U.S. flag

An official website of the United States government

NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.

National Clinical Guideline Centre (UK). Intravenous Fluid Therapy: Intravenous Fluid Therapy in Adults in Hospital [Internet]. London: Royal College of Physicians (UK); 2013 Dec. (NICE Clinical Guidelines, No. 174.)

  • April 2017: Key research recommendations 2 to 4 have been stood down and deleted by NICE from the short version of this guideline. These recommendations have been greyed out in this document. December 2016: A footnote was added to recommendations 19 and 22, tables 4.2.3 and 4.2.4 and the algorithms giving more information on weight-based potassium prescriptions. October 2016: A correction was made to the IV fluid therapy algorithm to clarify the period of time over which a fluid bolus can be given. A missing entry was corrected in the table showing the composition of commonly used crystalloids.

April 2017: Key research recommendations 2 to 4 have been stood down and deleted by NICE from the short version of this guideline. These recommendations have been greyed out in this document. December 2016: A footnote was added to recommendations 19 and 22, tables 4.2.3 and 4.2.4 and the algorithms giving more information on weight-based potassium prescriptions. October 2016: A correction was made to the IV fluid therapy algorithm to clarify the period of time over which a fluid bolus can be given. A missing entry was corrected in the table showing the composition of commonly used crystalloids.

Cover of Intravenous Fluid Therapy

Intravenous Fluid Therapy: Intravenous Fluid Therapy in Adults in Hospital [Internet].

Show details

5Principles and protocols for intravenous fluid therapy

Hospitalised patients need intravenous (IV) fluid and electrolytes for one or more of the following reason (the 4Rs):

  • Fluid resuscitation
    IV fluids may need to be given urgently to restore circulation to vital organs following loss of intravascular volume due to bleeding, plasma loss, or excessive external fluid and electrolyte loss, usually from the gastrointestinal (GI) tract, or severe internal losses (e.g. from fluid redistribution in sepsis).
  • Routine maintenance
    IV fluids are sometimes needed for patients who simply cannot meet their normal fluid or electrolyte needs by oral or enteral routes but who are otherwise well in terms of fluid and electrolyte balance and handling i.e. they are essentially euvolaemic, with no significant deficits, ongoing abnormal losses or redistribution issues. However, even when prescribing IV fluids for more complex cases, there is still a need to meet the patient’s routine maintenance requirements, adjusting the maintenance prescription to account for the more complex fluid or electrolyte problems. Estimates of routine maintenance requirements are therefore essential for all patients on continuing IV fluid therapy.
  • Replacement
    In some patients, IV fluids to treat losses from intravascular and or other fluid compartments, are not needed urgently for resuscitation, but are still required to correct existing water and/or electrolyte deficits or ongoing external losses. These losses are usually from the GI or urinary tract, although high insensible losses occur with fever, and burns patients can lose high volumes of what is effectively plasma. Sometimes, these deficits have developed slowly with associated compensatory adaptations of tissue electrolyte and fluid distribution that must be taken into account in subsequent replacement regimens (e.g. cautious, slow replacement to reduce risks of pontine demyelinosis).
  • Redistribution
    In addition to external fluid and electrolyte losses, some hospital patients have marked internal fluid distribution changes or abnormal fluid handling. This type of problem is seen particularly in those who are septic, otherwise critically ill, post-major surgery or those with major cardiac, liver or renal co-morbidity. Many of these patients develop oedema from sodium and water excess and some sequester fluids in the GI tract or thoracic/peritoneal cavities.

Deciding on the optimal amount, composition and rate of administration of IV fluids to address these often complex needs is inherently difficult yet assessment, prescribing and monitoring of IV fluids in general admission and ward areas of hospitals, is often left to junior doctors and hard-pressed nurses who may lack required training and competence.56,57,86,87,93 Evidence suggests that mismanagement of fluids is common, particularly in general ward areas with the potential for adverse outcomes including excess morbidity and mortality, prolonged hospital stays and increased costs.6,39,73,105,116,117

There is, therefore, a clear need for guidance on IV fluid prescribing applicable to general ward areas but since most randomized controlled trials of IV fluid therapy have examined narrow clinical questions in intensive care or intra-operative settings, many recommendations for more general use must be based on first principles. All health professionals involved in prescribing and administering IV fluids need to understand these principles if they are to prescribe and manage IV fluid therapy safely and effectively.

5.1. The principles of fluid prescribingb

The knowledge needed to underpin safe and effective IV fluid and electrolyte prescribing lies in four areas:

  • The physiology of fluid balance in health;
  • Pathophysiological effects on fluid balance;
  • Clinical approaches to assessing IV fluid needs;
  • The properties of available IV fluids.

5.1.1. The physiology of fluid balance in health

When primitive marine unicellular organisms evolved into multicellular organisms and emerged onto land, they carried with them their own internal sea or extracellular fluid (ECF), in which their cells could bathe in a constant chemical environment. The French physiologist Claude Bernard called this the ‘milieu interieur’, 10 an environment in which the cells retain their energy consuming capacity to pump sodium out and retain potassium in order to neutralise the negative charges of proteins and other ions.

While fluid balance is usually considered as that between the body and its environment, i.e. external balance, disease also affects the internal balance between the various body fluid compartments, e.g. between the intravascular and interstitial components of the extracellular fluid compartment (ECF), between the intracellular fluid (ICF) and the ECF, and between the ECF and the gut and other internal spaces.58,60 Appropriate IV fluid therapy depends on an understanding of the underlying physiology and pathophysiology and a consideration not only of external but internal fluid balance.58 Normal anatomy and physiology

Water comprises approximately 60% of the body weight of an average adult (about 40L in a 70kg man).29 The percentage is lower in obesity, since adipose tissue contains less water than lean tissue. It is also lower in women than in males because of the relatively greater amount of adipose tissue in women. The total body water is divided functionally into the extracellular (ECF=20% of body weight, about 14L in a 70kg man ) and the intracellular fluid spaces (ICF= 40% of body weight, about 28L in a 70kg man) separated by the cell membrane with its active sodium pump, which ensures that sodium remains mainly in the ECF. The cell, however, contains large anions such as protein and glycogen, which cannot escape and, therefore, draw in K+ ions to maintain electrical neutrality (Gibbs-Donnan equilibrium). These mechanisms ensure that Na+ and its balancing anions, Cl- and HCO3-, are the mainstay of ECF osmolality, and K+ has the corresponding function in the ICF. The ECF is further divided into the intravascular (within the circulation) and the interstitial (extravascular fluid surrounding the cells) fluid spaces. The intravascular space (blood volume = 5–7% of body weight, approx. 4 – 5L) has its own intracellular component in the form of red (haematocrit = 40–45%) and white cells and an extracellular element in the form of plasma (55–60% of total blood volume). The normal distribution of fluids in the different body compartments is shown in Figure 2 which also shows the likely compartmental distribution of some different types of IV fluids (see section 5.1.4).

Figure 2. Body water compartments and approximate distribution of commonly used IV fluids.

Figure 2

Body water compartments and approximate distribution of commonly used IV fluids. Source: Adapted from diagram(copyright obtained) by Dileep Lobo

The intravascular and extravascular components of the ECF are separated by the capillary membrane, with its micropores. The intravascular volume depends on plasma oncotic (colloid) pressure (POP) with plasma proteins retaining water in the circulation. POP is normally ~3.4kPa (26mmHg) with 75% of the effect due to albumin, 20% haemoglobin and 5% globulins. The plasma albumin concentration is ~35–52g/L, total body albumin is ~270g (120g intravascular, 150g ISF) and Figure 2 (see below) illustrates the albumin cycle. A gram of albumin ‘binds’ ~18mls of water, thus normal plasma albumin concentrations bind ~2.25L (18mls × 120g) of intravascular ‘plasma’ water. Normally, the capillary micropores only allow a slow escape rate of albumin (5%/hr, 120g/day), which is then returned to the circulation via the lymphatics at the same rate, maintaining equilibrium.30 While the hydrostatic pressure within the circulation drives fluid out, the oncotic pressure of the plasma proteins, e.g. albumin, draws fluid in. This maintains the relative constancy of the plasma volume as a proportion of the ECF (Starling effect). There is also a clinically important flux of fluid and electrolytes between the ECF and the GI tract involving active secretion and reabsorption of digestive juices. In health there is a constant flux between these various spaces and important physiological mechanisms ensure a constant relationship between them, which is termed the internal fluid balance.58 Fluid Balance

The external fluid and electrolyte balance between the body and its environment refers to the intake of fluid and electrolytes versus the output from kidneys, GI tract and the skin and lungs (insensible loss). The normal average daily intake and output of fluid and electrolytes are shown in Table 8 and Table 9 although these are very approximate and are modified greatly in the presence of excessive insensible losses e.g. of water and sodium in hot climates.

Table 8. Approximate daily water balance in health.

Table 8

Approximate daily water balance in health.

Table 9. Average daily intake.

Table 9

Average daily intake. Intake

Under normal circumstances most of our fluid intake is in the form of drinks but food also contains fluid and electrolytes, and water is also an end product of its oxidation which makes a further small but significant additional contribution to fluid intake. Drinking is governed by thirst, which is triggered when water balance is negative through insufficient intake or increased loss. It is also triggered by high sodium intake, since extra water is then needed to keep the ECF sodium concentration in the normal range.

Although, in the elderly, thirst may be blunted, in general it ensures that intake matches the bodily needs, maintaining zero balance and a steady physiological osmolality of 280–290mOsm/kg.

Claude Bernard coined the term ‘volume obligatoire’ to describe the minimum volume of urine needed to excrete waste products, e.g. urea, in order to prevent accumulation in the blood. This concept implies that, if sufficient fluid has been drunk or administered to balance insensible and other losses, and to meet the kidney’s needs, there is no advantage in giving more. Indeed, excessive intakes of fluid and electrolytes may be hazardous under certain circumstances (see below) since they can overwhelm the kidneys’ capacity to excrete the excess and maintain normal balance. Sodium and water excess in particular can cause oedema, although this only becomes an issue when the ECF has been expanded by at least 2–3 litres.61 Output

Insensible loss: evaporation of water from the lungs and skin occurs all the time without us being aware of it. In the UK climate, the amount lost is 0.5–1 litre/day but in hot climates, during fever or with exertion, losses of several litres of sweat can occur, containing up to 50 mmol/l of sodium.

Gastrointestinal losses: normally, the intestine absorbs water and electrolytes efficiently so that stool fluid loss is as little as 100–150 ml/day. However, in the presence of disease this may be greatly increased (see Section 5.1.2 and section on Intravenous fluid therapy for replacement and redistribution).

Kidneys: These are the main organs for fluid and electrolyte regulation and excretion of waste products from metabolism, e.g. urea. Their activity is controlled by pressure and osmotic sensors which result in changes in the secretion of hormones. The modest daily fluctuations in water and sodium intake cause small changes in plasma osmolality which trigger osmoreceptors. This in turn causes changes in thirst and the renal excretion of water and sodium. If blood or ECF volumes are subject to abnormal losses, volume receptors are triggered (see below) which override the osmoreceptors. In the presence of large volume changes, therefore, the kidney is less able to adjust osmolality. This can be important in some clinical situations. Water regulation

Osmoreceptors which sense changes in plasma osmolality, are located in the hypothalamus and signal the pituitary to increase or decrease secretion of vasopressin or antidiuretic hormone (ADH). Dilution of the ECF, including plasma, by intake of water or fluid of lower osmolality than plasma, causes ADH secretion to fall, so that the kidneys excrete more free water and produce a dilute urine). Conversely, dehydration causes the ECF to become more concentrated, ADH secretion rises and the renal tubules reabsorb more water, producing concentrated urine. In response to dehydration, the normal kidney can concentrate urea in the urine up to a hundred-fold, so that the normal daily production of urea related to protein metabolism in health can be excreted in as little as 500 ml of urine.

In the presence of water deficit, the urine to plasma urea or osmolality ratio is, therefore, a measure of the kidney’s concentrating capacity. Age and disease can impair the renal concentrating capacity so that a larger volume of urine is required in order to excrete the same amount of waste products. Also if protein catabolism increases due to a high protein intake or increased catabolism, a larger volume of urine is needed to clear the resulting increase in urea production.

To assess renal function, therefore, measurement of both urinary volume and concentration (osmolality) are important, and the underlying metabolic circumstances taken into account. If serum urea and creatinine concentrations are unchanged and normal, then, urinary output over the previous 24 hours has been sufficient, fluid intake has been adequate, and the urinary ‘volume obligatoire’ has been achieved. Sodium (Na+) regulation

Since the integrity of the ECF volume and its proportion of the total body water are largely dependent on the osmotic effect of Na+ and its accompanying anions, it is important that the kidneys maintain Na+ balance within narrow limits. If sodium depletion occurs, the ECF and plasma volumes fall. Pressure sensors in the circulation are then stimulated and these excite renin secretion by the kidney. This, in turn, stimulates aldosterone secretion by the adrenal gland, which acts on the renal tubules, causing them to reabsorb and conserve sodium.

Conversely, if the intake of Na+ is excessive, the renin-aldosterone system is supressed, allowing more Na+ to be excreted, until normal balance is restored. The mechanism for sodium conservation is extremely efficient and the kidney can reduce the concentration of Na+ in the urine to <5 mmol/l. On the other hand, even in health, we are slow to excrete an excess sodium load, possibly because human physiology evolved in the context of the hot, low sodium environment of Africa and has not until modern times been exposed to excessive sodium intake. The response of atrial natriuretic peptide to fluid infusions seems to be related more to volume (stretching of the right atrium) than sodium load per se.

The mechanism for maintaining sodium balance may be disturbed in disease, leading to Na+ deficiency or, more commonly, to excessive sodium retention, with consequent oedema and adverse clinical outcome. Potassium (K+) regulation

Although only a small proportion of the body’s K+ is in the extracellular space, its concentration has to be maintained within narrow limits (3.5–5.3 mmol/l) to avoid the risk of muscular dysfunction or potentially fatal cardiac events. This is achieved by exchange of K+ in the renal tubules for Na+ or H+, allowing more or less K+ to be excreted. In the presence of K+ deficiency, H+ ion reabsorption is impaired, leading to hypokalaemic alkalosis and a decrease in the kidneys’ ability to excrete a sodium load.

5.1.2. Pathophysiological effects on fluid balance

Illness and injury alter fluid and electrolyte balance and distribution needs in many ways including:

  • Non-specific metabolic responses to stress (especially in the seriously ill or injured);
  • Changes in fluid or electrolyte handling directly attributable to specific organ or system dysfunction or the effects of drugs or other IV therapies used to treat such problems;
  • Changes in fluid or electrolyte handling due to very restricted recent food intake or malnutrition. Non-Specific responses to illness and injury

In the 1930’s, Cuthbertson22 described the metabolic changes, which occur in response to injury (including surgery and sepsis), as an increase in metabolic rate and protein breakdown to meet the requirements for healing. These changes were later shown to be due to neuroendocrine and cytokine changes and to occur in three phases. The ebb or shock phase is brief and is modified by resuscitation. This gives way to the flow or catabolic phase, the length and intensity of which depends on the severity of injury and its complications. As inflammation subsides, the convalescent anabolic phase of rehabilitation begins. In parallel with these metabolic changes, there are changes in water and electrolyte physiology. During the flow phase, there is an increase in ADH, cortisol and aldosterone secretion, especially if there has been any reduction in blood or ECF volume. These lead to retention of sodium and water with loss of potassium.121,122 The normal, if somewhat sluggish, ability to excrete an excess of sodium and water load is then further diminished, leading to ECF expansion and oedema. 58

These non-specific responses imply that a degree of oliguria is normal in the context of serious illness or injury, 110 and hence that the presence of oliguria does not necessarily indicate a need to increase administration of sodium and water or plasma expanders unless there are also indications of intravascular volume deficit, e.g. from postoperative bleeding. Indeed, sodium and water retention after injury can be seen as nature’s way of trying to protect the ECF and circulating volume at all costs. It also explains why sick patients can be so easily overloaded with excessive IV sodium and water administration during the flow phase. Since water as well as sodium is retained, it is also easy to cause hyponatraemia by giving excess water or hypotonic fluid. It is important, therefore, to administer crystalloids, not only in the correct volume but also in the appropriate concentration especially as, in the presence of these responses to illness or injury, the kidneys are unable to correct for errors in prescribing, even in the absence of significant acute kidney injury (AKI) or other renal pathology.

The convalescent phase of serious illness or injury is not only characterised by the return of anabolism but also by a returning capacity to excrete any excess sodium and water load that has been accumulated. These periods have been termed the ‘sodium retention phase’ and the ‘sodium diuresis phase’ of injury.

Transcapillary escape rate of albumin

The responses to serious illness of injury also includes an increase in the size of the pores in the capillary membrane and the transcapillary escape rate of albumin increases by up to 300% from about 5%/h in health to 13–15%/h.30 Subsequent falls in plasma albumin then reduce POP and intravascular volume, whilst increases in ISF albumin promote oedema. This phenomenon can last from several hours to days. Albumin and other plasma proteins leak out from the intravascular compartment into the interstitial space and water and sodium also move into that space. This results in a net contraction of the intravascular compartment and expansion of the interstitial space. As the return of albumin to the circulation via the lymphatics is unchanged, the net result is an intravascular hypovolaemia with oedema.


Potassium losses during serious illness and injury are not only secondary to increased excretion from high cortisol and aldosterone levels, but also to protein and glycogen catabolism. As intracellular protein is broken down and its constituent amino acids are released from cells, so intracellular negative charges are lost and K+, with its balancing positive charges, passes out into the ECF to be excreted. In situations where catabolism is extreme and renal function is impaired, the outflow of K+ from the cells may exceed the kidney’s capacity to excrete it, causing dangerous hyperkalaemia. Conversely, in the convalescent phase, as net intracellular protein and glycogen anabolism is restored, the cells take up again and the patient’s K+ intake has to be increased to prevent the development of hypokalaemia and to help with the excretion of a likely total excess in body sodium.

Malnutrition is common in hospital patients since it is both a cause and a consequence of illness and injury. When present, it can have non-specific effects on fluid and electrolyte status and handling since starvation is accompanied by reductions in cell membrane pumping, with consequent movement of more sodium and water into cells than usual, while simultaneously potassium, magnesium, calcium and phosphate move out of cells and are excreted by the kidneys. A malnourished individual therefore tends to have a degree of total body sodium and water overload, coupled with depletion of total body potassium, phosphate, magnesium and calcium. These changes are often unrecognized as plasma levels may remain normal. The most important problems caused by these changes in relation to IV fluid and electrolyte prescribing, occur when a malnourished individual is fed, even if that feeding is only in the form of glucose from IV infusions. The arrival of the glucose, coupled with the release of insulin it triggers, can reverse the depression of the membrane pumps, leading to cellular uptake of potassium, phosphate, magnesium and calcium with potentially dangerous falls in plasma levels.104 At the same time, there is a net movement of sodium and water out of cells into the circulation, a redistribution change that is effectively added to any IV fluids being administered but is frequently unaccounted for. Since malnourished individuals may have diminished cardiac reserve and/or hidden infection with high capillary escape rates, the consequence of all the above may be potentially lethal fluid overload and cardiac instability. These problems are known as the refeeding syndrome and specific advice on the prevention and management of these problems is provided in the NICE guideline on Nutrition Support in adults. Effects of specific organ or system dysfunction

Many specific medical conditions can alter the body’s fluid and electrolyte handling, as can many of the therapies used to treat such problems. Detailed discussions of such changes are clearly not possible within this guidance but examples of issues that might influence IV fluid prescriptions are shown in Table 10. The organ or system dysfunction may be the either the primary problem that has brought the patient into hospital or a significant co-morbidity)

Table 10. Issues influencing IV fluid prescriptions.

Table 10

Issues influencing IV fluid prescriptions. Effects due to very restricted recent food intake or malnutrition

Some degree of starvation is common in individuals who are ill or injured, especially those who might need IV fluid therapy. Reduced or absent food intake leads quite swiftly to alterations in cell function which include a reduction in membrane pumping so that potassium leaks out of the cells and is then lost in the urine, while sodium and water move into cells. Malnourished individuals, and even those who are overweight but have a history of recent starvation, may therefore have lower than expected total body potassium and higher total salt and water content. This makes them potentially vulnerable to fluid mismanagement, especially since malnutrition can also cause a decrease in cardiac reserve, a decrease in renal capacity to clear salt and water, and deficiencies of specific vitamins. Furthermore. This vulnerability is further enhanced if significant feeding is introduced at the same time as IV fluids withy the potential for inducing low phosphate, potassium or magnesium as part of the refeeding syndrome (see Guidance of Refeeding syndrome in NICE CG32 – Nutrition Support in Adults).

5.1.3. The clinical approach to assessing IV fluid needs

The most appropriate method of fluid and electrolyte administration is the simplest, safest and effective. The oral route should be used whenever possible and IV fluids can usually be avoided in patients who are eating and drinking. The possibility of enteral tube administration should also be considered if safe oral intake is compromised but there is enteral tube-accessible GI function.

Figure 3 illustrates the ‘4 Rs’ that underpin the clinical approach to deciding IV fluid needs: Resuscitation, Routine maintenance, Replacement and Redistribution. There is also a ‘5th R’ for Reassessment.

Figure 3. The 4 Rs - Resuscitation, Routine maintenance, Replacement and Redistribution.

Figure 3

The 4 Rs - Resuscitation, Routine maintenance, Replacement and Redistribution. A 5th R – Reassessment is also a critical element of care. Source: Adapted from diagram(copyright obtained) by Dileep Lobo

Clinical considerations around the ‘4Rs’ can be complex and so decisions on the optimal amount, composition and rate of IV fluid administration must be based on careful, individual patient assessment. However, the clinical principles underlying these decisions can be approached as a series of questions.

Does my patient need IV fluid resuscitation?

This is the first question, since urgent IV fluid therapy is a critical element in the management of most shocked patients. For details on prescribing for routine maintenance see section Intravenous fluid therapy for fluid resuscitation.

Can my patient meet fluid and electrolyte needs by the oral or enteral route?

The unnecessary use of IV fluids should be avoided. When they are needed, they should be stopped as soon as possible.

What is my patient’s current fluid and electrolyte status?

Assessment must be informed by all information available including a focussed history and examination along with results of clinical monitoring (e.g. NEWS, fluid balance and body weight) and laboratory results. For details on assessment and monitoring, see section Assessment and monitoring of patients receiving intravenous fluid therapy.

What are my patient’s routine maintenance needs for fluid and electrolytes?

The average person requires 25–30 ml/kg water per day and about 1 mmol/kg of Na+ and K+. For details on prescribing for routine maintenance see section Intravenous fluid therapy for routine maintenance.

Does my patient have existing fluid or electrolyte deficits or abnormal ongoing losses?

All IV fluid prescriptions should add enough fluid and/or electrolytes to correct any existing deficits or meet abnormal ongoing losses, to estimates of routine maintenance requirements. Recommendations and more details on fluid prescription for replacement are covered in the section Intravenous fluid therapy for replacement and redistribution.

Does my patient have problems with internal redistribution of fluid or other fluid handling issues from either their primary problem or significant co-morbidities?

IV fluid prescriptions must aim to account for both non-specific responses to illness or injury described in Section 5.1.2 as well as the more complex problems of fluid distribution or handling caused by specific organ or system dysfunction and/or malnutrition. Recommendations and more details on these issues are also covered in the section Intravenous fluid therapy for replacement and redistribution.

Consideration of all questions above allows estimates of the total volume of IV fluid and amounts of electrolytes that should be given, before deciding on the best rate at which to administer the fluids. Often, that rate needs to be slow in order not to overload the circulation or to cause acute electrolyte problems, since time is needed for transmembrane (i.e. ECF/ICF) physiological equilibrations to occur. The best IV fluid (or mix of fluids) to use can then be chosen although, before completing the prescription, allowance must be made for any fluid and electrolytes intake from other sources. These include any food and drinks, enteral tube provision and other IV therapies. Blood or blood products, in particular, contain large amounts of electrolytes as do some IV drugs, especially those given in larger volume diluents, several times a day. Patients on artificial parenteral or enteral nutrition usually receive adequate fluid and electrolytes from their feed to meet at least routine maintenance needs and prescription of unnecessary additional IV fluids in such patients is a common mistake.

5.1.4. The properties of available IV fluids

Many different crystalloids, artificial colloids and albumin solutions are available for IV fluid therapy. The aim is to meet estimates of total fluid and electrolyte requirements. There are theoretical advantages to giving a colloid instead of a crystalloid when resuscitating the hypovolaemic patient because colloid-based fluids generally remain for longer in the circulation. Crystalloids are distributed throughout the ECF and traditional teaching is that their infusion has relatively limited and transient effects on plasma volume. However, such considerations are based on data derived from studies undertaken in euvolaemic human volunteers who have no illness-induced abnormalities in fluid distribution and capillary permeability, and in hypovolaemic patients, crystalloids have much better intravascular retention than these studies have suggested. The actual benefits, if any, of colloids over crystalloids when intravascular volume expansion is required are therefore unclear.

A review of all the available IV fluids in the UK is beyond the remit of this guidance but understanding the composition and properties of some of those more commonly used provides much of the understanding needed to prescribe any fluid appropriately. Furthermore, it helps understanding of the issues in fluid prescribing which are of debate in current practice.

See Appendices P.1 and P.2 for details on the composition of commonly used crystalloids and colloids which have been reviewed as part of the evidence for this guideline. A brief description of some of the available fluids highlighting their properties and potential pros and cons of their usage is detailed below.

Isotonic saline

Sodium chloride 0.9% with or without additional potassium is one of the most commonly used IV fluids in UK practice. However, questions have been raised in relation to its appropriate use. As with all crystalloids, sodium chloride 0.9% is distributed throughout the ECF and infusion usually has a more transient effect on plasma volume than colloids. Traditionally sodium chloride 0.9% infusion has been considered to expand blood volume by only a quarter to a third of the volume infused, the remainder being sequestered in the interstitial space. 18,59,62,89 In practice, for the reasons given above, intravascular retention of sodium chloride 0.9% is likely to better than this in hypovolaemic and stressed patients. Theoretically, use of sodium chloride 0.9% for plasma volume expansion might cause more oedema than would occur with use of a colloid but such a difference is seldom realised in practice.

In addition, it is also possible that a significant albeit lesser degree of unnecessary sodium and water retention, is a problem when sodium chloride 0.9% is used for routine maintenance. The normal daily requirements of sodium are only 70–100mmol but one litre of normal saline contains 154mmol, so it is easy to give an excess. This will then need to be excreted but the ability to clear a solute load is limited even in health and may be further impaired during illness or injury.

Another issue that raises questions about the widespread usage of sodium chloride 0.9% is the fact that it produces a degree of hyperchloraemia due to its high chloride content compared with plasma. This in turn could lead to significant reductions in renal blood flow and glomerular filtration 18 as well as hyperchloraemic acidosis, gastrointestinal mucosal acidosis and ileus.59

Some GI fluid losses and occasionally renal losses are very high in sodium chloride and hence sodium chloride 0.9% use may well be appropriate in situations where there are ongoing high sodium losses or deficits of sodium, chloride and water from earlier losses. It is important to recognize, however, that many of these losses will be high in potassium, calcium and magnesium and so a balanced crystalloid might have advantages over sodium chloride 0.9% with added potassium.

Balanced crystalloid solutions

Balanced crystalloids are also distributed throughout the ECF and are therefore of similar efficacy to sodium chloride 0.9% in terms of plasma volume expansion. However, they do have theoretical advantages in that they contain somewhat less sodium and significantly less chloride, and they may already have some potassium, calcium and magnesium content. The use of balanced crystalloids could therefore have advantages over sodium chloride 0.9% when used for resuscitation or routine maintenance and preparations with more specialized ‘resuscitation’ and ‘maintenance’ versions, with content tailored to meet more closely the theoretical requirements for these different circumstances, are likely to become increasingly available in future. Balanced solutions containing lactate or other buffers might also grant advantages in situations of significant acidosis which is often seen when resuscitation is needed.

Glucose and glucose salines

Solutions such as 5% glucose and glucose/saline with or without potassium are not meant for resuscitation or replacement of electrolyte rich losses. They are however, useful means of providing free water for, once the glucose is metabolised, they are largely distributed through total body water with very limited and transient effects on blood volume. They should therefore be useful in correcting or preventing simple dehydration, and the administration of appropriate glucose saline with potassium solutions may provide a good means of meeting routine maintenance needs. However, the use of these fluids will increase risks of significant hyponatraemia, especially if too much fluid is given or the infusion is given too rapidly. Such risks are particularly high in children, the elderly, patients on diuretics and those with SIADH problems which are seen quite frequently in hospitalized patients. It is also important to appreciate that the calorie content of 5% glucose is very low and provides little contribution to the nutrition support which may be needed in some patients.

Synthetic Colloids

Synthetic colloids contain non-crystalline large molecules or ultramicroscopic particles dispersed through a fluid which is usually a crystalloid. The colloidal particles are large enough that they should be retained within the circulation and so exert an oncotic pressure across capillary membranes. In theory, colloids that are iso-oncotic with plasma should therefore expand blood volume by the volume infused but in practice, the volume expansion achieved is closer to 60–80%7,62 and it may be much less in sicker patients with high transcapillary escape rates. The actual advantages of colloids over crystalloids when used for either intravascular volume expansion in patients requiring fluids for resuscitation or to help with the resolution of oedematous redistribution problems are therefore uncertain and with some preparations, there have been concerns that any potential advantages may be offset by problems including renal dysfunction or disturbed coagulation. It is important to note, that older preparations of hydroxyethyl starch are suspended in sodium chloride 0.9% while some newer preparations are suspended in balanced solutions which should make them more ‘physiological’. Nevertheless, all currently available semi-synthetic colloids contain 140–154 mmol sodium which could contribute to positive sodium balance in sicker patients in the same way as for sodium chloride 0.9%, although colloids do contain less chloride.

In the UK, synthetic colloids commonly used in admission and general ward areas include; hydroxyethyl starch, succinylated gelatin, urea-linked gelatin, whilst dextrans and high molecular weight penta- and hexa-starches are used seldom or not at all.

Albumin solutions

As with synthetic colloids, infusion of albumin solutions might theoretically grant potential benefits from better intravascular volume expansion although costs would be very high. Concentrated (20–25%) sodium poor albumin could also be valuable in fluid redistribution problems especially when oedema from total sodium and water overload is present in post-severe illness or injury patients who still have low plasma volumes.3,4,54 Albumin is also used in some patients with hepatic failure and ascites although its use in this setting is beyond the scope of this guidance.

5.1.5. Recommendations based on fluid prescribing principles

RecommendationsThe assessment and management of patients’ fluid and electrolyte needs is fundamental to good patient care.

Assess and manage patients’ fluid and electrolyte needs as part of every ward review. Provide intravenous (IV) fluid therapy only for patients whose needs cannot be met by oral or enteral routes, and stop as soon as possible.


Skilled and competent healthcare professionals should prescribe and administer IV fluids, and assess and monitor patients receiving IV fluids (see recommendations 26–28).


When prescribing IV fluids, remember the 5 Rs: Resuscitation, Routine maintenance, Replacement, Redistribution and Reassessment.


Include the following information in IV fluid prescriptions:

  • The type of fluid to be administered.
  • The rate and volume of fluid to be administered.

Patients should have an IV fluid management plan, which should include details of:

  • the fluid and electrolyte prescription over the next 24 hours
  • the assessment and monitoring plan.

Initially, the IV fluid management plan should be reviewed by an expert daily. IV fluid management plans for patients on longer-term IV fluid therapy whose condition is stable may be reviewed less frequently.


When prescribing IV fluids and electrolytes, take into account all other sources of fluid and electrolyte intake, including any oral or enteral intake, and intake from drugs, IV nutrition, blood and blood products.


Patients have a valuable contribution to make to their fluid balance. If a patient needs IV fluids, explain the decision, and discuss the signs and symptoms they need to look out for if their fluid balance needs adjusting. If possible or when asked, provide written information (for example, NICE’s Information for the public), and involve the patient’s family members or carers (as appropriate).

Relative values of different outcomesMortality and morbidity were identified as the most critical outcomes. The other outcome considered important for decision making was length of stay in hospital.
Trade-off between clinical benefits and harmsGiven the morbidity associated with injudicious prescription of intravenous fluids, particularly the consequences of fluid overload (e.g. pulmonary oedema), the GDG agreed that emphasis should be placed on careful assessment and reassessment of the need for intravenous fluid therapy.
Economic considerationsThere was no cost-effectiveness evidence. However, the principle of only using intravenous fluids when necessary and stopping them as early as possible is likely to be highly cost-effective, since it should both reduce the cost of administering unnecessary IV fluids and should reduce the cost of treating avoidable fluid overload as well as improving other clinical outcomes.
Quality of evidenceThe GDG drafted these recommendations based on physiological, pathophysiological and clinical principles using consensus.
Other considerationsIt was acknowledged that that it was not possible to undertake clinical evidence reviews for certain areas of the guideline and the principles of fluid prescribing was one such exception to the normal systematic review process.
Here, the GDG took into consideration the principles of physiology and pathophysiology of intravenous fluids and other accepted standard clinical guidance and drafted recommendations based on expert consensus in a format intended to be useful to a clinician. The GDG discussed and agreed that as the recommendations were fundamental to fluid prescribing, the wording of the recommendations should reflect the strength of the recommendations.
Clinical assessment and diagnosis of the volume status of the patient was judged to be key to prescribing safe, appropriate IV fluid therapy for a patient. The GDG discussed the four states where intravenous fluid was given, that is, (i) resuscitation, (ii) routine maintenance, iii) replacement of existing deficits or abnormal ongoing losses and iv) complex issues of redistribution. They agreed that clear identification of the reason for giving IV fluid therapy should always precede administration.
Recommendations 3 and 5 were identified as key priorities for implementation by the GDG as they have a high impact on outcomes that are important to patients and have a high impact in reducing variation in care and outcomes.

5.2. Use of algorithms in IV fluid therapy

An approach to IV fluid prescribing based on physiological, pathophysiological and clinical principles can potentially be described in protocols and algorithms. Since it is well recognized that adoption of protocol-driven care has improved clinical standards in other areas, a review of the clinical and cost effectiveness of any published clinical algorithms or defined protocols for assessment, monitoring and/or management of IV fluid prescriptions was undertaken.

5.2.1. Review question

What is the clinical and cost effectiveness of clinical algorithms or defined protocols for the assessment, monitoring and/or management of intravenous fluid and electrolyte requirement in hospitalised adult patients?

The objective of this review was to compare outcomes in hospitalised patients who received IV fluid therapy as part of a protocol to those who received IV fluids without any protocol.

For the review protocol see C.1, Appendix C.

5.2.2. Clinical evidence

We searched for randomised controlled trials comparing the effectiveness of using algorithms or defined protocols compared to no protocols or usual care for the management of hospitalised adult patients on IV fluid therapy.

No Cochrane reviews relevant to the review question were identified.

Six randomised controlled studies were identified.9,32,40,53,78,91 The studies included different populations and settings, for example; surgical patients, sepsis patients, burn patients and patients in intensive care units. Some of these studies did not meet the criteria set in the protocol for our target population, but in view of the paucity of directly relevant literature data, they were still extracted and extrapolated to our target groups, with the evidence downgraded for indirectness (see clinical evidence profile in Table 12).

Table 12. Clinical evidence profile: Protocol vs. No protocol.

Table 12

Clinical evidence profile: Protocol vs. No protocol.

All 6 studies compared protocol directed care with no protocol. The components of the protocols varied across the studies. Three studies focused on early goal directed therapy.48,53,91 Table 11 details the summary characteristics of included studies.

Table 11. Summary of included studies - Protocol vs. no protocol.

Table 11

Summary of included studies - Protocol vs. no protocol.

Since the evidence came from different populations and settings, pooling of results across all studies was not considered to be appropriate. The evidence is therefore presented with respect to the different population sub-groups as identified in the review protocol.

See flow diagram for clinical article selection in J.1, Appendix J and economic article selection K.1, Appendix K, forest plots in G.1, Appendix G, clinical evidence tables in E.1, Appendix E, economic evidence tables in F.1, Appendix F and excluded studies list in H.1, Appendix, H.

5.2.3. Economic evidence

Three studies were included that made relevant comparisons. These are summarised in the economic evidence profile below (Table 13 and Table 14)

Table 13. Protocol vs No Protocol Economic Evidence.

Table 13

Protocol vs No Protocol Economic Evidence.

Table 14. Protocol vs. No Protocol -- Economic summary of findings.

Table 14

Protocol vs. No Protocol -- Economic summary of findings.

See also the full study evidence table in F.1, Appendix F.

5.2.4. Evidence statements

Clinical evidence

Patients with sepsis
  • Evidence from two studies in patients with sepsis suggested that patients receiving IV fluid therapy as part of a protocolised care package had less mortality, decreased length of hospital stay, and fewer renal complications compared to patients who received IV fluids not as part of any protocol. The quality of evidence was of low to very low quality.
Intra-operative patients
  • Evidence from two studies in intra-operative patients suggested that patients receiving IV fluid therapy as part of a protocolised care package may have decreased mortality and decreased length of stay in hospital compared to patients who received IV fluids not as part of any protocol. The evidence was of very low quality.
Post-operative patients
  • Evidence from one study in post-operative patients showed that patients receiving IV fluid therapy as part of a protocolised care package have decreased length of stay in hospital and intensive care unit compared to those receiving IV fluids not as part of any protocol. However, there was no difference with respect to number of renal complications between the two groups. The evidence was of very low quality.
Trauma/shock patients
  • Evidence from one study in patients with trauma or shock suggested that there was no difference with respect to mortality, length of stay in hospital and length of stay in intensive care unit when comparing patients receiving IV fluid therapy as part of a protocolised care package with those who receive IV fluids not as part of any protocol. The evidence was of very low quality.

Economic evidence

  • Three studies found that compared to conventional, non- protocolised care, IV fluid therapy as part of a protocolised care package for patients presenting with sepsis and septic shock was cost effective (from cost saving up to £10,312 per QALY gained). This analysis was assessed as partially applicable with potentially serious limitations.

5.2.5. Recommendations and link to evidence


Offer IV fluid therapy as part of a protocol (see Algorithms for IV fluid therapy):

Relative values of different outcomesMortality and morbidity were identified as the most critical outcomes. Length of stay in hospital was also considered important for decision making was.
Trade-off between clinical benefits and harmsThe clinical evidence review found that on the whole, outcomes, including survival were more favourable in patients receiving IV fluids as part of a protocol-based care package, irrespective of different patient population groups, that is, patients with sepsis or intra/post-operative patients. It was recognised that components of individual protocols influence outcomes differently in different populations and this should be kept in mind when following any particular protocol. The GDG agreed that emphasis should be placed on accurate assessment and reassessment of volume and electrolyte status when administering IV fluid therapy to any patient.
Economic considerationsIn patients with sepsis, IV fluid therapy as part of a protocolised care package was found to be cost-effective for sepsis patients in two studies and cost saving in a third study.
There was no cost-effectiveness evidence for patients without sepsis. However, given that the health improvements observed in the review of clinical effectiveness evidence were just as pronounced for intra-operative care the GDG felt that the economic benefits of protocols are very likely to be achievable across all settings. However, there were issues of applicability and quality – see below.
Quality of evidenceThe quality of the clinical evidence varied from low to very low. The studies included in the clinical evidence review have several limitations and are at risk of bias. Since our target population is all hospitalised patients, the clinical evidence available from the studies found for specific population groups has limited applicability and the evidence has been downgraded for indirectness.
The three cost-effectiveness evidence studies were all in a US setting and therefore may not be transferable to a UK NHS setting since clinical practice resource use and unit costs are all likely to be different. In addition there were some potentially serious limitations. For example, not all health and cost outcomes of interest were included and unlike the clinical evidence reviewed above all three were based on observational evidence.
Other considerationsThe GDG discussed that evidence was only available for specific population groups which may not applicable to all hospitalised patients, particularly older patients with multiple co-morbid chronic diseases. The GDG also discussed the extreme heterogeneous nature of the target population and agreed that it would not be meaningful to pool the evidence across different population groups. Results are therefore presented separately. Nevertheless, the evidence favoured the use of protocolised care when giving IV fluids, irrespective of the population group, and the GDG were not only aware that following of protocols has been shown to be of value in several other areas of complex decision making in healthcare, but felt that algorithms were the best way for the guidance to be implemented across hospital settings. The GDG therefore made a consensus decision to advocate the use of algorithms for IV fluid therapy.
In view of the above, the GDG drafted four algorithms to be used for management of IV fluid therapy in hospitalised patients covering: assessment (algorithm 1); fluid resuscitation (algorithm 2); routine maintenance (algorithm 3); and replacement and redistribution (algorithm 4). Available evidence and discussion underpinning steps in each of the individual algorithm is presented in the relevant sections. This recommendation was identified as a key priority for implementation by the GDG.

5.2.6. Algorithms for IV fluid therapy

Flowchart Icon

Algorithms for IV fluid therapy (PDF, 357K)



Some sections of the text in the introduction of this chapter are written by two GDG members who are also co-authors of a textbook on intravenous fluid therapy. The text book was commissioned by a pharmaceutical company (B.Braun) who own the copyright permissions. The company did not have any intellectual or editorial input into the text and the intellectual content of the text is the property of the GDG members. If the wording and sentiment of the text is similar, it can be attributed to the GDG members’ direct involvement in both pieces of work.

Copyright © 2013, National Clinical Guideline Centre.
Bookshelf ID: NBK333103


  • PubReader
  • Print View
  • Cite this Page
  • PDF version of this title (5.5M)

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...