MetaPlus Shows That Less Is More in Critical Care Nutrition
MetaPlus Shows That Less Is More in Critical Care Nutrition
This is Andy Shorr from Washington, DC, with the pulmonary and critical care literature update. I want to discuss an article describing the MetaPlus study, which was in the August 6 issue of JAMA. This study focused on the role of immunonutrition in critically ill patients.
Immunonutrition is a controversial topic. One of the difficulties in studying immunonutrition as a way to improve outcomes in the intensive care unit is that not everyone agrees on what actually constitutes immunonutrition. Immunonutrition may focus only on antioxidants. It may focus on supplementation with certain vitamins and minerals, such as selenium. When you look at immunonutrition studies, it is often unclear whether researchers are comparing apples with apples. There also has been some suggestion in the literature that the impact of immunonutrition on surgical patients may be vastly different from the impact on medical patients.
The MetaPlus investigators tried to get at the crux of the issue by conducting a large, randomized, controlled trial of approximately 300 patients. They looked at a very specific immunonutrition that included omega-3 fatty acids, antioxidants, and other nutrients. They randomly assigned these patients to receive this immunonutrition or standard enteral nutrition. The primary endpoint for the study was infections in the critical care unit (CCU), and the secondary endpoints focused on safety in terms of overall mortality.
These authors carefully matched total caloric intake between the two groups. Many past immunonutrition studies have not addressed the fact that, depending on the product being used, not everyone is getting the same number calories. This difference can be a major confounder because both under- and overfeeding, in terms of caloric intake, may have implications for outcomes. The study included a mixed cohort of medical, surgical, and trauma patients. About one third of the cohort was made up of purely medical CCU patients; based on their severity of illness using either Sequential Organ Failure Assessment (SOFA) or Acute Physiology and Chronic Health Evaluation (APACHE) II scores, the overall predicted mortality from this cohort was rather severe, in the 30% to 40% range. Thus, the researchers studied patients who were quite sick. All of these patients were on mechanical ventilation for at least 72 hours and were thought to require enteral nutrition for some period of time thereafter.
One of my concerns with this study is the definition they used for their primary endpoint, which was, essentially, all nosocomial infections. They used criteria from the Centers for Disease Control and Prevention (CDC). Now, that certainly makes it very easy in terms of standardization and recording, but some of the CDC criteria are imprecise and include considerable assessment bias, particularly those for ventilator-associated pneumonia. This study was blinded; therefore, those problems should not have been an issue. The study did not require microbiologic confirmation for a number of infections except, perhaps, with bloodstream infection, where you need a culture to make the diagnosis in the first place.
In terms of the primary endpoint, the researchers found absolutely no difference in the rate of infections between those given immunonutrition and those given standard nutrition. They also looked at infection subtypes, which were very important. They did not just pool all of the infections together as their only effort to analyze this because, of course, a urinary tract infection does not have the same implications as a ventilator-associated pneumonia event. When they broke down the infectious complications by type—bloodstream infection, ventilator-associated, and so forth—they found no difference between patients who received immunonutrition vs standard nutrition. Likewise, when they looked at the interaction between type of patients—medical, surgical, or trauma—and the site of infection, they saw no differences.
Thus, there was no suggestion of benefit from immunonutrition in any subgroup for a particular type of infection. For these exploratory subgroup analyses and interaction assessments, everything is underpowered because the study started with only 300 patients overall, but they did not see a difference or even a suggestion that would support a hypothesis.
More troubling is what they saw when they looked at mortality. Overall mortality was relatively well balanced, but when they created a model to look at predictors of mortality, they saw some suggestion that immunonutrition was associated with greater mortality, and that signal became particularly evident in the medical patients.
These authors have provided a tremendous service by looking specifically at the safety question and raising concerns that, in terms of the risk/benefit calculus, there seems to be no suggestion of a benefit and actually a suggestion of harm with immunonutrition. When you add the fact that immunonutrition is uniformly more expensive than standard nutrition, clearly this study augments others in the critical care literature that document that less is more.
We do not need fancy things to take care of patients. We need to focus on the basics, such as fluids, antibiotics, and source control at the beginning of sepsis. The same is true with nutrition and good standard nutrition practices. Following on the EDEN study from the ARDS Network investigators, this is another study that focused on nutrition in a high-risk population, and it indicates that aggressive standard feeding is the way to go. Immunonutrition does not offer a benefit.
This is Andy Shorr from Washington, DC.
This is Andy Shorr from Washington, DC, with the pulmonary and critical care literature update. I want to discuss an article describing the MetaPlus study, which was in the August 6 issue of JAMA. This study focused on the role of immunonutrition in critically ill patients.
Immunonutrition is a controversial topic. One of the difficulties in studying immunonutrition as a way to improve outcomes in the intensive care unit is that not everyone agrees on what actually constitutes immunonutrition. Immunonutrition may focus only on antioxidants. It may focus on supplementation with certain vitamins and minerals, such as selenium. When you look at immunonutrition studies, it is often unclear whether researchers are comparing apples with apples. There also has been some suggestion in the literature that the impact of immunonutrition on surgical patients may be vastly different from the impact on medical patients.
A Study Designed to Settle the Controversy
The MetaPlus investigators tried to get at the crux of the issue by conducting a large, randomized, controlled trial of approximately 300 patients. They looked at a very specific immunonutrition that included omega-3 fatty acids, antioxidants, and other nutrients. They randomly assigned these patients to receive this immunonutrition or standard enteral nutrition. The primary endpoint for the study was infections in the critical care unit (CCU), and the secondary endpoints focused on safety in terms of overall mortality.
These authors carefully matched total caloric intake between the two groups. Many past immunonutrition studies have not addressed the fact that, depending on the product being used, not everyone is getting the same number calories. This difference can be a major confounder because both under- and overfeeding, in terms of caloric intake, may have implications for outcomes. The study included a mixed cohort of medical, surgical, and trauma patients. About one third of the cohort was made up of purely medical CCU patients; based on their severity of illness using either Sequential Organ Failure Assessment (SOFA) or Acute Physiology and Chronic Health Evaluation (APACHE) II scores, the overall predicted mortality from this cohort was rather severe, in the 30% to 40% range. Thus, the researchers studied patients who were quite sick. All of these patients were on mechanical ventilation for at least 72 hours and were thought to require enteral nutrition for some period of time thereafter.
Surprising and Disturbing Outcomes
One of my concerns with this study is the definition they used for their primary endpoint, which was, essentially, all nosocomial infections. They used criteria from the Centers for Disease Control and Prevention (CDC). Now, that certainly makes it very easy in terms of standardization and recording, but some of the CDC criteria are imprecise and include considerable assessment bias, particularly those for ventilator-associated pneumonia. This study was blinded; therefore, those problems should not have been an issue. The study did not require microbiologic confirmation for a number of infections except, perhaps, with bloodstream infection, where you need a culture to make the diagnosis in the first place.
In terms of the primary endpoint, the researchers found absolutely no difference in the rate of infections between those given immunonutrition and those given standard nutrition. They also looked at infection subtypes, which were very important. They did not just pool all of the infections together as their only effort to analyze this because, of course, a urinary tract infection does not have the same implications as a ventilator-associated pneumonia event. When they broke down the infectious complications by type—bloodstream infection, ventilator-associated, and so forth—they found no difference between patients who received immunonutrition vs standard nutrition. Likewise, when they looked at the interaction between type of patients—medical, surgical, or trauma—and the site of infection, they saw no differences.
Thus, there was no suggestion of benefit from immunonutrition in any subgroup for a particular type of infection. For these exploratory subgroup analyses and interaction assessments, everything is underpowered because the study started with only 300 patients overall, but they did not see a difference or even a suggestion that would support a hypothesis.
Assessing Risk/Benefit: Less Is More
More troubling is what they saw when they looked at mortality. Overall mortality was relatively well balanced, but when they created a model to look at predictors of mortality, they saw some suggestion that immunonutrition was associated with greater mortality, and that signal became particularly evident in the medical patients.
These authors have provided a tremendous service by looking specifically at the safety question and raising concerns that, in terms of the risk/benefit calculus, there seems to be no suggestion of a benefit and actually a suggestion of harm with immunonutrition. When you add the fact that immunonutrition is uniformly more expensive than standard nutrition, clearly this study augments others in the critical care literature that document that less is more.
We do not need fancy things to take care of patients. We need to focus on the basics, such as fluids, antibiotics, and source control at the beginning of sepsis. The same is true with nutrition and good standard nutrition practices. Following on the EDEN study from the ARDS Network investigators, this is another study that focused on nutrition in a high-risk population, and it indicates that aggressive standard feeding is the way to go. Immunonutrition does not offer a benefit.
This is Andy Shorr from Washington, DC.