The Nutritional Status of Patients Hospitalized in the ICUs of Iranian Hospitals: A Systematic Review

Introduction

Literature review indicates that the frequency of undernutrition among hospitalized patients ranges from 18.2%-40% (1). Undernutrition incidence among critically ill patients is anticipated to be considerably higher due to a sharp rise in basal metabolic rate (2,3) and the existence of several circumstances that lead to the administration of enteral diets (1). In the Intensive Care Units, patients receive continuous medical attention in an effort to reduce the risk of serious outcomes, such as malnutrition, infection, and ultimately death (4). Nutritional therapy is crucial in the medical treatment of critically ill patients since it provides the targeted supply of energy and nutrients, avoids or minimizes the development of undernutrition, and corrects nutritional changes in undernourished patients (5). In addition, nutritional therapy has lately been shown to serve a fundamental therapeutic function, directly influencing disease pathophysiology and, consequently, the clinical result (6), and play a vital role in reducing and managing patients' morbidities (7,8).

Recent evidence supports the hypothesis that nutrients and nutritional treatments affect the underlying illness process and patient outcomes (9); therefore, the focus has shifted from nutrition as support to nutrition as therapy. Without compulsory feeding, such as enteral nutrition (EN) through a tube put into the GI tract or parenteral nutrition (PN) straight into the bloodstream, ICU patients who are unable to nourish themselves orally quickly become malnourished (10). Unless a serious problem would delay the start of enteral feeding, it is preferable to use enteral nutrition rather than parenteral nutrition. Early enteral nutrition (starting 24-48 hours after admission) is the first line of recommended nutrition support since it lowers the risk of infection and mortality compared to late enteral nutrition (EN) and early parenteral nutrition (PN) (11-15). Malnutrition is linked to higher morbidity and mortality rates, nosocomial infections, lengthier hospital stays, a lower functional level upon ICU discharge, and higher hospital costs (16,17). The body's functional and metabolic abnormalities, which serve to explain the preceding events, are predicated on the idea that malnutrition affects virtually every organ and/or system in the human body. Malnourished patients exhibit poor immunological response, digestion, and absorption in theguts (18,19).

Malnutrition also exerts a negative impact on how quickly wounds heal (20). Researchers use several tools and criteria to evaluate and score healthy and adequate nutritional status. Numerous factors, including food/nutritional consumption, physical examination, disease severity, anthropometric data, and functional assessment are included in the methods used to assess nutritional risk (21). NUTRIC score, APACHE II score, and the Sequential Organ Failure Assessment (SOFA) score are some of these tools (22). The rate of hospital admissions in Iran are over two million each year, and 20%-30% of patients need intensive care (23). According to Iranian studies, the overall malnutrition rate has been reported as 20%-50% (24). Therefore, the present study was performed with aim to investigate nutritional support in ICU admitted patients in Iran by reviewing the studies conducted in this regard.

Methods

This systematic review study was conducted using the methodology described in the Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) and Cochrane Handbook for Systematic Reviews. This review presents its findings by using inclusion and exclusion criteria, searching a comprehensive database, removing irrelevant studies, evaluating the quality of included studies, and interpreting the data. Adult patients over 18 years admitted to ICUs in Iranian hospitals were the key inclusion criterion. All of the included studies were written in English or Persian. This study excluded all meta-analysis review articles, narratives and systematic reviews, editorial letters, conference/poster presentations, case studies, brief studies, and animal studies.

All research publications containing the keywords “intensive care unit”, “nutritional support”, “Iran”, “parenteral nutrition” and “enteral nutrition” were searched in five online databases (Web of Sciences, Medline, PubMed, Scopus, and SID) until April 9, 2023. Both the titles and the abstracts of were evaluated for relevance. The process of screening was carried out with the eligible criteria. Two researchers checked the inclusion and exclusion criteria, taking into account the required qualifications. The full-text editions of the publications were obtained for the final screening of the extracted papers. The papers that were duplicates, presented in languages other than English and Persian, or lacked adequate information were eliminated. The publications written in a gray literature format were also removed. Two researchers read and examined the abstracts and titles of the articles individually. In case of any ambiguity or question, they discussed together. The researchers gathered all relevant data after deciding their goal. The collected information was utilized to create a checklist. Figure 1 depicts a PRISMA flowchart of the selected articles and procedures.

Figure 1: The PRISMA flowchart of the procedure for selecting studies

Cochrane's risk of bias tool was used to assess the potential bias in the published studies by considering eight criteria: random sequence generation, allocation concealment, blinding of participants, blinding of researchers, blinding of the evaluation process, attrition bias, adequate outcome data, selective reporting, and the absence of other biases. The level of bias risk was indicated by the responses "Yes," "No," and "Unclear", which correspond to low, high, and unclear rates of bias, respectively (25).

Ethical Consideration

Since the article was a systematic review study, no ethical consideration is applied.

Results

The search in five websites revealed a total of 1,859 documents. The initial step was to eliminate duplicate articles (n=4). After analyzing the titles and abstracts of the publications, 52 articles that were irrelevant to the research goals were deleted. Other studies included narrative and systematic articles (n=3) and letters to the editor (n=1) were excluded. The full text of four articles could not be found. Finally, there were 19 studies left.

These studies (n=19) observed a total of 5,672 patients. Women comprised 40.67% (n=2,307) of the total sample. The patients were admitted to ICUs in the hospitals throughout Iran, including Mazandaran, Amol, Kermanshah, Mashhad, Tehran, Hamedan, Isfahan, Shiraz, Semnan, Rasht, and Urmia. The patients in medical intensive care units (MICU), neurosurgery intensive care units (NICU), surgery intensive care units (SICU), and general intensive care units (GICU) were monitored. Various tools were used to check the nutritional status of ICU patients in these 19 studies. NUTRIC and APACHE II used in five studies (25%), NRS (20%), SOFA (20%), and SGA (15%) scoring systems were placed in the next levels. BMI were recorded in 35% of the studied articles. Moreover, in 14 articles, the level of calorie and protein intake was significantly lower than the daily needs of the patients (23,26-37). Three studies (28,32,38) reported that the level of blood factors, such as albumin, significantly decreased among patients during hospitalization. In all studies, except for one (23), the results pointed to poor nutritional status among patients hospitalized in ICUs. Data extracted from the studies were presented in Tables 1 and 2.

Quality assessment of methodological approaches

Five of the submitted studies had an intermediate risk of bias, while others (14) had a low risk. An assessment of the quality of articles was displayed in Table 3 and Figure 2.

Figure 2. Quality of the articles to determine their bias

Discussion

The purpose of the present study was to assess the nutritional health of adult patients hospitalized in ICUs of Iranian hospitals. According to the findings of this systematic review, the risk of malnutrition in ICU patients is very high. In most cases, patients hospitalized in these units were at high risk of malnutrition, which can generate a myriad of problems for the patient and ultimately, the healthcare system.

Types of feeding methods in critically ill patients

Nutrition support therapy administers EN and/or PN (39). In ICUs, nutritional support is recognized as an important aspect for management of critically ill patients (40,41). Enteral nutrition utilizing nasogastric tubes (NGTs) is typically used to address the nutritional demands of the majority of critically ill patients who have a functional gastrointestinal system but are unable to do this orally (42,43). Even though EN has shown promising clinical results and lowers the mortality of malnourished critically ill patients. The majority of patients receiving tube feedings have an intolerance to feeding tubes, including vomiting, diarrhea, constipation, abdominal distention, regurgitation, and high gastric residual volume (GRV) (44,45). In fact, a drawback of EN is a possible reduction in nutritional sufficiency, especially during the acute illness phase and in the presence of gastrointestinal dysfunction (6,46). On the other hand, the PN has repeatedly been demonstrated in previous reviews to be associated with more serious infectious problems, although potentially better securing the intended nutritional intake (47-51).

Factors influencing the measurement and diagnosis of nutritional risk

Various criteria, including food/nutritional intake, physical examination, disease severity, anthropometric data, and functional assessment, are applied to identify nutritional risk. Some tools to evaluate nutritional risk have been confirmed in the hospital setting. Since almost all critically ill patients require mechanical ventilation and sedation, achieve many of these requirements are challenging (52). The articles examined in this study include several methods for measuring and estimating the risk of malnutrition, such as measuring calories, protein, carbohydrates, and fat taken daily, recording changes in weight and BMI, or the amount of arm circumference or thickness of skin folds. Considering the enormous quantities required to maintain hemodynamic stability, changes in weight might be impacted by fluid status, making it more challenging to evaluate muscle and fat loss. Numerous conventional techniques lack information on inflammatory status, which is essential in an ICU patient since it is one of the causes of hypermetabolic status and consequently, muscle wasting (21). Hospital malnutrition should be identified early, and effective nutritional intervention procedures must be determined through appropriate instruments (53). For instance, Valizade (33) and Ramezani (27) suggested that the presence of nutritionists in the ICU is very necessary, and the patients who were under the supervision of nutritionists had a higher protein and energy intake compared to other patients, while the frequency of their nutrition was less.

Malnutrition assessment tools

Researchers use several tools to measure nutritional status, such as the Nutrition Risk Screening 2002 (NRS e2002) (54) and the NUTRIC score (22) for identifying critically ill patients who would benefit from nutritional therapy (NT). The NUTRIC score, a tool developed by Heyland et al., is the first nutritional risk assessment tool developed especially for the ICU patients, linking starvation, inflammation, and outcomes. It can identify patients who need more aggressive nutritional treatment and extensive nutritional support depending on their nutritional risk (22).

Among the advantages of this privilege, we can refer to the fact that this score improves the death rates and ventilation times of critically ill patients who benefit from intensive protein-energy provision during ICU stay. In order to stratify effects according to baseline risk, the effects of nutritional therapies for ICU patients with particular baseline characteristics were assessed by the NUTRIC score (55). Mishamandani et al. reported that patients' NUTRIC scores increased significantly with age (23). The NUTRIC score assess nutritional risk and associated outcomes (mortality and length of ventilation) through combining pre-hospitalization parameters, such as chronic BMI and acute starvation (pre-hospital admission duration) with acute Interleukin-6 and chronic inflammatory parameters (number of comorbidities), as well as the severity of illness based on APACHE-II classification system and SOFA-scores, on ICU admission.

The NUTRIC score has been validated and a high score is linked to a longer time spent on mechanical ventilation and a greater 28-day death rate. On the other hand, although the NUTRIC score was validated within the same database, limiting its applicability outside of that database (22). Moreover, this method lacks basic nutrition risk indicators, including BMI, weight status, oral intake, or physical examination, and its lack of nutrition history data may restrict its clinical applicability (56). Furthermore, PANDORA (57), recently introduced the Patient and Nutrition Derived Outcome Risk Assessment Score. Nonetheless, this score has not yet been validated in an ICU context.

The Malnutrition Universal Screening Tool (MUST) is another nutritional risk assessment tool frequently used for hospitalized patients; however, it has limited efficacy in critically ill patients (56,58,59). Nevertheless, the Netherlands, for example, has been used MUST score as a quality measure for measuring hospitals for years. In Iran, Majari et al. (2021), in addition to the NUTRIC score, also used the MUST score to diagnose the risk of malnutrition among patients (26). It should also be highlighted that the MUST score was commonly utilized for critically ill patients until recently, when no other tool was available to assess nutritional risk in ICU patients (60). However, studies illustrated that the MUST score is less effective at predicting the future than the NUTRIC score. Furthermore, since it has not been validated in the ICU patients, the researchers suggest that the MUST score should not be used to assess nutritional risk in the ICU (55). The study by Majari also found that more energy intake was associated with a lower 28-day mortality rate in patients with a higher NUTRIC score (26). In the same context, Coltman et al. demonstrated that despite some restrictions, the SGA was thought to be the most useful instrument for evaluating nutrition status in the ICU population. It is worth noting that there was a considerable amount of overlap between various patient groups; moreover, the SGA and the NUTRIC score were able to identify nutritional risk or malnutrition despite their differences (56). In addition, Rosa et al. stated that in terms of patients requiring additional rehabilitation after discharge and ICU length of hospital stay, NUTRIC Score as a tool for identifying nutritional risk or malnutrition demonstrated similar outcomes to SGA; nonetheless, it was associated with high rate of mortality when compared to SGA (53).

The APACHE diagnostic scoring system was developed in 1981 to measure the severity of a condition. The tool used 34 physiologic variables to assess the level of illness. The APACHE scoring system has been the gold standard in intensive therapy. The APACHE score is directly correlated with hospital mortality. From time to time, new versions have been released (61). The 1985 release of APACHE II was a simplified improvement over APACHE (62). Three factors can be used to estimate the probability of death: patient's age, 12 acute physiological features (acute physiology score [APS]), chronic illnesses, and surgical interventions. Some applications of APACHE II are risk stratification, comparing the quality of care, and prognosis (63). Today, other versions of this tool have been introduced with some modifications. The studies conducted in Iran that were evaluated in this review also used different versions of the APACHE score (II and IV) (23,27,32,34,36). Researchers believe that these scoring systems have limitations in evaluating critically ill patients. For example, ICU mortality appeared to be the best outcome to discriminate between ICUs with good quality treatment (low risk-adjusted mortality) and ICUs with poor quality treatment (high risk-adjusted mortality). Prognostic indicators were established to compare the quality of health care provided by various ICUs. On the other hand, some authors have underlined that the prognostic significance of the grading systems that are currently available is heavily dependent on the treatment. Two patients who have a similar severity of disease may have significantly different scores based on whether or not they received adequate resuscitation (64). Another reason is that the prognosis assessment based on scores is only applicable to a research population, not to a specific patient. Furthermore, some parts of the scores are difficult to determine. A perfect diagnostic index should be based on factors identified in common regular examinations, applicable to a variety of patient populations, easily and broadly measured, and unrelated to treatment or individual opinions (65). Finally, the severity scores used nowadays are typically based on a prediction of death. Recent studies have demonstrated a new method for evaluating organ failure that emphasizes on morbidity than mortality (66).

Organ dysfunction/failure in critically ill patients can be described using the SOFA score, as demonstrated by Vincent et al., patients' conditions and disease progression can be observed and clarified with the help of regularly repeated scoring. The SOFA score could facilitate patient comparison, which would be useful for clinical trials (67). Respiratory, cardiovascular, renal, hematological, hepatic, and central nervous systems function are used to calculate the SOFA score, which evaluates organ dysfunction, morbidity, and mortality in the ICU patients (68). Some studies demonstrated that an increasing score of the SOFA(within the range of 0-4) indicates growing organ dysfunction (69). The mean score of SOFA was 6 in the studies evaluated in this paper, indicating that patients admitted to ICUs across the country have poor health status and receive inadequate and poor-quality nutrition (23,26,34,38).

One of the advantages of this systematic review study was the large number of cross-sectional longitudinal studies conducted in various cities around Iran that considered various hospitals and patients with severe diseases. On the other hand, the number of various assessment systems utilized in the studies can present problems with clarity of the results, considering that approved and recognized procedures were used in all of the studies around the world. However, the limitation of this study is the lack of a standardized measurement scale. The articles reviewed in this study utilized various methods for assessing and quantifying the risk of malnutrition, as well as measuring changes in weight, BMI, and other validated criteria. Furthermore, an additional crucial issue to consider is the varying lengths of stay for patients in the intensive care unit.

Implications for practice

The nutritional care services were in poor and inadequate condition in the hospitals included in this study. Furthermore, the patient's daily intake of calories, protein, and other nutrients was drastically less than the recommended amount, exerting adverse effects on patients and causing negative consequences. Researchers concluded that the presence of nutritionists is vital and necessary for each patient hospitalized in the ICU of Iranian hospitals. Nutritionists can contribute greatly to accelerated patient recovery and improving treatment progress by monitoring their health conditions and assessing their energy and nutritional requirements.

Acknowledgments

This study was approved by the Mashhad University of Medical Sciences with grant number 4020458. The authors would like to appreciate the cooperation of Nooshin Abdollahpour, who provided technical help and writing assistance.

Conflicts of interest

The authors declared no conflict of interest.

Funding

There was no financial support for this study.

Authors' Contributions

Zahra Feyzabadi: Conception and design, Saeid Jamalie: Data collection, Benyamin Fazli: Data collection, Seyyed Saeed Shams: Data analysis, Mohaddeseh Badpeyma: Data collection, Alireza Sedaghat: Conception and design. All authors contributed to the writing of the manuscript and discussed on the manuscript.