Transient Delayed Hyponatremia after Transsphenoidal Surgery: Attempting to Enlighten the Epidemiology and Management of a Still-Obscure Complication

Although the occurrence of transient delayed hyponatremia after transsphenoidal surgery is well known, the need for more and better knowledge of this phenomenon remains. The vast majority of patients have asymptomatic hyponatremia, which explains the great disparities reported in the prevalence of the disorder, from 2% to 35% on the basis of different inclusion criteria.1-12 The more the patients perform daily venous samples for serum sodium assay during the postoperative period, the more frequent the occurrence of asymptomatic hyponatremia becomes, because a large number of cases will be discovered.4,11 It is well known that delayed hyponatremia after transsphenoidal surgery, defined as a serum sodium level of less than 135 mmol/L on or after postoperative day 3, usually occurs around postoperative day 7 (between 3 and 11 days after surgery) as an isolated entity or, much more rarely, as an interphase in the triphasic postoperative diabetes insipidus. As a result, hyponatremia often occurs when the patient already has been discharged, making the postoperative care troublesome. Furthermore, in the case of readmission because of delayed hyponatremia after transsphenoidal surgery, the cost of overall care of this patient population increases relevantly. For a long time after Cusick et al.1 reported the inappropriate secretion of antidiuretic hormone (SIADH) as the possible cause of delayed symptomatic hyponatremia after transsphenoidal surgery in 1984, the debate regarding the pathophysiology of the disorder had a checkered history. Given the whole body of evidence collected during the last 3 decades, the syndrome of SIADH would seem the more plausible cause of hyponatremia in this setting1-12; however, some studies in which the authors suggest cerebral salt wasting syndrome (CSWS) as another possible cause have also been published.13,14 The most likely explanation for these puzzling findings is the difficulty from the patient bedside in evaluating the extracellular volume status, which highlights the real difference between these 2 conditions. On the other hand, when faced with a hyponatremic patient, the physician has to bear in mind that the clinical evaluation of the extracellular volume status is mandatory for a correct diagnosis and an effective treatment. Both SIADH and CSWS share similar laboratory tests (plasma osmolality, urinary osmolality, and urinary sodium concentrations) but differ in the extracellular volume status, which is expanded in the former (despite the patient is clinically euvolemic) and depleted in the latter. Some retrospective studies have identified factors associated with increased risk for delayed hyponatremia after transsphenoidal surgery, including sex (female), tumor size, hypopituitarism, transient diabetes insipidus, and Cushing disease.5,6,8,10,13,15,16 Concerns have been raised, however, about the real impact of these findings, and no consensus exists. The group of patients affected with Cushing disease, for example, could represent a distinct cohort, in which the postoperative hyponatremia may have a different pathophysiology, namely the secondary adrenal insufficiency occurring after remission of Cushing disease. Keeping in mind, however, the relevant role of delayed hyponatremia in the postoperative management and the additional cost of this complication, there is no doubt that the identification of predictive factors may be relevant to identify patient subpopulations at greater risk and keep them under a strict surveillance after discharge. In January 2016, Bohl et al.17 identified delayed symptomatic hyponatremia as the primary cause of unplanned readmission after transsphenoidal surgery for pituitary tumors. In their retrospective cohort of 303 patients, 27 (8.9%) were readmitted within 30 days. Of these, 15 (55.6%) had delayed hyponatremia. As a result, of all the patients discharged after pituitary surgery, 5% (15/303) were readmitted because of symptomatic hyponatremia. Other less-frequent causes of unplanned readmission were diabetes insipidus (4 cases), secondary adrenal insufficiency (2 cases), cerebrospinal fluid leak, epistaxis, cardiac arrhythmia, pneumonia, urinary tract infection, and hypoglycemia (1 case each). Another recent retrospective cohort of 466 consecutive patients who underwent endoscopic transsphenidal surgery showed a slightly greater readmission rate for epistaxis than hyponatremia (2.1% and 1.5%, respectively).18 Some years ago, we evaluated retrospectively the prevalence, risk factors, course, and management of symptomatic delayed postoperative hyponatremia in our cohort of 1098 consecutive patients who underwent pituitary surgery by endoscopic endonasal approach from 1998 to 2010, and the results of the study were presented in 2011 at the Italian Society of Endocrinology's 35th Meeting in Pescara, Italy. Ninety-five patients were excluded on the basis of the following exclusion criteria: postoperative triphasic diabetes insipidus, diuretic treatment, drug-induced SIADH, and scant laboratory data. Therefore, a total of 1003 patients (median age, 50 years, range, 3–89 years; 52% females; 87% pituitary adenomas; 5% craniopharyngiomas; 5% Rathke cleft cysts; 2% metastases; 1% meningiomas) were eligible for the study. Forty-one patients (4%) developed symptomatic delayed hyponatremia, which occurred after discharge in all patients but 3. Nadir of medians of serum sodium (124 mmol/L) occurred on postoperative day 8. Both the clinical assessment of extracellular volume status and laboratory tests were consistent with the diagnosis of hypotonic hyponatremia due to SIADH in all cases (no case of CSWS was recorded in our cohort). Age, tumor size, and tumor type (including corticotroph cell adenoma associated with Cushing disease) did not correlate with the development of symptomatic hyponatremia. Interestingly, female patients were more likely to develop symptomatic hyponatremia than males (female to male ratio: 3.1; P < 0.05). Fluid intake restriction was successfully used in 16 patients (39.1%); the remaining 25 (60.9.%) were treated with 3% NaCl hypertonic saline infusion (0.5–2 mL/kg/hour). No case of osmotic demyelination syndromes occurred. The recommended limits of correction in serum sodium concentration in the first 24 hours were 6–8 mmol/L. The apparently high percentage of patients treated with hypertonic saline was attributable to the inclusion criteria. Indeed, we evaluated only patients with symptomatic hyponatremia who were readmitted after transsphenoidal surgery. We continue to be successful in using the same protocol for the management of transient delayed hyponatremia after transsphenoidal surgery. As a matter of fact, the choice of treating patients with severe symptoms of hyponatremia (vomiting, abnormal somnolence, cardiorespiratory distress) with 3% NaCl hypertonic saline independently by the biochemical degree of hyponatremia is now accepted widely and the recommended limits of correction of serum sodium as well.19 To date, the systematic review recently published in WORLD NEUROSURGERY by Cote et al.20 represents a more accurate and updated review of the literature on the delayed hyponatremia after transsphenoidal surgery. The authors have chosen 10 studies for a total of 2947 patients. Despite some limitations of the studies, mainly from the heterogeneity of the including criteria, which makes the pool analysis not so meaningful as expected, the authors have found a number of predictive factors for developing delayed postoperative hyponatremia: young age, female sex, greater tumor size, greater rate of decline of serum sodium, and Cushing disease. Again, the heterogeneity of the inclusion criteria remains the main concern regarding the possibility of considering their findings to be conclusive. However, the systematic review of Cote et al. represents a key step toward improving the postoperative management of postoperative hyponatremia in patients undergoing transsphenoidal surgery. Furthermore, the authors encourage the research in this field, with the aim to recognize the causes and the mechanisms of this fascinating electrolyte disorder. One way to do so is to design a large prospective cohort study that allows investigators to avoid confounding factors, such as uncertainties in the postoperative management of patients with Cushing disease, lack of assessment of extracellular volume status, and scant laboratory data (urinary osmolality and urinary sodium, for example, are mandatory for a correct diagnosis of hyponatremia, but these data sometimes have not been recorded in previous studies), in order to ascertain the epidemiology, the proper management, and the causes of this still-obscure and challenging postoperative complication