Nutrition Risk in the Critically Ill (NUTRIC) Score

Why did you develop the NUTRIC Score? Was there a particular clinical experience or patient encounter you had that inspired you to start this research?

It was more the recognition that clinicians were not able to discriminate patients who would benefit the most from being aggressive with nutrition therapy versus those where it didn’t seem to really matter. We talked a lot about nutrition risk assessments and malnutrition screening, and most of the tools that are out there were developed and validated for non-ICU patients or said differently, people in hospitalized settings or outpatient settings. So there was a need for a tool for critically ill patients that helped discriminate who’s nutritionally high risk and therefore would benefit more from optimal nutrition.

What pearls, pitfalls and/or tips do you have for users of the NUTRIC Score? Do you know of cases when it has been applied, interpreted, or used inappropriately?

That’s a great question. I have observed people using it as a decision tree to figure out when to start enteral nutrition versus not, the idea being that a low risk patient doesn’t need to be fed whatsoever, and that is not how we conceptualized or validated the tool. My assumption is, you’re talking about a mechanically ventilated ICU patient in whom you have already started some form of early enteral nutrition within 24 or 48 hours, and I don’t think you need any tool to tell you if a critically ill patient on mechanical ventilation needs to be tube-fed. How we’ve used the NUTRIC Score is on day 3, we do the risk assessment to see if they’re high risk versus low risk, and it’s the high risk patients who we need to optimally feed. The low risk patients you can putz along with hypocaloric feeding for a few more days, but in those high risk patients that you see on day 3 who are still on your unit and not getting at least 80% of the prescribed amount -- that’s the one where you have to get more nutrition in.

Where did the Day 3 time point come from?

We thought clinically it made sense to evaluate on day 3, since you don’t need to do a nutritional risk assessment in mechanically ventilated ICU patients right away—you want to give both the nutritional and non-nutritional benefit of early feeding. But as time evolves, if the patient is still on nutrition on day 3, obviously they’re more complicated, they’re sicker, and they’re there longer, so is this a high risk patient or not? It’s around day 3, plus or minus a day or so, where you’re making the judgment, so that’s what we validated it for.

You touch on one of the broader challenges in research, which is translating evidence for bedside use? To that end, can you talk a little bit about IL-6? Your research shows that it’s a contributing risk factor to the NUTRIC Score, but you also note in your paper that it can be left out because it’s sometimes not commonly available.

Correct, even from a psychometric point of view, it doesn’t contribute much to the predictive validity of the tool, and from a practical and feasible point of view, many sites are just not going to get an IL-6 commonly, so in our subsequent publication, we studied what we call the modified NUTRIC Score, and the tool is still valid.  

You mention that other risk assessments are not specifically validated in ICU patients. From your perspective both clinically and as a researcher, how does the NUTRIC Score compare to other risk assessments like the NRS 2002, which is not specific to ICU patients?

The problem with the NRS is that it hasn’t been validated to the extent that NUTRIC has in ICU patients, where it identifies the subpopulation of people who will have a better outcome with more nutrition. The way it was developed and validated is very different from what we’ve done, and it was actually derived from the literature, by looking at many studies, which say that patients in the ICU have more positive studies and therefore ICU patients are high risk—the way they created the NRS, all ICU patients are high risk. So it doesn’t really help with the further discrimination that of ICU patients, these ones will benefit more than these other ones, and we can point to validated data that prove that point. For NRS, that data doesn’t exist, but there are multiple validations with NUTRIC.  

Finally, any upcoming studies that you or your research group are working on that you’re particularly excited about?

Yes! One is the EFFORT study, which is a multicenter randomized trial of high versus low protein dose in nutritionally at-risk critically ill patients, where one of those risk stratification variables will be the NUTRIC Score, so we’re continuing to try and make the point that not all ICU patients are the same, and that high risk patients will respond better to optimal nutrition therapy. So the only people we’re going to recruit are nutritionally high risk patients, using the NUTRIC Score and some other traditional criteria for malnourishment. The way NUTRIC ended up, there’s not a lot in it related to traditional malnutrition, like low BMI, or history of weight loss, or reduced oral intake, that isn’t reflected in the NUTRIC Score.

Daren K. Heyland, MD, MSc, FRCPC, is a professor of medicine and epidemiology at Queen’s University in Ontario, Canada. He is a critical care physician and director of the Clinical Evaluation Research Unit at the Kingston General Hospital. Dr. Heyland’s research interests include nutrition support and gastrointestinal function in patients suffering from critical illnesses.