Hyperinsulinism hyperammonaemia (HI/HA) syndrome due to GLUD1 mutation: phenotypic variations ranging from late presentation to spontaneous resolution.

Background The hyperinsulinism/hyperammonaemia (HI/HA) syndrome is the second most common cause of hyperinsulinaemic hypoglycaemia, caused by activating mutations in GLUD1. In this article, we report a series of three unrelated patients with HI/HA syndrome who demonstrated variable phenotypes, ranging from delayed presentation to spontaneous resolution of hypoglycaemia, thereby expanding the current knowledge and understanding of GLUD1 mutations. Case presentation This paper is a retrospective analysis of patients with HI/HA syndrome who demonstrated a variable disease course. Patient 1 presented with hypoglycaemic seizures at the age of 7 months and was diagnosed with HI/HA syndrome. Patient 2, a 5-year-old boy, on anti-convulsants since 8 months of age, was diagnosed with HI/HA at the age of 4 years. Patient 3, an 11-year-old girl with a history of transient neonatal hypoglycaemia, was diagnosed with HI/HA at the age of 12 months following evaluation for absence seizures. Patients 1 and 2 had raised ammonia levels, whilst patient 3 had normal ammonia level. The genetic analysis in all three patients confirmed GLUD1 mutation. Good glycaemic control was observed in all following diazoxide treatment. All patients have learning difficulties. Patient 1 demonstrated spontaneous resolution of hypoglycaemia at the age of 8 years, enabling discontinuation of diazoxide. Conclusions The cases highlight the diagnostic challenges in HI/HA syndrome due to a highly variable presentation. Knowledge of variable phenotypes would enable early diagnosis, thereby decreasing the risk of long-term neurological damage. Spontaneous resolution of hyperinsulinism could occur, and it is important to consider a trial off diazoxide therapy especially if the patients are on a small dose of diazoxide.

Continuous Flash Glucose Monitoring in children with Congenital Hyperinsulinism; first report on accuracy and patient experience.

BACKGROUND: The factory calibrated FreeStyle Libre (FSL) flash glucose monitoring system has been recently introduced for use in patients with diabetes mellitus. There are no reports available regarding its use in patients with congenital hyperinsulinism (CHI). We have assessed the accuracy of FSL compared to the finger prick capillary blood glucose (CBG) over 2 weeks period in patients with CHI and evaluated the parents' experience of using FSL. METHODS: Four hundred sixty-seven episodes of CBG along with corresponding swipe FSL readings were available from 11 children with CHI (0.5-5 years). A detailed questionnaire was completed by the parents. RESULTS: The mean variation between the two methods was 0.29 mmol/l (SD ±1.07), higher readings by FSL compared to CBG. The FSL sensors stayed in-situ for an average period of 11.5 days. There was a positive correlation between the two methods (r = 0.7). The FSL tended to overestimate compared to CBG (bias = 0.29 mmol/l; 95% CI: 0.19 to 0.38). Only 70% of values were within the reference standard (±0.83 mmol/l) at glucose concentrations less than 5.6 mmol/l. The overall Mean Absolute Relative Difference (MARD) was 17.9%. Forty two episodes of hypoglycaemia (CBG < 3.5 mmol/l) were noted but FSL identified only 52% of these episodes. The Bland Altman analysis showed the 95% limits of agreement between the two methods ranging from - 1.8 (95% CI: -1.97 to - 1.64) to 2.37 (95% CI: 2.21 to 2.54). Majority of the parents found the glucose trend on FSL to be useful to detect and prevent hypoglycaemic episodes. All parents felt that FSL is a very easy and convenient method to measure the glucose especially during sleep. A significant proportion of parents felt that FSL readings were not accurate and 56% of parents expressed interest to continue using FSL after the trial period. CONCLUSION: Noticeable variability between the two methods of measuring the glucose was noted. Despite the ease of using the FSL system, concerns related to accuracy, especially at low glucose values do remain although parents find the glucose trend to be very useful. Further larger trials are needed in CHI patients before FSL is recommended as a routine alternative method for measuring glucose levels.

Fluoxetine-Induced Hypoglycaemia in a Patient with Congenital Hyperinsulinism on Lanreotide Therapy.

Antidepressant drugs are reported to cause alterations in blood glucose homeostasis in adults with diabetes mellitus. We report a patient with persistent congenital hyperinsulinism (CHI) who developed recurrent hypoglycaemia following fluoxetine therapy. This 15-year-old girl was initially managed with diazoxide therapy. She developed troublesome hypertrichosis, which affected her quality of life adversely. Diazoxide was then slowly weaned and stopped with the introduction of octreotide, to which she responded well. Subcutaneous lanreotide (long-acting somatostatin analogue) was subsequently commenced (30 mg, once monthly) as injecting octreotide multiple times a day was proving to be difficult for the patient. The continuous blood glucose monitoring on monthly lanreotide injections revealed good glycaemic control. Six months later, she developed depression due to psychosocial problems at school. She was started on fluoxetine by the psychiatry team. She subsequently developed recurrent symptomatic hypoglycaemic episodes (blood glucose <3.5 mmol/L) and fluoxetine was discontinued, following which the hypoglycaemic episodes resolved within a week. Fluoxetine has been associated with hypoglycaemia in patients with diabetes mellitus. We report, for the first time, hypoglycaemia secondary to fluoxetine in a patient with CHI.