Perioperative fluid therapy
BMJ 2012; 344 doi: https://doi.org/10.1136/bmj.e2865 (Published 26 April 2012) Cite this as: BMJ 2012;344:e2865
- 1Barts and The London School of Medicine and Dentistry, Queen Mary’s University of London, London EC1M 6BQ, UK
- 2University College London/University College Hospitals NHS Trust, Wolfson Institute for Biomedical Research, London WC1E 6BT
- Correspondence to: R Pearse Adult Critical Care Unit, Royal London Hospital, London E1 1BB r.pearse{at}qmul.ac.uk
Patient outcomes after major non-cardiac surgery can be improved considerably through more effective perioperative care.1 Factors such as advancing age, comorbidity, and complex surgical procedures can result in postoperative morbidity and mortality rates similar to those found with common acute medical emergencies.1 2 3 Patients who survive postoperative complications experience functional limitation and reduced long term survival.1 2 3 Doctors often prescribe intravenous fluid with limited knowledge of the benefits and risks of this treatment. Doctors in training commonly express frustration at the lack of clear guidance on the optimal approach to fluid therapy. The debate that followed recent UK guidelines aimed at standardising best practice highlights the uncertainty in this area, even among experienced practitioners.4 5
We review the evidence from clinical studies, systematic reviews, and practice guidelines to provide a summary of current best practice in the prescription of fluid for patients undergoing major non-cardiac surgery.
Methods
We searched various databases, including Clinical Evidence and the Cochrane Collaboration, for articles which would define clinical practice by using the search terms: surgery; fluid, intra-venous; perioperative; blood transfusion. We also consulted several UK and international experts in formulating the contents of this review and sought comments on the final version.
We recognise the considerable variation in clinical practice in this area. The article focuses on key aspects of the topic affecting the majority of patients who would benefit from a standardised and individual approach to fluid prescribing. Our aim is to provide simple guidance for the less experienced or less expert clinician who prescribes fluid on a regular basis. We stress the importance of recognising that fluids are drugs and accordingly require expert guidance in their use.
What are the principles behind fluid therapy?
In health, 60% of total body mass consists of water. Most water resides within the intracellular compartment, separated from extracellular water, which comprises the interstitial and plasma volumes. The neuroendocrine response to surgery results in retention of sodium and water with a reduction in maintenance requirements. Conversely, absolute hypovolaemia (blood loss) and relative hypovolaemia (such as epidural or inflammation mediated vasodilatation) commonly result in a fluid deficit. For most patients fluid losses are replaced during surgery and oral intake of fluid is rapidly resumed after surgery. However, for some procedures (such as gastrointestinal surgery, proximal femoral fracture repair), the preoperative deficit and losses during surgery vary widely and may not be adequately replaced.
Inadequate fluid replacement leads to reduced cardiac output and oxygen delivery to injured tissues, which is associated with an excess of postoperative complications. Excessive fluid administration may also have adverse effects, including acidosis, coagulation defects, and oedema of both lungs and peripheral tissues (fig 1⇓).6 It is also worth noting that postoperative adverse events may be attributed to fluid prescribing when associated factors are to blame. The tissue injury of surgery results in a systemic inflammatory response associated with both tissue oedema and hypovolaemia. Negative fluid balance after surgery is associated with reduced mortality (odds ratio 0.50 (95% confidence interval 0.28 to 0.89)),7 although this may reflect the degree of inflammatory response as well as excessive fluid administration.8 Fluid restriction and diuresis may decrease oedema in patients with poor ventricular function but also increase the incidence of acute kidney injury.
Fig 1 Optimal perioperative fluid therapy requires a balance of the beneficial and adverse effects of intravenous fluid. This requires an individualised approach to prescribing that is often neglected, resulting in poor patient outcomes
How should we select the dose of intravenous fluid?
Perioperative maintenance fluid
The normal daily dietary requirements for water and electrolytes are listed in table 1⇓. However, retention of sodium and water after surgery reduces their requirements. Additional amounts should be given only to correct deficit or continuing losses. Monitoring should include clinical examination, fluid balance charts, regular weighing, and biochemical analyses (urea, electrolytes, creatinine, bicarbonate).
Normal daily requirements for fluid and electrolytes for a healthy adult
It is often helpful to calculate the quantity of water, sodium, and potassium prescribed in a given fluid regimen (table 2⇓). The optimal daily dose of water and electrolytes cannot be provided by a single crystalloid solution, and a suitable daily prescription for maintenance fluid will normally include more than one formulation. Many fluid solutions contain large quantities of sodium and chloride, which may result in nausea and vomiting, metabolic acidosis,9 and impaired renal blood flow.10 However, although hypotonic solutions are important sources of water, they may cause severe hyponatraemia and neurological impairment when used as replacement fluids. Analysis of 38 randomised trials involving 1589 patients undergoing major abdominal aortic surgery failed to identify specific superior fluid regimens.11 Although crystalloid solutions that contain bicarbonate may limit the deleterious effects of chloride, the clinical and metabolic consequences of limiting chloride-rich fluids remain unclear.9
Constituents of commonly used crystalloid and colloid based intravenous fluids
Replacement of perioperative fluid deficit
The most important information required to assess intravascular volume is provided by the clinical scenario; is it likely that the patient is hypovolaemic? Estimates of fluid deficit based on traditional physiological parameters such as heart rate, blood pressure, and central venous pressure are not reliable. A systematic review of 24 studies showed that central venous pressure is a poor measure of fluid deficit (pooled correlation between central venous pressure and change in cardiac output 0.11 (95% confidence interval 0.02 to 0.21)).12
Assessing the dynamic response of physiological variables to a fluid challenge is a more instructive approach. The response to the rapid administration of a fluid bolus (in our practice, 250 ml colloid solution) may be evaluated during surgery by monitoring cardiac output and is best guided by an algorithm for perioperative fluid and inotropic therapy (fig 2⇓). This treatment approach is associated with a mortality reduction of 37% and a two or three day reduction in length of hospital stay.13 The National Institute for Health and Clinical Excellence (NICE) has endorsed the use of perioperative cardiac output monitoring while acknowledging the need for further research.14
Fig 2 Use of cardiac output monitoring to guide replacement fluid therapy during and immediately after major surgery is much more reliable than use of venous pressure. A physiological “challenge” with a small bolus of intravenous fluid will result in an increase in venous return and hence stroke volume, but only in patients with a fluid deficit. Thus both inadequate and excessive fluid administration may be avoided.
For most patients, resumption of oral fluid and light diet can begin shortly after surgery. Patients undergoing major gut or vascular surgery require individualised prescription of intravenous fluid, and the clinical team should communicate to ensure optimal fluid administration during and after surgery. These patients should be reassessed on a regular basis by suitably trained staff.
After major surgery, a substantial proportion of patients will become critically ill. In this scenario decisions regarding fluid and inotropic therapy should be guided by senior medical advice. This is best determined by objective, dynamic assessment of cardiovascular performance in response to a fluid challenge and may also be guided by a basic transthoracic echocardiogram to assess left ventricular function. Lactate and central venous saturation measured from a blood sample drawn from a central venous catheter may also indicate hypovolaemia. This approach is currently being evaluated in clinical trials. Initial care of patients who become critically ill on a standard surgical ward can be improved by input from critical care outreach staff. However, this resource cannot replace prompt admission to a critical care unit.2 3
What are the differences between crystalloid and colloid solutions?
Intravenous fluids should be considered as conventional drugs, with both beneficial and adverse effects. Differences in the chemical structure of colloid and crystalloid solutions may explain their diverse metabolic effects (table 2⇑). There is no evidence that resuscitation with colloids (including albumin) reduces the risk of death or morbidity compared with resuscitation with crystalloids (pooled relative risk 1.00 (0.91 to 1.09)).15 Safety concerns have emerged from several studies regarding a potential increase in the risk of bleeding and acute kidney injury with different colloids. However, systematic review of these data is limited by the lack of statistical power and inconsistent definitions of kidney injury.16 There are several ongoing clinical trials comparing the effects of crystalloid and colloid solutions. Until these studies are complete, wide international variations in fluid prescribing practice are unlikely to change.17
When should we transfuse blood to a surgical patient?
Increased mortality among patients undergoing non-cardiac surgery is associated with both preoperative anaemia (odds ratio 1·42 (1·31 to 1·54)) and perioperative haemorrhage.18 19 However, the benefits of blood transfusion are uncertain, and further research is required in this area.20 The adverse effects associated with transfusion are well described,21 and, for surgical patients, include increased risks of postoperative infection and increased recurrence after cancer surgery.22 23 Current guidelines recommend transfusion when haemoglobin values fall below 70 g/L during the perioperative period but do not recommend transfusion to achieve values >100 g/L.24 A recent large randomised trial in patients undergoing proximal femoral fracture repair, which randomised patients to transfusion triggers of 100 g/L versus 80 g/L, did not show any differences in functional outcome (odds ratio 1.01 (0.84 to 1.22)).25 Consideration should also be given to the use of techniques which reduce transfusion requirements, including iron and erythropoietin, antifibrinolytic therapy, and red cell salvage devices.23 Systematic reviews indicate that use of autologous transfusion of the patient’s own blood (donated before surgery) is not associated with benefit and may increase transfusion rates.26
How should we optimise perioperative nutrition?
Surgical patients should be screened for nutritional deficit and managed according to published guidelines for perioperative nutritional support.27 In the small proportion of patients who present for surgery with a severe nutritional deficit, caution should be taken with re-feeding after surgery either by enteral or parenteral routes. The advice of a specialist dietician should be sought. Most patients without disorders of gastric emptying who are undergoing elective surgery should be fasted for six hours for solids (including milk), but clear non-particulate oral fluids should not be withheld for more than two hours before induction of anaesthesia.28 Preoperative administration of carbohydrate-rich drinks two to three hours before induction of anaesthesia may enhance recovery from surgery.29 These approaches should be considered for most patients in the preparation for elective surgery. Meanwhile, routine use of preoperative mechanical bowel preparation is not beneficial and should be avoided, unless indicated surgically.29 30 Most patients do not require intravenous fluid once surgery is complete and can rapidly resume oral fluids and light diet.
Conclusions
Perioperative fluid therapy may have important beneficial effects on outcome after surgery. Inappropriate fluid management is likely to harm patients, but fluid prescribing practice still varies widely. Forthcoming guidelines from NICE will encourage a standardised approach to fluid prescription and management. Such initiatives are welcome and should be widely implemented to ensure the highest standards of patient care.
Learning points
Perioperative fluid therapy is associated with potential benefits and harm, yet fluid prescribing practice varies much more widely than would be accepted for other drugs
The inflammatory response generated by tissue injury causes complex changes in fluid and electrolyte balance. Although maintenance requirements are reduced, many patients require replacement of fluid losses
Careful monitoring should be undertaken using clinical examination, fluid balance charts, and regular weighing when possible
For most patients fluid losses during surgery can be replaced with resumption of oral fluid and light diet after surgery
Patients undergoing complex major surgery (such as major gastrointestinal surgery, proximal femoral fracture repair) experience variable fluid losses due to tissue injury, inflammation, and blood loss. For these patients, evidence supports an individualised approach to fluid therapy often involving cardiac output monitoring during surgery
Fluid restriction and diuresis may decrease oedema in patients with poor ventricular function but may increase the incidence of acute kidney injury. It is better to avoid excess fluid administration than to treat the consequences
Notes
Cite this as: BMJ 2012;344:e2865
Footnotes
Contributors: All authors contributed to each stage of the preparation of this manuscript and approved the final version. RMP is the guarantor.
Funding statement: RMP is a National Institute for Health Research (NIHR) clinician scientist. GLA is supported by an Academy of Medical Sciences/Health Foundation clinician scientist award. This work was undertaken at University College London Hospitals NHS Trust, University College London, which received a proportion of funding from the Department of Health UK NIHR Biomedical Research Centre funding scheme.
Competing interests: All authors have completed the unified competing interest form at www.icmje.org/coi_disclosure.pdf (available on request from the corresponding author) and declare that we have received no support from any company for the submitted work; RMP has received a research grant from LiDCO and honorariums from Covidien, Pulsion Medical Systems, and Edwards Lifesciences—companies that might have an interest in the submitted work. The authors have no other relationships or activities that could appear to have influenced the submitted work.