Abstract
Objective
Mutations in the KATP channel genes, ABCC8 and KCNJ11, are the most common cause of congenital hyperinsulinism. The diagnosis of KATP-hyperinsulinism is important for the clinical management of the condition. We aimed to determine the clinical features that help to identify KATP-hyperinsulinism at diagnosis.
Design
We studied 761 individuals with KATP-hyperinsulinism and 862 probands with hyperinsulinism of unknown aetiology diagnosed before 6 months of age. All were referred as part of routine clinical care.
Methods
We compared the clinical features of KATP-hyperinsulinism and unknown hyperinsulinism cases. We performed logistic regression and receiver operator characteristic (ROC) analysis to identify the features that predict KATP-hyperinsulinism.
Results
Higher birth weight, diazoxide unresponsiveness and diagnosis in the first week of life were independently associated with KATP-hyperinsulinism (adjusted odds ratio: 4.5 (95% CI: 3.4–5.9), 0.09 (0.06–0.13) and 3.3 (2.0–5.0) respectively). Birth weight and diazoxide unresponsiveness were additive and highly discriminatory for identifying KATP-hyperinsulinism (ROC area under the curve for birth weight 0.80, diazoxide responsiveness 0.77, and together 0.88, 95% CI: 0.85–0.90). In this study, 86% born large for gestation and 78% born appropriate for gestation and who did not respond to diazoxide treatment had KATP-hyperinsulinism. In contrast, of those individuals born small for gestation, none who were diazoxide responsive and only 4% of those who were diazoxide unresponsive had KATP-hyperinsulinism.
Conclusions
Individuals with hyperinsulinism born appropriate or large for gestation and unresponsive to diazoxide treatment are most likely to have an ABCC8 or KCNJ11 mutation. These patients should be prioritised for genetic testing of KATP channel genes.
Congenital hyperinsulinism (CHI) is a potentially life-threatening disorder characterised by inappropriate high levels of insulin at the time of hypoglycaemia. The incidence of CHI is estimated at 1 in 28 000 to 1 in 50 000 live births in European populations but rises to approximately 1 in 2500 in countries with high rates of consanguineous unions (1, 2, 3).
Mutations in approximately 15 genes have been reported to cause isolated CHI or a multi-system syndromic disease, where CHI is a rare feature (4). Screening these genes identified a mutation in 36–69% of the cases (5, 6). Inactivating mutations in the ABCC8 and KCNJ11 genes, encoding the SUR1 and Kir6.2 subunits of the pancreatic ATP-sensitive potassium (KATP) channel, are responsible for 80–84% of confirmed monogenic cases of CHI (5, 6).
Rapid screening of the KATP channel genes is critical for informing surgical and medical management of CHI (7). Identifying a dominantly-acting or bi-allelic ABCC8 or KCNJ11 mutation confirms a diagnosis of diffuse pancreatic disease, which is preferentially managed with medical treatment, whereas a paternally inherited recessively acting ABCC8 or KCNJ11 mutation suggest a focal pancreatic lesion which can be cured by lesionectomy (8).
In this study, we undertook genetic testing of the KATP channel genes in a large cohort of children with CHI presenting before the age of 6 months with the aim to assess whether clinical features at presentation could predict which individuals were most likely to have monogenic CHI due to a KATP channel mutation.
Methods
Study population and genetic analysis
We studied 761 probands with a pathogenic or likely pathogenic mutation in ABCC8 (n = 665) or KCNJ11 (n = 96) identified at the Exeter Genomics Laboratory between 2002 and 2018. We also reviewed 862 patients with CHI of unknown genetic aetiology who were referred to our laboratory within the same period. In all of these patients, mutations in the ABCC8 and KCNJ11 genes had been excluded by Sanger sequencing or gene panel testing.
Analysis of the coding regions and intron–exon boundaries of ABCC8 and KCNJ11 was performed using previously described methods (9, 10). When available, parental samples were tested to confirm the inheritance of ABCC8/KCNJ11 variants identified in the proband. Pathogenicity of variants was assessed according to the American College of Medical Genetics and Genomics guidelines (11).
Clinical information was provided at referral for genetic testing using a standardised request form and included sex, ethnicity, birth weight, gestational age at birth, age at diagnosis of CHI, biochemical measurements at diagnosis (including insulin and glucose), current treatment, response to diazoxide treatment if attempted and the presence of additional features. Although a precise, consensus definition of diazoxide responsiveness is lacking, diazoxide unresponsiveness in this study can be broadly defined as persistent hypoglycaemia despite treatment with maximal dose diazoxide indicating the need for additional therapies to achieve euglycaemia. Consanguinity was defined as the parents of the proband being related as second cousins or closer or whether the proband was referred from a country with high reported rates of consanguinity (12).
Informed consent was obtained from the parents or guardians of all probands. This study was approved by the North Wales Research Ethics Committee (517/WA/0327).
Statistical analysis
Birth weight Z score and corrected birth weight for sex and gestation were generated using World Health Organization (WHO) standards accessed through the Zanthro package in Stata (13). Small for gestational age (SGA) was defined as a birth weight lower than the 10th centile, and large for gestational age (LGA) was defined as a birth weight greater than the 90th centile. Age at diagnosis followed a skewed distribution, so it was analysed categorically between those diagnosed in the first week of life and those diagnosed after the first week. As insulin (pmol/L) level followed a right skewed distribution, values were log transformed for the statistical analysis.
Individuals with an ABCC8 or KCNJ11 mutation were combined (referred to as KATP channel mutations hereafter) as no difference in clinical features was observed between the two genetic subgroups (Supplementary Table 1, see section on supplementary materials given at the end of this article). Statistical analyses were performed to determine differences in the clinical features of patients with an unknown aetiology compared to patients with a KATP channel mutation. Two-tailed P values were calculated to determine statistical significance using Pearson’s chi-squared test for categorical variables, and Student’s t-test for continuous variables.
Univariate and multivariable logistic regression were used to assess which clinical features were independently predictive of having a KATP channel mutation. The area under the curve (AUC) of the receiver operator characteristic (ROC) was used to assess the discriminatory ability of the clinical features to identify individuals with a KATP channel mutation from those without a mutation. Stata/SE 16.0 (Stata Corp) was used to perform statistical analyses.
Results
Seven of the clinical features studied were different between individuals with KATP CHI and those with CHI of unknown aetiology
Individuals with KATP channel mutations (n = 761) had higher birthweights (mean 4333g vs 3512g, P = 6 × 10−94, mean difference 821 g, 95% CI: 748–894g), were more likely to be diagnosed in the first week of life (85% vs 72%, P = 1 × 10−9), had a higher insulin level at diagnosis (mean 162.2pmol/L vs 115.4pmol/L, P = 1 × 10−8) and were less likely to respond to diazoxide, the mainstay treatment for CHI (32% vs 88%, P = 2 × 10−84) compared to individuals with CHI of unknown genetic aetiology (n = 862) (Table 1). They were more likely to be female (46% vs 36%, P = 5 × 10−5), have consanguineous parents (52% vs 34%, P = 2 × 10−12) and were less likely to be Caucasian (30% vs 52%, P = 3 × 10−18) compared to CHI of unknown genetic aetiology. Glucose levels at presentation and the number of individuals with extra-pancreatic features at referral were similar between the two groups.
Characteristics of individuals with congenital hyperinsulinism (CHI) with confirmed KATP channel mutations and of unknown aetiology. Categorical data are shown as n (%) whereas continuous data are shown as mean ± s.d. Number of patients (n) is presented where it differs from the total number in the cohort.
| Characteristics | CHI with confirmed KATP channel mutations | CHI of unknown aetiology | P value | ||
|---|---|---|---|---|---|
| Values | n | Values | n | ||
| Total n | 761 | 862 | – | ||
| Age at diagnosis | – | – | 1 × 10−9 | ||
| ≤7 days | 85% | 644 | 72% | 622 | – |
| >7 days | 15% | 117 | 28% | 240 | – |
| Female sex | 46% | 352 | 36% | 313 | 5 × 10−5 |
| Corrected birth weight (g) | 4333 ± 718 | 746 | 3512 ± 762 | 843 | 6 × 10−94 |
| Birth weight Z score | 1.64 ± 1.50 | 746 | −0.13 ± 1.66 | 843 | 1 × 10−95 |
| Birth weight categories | – | – | 8 × 10−93 | ||
| LGA | 56% | 415/746 | 18% | 151/843 | – |
| AGA | 44% | 329/746 | 57% | 478/ 843 | – |
| SGA | 0.3% | 2/746 | 25% | 214/843 | – |
| Additional features | 13% | 102 | 18% | 152 | 0.02* |
| White ethnicity | 30% | 224/736 | 52% | 438/841 | 3 × 10−18 |
| Consanguineous parents | 52% | 394 | 34% | 296 | 2 × 10−12 |
| Glucose (mmol/L) | 1.6 ± 0.7 | 653 | 1.7 ± 0.8 | 740 | 0.02* |
| Insulin (pmol/L) | 162.2 ± 3.0 | 666 | 115.4 ± 2.9 | 715 | 1 × 10−8 |
| Diazoxide responsive | 32% | 160/495 | 88% | 521 | 2 × 10−84 |
AGA, appropriate for gestational age; LGA; large for gestational age; SGA, small for gestational age.
*P value above the multiple comparison threshold = 0.05/11
Birth weight and diazoxide unresponsiveness are the most discriminative, independent and additive for identifying children with KATP CHI
To identify the independent features that can help in discriminating KATP CHI from unknown cases, we performed a multivariable logistic regression analysis. We showed that higher birth weight, diazoxide unresponsiveness and a diagnosis within the first 7 days of life were independent predictors of KATP CHI after adjustment for all variables (Table 2). We next performed ROC AUC analysis to assess the discriminatory ability of these variables. The ROC AUC for birth weight was 0.80 (95% CI: 0.77–0.82), 0.77 for diazoxide responsiveness (95% CI: 0.74–0.80) and 0.59 for age at diagnosis (95% CI: 0.57–0.62). Combining birth weight with diazoxide unresponsiveness increased the ROC AUC to 0.88 (95% CI: 0.85–0.90). The addition of other factors (age at diagnosis, sex, ethnicity, consanguinity and insulin level) only marginally increased ROC AUC (0.89, 95% CI: 0.87–0.91) over the combination of birth weight and diazoxide unresponsiveness (Fig. 1 and Table 2). To establish if our results were consistent across institutions that may have different standards in measuring these clinical features, we performed a sensitivity analysis with the two institutions that referred the most patients, along with the two countries with the most referred patients. We observed similar results in this analysis, suggesting the presence of homogeneity in clinical features despite different patient populations and clinical practice (Supplementary Table 2).
Receiver operating curves analysis showing the discriminating ability of clinical features to identify individuals with KATP congenital hyperinsulinism (CHI) from those with unknown aetiology. All includes birth weight, diazoxide responsiveness, age at diagnosis, sex, insulin level at diagnosis, ethnicity and consanguinity. DR, diazoxide responsiveness; BW, birth weight.
Citation: European Journal of Endocrinology 185, 6; 10.1530/EJE-21-0476
Receiver operating curves analysis showing the discriminating ability of clinical features to identify individuals with KATP congenital hyperinsulinism (CHI) from those with unknown aetiology. All includes birth weight, diazoxide responsiveness, age at diagnosis, sex, insulin level at diagnosis, ethnicity and consanguinity. DR, diazoxide responsiveness; BW, birth weight.
Citation: European Journal of Endocrinology 185, 6; 10.1530/EJE-21-0476
Receiver operating curves analysis showing the discriminating ability of clinical features to identify individuals with KATP congenital hyperinsulinism (CHI) from those with unknown aetiology. All includes birth weight, diazoxide responsiveness, age at diagnosis, sex, insulin level at diagnosis, ethnicity and consanguinity. DR, diazoxide responsiveness; BW, birth weight.
Citation: European Journal of Endocrinology 185, 6; 10.1530/EJE-21-0476
Odds ratios for clinical characteristics from logistic regression model for identifying individuals with KATP CHI from CHI of unknown aetiology. Table shows odds ratio (95% CI) for univariate and multivariate analysis after adjusting for all variables.
| Characteristics | Unadjusted odds ratio (95% CI) | P value | Adjusted odds ratio (95% CI) | P value |
|---|---|---|---|---|
| Diagnosed in first week of life | 2 (1.6–2.5) | 3 × 10−10 | 3.3 (2.0–5.0) | 2 × 106 |
| Female sex | 1.5 (1.2–1.8) | 5 × 10−5 | 0.9 (0.6–1.3) | 0.6 |
| Corrected birth weight | 4.4 (3.7–5.2) | 3 × 10−68 | 4.5 (3.4–5.9) | 2 × 10−26 |
| Caucasian ethnicity | 2.4 (2.0–3.0) | 9 × 10−18 | 1.5 (0.9–2.6) | 0.1 |
| Consanguinity | 2.0 (1.6–2.4) | 7 × 10−12 | 1.9 (1.1–3.3) | 0.02 |
| Insulin at diagnosis | 1.9 (1.5–2.4) | 9 × 10−9 | 0.7 (0.5–1.0) | 0.08 |
| Diazoxide responsive | 0.08 (0.06–0.11) | 2 × 10−65 | 0.09 (0.06–0.13) | 2 × 10−36 |
Individuals born appropriate or large for gestation who did not respond to diazoxide had the highest likelihood of KATP CHI
Of the 220 individuals born large for gestational age who were unresponsive to diazoxide, 86% (95% CI: 81–91%) had KATP CHI, while no mutations were detected in those born small for gestational age who responded to diazoxide (Fig. 2). Of the 179 individuals born appropriate for gestational age (AGA) who were unresponsive to diazoxide, 78% (95% CI: 71–84%) had KATP CHI, while of the 389 individuals born AGA who were responsive to diazoxide, only 18% (95% CI: 14–22%) had KATP CHI.
Proportion of KATP CHI by diazoxide responsiveness and birth weight categories. Light grey bars represent the percentage number of individuals without a KATP channel mutation, dark grey bars represent the percentage number of cases with a KATP channel mutation. SGA, small for gestational age; AGA, appropriate for gestational age; LGA, large for gestational age.
Citation: European Journal of Endocrinology 185, 6; 10.1530/EJE-21-0476
Proportion of KATP CHI by diazoxide responsiveness and birth weight categories. Light grey bars represent the percentage number of individuals without a KATP channel mutation, dark grey bars represent the percentage number of cases with a KATP channel mutation. SGA, small for gestational age; AGA, appropriate for gestational age; LGA, large for gestational age.
Citation: European Journal of Endocrinology 185, 6; 10.1530/EJE-21-0476
Proportion of KATP CHI by diazoxide responsiveness and birth weight categories. Light grey bars represent the percentage number of individuals without a KATP channel mutation, dark grey bars represent the percentage number of cases with a KATP channel mutation. SGA, small for gestational age; AGA, appropriate for gestational age; LGA, large for gestational age.
Citation: European Journal of Endocrinology 185, 6; 10.1530/EJE-21-0476
Discussion
In this large study, we showed that simple clinical features such as birth weight and diazoxide responsiveness were independent and highly predictive for identifying individuals with KATP CHI from those with CHI of unknown aetiology.
We show that patients with KATP CHI were ~830 g heavier compared to CHI due to an unknown cause. Higher birth weight has been reported previously in patients with KATP CHI (14, 15) and with persistent CHI (16). However, previous studies generally did not include control cases without KATP mutations, were smaller in size, and were limited to single centres impacting on the ability to statistically assess the importance of birth weight for identifying KATP CHI. We also show that birthweight and other clinical features were comparable between ABCC8 and KCNJ11 CHI, thus this finding is applicable to both KATP CHI subtypes (Supplementary Table 1). The higher birth weight in the KATP CHI patients is in keeping with the onset of hyperinsulinism with KATP CHI in utero, as insulin acts as a growth factor in pregnancy (17). This is supported by the observation that most of the children with CHI are diagnosed in the first week of life.
Diazoxide is the most common medical treatment for CHI which suppresses insulin secretion by binding to and opening the KATP channel (18). 78% of patients in our cohort who were unresponsive to diazoxide treatment had a KATP channel mutation, in keeping with the mutations leading an absence of channels or channels with disrupted function. Previous studies reported similar levels of diazoxide unresponsiveness in patients with KATP CHI (5, 6). Therefore, our finding is not unexpected, but due to large control cases, we were able to quantify the importance of lack of diazoxide response in identifying KATP CHI. Additionally, there was no difference in diazoxide responsiveness between patients with mutations in ABCC8 and KCNJ11, thus our findings are applicable to both genetic subtypes (Supplementary Table 1). Given that diazoxide is not universally available, it will be important to perform further studies to assess whether responsiveness to other treatments (including somatostatin analogues) can help to predict KATP channel hyperinsulinism.
Our study has important clinical and research implications. Using the largest cohort of CHI cases referred from routine clinical practise, we robustly show that birth weight and response to diazoxide treatment can be used to help and guide genetic testing in patients with CHI. Our cohort also included patients from multiple centres, of different ethnicities, and with different modes of inheritance for mutations in the KATP channel genes suggesting that our findings are applicable to patients worldwide. More than 80% of the cases born LGA or AGA who were unresponsive to diazoxide had KATP channel mutations, and therefore every effort should be made to prioritise genetic testing in these individuals as finding a KATP mutation can guide clinical management (7). Current Pediatric Endocrine Society guidelines on hypoglycaemia lack recommendation on criteria that can be used to prioritise patients for genetic testing for CHI (19). We believe our study provides important evidence in this regard which is especially relevant for resource-poor countries, where access to genetic testing is limited (19). Furthermore, the findings from this study will help to prioritise patients for genetic discovery studies for patients with an unknown aetiology. Patients born large for gestational age and who do not respond to diazoxide have an 86% chance of harbouring a mutation in a KATP channel gene (Fig. 1). Therefore, patients in this category without a monogenic cause are good candidates to screen for pancreas-specific mosaic mutations and regulatory mutations affecting the KATP channel genes (20).
We acknowledge that our study has some limitations. We used clinician reported assessment of diazoxide responsiveness, thus it is likely to vary by referring clinician. However, the rate of diazoxide unresponsiveness in KATP CHI is similar to studies that used a single definition for this feature (5, 6). Furthermore, the lack of consistent definition would have reduced the discriminatory ability rather than inflated the estimate suggesting that a robust definition will provide improved or similar results to our study. Our study did not include comprehensive data on whether hyperinsulinism was transient or persistent in each patient as we wanted to identify features at diagnosis which could help predict a KATP channel mutation. However, enrichment of male and SGA patients in the unknown aetiology group suggests the presence of transient hyperinsulism as reported by a recent study (21). We used clinician reporting and country of referral to infer consanguinity. Although this approach has been used previously (22), we recognise that this has likely increased the proportion of consanguinity reported in our cohort. Despite this, consanguinity was significantly enriched in patients with KATP-hyperinsulinism. This suggests that a more accurate definition will further increase its utility in identifying KATP-hyperinsulinism. Finally, our study finding is not applicable to cases of non-KATP monogenic CHI. We chose to focus on KATP CHI as they are responsible for more than 80% monogenic CHI and finding these mutations is of great clinical importance.
In conclusion, this study shows that birth weight, diazoxide responsiveness and presenting with CHI in the first week of life are independent and highly predictive for KATP channel mutations in patients with congenital hyperinsulinism. Patients born normal or large for gestational age who are unresponsive to diazoxide are most likely to have KATP channel mutations and should be referred for genetic testing. Our study provides robust results that will help to shape future guidelines on genetic testing for CHI and is applicable to patients worldwide.
Supplementary materials
This is linked to the online version of the paper at https://doi.org/10.1530/EJE-21-0476.
Declaration of interest
The authors declare that there is no conflict of interest that could be perceived as prejudicing the impartiality of this study.
Funding
T W L is supported by the UKRI Expanding Excellence in England award. S E F has a Sir Henry Dale Fellowship jointly funded by the Wellcome Trust and the Royal Society (Grant Number: 105636/Z/14/Z). K A P has a Career Development fellowship funded by the Wellcome Trust (219606/Z/19/Z). This study represents independent research supported by the National Institute of Health Research Exeter Clinical Research Facility. The views expressed are those of the author(s) and not necessarily those of the Wellcome Trust, Department of Health, NHS or NIHR. For the purpose of open access, the author has applied a CC BY public copyright licence to any Author Accepted Manuscript version arising from this submission.
References
- 1↑
Yau D, Laver TW, Dastamani A, Senniappan S, Houghton JAL, Shaikh G, Cheetham T, Mushtaq T, Kapoor RR & Randell T et al.Using referral rates for genetic testing to determine the incidence of a rare disease: the minimal incidence of congenital hyperinsulinism in the UK is 1 in 28,389. PLoS ONE 2020 15 e0228417. (https://doi.org/10.1371/journal.pone.0228417)
- 2↑
Bruining GJ Recent advances in hyperinsulinism and the pathogenesis of diabetes mellitus. Current Opinion in Pediatrics 1990 2 758–765. (https://doi.org/10.1097/00008480-199008000-00024)
- 3↑
Mathew PM, Young JM, Abu-Osba YK, Mulhern BD, Hammoudi S, Hamdan JA, Sa’di AR. Persistent neonatal hyperinsulinism. Clinical Pediatrics 1988 27 148–151. (https://doi.org/10.1177/000992288802700307)
- 4↑
Gϋemes M, Rahman SA, Kapoor RR, Flanagan S, Houghton JAL, Misra S, Oliver N, Dattani MT, Shah P. Hyperinsulinemic hypoglycemia in children and adolescents: recent advances in understanding of pathophysiology and management. Reviews in Endocrine and Metabolic Disorders 2020 21 577–597. (https://doi.org/10.1007/s11154-020-09548-7)
- 5↑
Kapoor RR, Flanagan SE, Arya VB, Shield JP, Ellard S, Hussain K. Clinical and molecular characterisation of 300 patients with congenital hyperinsulinism. European Journal of Endocrinology 2013 168 557–564. (https://doi.org/10.1530/EJE-12-0673)
- 6↑
Snider KE, Becker S, Boyajian L, Shyng SL, MacMullen C, Hughes N, Ganapathy K, Bhatti T, Stanley CA, Ganguly A. Genotype and phenotype correlations in 417 children with congenital hyperinsulinism. Journal of Clinical Endocrinology and Metabolism 2013 98 E355–E363. (https://doi.org/10.1210/jc.2012-2169)
- 7↑
Banerjee I, Skae M, Flanagan SE, Rigby L, Patel L, Didi M, Blair J, Ehtisham S, Ellard S & Cosgrove KE et al.The contribution of rapid KATP channel gene mutation analysis to the clinical management of children with congenital hyperinsulinism. European Journal of Endocrinology 2011 164 733–740. (https://doi.org/10.1530/EJE-10-1136)
- 8↑
Adzick NS, De Leon DD, States LJ, Lord K, Bhatti TR, Becker SA, Stanley CA. Surgical treatment of congenital hyperinsulinism: results from 500 pancreatectomies in neonates and children. Journal of Pediatric Surgery 2019 54 27–32. (https://doi.org/10.1016/j.jpedsurg.2018.10.030)
- 9↑
Ellard S, Lango Allen H, De Franco E, Flanagan SE, Hysenaj G, Colclough K, Houghton JA, Shepherd M, Hattersley AT & Weedon MN et al.Improved genetic testing for monogenic diabetes using targeted next-generation sequencing. Diabetologia 2013 56 1958–1963. (https://doi.org/10.1007/s00125-013-2962-5)
- 10↑
Flanagan SE, Dung VC, Houghton JAL, De Franco E, Ngoc CTB, Damhuis A, Ashcroft FM, Harries LW, Ellard S. An ABCC8 nonsense mutation causing neonatal diabetes through altered transcript expression. Journal of Clinical Research in Pediatric Endocrinology 2017 9 260–264. (https://doi.org/10.4274/jcrpe.4624)
- 11↑
Richards S, Aziz N, Bale S, Bick D, Das S, Gastier-Foster J, Grody WW, Hegde M, Lyon E & Spector E et al.Standards and guidelines for the interpretation of sequence variants: a joint consensus recommendation of the American College of Medical Genetics and Genomics and the Association for Molecular Pathology. Genetics in Medicine 2015 17 405–424. (https://doi.org/10.1038/gim.2015.30)
- 12↑
Bittles AH A community genetics perspective on consanguineous marriage. Community Genetics 2008 11 324–330. (https://doi.org/10.1159/000133304)
- 13↑
Vidmar S, Carlin J, Hesketh K, Cole T. Standardizing anthropometric measures in children and adolescents with new functions for Egen. Stata Journal: Promoting Communications on Statistics and Stata 2004 4 50–55. (https://doi.org/10.1177/1536867X0100400104)
- 14↑
Pinney SE, MacMullen C, Becker S, Lin YW, Hanna C, Thornton P, Ganguly A, Shyng SL, Stanley CA. Clinical characteristics and biochemical mechanisms of congenital hyperinsulinism associated with dominant KATP channel mutations. Journal of Clinical Investigation 2008 118 2877–2886. (https://doi.org/10.1172/JCI35414)
- 15↑
Flanagan SE, Kapoor RR, Hussain K. Genetics of congenital hyperinsulinemic hypoglycemia. Seminars in Pediatric Surgery 2011 20 13–17. (https://doi.org/10.1053/j.sempedsurg.2010.10.004)
- 16↑
Falzone N, Harrington J. Clinical predictors of transient versus persistent neonatal hyperinsulinism. Hormone Research in Paediatrics 2020 93 297–303. (https://doi.org/10.1159/000511139)
- 17↑
Pedersen J Diabetes and pregnancy; blood sugar of newborn infants during fasting and glucose administration. Ugeskrift for Laeger 1952 114 685.
- 18↑
Ashcroft FM ATP-sensitive potassium channelopathies: focus on insulin secretion. Journal of Clinical Investigation 2005 115 2047–2058. (https://doi.org/10.1172/JCI25495)
- 19↑
Thornton PS, Stanley CA, De Leon DD, Harris D, Haymond MW, Hussain K, Levitsky LL, Murad MH, Rozance PJ & Simmons RA et al.Recommendations from the pediatric Endocrine Society for evaluation and management of persistent hypoglycemia in neonates, infants, and children. Journal of Pediatrics 2015 167 238–245. (https://doi.org/10.1016/j.jpeds.2015.03.057)
- 20↑
Houghton JA, Banerjee I, Shaikh G, Jabbar S, Laver TW, Cheesman E, Chinnoy A, Yau D, Salomon-Estebanez M & Dunne MJ et al.Unravelling the genetic causes of mosaic islet morphology in congenital hyperinsulinism. Journal of Pathology: Clinical Research 2020 6 12–16. (https://doi.org/10.1002/cjp2.144)
- 21↑
Hoermann H, Roeper M, Salimi Dafsari R, Koestner F, Reinauer C, Mayatepek E, Meissner T, Kummer S. Challenges in management of transient hyperinsulinism – a retrospective analysis of 36 severely affected children. Journal of Pediatric Endocrinology and Metabolism 2021 34 867–875. (https://doi.org/10.1515/jpem-2020-0639)
- 22↑
De Franco E, Flanagan SE, Houghton JA, Lango Allen H, Mackay DJ, Temple IK, Ellard S, Hattersley AT. The effect of early, comprehensive genomic testing on clinical care in neonatal diabetes: an international cohort study. Lancet 2015 386 957–963. (https://doi.org/10.1016/S0140-6736(1560098-8)
