The incidence of non-alcoholic fatty liver disease (NAFLD) has increased in recent years, with a global prevalence of 25%. The capacity of the liver to process primary metabolic energy substrates diminishes following the development of NAFLD. This results in the accumulation of toxic lipids and the induction of liver cell stress, which gradually leads to inflammation, tissue damage, and even death. Although NAFLD is the leading cause of chronic liver disease, most patients are diagnosed with mild or mild-to-moderate fatty liver disease, without further progression to fibrosis or advanced cirrhosis. However, similar to diabetes, hypertension, and metabolic syndrome, the high prevalence of NAFLD and the possibility of disease progression still pose a potential risk to patient health [1, 2].

The majority of NAFLD patients suffer obesity, and generally present with a heavier body weight and higher body mass index (BMI). In contrast, malnutrition is generally associated with lower BMI. Nevertheless, recent studies have emphasized that BMI is not a good indicator of nutritional status, and some obesity may actually be due to poor nutritional quality [3, 4]. Malnutrition and changes in the intestinal microbiota can also affect the progression of NAFLD [1, 5]. Obesity, usually defined as having too much body fat, is damaging to human health. At present, the incidence of obesity is still on the rise worldwide [6]. In recent years, people have gained a lot of knowledge related to the biology of obesity, but the information is still limited in controlling the prevalence of obesity. More and more scholars believe that BMI alone cannot be relied on to assess obesity, and that insulin resistance and genetics seem to play a more important role in determining the risk of obesity comorbidities. There is also much unknown about the effects of interactions between individual behavior and life circumstances. Sarcopenic obesity has become an important research area to verify the poor prognosis of malnutrition-related diseases in the future [7]. However, BMI is still used as easy tool to quickly and directly classify patients for risk and monitor changes in obesity [8].

Malnutrition is a common late-stage clinical consequence during the course of organ dysfunction and death in critically ill patients. It is often caused by inadequate nutrient intake, malabsorption, or disproportionality. However, early nutritional status is a significant predictor of disease progression and long-term prognosis in such patients. Severe malnutrition is strongly linked to prognosis and complications, but such relationships are still unclear in NAFLD patients, especially the early nutritional status.

Recently, several malnutrition scales have attracted increasing attention and are now widely used in the clinical setting. These include the controlling nutritional status (CONUT) score, prognostic nutrition index (PNI), and nutrition risk index (NRI), all of which provide an accurate evaluation of the nutritional status and malnutrition. These three objective scales provide a better reflection of the patients’ nutritional status than BMI. Specifically, the malnutrition status determined by the scales has been correlated with an elevated risk of adverse long-term consequences, including death and severe disability [9]. Malnutrition also contributes to a higher incidence of dysfunction and mortality in patients with diabetes, chronic kidney disease (CKD) and cardiovascular disease. The aim of this study was therefore to explore the prevalence of malnutrition among NAFLD patients in the United States, as well its association with adverse clinical events. In addition, we explored the predictive significance of malnutrition for the prognosis of NAFLD patients, as evaluated by CONUT, PNI and NRI.

NAFLD is the most common chronic disease in both adults and children globally, with a prevalence of approximately 30% in Western countries. More than one quarter of the United States population shows a positive diagnosis for NAFLD with ultrasound [17]. This varies from steatosis or simple fat accumulation exceeding the standard ratio, to non-alcoholic steatohepatitis (NASH), and to more severe inflammation and hepatocyte death. These conditions accelerate the progression to liver fibrosis or cirrhosis, making NAFLD the most common cause of liver disease [2, 14]. In the present study, VCTE was used as a non-invasive diagnostic tool to obtain a relatively accurate assessment of hepatic steatosis, thereby allowing the prevalence of NAFLD to be determined through CAP. It is well known that ultrasound propagation is significantly attenuated in liver adipose tissue. The VCTE technique exploits this feature to assess the degree of hepatic steatosis by CAP quantitative analysis, thus providing a novel technique for the diagnosis of fatty liver. VCTE can quantitatively assess the liver tissue in a non-invasive way and at multiple angles to provide reliable, accurate and intuitive results. Although it may not be completely sensitive, fatty liver can be detected when hepatic steatosis reaches 20%. Compared with invasive liver biopsy, VCTE is safe and has a wider detection area, thus making it more suitable for large-scale studies and minimizing the interference caused by sampling error [18].

NAFLD is generally described as a hepatic manifestation of metabolic syndrome [19]. In October 2020, the Asia Pacific Society for the Study of the Liver (APASL) issued guidelines for the diagnosis and management of metabolically-associated fatty liver disease (MAFLD). The specific diagnostic criteria are the presence of hepatic steatosis with one of the following three conditions: overweight/obesity, type 2 diabetes, or metabolic dysfunction. Therefore, most participants in the current study met the diagnostic criteria for both MAFLD and NAFLD. However, some NAFLD subjects with normal weight and no diabetes could not be classified as MAFLD [20, 21, 22]. A growing number of studies have reported the presence of nutrient deficiencies in NAFLD patients that are likely to be related to prognosis. For example, low serum vitamin D levels correlate with both the severity of liver inflammation and the progression of NAFLD [23, 24, 25]. Vitamin D has anti-proliferative, anti-inflammatory, and also anti-fibrotic effects [26]. It is still unclear whether vitamin D deficiency is a contributing factor for NAFLD, or for the impaired metabolic capacity following hepatic steatosis. However, vitamin D deficiency is prevalent in both NAFLD adolescents and adults [23]. A study of Chinese stroke patients reported that a large proportion of those who were obese or overweight, as defined by BMI, also had malnutrition and a higher mortality rate [27]. Hence, the nutritional status of the overweight or obese population should not be overlooked, especially in patients with NAFLD.

In 2018, the Global Leadership Initiative on Malnutrition proposed diagnostic criteria for malnutrition in order to build a global consensus on the concept of adult malnutrition in clinical settings. Multiple assessment tools for nutritional status were recommended so that researchers could choose the appropriate tools for screening based on nutritional assessment objectives, subject characteristics, and clinical needs. Nutritional risk screening 2002 (NRS-2002) and the malnutrition universal screening tool (MUST) can effectively identify malnutrition in hospitalized patients [28, 29]. For cerebral stroke patients, malnutrition evaluated by the geriatric nutritional risk index (GNRI) has been linked to long-term all-cause mortality [27]. For heart failure patients, mini nutritional assessment (MNA) and MNA-short form (MNA-SF) are the optimal nutritional assessment and screening tools used to predict outcome [30]. CONUT is also a predictor of mortality from cardiovascular disease and chronic liver disease [31, 32, 33, 34].

Unfortunately, early nutrition issues among NAFLD patients are not yet well recognized in clinical practice, and relevant nutritional therapy interventions have not been fully utilized. The incidence of malnutrition in patients with advanced liver disease is high, but their nutritional grade varies greatly between different studies. The impact of late complications and the choice of screening tools also pose major challenges for clinical nutritional assessment. Early nutritional assessment therefore appears to be increasingly important. CONUT, PNI and NRI were used in the present study to evaluate the nutritional status of NAFLD patients. All three scales use serum albumin to assess body protein storage. Lymphocyte count is also included in CONUT and PNI, while height and weight are included in NRI. Albumin is the most frequently used blood biomarker for the screening of malnutrition. A previous study on the relationship between NAFLD and hemodialysis (HD) reported a negative correlation between the severity of hepatic steatosis and serum albumin levels that may affect the prognosis of HD patients [35]. Another study found that NAFLD patients with low serum albumin levels had significantly higher mortality from COVID-19 infection [36]. In addition to serum albumin, TC and lymphocyte count are two essential biomarkers of malnutrition in the elderly population [37] as well as having prognostic value [38, 39]. In summary, the CONUT, PNI and NRI tools used in the present study include serum albumin, TC and lymphocyte count, thereby allowing reliable evaluation of nutritional status.

The incidence of malnutrition determined by CONUT, PNI and NRI in our study population was 18%, 11% and 1%, respectively. The much lower incidence found with NRI might be due to the influence of body weight on nutritional assessment with this scale. The different results suggest that each method is dependent on the characteristics of subjects. Nevertheless, the three scales showed similar prognostic performance in our study, with each finding that all-cause mortality was higher in participants with malnutrition. This association was even stronger in those with NAFLD, highlighting that more attention should be paid to the assessment of early nutritional status and to nutritional grade stratification in such patients. Additional health education, preventive intervention and targeted treatments should be offered to individuals with NAFLD.

A total of 2848 participants in this study were obese or overweight, with 40% having a BMI of $\ge$30. In the NAFLD group (n = 1737), 91% of participants had a BMI $\ge$25, and $>$60% were overweight. A heavier body weight has traditionally been considered to indicate over-nutrition, or good nutritional status. The proportion of obese or overweight NAFLD patients (n = 1586) with malnutrition according to CONUT, PNI and NRI was 18% (n = 287), 22% (n = 335), and 0.1% (n = 2), respectively. Moreover, these patients had increased all-cause mortality (Fig. 3). The concept of high body weight being indicative of good nutritional status should be reviewed, since the BMI value alone does not appear to be a good indicator of nutritional status in humans. More comprehensive nutritional assessment methods need to be incorporated into clinical practice. Although many studies have recommended different methods, a consensus has yet to be reached.

Fig. 3.

The prevalence of malnutrition in NAFLD patients stratified by BMI. Abbreviations: BMI, body mass index; NAFLD, non-alcoholic fatty liver disease.

The CONUT, PNI and NRI scales are easy to perform and analyze, and can rapidly detect malnutrition in NAFLD patients. Nevertheless, there are several limitations to this study. First, cholesterol levels are generally higher in NAFLD patients, and the CONUT scoring criteria could underestimate the nutritional status. Similarly, the NRI method might also underestimate the nutrition status in some underweight patients. Second, the number of participants with moderate or severe malnutrition in this study was limited and could not be further examined using a more meticulous nutritional stratification method. Finally, the effect of improving the nutritional status on survival outcome was not examined. Further studies are therefore required to better understand the impact of nutritional intervention on patients with malnutrition, as well as to generalize the current findings to other populations.

In summary, malnutrition is frequently observed in NAFLD patients, including those who are overweight or obese. NAFLD patients with malnutrition have significantly higher all-cause mortality compared to those with a normal nutritional status. As NAFLD becomes more common, greater attention should be paid to nutritional status. Clinicians should also provide nutritional guidance and supplementation in a timely manner according to individual circumstances. Finally, the three nutrition assessment tools employed in this study were shown to be feasible and effective for the prediction of mortality among NAFLD cases.

Datasets are not available to the public but are available upon reasonable request from the corresponding author.

XJ, Conceptualization; writing-original draft; data analysis and interpretation. LJ, Conceptualization; writing-review and editing; data acquisition and curation. CHH, Writing-original draft; data analysis; software. LQ, Design; Writing-review and editing; data analysis. WYY, Writing-review and editing; data acquisition and analysis. CXJ, Writing-review and editing; data acquisition and analysis. KXJ, Conceptualization; design; writing-review and editing. All authors contributed to editorial changes in the manuscript. All authors read and approved the final manuscript. All authors have participated sufficiently in the work and agreed to be accountable for all aspects of the work.

The NHANES survey data is publicly available. The survey protocol was approved by the National Center for Health Statistics Ethics Review Board. The studies involving human participants were reviewed and approved by Ethical approval for the use of the NHANES survey data from 2017 to 2018 were obtained from the National Center for Health Statistics (NCHS) Research Ethics Review Board (ERB) through Protocol Number #2018-01. All methods were undertaken in accordance with the relevant guidelines and regulations. NHANES has previously obtained written informed consent from all participants.

Medical Science and Technology Project of Zhejiang Province (grant 2020KY869).

The authors declare no conflict of interest.

Supplementary material associated with this article can be found, in the online version, at https://doi.org/10.31083/IJVNR26099.