Why does glucocorticoid deficiency cause hyponatremia

Glucocorticoid inhibition of neurohypophysial vasopressin secretion. Erkut, Z. Glucocorticoids suppress corticotropin-releasing hormone and vasopressin expression in human hypothalamic neurons. Diederich, S. Severe hyponatremia due to hypopituitarism with adrenal insufficiency: report on 28 cases.

Download references. You can also search for this author in PubMed Google Scholar. Correspondence to Stefan Reuter. Reprints and Permissions. Reuter, S. Hyponatremia and seizures caused by triamcinolone-induced adrenal insufficiency. Nat Rev Nephrol 6, — Download citation. Issue Date : February Anyone you share the following link with will be able to read this content:. Sorry, a shareable link is not currently available for this article. Provided by the Springer Nature SharedIt content-sharing initiative.

Nature Reviews Nephrology Advanced search. Skip to main content Thank you for visiting nature. Subjects Acid, base, fluid, electrolyte disorders Adrenal gland diseases Hormonal therapies. Abstract Background. Access through your institution.

Buy or subscribe. Rent or Buy article Get time limited or full article access on ReadCube. Figure 1: Hormonal interactions that regulate the hypothalamic—pituitary—adrenal axis. References 1 Jansen, T. Article Google Scholar 14 Rai, A.

Article Google Scholar 15 Ayus, J. Article Google Scholar 18 Hoorn, E. Article Google Scholar 19 Rivkees, S. Article Google Scholar 20 Sherlock, M. Article Google Scholar 21 Decaux, G. Article Google Scholar 24 Ayus, J. Article Google Scholar 31 Raff, H. Coolens' equation was used to calculate free cortisol estimate in both TBI patients and comparisons We wanted to show that PC and free cortisol correlate closely, which would mean that PC is a good surrogate marker for in vivo free cortisol levels.

We believed that it would be ethically inappropriate to withhold steroid replacement in patients with acute hypocortisolemia. If a patient whom we had started on hydrocortisone needed to have their PC and CBG measured as part of the study protocol, their hydrocortisone was held for 2 doses preceding venesection, meaning that they would not have received any hydrocortisone for 15 h preceding venesection, allowing full clearance of the previous hydrocortisone dose Patients who were commenced on hydrocortisone and had persistent hypocortisolemia remained on it after discharge; their dose was weaned to a standard oral maintenance dose of 10 mg twice daily, which is the standard oral replacement dose used in our unit Our first-line test was the insulin tolerance test 29 , which involves the measurement of serum cortisol and serum GH before and at 15, 30, 45, 60, 90, and minutes after the iv administration of 0.

Because the glucagon test can overestimate the incidence of ACTH deficiency, patients who had evidence of ACTH deficiency on a glucagon test underwent confirmatory testing, using the short synacthen test.

Secondary hypothyroidism was diagnosed if patients had low serum FT4 with low or inappropriately normal serum TSH. In males and postmenopausal females, central hypogonadism was defined as low serum estradiol or T with low or inappropriately normal LH and FSH.

Premenopausal females were defined as having central hypogonadism if they were amenorrheic with low or inappropriately normal LH and FSH. Serum IGF-I and prolactin were measured in all patients. Plasma sodium was measured daily. If plasma sodium was abnormally high or low, urine sodium and osmolality were measured. Fluid balance was also recorded daily. Those diagnosed with CDI were initially treated with as-needed doses of desmopressin due to the often transient nature of CDI in these patients.

Those with nonrecovering CDI were maintained on oral desmopressin after discharge. When patients returned for dynamic pituitary testing, a detailed history and clinical examination were performed and plasma sodium was measured. If patients had clinical or biochemical evidence of CDI, they proceeded to a formal evaluation using a standard 8-hour water deprivation test Plasma sodium was measured daily, as outlined above. Those with hyponatremia underwent a careful clinical evaluation to determine the underlying cause of hyponatremia and were treated accordingly see Tables 1 and 2 When patients returned for dynamic pituitary testing, the plasma sodium was again measured.

Data are modified from elsewhere Fluid balance and other key clinical and radiological parameters such as GCS, mean arterial pressure, pressor requirements, sedation requirements, computed tomography results, and medications were recorded daily by the authors M.

The purpose of the study was explained carefully to patients and relatives, who were provided with written information on the background to the study. Due to the severity of most patients' illness, written consent was usually given by next of kin. If patients' neurological status improved, they were reconsented and given the choice to withdraw from the study. The interassay CV is 5.

Free cortisol was calculated using Coolens' equation as follows: [ 0. This equation has been previously validated as a surrogate measure of free cortisol Data are nonparametric and are expressed as median range unless otherwise stated. One hundred TBI patients [84 males, median age 33 years range 18—75 years , 85 male] were recruited.

None had major abdominal trauma, severe hypotension, or multiple organ failure. Overall median length of stay LOS was 17 days range 2— days. Fifteen comparison subjects were also recruited [13 males, median age Ten comparison subjects were admitted for elective abdominal aortic aneurysm repair and 5 were admitted for femoral-popliteal bypass surgery.

Most patients 63 of 78 developed these inappropriately low PC measurements on days 1—3 the days in which comparisons had the highest PC , with recovery occurring in 48 of 78 In contrast, none of the comparison subjects died.

Length of stay was higher in those with transient Points, individual PC measurements; bars, median PC. A free cortisol measurement was mathematically calculated for these samples. Two died after their discharge from the hospital. This left 79 patients eligible for dynamic pituitary assessment.

Thirty-two of 79 Of the remaining 47 patients, 9 were not from Ireland and returned to their home country upon discharge from the hospital. Ten patients were found to be in long-term residential care and unable to attend for testing.

Nine others refused testing and 19 were lost to follow-up. Fifteen of 32 Nine of 32 with a history of seizure Of those who underwent glucagon testing, 5 of 9 also had a short synacthen test to further evaluate their ACTH reserve because the glucagon test can overdiagnose ACTH deficiency.

The median time to dynamic pituitary testing was 14 months range 6—24 months. Six of 32 patients Three of 6 previously had acute cortisol deficiency, 2 of 6 previously had acute cortisol deficiency and CDI, and 1 previously had acute CDI but a normal acute cortisol response to illness. Six of 32 had severe GH deficiency, all of whom had had acute cortisol deficiency and 2 of 6 One patient male was gonadotropin deficient; he previously had acute cortisol deficiency and CDI but not acute gonadotropin deficiency.

No patients were TSH or prolactin deficient. Overall, 11 of 32 Therefore, far fewer patients had chronic hypopituitarism compared with those who had acute hypocortisolemia or CDI. However, all patients who were later found to have a chronic pituitary deficit had suffered from either acute hypocortisolemia or acute CDI during their admission with TBI.

In 11 of 51 The median day of onset of transient DI was 6 days range 1—9 days. The median duration of transient CDI was 4 days range 1—13 days. Patients who developed CDI had a significantly higher mortality than those who did not [17 of 51 There were no cases of dysnatremia in the comparison group.

Fifteen percent developed hyponatremia in the acute phase after TBI, which was transient in all patients. The median day of onset of hyponatremia was 3 days range 1—9 days. There were no cases of cerebral salt-wasting syndrome. There were no differences in mortality or LOS in those patients who developed hyponatremia and those who did not. No patients had hyponatremia when reevaluated 6 or more months after TBI. There were no significant differences between those patients with acute cortisol deficiency and those with no pituitary deficits in terms of age, gender, mean arterial pressure, initial GCS, or initial computed tomography appearance.

There were also no significant differences in these parameters between those patients with CDI and those with normal posterior pituitary function.

The findings were similar when those patients with long-term anterior or posterior pituitary deficits were examined: there were no significant differences in the above-mentioned parameters between those patients with long-term anterior pituitary insufficiency and long-term CDI and those with no long-term pituitary deficits. The data from this study show a number of novel findings.

Low plasma cortisol concentrations, inappropriate for an acutely sick patient, were extremely common in TBI patients. There was a clear association between acute cortisol deficiency and subsequent mortality, and acute hypocortisolemia was also found to be predictive of chronic hypopituitarism.

The results have additional potential clinical importance, given that most low readings occurred during the first 3 days after admission, at the time when the patients were most unwell, and at a time when plasma cortisols in the comparison group were at their highest. The comparisons were significantly older than the TBI cohort; this is unsurprising, given that most TBI patients are young healthy males for which there is no readily available unwell in-hospital comparison group.

The diagnosis of cortisol deficiency in the acutely unwell patient is not easy. Some guidelines recommend evaluating cortisol reserve with a short synacthen test before commencing steroid therapy in the critically ill 33 , but this is clearly inappropriate in the setting of TBI because the acute nature of the ACTH deficiency will not have had time to cause secondary adrenal deficiency, and patients will therefore mount a normal cortisol response to synacthen.

For this reason, we selected the measurement of sequential am PC, with comparison with multiple measurements in a comparison group, to assess cortisol secretion. There has been concern that total cortisol measurement may overestimate the true burden of hypocortisolemia in the critically ill patient However, much of the previous data in this area focused on patients with septic shock and consequent hypoalbuminemia 36 , 37 ; in contrast, our TBI patients are mostly young, previously healthy males, capable of maintaining normal plasma concentrations of hormone binding proteins.

Savaridas et al 38 suggested that plasma CBG concentrations fall after TBI; however, a prospective study of TBI patients found that free cortisol rose in parallel with total cortisol as severity of TBI increased 5 , and it has been suggested that discrepancies will not emerge between PC and free cortisol unless patients have low serum albumin concentrations Serum albumin and CBG did not fall in our group, and we found a close correlation between plasma total cortisol and free cortisol across all patients Figure 3.

These data confirm that the measurement of total cortisol is a good marker of free cortisol level in TBI patients and therefore reliably detects cortisol insufficiency. In addition, the lowest PC quartile had the highest mortality, despite empirical treatment with iv hydrocortisone.

There are a number of potential explanations for this. The first is that the inappropriately low PC was simply a marker for more serious TBI and that treatment with hydrocortisone, although consistent with routine management of hypopituitarism, could not reverse the fatal nature of the severe brain injury. The second possibility is that the treatment regimen was insufficient. However, treatment corresponded to the maximal stressed daily production of cortisol Although there is no consensus with regard to what treatment regimen should be used in these patients 41 , it seems unlikely that treatment was insufficient.

Finally, glucocorticoid therapy itself may have contributed to mortality. Many physicians are reluctant to treat critically ill patients with corticosteroids because previous studies have suggested that giving supraphysiological steroid doses to critically ill patients may increase mortality 42 — However, the patients in previous studies mentioned had no evidence of hypocortisolemia, whereas it would be counterintuitive to withhold corticosteroid treatment in patients with inappropriately low PC in a setting known to be associated with hypopituitarism.

A randomized controlled trial will be necessary to address this uncertainty however. One of the key aims of the study was to identify predictors of chronic hypopituitarism because untreated hypopituitarism is associated with excess morbidity and premature mortality after TBI 44 , Our new finding was that acute hypocortisolemia is strongly associated with long-term hypopituitarism. Dynamic pituitary testing is expensive and labor intensive and not applicable to the large numbers of TBI patients who present to overburdened health care providers.

No hemodynamic or radiological criteria reliably predict the risk of long-term anterior hypopituitarism after TBI because only 3 studies have looked at both acute and long-term anterior pituitary failure in the same patient cohort. The first found that early neuroendocrine abnormalities after TBI may be transient, but new deficits may emerge during the chronic phase after TBI Although autoimmune Addison disease most often occurs sporadically, it can occur as a component of autoimmune polyendocrinopathy syndromes, consisting of a constellation of disorders such as candidiasis and autoimmune conditions hypothyroidism, hypoparathyroidism, ovarian failure, vitiligo, gastritis, type 1 diabetes mellitus, hepatitis which should be suspected and ruled out in presence of acquired adrenal insufficiency in children [ 2 ].

Diagnosis of AI in children and young people can be challenging, especially for the chronic form of the disease, as it presents with vague and non-specific symptoms. Clinical presentation of chronic AI comprise fatigue, weight loss, nausea, vomiting, abdominal pain, salt craving, muscle and joint pain.

More specific signs are hypotension and skin hyperpigmentation; the latter is generally more prominent on mucosae and on sun-exposed areas or over pressure points such as elbows and knees, and result from enhanced activation of skin melanocortin 1 receptors [MC1R]. In our patient the initial diagnosis of SIADH seemed unlikely due to the presence of syncope and dizziness on standing, increased thirst, no weight gain and normal urinary output.

Of note, absence of hyperkalaemia and hypoglycemia does not rule out adrenal insufficiency, so initial presentation with hyponatraemia and high urinary sodium may actually mimic SIADH [ 1 , 5 ].

While intravenous fluid replacement in the suspect of a gastroenteritis would be of benefit, fluid restriction due to a SIADH misdiagnosis can be harmful and even potentially life-threatening.

Intravenous hydrocortisone sodium succinate is urgently required for therapy. A rapid correction of severe hyponatraemia may causes pontine myelinolysis and even death [ 6 , 7 ]. Intramuscular Hydrocortisone may be necessary, increasing by two or threefold the dose during episodes of stress, if the patient is not able to take the treatment orally. Hence emergency physicians and pediatricians should be familiar with this diagnosis to enhance early recognition of this potentially life-threatening condition.

Severe hyponatraemia with absence of hyperkalaemia in rapidly progressive Addison's disease. BMJ Case Rep. Primary adrenal insufficiency in children: twenty years experience at the Sainte-Justine Hospital, Montreal.

J Clin Endocrinol Metab. June ;90 6 — The diagnosis and treatment of adrenal insufficiency during childhood and adolescence. Arch Dis Child. Article Google Scholar. Hsieh S, White PC. Presentation of primary adrenal insufficiency in childhood. Yadav D, Akhtar A. A case of primary and secondary adrenal insufficiency in children. Clin Pediatr Phila. Zieg J.

Evaluation and management of hyponatraemia in children. Acta Paediatr. The importance of re-evaluation, re-investigation and follow-up of adrenal insufficiency.