Analysis of water and electrolyte imbalance in a patient with adipsic hypernatremia associated with subfornical organ-targeting antibody

Abstract

Patients with adipsic hypernatremia present with chronic hypernatremia because of defects in thirst sensation and dysregulated salt appetite, without demonstrable hypothalamic structural lesions. The involvement of autoantibodies directed against the sodium channel, Na_x in the subfornical organ (SFO) has recently been reported. However, the pathophysiology of water and electrolyte imbalance underlying the disease has yet to be elucidated. We describe the case of a 5-year-old boy who complained of headaches and vomiting that gradually worsened. Brain magnetic resonance imaging detected no abnormal lesions. Blood laboratory testing revealed a serum sodium (Na) concentration of 152 mmol/L and a serum osmolarity of 312 mOsm/L. His body weight had slightly decreased, and his thirst sensation was absent. His plasma vasopressin concentration was 0.9 pg/mL, despite the high serum osmolarity. He was encouraged to drink water, and oral 1-deamino-8-D-arginine-vasopressin was administered. When serum sodium concentrations were normalized, plasma vasopressin concentrations were apparently normal and ranged from 0.8 to 2.0 pg/mL. He did not present with polyuria at any time. Immunohistochemical study using mouse brain sections and the patient’s serum revealed the deposition of human immunoglobulin G (IgG) antibody in the mouse SFO. In conclusion, our observations suggested that water and electrolyte imbalance in adipsic hypernatremia is characterized by a certain amount of vasopressin release regardless of serum sodium concentrations with no response to hyperosmolarity.

Introduction

Patients with adipsic hypernatremia present with chronic hypernatremia because of defects in their thirst sensation and excessive salt ingestion, but do not have demonstrable hypothalamic structural lesions [1]. Recently, the involvement of autoantibodies directed against the sodiumchannel, Na_x in the subfornical organ (SFO), one of the circumventricular organs (CVOs), has been reported [2–4].

However, the pathophysiology of water and electrolyte imbalance underlying the disease has yet to be elucidated. Upward resetting of both the osmotic set point for vasopressin release and the threshold for thirst sensation is a possible mechanism [4], but this may not explain the absence of polyuria when serum sodium levels are within the normal range. Here, we report a case of a 5-year-old boy who was diagnosed with adipsic hypernatremia, with anti-SFO antibody. We analyzed the relationships between serum sodium levels and plasma vasopressin levels with consideration of volume status.

Case report

A 5.5-year-old Japanese boy was admitted to our hospital because of headaches and vomiting over the preceding 6 months without a history of preceding infections. His medical history revealed no perinatal problems. No developmental delay was noted, and his growth was normal. He was prescribed acetaminophen and ibuprofen when he had a headache, but his symptoms worsened, and the frequency of the headaches gradually increased. He did not present with polydipsia and polyuria. Physical examination on admission revealed normal height (113.5 cm + 1.0 standard deviation (SD)) and body weight (18.5 kg) for his age and sex, along with normal blood pressure (100/64 mmHg), normal heart rate (100 beats/min), prompt capillary refill, and normal skin turgor. He had no pretibial pitting edema. There were no clinical features of precocious puberty, and neurological examination findings were unremarkable. Finally, brain magnetic resonance imaging detected no abnormal lesions.

Laboratory data

As shown in Table 1, the patient’s serum sodium concentration and osmolarity were elevated. Plasma vasopressin concentrations was low at 0.9 pg/mL, considering his high serum sodium concentration of 152 mmol/L. Thyroid function was normal and hypernatremia due to Cushing’s syndrome and primary aldosteronism were ruled out, because his cortisol and adrenocorticotropic hormone concentrations were within the normal ranges and his aldosterone concentration was not elevated. Serum prolactin concentration was slightly high. Urine-specific gravity and osmolarity were elevated, which suggested intact urine concentrating capacity (Table 1).

Table 1.

The results of laboratory tests on admission

	Value	Reference range
Serum biochemistry
Sodium	152 mmol/L
Potassium	3.8 mmol/L
Chloride	107 mmol/L
Creatinine	0.42 mg/dL
Blood urea nitrogen	10 mg/dL
Osmolarity	312 mOsm/L
Blood endocrinological tests
Vasopressin	0.9 pg/mL
Thyroid stimulating hormone	2.9 mIU/mL	0.7–6.4
Free triiodothyronine	3.5 pg/mL	2.3–6.6
Free thyroxine	1.4 ng/dL	0.8–2.2
Adrenocorticotropic hormone	10.5 pg/mL	7.2–66.3
Cortisol	11.5 μg/dL	7–25
Plasma renin activity	0.6 ng/mL/hr	0.3–5.4
Aldosterone	41.0 pg/mL	38.9–306.8
Prolactin	19.2 ng/mL	1.2–12
Growth Hormone	2.6 ng/mL	0.3–3.4
Insulin-like growth factor-1	110 ng/mL	44–193
Luteinizing hormone	0.14 mIU/mL	< 0.1–0.4
Follicle-stimulating hormone	1.4 mIU/mL	0.4–2.0
Urinalysis
Urine-specific gravity	1.030
Urinary creatinine	120 mg/dL
Urinary sodium	235 mmol/L
Urinary potassium	60.3 mmol/L
Urinary chloride	205.4 mmol/L
Urine osmolarity	898 mOsm/L

Clinical course

The patient’s clinical course is shown in Fig. 1. Intravenous hypotonic fluid was administered, and salt intake was restricted to < 5 g/day. His water intake was 600 mL/day, and his urine volume was approximately 1000 mL/day. Three days after admission, the serum sodium concentration decreased to 140 mmol/L, and his plasma vasopressin level was 1.0 pg/mL. Intravenous fluid administration was discontinued 5 days after admission, and he was encouraged to drink 1500 mL water per day. His body weight was 18.9–19.0 kg after adequate rehydration, which was considered his appropriate body weight with normal fluid status. Urine volume increased to approximately 1500 mL/day. However, after discharge, he again complained of headaches and his serum sodium concentration was increased to 148 mmol/L, and his body weight was decreased to 18.3 kg, 25 days after admission. Therefore, 120 µg of oral 1-deamino-8-D-arginine-vasopressin (dDAVP) per day was started, which resolved the headaches. Two weeks after starting dDAVP, the serum sodium concentration decreased to 144 mmol/L, and his body weight increased to 19.1 kg. Since then, hypernatremia and body weight loss have not recurred.To characterize the pathophysiology of the dysregulation in sodium and water balance in the present case, we analyzed the relationships between serum sodium concentrations and plasma vasopressin concentrations before we initiated dDAVP. As shown in Fig. 2, plasma vasopressin concentrations did not increase in response to increased serum sodium concentrations, and body weight and urine volume decreased when serum sodium concentrations increased. Urine osmolarity increased when serum sodium concentrations increased and urine volume decreased, which reflected intact urine concentrating capacity. In contrast, urine sodium excretion did not change significantly when serum sodium concentrations increased. A 24-h urine collection during outpatient management showed that the patient’s urinary sodium excretion level was 16.3 g/day, which reflected excessive salt intake.

Fig. 1.

The clinical course of the patient. Intravenous hypotonic fluid was given until 5 days after admission. Serum sodium concentration decreased to 140 mmol/L and his plasma vasopressin concentration was 1.0 pg/mL at 3 days after admission. His body weight increased to 18.9–19.0 kg and urine volume was around 1500 mL/day after adequate rehydration. After discharge, he complained of headaches and his serum sodium concentration was increased to 148 mmol/L with decreased body weight to 18.3 kg at 25 days after admission. Oral dDAVP was started, which ameliorated hypernatremia. dDAVP 1-deamino-8-D-arginine-vasopressin

Fig. 2.

The relationships between serum sodium concentrations and plasma vasopressin concentrations, body weight, urine volume, UNa/UCr and urine osmolarity. Plasma vasopressin concentrations were maintained at almost the same levels and did not increase in response to elevated serum sodium levels. The shaded area is the normal range of plasma vasopressin levels [5]. Body weight and urine volume decreased and urine osmolarity increased when serum sodium levels increased. In contrast, UNa/UCr did not significantly change. UNa/UCr Urinary sodium-to-creatinine ratio

We attempted to detect an autoantibody against a protein in the SFO in the patient’s serum, as previously reported [5]. Perfusion-fixation of the brains of male 8–16-week-old C57BL/6 J mice (CLEA Japan Inc., Tokyo, Japan) was performed.The patient’s serum was diluted 1:250 with blocking solution and was added to the brain sections for 48 h. After washing, human immunoglobulin G (IgG) antibody and blocking solution (1:500) were added, and the sections were incubated for an additional 24 h. Then, the sections were washed and examined using a confocal microscope (Keyence BZ-9000; Keyence Corp., Osaka, Japan; ×20 magnification). Human IgG deposition was observed in the SFO, suggesting the presence of autoantibodies in the patient’s serum that targeted a protein in the SFO (Fig. 3a–c), which is located at the undersurface of the fornix in the upper part of the third ventricle (Fig. 3d).

Fig. 3.

Immunofluorescent study of the SFO with anti-human IgG. Brain tissues obtained from wild-type male mice (C57Bl/6 J CLEA Japan, Tokyo, Japan) were dissected and embedded in OCT compound (Sakura, Tokyo, Japan) and sectioned at 30–40 μm with a cryostat (CM3050S, Leica Microsystems, Wetzlar, Germany). The sections were then incubated with serum from a previously reported patient (positive control) (a) [7], the present patient (b), and a normal control (c). The bound antibodies were detected with goat anti-human IgG (H + L) cross-adsorbed secondary antibody conjugated with Alexa Fluor 488 (1:500). Sections were analyzed by confocal fluorescence microscopy (OlympusFV-1000D, Tokyo, Japan). Human IgG deposition was observed in the SFO of murine brains incubated with the present patient’s serum (b). Original magnification, 40 × . d SFO is located at the undersurface of the fornix in the upper part of the third ventricle. SFO, subfornical organ

To detect the anti-Na_x antibody in the patient’s serum, HEK293 cells expressing human Na_x (SCN7A gene) were generated by transfection using an expression vector and were compared with untransfected control cells. Western blot analyses were performed using membrane fractions prepared from lysates of each HEK293 cell type, an aliquot of the patient’s serum as the primary antibody, and anti-human IgG antibody as the secondary antibody. The results were negative (data not shown).

Discussion

Serum sodium concentration and osmolarity are monitored and controlled by specific sensors in the brain that are responsible for the sensation of thirst, salt appetite, and vasopressin release. The sensation of thirst and the suppression of salt appetite are mediated by neurons in the SFO [6]. Hiyama et al. reported that autoantibodies directed against Na_x which is an atypical Na channel that expresses in the SFO and CVOs, can be responsible for adipsic hypernatremia [5, 7, 8]. The injection of immunoglobulin obtained from a patient with adipsic hypernatremia into mice led to inflammation in the SFO [7]. Consistent with this finding, in the present case, IgG antibody against a protein in the SFO could explain the defects in thirst sensation and the dysregulated salt appetite.

To better understand the pathophysiology of water and electrolyte imbalance underlying the disease, we analyzed the relationships between plasma vasopressin concentrations and serum sodium concentrations with consideration of volume status. As a result, the plasma vasopressin levels were maintained at almost the same levels and did not increase in response to hyperosmolarity. When serum sodium levels were within normal ranges, apparently normal secretion of vasopressin was observed, which was compatible with the lack of polyuria in our patient. Additionally, as previous reports have described [4, 5], urine concentrating capacity was intact, which was also compatible with a certain amount of vasopressin release. These findings suggested that water and sodium imbalance in adipsic hypernatremia is explained by dysregulation of vasopressin release owing to a lack of response to hyperosmolarity rather than upward resetting of the osmotic set point. However, we cannot entirely exclude a possibility of upward resetting of the osmotic set point, because measurements of plasma vasopressin levels were limited to a few time points. The SFO is a center for the monitoring of body fluid conditions and the control of water and salt intakes, and the SFO contains neurons innervating hypothalamic nuclei for vasopressin release [9, 10]. In the SFO, osmolarity sensors, such as transient receptor potential vanilloid 4 (TRPV4), are expressed in addition to the Na_x channel [11]. Liedtke et al. reported that TRPV4-knock out mice showed reduced fluid intake and an impaired vasopressin response to a hyperosmolar stimulus [12]. However, it has not been clearly demonstrated that TRPV4 is involved in the pathogenesis of adipsic hypernatremia [12].

Apoptosis of neurons in the SFO caused by anti-SFO antibodies may lead to defects sensing serum sodium concentrations and osmolarity [5]. Other pathways stimulating vasopressin release include arterial baroreceptors that communicate with the hypothalamus and modify vasopressin release through cranial nerves IX and X [13]. This pathway may have contributed to maintaining a certain amount of vasopressin release in our patient with impaired vasopressin release via the SFO pathway.

We further analyzed serial changes in urine volume, urinary chemistry and volume status to determine the factors contributing to hypernatremia. Body weight and urine volume decreased, and urine osmolarity increased when serum sodium concentrations were elevated (Fig. 2). Urinary sodium excretion did not change significantly when serum sodium concentrations increased. Moreover, a 24-h urine collection indicated excessive salt intake. These observations suggested that a decreased free water intake under the condition of persistent excessive salt intake and a certain amount of vasopressin release contributed to developing hypernatremia and elevated urine osmolarity in this patient. These findings are consistent with previous experiments indicating that mice with anti-SFO antibodies show excessive salt intake in the presence of hypernatremia, which is explained by defective of serum sodium concentration sensing caused by anti-SFO antibodies [5]. Therefore, adequate supplementation of free water and salt restriction are required for the management of adipsic hypernatremia. The clinical course of our patient suggested that administering dDAVP is also effective in the management of serum sodium levels.

In conclusion, we described a case of adipsic hypernatremia with anti-SFO antibody. A certain amount of vasopressin release regardless of serum sodium concentrations was observed, which contributed to a better understanding of the pathophysiology of water and sodium imbalance in adipsic hypernatremia.

Abbreviations

SFO

Subfornical organ

CVOs

Circumventricular organs

SD

Standard deviation

dDAVP

1-Deamino-8-D-arginine-vasopressin

Declarations

Conflict of interest

The authors confirm that they have nothing to declare.

Ethical approval

All procedures performed in studies involving human participant and animals were in accordance with the ethical standards of the institutional research committee at which the studies were conducted (IRB approval number R2-8-1 and A17-157) and with the 1964 Helsinki Declaration and its subsequent amendments or comparable ethical standards.

Informed consent

Written informed consent was obtained from the patient’s guardian.