Endocrine Abstracts

Introduction: Central diabetes insipidus (DI) is observed in up to 34% of patients undergoing pituitary surgery. Early postoperative diagnosis of DI is critical because its detection allows early treatment, reducing morbidity and hospital stay. AVP levels measurement is not commonly used to diagnose DI. As an alternative, copeptin determination has been proposed, but its usefulness has only been validated in one series.

Objective: To define the threshold and optimal measurement time of serum copeptin for early prediction of DI after pituitary surgery in an independent series.

Design: Observational, prospective study in a unique center. Subjects: Patients operated by endonasal surgery from 2017 to 2018, excluding those with pre-surgery DI. Variables: serum copeptin, sodium, potassium, glucose, creatinine, urea, plasma and urinary osmolality, and glomerular filtration before and at 4-6h, 24h and 48h after surgery, preoperative basal serum cortisol, TSH and free thyroxine, sex and tumor features (size, invasion, chiasmatic compression and functionality). Statistical analysis: SPSSv.21. Kolmogorov-Smirnov test to evaluate normality of the distribution. Descriptive of quantitative variables: mean and standard deviation in parametric distributions, median and P25–75 in nonparametric ones. Calculation of the best cut-off point by Receiver Operating Characteristic (ROC) curves in each time. Calculation of sensitivity (Se), specificity (Sp), negative (NPV) and positive predictive value (PPV).

Results: 27 patients, 32% women. Age 49±17years. Maximum diameter 20.3±9mm. Clinical diagnosis: 55% non-functioning adenomas, 15% Prolactinomas, 19% acromegaly, 7% Cushing’s disease, 4% pituitary apoplexy. Eight (30%) patients developed post-surgery DI. Copeptin concentrations at 4–6h after extubation was significantly lower (7.7 vs 16.5 pmol/L, P=0.037) in patients who developed postoperative DI than those who did not. No other significant differences between groups were found in any of the studied variables, except for the development of cerebrospinal fluid leak (P=0.006). The greater diagnostic validity of copeptin for the detection of DI in the postoperative period was obtained 24h after surgery, with a threshold point of 3 pmol/L (Se 62.5%, Sp 95%, PPV 83.3%, NPV 86.3%). All patients who presented DI normalized their neurohypophyseal function during the follow-up, except one with persistent DI two months after surgery.

Conclusions: The prevalence of post-surgery DI in the studied series (30%) agrees with that described in the literature. Our preliminary results suggest that copeptin measured 24h postoperatively may be a useful tool for early diagnosis of DI, although the optimal threshold differs from that previously reported.