Simultaneous Occurrence of Hyponatremia and Hypokalemia in a Patient with Herpes Zoster: A Case Report with a Review of the Literature

Abstract

We herein report a patient with herpes zoster (HZ), severe hyponatremia, and hypokalemia. Syndrome of inappropriate antidiuresis (SIAD) leads to euvolemic hyponatremia and hypoosmotic plasma due to inadequate diuresis. Hyponatremia in the current patient was caused by SIAD and associated with HZ of the trigeminal facial nerve (V1). The patient also had hypokalemia, with excessive urinary potassium excretion and elevated cortisol levels. Hypokalemia is caused by hypercortisolemia, which is stimulated by HZ pain. Adequate treatment for HZ and comprehensive pain control play pivotal roles in improving SIAD, cortisol hypersecretion, and the subsequent electrolyte abnormalities.

Introduction

Hyponatremia is the most common electrolyte disorder in clinical practice (1), and the syndrome of inappropriate antidiuresis (SIAD) constitutes one of the most common causes of hyponatremia (2-4). SIAD is characterized by dilutional hyponatremia (＜135 mEq/L) with inappropriate arginine vasopressin (AVP) activity relative to plasma osmolality (＜275 mOsm/kg); inappropriate urine osmolality (＞100 mOsm/kg H₂O); elevated urine sodium concentration (＞20-30 mEq/L); absence of edema or volume depletion; and normal renal, adrenal, and thyroid functions (5). SIAD is caused by several factors and diseases, including malignancy, central nervous system disorders, respiratory infections, pain, and idiopathy.

Herpes zoster (HZ) is caused by reactivation of endogenous latent varicella-zoster virus (VZV) within the sensory nerve ganglia after chickenpox infection during childhood (6). The ophthalmic division of the trigeminal facial nerve (V1) is the most common cranial nerve affected by VZV, accounting for 10%-25% of all HZ cases (7). SIAD rarely occurs in cases of HZ (8).

We herein report a patient with hyponatremia caused by SIAD associated with HZ in the V1 region. HZ-induced pain also contributed to hypercortisolemia and hypokalemia in this patient.

Case Report

A 59-year-old man developed HZ in the V1 region of the trigeminal nerve and was treated with famciclovir (750 mg daily) for 7 days at a dermatology clinic. He had an 8-year history of diabetes mellitus with a hemoglobin A1c level of approximately 6%, which was managed with dietary therapy only. He exhibited diabetic neuropathy, characterized by decreased vibration sensation and loss of the Achilles tendon reflex.

Fourteen days after the onset of HZ (seven days after famciclovir treatment), the patient visited a local internal medicine clinic with cough and nausea and was treated with an antitussive agent. Four days later, the patient presented to the same clinic complaining of dizziness and fatigue. He was found to have severe hyponatremia (111 mEq/L) and hypokalemia (3.2 mEq/L); therefore, he was sent to our hospital for a further evaluation. There was no history of analgesics or any Chinese herbal medicines, health products, or supplements containing licorice before admission.

On admission, he was alert and conscious, although he reported intense pain in the left forehead area due to HZ. The affected area exhibited crusted skin with blisters. There were no edematous findings suggestive of cardiac or renal failure or nephrotic syndrome. No cushingoid features were observed. The patient showed no signs of dehydration or hypovolemia. His blood pressure was 138/87 mmHg and heart rate was 76 beats/min. The patient reported no current medication.

Laboratory findings on admission revealed hyponatremia (110 mEq/L), low plasma osmolality (224 mOsm/kg H₂O), and hypokalemia (2.9 mEq/L) (Table 1). Despite the low blood electrolyte levels, urinary sodium (101 mEq/L) and potassium (53 mEq/L) concentrations were elevated. An arterial blood gas analysis (pH, 7.525) indicated metabolic alkalosis. The plasma AVP concentration was increased (4.7 pg/mL), despite the low plasma osmolality and high urinary osmolality. Based on these findings, the patient was diagnosed with SIAD (5). Plasma levels of adrenocorticotropic hormone (ACTH) and cortisol at 9:00 AM were 22.4 pg/mL and 24.9 μg/dL, respectively. The thyroid function was normal (Table 1). Chest radiography and computed tomography (CT) revealed no abnormal lesions, such as pneumonia or lung tumors. Blood levels of VZV immunoglobulin (Ig) G (＞128) and IgM (6.8) antibodies were elevated, although the VZV IgM (0.24) levels in the cerebrospinal fluid were normal. Brain magnetic resonance imaging (MRI) revealed no abnormal findings, including meningitis. These findings indicate that VZV was reactivated.

Table 1.

Laboratory Data on Admission and 3 Months after Discharge.

Blood biochemistry	On admission	At 3 months	(Reference range)
Osmolality (mOsm/kg H2O)	224	269	(275-295)
Na (mEq/L)	110	139	(135-142)
K (mEq/L)	2.9	3.8	(3.5-5.0)
Cl (mEq/L)	73	96	(96-108)
Uric acid (mg/dL)	1.3	5.3	(3.7-7.8)
Glucose (mg/dL)	183	137	(70-109)
BUN (mg/dL)	12	18	(8.0-20)
Creatinine (mg/dL)	0.45	0.66	(0.65-1.07)
HbA1c (%)	6.5	6.0	(4.6-5.8)
Hormones
TSH (μU/mL)	1.01	-	(0.5-5.0)
Free T3 (pg/mL)	2.27	-	(2.3-4.0)
Free T4 (ng/dL)	1.48	-	(0.9-1.7)
ACTH (pg/mL)	22.4	43.4	(7.2-63.3)
Cortisol (μg/dL)	24.9	14.3	(6.2-19.4)
AVP (pg/mL)	4.7	0.5	(0.0-4.2)
PAC (ng/dL)	12.4	22.8	(3.6-24)
PRA (ng/mL/h)	1.2	2.3	(0.3-2.9)
Urinary test
Osmolality (mOsm/kg H2O)	567	606	-
Na (mEq/L)	101	25	-
K (mEq/L)	53	31	-
Cl (mEq/L)	83	15	-
Free cortisol (μg/day)	148-101-73.3	-	(11.2-80.3)
Aldosterone (μg/day)	12.7-11.4-12.9	-	(3.6-24)
Arterial BGA
pH	7.525	-	(7.37-7.44)
PaO2 (mmHg)	107	-	(65-105)
PaCO2 (mmHg)	30	-	(34-44)
HCO3- (mmol/L)	24.7	-	(22-26)

PRA: plasma renin activity, TSH: thyroid stimulating hormone, ACTH: adrenocorticotropic hormone, PAC: plasma aldosterone concentration, PRA: plasma renin activity, BGA: blood gas analysis, BUN: blood urea nitrogen, HbA1c: hemoglobin A1c

Endocrinological findings revealed normal plasma aldosterone, renin activity, and urinary 24-h aldosterone concentration, although 24-h urinary free cortisol levels were elevated (148 μg/day) (Table 1). The plasma cortisol level (5.3 μg/dL) at 11:00 PM was ＞5.0 μg/dL. In the screening tests for subclinical Cushing's disease and syndrome, the 0.5- and 1-mg dexamethasone suppression tests revealed that plasma cortisol levels were not suppressed (Table 2). Whole-body CT and three-tesla pituitary MRI did not reveal abnormalities, allowing for the exclusion of ACTH-producing pituitary tumors or ectopic ACTH syndrome.

Table 2.

Endocrinological Examination Human CRH (100 µg) or DDAVP (4 μg) was Administered Intravenously. Dexamethasone (Dex) was Administered Orally.

	23:00	Dex (0.5 mg)	Dex (1.0 mg)	Dex (8.0 mg)
ACTH (pg/mL)	10.4	24.1	16.1	<0.1
Cortisol (µg/dL)	5.3	8.5	3.07	1.29
Salivary free cortisol (µg/dL)	0.23
DDAVP test	0’	15’	30’	60’	90’
ACTH (pg/mL)	17.6	35.5	32.8	21.9	12.2
Cortisol (µg/dL)	6.9	13.9	13.8	11.1	8.3
CRH test	0’	30’	60’	90’
ACTH (pg/mL)	30.7	56.4	48.0	31.6
Cortisol (µg/dL)	14.7	21.9	19.2	15.1

CRH: corticotropin-releasing hormone, DDAVP: l-deamino-8-D-arginine vasopressin, ACTH: adrenocorticotropic hormone

Administration of l-deamino-8-D-arginine vasopressin (DDAVP, 4 μg) and corticotropin-releasing hormone (100 μg) resulted in peak values of ACTH (35.5 pg/mL and 56.4 pg/mL, respectively) more than 1.5 times the baseline level (Table 2). Fluid restriction of 1,000 mL/day with a salt intake of 9 g/day was initiated on admission (Figure). Serum sodium and potassium levels at 5:00 PM on the day of admission were 103 mEq/L and 2.7 mEq/L, respectively. One week after admission, potassium L-aspartate (900 mg/day) was administered. During hospitalization, the patient's fluid intake was restricted to 500 mL/day, and his salt intake and potassium L-aspartate replacement dose were changed to 12 g/day and 1,800 mg/day, respectively. Carbamazepine (100 mg/day) was initiated on hospital day 6 and switched to pregabalin (75 mg/day) on hospital day 13.

Figure.

Changes in serum sodium and potassium levels. The black arrowhead indicates the onset of herpes zoster.

The patient was discharged on hospital day 35, with a serum sodium concentration of 129 mEq/L and a serum potassium concentration of 3.8 mEq/L. The patient's AVP level reduced to 0.5 pg/mL. Mirtazapine (15 mg/day) and pregabalin (75 mg/day) were administered for post-herpetic neuralgia (PHN). One month after discharge, oral sodium and potassium supplementation and fluid restriction were discontinued. Three months after discharge, the PHN in the forehead area disappeared. His plasma ACTH and cortisol levels at 8:00 AM improved to 43.4 pg/mL and 14.3 μg/dL, respectively, and the serum sodium (139 mEq/L) and potassium (3.8 mEq/L) levels normalized (Table 1 and Figure).

Discussion

The laboratory findings of this patient corresponded to SIAD, which may have been correlated with HZ, based on the clinical trajectory from the onset of HZ to the appearance of hyponatremia. SIAD associated with localized HZ in the V1 region of the trigeminal nerve has been reported in eight patients, including the current patient (Table 3) (9-14). In the patient presented in this report, diabetes mellitus may have contributed to an increased risk of developing HZ (15). The underlying mechanism of SIAD associated with HZ remains unclear; however, VZV induces AVP secretion from the paraventricular nucleus in the hypothalamus (PVH) and supraoptic nucleus (SON) via pathways involving the nucleus of the solitary tract or ventrolateral medulla in the brainstem (13). In addition, a study conducted in rats showed that direct electrical stimulation of the supraorbital nerve, a branch of the ophthalmic nerve, increased plasma AVP concentrations (16). Consequently, VZV reactivation in the dorsal root ganglia may directly and/or indirectly stimulate AVP secretions.

Table 3.

Previous Cases of Syndrome of Inappropriate Antidiuresis with Herpes Zoster in a Trigeminal Nerve Lesion.

Case report	Age/Sex	Dermatome	Latency (days)	Na (mEq/L)	Consciousness	Complications
(9)	78/M	V1	n.a	108	Altered	Death
(10)	77/M	V1	5	115	Altered	-
(11)	71/F	V1	4	120	Preserved	-
(12)	76/F	V1	15	112	Altered	-
(13)	58/M	V1	4	114	Preserved	PHN
(13)	38/M	V1	7	116	Preserved	PHN
(14)	87/F	V3	3	114	Preserved	-
Present report	58/M	V1	14	110	Preserved	PHN, hypokalemia

n.a: not available, PHN: postherpetic neuralgia, F: female, M: male

Severe pain may also trigger nonosmotic AVP secretion. VZV-induced pain is categorized into zoster-associated pain (ZAP) and PHN. ZAP includes clinical prodromal symptoms, such as pain occurring two to four days prior to rash onset, pain concurrent with the rash, and pain persisting after rash resolution. In contrast, PHN is a spontaneous pain and allodynia that persists for 90 days or longer following resolution of the skin eruption (17). AVP is produced in two distinct populations of neuroendocrine neurons in the PVH of the human hypothalamus: magnocellular AVP neurons and parvocellular corticotropin-releasing factor (CRF)/AVP co-expressing neurons (18). Chronic pain increases AVP secretion in the PVH of rodents, in addition to chronic restraint stress and glucocorticoid deficiency (19,20). Baba et al. reported that neuropathic pain increases the mRNA levels of AVP not only in the magnocellular PVH and SON but also in the parvocellular PVH (19). AVP in the parvocellular PVH is involved in the impairment of water diuresis and hyponatremia via regulation of renal aquaporin-2 protein levels and localization in epithelial cells of the renal collecting ducts in glucocorticoid deficiency (21). In the current patient, analgesia achieved using carbamazepine, pregabalin, and mirtazapine was effective in the treatment of SIAD. Although these drugs have also been reported to cause SIAD (22-24), they were not deemed the causative agents of SIAD in this case because hyponatremia occurred prior to the administration of these drugs. Taken together, HZ-induced neurogenic pain may have directly or indirectly stimulated AVP secretion in an osmolality-independent manner; therefore, pain control may have suppressed inappropriate AVP secretion from both parvocellular and magnocellular neurons in the PVH and SON, restored water diuresis, and resolved dilutional hyponatremia.

The patient also had hypokalemia. Bassi et al. reported a case of SIAD complicated by hypokalemia in HZ, although blood renin activity and aldosterone levels were not reported (8). Pain-induced stress is reflected in cortisol levels but not aldosterone levels (25). The current patient did not exhibit increased plasma or urinary aldosterone levels, although his 24-h urinary free cortisol level was elevated (Table 1). A plasma cortisol level at 11:00 PM suggested excess cortisol production (26). The 0.5- and 1-mg dexamethasone suppression tests revealed excess cortisol secretion (27) and autonomous cortisol secretion when the serum cortisol level cutoff value ≤1.8 μg/dL was used (28,29). In contrast, a higher dose (8 mg) of dexamethasone completely suppressed ACTH and cortisol secretion. These findings suggest ACTH-dependent autonomous cortisol secretion but not tumorigenic cortisol production. Together, these data correspond to the excess cortisol production caused by pain stress. Excess cortisol can bind to mineralocorticoid receptors (MR) in the renal tubule, as the specificity of MR for aldosterone conferred by 11β-hydroxysteroid dehydrogenase type 2 is compromised in the presence of excess cortisol (30). Overall, severe pain in the forehead induces excessive cortisol production, leading to increased urinary potassium excretion and hypokalemia.

The standard therapy for HZ includes acyclovir, valacyclovir, and brivudine (6). Although there is insufficient evidence to conclusively demonstrate that antiviral medications reduce the risk of PHN, famciclovir and valacyclovir may shorten the duration of PHN (31). Famciclovir was promptly administered following the onset of HZ in the present patient, relieving PHN for more than three months. Fluid restriction and oral electrolyte replacement also effectively improved hyponatremia in this patient (5,32). Carbamazepine, pregabalin, and mirtazapine were administered for pain control. As reported previously (13), pain management may help alleviate the hypersecretion of both AVP and ACTH/cortisol, thus improving both hyponatremia and hypokalemia in the current patient.

In conclusion, we reported a patient with hyponatremia caused by SIAD associated with HZ in the V1 region of the trigeminal nerve. The patient also had hypokalemia. This report highlights the potential involvement of inappropriate AVP synthesis in both parvocellular and magnocellular neurons in the PVH and SON. Hyponatremia and hypokalemia should be carefully monitored as complications of HZ. In addition, pain management mitigates not only the excessive secretion of AVP but also the activation of the HPA axis, resulting in improved hypokalemia. Adequate antiviral therapy and neurogenic pain control facilitated normalization of SIAD, cortisol hypersecretion, and subsequent electrolyte abnormalities in this case.

The authors state that they have no Conflict of Interest (COI).