Asymptomatic elevation of creatine kinase in patients with hyponatremia

Abstract

Elevated creatine kinase (hyper-CKemia) has been observed in small number of patients with hyponatremia. This study evaluated the features and outcomes of patients admitted with hyponatremia complicated by hyper-CKemia. Patients admitted with hyponatremia and concurrently found to have elevated creatine kinase (CK) of above 375 IU/L (male) or 225 IU/L (female), over a 5-year period were retrospectively reviewed. Those with myocardial injury (elevated CK-MB isoenzyme [CK-MB/CK percentage of >2.5%] or Troponin T [>0.02 μg/L]), traumatic or ischemic muscle damage, primary myopathic disorder, seizures prior to CK measurement or those taking medications which can cause myopathy, were excluded. Thirty-two patients with hyponatremia and hyper-CKemia were identified. All patients had no muscular symptoms or weakness. The commonest cause of hyponatremia in this cohort was related to diuretics (50%). The mean sodium level on presentation was 116.0 ± 6.9 mmol/L and the median peak CK was 895.5 (interquartile range: 610.8–1691.8) IU/L. Six (18%) patients developed acute kidney injury (AKI). The length of hospital admission of the entire cohort was 8.0 ± 5.8 days. Patients with hyper-CKemia in the setting of diuretic-associated hyponatremia were older and had longer hospital length of stay compared with primary-polydipsia-associated. Asymptomatic hyper-CKemia is an uncommon association with hyponatremia of various etiologies. Hyponatremia-associated hyper-CKemia can be complicated by AKI.

• Creatine kinase
• diuretic
• hyper-CKemia
• hyponatremia
• polydipsia

Introduction

Creatine kinase (CK) is a catalytic enzyme that combines creatine and adenosine triphosphate to form phosphocreatine and adenosine diphosphate. Elevated CK indicates myocardial damage (MB fraction is elevated) or neuromuscular disorders. Asymptomatic elevation of serum CK is often referred to as asymptomatic hyper-CKemia. The European Federation of Neurological Society guidelines defines hyper-CKemia as elevations of CK that are at least 1.5 times the upper limit of normal but acknowledging that this upper limit needs to be specified.1 A precise definition and clinical implications of hyper-CKemia is still a contentious subject.

Hyper-CKemia from non-traumatic muscular injury occurs with the use of some drugs or in several medical conditions such as infection, alcoholism and electrolyte disorders.2 Among electrolyte disorders, hypokalemia and hypophosphatasemia are known to cause myocyte destruction.3,4 However, less is known of hyponatremia as a cause of skeletal muscle damage.5 In past reports, the cause of hyponatremia associated with elevated CK has predominantly been caused by primary polydipsia.6 However, several cases and cohort studies have reported other causes of hyponatremia associated with hyper-CKemia or myopathy including post-prostate surgery, acute adrenal insufficiency and syndrome of inappropriate anti-diuretic hormone (SIADH) secretion.7–10

Severe skeletal muscle damage can lead to detrimental complications such as acute kidney injury (AKI) in 7–10% of cases, which can occur in both traumatic and non-traumatic causes.11,12 Risk for developing AKI is higher when the CK level is above 15 000 IU/L but can occur at lower CK values when coexisting conditions such as sepsis or dehydration are present.13 Long-term survival of patients with myopathy and AKI is lower than those without AKI.

Previous reports have not reported the clinical outcomes of patients with hyponatremia and concurrent hyper-CKemia. This study aimed to identify clinical features and outcomes of patients with hyper-CKemia in the setting of hyponatremia without associated seizures.

Methods

This is a retrospective cohort study of patients presenting with concomitant hyponatremia and hyper-CKemia. All patients with hyponatremia (sodium <135 mmol/L) admitted to the General Medicine Department of the Southern Adelaide Local Health Network between 1 January 2006 and 31 December 2010 were identified. This study was approved by the Human Research Ethics Committee of the Southern Adelaide Local Health Network.

The hyponatremia can be of any cause except pseudo-hyponatremia and only those with a rise in CK above 375 IU/L for male and 225 IU/L for female at close proximity to the time of the hyponatremia (within 3 days of diagnosing hyponatremia) were included. Patients with elevated CK MB isoenzyme (CK-MB/CK percentage of >2.5%) or Troponin T (>0.02 μg/L), traumatic or ischemic muscle damage, primary myopathic disorder, seizures prior to CK measurement or those taking medications which can cause elevation in CK, were excluded from the analysis.

Patients’ characteristics, the severity of hyponatremia, etiology of hyponatremia, associated serum biochemical abnormalities, temporal relationship between hyponatremia and hyper-CKemia, clinical features particularly related to symptomatic myopathy, and the clinical progress were evaluated. Outcomes evaluated were acute kidney injury (AKI), intensive care unit (ICU) admission, hospital length of stay and in-hospital mortality. AKI was defined in accordance with the AKI network criteria, which states that if there was a rise in serum creatinine of >26 µmol/L from a pre-existing value (within 12 months) or an increase of 1.5- to 2.0-fold from the admission value had occurred within 48 h of admission.14 Comparison between different etiologies of hyponatremia was made in relation to patient characteristics, degree of CK elevation and clinical progress.

All continuous values are expressed as median and interquartile range (IQR) or means ± standard deviations (SD). Categorical data is expressed in proportions. Where applicable, the values were analyzed by Student’s t-test for continuous data and Chi-square or Fisher’s exact probability test for categorical data to compare differences between groups. Non-normally distributed data were compared with non-parametric tests. The statistical package of social science (SPSS for Windows, version 12.0, SPSS Inc., Chicago, IL) was used in the analysis. A p value < 0.05 was considered significant.

Results

Forty-three patients with hyponatremia (sodium <135 mmol/L) and hyper-CKemia were initially identified. After excluding those with seizures and traumatic causes of elevated CK, 32 patients were included in this study. The mean age of these patients was 68.9 ± 14.2 years and a predominance of female (69%) in this cohort (Table 1). In this group of patients, the commonest cause of hyponatremia was diuretic-induced (n = 16; 50%), followed by primary polydipsia (n = 9; 28%) (Table 1). Among those with diuretic-induced hyponatremia, thiazide was taken by nine patients, indapamide by six and the combination of amiloride and hydrochlorothiazide by one. In all patients, no recent history (within 72 h of admission) of trauma, excessive exertion, excessive alcohol intake, illicit drug use, intramuscular injection or seizures was documented. No patient was taking lipid-lowering agents in the four weeks preceding presentation to hospital.

Table 1. Characteristics and clinical outcomes of patients with hyper-CKemia and hyponatremia.

	n = 32
Demographic data
Mean age ± SD, years	68.9 ± 14.2
Sex: Male/Female	10/22
Ethnicity – Caucasian (%)	32 (100)
Etiology of hyponatremia, n (%)
Diuretic induced	16 (50)
Primary polydipsia	9 (28)
SIADH	2 (6)
Severe diarrhea	2 (6)
Primary adrenal insufficiency	1 (3)
Cardiac failure	1 (3)
Viral meningitis	1 (3)
Clinical outcomes
Acute kidney injury, n (%)	6 (18)
ICU admission, n (%)	12 (38)
Mean length of stay ± SD, days	8.0 ± 5.8

Note: ICU: Intensive care unit; SIADH: Syndrome of inappropriate antidiuretic hormone secretion.

Table 2 shows the laboratory features of the subjects with hyponatremia-associated hyper-CKemia. The mean sodium level on presentation was 116.0 ± 6.9 mmol/L. All patients except for one had nadir sodium concentration of less than 125 mmol/L. Mean serum osmolality was also decreased (244.0 ± 17.0 mmol/L).

Table 2. Laboratory findings in patients with hyper-CKemia in the setting of hyponatremia.

	Mean ± SD or Median (Interquartile range)	Laboratory referenceintervals
Serum sodium, mmol/L	116.0 ± 6.9	137.0–145.0
Serum potassium, mmol/L	3.8 ± 0.6	3.2–4.3
Serum peak creatine kinase, IU/L	895.5 (610.8–1691.8)	<250 (Male) <150 (Female)
Serum urea, mmol/L	6.7 ± 6.6	2.7–8.0
Serum creatinine, µmol/L	57.0 ± 188.0	50.0–120.0
Serum osmolality, mmol/L	244 ± 17	265–285

The median peak CK elevation was 895.5 (IQR: 610.8–1691.8) IU/L. Three (9%) of the patients had peak CK values above 5000 IU/L. Elevation in CK was present at the same time as when the hyponatremia was detected in 69% (n = 22) of patients. In the rest of the patients, the CK rise occurred 24 to 48 h after presentation. None of the patients reported any symptom of muscular pain or weakness. No significant correlation was present between the severity of hyponatremia and the peak elevated CK concentration. After initial detection, the mean duration of CK elevation above reference interval was 1.7 ± 1.6 days. Hyponatremia improved with the appropriate treatment according to etiology and the CK level fell simultaneously. Urine myoglobin was tested and detected in 24 patients (75%) but it was not evaluated in the other patients. None of the patients had a muscle biopsy or electromyographic (EMG) study.

The mean peak creatinine concentration during the index admission was 57 ± 188 µmol/L. Six patients (23%) developed AKI. The kidney function in all six patients returned to baseline without the need for renal replacement therapy. No differences in age, severity of hyponatremia and peak CK level were found between patients who developed AKI and those who did not.

Thirty-eight percent (n = 12) of the patients needed to be admitted to the ICU because of the severity of hyponatremia. This subgroup had a mean length of stay of 1.5 ± 1.5 days in the intensive care unit. The overall length of admission of the entire cohort was 8.0 ± 5.8 days. No in-hospital death occurred in this cohort of patients.

The characteristics of two sub-groups of patients, diuretic-associated hyponatremia (n = 16) and primary-polydipsia-related hyponatremia (n = 9) were compared. Age was significantly higher (76.8 ± 10.7 vs. 55.9 ± 7.4 years; p < 0.001) and LOS significantly longer (9.1 ± 5.1 vs. 3.8 ± 2.1 days; p = 0.009) in the diuretic-associated cohort (Table 3). The severity of hyponatremia and peak concentration of CK was not significantly different between the two groups. One of the patients with diuretic-associated hyponatremia developed AKI but none from the primary polydipsia group.

Table 3. Comparison of laboratory findings and clinical outcomes of patients with hyper-CKemia in the setting of diuretic-associated and primary-polydipsia-associated hyponatremia.

	Diuretic-associated hyponatremia n = 16	Primary-polydipsia-associated hyponatremia n = 9	p Value
Mean age, years	75.9 ± 10.8	55.9 ± 7.4	<0.001
Sex, male, n (%)	1 (6)	5 (56)	0.01
Mean serum sodium ± SD, mmol/L	114.0 ± 5.6	117.0 ± 8.8	0.32
Median peak serum creatine kinase (IQR), IU/L	990.5 (580.5–1621.3)	1087.0 (833.0–2037.0)	0.27
Mean serum urea ± SD, mmol/L	4.6 ± 2.7	3.6 ± 2.4	0.13
Mean serum creatinine ± SD, µmol/L	52.3 ± 19.6	52.4 ± 10.9	0.59
Mean serum osmolality ± SD, mmol/L	238.9 ± 14.9	241.3 ± 21.9	0.92
Mean length of stay ± SD, days	9.1 ± 5.1	3.8 ± 2.1	0.009
Intensive care unit admission, n (%)	7 (44)	2 (25)	0.40

Note: IQR: Interquartile range.

Discussion

This study showed that asymptomatic hyper-CKemia can be associated with hyponatremia. Hyponatremia-associated hyper-CKemia was observed in two main groups of patients – elderly patients with diuretic-associated hyponatremia and patients with psychogenic polydipsia who were younger in age. In addition, the group with diuretic-associated hyponatremia had a longer hospital LOS compared to those with primary polydipsia.

In contrast to other electrolyte disturbances such as hypokalemia, hyponatremia is a less known association with hyper-CKemia or myopathy. From the present and past reports of hyponatremia-associated hyper-CKemia and myopathy, the causes of hyponatremia varied. In case reports of hyponatremia-associated CK elevation, many have occurred in patients with water intoxication who have drunk a large amount of fluid.15–17 Two previous cohort studies, with a total of 33 patients, reported hyponatremia-associated CK elevation in water intoxication.6,18 In one non-selective case series, the commonest cause of hyponatremia-associated hyper-CKemia was SIADH followed by primary polydipsia and congestive cardiac failure.8 Although hyponatremia-associated hyper-CKemia with water intoxication has been reported in several case reports, some of these presentations are confounded by the presence of coexisting causes of muscular injury such as seizures, intensive exercise and illicit drug use.19,20

In comparison to others, the present study found a larger number of patients who had hyper-CKemia in the setting of diuretic-related hyponatremia. The temporal observations would suggest that the hyper-CKemia was more likely related to hyponatremia or hypo-osmolality as opposed to the diuretic agents. Three previous case reports have been published on hyponatremia-associated hyper-CKemia or myopathy in the setting of diuretic-induced hyponatremia.10,21,22 Two were in the setting of thiazide diuretic use and one was with indapamide. No association has previously been reported on non-traumatic hyper-CKemia or myopathy and diuretic therapy in the absence of hyponatremia. However, this study is not designed to determine the incidence of myopathy in patients with hyponatremia or in specific etiological groups of hyponatremia. Further studies will be required to ascertain the incidence of hyponatremia-associated hyper-CKemia.

Even though the severity of hyponatremia did not differ significantly between the diuretic-associated group and primary polydipsia, the reason for the longer hospital LOS is uncertain. Perhaps the rate of correction of the sodium is achieved at a faster rate in the primary polydipsia group compared to the diuretic-associated cohort. Alternatively, patients taking diuretics may have medical comorbidities which lead to a longer hospital stay.

The elderly appears to be more predisposed to developing hyponatremia-associated hyper-CKemia. In a previous study, the slowly progressive muscle damage was only observed in the elderly subjects.8 However, this may be a reflection that the elderly were more prone to developing hyponatremia and not necessarily a risk factor for hyponatremia-associated hyper-CKemia.

Overall the rate of decline in CK concentration was relatively rapid in this study. This might be because the peak CK level in most patients was not extremely high with only three patients recording levels above 5000 IU/L. However, there may be other explanations that have not been documented accurately in this retrospective study such as intensive intravenous fluid therapy.

The clinical course of the patients in this cohort was relatively uncomplicated. Although several patients had to be admitted to ICU initially, it was primarily for close monitoring of the sodium correction. The outcome of hyponatremia-associated hyper-CKemia is usually good provided that there is no kidney injury. In the present study, a small number developed AKI but none required emergent dialysis. Previous literature on hyponatremia-associated rhabdomyolysis in small cohorts revealed an incidence of AKI of 9–40%.6,8,18 This is comparable with non-selective patients with rhabdomyolysis, the incidence of AKI is reported to be 13–50%.11,13,23 In this study, all patients survived including those with AKI. Furthermore, all patients who developed AKI recovered their renal function. The present study is not designed to address whether the hyper-CKemia is contributory to the development of AKI or if the underlying cause of hyponatremia is the primary contributing factor.

The mechanisms underlying hyponatremia-associated hyper-CKemia are not entirely known but one possibility is the acute disequilibrium between the intracellular and extracellular osmolality leading to breakdown of muscle cells.5 Another possible mechanism is through the activation of sodium–calcium exchange which causes an increase in intracellular calcium.24 When intracellular calcium ions reach a critical concentration, they activate neutral proteases and lipases that can destroy muscle cells.25 Past reports have suggested that the elevated CK may be related to the rapid correction rate of hyponatremia.26,27 Rapid correction of hyponatremia was identified as the main risk factor for hyper-CKemia in patients with self-induced water intoxication.6 However, in the current study, most of the patients presented with increased CK levels before the hyponatremia was corrected. Hyper-CKemia in settings other than hyponatremia have been associated with normal as well as non-specific findings on EMGs and muscle biopsies.2

This study has a number of limitations. Data collection was retrospective and subject to the limitations inherent in this study design. This study only included patients treated in a general medical department and has not included patients who developed hyponatremia after surgery such as after prostate surgical procedures. From this study, a definite causal relationship between hyponatremia and hyper-CKemia cannot be made. Nevertheless, a combined or more complex interaction between electrolyte disturbances and muscular damage cannot be ruled out.

This study has found that hyponatremia is associated with a reversible asymptomatic hyper-CKemia. The causes of hyponatremia complicated by hyper-CKemia are varied and can include diuretic-associated hyponatremia, primary polydipsia and SIADH. Patients with hyper-CKemia in the setting of diuretic-associated hyponatremia were older and required longer hospital LOS compared to those with primary polydipsia. Further studies are required to determine the incidence and effects of hyponatremia-associated hyper-CKemia on clinical outcomes.

Declaration of interest

The authors report no conflicts of interest. The authors alone are responsible for the content and writing of the paper.