An evidence-based approach to monitoring serum sodium in patients following non-pituitary cerebral neoplasm resection

Abstract Object: It is common for many neurosurgical services to routinely order daily laboratory studies on all inpatients. Here we investigate whether daily post-operative sodium (Na) is predictive of seizures experienced by patients admitted for non-pituitary tumor resection. Methods Three databases were constructed over a one-year period; a retrospective database inclusive of 916 inpatients admitted to the UCLA neurosurgery service, a prospective database of 644 inpatients admitted to the UCLA neurosurgery service, and a readmission database of 139 patients that were either readmitted or evaluated in the Emergency Department within 30 days of discharge. Sodium levels during admission were recorded in the retrospective database and the probability of a ≥ 5 mEq/L drop in Na over the first five days was calculated. The incidence of post-operative seizures was recorded in the prospective and readmission cohorts. For each seizure event, the diagnosis category and Na level were recorded. Finally, a subset of patients with borderline (130-134 mEq/L) Na levels at discharge were scheduled for outpatient Na checks to explore the safety of this practice. Results In the retrospective and prospective databases, the primary diagnosis of tumor was present in 227 (25 %) and 149 (23 %) patients respectively. Patients with a primary diagnosis of tumor showed an overall positive skewness of 0.187, with a probability of ≥ 5 mEq/L drop in Na level of 6.5%. Ten patients experienced early, and three patients experienced late postoperative seizures. Na level was within the normal range in all patients with a primary diagnosis of tumor at the time of seizure. Of the patients discharged with borderline serum Na, serum Na normalized during outpatient follow-up in all who presented to the appointment. Conclusions There is a mild trend towards hyponatremia following non-pituitary tumor surgery. However, this mild trend towards hyponatremia following brain tumor surgery is not associated with an increased incidence of seizure in either the early or late setting. Further, this hyponatremia is self-resolved and does not require aggressive treatment or prolonged hospital stays.