NUTRIC score use around the world: a systematic review

Objective To collect data on the use of The Nutrition Risk in Critically Ill (NUTRIC) score. Methods A systematic literature search was conducted in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) statement. Reviews, abstracts, dissertations, protocols and case reports were excluded from this review; to be included in the review, studies needed to specifically evaluate the NUTRIC score and to have been published in English, Spanish or Portuguese. Results We included 12 (0.8%) studies from our search in this review. Ten studies (83.3%) were observational, 1 was a pilot study (8.3%) and 1 was a randomized control trial (8.3%). All of the included studies (100%) chose not to use IL-6 and considered a high nutritional risk cutoff point ≥ 5. There were 11 (91.7%) English language studies versus 1 (8.3%) Spanish language study. Mechanical ventilation and a high NUTRIC score were significantly correlated in four studies. The association between intensive care unit or hospital length of stay and nutritional high risk was significant in three studies. Seven studies found a statistically significant association between the NUTRIC score and mortality. Conclusion The NUTRIC score is related to clinical outcomes, such as length of hospital stay, and is appropriate for use in critically ill patients in intensive care units.


INTRODUCTION
Malnutrition is common in hospitalized patients and highly prevalent in the population of critically ill individuals. (1,2) Malnutrition is associated with increased morbidity, mortality, occurrence of nosocomial infections, prolonged hospitalization, worse functional status at discharge from intensive care units (ICU) and increased hospital costs. (3,4) Most of the tools used to assess nutritional risk include a variety of criteria to identify nutritional risk, such as food/nutritional intake, physical examination, severity of illness, anthropometric data and functional assessment. (5) Many of these criteria are difficult to obtain in critically ill patients because almost all of these patients require mechanical ventilation (MV) and sedation. (5) Changes in weight can be influenced by fluid status, given the large volumes necessary to maintain hemodynamic stability. (5) Many traditional tools do not provide not to use IL-6 and considered a high nutritional risk cutoff point ≥ 5. There were 11 (91.7%) English language studies versus 1 (8.3%) Spanish language study. Mechanical ventilation and a high NUTRIC score were significantly correlated in four studies. The association between intensive care unit or hospital length of stay and nutritional high risk was significant in three studies. Seven studies found a statistically significant association between the NUTRIC score and mortality.
Conclusion: The NUTRIC score is related to clinical outcomes, such as length of hospital stay, and is appropriate for use in critically ill patients in intensive care units. information regarding inflammatory status, which is crucial in critically ill patients because it is one of the factors responsible for hypermetabolic status and muscle wasting. (5) In 2011, Heyland et al. presented a new screening tool called Nutrition Risk in Critically Ill (NUTRIC) score, which was validated for ICU patients. (6) This score evaluates adverse outcome risk (mortality, MV) modifiable by intensive nutritional intervention. (6) The variables incorporated in this score are: age, Acute Physiology and Chronic Health disease Classification System II (APACHE II) score, Sequential Organ Failure Assessment (SOFA) score, comorbidities, days in the hospital prior to admission to the ICU and Interleukin-6 (IL-6). (6) Proposed in 2016, a modified NUTRIC without IL-6 can be used considering a high nutritional risk cutoff point ≥ 5. (7) The purpose of this review is to collect data on the use of the NUTRIC score.

METHODS
A systematic literature search was conducted in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) statement (8) in December 2017. The search was carried out in four databases: Medical Literature Analysis and Retrieval System Online (MEDLINE), Latin American and Caribbean Health Sciences Literature (LILACS), Scientific Electronic Library Online (SciELO) and Cochrane Collaboration. The search strategy for these databases were defined by terms related to NUTRIC [NUTRIC, Nutrition Risk in Critically Ill score] and terms related to nutritional assessment [nutritional risk, nutritional status] in addition to "critical illness". The terms were enclosed in quotation marks, and the search operators "and" and "or" were used. Reviews, abstracts, studies protocols, dissertations and case reports were excluded from this review.
Moreover, to be included in the review, studies needed to specifically evaluate the NUTRIC score and to have been published in English, Spanish or Portuguese. Finally, articles were screened according to the following steps: at first, duplicates were excluded. Then, the remaining articles were screened by title, abstract and text in full. Articles were selected based on the eligibility criteria as outlined above. If eligibility could not be determined during the initial screening of the title and abstract, full-text articles were accessed to determine inclusion. Both the study selection and data extraction were concurrently performed by two of the authors (AR and AF). If there was any doubt regarding the eligibility criteria, a third evaluator (LFM), another author, made the final decision. MEDLINE, LILACS, Cochrane and SciELO provided 1189, 30, 179 and 89 articles, respectively. More details are shown in figure 1.

NUTRIC applicability
In Brazil, a pilot study was conducted. (13) Portuguese translation and adaptation were required to validate the NUTRIC score for use in Brazil. (13) The authors evaluated 50 individuals whose data were easily obtained from medical records, and neither nutritionists nor physicians reported difficulties in registering them. (13) All five healthcare professionals who participated in the pilot study reported that the new version of the NUTRIC score was easy and clear to understand as well as practical and fast to apply. (13)

Altered NUTRIC
Moretti et al. conducted a study that used C-reactive protein (CRP) instead of IL-6 (NUTRIC-2, ≥ 6 points) versus no inflammatory marker (NUTRIC-1, ≥ 5 points). (9) NUTRIC-2 used cut-off value of ≥ 6 points to define high risk, as suggested by Heyland et al., (6) and had a sensitivity and specificity of 37.76% and 88.95%, respectively. (9) A cut-off value of 3 points led to a sensitivity close to 70% and a specificity of 60%. (9) However, the sensitivity and specificity of the area under the receiver operator characteristic (ROC) curve were lower for predicting mortality than the original study (0.671 and 0.679 versus 0.783, respectively). (6,9)

Author, Country Main results
Rahman et al. (7) Canada Mortality at 28 days was multiplied by 1.4 for every point increase of the NUTRIC score. There is a strong positive association between nutritional adequacy and 28-day survival in patients with a high NUTRIC score, but this association decreases with the decreasing NUTRIC score. Higher NUTRIC scores are also significantly associated with higher 6-month mortality (p < 0.001).
Moretti et al. (9) Argentine Mortality increased in relation to the score (p < 0.001). The mean CRP was higher in mortality (p = 0.001) and VM time (p = 0.010), and the AUC increased in a similar way to IL-6 in the original work (0.008 and 0.007, respectively).
Lee et al. (10) Malaysia For patients with low nutritional risk, mortality was increased by approximately 6 times in the group that received ≥ 2/3 of prescribed than both < 2/3 (p = 0.032).
Rosa et al. (13) 2016 Brazil The Portuguese version was easily introduced into four Brazilian ICUs, and the prevalence of patients with a high score was 46%.
Coltman et al. (15) United States Patients determined to be at nutritional risk using the NUTRIC score alone or in combination with any other tool had the highest rates of death. A larger proportion of patients requiring additional rehabilitation after discharge was seen with NUTRIC score. Patients identified as being at nutritional risk or malnourished using NUTRIC had the longest hospital LOS and ICU LOS.
Özbilgin et al. (16) Turkey There was a positive correlation with mortality and the NUTRIC score (p=0.020) and pulmonary complications (p = 0.030).

DISCUSSION
This systematic review showed that many patients are at high nutritional risk on ICU admission. We also demonstrated that the NUTRIC score is becoming increasingly popular around the world. Application of the NUTRIC score in patients at the beginning of hospitalization in this sector has become relevant, and it is associated with MV, clinical complications, hospitalization time and death.
The NUTRIC score was validated by Heyland et al. and is the first nutritional risk assessment tool developed specifically for ICU patients that can identify patients at nutritional risk. (6) Heyland et al. considered the need for a more specific nutritional risk evaluation tool for ICU patients and found that inquiring about weight loss and their nutritional situation was insufficient, mainly due to the heterogeneous nature of ICU patients. (6) Thus, they incorporated different variables into the score (age, APACHE II, SOFA, comorbidities, days at hospital before ICU and IL-6). (6) Later, Rahman et al. validated the modified NUTRIC, which allows the exclusion of IL-6 levels, if not available, to assess nutritional risk at admission. (7) The NUTRIC score is based on a conceptual model designed around how to measure acute and chronic inflammation. (6) The importance of inflammation and illness severity are well recognized in the characterization of malnutrition, (1) such as its association with hospital length of stay. (20) Patients with a higher score have worse clinical outcomes, such as high mortality rates. (16) There are no traditional markers of nutritional risk, such as body mass index (BMI), weight loss, oral intake, or physical assessment, and the NUTRIC score only considers the severity of illness. (6) However, in the original study regarding validation, data such as BMI, percentage oral intake in the prior week, and percentage weight loss in the past three months were not associated with mortality. (5) Early identification of individuals at nutritional risk who may benefit from nutritional therapy is paramount in the hospital environment, including the ICU setting. (6) Heyland et al. considered that greater awareness of nutritional risk assessment tools, such as the NUTRIC score, and risk factors, such as BMI and duration of ICU stay, may enhance the delivery of calories and protein to patients who need them the most. (6) Although many instruments have indicated that all critically ill patients are at nutritional risk due to their clinical conditions, (21,22) they may not have the same risk of adverse events related to malnutrition. (6) The NUTRIC score shows the importance of developing specific scores for individuals with particular clinical conditions. (13) Additionally, the NUTRIC score is a fast, practical instrument that can be incorporated into the routine care of ICUs. (13) One clear advantage of the NUTRIC score is its applicability in situations in which patients are unable to respond verbally, as in MV, since the variables used in this scoring system are objectively obtained from data routinely registered in patients' medical records. (13) Regarding the future perspectives for NUTRIC, its use is promising for health professionals. The use of IL-6 in the score makes it difficult to use because no study has included it. We must value studies that seek to simplify NUTRIC with variables that are more commonly available, such as CRP inclusion. (9) We found that there are many observational studies relating the NUTRIC score to unfavorable clinical outcomes, but only one group has performed an interventionist study. (7) Therefore, is necessary to conduct studies that show the relationship of clinical outcomes through NUTRIC intervention. In addition, it is important that its use is not limited to nutritionist as the NUTRIC score is capable of pointing out relevant clinical outcomes, such as complications and death.
This was the first systematic review of the use of the NUTRIC score. The number of studies that evaluated the performance or application of the score is relatively low because of its recent validation. We consider this a limitation of our review.

CONCLUSION
The NUTRIC score is related to clinical outcomes, length of hospital stay and death and is appropriate for use in critically ill patients in intensive care units. More studies are needed to evaluate this tool for this particular population.