Maintenance Intravenous Fluids in Acutely Ill Patients: Review Article
Maintenance Intravenous Fluids in Acutely Ill Patients: Review Article
Maintenance Intravenous Fluids in Acutely Ill Patients: Review Article
Review Article
A
From the Department of Pediatrics and critical aspect of the care of acutely ill patients is the admin-
Division of Nephrology, Children’s Hos- istration of intravenous fluids. Intravenous fluids may be required as a
pital of Pittsburgh of the University of
Pittsburgh Medical Center (UPMC), Uni- bolus infusion for resuscitation or as a continuous infusion when sufficient
versity of Pittsburgh School of Medicine, fluids cannot be ingested orally.
Pittsburgh (M.L.M.); Renal Consultants The goal of maintenance intravenous fluids is to preserve the extracellular vol-
of Houston, Houston (J.C.A.); and the
Division of Nephrology, Hospital Italiano ume while maintaining a normal electrolyte balance. An appropriate maintenance
de Buenos Aires, Buenos Aires (J.C.A.). fluid provides an adequate quantity of both water and electrolytes to ensure good
Address reprint requests to Dr. Moritz at tissue perfusion without causing complications related to fluid overload or volume
the Division of Nephrology, Children’s
Hospital of Pittsburgh of UPMC, 1 Chil- depletion. It also prevents the development of hyponatremia, hypernatremia, and
dren’s Hospital Dr., 4401 Penn Ave., other electrolyte imbalances.
Pittsburgh, PA 15224, or at m oritzml@ Despite the almost ubiquitous need for intravenous fluids in acutely ill pa-
upmc.edu.
tients, there has been little consensus on the most appropriate rate of adminis-
N Engl J Med 2015;373:1350-60. tration and composition of intravenous fluids, and practice patterns with respect
DOI: 10.1056/NEJMra1412877
Copyright © 2015 Massachusetts Medical Society. to maintenance fluids vary widely.1-3 In addition, acutely ill patients frequently
have conditions that impair normal water and electrolyte homeostasis (Table 1),
and choosing the appropriate volume and composition of intravenous fluids requires
great care.4
Intravenous fluids can be classified, according to the concentration of sodium
plus potassium in the fluid, as being either isotonic (approximately equal to the
plasma sodium concentration) or hypotonic (less than the plasma sodium concen-
tration). The dextrose content of intravenous fluids has no effect on the tonicity,
since the dextrose is rapidly metabolized when it enters the bloodstream and
should not produce hyperglycemia.
A prevailing practice has been to administer hypotonic maintenance intrave-
nous fluids in both children and adults.1,5,6 This practice has been associated with
a high incidence of hospital-acquired hyponatremia and more than 100 reports of
iatrogenic deaths or permanent neurologic impairment related to hyponatremic
encephalopathy,7-11 since acutely ill patients may have disease states associated
with an excess of arginine vasopressin (AVP) (Fig. 1). This excess impairs excretion
of free water and may aggravate hyponatremia.15,16
Over the past decade, many studies have evaluated the association between the
composition and quantity of maintenance fluids and the development of hypona-
tremia. Recent data also suggest that rapid volume expansion with 0.9% saline
(normal saline), as compared with balanced electrolyte solutions, may result in
untoward complications.17 The current review considers the physiological princi-
ples that guide the appropriate selection of intravenous fluids, as well as the recent
literature evaluating the safety of various intravenous fluids with respect to their
composition and rate of administration.
AV P a nd Disor der s of Sodium Table 1. Conditions Requiring Special Considerations in Maintenance Fluid
a nd Water Homeos ta sis Therapy.
The human body has a remarkable ability to main- Free-water restriction for euvolemic states of AVP excess
tain a normal composition of body water and CNS disturbances
plasma osmolality, despite wide variations in Meningitis
fluid and electrolyte intake.18 Sodium and water Encephalitis
homeostasis is regulated through the actions of Brain tumors
AVP, the renin–angiotensin–aldosterone system, Head injury
and natriuretic peptides. Plasma osmolality is Cerebritis
regulated both by thirst and by excretion of free Subarachnoid hemorrhage
water. In a patient who can ingest nothing by Pulmonary disease
mouth, plasma osmolality is primarily under the Pneumonia
control of AVP release, which determines the Asthma
rate of free-water excretion.12 Any disease state Bronchiolitis
that results in either excess AVP or impaired AVP Tuberculosis
action will place a patient at risk for a plasma Cancer
sodium concentration that is too high or too low. Postoperative state
There are numerous hemodynamic and non- Fluid restriction for edematous states
hemodynamic stimuli for AVP secretion; thus, Congestive heart failure
virtually all acutely ill hospitalized patients are Nephrosis
at risk for hyponatremia (Fig. 1). Hemodynamic Cirrhosis
stimuli for AVP production include volume deple- Fluid and sodium restriction for oliguric states
tion; hypotension; edematous states such as con- Acute glomerulonephritis
gestive heart failure, cirrhosis, and the nephrotic Acute tubular necrosis
syndrome; and sepsis.13 Nonhemodynamic phys- End-stage renal disease
iological stimuli for AVP secretion include pain, Increased free-water requirements for renal concentrating defects
stress, nausea, vomiting, hypoxemia, hypercapnia, Congenital nephrogenic diabetes insipidus
hypoglycemia, and the perioperative state.12,14 Sickle cell disease
These physiological stimuli can result in in- Obstructive uropathy
creased AVP levels in the absence of volume de-
Reflux nephropathy
pletion or hyperosmolality.12 In addition, an ever-
Renal dysplasia
expanding list of conditions and medications is
Nephronophthisis
associated with the syndrome of inappropriate
Tubulointerstitial nephritis
antidiuresis, in which AVP excess occurs in the
Use of lithium
absence of any identifiable nonosmotic stimuli
Increased sodium and water requirements for solute diuresis
for AVP production.19 The conditions most often
Diuretic phase of acute tubular necrosis
associated with this syndrome are cancer, cen-
Postobstructive diuresis
tral nervous system (CNS) disorders, pulmonary
Immediate postoperative renal transplantation
disorders, and infections.19 Medications that are
frequently associated with this syndrome include Diabetic ketoacidosis
narcotics, the chemotherapeutic agents cyclophos- Bartter’s syndrome
phamide and vincristine, selective serotonin- Fanconi’s syndrome
reuptake inhibitors, the antiepileptic agent ox- Cerebral salt wasting
carbazepine, and the recreational drug “ecstasy” Adrenal insufficiency
(3,4-methylenedioxymethamphetamine [MDMA]). Increased free-water requirements for extrarenal free-water losses
Thiazide diuretics can also be associated with a Burns
condition that resembles the syndrome of inap- Prematurity in neonates
propriate antidiuresis. The syndrome of inappro- Fever
priate antidiuresis is now recognized as the most Infectious diarrhea
common cause of euvolemic hyponatremia.20
The most consistent symptoms of hyponatremic Table 2. Risk Factors for Hyponatremic Encephalopathy.
encephalopathy are headache, nausea, vomiting,
and generalized weakness.7,38,39 Advanced symp- Risk Factor Pathophysiological Mechanism
toms of hyponatremic encephalopathy include Acute hyponatremia Decreased time for brain adaptation
seizures, respiratory arrest, noncardiogenic pul- (<48 hr)
monary edema, and decorticate posturing.7,38,39 Age <16 yr Increased ratio of brain to intracranial volume
Symptoms can occur abruptly and do not always Female sex Sex steroids (estrogens) inhibit brain adaptation
correlate with the plasma sodium concentration Increased AVP levels
or the rapidity of development of hyponatre- Cerebral vasoconstriction
Hypoperfusion of brain tissue
mia.7,10,38,40
Hypoxemia Impaired brain adaptation
Failure to recognize and treat hyponatremic Decreased cerebral perfusion
encephalopathy with hypertonic saline results in
Brain injury Vasogenic cerebral edema
a poor neurologic prognosis.7,10 Fluid restriction Cytotoxic cerebral edema
alone, isotonic fluids, and vaptans have no role
in the immediate management of hyponatremic
encephalopathy.14,39 as either of the following: 1500 ml per 1.73 m2 of
Mounting data show that even chronic hypona- body-surface area per 24 hours or, as calculated
tremia, which develops over a period of 48 hours with the use of the Holliday–Segar formula, 100 ml
or longer, without apparent neurologic symp- per kilogram per 24 hours for the first 10 kg
toms, is deleterious. Chronic hyponatremia has of body weight, plus 50 ml per kilogram per
been shown to result in subtle neurologic im- 24 hours for a weight of greater than 10 to 20 kg,
pairment, which can cause gait disturbances and an additional 20 ml per kilogram per 24 hours
that lead to falls and associated bone fractures for a weight above 20 kg.6,52
in the elderly, with increased bone fragility re- Such recommendations are based on theoreti-
sulting from bone demineralization.41-45 cal calculations from the 1950s, before the syn-
Hyponatremia is an independent risk factor drome of inappropriate antidiuresis was recog-
for death in the hospital setting, particularly nized as a common clinical entity. In 1953,
among patients with end-stage liver disease, Talbot et al. made recommendations regarding
congestive heart failure, pneumonia, and end- the administration of fluids in adults according
stage renal disease, and it is associated with to a theoretical model of the maximal and
increased hospital costs, length of hospital stay, minimal amount of sodium and water in paren-
and rates of readmission.46-50 When hyponatre- teral fluids that could be administered without
mia develops, it can be difficult to correct, since adverse effects.53 In 1957, Holliday and Segar
most initial therapies, such as fluid restriction made similar recommendations for children ac-
and isotonic fluids, are relatively ineffective in cording to average caloric requirements for hos-
correcting euvolemic and hypervolemic hypona- pitalized children and electrolyte requirements
tremia.20 For all these reasons, strategies should based on dietary intake.52 However, these recom-
be implemented to prevent hyponatremia. mendations lack validation in clinical practice in
children or adults. All current recommendations
regarding the composition and rate of mainte-
C oncer ns a bou t H y p o t onic
M a in tena nce Fluids in nance fluid therapy in children and adults, regard-
Chil dr en a nd A dult s less of the patient’s condition, are predominantly
opinion-based.51
Hypotonic fluids continue to be recommended We previously suggested that isotonic fluids
as maintenance fluids in acutely ill patients de- should be administered in maintenance intrave-
spite the strong association between their use nous fluids to prevent hospital-acquired hypo-
and the development of hospital-acquired hypo- natremia in acutely ill children and that the
natremia.6,51 The prevailing practice has been to routine practice of administering hypotonic fluids
administer hypotonic fluids (e.g., 5% dextrose in (sodium concentration, <130 mmol per liter) be
a solution of 0.18 to 0.45% saline) at a rate of abandoned because of the increased risk of hypo-
approximately 2.0 to 3.0 liters per day in adults. natremia.54 We subsequently extended this rec-
The equivalent amount in children is estimated ommendation to apply to adult patients.16
Currently, to our knowledge, there are no overload with edema or hypertension, and hy-
formal society-sponsored guidelines for mainte- perchloremic acidosis.60
nance fluid therapy in hospitalized patients in More than 15 randomized, prospective trials
the United States. Attempts have been made to involving more than 2000 patients have evalu-
develop consensus guidelines for children and ated the safety and efficacy of isotonic fluids as
adults in the United Kingdom.6,32,55,56 compared with hypotonic fluids for the preven-
The first consensus guideline on fluid ther- tion of hyponatremia.61-66 Most of these studies
apy was issued by the U.K. National Health involved children, the majority of whom were
Service in 2007. The National Health Service is- surgical and critical care patients. Isotonic flu-
sued a patient-safety alert to remove 0.18% sa- ids were superior to hypotonic fluids for the
line from general-use areas (e.g., the emergency prevention of hyponatremia in all but 1 small
department and pediatric ward) for children and study,67 in which 37 patients were followed for
recommended 0.45% saline for maintenance approximately 12 hours. Isotonic fluids were not
fluids for the majority of children and 0.9% sa- associated with an increased risk of hypernatre-
line for children at high risk for hyponatremia.32 mia or fluid overload, and there were no appar-
These opinion-based recommendations were sub- ent serious complications. A meta-analysis of 10
sequently adopted for the care of children in of these studies, involving almost 1000 children,
Northern Ireland by the Regulation and Quality showed that hypotonic fluids were associated
Improvement Authority and in Australia by the with a relative risk of 2.37 for the development
Victorian Paediatric Clinical Network. Prelimi- of mild hyponatremia (sodium concentration,
nary data suggest that these recommendations <135 mmol per liter) and a relative risk of 6.2 for
may have been helpful in reducing the incidence the development of moderate hyponatremia (so-
of hyponatremia.57 dium concentration, <130 mmol per liter).61
Even though 0.18% saline was essentially Some physicians have argued that fluid restric-
banned throughout the United Kingdom for use tion with hypotonic fluids would be sufficient to
in children, consensus guidelines still recom- prevent hyponatremia.58 In 4 studies, fluid re-
mend it as a maintenance fluid in adults.6,55 The striction with hypotonic fluids was not effective
British Consensus Guidelines on Intravenous in preventing hyponatremia.68-71
Fluid Therapy for Adult Surgical Patients, which It has also been argued that isotonic fluids
were published in 2008 and updated in 2011, may not be necessary for the prevention of hypo-
recommend 0.18 to 0.45% saline for postopera- natremia in a general pediatric ward. A study
tive maintenance fluids.55 In 2013, the National involving almost 700 children showed that the
Institute for Health and Care Excellence guide- odds ratio for prevention of hyponatremia with
lines for intravenous fluid therapy in adults, an isotonic fluid was similar among surgical
which were published in the United Kingdom, patients and nonsurgical patients and among
recommended 0.18% saline for maintenance patients who were in intensive care and those
fluids.6 These guidelines were primarily opinion- who were not in intensive care.62
based and included low-quality evidence on which Taken together, these studies suggest that
to base recommendations. These consensus isotonic fluids are most appropriate for the vast
guidelines lack validation in clinical practice. majority of hospitalized patients who are at risk
for elevated AVP levels and hospital-acquired
hyponatremia. This does not mean that isotonic
Iso t onic M a in tena nce Fluids
for the Pr e v en t ion of Hospi ta l - fluids are appropriate in all clinical settings or
Ac quir ed H y p onat r emi a that they are without risk. The studies were of
short duration (usually <72 hours) and excluded
The use of isotonic fluids in maintenance fluids patients with renal disease, heart failure, and
for the prevention of hospital-acquired hypona- cirrhosis. Precautions were taken if complica-
tremia in acutely ill patients was initially contro- tions were developing. It is not certain whether
versial,58,59 owing mainly to the potential for the side-effect profile with isotonic fluids would
unintended consequences. These consequences be as favorable in other patient populations,
include the development of hypernatremia, fluid such as the elderly, as it is in children.
Isotonic fluids may result in hyponatremia in is perfectly matched with plasma (Table 3).
patients with CNS injury in whom cerebral salt Multiple electrolytes injection, type 1, USP, (an
wasting develops or in patients with the syn- infusion fluid) has a supraphysiologic buffer con-
drome of inappropriate antidiuresis in whom the centration of 50 mmol per liter. Ringer’s lactate
urine osmolality is greater than 500 mOsm per contains calcium and may be incompatible with
kilogram.72 Isotonic fluids could also result in blood products and some medications. There are
hypernatremia if there is a renal concentrating insufficient data to suggest that 0.9% saline is
defect or if there are large extrarenal free-water unsafe when used as a maintenance fluid.
losses. Any type of intravenous fluid can result
in fluid overload if administered in excessive Selecting the Rate of Administration
quantities (for example, in patients with end- and the Composition of Maintenance
stage renal disease or heart failure in whom Intravenous Fluids
both sodium excretion and water excretion are No single rate of administration or composition
impaired). In most circumstances, the adminis- of maintenance intravenous fluids is appropriate
tration of isotonic fluids should not lead to hyper- in all circumstances. Thus, intravenous fluids
natremia or fluid overload.73 may be viewed as medications that require care-
ful dose adjustment that is specific to the dis-
Balanced Salt Solutions as Compared ease state of each patient.
with 0.9% Saline The rate of administration and the composi-
An area of active investigation focuses on the tion of intravenous fluids need to be individual-
potential for deleterious effects associated with ized, and while patients are receiving intrave-
0.9% saline (normal saline) as compared with nous fluids, they require close monitoring with
those associated with balanced electrolyte solu- daily measurement of weight, frequent assessment
tions (Table 3).17,74 Normal saline (sodium con- of vital signs, strict measurements of intake and
centration, 154 mmol per liter) has the same output, and daily measurement of serum electro-
sodium concentration as the aqueous phase of lyte levels and physical examination. Many chil-
human plasma, but it has a supraphysiologic dren have died of hyponatremic encephalopathy
chloride concentration. Saline solutions may also within 24 hours after the initiation of hypotonic
have a pH that is far lower than that of water intravenous fluids, so even close observation may
(range, 3.5 to 7.0). The low pH of 0.9% saline be inadequate to prevent this complication.
appears to be related to the polyvinyl chloride Table 1 lists a broad range of conditions that
bags in which it is packaged, since the pH of require special consideration when maintenance
0.9% saline in a glass bottle is 7.0. fluid therapy is administered, and Figure 2
Deleterious consequences of 0.9% saline have shows a practical approach for adjusting mainte-
been associated primarily with studies (both in nance intravenous fluids according to these con-
animals and in humans) in which infusion is ditions. This approach is opinion-based, since
rapid and with clinical scenarios in which fluid data are lacking from studies evaluating the most
resuscitation or intraoperative fluid management appropriate fluid therapy for all conditions.
includes administration of a large volume of Dextrose is provided in maintenance fluids to
fluid.74 In such circumstances, 0.9% saline, as provide sufficient calories to prevent hypoglyce-
compared with balanced salt solutions, may pro- mia and limit tissue catabolism, but it does not
duce a hyperchloremic metabolic acidosis, re- provide complete nutritional support. Some dex-
nal vasoconstriction, delayed micturition, an in- trose-containing saline solutions are hyperosmo-
creased incidence of acute kidney injury requiring lar to plasma, but they are not hypertonic, since
renal-replacement therapy, and hyperkalemia. the dextrose is rapidly metabolized on entering
Whether any of these complications will oc- the bloodstream (Table 3). The default mainte-
cur when 0.9% saline is used at maintenance nance solution for adults is 5% dextrose in a
rates for less than 72 hours is unclear, but ran- solution of 0.9% saline administered at a rate of
domized trials that have been sufficiently pow- 100 to 120 ml per hour; this is based on the fact
ered to assess this possibility have not been con- that most hospitalized patients are at risk for
ducted. Currently, no balanced electrolyte solution euvolemic hyponatremia from AVP excess if they
Electrolyte-free
Fluid Glucose Sodium Chloride Potassium Buffer† Calcium Magnesium pH Osmolarity Osmolality Water
mOsm/ mOsm/
g/dl millimoles per liter liter kg %‡
Human plasma 0.07–0.11 135–144 95–105 3.5–5.3 23–30 2.2–2.6 0.8–1.2 7.35–7.45 308 288 0
5% Dextrose in water 5 0 0 0 0 0 0 3.5–6.5 252 100
4% Dextrose in 0.18% sa- 4 30 30 0 0 0 0 3.5–6.5 282 81
line
5% Dextrose in 0.2% saline 5 34 34 0 0 0 0 3.5–6.5 321 78
5% Dextrose in 0.45% sa- 5 77 77 0 0 0 0 3.5–6.5 406 50
The
line
5% Dextrose Ringer’s lac- 5 130 109 4 28 1.5 0 4.0–6.5 525 13
tate
5% Dextrose in 0.9% saline 5 154 154 0 0 0 0 3.5–6.5 560 0
5% Dextrose multiple elec- 5 140 98 5 50 0 1.5 4.0–6.5 547 6
trolytes injection, type 1,
USP
Ringer’s lactate 0 130 109 4 28 1.35 0 6–7.5 273 254 13
Ringer’s acetate 0 130 112 5 27 1 1 6–8 276 12
Hartmann’s solution 0 131 111 5 29 2 0 5.0–7.0 278 12
n e w e ng l a n d j o u r na l
* To convert the values for glucose to millimoles per liter, multiply by 0.05551. To convert the values for potassium to milligrams per deciliter, divide by 0.2558. To convert the values for
calcium to milligrams per deciliter, divide by 0.250. To convert the values for magnesium to milligrams per deciliter, divide by 0.4114. USP denotes United States Pharmacopeia.
† The buffer is bicarbonate in plasma, lactate in Ringer’s lactate and Hartmann’s solution, acetate in Ringer’s acetate, acetate (27 mmol per liter) and gluconate (23 mmol per liter) in
multiple electrolytes injection, type 1, USP, and acetate (23 mmol per liter) and maleate (5 mmol per liter) in isotonic electrolyte solution.
‡ This percentage is based on a sodium plus potassium concentration in the aqueous phase of plasma of 154 mmol per liter, assuming that plasma is 93% water with a plasma sodium
Downloaded from nejm.org on January 4, 2016. For personal use only. No other uses without permission.
concentration of 140 mmol per liter and a potassium concentration of 4 mmol per liter.
§ Multiple electrolytes injection, type 1, USP, is the generic name for Plasma-Lyte 148, Normosol, and Isolyte.
¶ Isotonic electrolyte solution is the generic name for Sterofundin and Ringerfundin. It has an electrolyte composition that is similar to that of plasma.
Maintenance Fluids in Acutely Ill Patients
5% dextrose in isotonic 5% dextrose in isotonic 5% dextrose in isotonic 5% dextrose in isotonic 5% dextrose in isotonic
solution solution solution solution (may need to be solution (may need to be
25 ml/hr (adults) 40–60 ml/hr (adults) ≤100–120 ml/hr (adults) adjusted to hypertonic adjusted to hypotonic
25% of amount calculated 40–60% of amount calcu- ≤100% of amount calcu- fluid to maintain plasma fluid)
with Holliday–Segar lated with Holliday– lated with Holliday– sodium level >140 mmol/ ≥120 ml/hr (adults)
formula (children) Segar formula (children) Segar formula (children) liter) ≥120% of amount cal-
100–120 ml/hr (adults) culated with Holliday–
100% of amount calcu- Segar formula
lated with Holliday– (children)
Segar formula (children)
receive hypotonic fluids. Provision of fluids at a ly addressed in this review article or in the algo-
rate greater than or less than this amount is rithm in Figure 2. A common practice is to add
determined according to whether there is a dis- 20 mmol per liter of potassium to maintenance
order in renal concentration or a fluid overload intravenous fluids, but data to support the safety
state, respectively. A hypotonic fluid may be re- of this approach are lacking, and it is not clear
quired if there is a clinically significant renal that adding more than a physiologic concentra-
concentrating defect with ongoing free-water tion of potassium (i.e., >5 mmol per liter) is nec-
losses or to aid in the correction of established essary when intravenous fluids are administered
hypernatremia.4 for less than 48 hours, unless hypokalemia or
Studies that evaluate the most appropriate malnutrition is present or diuretics are being used.
maintenance fluid for edematous states are lack-
ing. It is our opinion that an isotonic fluid ad- Sum m a r y
ministered at a restricted rate would be appro-
priate, since these patients frequently have The administration of intravenous fluids is an
hyponatremia. essential component of supportive care for
Administration of potassium is not specifical acutely ill patients. Because of limited evidence,
Box 1. Case 1.
Box 2. Case 2.
Box 3. Case 3.
recommendations regarding fluid have histori- Acutely ill patients have multiple stimuli for
cally been opinion-based. It has now become AVP that place them at risk for the development
clear that the administration of hypotonic main- of hyponatremia. Numerous prospective studies
tenance fluids (sodium concentration, <130 mmol involving children have shown that isotonic fluids
per liter) is associated with the development of are safe and effective in preventing hospital-
hospital-acquired hyponatremia as well as with acquired hyponatremia. Acutely ill patients can
deaths or permanent neurologic impairment have a variety of conditions that can alter body-
from hyponatremic encephalopathy. water homeostasis, so both the rate and the
composition of intravenous fluid should be pre- lacking. A solid evidence base from which con-
scribed carefully. Isotonic fluids are the most sensus guidelines for fluid therapy could be
appropriate maintenance fluid in the vast major- built is also needed to better standardize fluid
ity of situations. In Box 1, Box 2, and Box 3, we management.
provide three practical cases regarding the most Drs. Moritz and Ayus report receiving consulting fees from
appropriate rate and composition of mainte- Otsuka and a continuing medical education honorarium from
the Global Education Group. No other potential conflict of in-
nance intravenous fluid. terest relevant to this article was reported.
Data from studies to determine whether bal- Disclosure forms provided by the authors are available with
anced solutions are superior to saline solutions the full text of this article at NEJM.org.
We thank Karen Branstetter for editing an earlier version of
for maintenance fluids and to determine the most the the manuscript and Kelli Crowly and Kevin Ordons for as-
appropriate potassium concentration in fluids are sistance in preparing an earlier version of Table 3.
References
1. Freeman MA, Ayus JC, Moritz ML. clinical disorders of urinary dilution and Walker JL. High antidiuretic hormone lev-
Maintenance intravenous fluid prescrib- concentration. J Am Soc Nephrol 2006;17: els and hyponatremia in children with gas-
ing practices among paediatric residents. 1820-32. troenteritis. Pediatrics 2005;116:1401-7.
Acta Paediatr 2012;101(10):e465-e468. 14. Spasovski G, Vanholder R, Allolio B, 27. Aronson D, Dragu RE, Nakhoul F,
2. Davies P, Hall T, Ali T, Lakhoo K. In- et al. Clinical practice guideline on diag- et al. Hyponatremia as a complication of
travenous postoperative fluid prescrip- nosis and treatment of hyponatraemia. cardiac catheterization: a prospective
tions for children: a survey of practice. Nephrol Dial Transplant 2014;29:Suppl 2: study. Am J Kidney Dis 2002;40:940-6.
BMC Surg 2008;8:10. i1-i39. 28. Nair V, Niederman MS, Masani N,
3. Chawla G, Drummond GB. Textbook 15. Moritz ML, Ayus JC. Preventing neuro- Fishbane S. Hyponatremia in community-
coverage of a common topic: fluid man- logical complications from dysnatremias acquired pneumonia. Am J Nephrol 2007;
agement of patients after surgery. Med in children. Pediatr Nephrol 2005; 20: 27:184-90.
Educ 2008;42:613-8. 1687-700. 29. Au AK, Ray PE, McBryde KD, New-
4. Moritz ML, Ayus JC. Intravenous fluid 16. Moritz ML, Ayus JC. Hospital-acquired man KD, Weinstein SL, Bell MJ. Incidence
management for the acutely ill child. Curr hyponatremia — why are hypotonic par- of postoperative hyponatremia and com-
Opin Pediatr 2011;23:186-93. enteral fluids still being used? Nat Clin plications in critically-ill children treated
5. Intravenous fluid therapy in adults in Pract Nephrol 2007;3:374-82. with hypotonic and normotonic solutions.
hospital. London:National Institute for 17. Severs D, Hoorn EJ, Rookmaaker MB. J Pediatr 2008;152:33-8.
Health and Care Excellence, 2013 (https:// A critical appraisal of intravenous fluids: 30. Hanna M, Saberi MS. Incidence of
www.nice.org.uk/g uidance/cg174). from the physiological basis to clinical hyponatremia in children with gastroen-
6. Padhi S, Bullock I, Li L, Stroud M. In- evidence. Nephrol Dial Transplant 2015; teritis treated with hypotonic intravenous
travenous fluid therapy for adults in hos- 30:178-87. fluids. Pediatr Nephrol 2010;25:1471-5.
pital: summary of NICE guidance. BMJ 18. Knepper MA, Kwon TH, Nielsen S. 31. Rodríguez MJ, Alcaraz A, Solana MJ,
2013;347:f 7073. Molecular physiology of water balance. García A. Neurological symptoms in hos-
7. Ayus JC, Wheeler JM, Arieff AI. Post- N Engl J Med 2015;372:1349-58. pitalised patients: do we assess hypona-
operative hyponatremic encephalopathy 19. Ellison DH, Berl T. The syndrome of traemia with sufficient care? Acta Paedi-
in menstruant women. Ann Intern Med inappropriate antidiuresis. N Engl J Med atr 2014;103(1):e7-e10.
1992;117:891-7. 2007;356:2064-72. 32. National Patient Safety Agency (NPSA).
8. Halberthal M, Halperin ML, Bohn D. 20. Greenberg A, Verbalis JG, Amin AN, Patient safety alert: reducing the risk of
Lesson of the week: acute hyponatraemia et al. Current treatment practice and out- hyponatraemia when administering in-
in children admitted to hospital: retro- comes: report of the Hyponatremia Regis- travenous infusions to children. 2007
spective analysis of factors contributing try. Kidney Int 2015;88:167-77. (http://www.nrls.npsa.nhs.uk/resources/
to its development and resolution. BMJ 21. Oster JR, Singer I, Thatte L, Grant- ?entryid45=59809).
2001;322:780-2. Taylor I, Diego JM. The polyuria of solute 33. Hospital-acquired acute hyponatremia:
9. Steele A, Gowrishankar M, Abraham- diuresis. Arch Intern Med 1997;157:721-9. two reports of pediatric deaths. ISMP
son S, Mazer CD, Feldman RD, Halperin 22. Upadhyay A, Jaber BL, Madias NE. In- Canada Safety Bulletin 2009;9:1-4.
ML. Postoperative hyponatremia despite cidence and prevalence of hyponatremia. 34. Plain D5W or hypotonic saline solu-
near-isotonic saline infusion: a phenom- Am J Med 2006;119:Suppl 1:S30-S35. tions post-op could result in acute hypo-
enon of desalination. Ann Intern Med 23. Carandang F, Anglemyer A, Longhurst natremia and death in healthy children.
1997;126:20-5. CA, et al. Association between mainte- ISMP Medication Safety Alert 2009;7:1-4.
10. Arieff AI, Ayus JC, Fraser CL. Hypona- nance fluid tonicity and hospital-acquired 35. Grissinger M. Hyponatremia and
traemia and death or permanent brain hyponatremia. J Pediatr 2013;163:1646-51. death in healthy children from plain dex-
damage in healthy children. BMJ 1992; 24. Chung HM, Kluge R, Schrier RW, An- trose and hypotonic saline solutions after
304:1218-22. derson RJ. Postoperative hyponatremia: surgery. P T 2013;38:364-88.
11. Arieff AI. Hyponatremia, convulsions, a prospective study. Arch Intern Med 36. Ayus JC, Armstrong D, Arieff AI. Hy-
respiratory arrest, and permanent brain 1986;146:333-6. ponatremia with hypoxia: effects on brain
damage after elective surgery in healthy 25. Anderson RJ, Chung HM, Kluge R, adaptation, perfusion, and histology in
women. N Engl J Med 1986;314:1529-35. Schrier RW. Hyponatremia: a prospective rodents. Kidney Int 2006;69:1319-25.
12. Danziger J, Zeidel ML. Osmotic ho- analysis of its epidemiology and the 37. Moritz ML, Ayus JC. New aspects in the
meostasis. Clin J Am Soc Nephrol 2015; pathogenetic role of vasopressin. Ann In- pathogenesis, prevention, and treatment
10:852-62. tern Med 1985;102:164-8. of hyponatremic encephalopathy in chil-
13. Schrier RW. Body water homeostasis: 26. Neville KA, Verge CF, O’Meara MW, dren. Pediatr Nephrol 2010;25:1225-38.
38. Nigro N, Winzeler B, Suter-Widmer I, perin ML. How to select optimal mainte- travenous fluids in children: a random-
et al. Symptoms and characteristics of nance intravenous fluid therapy. QJM ized controlled trial. Indian Pediatr 2014;
individuals with profound hyponatremia: 2003;96:601-10. 51:969-74.
a prospective multicenter observational 52. Holliday MA, Segar WE. The mainte- 66. Bomberger RA, McGregor B, DePal-
study. J Am Geriatr Soc 2015;63:470-5. nance need for water in parenteral fluid ma RG. Optimal fluid management after
39. Ayus JC, Caputo D, Bazerque F, He- therapy. Pediatrics 1957;19:823-32. aortic reconstruction: a prospective study
guilen R, Gonzalez CD, Moritz ML. Treat- 53. Talbot NB, Crawford JD, Butler AM. of two crystalloid solutions. J Vasc Surg
ment of hyponatremic encephalopathy with Homeostatic limits to safe parenteral fluid 1986;4:164-7.
a 3% sodium chloride protocol: a case therapy. N Engl J Med 1953;248:1100-8. 67. Saba TG, Fairbairn J, Houghton F,
series. Am J Kidney Dis 2015;65:435-42. 54. Moritz ML, Ayus JC. Prevention of Laforte D, Foster BJ. A randomized con-
40. Ayus JC, Arieff AI. Chronic hypona- hospital-acquired hyponatremia: a case trolled trial of isotonic versus hypotonic
tremic encephalopathy in postmenopaus- for using isotonic saline. Pediatrics 2003; maintenance intravenous fluids in hospi-
al women: association of therapies with 111:227-30. talized children. BMC Pediatr 2011;11:82.
morbidity and mortality. JAMA 1999;281: 55. Powell-Tuck J, Gosling P, Lobo DN, 68. Coulthard MG, Cheater LS, Long DA.
2299-304. et al. British consensus guidelines on in- Perioperative fluid therapy in children. Br
41. Renneboog B, Musch W, Vandemergel travenous fluid therapy for adult surgical J Anaesth 2007;98:146-7.
X, Manto MU, Decaux G. Mild chronic patients. Redditch, United Kingdom: 69. Kannan L, Lodha R, Vivekanandhan S,
hyponatremia is associated with falls, un- BAPEN, 2011 (http://www.bapen.org.uk/ Bagga A, Kabra SK, Kabra M. Intravenous
steadiness, and attention deficits. Am J pdfs/bapen_pubs/giftasup.pdf). fluid regimen and hyponatraemia among
Med 2006;119(1):71.e1-78.e1. 56. Mythen MG, Swart M, Acheson N, children: a randomized controlled trial.
42. Ayus JC, Moritz ML. Bone disease as a et al. Perioperative fluid management: Pediatr Nephrol 2010;25:2303-9.
new complication of hyponatremia: mov- consensus statement from the enhanced 70. Neville KA, Sandeman DJ, Rubinstein
ing beyond brain injury. Clin J Am Soc recovery partnership. Perioper Med (Lond) A, Henry GM, McGlynn M, Walker JL.
Nephrol 2010;5:167-8. 2012;1:2. Prevention of hyponatremia during main-
43. Kinsella S, Moran S, Sullivan MO, 57. Drysdale SB, Coulson T, Cronin N, tenance intravenous fluid administration:
Molloy MG, Eustace JA. Hyponatremia in- et al. The impact of the National Patient a prospective randomized study of fluid
dependent of osteoporosis is associated Safety Agency intravenous fluid alert on type versus fluid rate. J Pediatr 2010;
with fracture occurrence. Clin J Am Soc iatrogenic hyponatraemia in children. Eur 156(2):313.e2-319.e2.
Nephrol 2010;5:275-80. J Pediatr 2010;169:813-7. 71. Yung M, Keeley S. Randomised con-
44. Verbalis JG, Barsony J, Sugimura Y, 58. Hatherill M. Rubbing salt in the trolled trial of intravenous maintenance
et al. Hyponatremia-induced osteoporosis. wound. Arch Dis Child 2004;89:414-8. fluids. J Paediatr Child Health 2009;45:
J Bone Miner Res 2010;25:554-63. 59. Taylor D, Durward A. Pouring salt on 9-14.
45. Ayus JC, Negri AL, Kalantar-Zadeh K, troubled waters. Arch Dis Child 2004;89: 72. Musch W, Decaux G. Treating the
Moritz ML. Is chronic hyponatremia a 411-4. syndrome of inappropriate ADH secretion
novel risk factor for hip fracture in the 60. Holliday MA, Ray PE, Friedman AL. with isotonic saline. QJM 1998;91:749-53.
elderly? Nephrol Dial Transplant 2012;27: Fluid therapy for children: facts, fashions 73. Moritz ML. Urine sodium composition
3725-31. and questions. Arch Dis Child 2007;92: in ambulatory healthy children: hypo-
46. Hoorn EJ, Zietse R. Hyponatremia and 546-50. tonic or isotonic? Pediatr Nephrol 2008;
mortality: moving beyond associations. 61. Foster BA, Tom D, Hill V. Hypotonic 23:955-7.
Am J Kidney Dis 2013;62:139-49. versus isotonic fluids in hospitalized chil- 74. Lobo DN, Awad S. Should chloride-
47. Mohan S, Gu S, Parikh A, Radhakrish- dren: a systematic review and meta-analy- rich crystalloids remain the mainstay of
nan J. Prevalence of hyponatremia and sis. J Pediatr 2014;165(1):163.e2-169.e2. fluid resuscitation to prevent ‘pre-renal’
association with mortality: results from 62. McNab S, Duke T, South M, et al. 140 acute kidney injury?: con. Kidney Int 2014;
NHANES. Am J Med 2013;126(12):1127. mmol/L of sodium versus 77 mmol/L of 86:1096-105.
e1-1137.e1. sodium in maintenance intravenous fluid 75. Moro N, Katayama Y, Igarashi T, Mori
48. Gankam-Kengne F, Ayers C, Khera A, therapy for children in hospital (PIMS): T, Kawamata T, Kojima J. Hyponatremia
de Lemos J, Maalouf NM. Mild hyponatre- a randomised controlled double-blind in patients with traumatic brain injury:
mia is associated with an increased risk trial. Lancet 2015;385:1190-7. incidence, mechanism, and response to
of death in an ambulatory setting. Kidney 63. McNab S, Ware RS, Neville KA, et al. sodium supplementation or retention ther-
Int 2013;83:700-6. Isotonic versus hypotonic solutions for apy with hydrocortisone. Surg Neurol
49. Corona G, Giuliani C, Parenti G, et al. maintenance intravenous fluid adminis- 2007;68:387-93.
Moderate hyponatremia is associated with tration in children. Cochrane Database 76. Schrier RW. Pathogenesis of sodium
increased risk of mortality: evidence from Syst Rev 2014;12:CD009457. and water retention in high-output and
a meta-analysis. PLoS One 2013; 8(12): 64. Friedman JN, Beck CE, DeGroot J, low-output cardiac failure, nephrotic syn-
e80451. Geary DF, Sklansky DJ, Freedman SB. drome, cirrhosis, and pregnancy. N Engl J
50. Amin A, Deitelzweig S, Christian R, Comparison of isotonic and hypotonic in- Med 1988;319:1127-34.
et al. Evaluation of incremental health- travenous maintenance fluids: a random- 77. Issa VS, Andrade L, Ayub-Ferreira SM,
care resource burden and readmission ized clinical trial. JAMA Pediatr 2015;169: et al. Hypertonic saline solution for pre-
rates associated with hospitalized hypo- 445-51. vention of renal dysfunction in patients
natremic patients in the US. J Hosp Med 65. Shamim A, Afzal K, Ali SM. Safety with decompensated heart failure. Int J
2012;7:634-9. and efficacy of isotonic (0.9%) vs. hypo- Cardiol 2013;167:34-40.
51. Shafiee MA, Bohn D, Hoorn EJ, Hal- tonic (0.18%) saline as maintenance in- Copyright © 2015 Massachusetts Medical Society.