam, and clobazam.

Case 4

A 7-year-old boy, a known case of postencephalitis sequalae (minimal conscious state and epilepsy) for 6 months, presented with generalized tonic–clonic status epilepticus for 6 days. Before visiting us, he had received lorazepam, levetiracetam, phenytoin sodium, oxcarbazepine, and clobazam. The child was not arousable. The family refused the use of intravenous anesthetic agents. He was started on KD (4:1); ketosis achieved on day 5, seizure cessation from day 7. The child is seizure free on KD, oxcarbazepine, and clobazam for 10 months.

DISCUSSION

The classic KD uses four-part fat to one-part protein and carbohydrate. The traditional way is to initiate KD after maintaining fasting for 1–3 days. Ketosis is measured by urinary ketones and serum beta-hydroxybutyrate levels.[2] The other diet forms with similar efficacy are modified Atkins diet (allows 20 g of carbohydrate and fat: protein in 60–70%:40–30%), low glycemic index treatment (allows 60 g carbohydrates of < 50% glycemic index), and medium chain fatty acid (MCT) variant of classic KD (uses MCT as fat source).[3] We preferred classic KD due to urgency of achieving ketosis. Contraindications to KD like persistent metabolic derangement, fatty acid oxidation disorders, pancreatitis, and severe dyslipidemia were excluded. Dextrose-free fluid was used and tablets replaced syrups [Figure 1]. We gave one-third of total required calories on first day, escalated to full KD dose by third day.

We compiled studies that used KD for SRSE in Table 2.[45678910111213141516171819] Response rate with KD in SRSE is not known due to lack of controlled trials. KD is postulated to be more effective in children than adults probably due to high levels of ketone-metabolizing enzymes in their brain, resulting in higher uptake of ketone bodies. With cerebral maturation, the number of monocarboxylic acid transporters in brain decreases. However, the ketone-dependent monocarboxylic acid transporters increase in adults in situations like ischemia, trauma, and sepsis. A multicenter study did not find relationship between KD outcome and age, sex, seizure type, or EEG findings.[18]

Table 2

Review of studies reporting the effect of ketogenic diet on refractory/superrefractory and nonconvulsive status epilepticus

Author (year)	n	Age (years)	Diagnosis (SE, RSE, NCSE)	Diet parameters					Etiology	Anesthetic drug duration	Adverse effects
				Diet type	Day of starting KD	Day of response	Duration of KD	Time to ketosis (days)
Francois et al. (2003)	6	49 M (mean)	RSE	KD	NA	NA	8 day-2 M	NA	NA	NA	NA
Bodenant et al. (2008)	1	54	RSE	KD	31	7	NA	NA	Epileptic encephalopathy with pneumonia	4 days	None
Nabbout et al. (2010)	9	5-8	RSE	KD	2-4	NA	NA	4-6	FIRES Frontal lobe epilepsy, subcortical band heterotropia	NA	None
Kumada et al. (2010)	2	4, 5	NCSE	MAD	NA	5, 10	NA	NA	NA	NA	None
Wusthoff et al. (2010)	2	29, 34	NCSE	KD	20, 101	6, 11	NA	8-10	Viral encephalitis	NA	Acidosis
Nam et al. (2011)	5	4, 8, 10, 14, 40	RSE	KD	15	8	NA	NA	Encephalitis	NA	None
Martikainen et al. (2012)	1	26	NCSE	LGIT	4	4	NA	NA	POLG	NA	None
Strzelczyk et al. (2013)	1	14	SRSE	KD	16	4	NA	4	Lafora disease	NA	None
Thakur et al. (2014)	10	33 (mean)	SRSE	KD	2-60	1-31	NA	1-7	NORSE, NMDA, LGI1, Anoxia, Neurocysticercosis, Band heterotopia	0-6 days	Acidosis Elevated TG
Lin et al. (2015)	1	6	SRSE	KD	1	70 h	NA	1	NORSE	Stopped prior to KD	Elevated TG, chol
Fung et al. (2015)	4	6, 8, 16, 16	SRSE	KD	12-21	NA	NA	NA	NORSE, FIRES, VGKC-positive AE	NA	NA
Cobo et al. (2015)	4	9 weeks—13.5 years	SRSE	KD	19-36	2-14	NA	NA	CPE, MMPSI, TS	2-14 days	Nephrolithiasis, asymptomatic hypoglycemia, constipation
Amer et al. (2015)	1	21	RSE	KD	21	14	NA	NA	NMDA-positive AE	NA	NA
Cervenka et al. (2017)	15	19-86 (mean 38)	SRSE	KD	2-21	0-10	NA	0-16	NORSE, LGS, ICH, anoxia, GBM, encephalitis, NAT	72 h before initiating KD	Acidosis, GI, hyperlipidemia, hypoglycemia, hyponatremia, weight loss, elevated TG
Park et al. (2018)	16	5-13.5 (mean 8)	SRSE	KD	3-420	NA	0.1-15.8 months	2-6	FIRES, encephalitis (enterovirus, HSV), hemimegalencephaly, cryptogenic FLE	1-28 days	Gastro intestinal disturbances, hypoprotenemia, hypercholesterolemia, lipoid aspiration pneumonia

AE—Autoimmune encephalitis, CPE—cryptogenic preexisting epilepsy, FIRES—febrile infection-related epilepsy syndrome, GBM—glioblastoma multiforma, GI—gastrointestinal side effects (including constipation), ICH—intracranial hemorrhage, KD—ketogenic diet, LGI1—leucine-rich, gliomainactivated 1 encephalitis, LGS—Lennox-Gastaut syndrome, MAD—modified Atkin diet, MMPSI—malignant migrating partial seizures of infancy, NA—not available, NAT—remote nonaccidental trauma resulting in epilepsy, NMDA—N-methyl D-aspartate receptor encephalitis, NORSE—new-onset refractory status epilepticus, POLG—mitochondrial polymerase c-related epilepsy, RE—refractory epilepsy, SE—status epilepticus, TG—triglycerides, TS—tuberous sclerosis, VE—viral encephalitis, VGKC—voltage-gated potassium channel antibody

The optimal timing of initiating KD and its duration in SRSE have not been established. KD showed similar efficacy, whether used early or after the seizure became refractory.[19]

A recent review found no significant difference between KD responders and KD nonresponders based on age of seizure-onset, etiology, seizure types, number of previously tried anti-epileptic drugs/anesthetic agents, duration of SRSE before initiating KD, time to ketosis, and KD duration.[17] Specifically, good response to KD was seen in children with febrile infection-related epilepsy syndrome.[917] KD has been successfully used in nonconvulsive status epilepticus.[679] Most people used classic KD. Only few used modified Atkins diet and low glycemic index diet.[69] The time to freedom from status epilepticus ranged from zero to 10th day after initiation of KD. Time to ketosis ranged from 1 to 16 days. Most authors used KD after failure to control status epilepticus with anesthetic agents. The common adverse effects of KD were gastrointestinal disturbances, hypoglycemia, hyponatremia, hypertriglyceridemia, hypoproteinemia, metabolic acidosis, nephrolithiasis, and weight loss.

(a) Cranial MRI diffusion-weighted sequence showing acute right middle cerebral artery infarct, (b) T2 FLAIR-weighted MRI of brain showing bilateral cerebral white-mater hyperintensity, and (c) T2 FLAIR-weighted MRI of brain showing gyral hyperintensities on right cerebral hemisphere

KD can be used in diabetics by closely monitoring for hypoglycemia and ketoacidosis. It can be rapidly withdrawn in diabetics after the control of status to prevent hyperlipidemia and weight loss/gain.

Use of KD for SRSE is a team effort. A trained dietician should institute and monitor KD. ICU staff should avoid glucose-containing fluids and syrups. KD poses a theoretical risk of renal stones if combined with zonisamide and topiramate. Poor palatability is other concern that can be managed by using 2:1 formulation or other KDs.

We conclude that KD is effective in controlling the SRSE irrespective of the age of the patient, etiology, duration of status epilepticus, and prior use anesthetic agents/antiepileptic drugs.

Declaration of patient consent

The authors certify that they have obtained all appropriate patient consent forms. In the form the patient(s) has/have given his/her/their consent for his/her/their images and other clinical information to be reported in the journal. The patients understand that their names and initials will not be published and due efforts will be made to conceal their identity, but anonymity cannot be guaranteed.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.