Objective To assess the efficacy and safety of two different protocols of ketogenic diet (KD)-eating on demand or eating at regular intervals for refractory epilepsy in children.Methods Sixty children with refractory epilepsy were randomly divided into eating on demand group (n =30) and eating at regular intervals group (n =30) by random number table method.After taking the whole amount of KD,the capillary blood ketone and glucose level and urine ketone were monitored every 6 hours in 72 continuous hours.Seizure frequency and onset time were recorded.Antiepileptic efficacy and diet tolerability of the two groups were evaluated on 4 weeks,12 weeks,24 weeks and 48 weeks after initiating the diet.Adverse effects were monitored.Results After treatment of 4 weeks,the complete seizure remission rates of eating on demand group and eating at regular intervals group were 33.3％ (10/30) and 30.0％ (9/30) respectively,which suggested a comparable efficacy for two groups (P ＞ 0.05).The day when KD started to work was averaged (6.18 ± 2.42) d and (8.63 ± 2.63) d respectively.The group of eating on demand showed a faster onset of action (P ＜0.05).After treatment of 12 weeks,24 weeks and 48 weeks,complete seizure remission rates of eating on demand group were 30.0％ (9/30),34.8％ (8/23) and 36.8％ (7/19) respectively;the eating at regular intervals group were 33.3％ (10/30),30.4％ (7/23) and 44.4％ (8/18) respectively.The two groups had no significant difference (P ＞ 0.05).One year later,the treatment retention rates of the two groups were 63.3 ％ (19/30) and 60.0％ (18/30) respectively.There was no significant difference (P ＞ 0.05).The adverse effects mainly including transient gastrointestinal symptoms and metabolic disturbances were mostly tolerable and curable.Conclusion The two different protocols of KD-eating on demand and eating at regular intervals are both effective and well-tolerated for refractory epilepsy in children.While protocol of eating on demand is more easier to achieve ketotic state and the effect is more quickly,so it can be more easily received by children.Therefore in clinical practice,we can choose flexible eating time according to children's eating habits,which can improve the therapeutic compliance.

Objective To explore the efficacy of ketogenic diet( KD) in the treatment of status epi-lepticus( SE) and whether KD could protect the brain,and propose a new thought on SE patients′reasonably individualized treatment, brain protection and prognosis improvement. Methods From Sep 2013 to Jan 2015,all the patients diagnosed as SE were advised to apply KD treatment; the patients who refused KD treatment were included in the control group,while the patients who accepted KD treatment were included in the treatment group. Based on the SE treatment principles,the control group applied traditional anti-convulsive therapy,while the treatment group applied traditional therapy combined with KD treatment. Before the treat-ment and after the epilepsy control,the patients′ serum was collected to test neuron specific enolase( NSE) and S100βlevels,and the duration of epilepsy control was recorded. Results The treatment group included a total of 10 patients; 3 patients had a good efficacy and obtained seizure-free after the treatment; clinical seizures declined significantly in 6 patients. The treatment group′s overall response rate was 9/10,which was higher than that of the control group(5/8)(P<0. 01). The treatment group′s duration to gain efficacy was shorter than that of the control group[(5. 2 ± 2. 9) d vs. (9. 8 ± 1. 5) d,P<0. 01]. After the treatment,the patients′NSE and S100β in both groups were significantly decreased than those before the treatment ( P<0. 001 or P<0. 05). After the treatment,the serum NSE and S100β of the patients in the treatment group were lower than those in the control group,with statistically significant difference(P<0. 05). Conclusion Frequent epileptic seizures and SE would impair the patient′s brain. Controlling the epileptic seizures actively could lower the severity of brain injury. KD could effectively control the epileptic seizure and had neuropro-tective effects.

Objective To analyze the clinical manifestations of familial acute necrotizing encephalopathy (ANE)and to improve the recognition of this disease. Methods The clinical data of a 25 - month - old girl with fa-milial and recurrent ANE with evidence of mutation in the RANBP2 gene were collected and analyzed,and the gene examination of their family members was performed. Results A previously healthy girl experienced recurrent ANE epi-sodes at the ages of 8 months,18 months and 25 months,respectively. At each beginning of each episodes the patient presented with lethargy and tremor of limbs following febrile illness of 3 - 4 days,even developed coma and convulsions in the last time. Brain magnetic resonance imaging showed bilateral and high T2 signal changes in thalamus,cerebellum and hippocampus. Abnormal signals also appeared in the brainstem,claustrum,corpus scallosum and cortex(temporal, parietal and cingulate)also appeared abnormal signals. Spinal MRI showed spinal cord involvement. The girl recovered after her first episode;she could speak but could not walk steadily after the second time;after the third episode,al-though she regained consciousness from coma,she could no longer speak or walk. The patient's sister died of encephali-tis at the age of 18 months. Her paternal uncle had suffered from dysnoesia from meningitis at his 17 months of age. The patient and her grandmother,father,uncle and one of her aunts harbored a mutation(c. 1754C ﹥ T)in RANBP2 gene. Conclusions Familial ANE has typical clinical manifestations and characteristic MRI findings. The patient with recur-rent history,especially with positive family history,should have the mutation in RANBP2 gene detected earlier in order to clarify the diagnosis of ANE.

Objective To compare the clinical and electroencephalogram(EEG)characteristics and therapeutic response prognosis of different age groups of children with epileptic spasm .Methods From January 2002 to October 2011 the clinical data ,EEG fea-tures of epileptic spasms children under 15 years old with unknown disease cause(cryptogenic) were retrospectively reviewed .74 of them were followed up for 12 to 92 months .All of them were divided into two groups on the basis of onset age :3 -12 months of onset as group infantile-onset spasms(group IOS ,n=60);and 12 months to 7 years old of onset as group late-onset epileptic spasms (group LOS ,n=14) .Clinical process ,seizure semiology and EEG features were compared between two groups .Results Semiologic features of two groups were similar ,but they showed differences in interictal EEG features including the background ,the location of discharges ;The response to drugs between the two groups are also different .Conclusion There are differences between group IOS and group LOS when comparing EEG features and response to drugs .