Purpose: Diabetes insipidus (DI) develops commonly after endoscopic transsphenoidal surgery (ETS). We retrospectively investigated the incidence, onset, duration and predictors of DI after ETS in patients with non-functioning pituitary adenoma (NFPA).
Methods: A total of 168 patients who underwent ETS to remove NFPAs were included. Various perioperative data on demographics, comorbidities, previous treatments, perioperative hormone deficiencies, tumor characteristics, surgery, anesthesia, intraoperative fluid balance, perioperative laboratory findings, postoperative complications, readmission and hospital length of stay were collected and analyzed. Patients were diagnosed with DI and treated with desmopressin when they showed urine output > 5 mL/kg/hr with a serum sodium concentration > 145 mmol/L or an increase ≥ 3 mmol/L in serum sodium concentration between two consecutive tests after surgery. DI was considered permanent when desmopressin was prescribed for > 6 months after surgery.
Results: Seventy-seven (45.8%) patients experienced postoperative DI and 10 (6.0%) patients suffered from permanent DI. The median onset of DI and the median duration of transient DI were postoperative day 1 and 5 days, respectively. In multivariable logistic regression analysis, cephalocaudal tumor diameter (odds ratio [95% confidence interval] 2.59 [1.05-6.36], P = 0.038) was related to postoperative DI. In receiver operating characteristic analysis, its area under the curve was 0.68 (95% confidence interval 0.59-0.76, P < 0.001). Its optimal cutoff value that maximized the sum of sensitivity and specificity for postoperative DI was 2.7 cm.
Conclusions: Postoperative DI was observed in 45.8% of patients undergoing ETS to remove NFPAs. A large cephalocaudal tumor diameter was predictive of postoperative DI in such patients.
Keywords: Cephalocaudal tumor diameter; Diabetes insipidus; Endoscopic transsphenoidal surgery; Non-functioning pituitary adenoma.