Commentary on studies investigating low‐dose diazoxide in hyperinsulinism

Diazoxide, utilised in hyperinsulinism since the 1960s,1 is the recommended first-line pharmacological therapy for children with hyperinsulinism.2 It blocks the sulfonylurea receptor 1 (SUR1) subunit of KATP-channel on pancreatic beta cells, increasing the permeability to potassium ions leading to hyperpolarization of the cells and inhibition of calcium-dependent insulin secretion.3 The recently published International Guidelines for the Diagnosis and Management of Hyperinsulinism2 recommends therapeutic diazoxide doses of 5–15 mg/kg/day aiming for plasma glucose concentration targets of 3.9–5.6 mmol/L (70–100 mg/dl). A lower hypoglycaemia threshold (3.5 mmol/L or 63 mg/dl) may be acceptable in children with severe forms of hyperinsulinism—where the recommended plasma glucose concentration targets may be difficult to achieve - depending on the severity and frequency of hypoglycemia and the availability other therapeutic options.2 Malhotra et al. and Ng et al. advocate for a shift to this paradigm, presenting their experience with the initial use of low-dose diazoxide (2–5 mg/kg/day) - escalating the dose if necessary—and adopting a plasma glucose concentration threshold for treatment of 3.5 mmol/L (63 mg/dl) as standard practice. These consensus parameters have been adopted in the UK.4 This sola dosis facit venenum approach aims to utilise the smallest effective dose to minimise the likelihood of adverse diazoxide-related side effects. Ng et al. reported that diazoxide doses of ≤5 mg/kg/day was sufficient to achieve therapeutic goals in preterm and term 34 neonates regardless of whether they were born small-for-gestational age (SGA). They reported 6 (17.6%) neonates having documented non-symptomatic hypoglycemia—the lowest being 3.3 mmol/L—with minimal adverse effects. In a larger cohort of 73 neonates, Malhotra et al. initiated low-dose diazoxide (2, 3 or 5 mg/kg/day) based on weight when commenced at a median age of 11 days (range 3–43) for a median duration of 4 months. They were able to achieve glycaemic stability in 65% of neonates without dose escalation, although 60% of those on 2 mg/kg/day (the 1000–2000 g cohort) required increased doses. Oedema (12%) and hyponatraemia (5%) were the most common adverse events but two neonates did develop necrotizing enterocolitis. The studies by Malhotra et al. and Ng et al investigating the effectiveness and safety of low-dose diazoxide for the management of hyperinsulinism in children adds to the literature first published by Chandran et al. who reported successful management of neonates born SGA with doses of diazoxide <5 mg/kg/day.5 Reports of the incidence rates of diazoxide-related adverse events vary, with the rate of pulmonary hypertension occurring in 2.4%–4.8% of infants with risk factors including being prematurity, and co-existing respiratory failure or structural heart defects.6, 7 Other reported rates of adverse events included oedema (18%), neutropenia (15.6%), thrombocytopenia (4.7%), and hyperuricemia (5.0%) (ref.11). A recent meta-analysis found similar rates of diazoxide-related adverse events.8 The rate of diazoxide-related necrotizing enterocolitis in preterm infants (33–37 weeks of gestation) is estimated to 4.4%, and 0.1%–0.7% in term infants.9 The rates of diazoxide-related adverse events reported by both Malhotra et al and Ng et al. were generally more favourable than the published literature. Although the factors associated with hyperinsulinism in neonates are well established, there are limited useful clinical or biochemical indicators allowing for the differentiation of transient vs persistent disease. A birth weight greater than the 90th percentile has been found to be associated with an eightfold increased risk of having a persistent form of HI (OR 8.8, 95% CI 2.5–30) and a 21-fold increased risk of being diazoxide unresponsive (OR 21.1, 95% CI 4.9–91.8).10 More research is required to allow for the accurate identification of children with transient forms of hyperinsulinism responsive to lower doses of diazoxide. The studies by Ng et al. and Malhotra et al. provide supplementary support for the safe initial use of low-dose diazoxide in hyperinsulinism. A lower hypoglycemia threshold is also an important feature of the suggested practice but must be predicated on adequate blood glucose monitoring, and prompt appropriate management of hypoglycemia. Long-term neurodevelopmental outcome data is lacking. The United Kingdom consensus standardising this practice4 will hopefully provide important additional information which may ultimately lead to a wider adoption. The authors declare no conflict of interest. Open access publishing facilitated by The University of Queensland, as part of the Wiley - The University of Queensland agreement via the Council of Australian University Librarians.