Choice of resuscitation fluids in critically ill adults: key messages from the European Society of Intensive Care Medicine 2024 clinical practice guidelines
Key words: critical care, critical illness, fluids, guidelines, intensive care
CC BY 4.0
Choice of resuscitation fluids in critically ill adults: key messages from the European Society of Intensive Care Medicine 2024 clinical practice guidelines
CC BY 4.0
The 2024 European Society of Intensive Care Medicine clinical practice guidelines provide clinicians with evidence‑based recommendations on intravenous fluids in critically ill adults across a range of common conditions. These guidelines aim to improve the practices of fluid therapy by adopting a global perspective that considers both clinical efficacy and resource utilization in diverse health care settings. The guidelines address 3 key areas: 1) albumin vs crystalloids, 2) balanced crystalloids vs isotonic saline, and 3) small‑volume hypertonic solutions vs isotonic crystalloids. The recommendations were developed using the Grading of Recommendations, Assessment, Development, and Evaluation approach, ensuring a rigorous and transparent evaluation of the evidence across critical outcomes. Albumin was generally not favored over crystalloids due to its lack of demonstrated mortality benefit, higher cost, and limited availability in resource‑limited settings. Balanced crystalloids were suggested over isotonic saline in most scenarios due to their potential to reduce mortality and mitigate harmful effects of hyperchloremia, although the certainty of evidence was low. Small‑volume hypertonic solutions were not shown to provide significant advantages over isotonic crystalloids, leading to a preference for the latter based on a very low certainty of evidence. This work provides an overview of the guideline development process and a detailed summary of their recommendations, highlighting key considerations for clinical practice. The guidelines also identify critical evidence gaps in fluid therapy research, underscoring the need for future studies to refine and optimize fluid management strategies in critically ill patients.
Introduction
Intravenous fluids are a common therapy in critically ill patients, especially during the acute resuscitation phase, where restoring intravascular volume is essential to maintain adequate oxygen delivery to peripheral tissues.1,2 Selecting their appropriate type and dose is essential to achieving the desired benefits while minimizing undesirable effects. This requires an understanding of human physiology and the patient’s unique characteristics.2
Intravenous fluids are broadly classified into crystalloids or colloids. Over the years, concerns about high chloride content in normal saline and its consequences have led to the development of solutions in which chloride is replaced with other anions to better mimic the composition of the extracellular fluid.3 These solutions, known as balanced or buffered crystalloids, include examples such as Ringer lactate / acetate and plasmalyte.3
Colloids, on the other hand, are either derived from human blood components—such as packed red blood cells, fresh frozen plasma, and albumin—or consist of synthetic molecules, such as starches and gelatin.3 Theoretical advantages of colloids include a longer plasma expansion time and lower volume requirements, which may limit interstitial edema. A systematic review of 31 observational studies (n = 31 706) found an association between positive fluid balance and volume overload and a risk of death in critically ill patients.4 Earlier studies reported a volume‑expansion ratio of 1:3 for albumin as compared with crystalloids.3 In contrast, data from the landmark randomized clinical trial (RCT) SAFE (Saline versus Albumin Fluid Evaluation) showed a ratio between 1:1.2 and 1:1.6.5 Despite these theoretical advantages, a Cochrane review of 69 studies (n = 30 020) did not find an effect on patient‑important outcomes.6
An international survey of 391 intensive care units (ICUs) showed significant variability in the choice of fluids, with colloids being prescribed more frequently than crystalloids, despite costs and concerns about harm.7 In addition, more than 60% of albumin use is not supported by guidelines.8
Recognizing the importance of this issue, the European Society of Intensive Care Medicine (ESICM) initiated a multipart, evidence‑based clinical practice guideline project. This project focused on 3 key aspects of fluid therapy in critically ill adults: fluid choice (part 1), fluid dose (part 2), and de‑escalation of fluids (part 3).9 In this paper, we discuss the methodology, recommendations, and supporting evidence from the first part of the guideline that focuses on fluid choice.9
Review of the guideline’s methodology
The ESICM guideline panel consisted of clinical chairs and a diverse group of international subject matter experts, supported by methodology expertise from ESICM and the Guidelines in Intensive Care, Development and Evaluation group. The panel adhered to the Grading of Recommendations, Assessment, Development, and Evaluation (GRADE) approach to structure questions in the population, intervention, comparator, and outcomes (PICO) format.10
The panel compiled a list of patient‑important outcomes, then prioritized them using a 9‑point Likert scale. Outcomes with an average rating of 7 or higher were considered critical for decision‑making and were included.10 The final list of prioritized outcomes included mortality, renal replacement therapy (RRT), ventilator‑free days or duration of mechanical ventilation, ICU- and hospital‑free days (or ICU and hospital length of stay), quality of life, and functional outcomes.
Methodologists conducted a comprehensive systematic literature search of Medline and Embase, abstracted relevant data, performed meta‑analyses, and provided an overall rating of the certainty of evidence for each outcome in accordance with the GRADE approach.11 The GRADE assessment of certainty of evidence evaluates the following domains: risk of bias, inconsistency, indirectness, imprecision, and publication bias. Methodologists may downgrade the certainty of evidence based on their assessment of these domains. In general, the overall certainty of evidence across all outcomes is that of the critical outcome with the lowest certainty.11 Certainty of evidence ranges from high to very low with different implications. High certainty implies a high confidence that the true effect of an intervention is close to the estimate provided. Moderate certainty, on the other hand, implies a lower degree of confidence in that the true effect most likely is the one provided by the estimate; however, it may be significantly different. Low certainty implies limited confidence, and the true effect may be significantly different than its estimate. Finally, very low certainty implies very little confidence in the estimate of the effect and the truth is very likely to be different than the estimate provided.12
For each recommendation, the evidence profile was integrated into the evidence‑to‑decision framework to help the panel reach consensus on the direction and strength of the recommendation.13 The strength of a recommendation depends on the balance between desirable and undesirable effects, the certainty of the effect estimates on the patient‑important outcomes, patient values and preferences, resource utilization, and other factors.13
A strong recommendation indicates that the intervention improves patient‑important outcomes (or vice‑versa) and should be implemented in most circumstances. Strong recommendations are worded as “we recommend.” Conversely, a conditional (or weak) recommendation reflects either less robust evidence of a benefit or significant uncertainty about the intervention’s utility. Weak recommendations are worded as “we suggest.”14
Ultimately, the guideline addressed 11 PICOs: 6 focused on albumin vs crystalloids, 4 on balanced crystalloids vs isotonic saline, and 1 on small‑volume hypertonic solutions vs isotonic crystalloids. Furthermore, a comparison with other relevant guidelines on specific PICOs is provided in Table 1.
Setting | ESICM 2024 Fluids Choice Guidelines9 | ESICM 2019 task force expert panel recommendation45 | ESICM 2018 Fluid therapy in neurointensive care patients29 | Surviving Sepsis Campaign Guidelines 202118,19 |
Abbreviation: ESICM, European Society of Intensive Care Medicine | ||||
Albumin vs crystalloids | ||||
All critically adults | Weak recommendation for crystalloids over albumin (moderate certainty) | Overall, no evidence to abandon isotonic saline | – | – |
Critically adults with sepsis | Weak recommendation for crystalloids over albumin (moderate certainty) |
| – |
|
Critically ill adults with acute respiratory failure | Weak recommendation for crystalloids over albumin (very low certainty) | – | – | – |
Critically ill adults with traumatic brain injury | Weak recommendation for crystalloids over albumin (very low certainty) | – |
| – |
Critically ill adults in perioperative period or at a high risk of bleeding | Weak recommendation for crystalloids over albumin (very low certainty) | – | – | – |
Critically ill adults with cirrhosis without other established indication for albumin | Weak recommendation for crystalloids over albumin (very low certainty) | – | – | – |
Balanced crystalloids vs isotonic saline | ||||
All critically ill adults | Weak recommendation for balanced crystalloids over isotonic saline (low certainty) | Balanced crystalloids should be preferred if large volumes of isotonic saline are used | – | – |
Critically ill adults with sepsis | Weak recommendation for balanced crystalloids over isotonic saline (low certainty) | – | – | Weak recommendation for balanced crystalloids over isotonic saline (low certainty) |
Critically ill adults with traumatic brain injury | Weak recommendation for balanced crystalloids over isotonic saline (very low certainty) | – | Weak recommendation for crystalloids as the first‑line fluid in neurointensive care patients with hypotension | – |
Critically ill adults with acute kidney injury | Weak recommendation for balanced crystalloids over isotonic saline (very low certainty) | – | – | – |
Small‑volume hyperoncotic solutions vs isotonic crystalloids | ||||
Weak recommendation for isotonic crystalloids over small‑volume hyperoncotic solutions (very low certainty) | – | – | – | |
Albumin vs crystalloids
The ESICM guidelines evaluated the use of albumin vs crystalloids in critically ill adults, both overall and in specific clinical contexts, including sepsis, acute respiratory failure, traumatic brain injury (TBI), perioperative or high risk of bleeding, and cirrhosis.
Albumin vs crystalloids in critically ill adults
The guidelines issued a weak recommendation, based on moderate certainty of evidence, favoring the use of crystalloids over albumin. The recommendation was driven by a lack of benefit of albumin on mortality, RRT, duration of mechanical ventilation, and ICU or hospital length of stay. In addition, the high cost of albumin was deemed prohibitive in the absence of a clear benefit, and data on its cost‑effectiveness were limited.15 The panel noted that the access to albumin is often challenging or nonexistent in resource‑limited settings, raising issues of inequity. Finally, as a human blood product, albumin may be unacceptable to certain patients due to cultural or personal reasons. These considerations are relevant to all subsequent albumin recommendations.
Albumin vs crystalloids in adults with sepsis
Similarly, the guidelines issued a weak recommendation favoring crystalloids over albumin in adults with sepsis, based on moderate certainty of evidence. While the justifications mirrored those for the general critically ill in the general population, a few points specific to sepsis were highlighted.
Multiple mechanisms are thought to be responsible for organ dysfunction in sepsis. Vascular endothelial injury plays an important role in the pathophysiology of sepsis and manifests as a loss of vessel wall permeability, hypotension, and excessive third‑spacing of extravascular fluid into interstitial tissues culminating in reduced oxygen delivery to the tissues and organ dysfunction.16,17 Theoretically, albumin could reduce tissue edema by increasing oncotic pressure. Consequently, the panel considered specific subgroups of patients who might benefit from albumin, such as those with hypoalbuminemia, those who received large volumes of crystalloids, and those with septic shock. The Surviving Sepsis Guidelines for Management of Sepsis and Septic Shock 2021, for example, issued a weak recommendation for adults with sepsis or septic shock who received large volumes of crystalloids.18,19 However, the panel found limited data to inform a recommendation in these subgroups, therefore refrained from making a recommendation for or against albumin, instead highlighting it as a research gap for future studies. A recent expert consensus document on the management of sepsis in resource‑limited setting highlighted the limited access to albumin and did not include it in any recommendation.20
Albumin vs crystalloids in critically ill patients with respiratory failure
Acute respiratory failure is a common condition, and patients often need invasive ventilatory support. The most common etiology is infectious, which when severe can lead to acute respiratory distress syndrome.21,22 It is characterized by acute inflammation with increased capillary membrane permeability and noncardiogenic pulmonary edema. Chronologically, the pathophysiology is broadly divided into 3 phases: exudative, proliferative, and fibrotic. In the exudative phase, immune cells damage the endothelium and alveolar membrane leading to accumulation of protein‑rich fluid within these compartments.23 The alveolar and interstitial fluid accumulation increases lung weight, reduces compliance, and results in hypoxemia.23 As such, avoiding worsening of the edema by reducing fluid amounts or administering fluids that shall achieve increased intravascular pressure at a smaller volume seem physiologically appealing and may ameliorate hypoxemia. Once again, albumin seems to be promising in achieving these objectives. However, the guideline issued a conditional recommendation favoring crystalloids over albumin in adults with acute respiratory failure, based on a very low certainty of evidence. The recommendation was based on the results of a small study (n = 197) showing a lack of benefit of albumin on mortality.9 Data on other patient‑important outcomes were lacking.
Albumin vs crystalloids in critically ill patients with traumatic brain injury
The pathophysiology of TBI involves both primary and secondary injuries. The primary injury occurs during the initial trauma, while the secondary injury develops later on, through mechanisms such as glial injury, which disrupts the blood‑brain barrier, and neuronal damage, impairing intracellular fluid regulation.24 Managing intracranial pressure, cerebral edema, and cerebral perfusion pressure is critical for improving outcomes.24 Furthermore, observational studies showed a correlation between positive fluid balance and worse outcomes in the TBI patients,25 which is why albumin seems theoretically appealing by increasing oncotic pressure at lower volumes than crystalloids.
Severe TBI is often accompanied by extracranial injuries and organ dysfunction, adding complexity to critical care management.24,26 Although no RCT has specifically compared albumin to crystalloids in individuals with TBI, a subgroup analysis from a landmark study on critically ill patients suggested potential harm with albumin, showing increased mortality and worse long‑term functional outcomes.27 In addition, a network meta‑analysis including data from 4 RCTs (n = 1970) showed increased mortality with albumin.28 Therefore, the guideline issued a conditional recommendation favoring crystalloids over albumin in adults with acute TBI. This is similar to the ESICM consensus and clinical practice recommendations for fluid therapy in neurointensive care patients that favored crystalloids over albumin for resuscitation.29
Albumin vs crystalloids in critically ill patients in perioperative period and in patients with bleeding or at a risk for bleeding
Critically ill patients often have anemia and coagulopathy, which increase the risk for adverse events and poor outcomes. Even in the absence of active bleeding, critically ill patients often receive blood and blood products to correct these abnormalities.30 In the perioperative setting, anticipated blood loss adds further complexity. Interventions to manage bleeding, including intravenous fluids, can lead to dilutional coagulopathy, coagulation substrate deficiencies, and altered clot stability.31 Hemorrhagic shock is associated with so‑called lethal triad of acidosis, hypothermia, and coagulopathy, which are further worsened by crystalloid administration.32 Damage control resuscitation mitigates the risk of the triad by minimizing early crystalloids, permissive hypotension, and transfusion of blood and blood products in equal ratios.32 In theory, albumin may be associated with less dilutional effects than crystalloids. However, data from RCTs failed to demonstrate a favorable effect of albumin over crystalloids. Therefore, the guidelines issued a conditional recommendation favoring crystalloids over albumin, based on a very low certainty of evidence.
Albumin vs crystalloids in critically ill patients with cirrhosis
Cirrhosis is associated with complex physiological changes, which can make the hemodynamic management challenging. Splanchnic validation, portal hypertension, and relative hypovolemia are key characteristics. In addition, hypoalbuminemia and associated decreased oncotic pressure lead to salt and water retention, and, in extreme situations, to reduced glomerular filtration due to renal vasoconstriction.33 Consequently, volume status assessment becomes challenging, and worsening edema and ascites can ensue from fluid administration, without improvement in intravascular volume.33 The ideal fluid should allow for increasing the effective circulating volume and offset pathophysiological changes associated with cirrhosis.34
Critically ill patients with cirrhosis often have established indications for albumin, such as acute decompensation, hepatic encephalopathy, spontaneous bacterial peritonitis, large volume paracentesis, and hepatorenal syndrome.35-37 Outside these established indications, the guidelines evaluated the use of albumin in comparison with crystalloids. The evidence showed no benefit of albumin over crystalloids for other clinical scenarios, albeit with a very low certainty of evidence. Therefore, the guidelines issued a conditional recommendation favoring crystalloids over albumin in critically ill adults with cirrhosis but without specific indication for albumin.
Balanced crystalloids vs isotonic saline
The guidelines evaluated the role of balanced crystalloids and isotonic saline in 4 subgroups: critically ill patients in general, and those with sepsis, TBI, or acute kidney injury (AKI). Despite being labeled as isotonic saline, this solution is hyperosmolar in comparison with human plasma (308 vs 275–297 mOsm/l) and contains more sodium and chloride (154 mmol/l each vs 135–145 mmol/l and 94–111 mmol/l, respectively). Large volumes of saline can lead to hyperchloremia, which has been associated with adverse cardiovascular and renal effects, including AKI, need for RRT, and death. Balanced crystalloids, on the contrary, have a biochemical profile closer to human physiology and are thought to mitigate these risks.38 However, their cost is significantly higher. Data from an international survey of 187 sites revealed that isotonic saline costs less than 1 USD per 100 ml, and balanced crystalloids cost 3–4 USD per 100 ml, with substantial variation between countries.15 In resource‑limited settings, a recent expert consensus document highlighted the significant resource utilization of newer balanced crystalloid solutions and recommended the use of generic solutions, such as Ringer lactate or Hartman solution instead.20 These considerations are relevant to all subsequent recommendations on balanced crystalloids.
Balanced crystalloids vs isotonic saline in critically ill adults
The guidelines issued a conditional recommendation favoring balanced crystalloids over isotonic saline, based on low‑certainty evidence. A large body of evidence informed this recommendation, including an individual patient data meta‑analysis of 6 RCTs (n = 34 653).39 Balanced crystalloids may result in a small reduction in mortality, as compared with isotonic saline, though no improvement was observed in other patient‑important outcomes.
The guidelines also provided practical guidance based on specific patient characteristics and resource availability. For patients with hypochloremia or metabolic alkalosis, isotonic saline is a reasonable alternative when access to balanced crystalloids is limited. Conversely, for patients with hyperchloremia or metabolic acidosis, or those requiring large volumes of fluid, the guidelines favored balanced crystalloids over isotonic saline.
Balanced crystalloids vs isotonic saline in sepsis
The guidelines issued a conditional recommendation favoring balanced crystalloids over isotonic saline in patients with sepsis, based on low‑certainty evidence. The recommendation was driven by the results of a subgroup analysis in an individual patient data meta‑analysis showing modest reduction in mortality.39
Balanced crystalloids vs isotonic saline in traumatic brain injury
As discussed earlier, the prevention and management of cerebral edema and maintenance of cerebral perfusion pressure are key priorities in patients with TBI. Given the blood‑brain barrier and neuronal dysfunctions, hypotonic solutions could precipitate cerebral edema in patients with TBI. In contrast, isotonic saline can help prevent and manage cerebral edema by creating an osmotic gradient between the intracellular and extracellular compartments. In addition, a network meta‑analysis including data from 4 RCTs (n = 1970) showed increased mortality with balanced crystalloids in this population, as compared with isotonic saline.28
The guidelines used data from 3 RCTs comparing balanced crystalloids with isotonic saline, which showed worse mortality with balanced crystalloids. Therefore, the panel issued a conditional recommendation favoring isotonic saline, based on a very low certainty of evidence. Additionally, the guidelines highlighted that most trials used balanced crystalloids with osmolarity close to normal plasma and advised against the use of hypotonic balanced solutions, such as Ringer lactate in this context. This is in agreement with ESICM consensus and clinical practice recommendations for fluid therapy in neurointensive care patients that recommended against the solutions with osmolality below 260 mOsm/l.29
Balanced crystalloids vs isotonic saline in critically ill patients with acute kidney injury
The high chloride content of isotonic saline has been associated with higher risks of AKI, RRT, and death.38 Balanced crystalloids, with their reduced chloride content, are theoretically beneficial in patients with AKI. However, direct evidence has been limited. The guidelines considered a small RCT (n = 38) enrolling patients with acute‑on‑chronic kidney injury, who failed to show a benefit of balanced crystalloids over isotonic saline.40 The panel considered indirect evidence from post‑kidney transplant patients, pooled from 8 RCTs (n = 1526).41 In this population, balanced crystalloids were associated with a lower risk for requiring RRT and mechanical ventilation. Therefore, the guidelines issued a conditional recommendation favoring balanced crystalloids over isotonic saline in critically ill adults with AKI. In resource‑limited settings, generic solutions, such as Ringer lactate / acetate or Hartman solution, are preferred, as mentioned earlier.20
Small‑volume hypertonic solutions vs isotonic crystalloids in critically ill patients
The benefits of hypertonic solutions stem from their ability to expand extravascular volumes at lower amounts. This can be of specific importance in patients with hemorrhagic shock at a risk of dilutional coagulopathy from large volumes of isotonic crystalloids, hypovolemia, or septic shock. However, pooled evidence from 17 RCTs (n = 2195) failed to demonstrate superiority of hypertonic solutions over isotonic crystalloids in any patient‑important outcome. In addition, the panel considered several limitations, such as limited availability of hypertonic solutions, increased costs, and potential issues of acceptability. Based on a very low certainty of evidence, the guidelines issued a conditional recommendation favoring isotonic crystaloids over small‑volume hypertonic solutions.
Conclusions
The 2024 ESICM clinical practice guidelines offer evidence‑based recommendations on the optimal choice of fluids for critically ill adults. These guidelines address the comparative use of albumin, crystalloids, balanced crystalloids, isotonic saline, and hypertonic solutions across a range of clinical contexts, including general critical illness, sepsis, TBI, AKI, the perioperative period, and cirrhosis. The recommendations emphasize patient‑specific considerations, balancing the potential benefits and harms of fluid choices with resource utilization and cost‑effectiveness. While crystalloids remain the preferred choice in most scenarios due to their safety profile, availability, and cost, the guidelines acknowledge theoretical advantages of albumin and balanced crystalloids in selected cases. However, the lack of robust evidence demonstrating consistent benefits for patient‑important outcomes limits strong recommendations in favor of these alternatives. The guidelines highlight critical gaps in the existing evidence, particularly for subgroups such as septic patients with hypoalbuminemia or those requiring large volumes of fluids. In addition, studies on 20%–25% albumin in TBI are warranted, given the negative findings with hypotonic 4% albumin in this subgroup.42 Furthermore, 2 recent international surveys explored current practices around albumin as well as acetated vs lactated buffered solutions in critically ill patients and explored respondents’ views on future RCTs and willingness to participate in them. The support for future RCTs was high in both studies.43,44 Altogether, these findings and efforts underscore the need for future high‑quality RCTs to fill knowledge gaps.
Ultimately, the 2024 ESICM guidelines aim to improve decision‑making in fluid therapy, ensuring that critically ill patients receive treatments that maximize benefits, minimize harm, and enhance overall outcomes.
- Myburgh JA, Mythen MG. Resuscitation fluids. N Engl J Med. 2013; 369: 1243‑1251. | Crossref
- Gordon D, Spiegel R. Fluid resuscitation: history, physiology, and modern fluid resuscitation strategies. Emerg Med Clin North Am. 2020; 38: 783‑793. | Crossref
- Finfer S, Myburgh J, Bellomo R. Intravenous fluid therapy in critically ill adults. Nat Rev Nephrol. 2018; 14: 541‑557. | Crossref
- Messmer AS, Zingg C, Muller M, et al. Fluid overload and mortality in adult critical care patients‑a systematic review and meta‑analysis of observational studies. Crit Care Med. 2020; 48: 1862‑1870. | Crossref
- Finfer S, Bellomo R, Boyce N, et al. A comparison of albumin and saline for fluid resuscitation in the intensive care unit. N Engl J Med. 2004; 350: 2247‑2256. | Crossref
ARTICLE INFORMATION