Extreme Hyponatremia Complicated by Osmotic Demyelination in a Previously Healthy Young Individual

Rationale: Severe hyponatremia can lead to dramatic complications whether it is treated or not. At times, it may be very severe (serum Na concentration: NaS < 115 mmol/L) or even extreme (NaS < 105 mmol/L)a and its cause difficult to identify, especially in younger individuals with no history of water disorders. The case presented herein illustrates these points quite eloquently and leads us to believe that the current recommendations for the treatment of very severe hyponatremia require some fine-tuning. Presenting Concerns: A 26-year-old man was admitted to our intensive care unit for a NaS of 88 mmol/L in the absence of obvious extracellular fluid volume contraction. He had been experiencing vomiting, diarrhea, fatigue, and excessive thirst for the past 6 weeks and minor neurological symptoms just before admission. Laboratory tests at presentation also showed a urine osmolarity of 697 mOsm/L and urine Na of 40 mmol/L. Diagnoses: The presenting concerns were consistent with syndrome of inappropriate antidiuretic hormone secretion (SIADH) manifesting as extreme, yet mildly symptomatic hyponatremia. At the same time, they did not point toward a specific cause initially. Interventions: The patient was treated through water restriction, subcutaneous desmopressin, and various intravenous (IV) fluids. Our goal had been to increase NaS at a rate of 4 to 6 mmol/L/day and required the amount of NaCl and free water perfused hourly to be readjusted constantly. Access to water also had to be opposed as the patient was unable to tolerate his thirst. Outcomes: During the first 6 days, the rate of NaS correction achieved was ~6 mmol/L/day. The patient improved initially but at the end of day 6, he experienced severe extrapontine osmotic demyelination (with widespread pyramidal and extrapyramidal deficits) that did not respond to intravenous immunoglobulin and NaS relowering. A little more than 3 weeks later, he began to develop low blood pressure and a subfebrile state that revealed secondary to severe Addison disease. The water disorder and insatiable thirst subsided gradually upon replacing the deficient hormones but the neurological disorder went on to become permanent and highly disabling. Teaching points: (1) Very severe hyponatremia should always be handled as an emergency and monitored stringently in view of its potential to cause irreparable damage. (2) Because it is a major risk factor for osmotic demyelination, it should probably be corrected at a rate of less than 4 mmol/L/day especially if it is in the extreme range, chronic, or of unknown duration. (3) It can be a presenting manifestation of Addison disease.


Introduction
Hyponatremia, defined as a serum Na concentration (Na S ) of 134 mmol/L or less a , is the most commonly encountered electrolyte abnormality in clinical practice. When it is hypotonic, it must be seen as a defect of water homeostasis that can translate into variable manifestations depending on its severity and chronicity. At times, hyponatremia can prove challenging as to its possible etiology or management. Herein, we describe such a case whose outcome was a devastating neurological complication.

Presenting Concerns
A formerly healthy and unmedicated 26-year-old man was admitted to our intensive care unit (ICU) for extreme hyponatremia (Na S of 88 mmol/L) identified in another hospital the day before. He had consulted initially for mild dysarthria, confusion, and lethargy of recent onset following a 6-week history of intermittent vomiting, diarrhea, fatigue, and intractable thirst. Before his transfer, there were no signs of focal neurological deficits or substantial extracellular fluid volume (ECFV) contraction based on physical examination.

Clinical Findings
On arrival at our ICU, the initial clinical status was as described. In particular, blood pressure was 102/76 mm Hg with still no evidence of substantial ECFV depletion. Relevant laboratory findings at presentation and imaging features were as shown in Figure 1. Besides hyponatremia, they revealed a urine osmolarity (OSM U ) of 697 mOsM/L, a urine Na (Na U ) of 40 mmol/L, a slight increase in serum thyroid stimulating hormone and evening cortisol, and a normal magnetic resonance imaging (MRI) of the head on day 5 (see panels A1, B, and C1). Note. Basal ganglia are diffusely hyperintense on C2 and more severely so on C3 (white arrows). A small portion of the internal capsules are also hyperintense. ALP = alkaline phosphatase; WBC = white blood cell; OSM U = urine osmolarity; ALT = alanine aminotransferase; TSH = thyroidstimulating hormone; Na U = urine Na; ACTH = adrenocorticotropic hormone; CRH = corticotropin-releasing hormone; OHase = hydroxylase; ODS = osmotic demyelination syndrome; MRI = magnetic resonance imaging; TPO = thyroid peroxidase. *Levothyroxine had already been started a few days postadmission for mild hypothyroidism of unknown cause initially. † Thus, first value corresponds to the difference between initial Na S at day 0 and mean Na S for the remainder of day 1.

Diagnostic Focus and Assessment
Based on the presenting concerns, it was concluded that the patient suffered from mildly symptomatic hyponatremia due to syndrome of inappropriate antidiuretic hormone secretion (SIADH). Whether he could have also suffered from mild ECFV contraction or been prone to excessive water intake as contributory factors could not be ruled out. At that time, the severity and etiology of hyponatremia were otherwise unclear to us.

Therapeutic Focus and Assessment
Given that the hyponatremia was both extreme and only mildly symptomatic, it was decided to correct Na S at a rate of 4 to 6 mmol/L/day. This goal was achieved by restricting water ingestion, preventing water diuresis with desmopressin (1-2 μg subcutaneously 2 or 3 times a day) and administering intravenous (IV) fluids at rates 0-150 mL/h and NaCl 0-513 mmol/L, which varied according to repeated Na S measurements. Several adjustments had to be made every day, and the patient had to be physically restrained from access to water.

Follow-up and Outcomes
During the first 6 days at the ICU, the overall clinical condition improved partially and serum HCO 3 concentration (with 3 underscripted as written) normalized progressively. As for the rate of Na S correction, it increased from 88 to 131 mmol/L over these first 6 days, that is, by a daily mean of 2 to 9 mmol/L/day for an all-around mean of ~6 mmol/L/day (see Figure 1B).
At the end of day 6, the patient became acutely stuporous while Na S were all 132 mmol/L or lower and while the mean Na S correction rate had increased by 5.5 mmol/L compared with the day before. A control MRI then showed that the basal ganglia had become diffusely hyperintense ( Figure  1C2), consistent with extrapontine osmotic demyelination syndrome (ODS). In an attempt to reverse this condition, a single dose of 20 g intravenous immunoglobulin was administered and Na S relowered with hypotonic IV fluids to an empirical target of ~115 mmol/L ( Figure 1B). The patient regained awareness a few days later but developed transient akinetic mutism followed by severe and widespread pyramidal and extrapyramidal deficits.
Less than a month after the onset of ODS, he became subfebrile and hypotensive (with blood pressures of 80-100/40-70 mmHg) while affected still by intractable thirst. He was eventually found to have autoimmune polyendocrine deficiency with thyroiditis and severe Addison disease ( Figure  1A2) for which he was initially treated with oral hydrocortisone (40 mg at 8 a.m., 20 mg at 12 p.m., 10 mg at 6 p.m.) and fludrocortisone (0.1 mg daily). After only 2 days under this regimen, his thirst became much less intense.
As the cause of SIADH had been elusive and was still uncertain despite the identification of an autoimmune polyendocrine disorder, additional tests were ordered including serum anti-AQP4, a focused genetic panel of water disorders b and a Ga-DOTATATE-positron emission tomodensitometry scan. However, no other abnormalities could be identified, and the water disorder went completely extinct over several days even when liberal access to fluids was eventually permitted.
Although extreme a , hyponatremia in this case probably occurred as a result of primary adrenal failure that had begun indolently 6 weeks before admission. A Na S of 88 mmol/L was not only unexpected in this setting unless associated with severe hyperglycemia. but had never been described in any other settings. In our opinion, the severity of hyponatremia was the main risk factor for the ODS and probably explains why the patient failed to recover in the long term ( Figure 1C3).

General Questions
This clinical vignette is that of a puzzling case of extreme hyponatremia (Na S < 105 mmol/L) a . It took us some time to figure out why the hypotonic water disorder had developed, why this condition had been so severe, and why it had led to such a drastic complication. It is by revisiting the differential diagnosis, risk factors, and treatment of severe hypotonic hyponatremia that we were able to find answers to several of our questions.

What was the Cause of Hyponatremia in Our Patient?
The causes of hyponatremia have been traditionally grouped into three categories based on the associated volume status (hypovolemic, euvolemic, or hypervolemic) to aid in their identification and treatment. 1 Yet, there are multiple reasons as to why such a classification should probably not be used. In particular, it calls for potentially inadequate therapeutic measures and revolves around a clinical parameter that is notoriously difficult to assess.
Many experts recommend classifying the causes of hyponatremia based on OSM U and Na U (Figure 2A) as a preferred approach to identify the mechanisms at play. 2 In the panel shown, Addison disease and hypothyroidism may appear to have been misplaced as they are often seen as exclusion criteria for a diagnosis of SIADH. 3 Yet, one must remember that in both endocrinopathies, antidiuretic hormone (ADH) secretion is partly inappropriate. 4,5 Looking again at Figure 2A, it would thus appear that our patient suffered from a SIADH-type of disorder (follow red arrows) as OSM U was 697 mOsm/L and Na U was 40 mmol/L in the absence of renal dysfunction and diuretic treatment. Given that an exhaustive imaging exploration had also failed to reveal abnormalities aside from ODS, a limited number of etiological possibilities could have accounted for the water disorder ( Figure 2B).
Among those listed, Addison disease would have certainly merited consideration early on. In particular, hyponatremia in this condition can be at times very severe a with Na S going down to as low as ~110 mmol/L based on a few case reports. 6 In addition, subacute Addison disease is often difficult to diagnose in the early stages unless it is searched for through a corticotropin stimulation test.
During hospitalization, it was found that the patient did suffer from Addison disease along with thyroiditis. Although adrenal failure was not grossly apparent initially, it was still the main cause of hyponatremia in our view given that a thorough investigation failed to reveal other possibilities (see Figure 1 and Note b) and that Na S normalized under hormonal therapy while water restriction could be lifted progressively.

Why was Hyponatremia So Severe?
As far as we know, a Na S of less than 90 mmol/L has never been reported before except if uncorrected for effective tonicity. This presentation was thus puzzling to us especially in the context of Addison disease. At the same time, it could have simply indicated that water homeostasis was perturbed excessively through a concomitant defect of maximal renal dilution capacity or through excessive water intake. In our case, such possibilities appeared likely based on the absence of overt Addison disease on admission.
When ADH levels are already inappropriately high regardless of the etiology at play, a decrease in effective circulatory volume or genetic predisposition is among the various factors that could further compromise renal dilution capacity. However, these factors did not appear to play a major role for our patient given that there were no obvious signs of substantial ECFV contraction at presentation and that the genetic tests failed to reveal a contributory defect.
Thiazide consumption is a well-known cause of mild hyponatremia and another factor that can compromise renal dilution capacity. From time to time, however, it can lead to severe reductions in Na S as if a coexisting factor was also at work. According to one study, this cofactor could be a preexisting tendency toward potomania, 7 a tendency that would be well-tolerated and clinically silent until unmasked by an iatrogenic reduction in renal dilution capacity.
We believe that our patient could have thus developed extreme hyponatremia in the setting of Addison disease as he was prone to excessive water ingestion to start with. That his thirst was very important throughout most of the hospital stay would be consistent with this hypothesis. The genetic tests conducted in this regard were not conclusive but the molecular mechanisms of thirst control are still ill-defined.

Why did the Patient Develop ODS?
The treatment of severe hyponatremia is still debated. However, the actual consensus is that Na S should be increased by less than 8 mmol/L/day (and even less than 6 based on some observations) 8,9 when it is chronic or of unknown duration 1,2,8,9 to prevent osmolyte-poor central neuroglial cells from undergoing shrinkage-induced demyelination. 10 In such situations, there is also consensus that the same rule should apply to very symptomatic hyponatremia but that the 6 to 8 mmol/L of increase allowed should be achieved more rapidly even if it requires Na S to be kept stable for many hours afterward. Note. Hyponatremia in Addison disease is also contributed for by urinary and gastrointestinal losses in sodium salts and by the resulting (appropriate) increase in ADH levels. OSM U = urine osmolarity; VOL U = urinary volume; N = normal; Na U = urine Na; ECF = extracellular fluid; ODS = osmotic demyelination syndrome; CNS = central nervous system; SIADH = syndrome of inappropriate antidiuretic hormone secretion; ADH = antidiuretic hormone. *If OSM U is above 150: mmol/L, renal dilution capacity is compromised, but if serum Na is very low in this context, the more OSM U is close to 150: mmol/L, the more excessive water ingestion contributes to hyponatremia. † Hyponatremia in heart failure and cirrhosis falls under this category. ‡ Low serum cotisol has been found to result in corticotropin hormoneinduced ADH release by parvocellular neurons and hypothyroidism to impair ADH metabolism. 4,5 In our case, ODS occurred even if Na S was increased at a mean rate of ~6 mmol/L/day and in the absence of obvious or known risk factors for ODS other than the severity of hyponatremia (see Figure 2C). For these reasons, we believe that it is the profoundness of the water disorder itself that led to the outcome observed. Although it was pretty much on par with recommendations, the rate of correction chosen could have also been simply too high in the context of a double-digit Na S value.
In the light of this case, we will continue to treat very severe or extreme hyponatremia as before-through the use of water restriction, desmopressin, and various IV fluids 8-10while measuring Na S every 2 hours during the first few days. From this moment on, however, we will try to aim for Na S correction rates of 2 to 4 mmol/L/day under such circumstances.

Conclusion
Hyponatremia of the degree experienced by our patient remains uncharted territory. It could imply that ODS is a near-inescapable aftermath. Unless hyponatremia is very symptomatic, it should probably be treated by raising Na S as slowly as possible until normalization. Desmopressin should also be administered concomitantly as the treatment of hyponatremia (or of its cause) can also lead to acute ADH suppression. The same reasoning should perhaps also apply to very severe hyponatremia to be on the safe side.

Author Contributions
All authors contributed to manuscript revision and PI wrote initial and final versions.

Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: Dr Isenring has received support from the Kidney Foundation of Canada and Canadian Institutes of Health Research. Dr Garneau is supported by a Banting Postdoctoral Scholarship from the Canadian Institutes of Health Research.

Ethics Statement
Written informed consent was obtained from the patient described in this report.

Data Availability and Ethics Statement
Data will be made available upon request provided that participant privacy is assured.