 |
|
 |
|
 |
 |
|
 |
 |
 |
 |
 |
 |
 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
PEDIATRICS Vol. 104 No. 3 September 1999, pp. 435-439
From the * Department of Pediatrics, Texas Children's Hospital,
Baylor College of Medicine, Houston, Texas; and the Department of
Medicine, Baylor College of Medicine, Houston, Texas.
 |
ABSTRACT |
|---|
Top
Abstract
Methods
Results
Discussion
Conclusion
References
|
|---|
Objectives. Past studies have revealed that hypernatremia occurs primarily in infants with diarrheal dehydration. With improved infant feeding practices and the advent of pediatric critical care medicine, the pattern of hypernatremia in children has likely changed. The purpose of this study was to evaluate the current pattern of hypernatremia in hospitalized children.
Methods. Medical records were reviewed for 68 patients admitted to a large urban children's hospital during a 3-year period, all with a serum sodium greater than 150 mEq/L. The etiologies, predisposing factors, and morbidity and mortality associated with hypernatremia were evaluated.
Results. The average patient age was 3.9 years (range, 1 day to 19.7 years), and the peak serum sodium concentration was 159 mEq/L (range, 151-184 mEq/L). Hypernatremia was hospital acquired in 60% of children. The majority of children (71%) were admitted for reasons other than hypernatremia. In 76% of the patients, inadequate fluid intake was the main cause of hypernatremia. Gastroenteritis contributed to the hypernatremia in only 20% (14 out of 68) of children. Eleven of these were infants <1 year of age with hypernatremia on admission. Eighty-eight percent of patients (60 out of 68) suffered from neurologic impairment, critical illness, chronic disease, or prematurity before developing hypernatremia. The overall mortality was 16%. Patients in whom hypernatremia was not corrected had a significantly higher mortality than those in whom hypernatremia was corrected (4 out of 8 [50%] vs 7 out of 60 [12%]). Peak serum sodium was no different for survivors than nonsurvivors. No deaths were attributable to cerebral edema caused by correction of hypernatremia. Neurologic complications related to hypernatremia occurred in 15% of patients.
Conclusions. Hypernatremia occurs in children of all ages, with the vast majority having significant underlying medical problems. Hypernatremia caused by gastroenteritis in infants has become much less common than previously reported. Hypernatremia is primarily a hospital-acquired disease, produced by the failure to administer sufficient free water to patients unable to care for themselves. Failure to correct hypernatremia may result in a high mortality rate. Key words: hypernatremia, sodium, pediatric, fluid, treatment, mortality.
Previous studies have revealed that hypernatremia occurs
primarily in infants with diarrheal dehydration.1-8 During the course of hypernatremic dehydration infants can develop central nervous system dysfunction, consisting of seizures, decreased consciousness and coma, and may suffer brain damage or die from intracranial hemorrhages, infarction, and venous
thrombosis.29-14 Improper rehydration has been shown to
result in brain dysfunction as a result of cerebral
edema.11,1215-18 Recognition of these serious problems
led to the institution of low solute infant formulas that resulted in a
decrease in the incidence of infantile hypertonic
dehydration.719-22
Since Rapoport's1 first description of hypernatremia in
childhood, few investigators have evaluated this condition in an
inpatient population.23 We reviewed the records of
children admitted to a large children's hospital to obtain an
up-to-date picture of the predisposing factors, the common etiologies,
and the associated morbidity and mortality of hypernatremia.
Study Population
The medical records department queried a database of 32 000
patients admitted to Texas Children's Hospital in Houston, Texas, from
January 1992 through December 1994 to identify all patients with a
discharge diagnosis of hyperosmolality. Of the 108 admissions identified with this discharge diagnosis, 68 patients were found to
have a serum sodium concentration >150 mEq/L. The remaining patients
had hyperglycemia or an incorrect discharge diagnosis. The incidence of
hypernatremia in hospitalized children by this method (0.22%) was less
than would be estimated by querying the department of pathology
chemistry lab database (1.4%) not excluding for values from outpatient
departments, the emergency department, and lab errors. The present
study population seems to be a representative sample with a peak serum
sodium (159 ± 5.1 mEq/L) similar to that found when querying the
lab database (157 ± 6.8 mEq/L).
Evaluation
A single physician reviewed all medical records. Patient age was
assigned as the time that serum sodium first became >150 mEq/L. Time
to correction of hypernatremia was determined from the time therapy was
instituted until the time serum sodium concentration was Patients were classified as having an associated medical problem before
the development of hypernatremia if they had a chronic disease or an
underlying neurologic impairment, or if they were critically ill,
premature, or expremature infants. Patients were classified as
critically ill if they were intubated for reasons other than
prematurity before developing hypernatremia. Infants were classified as
being premature if their birth weight was <2500 g, their gestational
age was <35 weeks, and they had not been discharged to home since
birth. Expremature infants had been discharged to home before
admission.
Patients were classified as having underlying neurologic impairment if
there was documentation of structural brain damage, increased
intracranial pressure or hemorrhage, developmental delay, mental
retardation, or fixed neurologic deficits before the development of
hypernatremia. When present, the underlying neurologic status and
reason for impairment were recorded. The final neurologic status was
recorded at discharge and was classified as impaired if the above
criteria were met, or if there were any focal abnormalities on
neurologic examination.
The causes contributing to hypernatremia were recorded for each
patient. These were: limited fluid intake, increased extrarenal or
urinary water losses, and excess sodium administration. Limited fluid
intake was a contributing factor if there was documentation of
decreased oral fluid intake, fluid restriction of <800
mL/m2/24 h,24 or inadequate access to fluids
in patients who were otherwise able to tolerate oral fluids. Urinary
water losses were considered increased if there was a concentrating
defect, such as nephrogenic or central diabetes insipidus, or if a
solute diuresis because of glucose, urea, or a loop diuretic was
present. A quantity of sodium Statistical Analysis
Data were analyzed using the SigmaStat statistical software
package (Jandel Scientific Software, San Rafael, CA). Descriptive data
were reported as the mean ± SD or as mean + range. Ordinal data
were analyzed using the Mann-Whitney rank sum test. Nominal data were
analyzed using the Fisher's exact test. A P value < .05 was considered significant.
Hypernatremia occurred in 68 children of all ages (Table
1). Twelve children (18%) were premature
infants. In the remaining children, only 19 of 56 (34%) were <1 year
of age. Hypernatremia was hospital acquired in 41 of 68 patients
(60%). The majority of children (71%) were admitted for reasons other
than hypernatremia.
TABLE 1
METHODS
Top
Abstract
Methods
Results
Discussion
Conclusion
References
150 mEq/L
for at least 24 hours. The reason for hospitalization, the presence of
significant underlying medical problems, and associated problems during
hospitalization were recorded for each patient.
5 mEq/kg/d administered in a
concentration exceeding 150 mEq/L was considered a contributing factor
to hypernatremia. Complications of hypernatremia were recorded, as were
the cause of death and pertinent autopsy findings when available.
RESULTS
Top
Abstract
Methods
Results
Discussion
Conclusion
References
Demographic Characteristics of Pediatric Patients With Hypernatremia
The primary factors that contributed to hypernatremia were inadequate fluid intake (76%, 52 out of 68), followed by gastrointestinal losses (44%, 30 out of 68), high urinary water losses (44%, 30 out of 68), and sodium excess (26%, 18 out of 68). Gastroenteritis, historically the major cause of hypernatremia in children,1-5,9,25,26 was a contributing factor in only 14 children. It predominantly affected full-term infants <1 year of age (57%, 11 out of 19), and patients with hypernatremia on admission (48%, 13 out of 27). Excessive urinary water losses, on the other hand, predominantly affected children >2 years of age (52%, 15 out of 29), and patients with hospital-acquired hypernatremia (54%, 22 out of 41). Diabetes insipidus was the major reason for excessive urinary water losses (43%, 13 out of 30), occurring in 9 children with central diabetes insipidus and 3 with nephrogenic diabetes insipidus. Excess intake of sodium predominately contributed to hypernatremia in the critically ill (53%, 9 out of 17). There were no cases of hypernatremia as a result of errors in infant formula preparations, pharmacy errors, salt poisoning, or breastfeeding malnutrition.
Associated Medical Problems
Sixty patients (88%) had an associated medical problem before developing hypernatremia. A chronic disease was the most common associated medical problem (57%, 39 out of 68). Neurologic impairment was present in 26 patients (38%). In 22 of these patients, limited fluid intake contributed to hypernatremia. Many of these children were profoundly disabled, and thereby dependent on caregivers to meet their fluid needs. Seventeen children were critically ill (22%); 11 of them had high urinary water losses.
The birth weight in premature infants was 844 ± 336 g, (range, 485-1730 g). Eleven were born after <28 weeks of gestation. In 10 of the 12 premature infants, insufficient fluid intake contributed to hypernatremia. Eight ex-premature patients accounted for 11 admissions. Six of these patients had complications of prematurity, such as short gut syndrome (n = 4), severe gastroesophageal reflux (n = 2), or neurologic impairment (n = 4), that either increased free water losses or decreased fluid intake.
Morbidity and Mortality
The overall mortality was 16% (11 patients) (Table 2). Four children died while hypernatremic; all were critically ill. Two of these children became hypernatremic on the day of death as a result of sodium bicarbonate administration for acidosis. The other 2 were hypernatremic for 9 days and 46 days because of fluid restriction. The mortality was substantially higher among patients who remained hypernatremic, than among those in whom hypernatremia was corrected (4 out of 8 vs 7 out of 60; P < .05).
|
Seven patients died after hypernatremia was corrected (Table 2), 5 were premature infants (birth weight, 700 ± 59 g) and 2 were older children. Hypernatremia was corrected at a rate of 0.29 ± 0.25 mEq/L/h (range, 0.13-0.79 mEq/L/h) for all patients with corrected hypernatremia. The fall in serum sodium was 12 ± 5 mEq/L (range, 8-19 mEq/L). Death occurred 7 to 90 days after the correction of hypernatremia. No deaths were attributable to cerebral edema because of correction of hypernatremia. The peak serum sodium concentration was similar in survivors and in those who died (159 mEq/L). Diarrheal dehydration was not the cause of death in any of the patients.
Ten patients (15%) developed neurologic complications related to hypernatremia (Table 3). Four patients developed seizures that resolved after the correction of hypernatremia. Eight patients had long-term neurologic sequelae. In 6 patients the neurologic sequelae were most likely caused by their underlying disease rather than to hypernatremia. Two infants with diarrheal dehydration who were previously normal had an impaired neurologic status with hypertonicity at discharge.
|
| |
DISCUSSION |
|---|
|
Top
Abstract
Methods
Results
Discussion
Conclusion
References
|
|---|
Hypernatremia has been reported to be a disease that primarily occurs in infants with diarrheal dehydration.1-5,16,25 Our study showed that, currently, in a tertiary care center, hypernatremia occurs in children of all ages. When excluding premature infants, it primarily occurs in children >1 year of age (37 out of 56, 66%). Only 11 children in our study were infants with gastroenteritis. The reason that the average age of patients in our study was greater than would be expected from previous reports could be explained by a decrease in the incidence of infantile hypernatremic diarrheal dehydration.7,19,21 This decrease has been attributed to widespread use of low-solute infant formulas, and the improved availability of oral rehydration solutions.27
Most of the patients described in our study developed hypernatremia
after admission to the hospital. This is similar to recent reports in
adults and children. In a large study of 162 elderly patients, 57%
developed hypernatremia while being in the hospital. In another study
of 103 adults, 83% became hypernatremic after hospitalization.28,29 In a series of 29 children with a
serum sodium concentration 165 mEq/L, 56% percent developed hypernatremia while in the hospital.23 Most of the
children in our study who developed hypernatremia after admission to
the wards were either critically ill or premature.
The majority of patients in our series had an associated medical problem. Neurologic impairment was the major predisposing factor, occurring in 26 of the 56 patients who were not premature. This observation is not without precedent.1,3,4,11,12,30 Franz and Segar4 showed that 12 of 68 patients with hypernatremic dehydration had preexisting brain damage, whereas Macaulay and Watson11 showed that 14 of 89 infants who survived hypernatremia had preexisting brain damage. Inadequate access to water is the factor that most likely accounts for this high incidence. Greater attention to providing adequate free water to these children, and assessing their fluid balance and serum sodium concentration during times of illness, is clearly required.
In 8 of our patients, the neurologic status was worse at the time of
discharge than on admission. Six of these patients had other medical
problems, such as prematurity and metabolic diseases, making it
difficult to determine what role hypernatremia played in their
neurologic deterioration. Previous reports have shown that children
with hypernatremia can develop neurologic impairment. Macaulay and
Watson11 observed that 33 of 114 infants with
hypernatremia (29%) went on to show evidence of brain damage. In 17 of
these patients, the brain damage was unrelated to or antedated
hypernatremia. Dunn and Butt23 found that 7 of 17 children
who survived severe hypernatremia (Na 165), had a worsened neurologic
status at follow-up. These observations are consistent with
experimental data in rats that revealed that acute hypernatremia can
lead to permanent histologic brain damage, consisting of myelinolysis
of the white matter and necrosis of neurons.31 In our
study, 12% of patients had long-term neurologic damage. The percentage
is lower than that reported previously.5,23 This may be
attributable to the relatively few infants in our series with
infectious diarrhea and the lower average peak serum sodium in our
series than in those of others.
The overall mortality of 16% (11 out of 68) in our series is 16 times higher than the 1% overall mortality of Texas Children's Hospital in 1996. This mortality is, however, similar to that reported by others in patients with hypernatremia.2,5,27 Our study is unique in that there were no deaths related to diarrheal dehydration. All the deaths in our study occurred either in premature or critically ill infants. Three of 7 (42%) premature infants with birth weights <750 g died, a percentage similar to that for the institution as a whole, and 1 in 4 (25%) premature infants with a birth weight between 750 g to 1000 g died compared to 22% for the institution. The mortality among critically ill patients with hypernatremia was 10 times higher than the mortality among patients without hypernatremia admitted to the intensive care unit during a 1-year period (41% [7/17] vs 7.3% [15/346]; P < .001). As already indicated, the mortality was the highest in children with uncorrected hypernatremia: 4 of our 6 deaths among nonpremature infants occurred while hypernatremia was present. Although hypernatremia was not the sole factor that contributed to their death, prolonged uncorrected hypernatremia seemed to be a contributing factor to their demise. This observation is consistent with that made by other investigators. Mandal et al32 found that in hospitalized adults with hypernatremia, the average serum sodium at the time of death was 152 mEq/L. Dunn and Butt23 found that of 10 hypernatremic children who died, 5 died with a serum sodium >150 mEq/L. Moritz et al33 observed that delayed or inadequate treatment of hypernatremia was a significant comorbid factor that contributed to death. Thus, although hypernatremia does not increase the mortality in premature infants, the development of hypernatremia seems to increase the mortality in the critically ill child with hypernatremia compared with the one without.
No deaths among our patients appeared to result from rapid correction
or overcorrection of hypernatremia. Animal data revealed that rapid
correction (24 hours) of both acute or chronic hypernatremia can
result in cerebral edema.31 This may not apply to our
patients who had a lower peak serum sodium concentration and an average
time to correction of hypernatremia of 48 hours.
| |
CONCLUSION |
|---|
|
Top
Abstract
Methods
Results
Discussion
Conclusion
References
|
|---|
In summary, hypernatremia occurs in children of all ages. It is primarily a hospital-acquired disease. The majority of these children have significant underlying medical problems such as a chronic disease, neurologic impairment, a critical illness, or prematurity. The major cause of hypernatremia is failure to administer sufficient free water to patients who are unable to care for themselves. Critically ill children who develop hypernatremia have a significantly increased mortality, with most deaths occurring before correction of hypernatremia. Few complications seem to be related to therapy. In most children, hypernatremia could be prevented by frequently assessing fluid and electrolyte balance and by providing adequate free water to these patients.
| |
ACKNOWLEDGMENTS |
|---|
We thank Lee Boletter for his assistance in constructing the computer database and Adrian Spitzer, MD, for his review and critique of the manuscript.
| |
FOOTNOTES |
|---|
Dr Moritz worked on this manuscript as a fellow in Pediatric Nephrology at Texas Children's Hospital. He is currently in the Department of Pediatrics, Division of Nephrology at the Children's Hospital of Pittsburgh, University of Pittsburgh School of Medicine.
Received for publication Nov 10, 1998; accepted Mar 23, 1999.
Presented in part at the 29th annual meeting of the American Society of Nephrology; November 3-6, 1996; New Orleans, LA.
Reprint requests to (M.L.M.) Children's Hospital of Pittsburgh, 3705 Fifth Ave, Pittsburgh, PA 15213-2583. E-mail: moritz613{at}hotmail.com
| |
REFERENCES |
|---|
|
Top
Abstract
Methods
Results
Discussion
Conclusion
References
|
|---|
This article has been cited by other articles:
 |
|
 |
|
  E. Koklu, T. Gunes, M. A. Ozturk, M. Kose, S. Kurtoglu, and F. Yuksel A Review of 116 Cases of Breastfeeding-Associated Hypernatremia in Rural Area of Central Turkey J Trop Pediatr, October 1, 2007; 53(5): 347 - 350. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. L. Moritz, M. D. Manole, D. L. Bogen, and J. C. Ayus Breastfeeding-Associated Hypernatremia: Are We Missing the Diagnosis? Pediatrics, September 1, 2005; 116(3): e343 - e347. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. L. Moritz and J. C. Ayus Disorders of Water Metabolism in Children: Hyponatremia and Hypernatremia Pediatr. Rev., November 1, 2002; 23(11): 371 - 380. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
 |
||
|
||
Home | My Pediatrics | Journal Information | Current Issue | Past Issues | Subscriptions & Services | Contact Us Click here for faster international access © 1999 American Academy of Pediatrics. All rights reserved. |
||
|
||
