Gabor Szinnai

PD Gabor
Szinnai, MD PhD

Deputy Head, Pediatric Endocrinology / Diabetology

University Children's Hospital Basel

Phone +41 61 704 29 22



Research Focus

Endocrinology, Metabolism & Developmental Disorders

Area of Research

Thyroid diseases with special focus on genetic diagnosis and optimized therapy in the neonate, the child and the adolescent. Water and electrolyte disorders with special focus on diagnosis in pediatric patients with polyuria / polydipsia syndrome and therapy and outcome in children with type 1 diabetes and ketoacidosis.

Approved Research Projects

2011 - 2015 PI: Swiss National Science Foundation (SNF), Title: “Epigenetic regulation of thyroid development”.


2012 - 2014 PI: S. Rückert Stiftung zur Krebsbekämpfung, Title: “PI3K-AKT in thyroid development“.

2015 - 2016 PI: G und J Bangerter-Rhyner-Stiftung, Title: “Epigenetic regulation of thyroid development”.

2015 - 2016 PI: Forschungsfonds Universität Basel, Title: “Epigenetic regulation of thyroid development”.

2018 - 2019 Co-I: Joint Grant University of Basel/ETH Zürich: Title: “Accounting for physical activity to personalize insulin treatment in children with type-1 diabetes mellitus”.

2017 - 2020 PI: Thermo-Fisher unrestricted grant: Title: “Use of copeptin for differential diagnosis of diabetes insipidus in pediatric patients (COPED) - a prospective multicentre study”.

2017 - 2020 PI: European Society for Paediatric Endocrinology (ESPE), Title: “Next generation sequencing in congenital hypothyroidism” (International Multicenter Study)


    National Collaborations


    • Prof. Myriam CHRIST-CRAIN, Deputy Head, Endokrinologie, Diabetologie und Metabolismus, Universitätsspital Basel
    • Prof. Daniel KONRAD, Head, Päd. Endokrinologie/Diabetologie, Universitäts-Kinderspital Zürich
    • Prof. Christa FLÜCK, Head, Päd. Endokrinologie/Diabetologie und Metabolismus, Universitäts-Kinderspital Bern
    • Prof. Valérie SCHWITZGEBEL, Head, Unité d'endocrinologie et diabétologie pédiatriques, Hôpital universitaire Genève
    • Prof. Dagmar L'ALLAMAND, Head, Päd. Endokrinologie/Diabetologie, Ostschweizer Kinderspital
    • Prof. Jörg STELLING, Computational Systems Biology Group, D-BSSE ETH Zürich
    • PD Dr. Michael KALTENBACH, Computational Systems Biology Group, D-BSSE ETH Zürich
    • Prof. Philipp LATZIN, Head, Päd. Pneumologie, Universitäts-Kinderspital Bern
    • PD Dr. Matthias BETZ, Endokrinologie, Diabetologie und Metabolismus, Universitätsspital Basel

    International Collaborations

    • Prof. Michel POLAK, Head, Service d'endocrinologie, gynécologie et diabétologie pédiatriques, Hôpital Universitaire Necker Enfants-Malades, Paris
    • Prof. Heiko KRUDE, Head, Institut für Experimentelle Pädiatrische Endokinologie, Charité, Universitätsmedizin, Berlin
    • Prof. Johnny DELADOËY, Divison of Paediatric Endocrinology and Research Center CHU, Ste-Justine, University of Montreal, Montreal
    • Prof. Paul VAN TROTSENBURG, Head, Department of Paediatric Endocrinology, Emma Children's Hospital, Academic Medical Center AMC, Amsterdam
    • Prof. Tim JONES, Head, Department of Endocrinology and Diabetes, Princess Margaret Hospital for Children, University of Western Australia, Perth
    • Prof. Johannes SCHROPP, Fachbereich Mathemathik und Statistik, Universität Konstanz

    Ongoing Research Projects

    • NGS Thyroid Gene Panel Study (Multicenter Study Basel, Berlin, Paris, Amsterdam, Montreal) Title: Improving the diagnostic work-up for congenital hypothyroidism (CH) by a multigene NGS panel: A systematic genetic, clinical and economic analysis in the context of an international CH research consortium. Background: Congenital hypothyroidism with eutopic gland (CH-EU) represents 30-47% of all CH cases in 6 different cohorts. In contrast to athyreosis or ectopy, the origin of CH-EU is heterogeneous. It includes mild forms of thyroid dysgenesis (TD; mutations in PAX8, NKX2-1, FOXE1), TSH-resistance (mutations in TSHR, GNAS), dyshormonogenesis (THDG; mutations in TG, TPO, SCL5A5, SCL26A4, DUOX2, DUOXA2, DEHAL). Next generation sequencing (NGS) gene panels allow rapid parallel mutational screening in a defined set of disease causing genes at low costs. Based on current genetic knowledge, NGS mutational screening would provide highest yield in CH-EU compared to athyreosis or ectopy. Hypotheses: The use of the NGS CH-panel in CH-EU patients would 1) provide the true frequency of all known genetic defects in CH-EU and a genetic diagnosis in 50% of patients, 2) provide information to optimize patient management and family counseling, 3) might be used in a cost-effective way for the diagnostic work-up of CH-EU. Study design: Multicenter observational study: A) Retrospective NGS analysis of known patients with CH-EU. B) Prospective study with NGS analysis of newborns with CH-EU. Aims: 1) define the prevalence of genetic defects in CH-EU, 2) compare management and counseling +/- NGS, 3) compare costs of different diagnostic algorithms for CH +/- NGS. Clinical significance: Expected results cover three areas: 1) systematic description of molecular genetics of CH-EU, 2) impact of genetic results on management and counseling, 3) evaluation, in which diagnostic algorithm of CH routine use of NGS might be cost-effective.
    • CoDIAB Study (Basel and Perth, Australia) Title: Copeptin in children with type 1 diabetes and ketoacidosis. Background: In new-onset type 1 diabetes children present variably disordered water homeostasis characterized by hyperosmolal hypovolemia. Diabetic ketoacidosis is frequent in children. Cerebral edema complicating diabetic ketoacidosis occurs in 1% of children and remains the major cause of morbidity and mortality in children with type 1 diabetes. Elevated levels of copeptin, the C-terminal preprovasopressin, were found in disordered water homeostasis (isoosmlaor dehydration, hyperosmolar normovolemia; Szinnai and Christ-Crain, J Clin Endocrinol Metab 2007), but no data are available of copeptin changes during ketoacidosis neither in adults nor in children. Further, copeptin has been shown to be a stress marker being increased in patients with cerebral haemorrhage and brain edema. Hypotheses: We hypothesize, that increased copeptin – a combined water balance regulating hormone as well as stress marker - could 1) be a sensitive parameter to guide adequate water replacement, and 2) be a preclinical marker of cerebral edema. During rehydration and insulin therapy, it is well known that plasma osmolality decreases. It is expected that copeptin would decrease along the physiological exponential correlation with plasma osmolality. In case of developing cerebral edema, a stress response of the body is occurring. It is expected that as soon as increasing intracranial pressure occurs during rehydration in the first 12 to maximal 24 hours, a sudden, stress-induced increase of copeptin would override the osmotic regulation of copeptin and result in copeptin value deviation above the normal range of the copeptin-osmolality normogramm (Szinnai and Christ-Crain, J Clin Endocrinol Metab 2007). Aims: 1) Characterize the physiological exponential correlation of copeptin and osmolality in patients with highly hyperosmolar states in ketoacidosis (>300 mosmol/kg). 2) Document copeptin in patients with suspected subclinical cerebral edema. 3) Modelling the interplay of the different known water regulating hormones during correcting of hyperosmolar dehydration. Study design: Observational study of 24 pediatric patients with diabetic ketoacidosis at the University Children’s Hospital Basel and in collaboration with Prof. Tim Jones, Head, Department of Endocrinology and Diabetes, Princess Margaret Hospital for Children, University of Western Australia, Perth. Copeptin, osmolality, aldosterone, and Pro-ANP are measured sequentially, at 0,1,2,3,4,6,8,10,12,18,24,36,48 and 72 hours after start of rehydration and insulin therapy together with clinical (weight, heart rate, blood pressure, GCS) and further laboratory parameters (glucose, blood gases. Clinical significance: Clinical assessment of dehydration is imprecise and generally shows only fair to moderate agreement among examiners. It is based on a combination of physical signs such as prolonged capillary refill time, abnormal skin turgor, dry mucus membranes and hyperpnea in the young child. Beside the blood gas analysis and the analysis of electrolytes we have no sensitive markers for monitoring the changes of water homeostasis in ketoacidosis. Further, clinical neurological evaluation with GCS is especially difficult in young children. Thus, copeptin could optimize monitoring and treatment in two ways: First, as a water regulating hormone, copeptin measurements could guide and optimize rehydration therapy during ketoacidosis, especially in the young child with high risk for complications. Second, as a marker for endocrine stress response, copeptin could be useful as a sensitive laboratory parameter for subclinical cerebral edema.
    • CoPED study. Title: Use of copeptin for differential diagnosis of Diabetes insipidus in pediatric patients - a prospective multicenter study. Background: Polyuria and polydipsia syndrome (PPS) may result from nephrogenic diabetes insipidus (DI), central DI, and primary polydipsia (PP), whose careful differentiation is mandatory as treatment strategies vary substantially. The diagnostic gold standard for PPS in children is the water deprivation test. Its diagnostic accuracy in adults is low. The measurement of copeptin, the C-terminal preprovasopressin, has been recently shown to increase diagnostic accuracy of the water deprivation test in adults. No data are available so far in infants, children and adolescents. Water deprivation testing is invasive and highly stressfull for children, its duration should be limited until definitive diagnosis is reached by the most sensitive parameter. Hypotheses: 1) the adult cut-off levels for copeptin are also applicable in the pediatric age group, 2) serial copeptin measurements during water deprivation test provide a more accurate and earlier diagnosis than use of urine osmolality. Aims: 1) To demonstrate that adult copeptin cut-off values at basal state or baseline of the water deprivation test are able to diagnose nephrogenic DI versus central DI and primary polydipsia in children. 2) To demonstrate that adult copeptin cut-off values at the end of the water deprivation test are able to differentiate between central DI and primary polydipsia. 3) To explore, whether serial copeptin measurements during water deprivation in children allow earlier final diagnosis (central DI versus primary polyuria) than the standard serial urine osmolality measurements. Study design: Non-interventional diagnostic accuracy study. The water deprivation test is performed according international standards. Copeptin is additionally measured every hour in parallel with routine measurement of serum and urine osmolality. Copeptin values are not available to the responsible clinician during the test and are not used for definitive diagnosis. Clinical significance: Water deprivation testing in children is highly invasive and stressfull for the patient and the family. It exposes the child to relevant dehydration. Due to its risk profile the test can only be performed in experienced tertiary care centers. Integration of serial measurements of copeptin during the water dehydration test could A) increase diagnostic accuracy and B) result in earlier diagnosis and shorter test procedure.
    • ThyMOD study: Title: The ThyMOD Study - Personalized Dosing in Children with Hyper- or Hypothyroidism Computed by MathematicalModeling. Background: Thyroid hormones are essential for normal brain development in the foetus and the infant, and for normal cognitive functions, and normal growth in children and adolescents. Treatment of thyroid diseases (congenital and acquired hyper- or hypothyroidism) is difficult in neonates, infants and children. First, a carefully selected initial dose based on clinical experience is necessary. Second, after reaching a physiological balance of thyroid hormones, a continuous adjustment of the individual dose depending on age and severity of thyroid disease is required to maintain euthyroidism during childhood and adolescence. To mitigate the risk of negative neurological and developmental outcome, it is essential to establish an optimal, personalized dosing that is continuously fine-tuned in neonates, infants and children with hyper- or hypothyroidism. Hypotheses: Combination of dynamical mathematical modelling based on population pharmacokinetic / pharmacodynamic (PKPD) principles with individual covariate effects together with algorithms from optimal control theory and longitudinal disease measurements is able to predict an optimal personalized dosing strategy that maintains thyroid hormones in the normal reference range. Aims: Aims of this research project are (i) to develop mathematical population PKPD models that are able to describe individual thyroid disease progression during treatment with thyroid hormone replacement therapy or anti-thyroid drugs based on routine clinical data such as thyroid-stimulation hormone, free/total triiodothyronine (T3) and thyroxine (T4), and key covariates such as age and weight, (ii) to combine population PKPD models with optimal control theory to compute personalized optimal doses, (iii) to develop optimal control feedback algorithms that account for patient specific factors such as disease severity and supports clinicians to maintain thyroid hormones in the normal reference range. Study design: Longitudinal routine clinical data of individual patients affected by one of the four paediatric thyroid diseases (congenital hypo- and hyperthyroidism, acquired hypothyroidism (Hashimoto thyroiditis) and hyperthyroidism (Graves’ disease) are collected retrospectively in 5 Swiss paediatric hospitals (starting from the year 1995) with a follow-up period of at least 2 years. Each patient has up to 20 visits and data at each visit contains clinical and laboratory parameters. Expected value of project: Mathematical models and optimal control feedback algorithms that support clinicians to optimally fine-tune dosing in paediatric patients. Specific deliverables include (i) mathematical models with optimal control will be able to compute the individual optimal dosing strategy. For congenital hyper- and hypothyroidism optimal dosing can be computed to bring the patient as soon as possible into the normal range by reducing over- and undershooting of the normal range. For acquired hyper- and hypothyroidism the dosing strategy can be computed to follow an individual hormone reference progression in order to avoid too abrupt changes of thyroid hormone levels, (ii) models and algorithms to compute optimal personalized dosing in neonates, infants and children are a medical necessity to optimize long-term neurological outcome and consequently enhance performance in school and professional education, (iii) optimal control algorithms will be made freely available for educational purposes, and (iv) established retro- and prospective databases will be made available under the FAIR Guiding Principles.

    Research Team

    Tatjana Welzel, MD

    Research Fellow

    Phone +41 61 704 22 52

    Melanie Hess, MD

    Medical specialist

    Phone -
    Email -

    Katrin Gerber-Windisch

    Study Nurse

    Phone +41 61 704 29 42

    Marie-Anne Burckhardt, MD

    Senior Physician

    Phone -
    Email -

    Noemi Züger

    Physician / Doctoral Student

    Phone -
    Email -

    Sara Bachmann, MD

    Medical specialist

    Phone -
    Email -