Overcorrection of Hyponatremia: A Complicated Story

Abstract

Overcorrection is a frequently encountered issue in the treatment of severe hyponatremia, which is defined as a serum sodium concentration below 120–125 mmol/L in symptomatic patients. Severe hyponatremic symptoms, including persistent vomiting, cardiorespiratory arrest, seizures, and reduced consciousness, clearly indicate the use of hypertonic (e.g., 3%) saline infusion to immediately elevate serum sodium levels. In the chronic phase, however, overcorrection is a major concern because an overly rapid increase in serum sodium concentration is an important cause of osmotic demyelination syndrome (ODS). Although risk groups for ODS are well known, saline infusion and water diuresis are 2 common factors underlying the overcorrection of hyponatremia. We propose that isotonic saline initially be infused (1 mL/kg/h) when encountering hyponatremic patients with moderately severe symptoms, such as confusion, headache, or nausea. Although this empirical therapy may not be consistent with current guidelines, it may reduce the risk of overcorrection and be effective in cases with unrecognized hypovolemia or waning vasopressin hyperactivity. We believe that hypertonic saline should be reserved for patients with severely symptomatic hyponatremia. In practice, urine sodium and osmolality should then be measured, along with follow-up serum sodium levels. Subsequent intravenous fluids can be selected according to the response of serum sodium levels and the differential diagnosis of hyponatremia. Because water diuresis is a critical contributor to overcorrection, urine output and urine sodium, potassium, and osmolality should be monitored during fluid therapy. Appropriate use of 5% dextrose in water and/or desmopressin may be necessary to re-lower serum sodium levels.

TREATMENT OF SEVERE HYPONATREMIA

Hyponatremia is a low serum sodium concentration resulting from an excess of extracellular water relative to sodium. Based on changes in sodium balance, hyponatremia is classified as hypovolemic, euvolemic, or hypervolemic. This classification is important because treatment principles are different depending on whether to restore sodium balance.

Irrespective of the sodium status, however, acute treatment of the severe hyponatremia aims to elevate the serum Na⁺ concentration. Usually, the hyponatremia is considered severe when the serum Na⁺ level is below 120–125 mmol/L and the patient is symptomatic [1,2]. For acute treatment of severe hyponatremia, guidelines recommend 3% NaCl solution using either bolus (100–150 mL over 10–20 minutes) therapy or continuous infusion (0.5–1 mL/kg/h) until a 5 mmol/L elevation of serum Na⁺ over 1–4 hours is reached [2]. The SALSA randomized clinical trial reported that rapid intermittent administration of hypertonic saline was preferred because bolus therapy had a lower incidence of therapeutic relowering treatment and tended to have a better efficacy in achieving serum Na⁺ within 1 h than continuous infusion [3].

Here, we present 2 cases with severe hyponatremia and describe their courses after an intravenous bolus of 100 mL 3% NaCl. With a literature review, strategies to avoid overcorrection of hyponatremia are discussed.

CASE VIGNETTE

Case 1

A 71-year-old female visited the emergency room (ER) due to confusion. She was previously diagnosed to have senile dementia and was taking an uncertain antihypertensive medication. Her oral intake was poor during the past week, but drinking water was not limited. The initial laboratory results were as follows: serum Na⁺ 110 mmol/L, K⁺ 2.4 mmol/L, Cl⁻ 77 mmol/L, total CO₂ 12.5 mmol/L and creatinine 0.56 mg/dL, and blood urea nitrogen (BUN) 3.3 mg/dL. Her brain imaging (computed tomography [CT] and magnetic resonance [MR]) revealed mild diffuse brain atrophy. After 100 mL 3% saline was infused over 20 min, 0.9% saline containing potassium chloride (KCl) was continued. The follow-up labs taken 5 hours after the initial sampling were as follows: serum osmolality 236 mOsm/kg H₂O, Na⁺ 116 mmol/L, K⁺ 2.6 mmol/L, Cl⁻ 85 mmol/L, total CO₂ 18.0 mmol/L and uric acid 3.4 mg/dL, and urine osmolality 151 mOsm/kg H₂O, Na⁺ 40 mmol/L, K⁺ 14 mmol/L, Cl⁻ 39 mmol/L and creatinine 10.6 mg/dL. In the meantime, urine output was 1,600 mL.

Follow-up laboratory data were reported nine hours later when the patient was admitted to a ward: serum Na⁺ 123 mmol/L, K⁺ 2.2 mmol/L, Cl⁻ 90 mmol/L, and total CO₂ 19.0 mmol/L. She was still confused or stuporous despite an increase in serum Na⁺ level. The tonicity of intravenous fluid was lowered to half-isotonic. However, the follow-up serum Na⁺ taken eight hours later was 129 mmol/L. Therefore, the increase in serum Na⁺ was 19 mmol/L over 22 hours. To lower the serum Na⁺ level, the half-isotonic fluid was switched to 5% dextrose in water (40 mL/h). The daily intake/output was assumed 2,400/5,050 mL, and the follow-up urine data were as follows: osmolality 104 mOsm/kg H₂O, Na⁺ 12 mmol/L, K⁺ 9.5 mmol/L, Cl⁻ 16 mmol/L and creatinine 17.6 mg/dL.

The consciousness of the patient was improved, and verbal communication became possible. Nonetheless, the follow-up serum Na⁺ was not lowered and was 132 mmol/L nine hours later. The duty physician decided to use desmopressin, and the serum Na⁺ was lowered to 127 mmol/L in 10 hours.

Case 2

Another 71-year-old female visited the ER due to a seizure. The seizure movement subsided when she arrived at the ER. However, she was confused and nauseated. The initial laboratory results were as follows: serum Na⁺ 116 mmol/L, K⁺ 3.5 mmol/L, Cl⁻ 84 mmol/L, total CO₂ 13.3 mmol/L, and creatinine 0.49 mg/dL, and BUN 6.8 mg/dL. Her brain imaging (CT and MR) revealed no remarkable findings. After 100 mL 3% saline was infused over 20 min, 0.9% saline was continued a rate of 60 mL/h.

She became alert when she was admitted to the intensive care unit (ICU). With a Foley catheter inserted, the urine output was initially 1,100 mL and subsequently 500 mL over one hour. The follow-up serum data taken six hours after the initial sampling were Na⁺ 126 mmol/L and uric acid 3.0 mg/dL. When another six hours had passed, serum Na⁺ reached 137 mmol/L. Therefore, the increase in serum Na⁺ was 21 mmol/L over 12 hours. The isotonic saline was switched into half-isotonic (60 mL/h), and her consciousness became clear.

Eight hours later, however, she complained of headache and nausea. The serum Na⁺ level was stationary (138 mmol/L). Although the half-saline was replaced by 5% dextrose in water (80 mL/h), the serum Na⁺ was not lowered. The daily urine output exceeded intake, and urine osmolality fluctuated between 562 and 150 mOsm/kg H₂O. Only after the use of desmopressin, the serum Na⁺ was lowered to 129 mmol/L over 6 hours. Her headache and nausea were relieved along with the reduction of urine output.

MORBIDITY AND MORTALITY OF HYPONATREMIA

In addition to the level of plasma Na⁺, the rapidity of hyponatremia development is critical in determining the severity of hyponatremia. Therefore, hyponatremia is classified as acute or chronic depending on whether it develops within 48 hours or not. Acute hyponatremia is usually symptomatic and requires urgent therapy because patients with acute hyponatremia have a risk of cerebral edema [4]. The symptoms of hyponatremia can be grouped as moderately severe and severe; whereas moderately severe symptoms are headache, nausea, and confusion, severe symptoms include vomiting, cardiorespiratory distress, abnormal and deep somnolence, seizures, and coma [1]. Our cases seemed to have moderately severe symptoms when they arrived at the ER, and their hyponatremia may be considered chronic because of unclear onset. Although the seizure had ceased upon arrival at the ER, case 2 was eligible for treatment with 3% hypertonic saline because of the severity of hyponatremia. This case suggests that avoiding overcorrection of hyponatremia may be difficult when using the traditional approach.

After the acute phase, chronic changes in extracellular tonicity occur for long-term adjustment through the efflux of organic osmolytes [5]. This compensation should be considered when chronic hyponatremia is corrected because rapid correction of extracellular tonicity can lead to a neurological disorder called osmotic demyelination.

Gait disturbances and cognitive impairment are frequently noted in patients with chronic hyponatremia due to altered neuronal activity in brain [6]. Chronic hyponatremia also induces calcium release from bone cells and lead to bone demineralization, osteoporosis and an increased risk of fractures [7].

Hyponatraemia, even when mild, is associated with increased mortality [8]. Acute hyponatremia can be life-threatening due to neurological deficits, and chronic hyponatremia is also associated with increased all-cause and cardiovascular mortality [9]. Recently, hyponatraemia was found to promote vascular calcification through Rac1-Akt pathway activation [10]. Although patients with hyponatremia usually have comorbid conditions, hyponatremia is consistently identified as an independent contributor to increased mortality [11].

OVERCORRECTION OF HYPONATREMIA AND OSMOTIC DEMYELINATION SYNDROME (ODS)

Temporary or permanent brain damage associated with rapid correction of chronic hyponatremia clinically presents 1 to 7 days after treatment [12]. This delayed onset of neurologic symptoms has been called ODS, and pontine and extrapontine myelinolysis can be demonstrated by magnetic resonance images or autopsy in severely affected patients. Brain astrocyte death and disruption of the blood–brain barrier precedes the development of ODS [13].

The first report of central pontine myelinolysis, a major component of ODS, was documented by autopsy in two middle aged women with severe, diuretic-induced hyponatremia (serum Na⁺, 96 and 100 mmol/L) and a history of protracted vomiting and drowsiness, who deteriorated neurologically after treatment with 3% saline increased their serum sodium concentration by 25 and 32 mmol/L over 48 hours [14]. Subsequent clinical and experimental studies confirmed that overly rapid correction of serum sodium can be associated with ODS [12]. The risk of ODS is increased in patients with liver disease, alcohol use, hypokalemia, severe hyponatremia, and malnutrition [15]. Our cases of severe hyponatremia were malnourished and had low levels of serum potassium.

However, the incidence of ODS is very low even in severe hyponatremia, whereas overcorrection is frequently associated with the treatment of hyponatremia. A bolus of hypertonic saline has the advantage of an immediate effect with little risk of overcorrection [16], but in the SALSA clinical trial, cumulative overcorrection occurred in 15 of 87 (17.2%) patients who were treated with bolus therapy [3].

In a large, a multicenter cohort study of 22,858 hospital admissions of adult patients whose serum Na⁺ < 130 mmol/L, rapid correction of serum Na⁺ (>8 mmol/L over 24 hours) was frequent (17.7%), but ODS was rare (0.05%) [17]. ODS occurred in 3 of 1,000 who had an initial serum sodium level of less than 120 mmol/L, and the incidence of ODS was 0.14% even when the analysis was limited to the patients with the rapid correction of serum Na⁺. Seven (58%) patients who developed ODS did not have rapid correction of serum Na⁺ [17], compatible with the notion that ODS is of multifactorial nature.

Moreover, ODS is infrequent even when correction rates exceed current guidelines, and faster correction may actually lower in-hospital mortality. In a recent systematic review and meta-analysis involving 16 studies and 11,811 patients, slow (< 8 or 6–10 mmol/L over 24 hours) and very slow (< 4–6 mmol/L over 24 hours) correction of severe hyponatremia (serum sodium < 120 or < 125 mmol/L plus severe symptoms) were associated with an increased risk of mortality and hospital length of stay compared to rapid correction. Notably, rapid correction was not associated with a statistically significant increase in the risk of ODS [18]. Healthier patients may tolerate faster correction better, while sicker patients may be more prone to both ODS and mortality.

CAUSES FOR OVERCORRECTION OF HYPONATREMIA

Treatment of severe hyponatremia is delicate because of the 2 sequential strategies; the initial correction must be rapid during the first few hours to decrease cerebral edema, and a slow correction < 10 mmol/L over 24 hours should be followed to avoid the development of ODS [19]. The strict goal of serum Na⁺ lowering was proposed as 4 to 8 mmol/L per day for patients at high risk of overcorrection or ODS (Table 1) [15,20,21,22].

Table 1. Risk groups for overcorrection and ODS in hyponatremia.

Overcorrection	ODS
Low solute intake	Serum sodium < 105 mmol/L
Hypovolemia	Hypokalemia
Adrenal insufficiency	Malnutrition
Transient SIADa	Alcoholism
Thiazide diuretics	Advanced liver disease
Prior use of desmopressin	V2R antagonist therapy
Body weight < 60 kg

ODS, osmotic demyelination syndrome; SIAD, syndrome of inappropriate antidiuresis; V2R, vasopressin-2 receptor.

^aTransient SIAD may be caused by nausea, hypoxia, surgery, pain, or stress.

The use of hypertonic saline solution is advantageous for the initial rapid correction, but is prone to later overcorrection. Repeat doses of intravenous bolus therapy would increase the risk of overcorrection. In our cases, however, serum Na⁺ levels were too rapidly corrected despite a single bolus infusion of 100 mL 3% NaCl. In terms of sodium load, the risk of overcorrection can be lessened by the use of isotonic saline. In a Japanese single-center retrospective study on the patients with severe hyponatremia visiting the ER, the lower risk of overcorrection was associated in the 0.9% saline users compared with 3% saline users [23].

In principle, intravenous administration of isotonic saline solution is indicated for treating hypovolemic hyponatremia. Continuous infusion of isotonic saline at a rate of 1 mL/kg/h seems to be safe and effective, aiming an increase in serum Na⁺ between 4–7 mmol/L over the first 24 hours in patients with hypovolemic hyponatremia [24]. In a typical syndrome of inappropriate antidiuresis (SIAD), the serum Na⁺ concentration may actually decrease during the infusion of isotonic saline if the urine Na⁺ plus K⁺ concentration is much higher than 154 mmol/L [12]. Patients given large volumes of isotonic saline may be more likely to overcorrect than patients given low volumes of hypertonic saline [25].

In addition to an excess sodium load to the extracellular fluid, spontaneous water diuresis is an important cause for overcorrection of hyponatremia [26]. Appropriate and inappropriate stimuli of vasopressin release can be attenuated or ceased by the administration of saline, volume repletion, discontinuing hyponatremia-inducing drugs or stress relief, leading to hypoosmotic polyuria. As shown in our cases, water diuresis, defined as urine flow ≥ 2 mL/kg/h and urine Na⁺ plus K⁺ concentration ≤ 50 mmol/L, is common during the treatment for severe hyponatremia and typically occurs within the first 24 hours [27]. Polyuria was remarkable in our cases, and their urine chemistry was compatible with water diuresis. Therefore, physicians should pay attention not only to changes in serum Na⁺ levels but also to the diuretic response. Electronic alert systems may provide rapid notifications during the treatment of hyponatremia, enabling timely intervention by physicians and nursing staff.

Hypertonic saline induces more water diuresis than isotonic saline during the treatment of hyponatremia because hypertonicity acts as a potent osmotic stimulus to pull water into the intravascular space, facilitating the excretion of free water in excess of sodium [20,28]. We could not avoid overcorrection in our cases although isotonic saline infusion was followed by a single bolus of 3% hypertonic saline. Close monitoring of serum Na⁺ is advisable, initially every 1–2 hours and then every 4 hours, particularly if urine output is high (> 150 mL/h or 2 mL/kg/h) [19,28]. Overcorrection might have been lessened if we had monitored urine output hourly from the beginning of treatment.

Intracellular free water uptake is another cause for overcorrection of hyponatremia [26]. Some patients with hyponatremia accompanies K⁺ deficiency (e.g., thiazide-induced hyponatremia), and K⁺ administration can increase serum Na⁺ concentration primarily through osmotic water shifts; as potassium is taken up into cells, it creates an osmotic gradient that draws water out of the extracellular space. Additionally, to maintain electrical neutrality, intracellular K⁺ uptake can lead to an equivalent movement of Na⁺ out of cells into the extracellular fluid. According to the Edelman equation, serum Na⁺ concentration is determined by the total amount of exchangeable Na⁺ and K⁺ divided by total body water [29]. We had to intravenously supplement potassium to correct hypokalemia, and thiazide-induced hyponatremia might have underlain case 1 [30]. To treat thiazide-induced hyponatremia, the intravenous administration of 1–2 L of isotonic saline containing 40 mmol KCl over 24 hours may be optimal in conjunction with stopping the thiazides [19].

The Adrogué-Madias formula is typical in aiding fluid prescription to correct hyponatremia [31]. However, it frequently underestimates the increase in serum Na⁺ after hypertonic saline infusion because of ignoring renal water losses and variable active water excretion, such as sudden suppression of vasopressin release (e.g., in syndrome of inappropriate antidiureis or volume depletion) [32].

Finally, we need to be more cautious about overcorrection of hyponatremia when the patients are old females like our cases with a small body size. According to a single-center, retrospective study, bolus infusion of 100 or 150 mL 3% NaCl may expose patients with low (≤ 60 kg) and high (≥ 100 kg) body weight to more overcorrection and undercorrection, respectively. Table 2 summarizes the major causes for overcorrection of hyponatremia.

Table 2. Major causes for overcorrection of hyponatremia.

Major causes
Sodium load to extracellular fluid
Spontaneous water diuresis
Potassium administration
Application of the Adrogué-Madias formula
Small body size

USE OF DESMOPRESSIN TO PREVENT OVERCORRECTION IN SEVERE HYPONATREMIA

Hyponatremia is often spontaneously improved by water diuresis when the causative factors such as hypovolemia, medications, and pain or stress, are reversed. We saw in our cases that spontaneous excretion of dilute urine raised the serum Na⁺ by > 2 mmol/L/h, exceeding therapeutic limits within a few hours. Three regimens of therapeutic re-lowering of serum Na⁺ are feasible: intravenous 5% dextrose in water, parenteral desmopressin, or both. We used less amount of 5% dextrose solution compared with the guidelines (3 mL/kg/h or 10 mL/kg over 1 hour) [1,15]. Therefore, we had to use desmoressin for rescue.

Three different strategies of demopressin administration for prevention or treatment of overcorrection in hyponatremia [20,22]: 1) proactive, in anticipation of water diuresis; 2) reactive, in response to water diuresis and impending overcorrection; and 3) rescue, after overcorrection had already occurred. The proactive strategy starts immediately when the patients whose serum Na⁺ is less than 115 mmol/L have risk factors for overcorrection or ODS (Table 1). Along with hypertonic saline bolus or continuous infusion, desmopressin 2–4 µg is administered intravenously or subcutaneously every 6–8 hours. The reactive strategy starts when a high urine flow (> 1 mL/kg/h or > 100 mL/h) is noted in patients whose serum Na⁺ increase was ≤ 6 mmol/L in a 24-hour period. It also uses intravenous or subcutaneous desmopressin 2–4 µg every 6–8 hours. The rescue strategy is applied when overcorrection (serum Na⁺ elevation ≥ 8 mmol/L in a 24-hour period) has occurred. Along with 5% dextrose in water (3 mL/kg/h), desmopressin 2–4 µg is administered intravenously or subcutaneously every 6–8 hours.

In our cases, the rescue desmopressin was used. Earlier applications of desmopressin might have prevented or relieved overcorrection of hyponatremia. Although favored by some experts, the proactive strategy may prolong the hospital stay [20]. We think the reactive strategy should be optimal [33], and vigilant urine output monitoring particularly during the first 24–48 hours is necessary.

CONCLUSION

Severe hyponatremia is vulnerable to overcorrection during saline infusion, and spontaneous water diuresis is a critical mechanism leading to overcorrection. The greater the amount of sodium administered to the plasma, the higher the risk of overcorrection. Therefore, hypertonic saline should be reserved for patients with severely symptomatic hyponatremia. We prefer the use of isotonic saline rather than hypertonic saline in patients with moderately severe symptoms who present to the ER. Unrecognized hypovolemia may respond to isotonic saline infusion, and some patients with transient SIAD may also improve with isotonic saline. The decision to switch to hypertonic saline can be made after evaluating follow-up changes in serum Na⁺ and urine parameters, including hourly urine flow, Na⁺, K⁺, and osmolality (Fig. 1). During the first 24–48 hours of treatment, vigilant monitoring of urine output is essential to avoid overcorrection of hyponatremia.

Fig. 1. Exemplary therapeutic approach for patients with moderately severe symptomatic hyponatremia who present to the emergency room and are awaiting admission. The initial serum sodium concentration is usually below 125 mmol/L, and patients typically present with confusion, headache, or nausea, but without persistent vomiting, cardiorespiratory arrest, seizures, or reduced consciousness. This scheme is intended to avoid overcorrection of hyponatremia.

D5W, 5% dextrose in water; U/O, urine output.