Matthew W. Semler, MD, MSc

If you work in an ICU, chances are good that you frequently order IV fluids (IVF). Between resuscitation, maintenance, and medication carriers, nearly all ICU patients receive IVF. Historically, much of this IVF has been 0.9% sodium chloride (“saline” or “normal saline”). Providers in the United States alone administer more than 200 million liters of saline each year (Myburgh JA, et al. N Engl J Med. 2013;369[13]:1243). New evidence, however, suggests that treating your ICU patients with so-called “balanced crystalloids,” rather than saline, may improve patient outcomes.

For over a century, clinicians ordering IV isotonic crystalloids have had two basic options: saline or balanced crystalloids (BC). Saline contains water and 154 mmol/L of sodium chloride (around 50% more chloride than human extracellular fluid). In contrast, BCs, like lactated Ringer’s (LR), Hartman’s solution, and others, contain an amount of chloride resembling human plasma (Table 1). BC substitute an organic anion such as bicarbonate, lactate, acetate, or gluconate, in place of chloride – resulting in lower chloride level and a more neutral pH.

Over the last 2 decades, evidence has slowly accumulated that the different compositions of saline and BC might translate into differences in patient physiology and outcomes. Research in the operating room and ICU found that saline administration caused hyperchloremia and metabolic acidosis. Studies of healthy volunteers found that saline decreased blood flow to the kidney (Chowdhury AH, et al. Ann Surg. 2012;256[1]:18). Animal sepsis models suggested that saline might cause inflammation, low blood pressure, and kidney injury (Zhou F, et al. Crit Care Med. 2014;42[4]:e270). Large observational studies among ICU patients found saline to be associated with increased risk of kidney injury, dialysis, or death (Raghunathan K, et al. Crit Care Med. 2014 Jul;42[7]:1585). These preliminary studies set the stage for a large randomized clinical trial comparing clinical outcomes between BC and saline among acutely ill adults.

Between June 2015 and April 2017, our research group conducted the Isotonic Solutions and Major Adverse Renal Events Trial (SMART) (Semler MW, et al. N Engl J Med. 2018;378[9]:819). SMART was a pragmatic trial in which 15,802 adults in five ICUs were assigned to receive either saline (0.9% sodium chloride) or BC (LR or another branded BC [PlasmaLyte A]). The goal was to determine whether using BC rather than saline would decrease the rates of death, new dialysis, or renal dysfunction lasting through hospital discharge. Patients in the BC group received primarily BC (44% LR and 56% another branded BC [PlasmaLyte A]), whereas patients in the saline group received primarily saline. The rate of death, new dialysis, or renal dysfunction lasting through hospital discharge was lower in the BC group (14.3%) than the saline group (15.4%) (OR: 0.90; 95% CI, 0.82-0.99; P=0.04). The difference between groups was primarily in death and new dialysis, not changes in creatinine. For every 100 patients admitted to an ICU, using BC rather than saline would spare one patient from experiencing death, dialysis, or renal dysfunction lasting to hospital discharge (number needed to treat). The benefits of BC appeared to be greater among patients who received larger volumes of IVF and patients with sepsis. In fact, among patients with sepsis, mortality was significantly lower with BC (25.2%) than with saline (29.4%) (P=.02).

 

Another trial was conducted in parallel. Saline against LR or another branded BC (PlasmaLyte) in the ED (SALT-ED) compared BC with saline among 13,347 non-critically ill adults treated with IVF in the ED (Self WH, et al. N Engl J Med. 2018;378[9]:829). Like the SMART trial, the SALT-ED trial found a 1% absolute reduction in the risk of death, new dialysis, or renal dysfunction lasting to hospital discharge favoring BC.

The SMART and SALT-ED trials have important limitations. They were conducted at a single academic center, and treating clinicians were not blinded to the assigned fluid. The key outcome was a composite of death, new dialysis, and renal dysfunction lasting to hospital discharge – and the trials were not powered to show differences in each of the individual components of the composite.

Despite these limitations, we now have data from two trials enrolling nearly 30,000 acutely ill patients suggesting that BC may result in better clinical outcomes than saline for acutely ill adults. For clinicians who were already using primarily BC solutions, these results will reinforce their current practice. For clinicians whose default IVF has been saline, these new findings raise challenging questions. Prior to these trials, the ICU in which I practice had always used primarily saline. Some of the questions we faced in considering how to apply the results of the SMART and SALT-ED trials to our practice included:

1. Recent data suggest BC may produce better clinical outcomes than saline for acutely ill adults. Are there any data that saline may produce better clinical outcomes than BC? Currently, there are not.

2. Cost is an important consideration in critical care, are BC more expensive than saline? The cost to produce saline and BC is similar. At our hospital, the cost for a 1L bag of saline, LR, and another branded BC (PlasmaLyte A ) is the exactly the same.

3. Is there a specific population for whom BC might have important adverse effects? Because some BC are hypotonic, the safety of administration of BC to patients with elevated intracranial pressure (e.g., traumatic brain injury) is unknown.

4. Are there practical considerations to using BC in the ICU? Compatibility with medications can pose a challenge. For example, the calcium in LR may be incompatible with ceftriaxone infusion. Although BC are compatible with many of the medication infusions used in the ICU for which testing has been performed, less data on compatibility exist for BC than for saline.

5. Are BC as readily available as saline? The three companies that make the majority of IVF used in the United States produce both saline and BC. Recent damage to production facilities has contributed to shortages in the supply of all of them. Over the long term, however, saline and BC are similar in their availability to hospital pharmacies.

After discussing each of these considerations with our ICU physicians and nurses, consultants, and pharmacists, our ICU collectively decided to switch from using primarily saline to BC. This involved (1) our pharmacy team stocking the medication dispensing cabinets in the ICU with 90% LR and 10% saline; and (2) making BC rather than saline the default in order sets within our electronic order entry system. Based on the results of the SMART trial, making the change from saline to BC might be expected to prevent around 100 deaths in our ICU each year.

Many questions regarding the effect of IV crystalloid solutions on clinical outcomes for critically ill adults remain unanswered. The mechanism by which BC may produce better clinical outcomes than saline is uncertain. Whether acetate-containing BC (eg, PlasmaLyte) produced better outcomes than non-acetate-containing BC (eg, LR) is unknown. The safety and efficacy of BC for specific subgroups of patients (eg, those with hyperkalemia) requires further study. Two ongoing trials comparing BC to saline among critically ill adults are expected to finish in 2021 and may provide additional insights into the best approach to IVF management for critically ill adults. An ongoing pilot trial comparing LR to other branded BC (Plasmalyte/Normosol )may inform the choice between BC.

In summary, IVF administration is ubiquitous in critical care. For decades, much of that fluid has been saline. BC are similar to saline in availability and cost. Two large trials now demonstrate better patient outcomes with BC compared with saline. These data challenge ICU providers, pharmacies, and hospital systems primarily using saline to evaluate the available data, their current IVF prescribing practices, and the logistical barriers to change, to determine whether there are legitimate reasons to continue using saline, or whether the time has come to make BC the first-line fluid therapy for acutely ill adults.

Dr. Semler is with the Department of Medicine, Division of Allergy, Pulmonary, and Critical Care Medicine –Vanderbilt University Medical Center, Nashville, Tennessee.

 

Editor’s Comment

For a very long time, normal saline has been the go-to crystalloid in most ICUs around the globe. In the recent past, evidence started mounting about the potential downside of this solution. The recent SMART trial, the largest to date, indicates that we could prevent adverse renal outcomes by choosing balanced crystalloids over normal saline. These results were even more marked in patients who received a large amount of crystalloids and in patients with sepsis. Dr. Matthew Semler presents solid arguments to consider in changing our practice and adopting a “balanced approach” to fluid resuscitation. We certainly should not only worry about the amount of fluids infused but also about the type of solution we give our patients. Hopefully, we will soon learn if the different balanced solutions also lead to outcome differences.

Angel Coz, MD, FCCP – Section Editor

If you work in an ICU, chances are good that you frequently order IV fluids (IVF). Between resuscitation, maintenance, and medication carriers, nearly all ICU patients receive IVF. Historically, much of this IVF has been 0.9% sodium chloride (“saline” or “normal saline”). Providers in the United States alone administer more than 200 million liters of saline each year (Myburgh JA, et al. N Engl J Med. 2013;369[13]:1243). New evidence, however, suggests that treating your ICU patients with so-called “balanced crystalloids,” rather than saline, may improve patient outcomes.

For over a century, clinicians ordering IV isotonic crystalloids have had two basic options: saline or balanced crystalloids (BC). Saline contains water and 154 mmol/L of sodium chloride (around 50% more chloride than human extracellular fluid). In contrast, BCs, like lactated Ringer’s (LR), Hartman’s solution, and others, contain an amount of chloride resembling human plasma (Table 1). BC substitute an organic anion such as bicarbonate, lactate, acetate, or gluconate, in place of chloride – resulting in lower chloride level and a more neutral pH.

Over the last 2 decades, evidence has slowly accumulated that the different compositions of saline and BC might translate into differences in patient physiology and outcomes. Research in the operating room and ICU found that saline administration caused hyperchloremia and metabolic acidosis. Studies of healthy volunteers found that saline decreased blood flow to the kidney (Chowdhury AH, et al. Ann Surg. 2012;256[1]:18). Animal sepsis models suggested that saline might cause inflammation, low blood pressure, and kidney injury (Zhou F, et al. Crit Care Med. 2014;42[4]:e270). Large observational studies among ICU patients found saline to be associated with increased risk of kidney injury, dialysis, or death (Raghunathan K, et al. Crit Care Med. 2014 Jul;42[7]:1585). These preliminary studies set the stage for a large randomized clinical trial comparing clinical outcomes between BC and saline among acutely ill adults.

Between June 2015 and April 2017, our research group conducted the Isotonic Solutions and Major Adverse Renal Events Trial (SMART) (Semler MW, et al. N Engl J Med. 2018;378[9]:819). SMART was a pragmatic trial in which 15,802 adults in five ICUs were assigned to receive either saline (0.9% sodium chloride) or BC (LR or another branded BC [PlasmaLyte A]). The goal was to determine whether using BC rather than saline would decrease the rates of death, new dialysis, or renal dysfunction lasting through hospital discharge. Patients in the BC group received primarily BC (44% LR and 56% another branded BC [PlasmaLyte A]), whereas patients in the saline group received primarily saline. The rate of death, new dialysis, or renal dysfunction lasting through hospital discharge was lower in the BC group (14.3%) than the saline group (15.4%) (OR: 0.90; 95% CI, 0.82-0.99; P=0.04). The difference between groups was primarily in death and new dialysis, not changes in creatinine. For every 100 patients admitted to an ICU, using BC rather than saline would spare one patient from experiencing death, dialysis, or renal dysfunction lasting to hospital discharge (number needed to treat). The benefits of BC appeared to be greater among patients who received larger volumes of IVF and patients with sepsis. In fact, among patients with sepsis, mortality was significantly lower with BC (25.2%) than with saline (29.4%) (P=.02).

 

Another trial was conducted in parallel. Saline against LR or another branded BC (PlasmaLyte) in the ED (SALT-ED) compared BC with saline among 13,347 non-critically ill adults treated with IVF in the ED (Self WH, et al. N Engl J Med. 2018;378[9]:829). Like the SMART trial, the SALT-ED trial found a 1% absolute reduction in the risk of death, new dialysis, or renal dysfunction lasting to hospital discharge favoring BC.

The SMART and SALT-ED trials have important limitations. They were conducted at a single academic center, and treating clinicians were not blinded to the assigned fluid. The key outcome was a composite of death, new dialysis, and renal dysfunction lasting to hospital discharge – and the trials were not powered to show differences in each of the individual components of the composite.

Despite these limitations, we now have data from two trials enrolling nearly 30,000 acutely ill patients suggesting that BC may result in better clinical outcomes than saline for acutely ill adults. For clinicians who were already using primarily BC solutions, these results will reinforce their current practice. For clinicians whose default IVF has been saline, these new findings raise challenging questions. Prior to these trials, the ICU in which I practice had always used primarily saline. Some of the questions we faced in considering how to apply the results of the SMART and SALT-ED trials to our practice included:

1. Recent data suggest BC may produce better clinical outcomes than saline for acutely ill adults. Are there any data that saline may produce better clinical outcomes than BC? Currently, there are not.

2. Cost is an important consideration in critical care, are BC more expensive than saline? The cost to produce saline and BC is similar. At our hospital, the cost for a 1L bag of saline, LR, and another branded BC (PlasmaLyte A ) is the exactly the same.

3. Is there a specific population for whom BC might have important adverse effects? Because some BC are hypotonic, the safety of administration of BC to patients with elevated intracranial pressure (e.g., traumatic brain injury) is unknown.

4. Are there practical considerations to using BC in the ICU? Compatibility with medications can pose a challenge. For example, the calcium in LR may be incompatible with ceftriaxone infusion. Although BC are compatible with many of the medication infusions used in the ICU for which testing has been performed, less data on compatibility exist for BC than for saline.

5. Are BC as readily available as saline? The three companies that make the majority of IVF used in the United States produce both saline and BC. Recent damage to production facilities has contributed to shortages in the supply of all of them. Over the long term, however, saline and BC are similar in their availability to hospital pharmacies.

After discussing each of these considerations with our ICU physicians and nurses, consultants, and pharmacists, our ICU collectively decided to switch from using primarily saline to BC. This involved (1) our pharmacy team stocking the medication dispensing cabinets in the ICU with 90% LR and 10% saline; and (2) making BC rather than saline the default in order sets within our electronic order entry system. Based on the results of the SMART trial, making the change from saline to BC might be expected to prevent around 100 deaths in our ICU each year.

Many questions regarding the effect of IV crystalloid solutions on clinical outcomes for critically ill adults remain unanswered. The mechanism by which BC may produce better clinical outcomes than saline is uncertain. Whether acetate-containing BC (eg, PlasmaLyte) produced better outcomes than non-acetate-containing BC (eg, LR) is unknown. The safety and efficacy of BC for specific subgroups of patients (eg, those with hyperkalemia) requires further study. Two ongoing trials comparing BC to saline among critically ill adults are expected to finish in 2021 and may provide additional insights into the best approach to IVF management for critically ill adults. An ongoing pilot trial comparing LR to other branded BC (Plasmalyte/Normosol )may inform the choice between BC.

In summary, IVF administration is ubiquitous in critical care. For decades, much of that fluid has been saline. BC are similar to saline in availability and cost. Two large trials now demonstrate better patient outcomes with BC compared with saline. These data challenge ICU providers, pharmacies, and hospital systems primarily using saline to evaluate the available data, their current IVF prescribing practices, and the logistical barriers to change, to determine whether there are legitimate reasons to continue using saline, or whether the time has come to make BC the first-line fluid therapy for acutely ill adults.

Dr. Semler is with the Department of Medicine, Division of Allergy, Pulmonary, and Critical Care Medicine –Vanderbilt University Medical Center, Nashville, Tennessee.

 

Editor’s Comment

For a very long time, normal saline has been the go-to crystalloid in most ICUs around the globe. In the recent past, evidence started mounting about the potential downside of this solution. The recent SMART trial, the largest to date, indicates that we could prevent adverse renal outcomes by choosing balanced crystalloids over normal saline. These results were even more marked in patients who received a large amount of crystalloids and in patients with sepsis. Dr. Matthew Semler presents solid arguments to consider in changing our practice and adopting a “balanced approach” to fluid resuscitation. We certainly should not only worry about the amount of fluids infused but also about the type of solution we give our patients. Hopefully, we will soon learn if the different balanced solutions also lead to outcome differences.

Angel Coz, MD, FCCP – Section Editor

If you work in an ICU, chances are good that you frequently order IV fluids (IVF). Between resuscitation, maintenance, and medication carriers, nearly all ICU patients receive IVF. Historically, much of this IVF has been 0.9% sodium chloride (“saline” or “normal saline”). Providers in the United States alone administer more than 200 million liters of saline each year (Myburgh JA, et al. N Engl J Med. 2013;369[13]:1243). New evidence, however, suggests that treating your ICU patients with so-called “balanced crystalloids,” rather than saline, may improve patient outcomes.

For over a century, clinicians ordering IV isotonic crystalloids have had two basic options: saline or balanced crystalloids (BC). Saline contains water and 154 mmol/L of sodium chloride (around 50% more chloride than human extracellular fluid). In contrast, BCs, like lactated Ringer’s (LR), Hartman’s solution, and others, contain an amount of chloride resembling human plasma (Table 1). BC substitute an organic anion such as bicarbonate, lactate, acetate, or gluconate, in place of chloride – resulting in lower chloride level and a more neutral pH.

Over the last 2 decades, evidence has slowly accumulated that the different compositions of saline and BC might translate into differences in patient physiology and outcomes. Research in the operating room and ICU found that saline administration caused hyperchloremia and metabolic acidosis. Studies of healthy volunteers found that saline decreased blood flow to the kidney (Chowdhury AH, et al. Ann Surg. 2012;256[1]:18). Animal sepsis models suggested that saline might cause inflammation, low blood pressure, and kidney injury (Zhou F, et al. Crit Care Med. 2014;42[4]:e270). Large observational studies among ICU patients found saline to be associated with increased risk of kidney injury, dialysis, or death (Raghunathan K, et al. Crit Care Med. 2014 Jul;42[7]:1585). These preliminary studies set the stage for a large randomized clinical trial comparing clinical outcomes between BC and saline among acutely ill adults.

Between June 2015 and April 2017, our research group conducted the Isotonic Solutions and Major Adverse Renal Events Trial (SMART) (Semler MW, et al. N Engl J Med. 2018;378[9]:819). SMART was a pragmatic trial in which 15,802 adults in five ICUs were assigned to receive either saline (0.9% sodium chloride) or BC (LR or another branded BC [PlasmaLyte A]). The goal was to determine whether using BC rather than saline would decrease the rates of death, new dialysis, or renal dysfunction lasting through hospital discharge. Patients in the BC group received primarily BC (44% LR and 56% another branded BC [PlasmaLyte A]), whereas patients in the saline group received primarily saline. The rate of death, new dialysis, or renal dysfunction lasting through hospital discharge was lower in the BC group (14.3%) than the saline group (15.4%) (OR: 0.90; 95% CI, 0.82-0.99; P=0.04). The difference between groups was primarily in death and new dialysis, not changes in creatinine. For every 100 patients admitted to an ICU, using BC rather than saline would spare one patient from experiencing death, dialysis, or renal dysfunction lasting to hospital discharge (number needed to treat). The benefits of BC appeared to be greater among patients who received larger volumes of IVF and patients with sepsis. In fact, among patients with sepsis, mortality was significantly lower with BC (25.2%) than with saline (29.4%) (P=.02).

 

Another trial was conducted in parallel. Saline against LR or another branded BC (PlasmaLyte) in the ED (SALT-ED) compared BC with saline among 13,347 non-critically ill adults treated with IVF in the ED (Self WH, et al. N Engl J Med. 2018;378[9]:829). Like the SMART trial, the SALT-ED trial found a 1% absolute reduction in the risk of death, new dialysis, or renal dysfunction lasting to hospital discharge favoring BC.

The SMART and SALT-ED trials have important limitations. They were conducted at a single academic center, and treating clinicians were not blinded to the assigned fluid. The key outcome was a composite of death, new dialysis, and renal dysfunction lasting to hospital discharge – and the trials were not powered to show differences in each of the individual components of the composite.

Despite these limitations, we now have data from two trials enrolling nearly 30,000 acutely ill patients suggesting that BC may result in better clinical outcomes than saline for acutely ill adults. For clinicians who were already using primarily BC solutions, these results will reinforce their current practice. For clinicians whose default IVF has been saline, these new findings raise challenging questions. Prior to these trials, the ICU in which I practice had always used primarily saline. Some of the questions we faced in considering how to apply the results of the SMART and SALT-ED trials to our practice included:

1. Recent data suggest BC may produce better clinical outcomes than saline for acutely ill adults. Are there any data that saline may produce better clinical outcomes than BC? Currently, there are not.

2. Cost is an important consideration in critical care, are BC more expensive than saline? The cost to produce saline and BC is similar. At our hospital, the cost for a 1L bag of saline, LR, and another branded BC (PlasmaLyte A ) is the exactly the same.

3. Is there a specific population for whom BC might have important adverse effects? Because some BC are hypotonic, the safety of administration of BC to patients with elevated intracranial pressure (e.g., traumatic brain injury) is unknown.

4. Are there practical considerations to using BC in the ICU? Compatibility with medications can pose a challenge. For example, the calcium in LR may be incompatible with ceftriaxone infusion. Although BC are compatible with many of the medication infusions used in the ICU for which testing has been performed, less data on compatibility exist for BC than for saline.

5. Are BC as readily available as saline? The three companies that make the majority of IVF used in the United States produce both saline and BC. Recent damage to production facilities has contributed to shortages in the supply of all of them. Over the long term, however, saline and BC are similar in their availability to hospital pharmacies.

After discussing each of these considerations with our ICU physicians and nurses, consultants, and pharmacists, our ICU collectively decided to switch from using primarily saline to BC. This involved (1) our pharmacy team stocking the medication dispensing cabinets in the ICU with 90% LR and 10% saline; and (2) making BC rather than saline the default in order sets within our electronic order entry system. Based on the results of the SMART trial, making the change from saline to BC might be expected to prevent around 100 deaths in our ICU each year.

Many questions regarding the effect of IV crystalloid solutions on clinical outcomes for critically ill adults remain unanswered. The mechanism by which BC may produce better clinical outcomes than saline is uncertain. Whether acetate-containing BC (eg, PlasmaLyte) produced better outcomes than non-acetate-containing BC (eg, LR) is unknown. The safety and efficacy of BC for specific subgroups of patients (eg, those with hyperkalemia) requires further study. Two ongoing trials comparing BC to saline among critically ill adults are expected to finish in 2021 and may provide additional insights into the best approach to IVF management for critically ill adults. An ongoing pilot trial comparing LR to other branded BC (Plasmalyte/Normosol )may inform the choice between BC.

In summary, IVF administration is ubiquitous in critical care. For decades, much of that fluid has been saline. BC are similar to saline in availability and cost. Two large trials now demonstrate better patient outcomes with BC compared with saline. These data challenge ICU providers, pharmacies, and hospital systems primarily using saline to evaluate the available data, their current IVF prescribing practices, and the logistical barriers to change, to determine whether there are legitimate reasons to continue using saline, or whether the time has come to make BC the first-line fluid therapy for acutely ill adults.

Dr. Semler is with the Department of Medicine, Division of Allergy, Pulmonary, and Critical Care Medicine –Vanderbilt University Medical Center, Nashville, Tennessee.

 

Editor’s Comment

For a very long time, normal saline has been the go-to crystalloid in most ICUs around the globe. In the recent past, evidence started mounting about the potential downside of this solution. The recent SMART trial, the largest to date, indicates that we could prevent adverse renal outcomes by choosing balanced crystalloids over normal saline. These results were even more marked in patients who received a large amount of crystalloids and in patients with sepsis. Dr. Matthew Semler presents solid arguments to consider in changing our practice and adopting a “balanced approach” to fluid resuscitation. We certainly should not only worry about the amount of fluids infused but also about the type of solution we give our patients. Hopefully, we will soon learn if the different balanced solutions also lead to outcome differences.

Angel Coz, MD, FCCP – Section Editor