Introduction
In recent years, there has been much interest in the role of nutrition therapy in critical illness. Increased awareness of clinical nutrition has been hypothesized to be extremely important for intensive care unit (ICU) patients. Critical illness is associated with a catabolic stress state and an altered inflammatory response that may contribute to complications such as increased infectious morbidity, multi-organ failure, and prolonged hospitalization [
1].
Careful supplementation and caloric and protein intake modulation can avoid under or overfeeding. Additionally, adequate nutritional interventions have been shown to attenuate the morbidity rate, decrease the length of hospital stay, and improve patient outcomes [
2].
International guidelines have been recently updated by the American Society of Parenteral and Enteral Nutrition/Society of Critical Care Medicine [
3] and the European Society of Clinical Nutrition and Metabolism (ESPEN) to integrate the best current knowledge and evidence from the literature with nutritional practices [
1].
Whereas the nutritional requirements vary according to the phase of critical illness and the heterogeneity of the ICU population, these guidelines provide a set of nutrition recommendations in the most frequent clinical situations encountered in daily practice in the ICU. However, translating evidence into practice is challenging, and there is an increasing need for protocol standardization based on the latest evidence to reduce practice variation and improve the overall quality of care. A robust nutrition stewardship program could gain a reputation if the concept spreads to various national programs and regulatory guidelines released recently [
4].
So far, despite these recommendations, studies have yet to assess the level of adherence to the ESPEN recommendations in the Italian context, except for a survey on nutrition support for critically ill patients during the COVID-19 pandemic [
5].
This survey aimed to provide a snapshot of the current clinical practice focusing on nutritional evaluation, management, and monitoring in Italian ICUs.
In this way, the Italian ICUs might confront policies based on their clinical practice and compare these to a worldwide reference database.
Methods
This was a nationwide online survey, developed by experts belonging to SIAARTI (the Italian Society of Anaesthesia Analgesia Resuscitation and Intensive Care) board of the Metabolism, Nutrition and Renal Therapies section, composed of five intensivists, in 2022.
The current report adheres to the Consensus-Based Checklist for Reporting of Survey Studies—CROSS reporting guideline [
6] (Supplemental material
1).
Population
In the first phase, SIAARTI distributed the questionnaire to all the Directors of Italian ICUs to be filled out by the referring physician for the nutrition and metabolism field. In the second phase, the board disseminated the questionnaire via social media to ICUs who did not answer previously with the same purpose. A short introduction and a link to the survey were available to share on social media. To avoid multiple answers from the same center, only one response was considered for each Intensive Care Unit. No monetary incentives were provided to the respondents.
Data was gathered from October 1 to December 31, 2022. No inclusion and exclusion criteria were applied for the participation to collect representative data of the national scenario. Participation was anonymous; respondents voluntarily provided clear indications about the purpose of the survey and the use of the collected data.
We used a convenience sampling strategy [
7].
Survey development
The questionnaire was based on a preliminary updated analysis of the literature on this topic and international guidelines; a panel member (AC) drafted the first version of the survey and spread it to the other members who revised and approved the final version. The board unanimously voted and approved the final formulation of the questions.
A closed structure was used to avoid multiple answers from the same respondent, and access to the questionnaire was protected by a unique, anonymous identifier assigned to each respondent. The questionnaire consisted of 30 questions in total with multiple answers.
The survey explored four domains:
1)
ICU size (number of beds), ICU type (medical, surgical, cardio-surgical, neurosurgical, pediatric), hospital size, and type of hospital (academic, non-academic, private, Scientific Institute for Research, Hospitalization, and Healthcare-IRCCS) were collected in questions from 1 to 5.
2)
Nutritional assessment in critically ill patients was contained in questions 6 to 13.
3)
Nutritional management was investigated by questions from 14 to 21.
4)
The nutritional monitoring section was collected in questions from 22 to 30.
Consistency and completion of all items were obtained using server-side techniques, such as displaying the questionnaire after submission and highlighting unanswered mandatory items The responses were reviewed and edited during a final step, displaying a questionnaire summary, and requesting confirmation until the final submission. Supplemental material
2 shows an adapted English version of the online survey.
The questionnaire was built using Survey Monkey Platinum (SurveyMonkey Inc., San Mateo, CA, USA). Respondents were instructed to answer questions from the perspective of their standard clinical practice. In case of additional questions, they could contact the board.
Data analysis
Data was downloaded as an Excel file (Microsoft Corp, Redmond, WA, USA) and analyzed using the Jamovi software (version 1.8.4.0) for descriptive statistics. Answers were included in the analysis if Sect. 1 and at least one question from the other questionnaire sections were answered. Missing answers were included in the analysis.
The exclusion criteria included duplicated answers from the same ICU participant.
Data are presented as numbers, mean ± standard deviation (SD).
Discussion
The main findings of the current nationwide survey can be summarised as follows:
in at least half of the included ICUs, there is a standardization of clinical nutritional practice based on the use of protocols that essentially provide for the administration of EN continuously and with a non-advanced metabolic evaluation of the patient. Furthermore, glycemic control is entirely the prerogative of the nursing staff in less than half of the ICUs.
Determination of nutritional status is not a straightforward process, and the recently developed Global Leadership Initiative on Malnutrition (GLIM) criteria consider the coexistence of phenotypic and etiologic criteria [
8]. However, in critically ill patients, the diagnosis of malnutrition is made difficult by the challenges of determining food intake and weight loss. Indeed, since nutritional support aims to preserve muscle mass in patients without malnutrition, the nutritional risk is even more important than nutritional status, which is often determined by the severity of the disease with no regard to nutritional status [
8].
Even though a general clinical assessment has been recommended to assess malnutrition in ICU patients [
2], most ICUs do not evaluate the nutritional risk at the ICU admission. A lack of awareness of the importance of early recognition and treatment of malnutrition among healthcare team members remains a significant challenge, particularly in the intensive care setting.
So far, several tools have been developed for nutrition screening and assessment of hospitalized patients and the modified Nutrition Risk in the Critically Ill (mNUTRIC) has been suggested for the nutritional risk assessment of critically ill patients [
2,
9,
10].
We found that most ICUs do not use any score. Among all screening tools, NRS and NUTRIC scores are mainly used, maybe because they are the easiest and quickest to calculate and have the most robust predictive value for mortality [
11,
12].
However, the indication to tailoring the nutritional therapy to minimize under or overfeeding is widespread in all Italian ICUs. Based on the results of this survey, predictive equations remain the most common resting energy expenditure (REE) estimation method. Commonly used in clinical practice is 25 kcal/kg.
Importantly, predictive equations tend to over or under-estimate REE with an accuracy rate, defined as % of patients where the predicted value by the equation is within 10% of the measured value by indirect calorimeters (IC), of 12% for 25 kcal/kg and 30% for Harris-Benedict in critically ill setting [
13].
IC is still unavailable in most Italian ICUs, whereas it is the gold standard for determining REE [
14]. Factors limiting the reliability and feasibility of IC measurements are agitation, fever, sedatives, and vasoactive adjustments. Likewise, air leakages in respiratory circuits, mechanical ventilation with PEEP > 10 or with FiO2 > 80%, non-invasive ventilation, ECMO, dialysis, or continuous renal replacement therapy [
14].
The accuracy of caloric intake evaluation can further decrease if propofol, citrate, and dextrose intake are not considered [
2,
15,
16].
The evaluation of nitrogen balance (NB) is generally performed once a week in ICUs. The NB could be considered an excellent marker to establish dietary protein requirements in critically ill patients whereas it did not appear to predict clinical outcomes [
17,
18]. The latest recent meta-analysis showed that improved NB was associated with all-cause mortality in critically ill patients [
19]. This highlights the requirement for dynamic monitoring of NB during nutrition treatment [
19].
The nutritional evaluation is personalized according to the potential organ failure of critically ill patients in most ICUs and the intensivist is the nutrition leader. According to a recent study, critical care physicians’ knowledge and understanding of nutritional therapy are limited, especially in supportive preparation [
20].
In the future, the continuing education of all intensivists, rather than only the leader, should emphasize the comprehensiveness and importance of nutritional management and encourage them to cooperate with dietitians to promote the development of protocols and standardization of therapy.
The use of the ICU nutritional protocol in Italy could be debated. Recommendations for medical nutritional therapy in critically ill patients vary among guidelines [
1‐
3,
21]. For these reasons, implementing specific recommendations into clinical routine remains often insufficient.
The scientific community still debates when to start nutrition in critically ill patients. According to guidelines, nutrition treatment usually begins within 2 days from admission [
1].
EN is the nutrition of choice in the first days from admission in the ICU for more than half of ICUs. Supplemental parenteral nutrition (SPN) is used in most cases when EN is insufficient. At the same time, in a small percentage of cases, PN is associated with EN in the first 24 h, and rarely, TPN is the first nutritional choice for ICU patients.
Regarding some particular issues in critical illness, the guidelines favor early EN in patients receiving ECMO, prone positioning, and muscle relaxants because it reduce infectious complications. Our survey showed that this indication seems fully respected and well-known [
1] and physician respondents know the indication of nutrition routes in shock conditions and hemodynamic alterations [
1].
Furthermore, the preferred way of nutrition administration is continuous infusion, as suggested by the guidelines [
1]. Another important item investigated by the survey is patients’ glycaemic status and its management by nurses. From the obtained answers, most ICUs use a glycaemic protocol according to which nurses correct glycemia on their own. The most used glycaemic target in non-diabetic critically ill patients is 140–180 mg/dl.) [
1].
Besides, blood glucose sample is drawn mainly from arterial blood gas analysis whereas the ESPEN Guidelines suggest how blood should preferentially be drawn from central venous or arterial blood, avoiding capillary pricks in critically ill patients as several sources of interference are likely [
2,
22,
23].
Other topics analyzed by the survey are the management of residual gastric volume and EN feeding intolerance. In several studies, the frequency of RGV measurement was every 6–8 h [
16,
17]. However, the need to be more consensus about the rate of RGV threshold persists [
24,
25]. In almost all ICUs, the dose of nutrition prescribed is the actual nutrition administered. In the case of RGV > 300–500 ml, the most common behavior in ICUs is to stop EN. ESPEN states enteral feeding should be delayed when RGV is > 500 mL/6 h. Furthermore, in patients with gastric feeding intolerance not solved with prokinetic agents, post-pyloric feeding (mainly jejunal) should be used, especially in patients at high risk for aspiration [
1]. In Italy, the post-pyloric approach is common in ICUs.
In our study, ICUs still lack confidence regarding the use of ultrasonography (US) to assess a patient’s nutritional state. Indeed, diagnosing both malnutrition and sarcopenia requires assessing lean body mass with validated methods. Critically ill patients can lose up to 15% of their total muscle mass in the first week of stay [
26], which has been associated with detrimental long-term effects. The monitoring of lean body mass with validated methods has then been suggested as a critical component of the assessment of critically ill patients, to assess the current muscle mass for the nutritional diagnosis and risk stratification, to monitor the progression of muscle loss and/or recovery of muscle mass and to evaluate the success or failure of therapeutic interventions [
27].
In a recent study, US muscle mass assessment was able to detect short-term changes in critically ill patients and it was also identified as a useful follow-up tool [
28‐
30].
Body composition assessment is a relatively new practice in the intensive care field. Despite some technical limitations in critically ill patients, their use is steadily increasing, and the survey findings pave the road for planning educational interventions to spread further the application of these tools besides the research field.
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.