Perioperative management of fluids and electrolytes in children

Learning objectives.

By reading this article, you should be able to:

• •Describe the clinical assessment of fluid status and fluid responsiveness in children.
• •Explain the clinical reasoning for the type and amount of i.v. fluids given to children during surgery.
• •Detail the reasons and benefits for promoting clear oral fluid intake in children before surgery.

Key points.

• •Fluid deficits are usually minimal when the intake of clear oral fluid is encouraged in healthy children until 1 h before elective surgery.
• •The assessment of fluid status and responsiveness is imprecise. A change in body weight is the most accurate measure of acute changes to fluid status.
• •Routine use of hypotonic maintenance fluids significantly increases the risk of iatrogenic hyponatraemia compared with isotonic solutions.
• •Balanced isotonic solutions should be used for maintenance fluid therapy during surgery in children. Glucose 1–2% may be added if there is risk of intraoperative hypoglycaemia.
• •Hyponatraemic encephalopathy is a rare but life-threatening condition requiring prompt treatment.

Fluid and electrolyte management is a component of the care provided to every child undergoing anaesthesia and surgery. Concepts surrounding fluid and electrolyte management, maintenance fluid and volume resuscitation have evolved over time, arising as much from first principles as they have from scientific evidence. The everyday familiarity of the subject combined with an incomplete evidence base has created an environment for ongoing debate and conjecture for the optimal fluid management for a given patient, with a given pathology, undergoing a given surgery. This is reflected in the wide variation in practice observed.¹

Fluid status depends on the complex interplay between physiology, pathology, surgery and the fluid therapies given to a patient. The aim for the anaesthetist is to assess and compensate for any perturbations to fluid homeostasis to maintain tissue perfusion, cellular metabolism and acid–base status within acceptable physiological limits. The major developments in perioperative fluid management in children over the past 20 yrs have been the liberalisation of fluid intake before surgery and the recognition of iatrogenic hyponatraemic encephalopathy secondary to the use of hypotonic fluids. This paper reviews our current state of knowledge for intraoperative fluid management in infants and children.

Physiology

Body composition

Body fluid compartments are collections of fluid with similar compositions and physiological roles. Total body water is subdivided into intracellular and extracellular fluid compartments, with the extracellular compartment further divided into interstitial and plasma volumes. Cell membranes separate the intracellular and extracellular compartments. These compartments and volumes behave predictably when fluids are given. I.V. free water (e.g. dextrose 5% or 10%) distributes evenly across total body water, isotonic crystalloid solutions distribute within the extracellular compartment and colloid solutions have a greater tendency to expand the plasma volume in the presence of an intact endothelial glycocalyx layer.

Although the electrolyte compositions within the body fluid compartments are the same throughout life and development, the relative size of the compartments varies with age. Most of total body water is extracellular in the fetus, and in newborn premature babies the extracellular compartment comprises 60–70% of total body weight. The extracellular compartment reduces to 40% of total body water in term babies and 30% in older infants.

I.V. fluids

Crystalloids

Crystalloids are solutions containing electrolytes and other low-molecular-weight substances. Fluids with ionic concentrations resembling extracellular fluid are called balanced solutions. Organic anions are added to balanced solutions to reduce chloride concentrations and the risk of iatrogenic hyperchloraemic acidosis. Lactate serves this purpose in lactated Ringer's solution, also acting as a buffer and substrate for hepatic bicarbonate synthesis. Acetate and gluconate serve a similar purpose in Plasma-Lyte solution.

Colloids

Colloids contain homogeneous non-crystalline particles suspended in balanced crystalloid or isotonic saline. These particles tend to remain within the bloodstream after infusion, thus increasing intravascular oncotic pressure and expanding plasma volume. Colloids are either artificial (hydroxyethyl starch [HES], gelatins and dextrans) or derived from donated blood (albumin and fractionated blood products). Although there are no designated clinical studies to guide the choice between albumin, HES or gelatin solutions in children, comment can still be made about safety profiles and common clinical practice.²

Dextrans are generally avoided, having a poor safety profile with the potential to cause iatrogenic coagulopathy and anaphylaxis.³ Gelatin solutions are also allergenic in adults.^2,3 This finding has not been specifically reproduced in paediatric studies, but this potential risk has made other solutions preferable. Studies of critically ill adults have repeatedly highlighted the risk of renal failure associated with HES infusions in patients with sepsis, but this finding has also not been reproduced in children. In an observational study of 1,130 children aged up to 12 yrs, no serious or severe adverse drug reactions were attributable to HES therapy.⁴ Specifically, there were no instances of HES-induced renal failure in this study.⁴ Caution should still be exercised surrounding the use of HES solutions in children, especially in those with renal impairment.

Albumin is the most prevalent plasma protein and is the primary determinant of plasma oncotic pressure. Albumin solutions are available in concentrations of 4%, 5% and 20–25%. Albumin solutions are less allergenic than HES but are more expensive to produce and store. Over the past 30 yrs, this balance between safety and expense has generated conjecture regarding the optimal colloid for rapid intraoperative volume expansion. Albumin and HES have been shown to be equivalent volume expanders in children aged 2–12 yrs undergoing surgery for congenital heart disease with no differences observed in haemodynamics, use of vasoactive or inotropic drugs, blood loss, transfusion or renal impairment.⁵ As discussed previously, the safety concerns surrounding HES and renal dysfunction have not been clearly demonstrated in children. Current practice is therefore dictated by national, institutional and practitioners' preferences.

Fractionated blood products are excellent plasma volume expanders and are indicated for the correction of anaemia and coagulopathy, which are often concurrent with volume depletion during haemorrhage. With the exception of albumin, their use is not indicated for volume expansion alone.

Preoperative fluid management

Preoperative fasting

Fasting is essential, as it decreases the risk of regurgitation; aspiration; and the associated life-threatening sequelae of airway obstruction, chemical pneumonitis and bacterial pneumonia. In paediatrics, the risk of aspiration is low (1–10:10,000), and morbidity and mortality are extremely rare.^6,7 This contrasts with adult patients who experience significantly increased morbidity and mortality from aspiration events under anaesthesia. This difference is likely in part to be attributable to rapid gastric emptying in healthy children, which is demonstrable with ultrasonography and MRI.⁷ Gastric emptying for solids follows zero-order kinetics, whereas it follows first-order kinetics for fluids.⁷ In practice, this means that solids are emptied at a constant rate from the stomach, whereas fluids initially empty very rapidly, then follow an exponential elimination curve. For solids, gastric antral area is smaller 4 h after a light breakfast than after an overnight fast.⁸ Gastric volumes essentially return to baseline concentrations in under an hour after drinking 5 ml kg⁻¹ of clear fluid.⁹ Fasting before surgery has the potential to induce dehydration and a number of neurohumoral and metabolic responses, resulting in ketosis and catabolism (glycogenolysis, lipolysis and proteolysis).¹⁰ Prolonged fasting can cause hunger, thirst, anxiety and distress. This adversely affects patient and guardian experiences.¹¹ International guidelines vary and debate is ongoing, but there is a trend towards increasing liberalisation of clear oral fluid intake before surgery. The Association of Paediatric Anaesthetists of Great Britain and Ireland, Australian and New Zealand College of Anaesthetists, European Society of Anaesthesiology and Intensive Care, European Society for Paediatric Anaesthesiology and Association Des Anesthésistes Réanimateurs Pédiatriques d'Expression Franèaise all recommend clear oral fluid fasting times of 1 h before elective surgery. Promoting clear oral fluid intake before surgery may also improve intraoperative haemodynamic stability and reduce nausea and delirium after surgery.¹²

Preoperative fluid deficits

Euvolaemia is expected in the elective surgical setting when clear oral fluids are promoted before surgery. For children undergoing emergency surgery, there may be significant fluid deficits caused by burns, fever, sepsis, bleeding or gastrointestinal losses (vomiting, gastric tube drainage and diarrhoea). Clinical signs and symptoms are an imprecise method for quantifying fluid deficits, although an approximation of the severity of dehydration can be estimated at the bedside (Table 1). The most precise means of quantifying dehydration is acute weight loss, which is often not attainable, as it requires an accurate pre-illness weight for comparison. Hypotension is a late and ominous sign of severe dehydration because of the impressive ability of children to compensate until physiological extremes are reached.

Table 1.

Assessment of dehydration.

Signs and symptoms	Mild	Moderate	Severe
Weight loss (%)	5	10	15
Deficit (ml kg−1)	50	100	150
BP	Normal	Low normal	Low
Pulse	Normal	Rapid	Rapid and weak
Respiration	Normal	Tachypnoea	Deep
Mucous membranes	Moist	Dry	Very dry
Skin	Normal	Prolonged capillary refill	Prolonged capillary refill
Anterior fontanelle	Normal	Sunken	Very sunken
Urine output (ml kg−1 h−1)	<2	<1	<0.5
Appearance	Alert and thirsty		Irritable and lethargic

Intraoperative fluid management

Maintenance infusions

The goal of giving fluid during surgery is to provide water for basal metabolism and to maintain circulating volume and electrolyte homeostasis. Balanced isotonic crystalloids (e.g. Plasma-Lyte) are the first-line fluids for intraoperative maintenance and volume replacement. In infants and children beyond 4 weeks of age, the National Institute for Health and Care Excellence (NICE) guidelines recommend using the Holliday and Segar equation to calculate basal daily fluid requirements (Table 2).¹³ This formula gives a reasonable starting point for intraoperative fluid prescription, but adjustments need to be made to replace the additional obligatory fluid losses of surgery. Examples of these losses include evaporation from open-cavity surgery, burns, fever and sepsis, gastrointestinal losses and bleeding. As a guide, insensible losses for superficial surgery amount to 1–2 ml kg⁻¹ h⁻¹ and typically do not require special replacement. Insensible losses during thoracotomy amount to 4–7 ml kg⁻¹ h⁻¹ and increase to 5–10 ml kg⁻¹ h⁻¹ during open abdominal surgery.¹⁴ To illustrate, a 30 kg child undergoing a peripheral superficial surgery will require 70 ml h⁻¹ maintenance fluids. The same child undergoing open abdominal surgery may require 70 ml h⁻¹ for maintenance and an additional 150–300 ml h⁻¹ to replace insensible losses, totalling 220–370 ml h⁻¹.

Table 2.

Holliday and Segar 4/2/1 calculation for basal fluid requirements.¹⁵

Weight (kg)	Fluids per hour	Fluids per day
0–10	4 ml kg−1 h−1	100 ml kg−1 day−1
10–20	40 ml h−1+2 ml kg−1 h−1 for each kg >10 kg	1,000 ml day−1+50 ml kg−1 day−1 for each kg >10 kg
>20	60 ml h−1+1 ml kg−1 h−1 for each kg >20 kg	1,500 ml day−1+20 ml kg−1 day−1 for each kg >20 kg

An alternative approach is to provide 10 ml kg⁻¹ h⁻¹ of balanced isotonic solution.² This tends to provide both maintenance fluid and replacement of insensible losses in one calculation. For our 30 kg child, this would equate to 300 ml h⁻¹. This infusion rate is likely to be appropriate for the child undergoing abdominal surgery but significantly more than what is required for a euvolaemic child undergoing a superficial peripheral surgery with minimal insensible losses.

Fluid responsiveness and volume replacement

Fluid responsiveness describes the ability of i.v. fluids to improve cardiovascular function, tissue perfusion and metabolic physiology. Infused fluids aim to increase preload, improve cardiac output, reduce peripheral vasoconstriction and optimise blood rheology. Autotransfusion manoeuvres are the most reliable means of predicting fluid responsiveness. These include passive leg raise (in children >5 yrs old), Trendelenburg positioning and external hepatic compression.^2,16 Autotransfusion dynamically alters central volume and cardiac preload, simulating the effect of an i.v. fluid bolus. Positioning and draping the patient usually prevent these manoeuvres being performed; therefore, determining fluid responsiveness is more often a dynamic clinical assessment based on response to treatment. Judicious i.v. ‘fluid challenges’ (10–20 ml kg⁻¹ balanced isotonic crystalloid or 5–10 ml kg⁻¹ colloid solution) are given whilst monitoring for favourable trends in physiological endpoints (BP, HR, Pe´co₂, urine output and skin perfusion) and point-of-care testing (pH, base excess, lactate, haemoglobin and echocardiography).

The static assessment of fluid status from clinical signs is unreliable for predicting fluid responsiveness. Variables, such as HR, systolic arterial pressure, central venous pressure and pulmonary artery occlusion pressure, have all been shown to have low predictive value for determining the usefulness of i.v. fluids to improve haemodynamics in children.¹⁷ Unfortunately, dynamic variables, such as pulse pressure variation, stroke volume variation, respiratory variation in inferior vena cava diameter, oesophageal Doppler indices and plethysmographic waveform techniques, also have inconclusive or conflicting evidence, rendering their clinical usefulness uncertain.16, 17, 18, 19

Large volume infusion of crystalloids during surgery (>40–60 ml kg⁻¹) may induce endothelial dysfunction, increased transfer of fluid to the interstitium and interstitial fluid overload. This reduces the efficacy of further isotonic fluid in bolstering plasma volume and replacing fluid deficits. Some consensus guidelines consider colloid solutions as second-line fluids for volume replacement in children in this setting, although there is insufficient evidence to definitively guide this practice.² Furthermore, dilutional anaemia may manifest after the infusion of large volumes of crystalloids. The use of red cell transfusion in this setting is a common strategy to mitigate this risk. Colloid solutions may be given in repeated doses of 5–10 ml kg⁻¹ (up to a daily maximum of 30 ml kg⁻¹ for HES 130).

Care must be exercised with intraoperative fluid management in the presence of hyponatraemia or hypernatraemia. Rapid correction of plasma sodium concentration may be dangerous, resulting in pontine demyelination or cerebral oedema. For the safe correction of sodium abnormalities, refer to the NICE guideline, ‘Algorithms for IV fluid therapy in children and young people in hospital’.¹³

Glucose

Glucose supplementation is not usually necessary for healthy children undergoing surgery. Beyond the neonatal period, children have sufficient metabolic and energy reserves to maintain normoglycaemia for the duration of most surgeries. Exceptions to this include children with malnutrition, hypermetabolism, liver failure, hypothermia, mitochondrial disorders, beta-blocker therapy or critical illness. In long surgery where glucose is not infused, blood sugar concentrations should be monitored closely.

Where glucose is provided, the addition of 1–2% glucose to intraoperative maintenance fluid is sufficient to prevent hypoglycaemia and ketosis, without causing hyperglycaemia.²⁰ Routine use of 5% glucose is associated with hyperglycaemia.²⁰ Nishina and colleagues compared three regimens of glucose administration in ASA 1 infants aged 1–11 months undergoing surgery. Children were provided maintenance fluid with either 1%, 2% or 5% glucose. No children experienced hypoglycaemia (plasma glucose concentration < 2.8 mmol L⁻¹) in any arm of the study. In the 5% glucose group, 30% of children had a plasma glucose concentration > 11 mmol L⁻¹ at the end of surgery.²⁰ Hyperglycaemia at this concentration results in glycosuria, osmotic diuresis and dehydration.

It is important to note that supplementation of maintenance fluid with glucose 1–2% is not a nutritional replacement, providing only 5–10% of basal metabolic energy requirements. Table 3 provides a guide for bedside formulation of glucose 1% and 2% solutions.

Table 3.

Glucose additives to create 1% and 2% solutions.

Glucose	Volume
	100 ml	500 ml	1,000 ml
1%	+ 2 ml glucose 50%	+ 10 ml glucose 50%	+ 20 ml glucose 50%
2%	+ 4 ml glucose 50%	+ 20 ml glucose 50%	+ 40 ml glucose 50%

Postoperative management

Early return to oral intake and enteral hydration should be encouraged wherever possible after surgery. I.V. fluid therapy is indicated when enteral hydration is insufficient to maintain fluid and electrolyte homeostasis. In our institution, the isotonic balanced salt solution Plasma-Lyte+dextrose 5% is the preferred fluid for this purpose, with maintenance rates calculated using the Holliday and Segar 4/2/1 formula (Table 2). After major surgery, given rates should be reduced by 30–50% to account for increased antidiuretic hormone (ADH) secretion and to mitigate the risk of hyponatraemia. Plasma electrolyte concentrations should be tested at least daily to monitor for electrolyte abnormalities during i.v. fluid infusions. Ongoing and careful clinical assessments of volume status, fluid losses and electrolyte status are imperative, with fluid administration rate, fluid type and additives titrated to requirements. Total duration of i.v. fluid therapy should be as short as possible, with nutritional requirements and potential need for total parenteral nutrition considered if enteral intake is not re-established within days of surgery.

Iatrogenic harms

Postoperative hyponatraemia

In 1957, Holliday and Segar described the fluid, electrolyte and energy requirements for well children.¹⁵ Water requirements were derived by balancing fluid to energy consumption at a ratio of 1 ml:100 kcal. Electrolyte requirements were derived from the electrolyte content of human and cow's milk.¹⁵ For 40 yrs, this paper shaped paediatric clinical practice, leading to the widespread use of hypotonic i.v. fluid therapy. In 1992, a series of 16 children with cerebral oedema (of which seven died) generated critical review of the use of hypotonic fluid in the management of acutely unwell children.²¹

Acute illness, hypovolaemia, nausea, pain, anxiety, vagal stimulation, opioid analgesia and surgery all stimulate ADH release from the posterior pituitary, which reduces water excretion from the body.²² Giving hypotonic fluids in the presence of increased ADH secretion creates a surplus of non-cleared free water. Hyponatraemia may be further exacerbated when combined with pathological extrarenal sodium loss (e.g. diaphoresis and gastrointestinal losses). Multiple studies have since corroborated the association between hypotonic fluids and hyponatraemia in paediatric patients.^23,24

Hyponatraemia causes osmotic stress to neurones and glial cells. This osmotic stress may progress to hyponatraemic encephalopathy, a life-threatening emergency involving cerebral oedema and intracranial hypertension.²¹ Hyponatraemic encephalopathy may develop extremely rapidly and be irreversible, leading to death from respiratory arrest even when symptoms are initially mild. Early treatment provides the best opportunity for a favourable outcome, but early detection of hyponatraemic encephalopathy is difficult. Early symptoms are non-specific and common after surgery. Headache, nausea, vomiting, lethargy and fussiness are at times ubiquitous in the surgical ward. With early detection difficult and consequences severe, prevention of hyponatraemic encephalopathy is therefore imperative.

The routine use of isotonic fluids for maintenance therapy significantly reduces the risk of hyponatraemia. In a Cochrane systematic review, McNab and colleagues combined data for 529 children from seven studies to determine a relative risk of 0.48 (95% confidence interval: 0.36–0.64) for hyponatraemia (<135 mmol L⁻¹) when comparing isotonic with hypotonic fluids for paediatric patients who undergo surgery.²⁵ The 2018 American Academy of Pediatrics Clinical Practice Guideline: Maintenance Intravenous Fluids in Children amassed primary outcome data from 2,313 children in 17 RCTs to estimate the number needed to treat (NNT) with isotonic fluids.¹ The NNT to prevent hyponatraemia (<135 mmol L⁻¹) was 7.5 and 27.8 for moderate hyponatraemia (<130 mmol L⁻¹).¹

The risk of hyponatraemia is increased in children with heart, liver or renal disease or humoral disorders of sodium and water homeostasis, such as syndrome of inappropriate secretion of antidiuretic hormone or adrenal insufficiency. Children with these conditions have been readily excluded from trials assessing the safety and efficacy of isotonic fluid in the mitigation of iatrogenic hyponatraemia. Clinical practice must therefore be guided by careful and considerate prescribing rather than robust trial data for the time being.

Treatment of hyponatraemic encephalopathy

Hyponatraemic encephalopathy is an emergency. Anticonvulsants are ineffective in the treatment of hyponatraemic seizures, and plasma sodium concentrations ([Na⁺]_plasma) must be rapidly increased. Seizures usually discontinue once [Na⁺]_plasma >125 mmol L⁻¹. A sodium chloride 3% solution contains sodium 500 mmol L⁻¹, and a dose of 1 ml kg⁻¹ will increase [Na⁺]_plasma by ∼1 mmol L⁻¹. A 3–5 ml kg⁻¹ dose of sodium chloride 3% solution should be given over 15–30 min with repeated doses until [Na⁺]_plasma >125 mmol L⁻¹ or seizures terminate. Acute normalisation of [Na⁺]_plasma is not necessary, as rapidly increasing [Na⁺]_plasma may cause demyelination. [Na⁺]_plasma measurement between boluses is therefore necessary to ensure correction does not exceed 2 mmol L⁻¹ h⁻¹ during the first 3–4 h.

Hypernatraemia, fluid overload and hyperchloraemic acidosis

The risk of sodium overload has been debated with the paradigm shift from hypotonic to isotonic fluid therapy in children. Maintenance fluid infusions using isotonic solutions deliver sodium doses beyond normal daily physiological requirements. There is a paucity of data on this issue, as outcome data have been limited in meta-analyses comparing the use of hypotonic and isotonic fluids.²⁵ McNab and colleagues conducted a systematic review of nine studies with data from 937 children and determined that isotonic fluid use had no effect on rates of hypernatraemia.²⁵ Data were of low quality, and the true risk of hypernatraemia (if present) is yet to be quantified. The risk of fluid overload when isotonic fluid is used instead of hypotonic fluid has similarly poor data. The risks of hyperchloraemia and hyperchloraemic acidosis are greatest when delivering supra-physiological concentrations of chloride ions in sodium chloride 0.9% solutions.²⁶

Summary

Preoperative fluid deficits in healthy children undergoing elective surgery are usually insignificant and do not require replacement. Intraoperative fluids are given to provide water for basal metabolism and to replace insensible and obligatory fluid losses from surgery. Balanced isotonic fluids are the i.v. fluids of choice for both maintenance infusions and initial volume resuscitation in paediatrics. Hypotonic i.v. fluids significantly increase the risk of iatrogenic hyponatraemia, exposing children to the life-threatening risk of hyponatraemic encephalopathy. Glucose supplementation may be justifiably added or omitted from maintenance fluids dependent on patient characteristics and duration of surgery. When indicated, glucose at a concentration of 1–2% added to balanced isotonic fluid is sufficient to prevent intraoperative hypoglycaemia. Accurate assessment of fluid status and fluid responsiveness is difficult in children, and fluid therapy should always be judicious.

Declaration of interests

The authors declare that they have no conflicts of interest.

MCQs

The associated MCQs (to support CME/CPD activity) will be accessible at www.bjaed.org/cme/home by subscribers to BJA Education.

Acknowledgements

The authors would like to extend special thanks to their colleagues who provided proofreading and writing assistance for this document: Dr Alexandra Cardinal, Dr Helena Stone, Dr Tom Burrows, Dr Ewa Johannsen and Dr Hugh Douglas.

Biographies

Nick Eaddy BSc FANZCA is a consultant paediatric anaesthetist at Waikato Hospital, New Zealand. He is an author for the Oxford Handbook of Anaesthesia. He has interests in fluid management, quality improvement and medical education.

Caleb Watene is a senior registrar and advanced Australian and New Zealand College of Anaesthetists trainee at Waikato Hospital.

Matrix codes: 1A01, 2A05, 3D00