Cirrhosis and liver cancer are currently the 11th and the 16th most common cause of death globally. In Latin America, cirrhosis is the 9th cause of death, representing 2.7% of the total deaths. Cirrhosis has an incidence of 15.6 per 100,000 inhabitants. It is within the top 20 causes of disability-adjusted life years and years of life lost, accounting for 1.6% and 2.1% of the worldwide burden. Excessive alcohol consumption is the most frequent cause of cirrhosis in the world; other causes are viruses, medication intake, and autoimmunity. Recently, the global epidemic of obesity and diabetes mellitus which affects 2 billion and 400 million subjects worldwide respectively, have significantly increased the incidence of non-alcoholic fatty liver disease (NAFLD) and related liver disease (cirrhosis and hepatocellular carcinoma) [1].

NAFLD is a metabolic disorder characterized by macro vesicular steatosis in more than 5% of hepatocytes in individuals without significant alcohol consumption, drug intake, or other known causes of steatosis [2]. It is closely associated with the metabolic syndrome (MS) which is a constellation of abnormalities, such as visceral obesity, insulin resistance, diabetes mellitus (DM), hypertriglyceridemia, and arterial hypertension. It is the hepatic manifestation of a disequilibrium in the metabolic status (MS, intestinal dysbiosis, adipokine changes, vitamin D deficiency, hypercortisolism) in which the central factor is insulin resistance [3]

NAFLD affects a third of the world's population and its incidence continues to increase. The risk of progression to cirrhosis in NAFLD seems to be low, with an incidence of 3.1% over a mean 7.6-year follow-up [4]. However, these figures can be underestimated, since in about 25–30% of patients, NAFLD is diagnosed in the inflammatory phase called non-alcoholic steatohepatitis (NASH), characterized by hepatocellular damage. Currently 30–40% of patients with NASH-cirrhosis succumb to a liver related death within 10 years and hospitalizations have increased 97% between 2000 and 2012 which have significantly increased the burden of this condition [5].

Advanced age, diabetes mellitus, and visceral obesity are the most important factors involved in the development of liver fibrosis in NAFLD. The stages of liver fibrosis are classified as F0 (without fibrosis), F4 (cirrhosis) and >F2 advanced or significant fibrosis [6]. Advanced fibrosis, (and not inflammation) is independently associated with a significant increase in the mortality rate from cardiovascular disease (CVD) and liver disease [7–9]. The progression rate from NASH to cirrhosis and hepatocellular carcinoma is from 5 to 18% when fibrosis is absent and of 40% when is present [10–12]. For this reason, the detection of advanced fibrosis is crucial for therapeutic decisions [13]. Previously, liver biopsy was the only available method to evaluate liver fibrosis. However, it is an invasive and expensive procedure, subjected to sampling errors and to severe complications. In recent years, several non-invasive methods, based on serological biomarkers and radiological methods have been developed [14].

The improvement of socioeconomic conditions, the migration to urban areas and the changes in habits and diet of the Mexican population in recent decades have resulted in a significant increase in the prevalence of obesity, CVD and DM [15,16]. According to the Ministry of Health of Mexico, the number one leading cause of mortality in 2008 was DM (14%), followed by CVD (11%) and stroke (5.6%) [17]. The International Diabetes Federation estimates that Mexico has the highest prevalence of DM of adults in OECD countries (15%) [18]. All the above discussed led us to speculate that screening for liver fibrosis in the general population of Mexico might detect naïf individuals affected by chronic liver disease in whom the application of therapeutic measures might reduce mortality, as well as the financial burden derived from cirrhosis and hepatocellular carcinoma.

In a first step, we determined the prevalence of advanced liver fibrosis in Mexican general population; in subjects without excessive alcohol consumption or a history of known liver disease using an inexpensive, readily available and extensively validated non-invasive method.

Although liver cirrhosis is a major health problem worldwide because of its high mortality and heavy financial burden, it has not received the required attention by policy-makers, public opinion, nor physicians.

Until now, most published screening studies for liver fibrosis have been performed in targeted population based on the presence of risk factors of NASH (DM, metabolic syndrome, visceral obesity). Only few screening studies in unselected population have been published worldwide [20–23] (Table 2). Fibro Test (FT) and transition elastography (TE) were used as screening tests. FT is a biomarker that includes α2 macroglobulin, apolipoprotein A1, haptoglobin, total bilirubin, and GGT [24]. TE involves measuring liver stiffness as a reflection of fibrosis [25]. Both methods have been extensively validated with histological confirmation of fibrosis of any etiology [26–28]. The prevalence of liver fibrosis obtained in unselected population ranged from 0.9 to 6.9% [20–23]. The studies showing a higher prevalence of fibrosis included individuals with a higher prevalence of DM, visceral obesity, metabolic syndrome or used lower cut-off points of the screening method. In the study with the lowest prevalence (0.9%), a high cut-off point of FT was used (>0.59) [22].

In other three studies that stratified members of the general population according to an age cutoff >40 and 45 years [29–31], the overall prevalence of fibrosis ranged from 2.6 to 7.5% (Table 2). The lowest estimate used a two-step approach with only half of the patients re-attending for the second test [29].

Finally, in a sub-analysis of one of these studies involving 832 patients with NAFLD (after excluding excessive alcohol consumption, viral infection and drug toxicity) the prevalence of fibrosis was higher than that observed in the total group (8.4 vs. 5.6%) (Table 2).

Our study is the first screening study to detect liver fibrosis in an unselected general population in Latin America. The epidemiological characteristics (high prevalence of obesity and DM) of the Mexican population made us to carry out this research. Besides, Mexican population have higher genetic susceptibility, than other ethnic groups, to develop NAFLD, since the PNPLA3 gene (which is closely related to the development of steatosis, NASH, and liver cirrhosis), has been found in 70% in Mexican population [32,33].

The prevalence of fibrosis obtained in our study (8.1%) was higher than that observed in other unselected population studies [20–23] probably due to the higher prevalence of DM and obesity of our population. This prevalence was also higher than that observed in the studies stratified according to an age >40 and 45 years [29,30] and similar to that observed in the sub-analysis of NAFLD individuals stratified by age >45 years [31]. Notwithstanding, our work included subjects aged between 20 and 75 years. In a sub-analysis of our study, with 372 individuals aged >45 years, the prevalence of fibrosis increased to 13.4% that was higher compared to all the above-reviewed screening studies (Table 2).

The FT (Fibro Test®) is a patented method and the TE (Fibroscan®) is a radiological procedure requiring specialized equipment and trained personnel, so they are expensive and do not have wide availability in many centers. Due to this, their use as screening tools for large populations may not be cost-effective.

In our study, we decided to use NAFLD fibrosis score due to its following characteristics:

• a.

  It uses an algorithm that includes clinical and biochemical parameters that can be easily obtained in the doctors’ office at a very low cost. In Mexico, the approximate calculated cost per test of NFS is $ 34 dollars. In contrast, the costs of Fibrotest® and Fibroscan® are $ 260 and $ 215 dollars respectively.

• b.

  It was designed for subjects with NAFLD [19].

• c.

  It has been extensively validated with liver biopsy showing AUROC values from 0.70 to 0.96, negative predictive values (NPV) from 81 to 96%, positive predictive values (PPV) of 90% and an accuracy from 66 to 90%. Therefore, it is considered as an excellent method to rule out fibrosis [19,34–43], so that the American Association for the Study of Liver Diseases (AASLD) recommends it as a useful tool for screening fibrosis in patients with NASH [44] (Table 3).

  Table 3.

  Results of studies of the NAFLD fibrosis score for diagnosis of advanced liver fibrosis (F>2) determined by biopsy in patients with NAFLD.

  Study (year, Country)	N	Design	AUROC	NPV (low cut-off)	PPV (high cut-off)	Accuracy
  Angulo P (2007, International)19	480	P, MC	0.88	0.93	0.90	0.90
  Ruffillo G (2011, Argentina)34	138	R, SC	0.70	0.81	1.00	0.82
  Demir M (2013, Germany)35	120	R, SC	0.96	0.96	1.00	0.84
  Pérez-Gutiérrez OZ (2013, México, Chile)36	228	R, TC	0.72	0.95	0.26	0.84
  Cales P (2009, France)38	235	P, TC	0.88	0.86	0.86	0.86
  Mc Pherson S (2010, United Kingdom)39	145	P, MC	0.81	0.92	0.79	0.88
  Wong VW (2010, France, China)40	274	P, TC	0.75	0.89	0.61	–
  Petta S (2015, Italy)41	321	P, TC	0.80	0.90	0.85	0.77
  Boursier J (2016, France)42	588	R, TC	0.71†	0.80	0.54	0.66
  Petta S (2017, Italy, France, China)43	761	P, MC	0.77	0.87	0.73	0.72

  †

  Cut off point: −1.036; P: prospective; MC: multicenter; TC: two centers; R: retrospective; SC: single center; NPV: negative predictive value; PPV: positive predictive value.
• d.

  It has a diagnostic performance mildly lower than the TE, which is currently the best noninvasive method for the diagnosis of fibrosis [45]. Besides, most published studies have reported that NFS has higher performance than other non-invasive biomarkers, such as APRI, FIB-4, BARD, AST/ALT ratio [46–49].

• e.

  It has outcome-predictive capacity in NAFLD patients. In studies with a follow-up from 5 to 14.5 years, patients with a high and indeterminate probability of fibrosis had significantly higher rate of death from any cause, due to CVD and liver disease compared to patients at low risk [37,47,50–52] (Table 4).

  Table 4.

  Results of studies of NAFLD fibrosis score for prediction of morbidity and mortality rates among patients with NAFLD.

  Author (year)	N	Design	Follow-up, yr.	Primary outcomes	Results
  Angulo P (2013)47	320	R, M	8.6	Liver related complications, death, or transplantation	The HRs for liver-related complications, deaths or transplantation in the indeterminate and high-risk groups, were significantly higher compared with the low-risk group. NFS was superior to APRI, BARD and Fib-4.
  Treeprasertsuk S (2013)50	302	R, SC	12	All cause, CVD and liver related death	A higher NFS at baseline and a new onset of CVD were significantly predictive of death. All-cause mortality: 6.6% (low NFS), 19.4% (indeterminate NFS) and 46.2% (high NFS).
  Kim WR (2013)37	4083†	P	14.5	All-cause mortality	High risk group showed 69% increased CVD mortality compared with low risk group for NFS, APRI and Fib-4.
  Xun YH (2014)51	180	R, SC	6.6	All-cause mortality	NFS was the only predictor of death (HR 2.74, 95% CI 1.67–4.5). NFS was superior to FIB-4, APRI and BARD.
  Sebastiani G (2015)52	148	R, SC	5	All-cause mortality	HR for all-cause mortality in the high-risk group (NFS>0.676) was: 11.9 (95% CI: 3.7–37.3).

  P: prospective; R: retrospective; SC: single center; HR: hazard ratio; CI: confidence interval; CVD: cardiovascular disease.

  †

  This analysis utilized the National Health and Nutrition Examination Survey conducted in 1988–1994.

The individuals detected here as having high probability of fibrosis, represent a relevant proportion of the general population which may be at risk of suffering a chronic liver disease, and with a well-documented probability of having complications and death from CVD, cirrhosis and hepatocellular carcinoma. It is important to highlight that 93% of these individuals had normal ALT blood levels. A similar prevalence of normal liver enzymes in high-risk subjects (43 to 91%) has been reported in other screening studies [20,23,29,30]. Considering that Mexico's population is 120 million inhabitants, of which 63.6% are over 20 years old (76, 320,000 inhabitants) [53], extrapolating the results of this study, it can be estimated that around 6,181,920 subjects (8.1%) may be at risk.

Our study has some limitations: (a) No reinvestigations in order to confirm advanced fibrosis were performed; (b) No follow-up to determine morbidity and mortality rates, in high-risk subjects was done; and (c) We did not take into account subjects with an indeterminate probability of fibrosis. The proportion of fibrosis in this group is highly variable (from 35 to 60%) [19,34–36]. However, these limitations were the result of the nature of our research conducted in the open population.

The results of this and other similar studies may support the need to perform screening for liver fibrosis in general populations as part of public health policies [54,55], particularly in those communities with a high risk of liver disease. Screening should be done ideally with a non-invasive and low-cost test, with adequate reliability for fibrosis detection, easy to apply and widely available. Whether or not one of the currently available tests may be able to gather these properties is unknown.

The justification for screening for liver fibrosis in general populations is because subjects with early stages of fibrosis are completely asymptomatic, most of them have normal transaminase blood levels and the fibrosis progression is quietly slow (7.7 years for each stage of fibrosis in NASH) [56]. In most cases, liver fibrosis manifests clinically in advanced stage, as decompensated cirrhosis, particularly in diseases in which serological markers for early detection are lacking (such as NAFLD). The detection of patients in the pre-cirrhotic phase of the process, would allow the application of preventive or corrective measures to reduce the progression of the disease and consequently, the mortality rate and the financial burden derived from the care of complications. Although the FDA has not approved any drug for the treatment of NASH, the substantial loss of body weight by bariatric surgery or lifestyle changes, reduces liver fibrosis and inflammation in patients with NASH and possibly improves the prognosis [57–59]. Besides, new drugs that promise to have important therapeutic effects (Obeticholic acid, selonsertif, elafibranor, cenicriviroc) are under intensive evaluation currently [60,61].

According to the results of our study in the adult Mexican general population, the prevalence of advanced liver fibrosis is 8.1%. In the future, studies that confirm or rule out fibrosis through complementary studies, and that include long-term follow up with the application of preventive and corrective measures and the analysis of the cost-benefit relationship will be decisive to determine the validity of systematic screening for liver fibrosis in general populations.

The study was approved by the Ethics Committee of the Faculty of Medicine of the Autonomous University of Nuevo León. Informed consent and assent were obtained from participants in the survey.

This study was funded by the Ministry of Health of the Government of the State of Nuevo León, Mexico and the Universidad Autonoma de Nuevo Leon, Mexico.

The authors have no conflicts of interest to declare.