OR20-06 Diagnosing Central Diabetes Insipidus Using Copeptin Upon Hypertonic Saline Versus Arginine Stimulation

Disclosure: J. Refardt: None. C. Atila: None. I.O. Chifu: None. E. Ferrante: None. Z. Erlic: None. J.B. Drummond: None. B. Mantovani: None. R. Drexhage: None. C.O. Sailer: None. A. Widmer: None. S. Felder: None. A.S. Powlson: None. N. Hutter: None. D. Vogt: None. M. Gurnell: None. B. Rocha: None. J. Hofland: None. F. Beuschlein: None. M. Fassnacht: None. B.F. Winzeler: None. M. Christ-Crain: None. Background: The main challenge in the diagnosis of central diabetes insipidus (cDI) is its distinction against primary polydipsia (PP). Hypertonic saline stimulated copeptin has a high diagnostic accuracy of 97%, but comprises discomfort for patients and requires close sodium monitoring. Arginine stimulated copeptin showed similar diagnostic accuracy of 93% with better tolerability, but a head-to-head comparison is lacking. We hypothesized that arginine stimulated copeptin is non-inferior (non-inferiority margin 10%) to hypertonic saline stimulated copeptin for the diagnosis of cDI. Methods: In this prospective randomized multicentre study conducted between 2018-2022 in seven tertiary medical centres, consecutive patients with cDI and PP underwent diagnostic evaluation with hypertonic saline and arginine stimulation. Serum copeptin levels were measured at sodium-level of >149 mmol/L after hypertonic saline and 60 minutes after arginine infusion, respectively. The final diagnosis was made after treatment response assessment at three-month follow-up blinded to copeptin levels. The main outcome measure was the overall diagnostic accuracy using the pre-defined copeptin cut-off of 4.9 pmol/L for hypertonic saline and 3.8 pmol/L for arginine stimulation. Results: 158 patients underwent both tests, of which 69 (44%) were diagnosed with cDI and 89 (56%) with PP. Forty-one (59%) patients had a complete and 28 (41%) patients had partial cDI. The diagnostic accuracy [95% CI] to differentiate patients with cDI from patients with PP was 95.6% [91.1,97.8] for hypertonic saline stimulated copeptin compared to 74.4% [67.0, 80.6] for arginine stimulated copeptin. Accordingly, arginine stimulation was inferior to hypertonic saline stimulation (estimated difference [95% CI] -21.2% [-28.7, -14.3]). Side effects were less frequent and less severe under arginine stimulation, resulting in a clear 72% of patients preferring the arginine stimulation over hypertonic saline stimulation. In addition, arginine stimulated copeptin of 3.0 pmol/L diagnosed cDI with a specificity of 90.9% (sensi