Before we go any further with this post, please do join us for this amazing meeting:

Screen Shot 2017-07-10 at 15.40.20.png


Register here

Submit your Abstract here



It is imperative to acknowledge that there are four main indications for fluid therapy. The United Kingdom’s National Institute for Health and Care Excellence (NICE) recently provided a complete set of guidelines, algorithms and instructions for intravenous fluid therapy in adult hospitalised patients . The somewhat older GIFTASUP-guidelines summarise evidence specifically for the management of surgical patients.

The four indications that will be further discussed are:

  1. Resuscitation fluids
  2. Maintenance solutions
  3. Replacement solutions
  4. Nutritional fluids.

This blog post dealing with “The 4 indications of fluid therapy” is the second in a total series of four, the others being:

  • “The 4 D’s of fluid therapy”
  • “The 4 questions of fluid therapy”
  • “The 4 phases of fluid therapy”


Introduction: between Scylla and Charybdis

The associated morbidity and mortality with poor fluid management is either related to hypovolemia with convective problems or hypervolemia and fluid overload with proven morbidity, due to interstitial oedema related diffusion problems (see IFAD abstract here and paper by Bellamy here). This was also the conclusion of a recent systematic review, reflecting the morbidity associated with more liberal fluid regimes, particularly on the development of intra-abdominal hypertension. Correct fluid resuscitation is all about finding the balance between Scylla and Charybdis. The sirens and sisters Charybdis and Scylla resided in the Sicilian Sea. They are thought to inhabit the Straits of Messina, the narrow sea between Sicily and the Italian mainland. Homer tells us that because Charybdis had stolen the oxen of Hercules, Zeus struck her with a thunderbolt and changed her into a whirlpool whose vortex swallowed up ships three times per day, as a metaphor for the risks of hypovolemia or under-resuscitation

Screen Shot 2017-07-24 at 12.10.22


Scylla was a supernatural creature, with 12 feet and 6 heads on long, snaky necks, each head having a triple row of shark-like teeth, while her loins were girt with the heads of baying dogs. From her lair in a cave she devoured whatever ventured within reach, including six of Odysseus’ companions, as a metaphor for the risks of hypervolemia or fluid overload. Odysseus avoids Charibdis in order not to lose all of his men. He chooses instead to get closer to Scylla and he only loses 6 of his men. In Ovid’s Metamorphoses, Books XIII–XIV, she was said to have been originally human in appearance but transformed out of jealousy through the witchcraft of Circe into her fearful shape. Being between Scylla and Charybdis is an idiom deriving from Greek mythology, meaning “having to choose between two evils”.

Nobody could look at her with delight, not even a god if he passed that way. She has twelve feet, all dangling in the air, and six long scrawny necks, each ending in a grisly head with triple row of fangs, set thick and close, and darkly menacing death. Up to her waist she is sunk in the depths of the cave, but her heads protrude from the fearful abyss, and thus she fishes from her own abode, groping greedily around the rock. (Odyssey, 12:87-95)


 Fluid Management

The last years have been very important with regard to evidence based medicine trying to close the colloid vs crystalloid fluid debate that has been going on for decades, at least for the time being. But is this really the case?

The publication of the CRYSTMAS study, comparing the use of hydroxyethylstarch (HES 130/0.4 waxy maize) vs saline in 196 patients with septic shock formed the kick-off for another series of multicentre studies on fluid management in the critically ill. While CRYSTMAS showed that less fluid was needed in the HES group (1370±886 vs 1709±1164 ml; p=0.02) to reach hemodynamic stability no differences were found in mean and cumulative fluid balance over the first 4 days and the same was true for renal and coagulation side effects.

This was followed by the 6S trial, a prospective state of the art study, comparing balanced HES (130/0.42 potato) vs Ringer’s acetate solution in 798 patients with severe sepsis. Albeit no differences in median trial fluid volumes (3000 ml in both arms) were observed the HES treated patients were more likely to die at day 90 and to require renal replacement therapy (RRT). This study was carefully designed, avoiding HES over dosage, using balanced solutions in both arms, with broad inclusion criteria and many patients exhibiting shock. However no data was provided on hemodynamic monitoring or if fluids were guided in a protocolized way.

The CHEST study concluded the series of big trials including 7000 general ICU patients randomized to either HES 130/0.4 vs saline. After the first 4 days, the average amount of study fluids per day was 526±425 ml (HES) vs 616±488 ml (NaCl)(p<0.001), while the amount of non-study fluids was 851±675 ml (HES) vs 1115±993 ml (NaCl) (p<0.001), resulting in a net fluid balance of 921±1069 ml (HES) vs 982±1161 ml (NaCl)(p= 0.03).

Crystalloids have a much better short-term effect on the plasma volume than previously believed. Their efficiency (i.e. the plasma volume expansion divided by the infused volume) is 50−80% as long as an infusion continues, while this fraction increases to 100% when the arterial pressure has dropped. Elimination is very slow during surgery, and amounts to only 10% of that recorded in conscious volunteers. Capillary refill further reduces the need for crystalloid fluid when bleeding occurs. These four factors limit the need for large volumes of crystalloid fluid during surgery

The publication of the big fluid trials however resulted in a scientific earthquake and cascade of reactions starting in November 2012 when the German medicines agency, the Federal Institute for Drugs and Medical Devices (Bundesinstitut für Arzneimittel und Medizinprodukte, BfArM) triggered a review on HES solutions by the European Medicines Agency (EMA) Pharmacovigilance Risk Assessment Committee (PRAC). The PRAC had initially recommended on June 13th last year, suspending marketing authorisations for infusion solutions containing HES in all patient populations.

This was followed by a re-examination request by stakeholders and on Oct 11th 2013, PRAC confirmed that HES solutions should no longer be used in patients with sepsis (bacterial infection in the blood) or burn injuries or critically ill patients, because of an increased risk of kidney injury and mortality. HES solutions may, however, continue to be used in patients to treat hypovolaemia (low blood volume) caused by acute blood loss, provided that appropriate measures are taken to reduce potential risks and that additional studies are carried out. On Oct 25th 2013 the Co-ordination Group for Mutual Recognition and Decentralised Procedures –Human (CMDh), endorsed the PRAC recommendations.

In late 2013, the results of the CRISTAL study became available showing that colloids – when given in patients with hypovolaemic shock – are lifesaving (significant reduction in 90-day mortality). The older gelatins, besides having a shorter half-life, smaller volume effect and a higher chance of developing anaphylactic reactions than HES, also lack actual scientific proof. Iso-oncotic human albumin was already shown not to be superior to saline in a general ICU setting many years ago in the SAFE study and there is no reason to assume this would be markedly different in a surgical population.

There is no evidence for the use of hyperoncotic albumin 20% either, a statement becoming even stronger after the recent ALBIOS trial that showed that in patients with severe sepsis, albumin replacement in addition to crystalloids, as compared with crystalloids alone, did not improve the rate of survival at 28 and 90 days. However, there was a significant beneficial effect observed in a post hoc subgroup analysis that included 1121 patients with septic shock, as compared with 660 without septic shock, at the time of enrollment (relative risk with septic shock, 0.87; 95% CI, 0.77 to 0.99; relative risk without septic shock, 1.13; 95% CI, 0.92 to 1.39; P=0.03 for heterogeneity).

The discussion continued with the publication of an open letter by Bellomo R et al. to the Executive Director of the EMA concerning the licensing of HES solutions for fluid resuscitation that was co-signed by 76 physicians. They concluded that it seems improbable that the PRAC recommendations: “that HES solutions should not be used for more than 24 hours and that patients’ kidney function should be monitored for at least 90 days” will guarantee patient safety. The adverse effects of HES appear to be generic to all HES classes and dose dependent, as such no safe dose for HES has been defined. The revised PRAC recommendation could mean that many thousands of patients with hypovolaemia and acute blood loss will continue to receive HES which will expose them to known risks of harm and offer no proven benefit.

So the discussion is going to a higher, almost religious level as this was followed by a counter-statement to the open letter to the Executive Director of the EMA concerning the licensing of HES solutions for fluid resuscitation by Coriat P et al. and co-signed by 77 physicians. They stated there is increasing evidence showing that there are relevant differences between the effects of the different products, with the best profile for the latest generation of starches and the authors emphasize that the conduct of further clinical studies is of high value to gain more information on the “best treatment” of surgical and trauma patients. Ultimately, it should be in everyone’s interest to interpret the existing data on medical topics objectively and neutrally, without rushing to premature, far-reaching conclusions, which could confuse physicians and even render future therapy with potentially life-saving drugs impossible. So, it seems that not making a choice is not an option either… But what about common sense, the majority of physicians are aware of the current status of knowledge of the risk-benefit assessment of HES, but how do they take this into account at the bedside?  For the time being, and awaiting new study results, the saga has stopped with the CMDh endorsement by majority of the recommendations of the European Medicines Agency’s PRAC, which concluded that HES solutions must no longer be used to treat patients with sepsis (bacterial infection in the blood) or burn injuries or critically ill patients because of an increased risk of kidney injury and mortality.


Resuscitation fluids

Resuscitation fluids are given to correct an intravascular volume deficit or in the case of absolute or relative hypovolemia. It is this particular indication that got by far the most scientific attention, especially in the light of the recent colloid-crystalloid debate. Therefore, it is sometimes overlooked that a large part of the total infused volume during a patient’s stay in the hospital does not fall into this category.

Acute intravascular volume depletion

This is most common related to polytrauma patients with hemorrhagic shock but it can also result from sepsis causing extreme inappropriate vasodilation and relative intravascular hypovolemia. Restoration of circulating blood volume is essential to improve survival in these patients.

Hypotensive resuscitation

The only reason for delayed resuscitation is a penetrating injury requiring surgical or angiographic intervention to obtain haemostasis, where mean arterial blood pressure should be maintained at a maximum of 60 mmHg (normally this corresponds to a palpable radial pulse) and a heart rate below 120 beats per minute. Restoration to pre-injury levels should be delayed until haemostasis has been secured. Low dose noradrenaline can be used while awaiting intervention to minimise clear fluid administration.

Small volume resuscitation

Hypertonic solutions of sodium chloride exist, and include 3%, 7% and 20% concentrations.  Hypertonic  saline solutions (HSS) can, related to his osmotic effect, attract free water from the interstitial space back into the intravascular space. In order to prevent the possible decline in effective arterial circulating volume related to hypovolemia, concomitant or preventive HSS administration can be given in order to preserve hemodynamics. This has the advantage of reducing endothelial and tissue oedema, and promoting perfusion. The effect can be prolonged by the addition of a colloid, although these are rarely used after the big fluid trials. These fluids are best used with haemorrhagic shock after achievement of haemostasis. Use in the prehospital environment cannot be recommended, even with associated traumatic brain injury, after the early termination of a study using prehospital hypertonic saline. During trauma resuscitation an initial bolus of 4ml/kg (approximately 250ml) should be given as early as possible after achieving haemostasis. This is often referred to as small volume resuscitation. Hypertonic sodium lactate has theoretical advantages and has been studied clinically in severe burns and postoperative patients. In summary, HSS simultaneously allows for rapid restoration of circulating intravascular volume with less administered fluid and attenuates the post-injury oedema at the microcirculatory level. It may also improve microvascular perfusion, but is not associated with an outcome advantage when used to restore systemic flow. Different effects, however, are noted when HSS is used as an adjunct to cerebral perfusion pressure management in those with traumatic brain injury and intracranial hypertension.


Maintenance fluids

Maintenance solutions are specifically given to cover the patient’s daily basal requirements of water and electrolytes. They are specifically intended to cover daily needs. These basic daily needs are:

  1. water – 25-30 mL/kg of body weight
  2. Sodium – 1 mEq/kg
  3. Potassium – 1 mEq/kg
  4. Glucose or dextrose 5 or 10% 1.4-1.6 g/kg

It is easily appreciated that with one litre of NaCl 0.9%, with a sodium content of 154 mEq/L the daily need for salt is already grossly exceeded. There is an ongoing debate, esp. in pediatric populations, about the tonicity of maintenance infusion. We strongly disagree that in their the recent review in the New England Journal of Medicine, Moritz and Ayus close a controversy that is far from settled. First, even in pediatrics it’s unfair to measure the quality of maintenance solutions mainly on an outcome parameter that can occur in only one treatment arm and thus proving that a solution containing 50 mEq/L of sodium entails a higher risk of lowering normal plasma sodium than a solution containing 154 mEq/L. Second, authors extrapolate their conclusions to adult care based on no more than one small 29-year-old trial in a specific subpopulation. In our opinion the majority of hyponatremia cases can be prevented by careful clinical judgement and follow-up by rigorously identifying (appropriate) ADH-secretion due to poorly corrected hypovolemia. Moreover it shouldn’t be neglected that hyponatremia often functions as indicator more than cause of severity of disease (e.g. cirrhosis, heart failure,…). And above all, quality evidence on deleterious effects of salt overload cannot be ignored. Instead of abandoning common practice and common sense, authors should realise their complaint of opinion-based scientific dissent is mutual. Although some specific maintenance solutions are commercially available in certain countries but they are far from ideal, none of them cover the precise daily needs of the average adult patient (mostly they contain too little potassium in relation to the salt and/or water content). Therefore, in practice it will be necessary to prepare them by adding electrolytes to readily available glucose or dextrose 5 or 10% preparations. Since the amount of fluids or electrolytes already administered by enteral intake or as part of total parenteral nutrition are to be subtracted from the maintenance prescription, the main strategies to reduce unnecessary fluid administration are the shortening of preoperative fasting regimens (see here & here) and the promotion of early postoperative feeding, even after abdominal surgery (see here & here). Pursuing this is imperative a first step in the reduction of fluid-related morbidity. In providing maintenance fluids care should be taken to avoid causing tissue oedema. This requires limitation of crystalloid administration.


Replacement solutions

Replacement solutions are prescribed to correct existing or developing deficits that cannot be compensated by oral intake, as seen in situations where fluids are lost via drains or stomata, fistulas, fever, open wounds (including evaporation during surgery), polyuria (salt wasting nephropathy or diabetes insipidus) among others.  Data on replacement fluids are sparse. Several recent guidelines advise to match the amount of fluid and electrolytes as closely as possible to the fluid that is being or has been lost. An overview of the composition of the different body fluids can be found in the NICE-guidelines above.  Most of the time isotonic balanced solutions will be just fine, although some diarrhoea can be hypotonic. An exception is the loss of gastric fluid (in case of nausea and vomiting), which is chloride rich and should be replaced by high chloride solutions, like normal saline. This may be the only indication for NaCl 0.9% administration. Replacement fluids are frequently overdosed in the perioperative setting due to the misconception that evaporation during surgery is always high. It was shown that even open abdominal wounds with liberal exposure of organs are associated with a fluid loss of no more than 30 ml per hour. Also, the practice of using diuresis as a trigger for fluid administration can easily lead to fluid overload since both anaesthesia and surgery slow down the rate of elimination of crystalloids. Oliguria is poorly correlated with hypovolemia in the perioperative setting and should not trigger fluid administration, although increased diuresis is a good indicator of hypervolaemia.


Nutritional and other fluids

Nutritional support is limited to patients with un-met nutrient requirements, documented inadequate oral intake, unpredictable return of GI function, or a prolonged period of bowel rest. In the latter case, every attempt should be made to provide early nutrition by the enteral route. Clear fluids containing electrolytes and both short chain polysaccharides and peptides are now available for preoperative hydration (up to 2 hours prior to elective surgery). After intra-abdominal surgery, a fine-bore nasojejunal feeding tube can be to allow immediate postoperative administration of semi-elemental enteral feed to maintain the integrity of the enterocytes and gut associated lymphoid tissue. Despite the beneficial effect of enteral nutrition (EN), EN fed critically ill patients often do not meet their nutritional targets, especially during the first days of ICU stay. Although adequate early nutrition is easier via the parental route, there is still a lot of controversy about the timing of the initiation (early vs late) of parenteral nutrition (PN) in critically ill adults in whom caloric targets cannot be met by EN alone, especially after the publication of the results of the EPaNIC trial. Casaer et al. found that there was no significant difference in mortality between late initiation and early initiation of PN among patients in the ICU who were at risk for malnutrition, despite the use of early EN plus micronutrients in a protocol that prevented hyperglycaemia. However, withholding of PN until day eight was associated with fewer ICU infections but a higher degree of acute inflammation. Late initiation of PN was also associated with a shorter duration of mechanical ventilation, a shorter course of RRT and a shorter ICU stay, despite a slight increase in hypoglycaemic episodes. Unlike the EPaNIC trial, which compared semi-starvation for one week to early glucose load followed by hypercaloric low protein PN within 48 hours, Heidegger et al. started the intervention on day four to maximise the potential for EN delivery, in keeping with ESPEN guidelines. Moreover, as opposed to the EPaNIC trial, their EN group was a true control group demonstrating cumulative increasing energy deficit (indirect calorimetry): 77±25% energy target vs. 104±16% (group with supplemental PN), and their population was composed exclusively of patients with a real indication of nutritional therapy, ie failure of EN on day three. Intravenous nutrition may be provided peripherally (5% dextrose) for a maximum of 5 days. After this period, or earlier with evidence of nutritional depletion, if there is absent or inadequate enteral nutrition, total parenteral nutrition should be provided. Phosphate levels should be checked prior to and after initiation of nutrition, if the period of starvation is longer than 48 hours. The refeeding syndrome is primarily due to acute hypophosphatemia. Crystalloids are required as vehicles for administration of medication, including antibiotics, sedation and inotropes / vasopressors. The fluid required for the administration of these medication and solutions together with those required for nutrition should ideally not exceed 2ml/kg/hour and their amount should be subtracted from the necessary maintenance fluids to provide the daily needs. The fluid chosen is usually 0.9% saline but 5% dextrose or glucose may be used if the serum electrolytes require.


The bottom line!

As of today, despite the recent large trials, we still don’t know which type of fluid is the best in each phase of shock in each type of patient. The ideal fluid simply does not exist. What we do know is that the best fluid is probably the one that has not been given unnecessarily and if we consider fluid administration to be like any other drug where the risks and benefits are weight up, then we are on the way to improving overall safety. Fluids must be guided via a protocolled approach and fluid responsiveness must be assessed upfront. Above all, fluid overload must be avoided and de-resuscitation needs to be considered as soon as is feasible Taking this into account, it seems that a dynamic approach starting with crystalloid maintenance, (covering daily needs for water, electrolytes and glucose) and resuscitation fluids, followed by colloids (sepsis and burns excluded) or albumin at a later stage is our best option. So, it’s all about the choice, the timing, the dose, and the speed of administration. Unfortunately, one size does not fit all and we should only give fluids for one of the four indications as discussed herein:

  1. Resuscitation
  2. Maintenance
  3. Replacement
  4. Nutrition


Post thought – Dr Jonny Wilkinson

The article you have just read, including some landmark references, highlights the complexity and relative indecision regarding what is right and what is wrong over IV fluid therapy. What is clear is that it requires an expert knowledge base in order to decrease the significant harm mal-prescrition and administration can cause. It is my firm belief that we should all attempt to be as ‘up to date’, as we can within this field, and have a ‘Go-to’, expert we can contact for controversial and complex decision making. What I also believe is that all trusts should have a clinical lead for IV fluid therapy (happens to be yours truly in ours!)

Having got to know more about this subject of late, as well as meeting Manu Malbrain, I certainly feel I am on the way! SO don’t let the battle between Scylla and Charybdis take place on your unit…well at least, not too often anyway!

Finally – why not come along to Antwerp and meet us all where you can find out more. Thanks so much to Manu again for all of his hard work and for choosing us to feature it!