1

Indications for perioperative intravenous fluid therapy

Ideally, a healthy child starts with undisturbed macrohaemodynamics and microcirculation, normal metabolic conditions, and a normal acid-base balance into an elective operation. The main goal of perioperative fluid therapy is to maintain this homeostasis. This delicate balance can be impaired due to prolonged fasting, intraoperative loss, shifting into the interstitium, or critical illness. In these cases, perioperative fluid therapy can help to preserve or restore homeostasis and to ensure optimal tissue perfusion. However, intravenous fluids should be considered as a medication and therefore the indication for intravenous fluid therapy must be considered carefully. Healthy children undergoing sedation for diagnostic procedures or anaesthesia for minor surgery with optimised preoperative fasting times and immediate postoperative fluid intake may not necessarily benefit from perioperative intravenous fluids. The omission of infusion therapy under the above conditions does not lead to negative effects such as hypotension, hypovolemia, hypoglycaemia, or a catabolic metabolic state [ ]. Whenever intravenous fluid therapy is indicated, the type of fluid should not interfere with homeostasis. Although there is increasing scientific evidence on the most appropriate type and amount of intravenous fluid, the choice is also based on physiological considerations on maintaining fluid, electrolyte, acid-base, and glucose balance.

2

Intraoperative maintenance infusion

For many decades, children’s fluid requirements have been calculated according to the weight-based 4-2-1 formula of Holliday and Segar [ ] ( Table 1 ). The formula is an estimate of daily energy expenditure and corresponding fluid requirements.

Holliday and Segar also calculated electrolyte requirements based on the composition of human and cow’s milk, which led to the use of hypotonic maintenance infusion in children. In the 1990s, the perioperative use of hypotonic solutions was increasingly questioned. Case reports and case series of hyponatraemia resulting in permanent brain damage or death due to the perioperative use of hypotonic infusions have been published [ ]. While hypotonic solutions match the requirements of children who are not acutely ill, in critically ill children or during perioperative stress, increased antidiuretic hormone (ADH) secretion leads to water retention and, in conjunction with the use of hypotonic infusion solutions, promotes hyponatraemia. Perioperative intravascular volume depletion is the most potent stimulus of ADH release. Conditions such as hypoxia, hypercapnia, drugs, pain, inflammation, surgical stress, sepsis, or organ dysfunction can potentiate the secretion of ADH, leading to the manifestation of the syndrome of inappropriate antidiuretic hormone secretion (SIADH). The risk of hyponatraemic encephalopathy is particularly high for prepubescent children due to their increased brain-size-to-cranial-vault ratio, decreased Na–K ATPase activity, and increased ADH levels in response to stress [ ]. In the meantime, numerous studies and meta-analyses have linked the use of hypotonic infusions in critical ill or perioperative children to the occurrence of hyponatraemia. International guidelines for fluid therapy in critically ill children recommend the use of isotonic solutions for maintenance fluid therapy [ ]. In 2014, a Cochrane review of 10 studies with 1106 children compared isotonic and hypotonic solutions as maintenance intravenous fluids. The risk of hyponatraemia was decreased by 52% in the group with the isotonic solution [ ]. In a prospective RCT by McNab the use of isotonic intravenous fluid with a sodium concentration of 140 mmol/l had a lower risk of hyponatraemia (4%) without an increase in adverse effects than did fluid containing 77 mmol/l of sodium (11%). This was accompanied by a decreased risk of hyponatraemic encephalopathy in hospitalised children [ ]. An updated systematic review and meta-analysis recently proposed the use of balanced isotonic maintenance fluid therapy in hospitalised children (even in neonates if the risk of hypernatraemia is not important) as it significantly reduced the risk of iatrogenic hyponatraemia and was better tolerated by the kidneys than 0.9% saline solution [ ].

Traditionally, 5% glucose is added to paediatric maintenance infusions. In the perioperative context, this can lead to hyperglycaemia [ ]. Surgical stress increases plasma counter-insulin hormone levels (e.g. cortisol, glucagon, epinephrine, and growth hormone) and decreases plasma insulin levels, which leads to a hyperglycaemia-induced catabolic state [ , ]. However, children in a catabolic state or with low glycogen reserves such as infants, children with liver disease, malnourishment or burns and children with prolonged fasting times are in danger of hypoglycaemia. A multicentre study recently defined intraoperative hypoglycaemia in children as blood glucose values less than 3.3 mmol/l (60 mg/dL) [ ]. Anaesthesia can mask typical symptoms such as tremor, weakness, fainting, change in mentation, and seizures. The reported incidence of intraoperative hypoglycaemia in children ranges from 0.7% to 28%. Its consequences can be neurologically devastating [ , ]. Even with still low normal plasma glucose concentrations, lipolysis, reduced base excess, elevated levels of ketone bodies and free fatty acids can occur [ , ]. Therefore, adding glucose to maintenance infusion is recommended for new-borns and infants, and at-risk populations prone to hypoglycaemia [ , ] ( Table 2 ). A glucose concentration of 1–2.5% is adequate to prevent lipolysis and hypoglycaemia without the risk of leading to hyperglycaemia [ , ]. Blood glucose should be routinely monitored in at-risk children or those who undergo prolonged surgical procedures. Glucose intake should be adjusted accordingly.

Starting the perioperative maintenance infusion with an infusion rate of 10 ml/kg/h can account for deficits due to fasting and assure a stable homeostasis [ , ]. There is also moderate-certainty evidence from a Cochrane analysis in children older than six months that maintenance infusion rates of 30 ml/kg/h reduce postoperative nausea and vomiting in ASA class I to II patients receiving general anaesthesia for ambulatory or short length of stay surgical procedures [ ]. However, after 1–2 h and especially in children with relevant fluid deficits or overload, the maintenance infusion rate should be adjusted to the actual requirements. The 4-2-1 formula by Holliday and Segar can be used as an estimate for dosing of the maintenance infusion. Fluid deficits beyond normal fasting (e.g. gastroenteritis, hypertrophic pyloric stenosis) should be compensated for before anaesthesia begins [ ].

3

Physiological considerations for perioperative fluids

The first solutions used to replace fluid losses were isotonic saline and Ringer’s solution, which was first described in the 19th century by Sydney Ringer. In 1929, Alexis Hartmann and his colleagues from St. Louis suggested using a solution with sodium lactate to prevent metabolic acidosis [ ]. The well-known Ringer’s solution was modified by adding lactate. Since then, infusions have been further adapted to better preserve homeostasis. The following requirements now apply to perioperative fluids for paediatric patients:

Isosmotic/-tonic: The effective osmolality of an infusion should be similar to plasma (288 mOsmol/kg H ₂ O). Due to absorption on proteins and cell membranes, a part of the infused electrolytes is not osmotically effective. The theoretical osmolarity (mOsmol/L), which is the sum of all osmotically effective particles in a solution, then increases by the osmotic coefficient. Tonicity or effective (= in vivo) osmolality is the osmotic effect of a solution in vivo and incorporates only osmotically effective particles that do not cross cellular membranes. Due to its plasma-adapted electrolyte content, tonicity is primarily determined by Na ⁺ concentration, whereas glucose that is rapidly metabolised does not contribute to effective osmolality. Isotonic means a solution with an effective osmolality similar to plasma and therefore no alteration of the cell volume. A solution is considered hypotonic if the cell swells, with a net movement of water from the solution into the cell, and hypertonic if the cell shrinks, with a net movement of water out of the cell [ ].

Isoionic: Fluids should contain physiological values of the most important plasma electrolytes: sodium (140 ± 5 mmol/l), potassium (4.5 ± 0.5 mmol/l), calcium (2.5 ± 0.5 mmol/l), and chloride (103 ± 3 mmol/l)] with corresponding concentrations of anions and cations.

Isohydric = balanced (potential base excess of 0 ± 10 mmol/l): Metabolisable bases, that is anions of organic acids such as acetate, lactate, gluconate, malate or hydrogen malate and citrate, which are converted into bicarbonate in the intact liver (mainly lactate) or muscles (mainly acetate + malate) result in greater stability in the acid-base balance. The potential base excess (BE _pot ) of a fluid describes the sum of hydrogen ions infused with the mostly acidic electrolyte solutions and the hydrogen ions consumed by the complete metabolisation of the organic anions in the electrolyte solution. An electrolyte solution with a BE _pot of 0 mmol/l has no effect on the acid-base balance, whereas a BE _pot of less than 0 mmol/l has an acidotic effect and a BE _pot of more than 0 mmol/l has an alkalotic effect [ ]. Unfortunately, the BE _pot is not declared for most infusion solutions and the changes in the acid-base balance during infusion therapy can only be monitored by blood gas analysis. Compared to lactate, acetate metabolism is significantly faster and more independent of hepatic function, with a lower increase in oxygen consumption and no interference with the diagnostic use of lactate as a marker of inadequate tissue perfusion. Using balanced solutions with metabolisable anions prevents acidosis due to dilution of the extracellular bicarbonate pool by bicarbonate-free infusion [ ].

To summarise, balanced crystalloid solutions (BCS) with a physiological electrolyte pattern (isoionic), an effective osmolality of approximately 288 mosmol/kg H2O (isotonic and isosmotic), and a bicarbonate precursor (e.g. acetate, malate) to achieve a potential base excess of approximately 0 mmol/l (isohydric) are recommended during the perioperative period from a physiological point of view.

4

Perioperative fluid replacement

In addition to maintenance fluid requirements, pathological fluid losses must be compensated perioperatively. The aim of perioperative fluid replacement is to maintain normovolaemia, stabilise circulatory function in case of anaesthetic-induced reduced vascular tone, preserve extracellular volume, and ensure an optimised oxygen supply to all tissues. Intravascular hypervolaemia and inflammation pose a risk of degradation to the endothelial surface layer (ESL), a layer composed of glycocalyx (proteoglycans + glycosaminoglycans) and plasma proteins crucial for the integrity of the vascular barrier. The degradation leads to a loss of barrier function and an increase in vascular permeability, allowing for an augmented influx of both crystalloid and colloid solutions into the interstitium [ ]. Therefore, hypervolaemia must be avoided. Perioperative fluid requirements are based on the duration and extent of the operative procedure as well as comorbidities (e.g. cardiac or renal insufficiency)–it can often only be estimated. Clinical parameters, standard monitoring, and blood gas parameters are frequently used in everyday clinical practice to control the fluid balance. Despite apparently normal parameters, such as a normal blood pressure and heart rate, tissue perfusion may already be impaired [ , ]. Therefore, in children with heart disease or major surgery with high fluid turnover, the use of extended haemodynamic monitoring can be considered ( Table 3 ).

The actual cardiac index can only be estimated imprecisely based on clinical parameters and without the use of additional invasive catheters [ , ]. Dynamic preload parameters are based on respiratory variations of arterial waveform (pulse pressure variation (PPV)), pulse oximetry (pleth variability index (PVI)), or ultrasound (aortic peak blood flow velocity, inferior vena cava diameter). While reliability of PPV is questionable even with high tidal volumes of 10 ml/kg or more, PVI and ultrasound measurements seem to be more useful for prediction of fluid responsiveness [ ]. Although the prediction of fluid responsiveness is an essential component in perioperative fluid management, fluid responsiveness is not the same as actual fluid requirement. Ultimately, not the optimisation of cardiac output but sufficient organ and tissue perfusion are the goals of perioperative fluid substitution. Near-infrared spectroscopy (NIRS) provides information on the ratio of oxygen supply and consumption in the tissue [ ]. It is mainly used to measure cerebral tissue oxygenation, which is actually the venous oxygen saturation due to the large venous blood volume in the brain [ ]. As no standard values or lower limit values for NIRS in children have yet been established in studies, a strategy involving measurement of a baseline value in an awake child in room air can be recommended [ , , ]. The NIRS value is influenced by heart rate, blood pressure, arterial oxygen saturation, carbon dioxide partial pressure, and haemoglobin concentration [ ]. These would be the target parameters that need to be optimised if the NIRS falls below the baseline value.

Finally, the combination of the various parameters to form an overall picture and the final estimation of fluid requirements depend on the experience of the attending anaesthesiologist. In any case, during perioperative fluid management, constant attention must be paid to signs of lack of volume responsiveness and hypervolaemia. It is recommended to start with a fluid bolus of 10 ml/kg and to repeat boluses depending on clinical and monitoring parameters with frequent re-evaluation throughout [ ].

Unfortunately, there is no ideal solution for perioperative fluid replacement that has the same composition as extracellular fluid, sustainably improves intravascular volume status, does not interfere with homeostasis, and does not accumulate in the body. Crystalloid and colloidal infusion solutions are available as options. Primarily, crystalloid infusions are used for perioperative fluid replacement. In many countries, 0.9% saline solution is most common owing to its ready accessibility and economical nature.

5

Saline and hyperchloremic acidosis

The first scientific support for the use of a 0.9% saline solution was the in vitro study conducted by Hamburger [ ]. Since the 1970s, it has been recognised that a 0.9% saline solution is neither “normal” nor “physiological” [ ]. Drawbacks are the high chloride content and the lack of a bicarbonate precursor. In healthy volunteers, this was associated with abdominal discomfort, pain, nausea, drowsiness, and decreased mental capacity to perform complex tasks [ ]. The consequences of an unphysiologically high chloride content (i.e., 154 instead of 103 mmol/l in 0.9 % saline solution) are renal vasoconstriction, reduced renal blood flow, reduced glomerular filtration rate and diuresis, suppression of the renin-aldosterone system and even hyperchloremia and dilutional acidosis [ ]. By inducing cytosolic acidification, membrane hyperpolarisation, inactivation of protein kinases, and disruption of phosphorylation, cellular dysfunction can occur [ ]. In RCTs and meta-analysis the use of BCS shows less postoperative chloride increase, decrease in base excess and frequency of hyperchloremic acidosis compared to 0.9% saline solution. These are better suited to perioperative maintenance of homeostasis [ , ]. A 2022 meta-analysis of three RCTs with 162 critically ill paediatric patients showed that metabolic acidosis and bicarbonate levels improved after 4–12 h of hydration with a BCS compared with a 0.9% saline solution [ ]. In critically ill children, hyperchloremia seems to have a negative effect on outcome. It was associated with increased mortality in the paediatric intensive care setting [ , ]. However, one recent study in critically ill adult patients with traumatic brain injuries showed a slight increase in mortality under balanced solutions [ ]. There is mixed evidence on the association of hyperchloremia and renal failure. Large multicentre studies showed no evidence for a protective effect of BCS vs. 0.9% saline solution but with the limitation of an overall low fluid intake of the infusion solutions studied [ ]. On the other hand, several single centre studies in intensive care and perioperative settings found an association of the use of 0.9% saline solution and renal impairment [ ]. The latest evidence in favour of BCS is a large double blind randomised multicentre trial in children with sepsis. The study reported a relative risk (RR) for acute kidney injury of 0.62 (95% CI, 0.49–0.80) for BCS vs. 0.9% saline solution [ ]. In summary, there is some evidence of improved outcome with the use of BCS, while no negative effects have been demonstrated to date. So far, however, only a few professional associations have explicitly recommended the use of BCS in their guidelines [ , ]. Children with vomiting and severe hypochloraemic alkalosis (e.g. pyloric stenosis, gastroenteritis) can be considered an exception. In these patients 0.9% saline solution can be used to balance chloride levels.

6

Colloids for fluid replacement

Colloid infusion solutions theoretically have greater effect on intravascular volume than crystalloid solutions. They remain intravascular to a greater extent and for longer, whereas crystalloid solutions are evenly distributed in the extracellular fluid [ ]. While the extracellular space is larger in small children relative to their body weight, the proportion of blood volume in the extracellular fluid is lower than in older children and adults (new-borns 10%, infants 20%, adults 25%). As a consequence, crystalloid solutions cause an even smaller effect on intravascular volume of infants and toddlers [ ]. Pure fluid losses without the loss of oncotically active macromolecules such as urine, insensible perspiration, or preoperative dehydration must be discriminated from the loss of oncotically active macromolecules. With blood loss, patients lose a sometimes-considerable amount of protein. The administration of crystalloid solutions may further reduce the intravascular colloid osmotic pressure whereas colloid solutions can help to quickly restore the colloid osmotic pressure.

However, the use of artificial colloids for perioperative fluid replacement is limited due to side effects like allergies, interference with coagulation, or impairment of renal function. All colloids show in vitro and in vivo effects on coagulation [ ]. Measurable changes of thromboelastographic parameters can occur from a dosage of 10–15 ml/kg of colloid solution [ , ], although the clinical impact remains unclear [ , ]. Hydroxyethyl starch (HES) has been shown to cause renal impairment requiring renal replacement therapy in adult patients with sepsis [ ]. Although an association between HES infusion and renal impairment could not be found in healthy adult patients undergoing major surgery [ ] or children [ , , ], it is now only used very restrictively for perioperative fluid replacement and in many countries, it is no longer available at all. To date there are no reports linking the infusion of gelatine or albumin with renal impairment [ , ]. Gelatine is said to have a high allergenic potential. However, a prospective observational study from 13 European centres with 601 children reported no allergic reactions [ ].

Human serum albumin (HSA) comprises more than 50% of plasma proteins, is responsible for 80% of the intravascular oncotic pressure, is the main regulator of the vascular barrier and, antioxidant in the plasma. It is also a transporter of nitric oxide, fatty acids, and drugs. HSA has anticoagulant and antithrombotic functions. However, few studies have shown that administration of isoncotic albumin has a clinical impact on these mechanisms and the use of albumin as perioperative fluid substitution remains controversial as there is a lack of evidence supporting the use of albumin in most cases of fluid resuscitation. The theoretical blood volume expansion efficacy of albumin can be achieved under the conditions of an intact vascular barrier and normal permeability. As critical illness and inflammatory responses are frequently associated with the degradation of the ESL and increased vascular permeability, the clinical intravascular volume expansion efficacy of albumin is often much less than theoretically expected. Albumin is extracted from human blood but considered as virus safe. It is the most expensive colloid besides plasma [ ].

In principle, the use of colloids should be considered carefully and restricted to cases of major blood loss if crystalloid infusions are not sufficiently effective and blood products not immediately available. They can be infused in repetition doses of 5–10 ml/kg until the desired effect is achieved [ ]. Whenever a colloid is used it should be a balanced solution to interfere less with acid-base balance [ ].

7

Perioperative infusion therapy in neonates

The kidneys of neonates have a limited physiological range for urine osmolality. This ranges from 50 mosmol/L to 600 mosmol/L in preterm and 800 mmol/l in term infants (adults: 40-1200mosm/L), which limits the capacity to excrete and to conserve sodium. Hypotonic infusion solutions can also lead to hyponatraemia in neonates, particularly if perioperative stress-induced ADH secretion inhibits the excretion of free water and most of the perioperative losses consist of isotonic extracellular fluids (e.g. sodium and chloride rich shed blood or interstitial fluid). Therefore, BCS should also be used perioperatively in neonates for maintenance infusion and fluid replacement. In a recent meta-analysis, an infusion of isotonic maintenance fluid without corresponding fluid loss increased the risk of hypernatraemia in hospitalised neonates with either medical or surgical conditions and may lead to renal dysfunction [ ]. Due to the metabolic demands of the developing brain, neonates have increased rates of glucose utilisation as compared to adults and require about 4–8 mg/kg/min of glucose for their brain development [ , ]. As undetected hypoglycaemia during anaesthesia can have devastating consequences, it was alarming that blood glucose was only measured in half of the most vulnerable patients according to the NECTARINE study [ ]. Perioperative administration of 10 ml/kg/h of glucose 1%, 2% and 4% in Ringer’s lactate solutions was found to be equivalent in new-borns (1.6–2.8 kg) in terms of glucose homeostasis, but 2–4% dextrose-containing fluids were more effective at preventing intraoperative catabolism, insulin resistance, rebound hyperglycaemia, and acidosis than a 1% dextrose-containing fluid [ ]. Highly concentrated electrolyte-free glucose solutions must be used with great caution and must never be used freely, as accidental over-infusions can lead to critical incidents (e.g. hyperosmolar/hyperglycaemic coma) [ ].

8

Pre- and postoperative fasting times

Prolonged preoperative fasting is not only associated with thirst, hunger, diminished well-being, stress, dehydration, and discomfort, but also with perioperative anxiety, nausea and vomiting, and metabolic changes, including ketoacidosis. Long fasting times (more than 8 h) are associated with an increased risk of a reduction in systolic or mean arterial blood pressure after induction [ ]. As early as 2015, Andersson was able to show that children who were allowed to consume clear fluids until called to the operating room had only a low incidence of pulmonary aspiration [ ]. In the following years, intake of clear liquids up to 1 h before surgery was endorsed by several professional associations [ ]. Since aspiration frequency and risk did not differ in several prospective studies with shortened fasting times for clear liquids [ ], this led to a new ESAIC guideline in 2022 [ ]. In the European guideline, it is recommended, that children should be actively encouraged to consume solid food up to 6 h, small meals/formula milk up to 4 h, breast milk up to 3 h, and clear liquids such as sweetened tea or juice without pulp up to 1 h before the operation. Whenever possible early oral hydration should be encouraged after surgery to maintain normovolaemia and electrolyte homeostasis. Favourable secondary effects of using a baby bottle or water ice include distraction, which is linked to reduced opioid consumption, nausea, and vomiting [ ]. When postoperative enteral hydration is insufficient, intravenous fluid therapy is indicated.

9

Summary of recommendations

Since the 1950s, there have been many new findings and further developments in perioperative fluid therapy. The use of BCS (isotonic, isoionic, isohydric) is recommended both on the basis of physiological considerations and increasing scientific evidence. One-third and one-half electrolyte solutions should no longer be used perioperatively to avoid hyponatraemia and consecutive encephalopathies. Glucose 1–2.5 % should be added to maintenance infusion for new-borns, infants, and children at risk for hypoglycaemia. Perioperatively, the infusion rate for maintenance infusion can be started at 10 ml/kg/h for 1–2 h and then adjusted to the actual fluid requirements to achieve and maintain normovolaemia and a normal extracellular volume. In order to maintain normovolaemia, stabilise circulatory function, preserve extracellular volume, and ensure an optimised oxygen supply to all tissues BCS without glucose supplement can be administered in repetition doses of 10 ml/kg. In cases of major blood loss, colloids (5–10 ml/kg) can be used repeatedly. However, colloids should be used with caution and artificial colloids should only be used as balanced solutions. The practical management of perioperative intravenous fluid therapy is summarised in a stepwise algorithm for minor, intermediate, and major surgery ( Fig. 1 ). Perioperative fluid replacement should be monitored using clinical parameters, standard monitoring, and blood gas parameters. The additional use of NIRS and other extended haemodynamic monitoring may be considered.

Only gold members can continue reading. Log In or Register to continue

Share this:

• Click to share on Twitter (Opens in new window)
• Click to share on Facebook (Opens in new window)

Related

Related posts:

Preoperative fasting in children. The evolution of recommendations and guidelines, and the underlying evidence Advances in foetal anaesthesia Lessons learned from big data (APRICOT, NECTARINE, PeDI) Advances in pediatric neuroanesthesia practices Pediatric regional anesthesia and acute pain management: State of the art What’s new in pediatric critical care?

Stay updated, free articles. Join our Telegram channel

Join