Endocrine Abstracts

Background: The water deprivation test serves as a key diagnostic test to differentiate between Arginine Vasopressin Deficiency (AVP-D), Arginine Vasopressin Resistance (AVP-R), and Psychogenic Polydipsia (PP) in patients presenting with symptoms of polydipsia and polyuria, when common causes such as diabetes mellitus have been excluded. This study evaluated the established diagnostic values for serum osmolality, serum sodium, and urine osmolality.

Methods: This retrospective analysis assessed 135 patients who underwent a water deprivation test between August 2014 and August 2023. All patient diagnoses were reviewed by an expert panel. Variability and receiver operating characteristic (ROC) curves were determined for serum osmolality, serum sodium and urine osmolality, to determine the effectiveness of the current interpretation reference ranges.

Results: A total of 120 patients were included in the study analysis. Use of serum sodium demonstrated reduced variability compared with serum osmolality (0.722% vs 1.16% respectively, 37.5% reduction; P <0.001). Use of the standard serum osmolality cut-off value of ≥300 mOsm/kg in diagnosing AVP-D, AVP-R, and PP achieved a sensitivity of 76.19% and specificity of 76.92%. Use of a serum sodium cut-off value of ≥148 mmol/l demonstrated 100% specificity in excluding PP. This cut-off was used in tandem with urine osmolality cut-off values of >630 mOsm/kg (for PP) and <383 mOsm/kg (for AVP-D and AVP-R). Review of post-desmopressin urine osmolality changes and clinical monitoring was performed in equivocal diagnostic cases (n=6), achieving both 100% sensitivity and 100% specificity within the study sample.

Conclusions: This study demonstrates that use of a serum sodium cut-off value of ≥148 mmol/l in combination with urine osmolality yields the best diagnostic accuracy to differentiate between arginine vasopressin deficiency, arginine vasopressin resistance, and psychogenic polydipsia. Serum sodium is superior to serum osmolality, demonstrating lower total variability (including biological and analytical variability).