Pattern Recognition versus Pathogenesis

Patient

A 29-year-old man with a history of migraines presents to urgent care with a headache and vomiting. He had been feeling poorly in recent months, which he attributed to anxiety from the coronavirus disease 2019 restrictions. The week prior to admission, he traveled to Texas, where he drank more alcohol than usual and “ate poorly.” On the day prior to admission, he played golf, and on the course, he had four 12-ounce beers. Afterward, he had a significant headache and drank 50 ounces of water. This made him feel worse; he vomited and decided to go to urgent care.

His medical history was notable for dyspepsia, migraines, and prior coronavirus disease 2019 vaccination. His medications included omeprazole and sumatriptan. The family history was notable for celiac disease in several cousins.

On examination, his BP was 114/78 mm Hg, and heart rate was 67 beats per minute. He had no edema.

Laboratory data (reference ranges) included sodium of 112 mEq/L, potassium of 4.5 mEq/L, chloride of 80 mEq/L, bicarbonate of 22 mEq/L, urea nitrogen of 12 mg/dl, creatinine of 1.3 mg/dl, serum osmolality of 226 mOsm/kg, cortisol of 8.5 µg/dl at 4 am (7–10 am: 6–18.4 µg/dl; 4–8 pm: 2.7–10.5 µg/dl), free T4 of 0.3 ng/dl (0.9–1.7), thyroid-stimulating hormone of 170 mIU/L (0.5–5.0), and adrenocorticotropic hormone (ACTH) of 852 pg/ml (2–60). Urine values were sodium of 178 mEq/L, potassium of 59 mEq/L, and urine osmolality of 774 mOsm/kg.

Question 1

Which of the following is the most likely cause of this patient’s hyponatremia?

• A.Baroreceptor-stimulated vasopressin release
• B.Low solute intake
• C.Hypothyroidism-induced vasopressin release
• D.ACTH-induced vasopressin release

The answer is A.

Question 2

Addison disease is associated with hyperkalemia, which is not seen here. How can this finding be explained?

• A.Aldosterone escape
• B.Increased gastrointestinal losses
• C.Ongoing collecting duct renal outer medullary K⁺ (ROMK) activity
• D.Pendrin upregulation

The answer is C.

Discussion

This patient presents with severe hyponatremia, and investigations revealed primary adrenal insufficiency with inappropriately low cortisol in the setting of acute illness and a very high ACTH. He also had severe primary hypothyroidism with low T4 and high thyroid-stimulating hormone. Most textbooks list adrenal insufficiency and hypothyroidism as causes of impaired water excretion, prompting a reflex assessment of both the corticosteroid and thyroid hormone axes in patients with hyponatremia. Yet, the pathogenesis of hyponatremia in endocrine disorders is often oversimplified.

The pathogenesis of hyponatremia in primary adrenal insufficiency is complex, and nonosmotic release of arginine vasopressin (AVP) occurs from both stimulation related to hemodynamic factors and the lack of inhibition of AVP by cortisol (Figure 1). Chronic mineralocorticoid deficiency leads to unchecked sodium loss from the distal nephron. Experimental studies in dogs following adrenalectomy show urine sodium losses that persist even in the setting of sodium restriction (1). Similarly, in observational human studies, a significant sodium deficit and negative sodium balance are observed until hormones were restored (2) (indeed, individuals with Addison disease typically describe salt cravings and diets of pickle juice and salt sachets). This sodium loss, although far less than the 5% of filtered sodium usually attributed to distal nephron sodium reabsorption, can result in significant reduction in plasma volume. The response to this sodium loss highlights the importance of angiotensin II in sodium reabsorption and concurrent baroreceptor-mediated AVP release to defend against hypovolemia (3). Studies in rats in which AVP is inhibited after adrenalectomy result in hypotension, reinforcing the notion that AVP release is mediated by baroreceptor tone. Restoration of only the mineralocorticoid axis, however, is not sufficient to reverse the impairment in water excretion observed with adrenalectomy. If adrenalectomy is followed by replacement of mineralocorticoid, hyponatremia still develops when the rats are challenged with a water load (4). In this model, a fall in stroke volume and an increase in heart rate are also present, consistent with corticosteroids being an important contribution to the maintenance of vascular tone (5). Indeed, cortisol plays a key role in the physiologic stress response (e.g., potentiation of vasoconstrictor response to catecholamines, production of epinephrine from the adrenal medulla, and a direct effect on vascular smooth muscles). Prolonged cortisol deficiency can also lead to vasopressin-independent impairment of water excretion related to a fall in glomerular filtration. This effect was identified in cortisol-deficient rats studied at 14 versus 1 day and further characterized after adrenalectomy in rats that have hypothalamic diabetes insipidus and are repleted with mineralocorticoid but not glucocorticoid. Here, the impairment in water excretion was independent of vasopressin and appears to be related to changes in kidney hemodynamics, including an increase in kidney vascular resistance and a fall in kidney blood flow resulting in a decrease in glomerular filtration (4).

Figure 1.

Multiple mechanisms can contribute to the development of hyponatremia in the setting of primary adrenal insufficiency. Primary adrenal insufficiency is characterized by both mineralocorticoid and glucocorticoid deficiency. Both ultimately lead to activation of baroreceptors and elaboration of arginine vasopressin (AVP) by the hypothalamus. Additionally, glucocorticoid deficiency releases negative feedback on corticotrophin-releasing hormone (CRH) and AVP production. AVP-independent impairment of water clearance can also occur from a decline in GFR. In the setting of fluid intake, hyponatremia develops from both AVP-dependent and AVP-independent mechanisms. Figure created with Biorender.com.

The finding that saline infusion in humans does not suppress vasopressin release in adrenal insufficiency led to the hypothesis that loss of negative feedback plays a fundamental role in the development of hyponatremia. Both corticotrophin-releasing hormone and AVP are synthesized in the paraventricular nuclei of the hypothalamus. In rats, absence of cortisol leads to a dramatic increase in corticotrophin-releasing hormone, and with this, AVP is also elaborated (6). Although it is difficult to precisely define the role of glucocorticoid deficiency on vasopressin release in animal models when baroreceptor-mediated AVP release dominates, in transfected cell lines, glucocorticoids have been shown to inhibit the vasopressin promoter (7). This direct transcriptional suppression, rather than synaptic or mechanoreceptor-mediated control of vasopressin release, explains earlier inferences regarding the normal basal inhibition of vasopressin by glucocorticoids, and it may explain the vasopressin-mediated syndrome of antidiuresis in patients with secondary glucocorticoid deficiency who appear euvolemic.

In this case, although this patient reported high fluid intake, the urine data are not consistent with hyponatremia from a low-solute diet. The latter is characterized by urine osmolality that is low, albeit not maximally dilute.

Hypothyroidism is often suggested as a cause for hyponatremia, yet this is extremely uncommon. Although T4 purportedly provides negative feedback for AVP, such that its absence could cause nonosmolar AVP release, this has not been demonstrated experimentally; instead, even in patients with chronic untreated hypothyroidism, normal osmoregulation can be demonstrated (8). Patients with severe hypothyroidism that is characterized by the slowing of function of multiple organs (myxedema) may develop hyponatremia. In those situations, AVP release is likely prompted by reduced cardiac output (e.g., by baroreceptor-mediated AVP release), but hyponatremia is not observed with milder thyroid dysfunction. Hyponatremia is also rare in the setting of acute hypothyroidism. In a series of 220 consecutive patients who underwent radioiodine-induced thyroid ablation following a 2-week sodium restricted diet, only four developed significant hyponatremia (<130 mEq/L), three of whom had either reduced kidney function or coincident diuretic use and the fourth drank 17 L during the 3 days of isolation (9).

Although hyperkalemia is a common expectation in primary adrenal insufficiency, it is present in only roughly 40% of cases. Aldosterone synthase–deficient mice hold the key to this puzzle. In these mice, there is normal potassium homeostasis except following a large potassium load. In these animals, potassium excretion is achieved by normal secretion through ROMK channels independent of aldosterone and in concert with an increase in urine sodium excretion. In contrast, colonic epithelia from the same rodents do not have increased ROMK activity or significant colonic K⁺ excretion, and increases in gastrointestinal potassium losses are unlikely in cases of mineralocorticoid deficiency (10). This complements our current understanding of the effect of high–blood potassium concentration on the early distal nephron. There, potassium effects on K_ir4.1/5.1 channels lead to depolarization of the cells and, ultimately, inhibition of sodium chloride cotransporters. This inhibition favors delivery of sodium to the collecting duct and passive loss of potassium by the principal cells (11).

Aldosterone escape occurs in the setting of high but not low aldosterone levels and refers to the observation that aldosterone-mediated volume expansion leads to increased kidney perfusion pressure, decreased proximal reabsorption, increased atrial natriuretic peptide, and, ultimately, the prevention of edema. Finally, pendrin plays an important role in bicarbonate excretion in alkalosis but not potassium homeostasis.

Additional evaluation of this patient was notable for the failure to respond to ACTH stimulation; his aldosterone was <1 ng/dl (reference range supine 8:00–10:00 am: 3–16), whereas plasma renin activity was 43.75 ml/min per hour (0.25–5.82). He was treated with hydrocortisone, fludrocortisone, and thyroid hormone. His serum sodium improved steadily, and he was discharged from the hospital. Testing for anti–21-hydroxylase antibodies was positive, consistent with autoimmune Addison disease. In follow-up, he reports feeling well and has normal electrolytes. He wears a medical alert bracelet and follows an “adrenal sick day protocol” as needed.

Disclosures

M.P. Hoenig reports receiving honoraria from the Pri-Med conference. The remaining author has nothing to disclose.

Funding

None.

Acknowledgments

For most American Society of Nephrology (ASN) Kidney Week attendees, case-based clinical nephrology talks are one of the most exciting venues. The Nephrology Quiz and Questionnaire is the essence of clinical nephrology and represents what drew all of us into the field of nephrology. The expert discussants prepared vignettes of puzzling cases, which illustrated some topical, challenging, or controversial aspect of the diagnosis or management of key clinical areas of nephrology. These cases were presented and eloquently discussed by our four expert ASN faculty. Subsequently, each discussant prepared a manuscript summarizing his or her case discussions, which serves as the main text of this article (Michael J. Ross and Ashita Tolwani, comoderators).

Author Contributions

M.P. Hoenig and S.H. Lecker conceptualized the report; M.P. Hoenig wrote the original draft; and M.P. Hoenig and S.H. Lecker reviewed and edited the manuscript.