2024 Volume 71 Issue 11 Pages 1087-1092
Adipsic diabetes insipidus (ADI) is characterized by central diabetes insipidus and an impaired thirst response to hyperosmolality, leading to hypernatremia. Hyponatremia observed in patients with ADI has been considered a complication of desmopressin therapy. Herein, we present a case of impaired thirst sensation and arginine vasopressin (AVP) secretion without desmopressin therapy, in which hyponatremia developed due to preserved non-osmotic AVP secretion. A 53-year-old woman with hypopituitarism, receiving hydrocortisone and levothyroxine, experienced hyponatremia three times over 5 months without desmopressin treatment. The first hyponatremic episode (120 mEq/L) was complicated by a urinary tract infection with a plasma AVP level of 33.8 pg/mL. Subsequent hyponatremia episodes occurred after administration of antipsychotic (124 mEq/L) and spontaneously (125 mEq/L) with unsuppressed plasma AVP levels (1.3 and 1.8 pg/mL, respectively). Hypertonic saline infusion did not affect AVP or copeptin levels. Regulating water intake using a sliding scale based on body weight prevented the recurrence of hyponatremia without the use of desmopressin. Except during infection, plasma AVP levels (1.3 ± 0.4 pg/mL) were not significantly correlated with serum sodium levels (rs = –0.04, p = 0.85). In conclusion, we present a unique case of impaired thirst sensation and AVP secretion in which hyponatremia developed without desmopressin therapy. Preserved non-osmotic AVP secretion, possibly stimulated by glucocorticoid deficiency, may contribute to the development of hyponatremia in patients with ADI.
Adipsic diabetes insipidus (ADI) is a rare disorder characterized by central diabetes insipidus and the deficiency or absence of a thirst response to hyperosmolality, leading to hypernatremia [1]. ADI can be caused by surgery for anterior communicating artery aneurysms, tumors, head injury [2], and measles encephalitis [3]. ADI is observed in 9.9% of patients with central diabetes insipidus [4]. Management of the disease is challenging because patients with ADI cannot sense rising osmolality and can quickly develop severe sodium imbalance, leading to marked increases in morbidity and mortality [1].
Patients with ADI usually require treatment with desmopressin, a synthetic analog of vasopressin, at least twice daily to ensure appropriate levels of antidiuresis [5, 6]. Plasma sodium level and body weight should be monitored to maintain the volume and osmolar status [7] and the requisite water intake is calculated [8]. However, despite careful monitoring, these patients frequently experience episodes of hypernatremia and hyponatremia [9]. Hyponatremia is generally explained as a condition of decreased water diuresis that results in water retention [10]. Indeed, a case of severe hyponatremia due to excessive water intake on desmopressin supplementation therapy has been previously reported [11]. Thus, the cause of hyponatremia in those patients has been explained by excessive water intake following desmopressin treatment [9]. Although arginine vasopressin (AVP) release is reportedly preserved in some patients with ADI [8], the desmopressin-independent mechanism of hyponatremia has not yet been studied.
We present a case of impaired thirst sensation and AVP secretion without desmopressin therapy in which hyponatremia developed due to preserved non-osmotic AVP secretion, which was successfully controlled by regulating water intake without desmopressin therapy.
A 53-year-old woman was admitted to our hospital with loss of consciousness. She complained of appetite loss for three days and was found somnolent and incontinent. She had a history of hypopituitarism and mental retardation that developed after measles encephalitis at the age of two years. Her family confirmed that she had been receiving 20 mg hydrocortisone and 75 μg levothyroxine. The patient had no other medications except atorvastatin 5 mg every other day in addition to daily hydrocortisone and levothyroxine. She developed her first episode of hypernatremia at the age of three years. At the time, the patient was diagnosed with hypopituitarism and mental retardation. She had received only hydrocortisone and levothyroxine for hypopituitarism.
Her height and weight were 129 cm and 55 kg, respectively. At admission, physical examination revealed no edema. Laboratory findings are presented in Table 1: serum sodium, 120 mEq/L; urinary sodium, 130 mEq/L; plasma AVP, measured using a radioimmunoassay kit (Yamasa Cat# 80114, RRID: AB_2801274), 33.8 pg/mL. As the patient had a urinary tract infection, the first episode of hyponatremia was suspected to have been caused by adrenal insufficiency. Treatment with 50 mg hydrocortisone and hypertonic saline gradually normalized her serum sodium level. The patient was discharged on day 26 with a serum sodium level of 139 mEq/L (Fig. 1).
| Serum sodium, mEq/L | 120 |
| Serum potassium, mEq/L | 4.2 |
| Serum chloride, mEq/L | 84 |
| Serum calcium, mg/dL | 9.0 |
| Serum creatinine, mg/dL | 0.49 |
| Estimated glomerular filtration rate, mL/min/1.73m2 | 100.1 |
| Plasma glucose, mg/dL | 165 |
| Hemoglobin A1c, % | 8.9 |
| Plasma arginine vasopressin, pg/mL | 33.8 |
| Urinary sodium, mEq/L | 130 |
| Urinary potassium, mEq/L | 13.2 |
| Urinary chloride, mEq/L | 90 |
Serum sodium level (closed circles and lines) fluctuations during treatment. In contrast, plasma AVP levels (open triangles and dotted lines) were primarily low and detectable, except at one point (bottom graph). The patient had free access to drinking water, and the water intake was approximately 500–1,500 mL/day before day 115 (top graph). *Missing values. Abbreviation: AVP, arginine vasopressin. IV, intravenous.
On day 51, the patient revisited our hospital because of loss of consciousness. The serum sodium level was 175 mEq/L, glucose level was 165 mg/dL, urinary sodium level was 218 mEq/L, and plasma AVP level was 2.0 pg/mL. She was admitted, diagnosed with hypernatremia due to a lack of water intake, and administered 5% glucose in water. The patient was discharged on day 59 after her serum sodium level had normalized.
Interestingly, on day 66, the patient was re-admitted for hypernatremia. Laboratory findings were as follows: serum sodium 178 mEq/L, glucose 666 mg/dL, HbA1c 9.5%, ACTH 14.5 pg/mL (reference, 7.2–63.3 pg/mL), cortisol 11.0 μg/dL (reference, 4.5–21.1 μg/dL), GH <0.03 ng/mL (reference, 0.13–9.88 ng/mL), IGF-1 <7 ng/mL (reference, 77–212 ng/mL), prolactin 5.6 ng/mL (reference, 6.1–30.5 ng/mL), LH <0.1 mIU/mL, FSH 0.4 mIU/mL, estradiol 31.0 pg/mL, progesterone <0.1 ng/mL, TSH 2.524 μIU/mL (reference, 0.50–5.00 μIU/mL), and free thyroxine 1.07 ng/dL (reference, 0.90–1.70 ng/dL). Treatment with 5% glucose in water and saline was initiated for hypernatremia and hyperglycemia, and perospirone, a serotonin-dopamine antagonist, was added for aggressive and agitated behaviors on day 68. Intravenous fluid treatment was discontinued on day 75, and her serum sodium level gradually decreased to 124 mEq/L with the plasma AVP level of 1.3 pg/mL on day 85 (the 2nd hyponatremia episode). Subsequently, the patient was diagnosed with perospirone-induced syndrome of inappropriate antidiuresis (SIAD). After discontinuing perospirone, her serum sodium level gradually increased to 150 mEq/L on day 96. When her serum sodium level increased from 124 to 150 mEq/L, she did not complain of dipsia, her urinary volume was approximately 1 L/day, and her plasma AVP level did not change (Fig. 1).
A hypertonic saline infusion test with continuous infusion of 5% NaCl at 0.05 mL/kg for 120 minutes on day 102 to assess AVP secretion in response to osmolality revealed a lack of dipsia and response in plasma AVP levels, as well as serum copeptin levels that were measured using an automated immunofluorescent assay (ThermoFisher Scientific B.R.A.H.M.S GmbH Cat# 857050N, RRID: AB_3073917) (Table 2). Based on the lack of dipsia and response to AVP secretion in both hypertonic and hypotonic situations, she was suspected to have impaired thirst sensation and AVP secretion response to osmolality. Brain magnetic resonance imaging revealed no mass in the hypothalamus-pituitary area, and measle encephalitis was suggested to be the cause of impaired thirst and AVP regulation.
| Time | Before | 30 min | 60 min | 90 min | 120 min |
|---|---|---|---|---|---|
| Serum sodium, mEq/L | 143 | 149 | 151 | 154 | 156 |
| Serum osmolality, mOsm/L | 297 | 305 | 309 | 316 | 320 |
| Plasma arginine vasopressin, pg/mL | 1.1 | 1.0 | 1.0 | 1.2 | 1.0 |
| Serum copeptin, pmol/L | 4.098 | 3.505 | 2.741 | 3.272 | 3.536 |
After the hypertonic saline infusion test, the serum sodium level spontaneously decreased to 125 mEq/L on day 115 (the 3rd hyponatremic episode), with a plasma AVP level of 1.8 pg/mL. Furthermore, hyponatremia was considered idiopathic SIAD, because hydrocortisone was supplemented, and no drugs, including perospirone or desmopressin, were used. Cerebral salt wasting was ruled out as there was no change in the neurological findings. Regulating water intake using a sliding scale, based on the differences between measured body weight and target weight, normalized serum sodium levels without desmopressin treatment (Fig. 1). For example, if the patient’s body weight exceeded the target weight by 1 kg, she was recommended to consume an additional 500 mL of water. Target weight was determined as the weight at which the sodium level was within the normal range. She was discharged one month after admission, and her sodium levels remained within the normal range for one year after discharge.
The plasma AVP and serum sodium levels observed in the present case are shown in Fig. 2. Plasma AVP levels during hyponatremia are higher than the normal range and those during hypernatremia are lower than the normal range [12]. When the data-point with a value of 33.8 pg/mL was excluded, plasma AVP levels were low, but detectable (1.3 ± 0.4 [mean ± standard deviation] pg/mL), and were not significantly correlated with serum sodium levels (rs = –0.04, p = 0.85).
Except at one point, plasma AVP levels were low and detectable, regardless of serum sodium levels. Abbreviation: AVP, arginine vasopressin.
We encountered a patient with impaired thirst sensation and AVP secretion under thyroid hormone and glucocorticoid supplementation therapies, who repeatedly developed not only hypernatremia but also hyponatremia without desmopressin use. AVP secretion in the case was preserved, increased during adrenal insufficiency due to infection, and was not suppressed during the subsequent two hyponatremia episodes. The hypertonic saline infusion had no effect on AVP and copeptin levels, suggesting that the preserved AVP secretion was regulated independently of osmolality. Indeed, plasma AVP levels were not significantly correlated with serum sodium levels except during infection. Regulating the patient’s drinking water intake using a sliding scale based on body weight effectively prevented the development of dysnatremia without the need for desmopressin.
Our case exhibited impaired thirst sensation and AVP secretion in response to hyperosmolality, which is usually called ADI [5]. However, our case did not meet the strict definition of ADI [13] due to the absence of polyuria. The lack of polyuria was considered to be attributed to the constant AVP secretion independent of osmolality. Similar cases with the osmolality-independent constant release of AVP were described as “adipsic hypernatremia” [14, 15]. Although the destruction of the osmoreceptor nuclei, the subfornical-organ and the organum-vasculosum of lamina terminalis in the hypothalamus [8], can impair thirst sensation and AVP secretion in response to hyperosmolality, it cannot explain the constant release of AVP under both hyperosmotic and hypoosmotic conditions in our case. Although the exact mechanisms remain unclear, the inactivation of inhibitory neurons such as γ-aminobutyric acid neurons, or activation of excitatory neurons such as glutamate neurons that connect to AVP neurons and regulate AVP release [16] might be associated with the constant release of AVP in our case.
We identified two mechanisms contributing to the development of hyponatremia in the case. First, preserved AVP secretion may have caused hyponatremia. Hyponatremia independent of desmopressin use has rarely been reported in patients with ADI—only one case of hyponatremia during rehydration has been previously reported [17]. In the current case, hyponatremia was observed three times without desmopressin use. The first episode was adrenal insufficiency due to infection. The second was drug-induced SIAD, and the third was idiopathic SIAD. The plasma AVP levels observed in the patient were consistently low and detectable (Fig. 2), similar to the slow, constant “leak” pattern of SIAD described by Robertson [18]. Indeed, AVP secretion was preserved in some patients with ADI [8]. Because the regulation of AVP release by changes in osmolality is blunted in ADI [8], the AVP secretion was not suppressed during hypoosmolar conditions, and relative water excess might cause hyponatremia similar to SIAD even in the absence of desmopressin therapy in the case.
Second, glucocorticoid deficiency might stimulate the preserved AVP secretion and contribute to the development of hyponatremia in the case. A high level of plasma AVP was observed in the first episode of hyponatremia, when the patient experienced secondary adrenal insufficiency. Glucocorticoids have been known to suppress AVP secretion [19]. The mechanism of hyponatremia in secondary adrenal insufficiency is explained as impaired free water excretion due to the inappropriately enhanced secretion of AVP [20, 21]. Considering that AVP secretion in our case was regulated independently of osmolality, glucocorticoid deficiency might have increased AVP secretion through osmolality-independent mechanisms. This hypothesis is consistent with a report that some patients with ADI respond normally to non-osmotic AVP release signals [13, 22]. The osmoreceptor nuclei send signals to the magnocellular neurons in the paraventricular nucleus (PVN) and supraoptic nucleus of the hypothalamus for AVP release into the systemic circulation in response to hyperosmolality [8]. On the other hand, AVP is also produced in the parvocellular corticotropin-releasing factor/AVP coexpressing neurons in the PVN, and it is released into the hypothalamic-pituitary circulation to promote ACTH secretion [23]. A recent in vivo study has shown that glucocorticoid deficiency may cause hyponatremia by stimulating AVP production of corticotropin-releasing factor/AVP coexpressing neurons in the PVN [24], while an indirect contribution of magnocellular neurons to the AVP production could not be ruled out. Considering these reports, secondary adrenal insufficiency might have increased AVP production, possibly in the parvocellular neurons in the PVN, independently of osmolality in our case.
These findings suggest the importance of evaluating preserved AVP secretion and the potential for desmopressin-independent management of patients with ADI. Desmopressin may cause iatrogenic hyponatremia through renal water retention [25]. Most patients with ADI receive desmopressin therapy, and hyponatremia is observed in 30% of the patients with ADI [9]. In our case, after careful assessment of the non-osmotic AVP secretion, regulating water intake using a sliding scale [8] without desmopressin therapy successfully maintained serum sodium levels. Although no studies have examined the differences in the risk of hyponatremia with or without desmopressin use in patients with ADI, this desmopressin-independent management may prevent hyponatremia in ADI patients with preserved AVP secretion.
In conclusion, we report a case of impaired thirst sensation and AVP secretion without desmopressin treatment in which hyponatremia developed through preserved non-osmotic AVP secretion, possibly stimulated by glucocorticoid deficiency. Non-osmotic AVP secretion, which is preserved in some patients with ADI, may play an important role in the development of hyponatremia, suggesting the usefulness of desmopressin-independent management.
The authors thank Editage (http://www.editage.jp/) for the English language editing.
The patient and the patient’s legal guardian provided written informed consent for publication of the results.
This research did not receive any specific grant from any funding agency in the public, commercial or not-for-profit sector.
The authors have no conflict of interest to disclose.
