Use of Albumin as a Resuscitative Fluid in the Intensive Care Unit.

Abstract

The advent of fluid resuscitation appears to date back to 1832 in response to a cholera outbreak in England.1 Crystalloid solutions have predominantly been used as initial resuscitative fluids in critically ill patients. Since the first clinical use of a human albumin solution was documented in 1941,2 clinicians and researchers have evaluated the use of albumin and other colloids as compared with crystalloids. Albumin has fallen out of favor for some clinicians because of the lack of clinical benefits over crystalloids in multiple large randomized controlled trials and the cost of albumin compared with alternative fluids.3–5 With over 35% of patients receiving fluid resuscitation on any given day in the intensive care unit,6 understanding the nuances between resuscitative fluids is paramount to caring for the critically ill patient because no ideal fluid exists. The purposes of this article are to review the role of albumin as a resuscitation fluid and to highlight the important considerations of its use for the critical care nurse.Restoration of intravascular volume through plasma volume expansion with fluid therapy is a cornerstone of hemodynamic resuscitation. Fluid resuscitation aims to increase cardiac output and improve organ perfusion to augment end-organ function.7,8 Although fluid therapy is universally used for critically ill patients, intravenous fluids for resuscitation are thought to have a narrow therapeutic index given the morbidity and mortality associated with positive fluid balance, negative fluid balance, and adverse effects of fluid therapy.9 Thus, critical evaluation of the type of fluid selected for resuscitation is important to optimize efficacy and minimize iatrogenic harm. Fluid pathophysiology and distribution are integral components of intravenous fluid selection. These concepts have previously been reviewed in other forums.10,11Resuscitative fluids can be broadly classified as crystalloid or colloid solutions. In this review, the discussion of fluid resuscitation excludes blood products (ie, whole blood, red blood cells, fresh frozen plasma, platelets, cryoprecipitate).Crystalloids are solutions containing water-soluble molecules such as electrolytes and dextrose that are freely permeable across the capillary membrane.7,8,11 The composition and quantity of ions within the solution vary depending on the specific crystalloid product. The Table compares common resuscitative fluids. Crystalloids can be described on the basis of their tonicity. Both isotonic and hypertonic solutions have been used as resuscitation fluids, whereas hypotonic solutions such as 0.45% sodium chloride (NaCl), dextrose 5% in water, and dextrose 5% to 0.45% NaCl are mainly used to restore free water deficits.11 Since hypotonic crystalloids primarily distribute into the intracellular and interstitial fluid compartments, their utility as resuscitation fluids is limited given the inadequate intravascular volume expansion. Alternatively, hypertonic solutions such as 3% NaCl, 7.5% NaCl, and 23.4% NaCl are considered plasma expanders because the higher tonicity of these solutions draws fluid from extravascular fluid compartments into the vascular space.10,11 Despite this theoretical advantage, few clinical trials have evaluated hypertonic crystalloid solutions for resuscitation. Currently, hypertonic saline solutions are primarily used to treat elevated intracranial pressure and severe symptomatic hyponatremia.8,11Isotonic crystalloids are the mainstay of fluid resuscitation efforts in critically ill patients. Normal saline, 0.9% NaCl, is the most commonly prescribed fluid worldwide.10 Isotonic crystalloid fluids can be further subclassified as balanced or unbalanced solutions. Normal saline—despite being coined normal—is not physiologically based. The assumed normal mammalian serum concentration of 0.92% was first observed in the 1880s and has remained the standard of care for around 150 years although the actual concentration is 0.6%.7 In fact, the concentrations of sodium and chloride in normal saline are approximately 1.1 times and 1.5 times the concentrations in human plasma, respectively. Because the concentration of chloride is much higher in normal saline than in plasma, the incidence of hyperchloremia is increased in patients who receive large-volume resuscitation with normal saline.10,11 Balanced solutions were created to more closely mimic the composition of human plasma. Some of the balanced solutions commonly used in clinical practice are lactated Ringer solution, Isolyte S (multi-electrolyte injection), Plasma-Lyte (multiple electrolytes injection), and Hartmann solution. Given the correlations between hyperchloremia and acute kidney injury, metabolic acidosis, and intensive care unit mortality, balanced crystalloid solutions may minimize some of the deleterious effects of large-volume resuscitation.10,11 The use of a balanced crystalloid as the initial resuscitation fluid has been increasing, as highlighted by the 2021 Surviving Sepsis Campaign guidelines recommending the use of balanced crystalloids rather than normal saline for resuscitation.15Colloid solutions are suspensions of largely insoluble molecules such as proteins and starches in a carrier fluid. Colloid solutions are relatively unable to permeate healthy capillary membranes because of the molecular weight and size of the specific molecules.7,8,11 Colloids can be categorized into 2 classes: semisynthetic and natural colloids. The semisynthetic colloids include hydroxyethyl starch, gelatin, and dextran products. Compared with crystalloids, hydroxyethyl starch solutions have a higher risk of mortality and nephrotoxicity in patients with sepsis.7,8,10 Similarly, compared with crystalloids, gelatin products lack clinical benefits and have increased adverse effects such as nephrotoxicity and anaphylaxis. Consequently, the 2021 Surviving Sepsis Campaign guidelines recommend against using starches or gelatins for resuscitation.15 The evidence for using dextran for resuscitation is more limited than for using hydroxyethyl starch or gelatin solutions. Nevertheless, the use of dextran is generally avoided because the available evidence shows a lack of benefit compared with crystalloid solutions and an increased risk of adverse effects, namely anaphylaxis and coagulopathy.10 At this time, dextran is primarily used as a priming fluid for extracorporeal membrane oxygenation circuits.16 Hence, until further compelling evidence is presented, the administration of semisynthetic colloids for resuscitation should be avoided.Albumin is the most abundant protein within human plasma and has many important roles during fluid resuscitation, including maintaining fluid balance, regulating colloid osmotic pressure, protecting the glycocalyx, binding and transporting molecules, and managing intracellular volume. It is synthesized endogenously in the liver and cannot be stored for on-demand release, but hepatocyte production of albumin can increase 2-fold to 3-fold in times of need.16 Hypoalbuminemia, generally defined as a serum albumin level of less than or equal to 3.0 g/dL, is common in critically ill patients and has clinically relevant implications. Decreases in the serum albumin concentration have been associated with increased mortality, morbidity, and hospital length of stay.2 Accordingly, the use of albumin solutions as resuscitative fluids may be logical for critically ill patients. Albumin is commercially available in the United States as 5% and 25% solutions. For these products, albumin is extracted from donated human plasma; therefore, it is considered a natural colloid. The theoretical advantage of the natural colloid, albumin, over crystalloid solutions as a plasma expander is that albumin can draw fluid from the interstitial space into the vascular space with a longer duration of expansion because of the creation of a larger oncotic gradient.10,11 Additionally, albumin solutions are considered to be volume sparing compared with crystalloid solutions.11Two albumin products, 5% albumin and 25% albumin, are commercially available in the United States. Internationally, 4% albumin and 20% albumin are also used in clinical practice. The 4% and 5% albumin solutions are considered iso-oncotic formulations, meaning they are oncotically equal to plasma, and the 20% and 25% albumin solutions are considered hyperoncotic formulations, meaning they have a higher colloid oncotic pressure than plasma.10,11 These relative differences in oncotic pressure portend theoretical differences in plasma volume expansion. Iso-oncotic formulations will minimize the volume that leaves the vascular space, and hyperoncotic formulations will mobilize interstitial volume into the vascular space. Theoretically, 500 mL of plasma is oncotically equal to 500 mL of 5% albumin and to 100 mL of 25% albumin.10,11According to classic prediction models, iso-oncotic albumin is expected to provide 3 to 4 times the volume expansion of an equal volume of an isotonic crystalloid solution. Likewise, hyperoncotic formulations are predicted to pull additional volume from the interstitial space approximating 3 to 4 times the volume administered.17 The expected intravascular effects of normal saline, 5% albumin, and 25% albumin are depicted in the Figure. Despite this common dogma of the 1:3 or 1:4 iso-oncotic colloid to crystalloid ratio, clinical trials have found this ratio to be much lower, with 2 landmark trials reporting ratios between 1:1 and 1:1.4.3,5 These findings have led to investigations of the impact of the endothelial glycocalyx on fluid distribution. The glycocalyx, a structure of membrane-bound proteoglycans and glycoproteins found on the surface of vascular endothelial cells, binds with plasma proteins, including albumin, to prevent extravasation of plasma components into the interstitial space. Comorbidities and conditions commonly seen in critically ill patients, including inflammation, sepsis, trauma, and surgery, may damage the integrity of the glycocalyx, causing plasma contents to leak into the interstitial space.7 Currently, it is thought that in healthy patients with an intact glycocalyx, colloids will remain intravascular and the 1:3 to 1:4 colloid to crystalloid ratio will hold true. However, in patients with a damaged glycocalyx, both colloids and crystalloids will migrate extravascularly, reducing the colloid to crystalloid ratio.18With the increasing emphasis on the use of balanced crystalloid solutions, it is important to contextualize the impact that albumin administration may have on electrolyte disturbances, specifically hyperchloremia. The specific composition of albumin products varies by manufacturer, and different electrolyte disturbances may be seen with various albumin products depending on the electrolyte composition of the solution.14 Unfortunately, not all manufacturers report chloride concentrations in their product labeling. Because the inventory of albumin within institutions is primarily determined by local availability and costs, consultation with a clinical pharmacist may elucidate the electrolyte composition of administered albumin solutions at your institution.Albumin is administered as an intravenous intermittent or continuous infusion. Both the 5% and 25% albumin products can be administered either peripherally or centrally. The recommended maximum infusion rate of albumin varies according to the specific product package insert. Most package inserts for 5% and 25% albumin products recommend an infusion rate of 1 to 2 mL/min in the absence of shock. They do not provide any specific guidance on the maximum infusion rate for patients with shock but note that the rate of administration should be adjusted in a patient-specific manner according to the patient’s clinical status and response.12,13 The highest recommended infusion rates per package inserts are 10 mL/min for 5% albumin and 3 mL/min for 25% albumin.19,20Hemodynamic parameters should be closely monitored to assess the efficacy and safety of an albumin infusion because of the potential risks of circulatory overload, hypervolemia, and pulmonary edema. Although rare, hypersensitivity or anaphylactoid reactions are the most common adverse effects of albumin administration.12,13 Cardiovascular overload and infusion reactions (eg, flushing, urticaria, fever, and nausea) may be mitigated by lowering the rate of infusion but may ultimately require discontinuation of the infusion if symptoms do not improve. The product is thermally sterilized by the manufacturer. Consequently, the risk of bacterial contamination is limited to a breach in the integrity of the product container; thus, it is recommended to administer albumin within 4 hours of entering the container. Viral or prion disease transmission is theoretically possible, but no cases of this have been reported to date.12,13Hemodynamically unstable, critically ill patients often have multiple intravenous medications infused simultaneously. Since albumin can act as a transport protein for many molecules, including ions, free fatty acids, and xenobiotics, the compatibility of other medications with albumin must be considered. Many antimicrobial agents, including ceftolozane/tazobactam, eravacycline, isavuconazonium sulfate, meropenem/vaborbactam, micafungin, plazomicin, tedizolid, and vancomycin, are specifically noted to be incompatible with albumin. Other medications that are incompatible with albumin include lipid emulsions, midazolam, total nutrient admixtures, and verapamil. In some clinical scenarios, albumin and blood products may be concurrently prescribed. Although no specific compatibilities are listed in clinical pharmaceutical databases, 1 manufacturer of albumin states that albumin is compatible with whole blood or packed red blood cells.20 Given the lack of standardized recommendations, local institutional guidance should be followed for coadministration of albumin with blood products through the same intravenous access site. Furthermore, patients who are Jehovah’s Witnesses might refuse transfusions of whole blood (red blood cells, platelets, unfractionated plasma, etc) but might accept transfusions of derivatives of primary blood components (cryoprecipitate, immunoglobulins, albumin, etc). The decision to administer albumin to a Jehovah’s Witness should be made jointly with the patient and/or the patient’s health care surrogate.Despite the clinical use of albumin for over 50 years, no large randomized controlled trials were amply powered to determine the impact of albumin as a resuscitative fluid on clinically relevant outcomes. In a 1998 meta-analysis, 32 randomized trials were systematically reviewed to determine the effect of albumin on survival when albumin was used as a therapeutic option in critically ill patients with hypovolemia, burns, or hypoproteinemia. When all 1204 patients were included in the analysis, the authors found that the relative risk of death with albumin use was 1.68 (95% CI, 1.26-2.23), meaning that for every 100 patients treated with albumin, 6 additional deaths would occur compared with crystalloid administration or no albumin administration.21 After this manuscript was published, the use of albumin sharply declined because of the possible increase in mortality associated with albumin, as reported in the study.The Saline Versus Albumin Fluid Evaluation (SAFE) study was subsequently conducted to determine the effect of albumin for fluid resuscitation on 28-day all-cause mortality in critically ill patients.3 This multicenter trial randomized over 6000 patients to receive either 4% albumin or 0.9% NaCl. The authors found no difference in mortality rate between the 2 groups; 79.1% of patients in the albumin group and 78.9% of patients in the NaCl group survived at day 28. They additionally reported no increased incidence of new-onset organ failure, as measured by the Sequential Organ Failure Assessment. Exploratory subgroup analyses of this cohort led to additional hypothesis-generating questions. In the included patients with trauma and associated traumatic brain injury, the risk of death was significantly higher in patients who received albumin. Furthermore, in a subgroup of patients with severe sepsis, those receiving albumin had a reduction in mortality that approached significance. This study dispelled the safety concerns for albumin and raised additional questions for future research.In response to the subgroup analysis from the SAFE study, the Albumin Italian Outcome Sepsis (ALBIOS) study was conducted to evaluate the effects of albumin and crystalloid administration compared with crystalloid administration alone in patients with severe sepsis.5 This trial randomized patients to receive either crystalloid solutions alone or 20% albumin and crystalloid solutions to maintain a serum albumin level of greater than or equal to 3.0 g/dL. The primary objective of this study was to compare the effect on 28-day all-cause mortality in the 2 groups. Like the SAFE study, the ALBIOS study found that albumin administration did not provide any additional survival advantage compared with crystalloids alone, with 68.2% and 68.0% of patients, respectively, surviving in each group. The lack of survival benefit remained even when the investigators extended the mortality evaluation period to 90 days. Although they found no differences in mortality, a tertiary analysis found that the albumin group had a shorter median time to suspension of vasopressor or inotropic agent administration despite a lower median cumulative net fluid balance. This result suggested a small—but possibly clinically relevant—hemodynamic advantage for albumin use in patients with severe sepsis. However, this finding was not confirmed; a retrospective, single-center, propensity score–matched cohort study found that patients who received 25% albumin had no difference in vasopressor-free days compared with those not receiving albumin.22 An additional finding of interest was discovered in a post hoc subgroup analysis within the ALBIOS study. Although mortality did not differ between patients receiving albumin and those receiving crystalloids alone, in the subgroup of patients with septic shock, mortality was significantly lower in the albumin group (relative risk, 0.87; 95% CI, 0.77-0.99). This analysis suggested that for every 16 patients treated with albumin, 1 death could be prevented. Ultimately, this study did not support the use of hyperoncotic albumin in addition to crystalloids to prevent hypoalbuminemia, but it did report findings that warrant further research.The concept of small-volume resuscitation with hyperoncotic albumin solutions has been gaining traction as a possible method to expand plasma volume while minimizing fluid accumulation. Small-volume resuscitation centers around the hypothesis that hyperoncotic albumin solutions can maximize intravascular plasma expansion and draw interstitial fluid into the vascular space via greater oncotic pressures.23 Despite the potential benefits of small-volume resuscitation with hyperoncotic albumin, a concern is that this treatment could induce a hyperoncotic state that might precipitate renal dysfunction, potentially nullifying the advantages of intravascular expansion. The Small Volume Resuscitation With Albumin in Intensive Care: Physiological Effects trial was conducted to compare the effects of 4% or 5% albumin with 20% albumin on fluid requirements and fluid accumulation within 48 hours of randomization in a mixed intensive care unit cohort.24 This study found that patients in the hyper-oncotic albumin group received a smaller amount of fluid and lower cumulative fluid balance than those in the iso-oncotic albumin group, without any noted impact on kidney function. This study lacked meaningful clinically relevant outcomes such as mortality data, and patients with sepsis encompassed just under 11% of the total population. Consequently, the results may be viewed as hypothesis generating, and further investigations of small-volume resuscitation should be conducted because the effects of small-volume resuscitation on mortality remain unknown. Additionally, comparisons of small-volume resuscitation should be conducted with alternative phased approaches to fluid resuscitation like the rescue, optimization, stabilization, and evacuation9 model that incorporates active deresuscitation principles such as the protocol proposed by Bissell and colleagues.25Currently, the 2021 Surviving Sepsis Campaign guidelines make a weak recommendation for albumin use in patients who have received large volumes of crystalloids. Unfortunately, the guidelines do not give any specific guidance on the volume of crystalloid infusion after which albumin should be considered.15 Further research surrounding the timing, concentration, and specific patient populations for resuscitation with albumin should be actively prioritized to enhance our understanding of how to optimize this therapeutic option. Two registered trials may provide more insight into the role of albumin in resuscitation. The Albumin Replacement Therapy in Septic Shock (ARISS) trial compares the effect of 20% albumin replacement with no albumin replacement on 90-day all-cause mortality in patients with septic shock (ClinicalTrials.gov identifier: NCT03869385). The Albumin Italian Outcome Septic Shock-BALANCED (ALBIOSS-BAL) trial randomizes patients with septic shock to 1 of 4 study groups—20% albumin and balanced crystalloids, 20% albumin and normal saline, balanced crystalloids, and normal saline—to determine the effect on all-cause mortality and incidence of acute kidney injury within 90 days of randomization (ClinicalTrials.gov identifier: NCT03654001). Investigators in the ARISS and ALBIOSS-BAL trials, like the ALBIOS trial, will attempt to correct hypoalbuminemia to a target serum albumin level of equal to or greater than 3.0 g/dL in patients in the albumin arm.Fluid resuscitation is a key intervention to improve the care of critically ill patients. Despite the theoretical advantages of albumin as a resuscitative fluid, clinical data do not support the routine use of albumin because of lack of a clinical benefit. Albumin infusions appear to be safe for most critically ill patients except those with traumatic brain injury. Because of the clinical controversies surrounding the selection of resuscitation fluids, it is imperative for the critical care nurse to understand the role of albumin in resuscitation as well as important considerations for the administration of albumin in critically ill patients.