1. Background

Hepatic steatosis or fatty liver is a significant public health issue owing to its association with metabolic diseases such as diabetes mellitus (DM), obesity, and metabolic syndrome (METS). They are recognized as major causes of morbidity and mortality. Hepatic steatosis may occur secondary to nonalcoholic fatty liver disease (NAFLD), alcoholism, chemotherapy, toxicity, and infectious agents. Although NAFLD is asymptomatic in its initial stages, it leads to an increased incidence of hepatic and nonhepatic comorbidities, including cirrhosis, hepatocellular carcinoma, cardiovascular disease, and other metabolic disturbances​ [1]. This is particularly true in developed countries [2]. Notwithstanding, the burden of metabolic diseases (type 2 DM and fatty liver) has increased worldwide at all levels of development [3].

NAFLD has many spectra, ranging from simple fat accumulation with no symptoms (NAFLD) to symptomatic nonalcoholic steatohepatitis (NASH). Simple steatosis has a silent and benign course in the early stages; however, if it progresses to NASH, it can lead to liver cirrhosis and hepatocellular carcinoma [4]. Genetically susceptible individuals under adverse environmental conditions are prone to developing NAFLD. Intrahepatic lipid accumulation induces the production of reactive oxygen species, leading to local and systemic inflammation that progressively evolves into NASH. If these processes persist, interstitial collagen deposition and fibrosis can occur, leading to cirrhosis. A persistent inflammatory microenvironment favors cancer development [5]. In these advanced scenarios, the only viable treatment is liver transplantation (LT) [6].

It is estimated that more than a quarter of the population has NAFLD. In Latin America, the prevalence of NAFLD ranges from 17% to 33.5% [7], and NASH accounts for 75% of all chronic liver diseases. The global prevalence of NAFLD is estimated to be 25% in the general population. Most epidemiological studies on NAFLD have been conducted in the United States, where the prevalence is 24.13%. African Americans had the lowest prevalence, followed by European Americans and Hispanic Americans. Despite the ability to diagnose NAFLD through noninvasive methods, the exact prevalence of NASH in the general population remains unknown because, to date, only liver biopsy can precisely confirm the diagnosis, and indirect data suggest that the prevalence of NASH in the general population ranges from 3% to 5%, using estimates from American population data [8].

In the United States, healthcare related to liver disease care is the most expensive, with an estimated total cost of 13.1 billion dollars (80.6% indirect costs and 19.3% direct costs) [9]. According to data from the European Liver Transplant Registry (ELTR) and the United Network for Organ Sharing (UNOS) databases, NAFLD has been the fastest-growing transplant indication in the last 20 years. It is currently the second leading cause of transplant-requiring liver disease, present in 1321 (21.5%) adult patients who underwent LT in 2018. A similar trend was observed in European countries, where NAFLD-related cirrhosis accounted for 1.2% of LT in 2002 and 8.4% in 2016 [10].

To control the progression from NAFLD to NASH, it is essential to understand the underlying mechanisms. Therefore, this review aimed to systematically compile and summarize the scientific literature on early risk factors for the development of NAFLD.

2. Methodology

This systematic review was conducted and reported following the Preferred Reporting Items for Systematic Reviews (PROSPERO) (CRD42022357262) and Meta-Analyses (PRISMA) statement. Analytical observational data that included social, demographic, and clinical risk factors associated with NAFLD development in the adult and pediatric populations were analyzed. The following databases were consulted to identify indexed publications: the Cochrane Central Register of Controlled Trials (CENTRAL), the Cochrane Library, MEDLINE through the PUBMED search engine, and Latin American and Caribbean Health Sciences Literature (LILACS). In addition, snowball searches were conducted based on references from published publications or those provided by clinical experts. The protocol for this systematic review is registered in the International Prospective Register of Systematic Reviews (PROSPERO).

3. Results

Figure 1 displays a PRISMA flowchart depicting the article selection process. The search yielded 987 records. A total of 196 manuscripts were selected after screening the titles and abstracts. Downs and Black quality analysis yielded 39 articles for full-text reading. Of these, 9 studies were excluded due to heterogeneity or inconclusive results. Ultimately, 30 publications were included in this systematic review (Figure 1).

Studies from Korea, Japan, the United States, Turkey, Brazil, the Netherlands, Germany, Sweden, the United Kingdom, Spain, and China were included, comprising data on over 21 million individuals. Among the eligible studies, twelve had a prospective cohort design, two had a retrospective cohort design, nine were cross-sectional, and six were case-control studies. All study samples were community-based general populations and some were defined through data linkage. Table 1 presents the risk factors identified in the review process. In 21 articles, the study population consisted of middle-aged adults (18–60 years); six included young adults and adolescents, and two involved middle-aged adults and school-age individuals. The most commonly used method for detecting hepatic steatosis was ultrasound, which was used in 22 of the included studies.

4. Discussion

In this systematic review of 29 studies including over 21 million people, we addressed early risk factors for NAFLD development. Several clinical, metabolic, hormonal, inflammatory, and lifestyle features are associated with the onset of NAFLD. NAFLD and METS share pathophysiological pathways, which explain why they have several common risk factors, both of which are triggered by IR and systemic inflammation [1]. In up to 73% of the cases, both entities coexist, although some authors proposed NAFLD as an early marker of METS [46, 47].

Central obesity and overweight are classically associated with NAFLD [3]. However, there are different assessments for estimating the risk of NAFLD beyond BMI [13, 14]. The 2017-2018 National Health and Nutrition Examination Survey found that, after 10 years, a weight increase larger than 9.07 kg increases the odds of NAFLD (OR: 2.63 and 95% CI: 1.72–4.02) and that, for every kilogram weight increase, this risk increases by 3% [47]. This review found that DM facilitated NAFLD. Similarly, a meta-analysis including 3901 patients with type I DM found that 19.1% of them had NAFLD [48]. IR is crucial for NAFLD: when type II diabetics are compared with healthy controls and type I diabetics, they are found to have NAFLD more often (62.8% vs. 13.4% vs. 4.7%, respectively). Interestingly, in type I diabetes patients, NAFLD has been identified in those with uncompensated disease and overweight/obesity [49].

An increased BMI at conception, excessive weight gain, or GD increases the odds of NAFLD [22, 50, 51]; however, causality is not always determined, as approximately 10% of women of childbearing age have NAFLD prior to pregnancy [51, 52]. Furthermore, children whose mothers develop NAFLD during pregnancy are also at risk of developing this condition, especially if their mothers have abnormal periconceptional BMI, METS, abnormal weight gain after pregnancy, early pregnancy, tobacco exposure during pregnancy, or if the person had low weight at birth [50, 53, 54].

People from developed countries have increased exposure to screens [55]; however, this also occurs in developing countries [36]. In conjunction with increased working time, screen exposure is associated with NAFLD because it is related to poor feeding habits; omitting breakfast; eating out of time; instant food, alcohol, and tobacco consumption; overfeeding; chronic stress; and sleep deprivation [56]. Similarly, poor sleep quality may cause NAFLD as it enhances IR and inflammation [57]. In a survey of 143,306 healthy people, poor sleep quality was observed in 20.1% of the cases; in this group, 45.8% slept less than 5 h daily. After a 4-year follow-up, the HR for NAFLD was 1.19 if a person sleeps for less than 5 h [57]. Similarly, a Japanese study of 2172 healthy individuals found an NAFLD prevalence of 29.6%; higher rates were found in those who slept less than 6 h daily (OR: 1.44 (CI 95% = 1.06–1.96)) [58]. Interestingly, another study suggested that this effect could be mitigated if people catch up on weekends [59].

Tobacco enhances NAFLD by impairing fatty acid synthesis, IR, inflammation, and glucose metabolism [42]. In this review, a study found that smoking was related to NAFLD with an HR of 1.988 (CI 95%: 1.057 ± 3.595) in 3860 participants and that the HR increased as the number of cigarettes increased. Furthermore, a meta-analysis of 20 observational studies involving 20,149 patients found that smoking had an OR of 1.11 (CI 95%: 1.028–1.199), while passive smoking had an OR of 1.380 (CI 95%: 1.199–1.588). Unfortunately, smoking cessation did not prevent the onset of NAFLD (OR = 1.316 and CI 95%: 1.158–1.496) [60].

With regard to chronic inflammatory conditions, we found that psoriatic patients had an increased prevalence of NAFLD. When compared with the general population, the prevalence of NAFLD in this population is 17%–66% vs. 8%–35% [61]. It also causes more severe steatosis, fibrosis, and progression to NASH [61]. Moreover, HIV coinfection magnifies this effect [62], mainly because antiretroviral therapy and the virus itself impair lipid metabolism [63, 64].

Lastly, this review revealed that hormonal disturbances increase the risk of NAFLD. A metanalysis of 44,140 patients with thyroid disturbances found that clinical hypothyroidism was related to NAFLD, whereas subclinical hypothyroidism was not [65]; other meta-analyses that included 61,548 patients affirms that both entities increased NAFLD risk [2]. Similarly, NAFLD is 2–3.5 more common and severe in women than in men. This is explained in part by the fact that sex hormones play key roles in lipid and protein metabolism. In a study of 159 men with NAFLD, 26% had decreased testosterone levels. A decrease in this hormone level was also more frequent in patients with NASH (88% vs. 67%) and advanced fibrosis (27% vs. 14%) [67]. Interestingly, androgen excess, rather than deficiency, causes adverse effects in women. This is supported by a meta-analysis in which androgen levels of 13,721 men and 5840 women were compared, finding that androgens were markedly decreased in males with NAFLD, whereas they were increased in women with steatosis they were increased [66]. This also explains why women presenting with PCOS have NAFLD more often than the healthy population (37%–70% vs. 14%–34%) (OR = 3.93 and 95% CI: 2.17–7.11) [67]. The prevalence of NAFLD in women with PCOS is estimated to be 27%–62% [68]. Obesity is not a necessary factor for NAFLD in PCOS but may exacerbate the effects of PCOS on the liver [69].

This study had several limitations. First, it should be noted that some studies, but not all, focused on data from the general population, which allows the results to be applicable to unselected populations in some cases. Second, the studies were heterogeneous in design, definitions, and type of measurement. Each study set different thresholds for NAFLD diagnosis according to diagnostic technique (biopsy, ultrasound, CT, and MRI). Third, in most cases, coexistence rather than causality was described. Fourth, other potential risk factors for NAFLD were not included in this review because the study design did not match the eligibility criteria.

5. Conclusion

Despite this, this review summarizes substantial evidence on the association between different clinical and lifestyle risk factors and NAFLD, supporting the need for a more structured approach for the prevention and detection of nonalcoholic liver disease. Understanding these early risk factors is crucial for proposing prompt preventive and therapeutic interventions in exposed patients so that further major morbidities such as NASH and cirrhosis could be prevented. Given the increasing prevalence of obesity, unhealthy lifestyles, and METS, particularly during early ages, it is essential to develop screening strategies and follow-up starting in childhood.

This could be accomplished through public health policies that promote healthy lifestyles and follow-up, particularly in vulnerable populations who lack access to proper healthcare programs. Longitudinal studies and population-based follow-up focusing on how to effectively intervene in NAFLD risk factors should be proposed so that their results can guide the development of national policies aimed at preventing chronic conditions.

Conflicts of Interest

The authors declare no conflicts of interest.

Author Contributions

Omaira Valencia: Conceptualization (C), Methodology (M), Project Administration (PA), Supervision (S), Data Curation (DC), Formal Analysis (FA), Writing – Original Draft (W-OD), Writing – Review & Editing (W-R&E), Validation (V). Carolina López: Conceptualization, Extraccion Data, Formal Analysis, Writing – Review & Editing W-R&E), Validation (V). Carolina Filizzola: Conceptualization (C). Carolina Fillizola contributed to the initial conceptual discussions of the study. Esteban Vanegas-Duarte: Investigation (I), Data Curation (DC), Writing – Review & Editing (W-R&E), Validation (V). Alonso Vera Torres, Nicol´s Andrés Cortés Mejía, Diana Fernanda Bejarano Ramírez: Data Curation (DC), Writing – Review & Editing (W-R&E), Validation (V).

Funding

The authors did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors for the preparation of this article.