ISSN 1470-3947 (print) | ISSN 1479-6848 (online)

Endocrine Abstracts (2019) 63 OC13.1 | DOI: 10.1530/endoabs.63.OC13.1

Arginine-stimulated copeptin measurements - a new test for diabetes insipidus

Bettina Winzeler1, Nicole Nigro-Cesana1, Julie Refardt1, Deborah R Vogt2, Cornelia Imber1, Benedict Morin1, Milica Popovic1, Michelle Steinmetz1, Clara Sailer1, Gabor Szinnai3, Irina Chifu4, Martin Fassnacht5 & Mirjam Christ-Crain1


1University Hospital of Basel, Basel, Switzerland; 2University Hospital of Basel, Department of Clinical Research, Basel, Switzerland; 3University Children’s Hospital Basel, Basel, Switzerland; 4Department of Internal Medicine, Division of Endocrinology and Diabetes, University Hospital, Würzburg, Germany; 5Department of Internal Medicine, Division of Endocrinology and Diabetes, University Hospital, Würzburg, Switzerland.


Background: The differential diagnosis of diabetes insipidus is challenging. The most reliable diagnostic approach is hypertonic saline-stimulated copeptin measurements. However, as this test is based on the induction of hypernatremia, it is associated with adverse effects and needs close sodium monitoring. We hypothesized that arginine-stimulated copeptin measurements provide an alternative, simple and safe diagnostic test for diabetes insipidus.

Methods: We recruited a development cohort (cohort 1, N=52) and a validation cohort (cohort 2, N=44) including patients with central diabetes insipidus (total N=38) and with primary polydipsia (total N=58). All patients and 92 healthy controls underwent arginine stimulation with 0.5 g L-Arginin-Hydrochlorid/kg body weight, infused over 30 minutes. Copeptin levels were measured at baseline and 30, 45, 60, 90, 120 minutes after arginine infusion. The primary objective in the first cohort was the diagnostic accuracy of copeptin levels at each measurement after stimulation whereas the second cohort should validate these results.

Results: Arginine infusion stimulated median [IQR] copeptin levels in healthy controls and in patients with primary polydipsia (3.6 pM/L [2.4, 5.7] to 7.9 pM/L [5.1, 11.8]), but not in patients with diabetes insipidus (2.1 pM/L [1.9, 2.7] to 2.5 pM/L [1.9, 3.1]). In the first cohort, a cutoff of 3.5 pM/L at 60 minutes provided the highest diagnostic accuracy to discriminate between diabetes insipidus and primary polydipsia: 0.94 [95% CI] [0.84, 0.98], sensitivity 91%, specificity 97%. The diagnostic accuracy of this cutoff in the second cohort was 0.86 [0.73, 0.94]. By pooling the data of both cohorts an optimal diagnostic accuracy was reached for a cutoff of 3.8 pM/L at 60 minutes: 0.93 [0.86, 0.97], (sensitivity 93%, specificity 92%). The test was safe and well tolerated: the test burden (median value on a visual analogue scale from 0 [no discomfort] to 10 [maximal discomfort]) was rated 3.5 and 3 in patients with diabetes insipidus and primary polydipsia and 1 in healthy controls.

Conclusion: Arginine is a potent stimulus of the neurohypophysis. Arginine-stimulated copeptin measurements are an innovative tool to discriminate between central diabetes insipidus and primary polydipsia with high diagnostic accuracy.