Hyponatremia After Placement of an External Lumbar Drain Following Transsphenoidal Surgery in a Patient with Preoperative Diabetes Insipidus

Introduction:
Hypernatremia due to diabetes insipidus (DI) is perhaps the most anticipated electrolyte abnormality after transsphenoidal surgery (TSS), but hyponatremia is more common and occurs in over 15% of patients undergoing TSS.  Postoperative placement of an external lumbar drain (ELD) is a well-known risk factor for hyponatremia following TSS. We present an unusual case of refractory hyponatremia occurring after ELD placement in a patient with preoperative DI.

Case Description:
A 68-year-old male underwent TSS for resection of a recurrent sella meningioma. History was notable for multiple recurrences and resections of the tumor, and the patient was treated for longstanding central DI and hypopituitarism ( >10 y). Surgery was complicated by bleeding, requiring placement of an ELD. Preoperative serum sodium levels were in the reference range (136-145 mM), but within 48 h of ELD placement, serum sodium level fell to 127 mM. Spot urine sodium level and osmolality were <10 mM and 162 mOsm/kg, respectively. The patient received stress dose glucocorticoids after surgery, and free T4 level was in the reference range (0.5-1.3 ng/dL). Other causes of hyponatremia and pseudohyponatremia, including hyperglycemia, renal tubular acidosis, paraproteinemia, and hyperlipidemia were excluded.  While the lumbar drain was in place, serum sodium levels remained 125-129 mM despite decreasing the amount and frequency of DDAVP doses for treatment of preoperative central DI. After the ELD was removed, hyponatremia resolved within 48 h, and the patient was discharged home with a serum sodium level of 140 mM on oral DDAVP (0.1 mg BID) prescribed prior to surgery.

Discussion:
In a series of over 300 adults undergoing TSS, placement of an ELD accounted for almost half of hyponatremia cases. Cerebral salt wasting syndrome (CSWS) and syndrome of inappropriate antidiuretic hormone (SIADH) are identified as the most common causes of ELD-induced hyponatremia, though undetectable urine sodium level in this case excludes CSWS. Spot urine osmolality was surprisingly high (>100 mOsm/kg) for the patient’s longstanding central DI, appearing to indicate SIADH as the etiology of hyponatremia. The patient likely had partial central DI, and persistent hyponatremia after placement of the ELD despite significant reductions of DDAVP doses appears to be best explained by release of endogenous ADH by surviving magnocellular neurons of the supraoptic nucleus in the hypothalamus superimposed on treatment with DDAVP. This case demonstrates that ELD-induced hyponatremia is possible in patients treated for central DI prior to TSS and likely indicates only a partial deficit of endogenous ADH secretion.