Trends in the Development of Postoperative Diabetes Insipidus and Prolonged Length of Stay After Pituitary Adenoma Resection: A Single Institution Experience

Introduction: Diabetes insipidus (DI) is a common postoperative occurrence after endoscopic resection of pituitary macroadenomas (PA). Occurring acutely in 15-30% of procedures, postoperative DI can adversely impact quality of life, increase length of hospital stay, and is associated with high morbidity. This study aimed to elucidate preoperative factors that are associated with postoperative DI and prolonged length of stay (PLOS) after resection of a pituitary tumor.

Methods: A single-center retrospective chart review was conducted to identify all patients who underwent PA resection from 2009-2020. DI was defined as postoperative polyuria and urine specific gravity < 1.005, and at least one of the following: excessive thirst, serum osmolality >300 mOsm/kg, and/or serum sodium levels >145 mmol/L. PLOS was defined as any stay longer than or equal to 96 hours. Univariate and multivariate logistic regression models were built to determine factors associated with postoperative DI and PLOS.

Results: A total of 253 patients were identified who underwent planned transsphenoidal macroadenoma resection, from which 25 patients (9.88%) developed DI. Tumor size was associated with increased odds of postoperative DI (OR=1.77; p=0.006), while older age was associated with decreased odds of postoperative DI (OR=0.96; p=0.047). Normal serum sodium at admission was associated with decreased odds of PLOS (OR=0.05; p=0.011). Tumor size (OR=2.12; p< 0.001) and DI (OR=4.62; p=0.006) were associated with PLOS.

Conclusion : This study suggests that normal serum sodium at admission may be protective for PLOS. Tumor size was associated with increased odds of postoperative DI while older age was associated with decreased odds, which is consistent with the current literature. This study further suggests the need for careful preoperative medical optimization prior to elective PA resection.