INTRODUCTION

Albumin is used throughout intensive care units (ICU) and medical/surgical floors. Albumin use is recommended over crystalloids in certain settings, such as large volume paracentesis, hepatorenal syndrome, and spontaneous bacterial peritonitis (SBP) treatment.¹ However, there is little data proving mortality benefit of albumin as the initial resuscitation fluid over less expensive crystalloids.²,³,⁴ Fluid therapy practice standards have considerable variation in terms of volume and choice of fluid administered.⁵ The Colloids Versus Crystalloids for the Resuscitation of the Critically Ill (CRISTAL) randomized trial determined there was no mortality benefit at 28 days when using colloids such as albumin, versus crystalloids such as normal saline or Ringer lactate solution in patients with hypovolemic shock.² Raghunathan et al. also determined giving colloids in addition to crystalloids resulted in increased cost per day without improved survival.³ The Surviving Sepsis Campaign guidelines reflect the findings of these studies by recommending crystalloids as the preferred fluid therapy.⁴ Inappropriate fluid selection can have a large economic impact on institutions.³,⁵,⁶,⁷,⁸,⁹ In a sequential multifaceted intervention to decrease albumin use in eight ICUs at two hospitals in an academic healthcare system, no statistically significant difference was found in ICU mortality or in-hospital mortality before and after the interventions.⁶ These findings reinforce the notion that aligning clinical practice with evidence-based literature can both lower the economic impact of albumin, and also maintain the integrity of patient-centered care.

One component lacking from many studies regarding institutional albumin utilization is an assessment of the appropriateness of albumin orders.⁶ Examples of inappropriate albumin prescribing, not supported by national guidelines or evidence-based literature, include first line for intradialytic hypotension (IDH)¹⁰,¹¹, paracentesis of less than 5 liters¹, and septic shock responsive to crystalloids.²-⁴,⁹ A randomized, double-blind, crossover trial in 72 chronic hemodialysis patients was conducted to determine whether 5% albumin was more effective than normal saline for the treatment of IDH. The results showed that 5% albumin was no more effective than normal saline, and the investigators recommended that normal saline should be used as the initial fluid treatment for IDH.¹⁰ A similar study by Emili et al evaluated the safety and efficacy of a stepwise protocol utilizing normal saline, mannitol, and albumin designed to minimize albumin use for IDH treatment. They found that of the 433 instances where the protocol was used, hypotension was reversed without the need for albumin in 91% of cases.¹¹

Enhanced diuresis and perioperative hypotension are two other controversial instances where albumin is used. Two studies, one in medical ICU patients and one in patients with cirrhosis and ascites, have shown no benefit to the addition of albumin to enhance diuresis.¹²,¹³ Regarding perioperative hypotension, The British Consensus Guidelines on Intravenous Fluid Therapy for Adult Surgical Patients lists either crystalloid or colloid as potential fluids to correct hypotension, but does not comment on which order they should be used in.¹⁴

Given the widespread use of albumin in different hospital areas, practice variation regarding fluid selection, and the high cost of albumin as compared to therapeutic alternatives, a unique opportunity for pharmacist intervention presents itself to align clinical practice with evidence-based literature and lower the economic impact of albumin by decreasing its use in inappropriate settings.

The purpose of this study was to assess adherence to current national guidelines in the literature for appropriate albumin use at an academic medical center. Albumin orders were assessed for appropriateness using Food and Drug Administration (FDA) approved labeling and national guidelines including the American Association for the Study of Liver Diseases (AASLD) and the Society of Critical Care Medicine’s Surviving Sepsis Campaign.

METHODS

This study was conducted as a retrospective chart review in an urban academic medical center in the United States.

RESULTS

Of the 192 patient charts that met the inclusion criteria, 150 were randomly selected to be included in the analysis. Patient ages ranged from 20 to 89 years, with a median (interquartile range) of 57 years (48 to 67 years). The majority of patients were male (n=97, 65%).

Regarding the primary outcome of interest, 68 patients (45%) received albumin that was not indicated based on guidelines or FDA approved indications. Of the 82 patients that received albumin based on an appropriate indication, 18 (22%) received the inappropriate dose, and 6 (7%) received the inappropriate concentration of albumin (Table 1). The majority of albumin orders were administered in the intensive care, medical/surgical, and perioperative units (Table 2). It was calculated that 3,545.7 grams of albumin was used inappropriately, and the cost (WAC) associated with this inappropriate albumin prescribing was USD 12,955.31 (Table 3).

DISCUSSION

In this study, we determined whether a patient received albumin for either an appropriate or an inappropriate indication. The distinction between appropriate and inappropriate indications was based on evidence-based literature and national guidelines. Appropriate indications included: SBP, large volume paracentesis, HRS, persistent hypotension despite adequate fluid resuscitation, plasmapheresis, and acute respiratory distress syndrome (ARDS). Inappropriate indications included paracentesis less than five liters, intradialytic hypotension, hypotension without adequate fluid resuscitation, enhanced diuresis, hypoalbuminemia, and perioperative hypotension.

Table 1 details the classification of the 150 orders into their respective appropriate or inappropriate indications. Sixty-eight (45%) of the orders were classified as inappropriate indications, with the three most frequent being hypotension without adequate fluid resuscitation, perioperative hypotension, and enhanced diuresis. Perioperative hypotension was an unexpected indication identified, making up almost one-fourth of the inappropriate orders. Of the appropriate indications, HRS and SBP were frequently ordered with an incorrect dose or with the incorrect concentration of albumin.

Limitations of this study include those that are inherent to a retrospective study, namely relying of the accuracy of documentation in the electronic medical record. One area where this lack of information was notable was the documentation of the amount of intravenous fluids used in the perioperative units. Another limitation was that this study was performed at a single medical center. The inclusion of additional sites may have yielded more indications for albumin not seen at our institution (i.e. ovarian hyperstimulation syndrome).

Defining adequate fluid resuscitation for the purposes of classifying albumin orders was another potential limitation in this study. The quantity of 30 mL/kg of crystalloid was derived from the Surviving Sepsis Campaign recommendations. However, a timeline for when this resuscitation occurred in relation to albumin administration needed to be established. There is little guidance in the literature defining this time period. The criteria used in this study to determine adequate resuscitation was for a patient to receive 30 mL/kg of crystalloid within 24 hours prior to albumin administration. However, there are some populations where risk and benefit should be weighed when administering these large amounts of fluid, such as heart failure patients. Although the Surviving Sepsis Campaign does not comment on these patient populations, perhaps a patient’s overall volume status may be a determinant of appropriate albumin use.

Using this information, a pharmacy driven protocol outlining appropriate albumin use was implemented at our institution to align clinical practice with evidence-based literature and lower the economic impact of albumin, while maintaining the integrity of patient-centered care. Implementation strategies used include a pocket reference card for physicians and pharmacists with the dosing regimens for SBP, HRS, and paracentesis, prescriber education, formulary restrictions, and criteria based ordering. Data will be collected after sufficient implementation of these strategies to assess the impact of this pharmacist driven initiative to decrease albumin use Table 4 shows a pocket reference card created to aid in dosing for these indications.

CONCLUSIONS

This study used evidence-based literature to assess adherence of albumin orders to current national guidelines. The results demonstrated a high rate of inappropriate albumin prescribing within our institution. These results reveal potential areas for pharmacist intervention to decrease the amount of inappropriate albumin prescribing, as well as to ensure that the correct dosing and concentration be used with frequently incorrectly dosed indications such as SBP, HRS, and paracentesis. This study also identified an unexpected indication with significant inappropriate albumin utilization, perioperative hypotension, which is an area for further intervention to monitor and decrease use.