This site is intended for health professionals only!

The Great Fluid Debate


Poor understanding of plasma volume expanders and fluid management hampers effective treatment for surgical patients. More rational prescribing and monitoring of treatment is needed

Christine Clark

Contributing Editor

In July 2008 in London, a satellite meeting of the Evidence Based Peri-Operative Medicine Conference (EBPOM) 2008 called The Great Fluid Debate – sponsored by B. Braun – provided a state-of-the-art review of plasma volume expanders and fluid management. Topics included the pharmacology of intravenous
fluids, physiological changes during the perioperative period, and common misunderstandings about the appropriate use of plasma expanders and crystalloids.

Many of the fluid management problems seen in surgical patients start at the preoperative stage, according to Gordon Carlson (professor of surgery, Hope Hospital, Salford, UK).

A typical patient undergoing bowel surgery (eg, a left hemicolectomy) would undergo “bowel preparation” that comprises three litres of Klean-Prep orally, no solid food for 48 hours, and no nutritional supplements. As a result, the patient is likely to be dehydrated by the time of surgery. Although bowel preparation is widely believed to reduce the occurrence of anastomotic leakage and infections, a recent systematic review has shown that, in fact, it significantly increases the rate of anastomotic leakage and does not reduce infections, said Professor Carlson. Moreover, preoperative fasting is not necessary; nonparticulate fluids and jelly can be taken up to two hours before surgery, and preoperative carbohydrate “loading” might also be beneficial, he added.

Fluid overload in postoperative patients is a common finding, because intravenous fluids tend to be prescribed without regard for patients’ actual needs. Patients are often oliguric at this time as part of the natural response to the stress of surgery. In healthy individuals, urine is normally produced at the rate of 0.5ml/kg/h, but in the first 24 hours after surgery urine production is slower. This prompts some doctors to prescribe fluids to restore urine output, explained Professor Carlson. In fact, during the postsurgery period there is increased aldosterone, angiotensin and antidiuretic hormone activity, along with a reduced glomerular filtration rate, and so sodium is actively retained and urine output is low. Studies have shown that patients can receive so much fluid in this period that they can have a positive fluid balance of as much as 11L. The sodium dose can range from 420 to 1,700mmol, when the normal daily requirement is 70mmol. There is a strong association between iatrogenic morbidity and sodium load, he said.

Provided that the patient is not hypovolaemic, intravenous fluids should not be given, but urine output should be monitored. Results should be evaluated over four to six hours, recommended Professor Carlson.

Although fluid loading increases the central venous pressure, it does not increase splanchnic blood flow. Indeed, if intravenous fluids are restricted to match patients’ needs there is better bowel recovery, but few surgical units have a policy of restricting fluids, he said. The traditional approach to surgery often involves a sequence of dehydration, over-rehydration and stress, together with pre- and postoperative starvation, all of which results in a longer hospital stay. Enhanced recovery can be achieved by reducing the amount of physiological interference to the minimum, Professor Carlson concluded.

Intravenous fluids should be prescribed in the same way as other medicines, and pharmacists should challenge inappropriate prescriptions more often, according to Dileep Lobo (associate professor and reader in gastrointestinal surgery, University of Nottingham, UK). Prescribing of intravenous fluids is usually left to junior doctors, and surveys have shown that the standard of prescribing is generally poor. A telephone survey
of 200 junior doctors (see Resources) found that in 89% of instances fluid prescribing was the responsibility of the most junior member of the team. Few respondents knew the sodium and potassium content of commonly used intravenous crystalloids and colloids. More than one-quarter were prescribing more than 300 mmol of sodium and chloride per day – which
is more than three times the daily requirement – but less than half prescribed any potassium.

The administration of 0.9% saline, which contains significantly more chloride than plasma, can lead to fluid overload and hyperchloraemia that lasts for more than six hours in normal volunteers. The administration of 0.9% saline causes weight gain and reduces albumin levels and haematocrit (packed cell volume) as a result of plasma dilution, whereas an equivalent
volume of Hartmann’s solution, which contains less sodium and chloride, does not, said Mr Lobo.


A survey of inpatients with serum sodium levels of greater than 149mmol/l found that such patients were almost always in intensive care areas and had received large amounts of intravenous saline.

Salt and water overload disrupts almost every body system. Critically, hyperchloraemia can also cause renal vasoconstriction and a reduced glomerular filtration rate, and this leads to further sodium retention. Evolution has geared the human body to retain sodium efficiently, but it does not have such effective mechanisms for excreting excess sodium, said Mr Lobo.

Recent studies have compared the results of standard postoperative fluid regimens with “restricted” regimens designed to match the patients’ needs. One study showed that the standard group received more fluids and experienced delayed gastrointestinal recovery (see Resources). Another study showed that when patients received more than 5.5L of fluid on the day of operation, more than 60% experienced complications
(see Resources).

“It is not restriction that makes the difference; it is overload,” he explained. Sodium and water overload can also give rise to splanchnic oedema, raised intraabdominal pressure and decreased mesenteric blood flow – which, in turn, are associated with complications such as wound dehiscence and ileus. It can take up to three weeks to excrete the excess fluid load, and, as
the weight falls, the serum albumin usually rises.

In conclusion, Mr Lobo said that it is very easy to give salt and water, but very difficult to remove them. He recommended that oral fluids should be given whenever possible, because the gut is an effective regulator of sodium and water intake that is bypassed by intravenous administration.

Saline infusion was sometimes described as “the poor man’s plasma expander”, although only 10-20% of the infused volume stays in the intravascular space, said Peter Gosling (consultant clinical scientist and honorary senior clinical lecturer, University Hospital Birmingham, UK). Both 0.9% sodium chloride and 5% dextrose are crystalloids, and their main uses are to replace electrolytes and to provide free water, respectively. Colloids, such as hydroxyethyl starch and albumin, are molecules that are large enough to be retained by the vascular endothelium and so trap water in the intravascular space. Albumin, for example, can hold 18 times its weight of water. Their effectiveness depends on molecular size, the rate of degradation and the permeability of the endothelium. Capillary permeability varies between tissues, with liver and lung being relatively “leaky” by virtue of larger pores in the capillary walls compared with those in skeletal muscle and subcutaneous tissue. Moreover, capillary permeability increases sharply in response to trauma, surgery or burns, and then returns to the normal level after about 12 hours. In general, colloids remain in the intravascular space, sodium chloride is distributed in the intravascular space and the interstitial space, and dextrose is distributed in intravascular, interstitial and intracellular compartments, explained Dr Gosling.

Modern colloid solutions are based on hydroxyethyl starches, which have high molecular weights (70,000-450,000 daltons) and can provide volume expansion for 6-24 hours. The hydroxyethyl groups impede degradation of the starch by amylases. The nomenclature describes the extent of hydroxyethyl substitution: for example, the prefix “tetra” indicates 40% substitution, and “penta” 50%.

Hydroxyethyl starches provide good, sustained volume expansion, and deliver less sodium and chloride than normal saline. Pentastarch (50% substituted hydroxyethyl starch) carries the risk of accumulation if repeated infusions are given, but tetrastarch, which is now available in a balanced salt solution, appears to provide volume expansion for 6-8 hours with little risk of accumulation or capillary leakage. However, if these products are given with insufficient water, then there is a high risk of hyperoncotic acute renal failure. This occurs because the oncotic pressure of plasma is raised to the point where it effectively opposes the filtration pressure in the kidneys, thereby impairing normal glomerular filtration.

It is important to avoid perioperative hypovolaemia, especially in patients who are at risk of developing acute kidney injury (AKI), Andrew Lewington (consultant renal physician, St James University Hospital, Leeds, UK) told the audience. Risk factors for AKI include chronic kidney disease, diabetes mellitus, vascular, cardiac or liver disease, and age over 65 years. Small rises in serum creatinine are now recognised to have a large impact on mortality. A rise of 25mmol/l, or a 1.5-fold increase from baseline, over an interval of 48 hours is the critical value, he explained.

Clinicians should “go back to first principles” and make a careful clinical evaluation before diagnosing hypovolaemia. Signs include decreased skin turgor, reduced peripheral perfusion and reduced jugular venous pressure. Reduced urine output is also a feature,
but this can be difficult to interpret in the first 24 hours after surgery, he acknowledged. Failure to give appropriate intravenous fluids to a hypovolaemic patient can lead to AKI. Fluid administration must be monitored carefully to avoid sodium and fluid overload with crystalloids and precipitation of a
hyperoncotic state with colloids.

The problems associated with AKI include impaired solute and water excretion, a loss of the kidneys’ ability to autoregulate their blood supply, complications affecting all cardiovascular, gastrointestinal, haematological and neurological systems, and depressed immune function.

Surgical patients are at risk because of the frequency of excess perioperative fluid administration. The resulting hyperchloraemia reduces glomerular filtration rate and thereby further impairs renal function.


Safe postoperative fluid management requires cardiac output measurement, argued Mike James (professor and head of surgery, Department of Anaesthesia, University of Capetown, and Groote Schuur Hospital, South Africa). There is an ideal fluid volume for everyone – too much and too little are both bad – and fluid optimisation is associated with more rapid recovery, he continued.

Measures of cardiovascular status, such as blood pressure and central venous pressure, are unreliable and are slow to indicate changes, whereas dynamic measurements, such as stroke volume, respond to changes early. This has led to the concept of “recruitable stroke work”, in which increasing fluid load leads
to an increase in cardiac output. In order to make use of this in practice, a reliable and easy measure of cardiac output is needed. A small volume of fluid can then be given and the response assessed to see whether there is “recruitable volume” or not.

“The objective is to get close to the top of the ‘Starling’ curve,” explained Professor James. In the event of an increase in cardiac output, further fluids can be given. “In this way we can move from [blood-] pressure-guided therapy to flow-guided therapy; flow is important because it is this that regulates most organs, and not blood pressure.”

Research shows that oesophageal Doppler ultrasound monitoring, which measures aortic blood flow, is associated with a highly significant reduction in complications and admissions to intensive care, explained Professor James (see Resources). Moreover, it is the only method that has been assessed by the NHS Centre for Evidence-Based Purchasing as having
“significant potential” (see Resources).

Evidence Based Peri-Operative Medicine (EBPOM)

Lobo DN, et al. Problems with solutions: drowning in the brine of an inadequate knowledge base. Clin Nutr 2001;20:125-30.

Lobo D, et al. Effect of salt and water balance on recovery of gastrointestinal function after elective colonic resection: a
randomised controlled trial. Lancet 2002;359:1812-18.

Brandstrup B, et al. Effects of intravenous fluid restriction on postoperative complications: comparison of two perioperative fluid regimens: a randomised assessor-blinded multicenter trial.
Ann Surg 2003;238:641-8.

Abbas SM, et al. Systematic review of literature for the use of esophageal Doppler monitor for fluid replacement in major abdominal surgery. Anaesthesia 2008;63(1):44-51.

Evidence Review: Oesophageal Doppler monitoring in patients undergoing high-risk surgery and in critically ill patients – CEP 08012 (Apr 08)

Most read

Latest Issue

Be in the know
Subscribe to Hospital Pharmacy Europe newsletter and magazine