Not always autoimmune hepatitis: Hepatic steatosis is associated with autoantibodies positivity in patients with hepatocellular dysfunction

Abstract

BACKGROUND

Autoantibodies, including anti-nuclear antibodies (ANA), are typically present in autoimmune hepatitis (AIH), but may be detected in other liver conditions, including metabolic dysfunction-associated steatotic liver disease (MASLD), even in the absence of an autoimmune disorder.

AIM

To identify predictive factors for the non-confirmation of AIH diagnosis in patients with liver dysfunction of unknown etiology and positive autoantibodies.

METHODS

A retrospective cohort study was conducted in a university-affiliated hospital, including consecutive adult patients with liver dysfunction of unknown etiology and positive autoantibodies typically present in hepatic autoimmune disorders, who underwent liver biopsy, between June 2016 and August 2023. Patients with other known liver diseases or contraindications to liver biopsy were excluded, as well as those who underwent liver biopsy and had a poor specimen quality.

RESULTS

A total of 81 patients were included, of whom 53.1% were diagnosed with AIH and 46.9% with MASLD. ANA had a high sensitivity (83.7%) in diagnosing AIH but the lowest specificity (18.4%). In patients with ANA positivity, male, diabetic and obese individuals were more likely to have a non-confirmed AIH (P = 0.022, P = 0.039 and P = 0.046, respectively). Higher controlled attenuation parameter (CAP) values in patients with ANA positivity were associated with non-confirmed AIH diagnosis (288 ± 56 vs 244 ± 60, P = 0.012). In multivariate analysis, male gender and higher CAP values were independent predictive factors for non-confirmed AIH diagnosis (P = 0.011 and P = 0.034, respectively).

CONCLUSION

AIH was not confirmed in 47% of patients with liver dysfunction and positive autoantibodies. Multivariate analysis identified male gender and elevated CAP values as independent predictive factors for a non-confirmed diagnosis of AIH in patients with liver dysfunction and ANA positivity. Further prospective and multicenter validation is required in order to corroborate our findings.

Key Words: Autoimmune hepatitis; Metabolic dysfunction-associated steatotic liver disease; Autoantibodies; Liver biopsy; Controlled attenuation parameter

Core Tip: This retrospective study included 81 consecutive adult patients with liver dysfunction and positive autoantibodies undergoing liver biopsy. Autoimmune hepatitis was not confirmed in 47% of patients. In patients with anti-nuclear antibodies positivity, male gender and higher controlled attenuation parameter values emerged as independent predictive factors associated with non-confirmed diagnosis of autoimmune hepatitis. These findings generate a valuable hypothesis that may inform future updates to clinical decision algorithms, particularly considering the rising prevalence of metabolic dysfunction-associated steatotic liver disease.

INTRODUCTION

Autoimmune hepatitis (AIH) is a chronic, progressive, immune-mediated liver disorder characterized by the presence of specific autoantibodies and elevated serum immunoglobulins, with a higher prevalence among women[1,2]. AIH can present in a variety of clinical presentations, ranging from acute hepatic dysfunction to established cirrhosis[1,2]. The diagnosis of AIH relies on a combination of histological findings, clinical and biochemical features and the detection of characteristic autoantibodies[1,2]. These include non-organ-specific autoantibodies, namely anti-nuclear antibodies (ANA) and smooth muscle antibodies (SMA), and the organ-specific autoantibodies, such as anti-liver kidney microsomal antibodies (anti-LKM), anti-liver cytosol type 1 (anti-LC-1) and antibodies against soluble liver antigen/liver pancreas autoantigen (anti-SLA/LP)[1].

Metabolic dysfunction-associated steatotic liver disease (MASLD) is defined as a steatotic liver disease in the presence of one or more cardiometabolic risk factors in the absence of harmful alcohol intake[3]. It is the leading cause of chronic liver disease globally, with a prevalence estimated to be 25%-30% of the general adult population[4]. MASLD also encompasses a spectrum of liver disease, ranging from steatosis and steatohepatitis to fibrosis and cirrhosis, in which the progression is due to worsening metabolism and severe hepatic inflammation and hepatocyte injuries[3]. MASLD is associated with an increased risk of mortality from various causes, including cardiovascular disease, type 2 diabetes mellitus (T2DM), cirrhosis and malignancies such as hepatocellular carcinoma[3,4].

Autoantibodies, including ANA, are typical of AIH; however, they are not disease-specific and may also be detected in other conditions, including non-alcoholic fatty liver disease (NAFLD) and the recently classified MASLD, even in the absence of an autoimmune disorder[5,6]. The prevalence of positive ANA in patients with NAFLD has been reported to be 13%-43%[7-9]. The presence of ANA in NAFLD leads to the activation of pro-inflammatory cytokines during immune responses, which in turn stimulates macrophages and hepatic stellate cells, contributing to liver damage progression[10]. It remains uncertain whether more advanced stages of steatosis are associated with a higher likelihood of positive autoantibodies in patients with NAFLD in the absence of autoimmune disorders or AIH[10].

Liver biopsy is indicated in patients with liver dysfunction of unknown etiology and is the cornerstone in the diagnosis of AIH[1,2,11]. Furthermore, liver biopsy can identify patients with liver dysfunction of other etiologies besides AIH[10].

The aim of this study was to identify predictive factors for the non-confirmation of AIH diagnosis in patients with liver dysfunction of unknown etiology and positive autoantibodies. Other secondary aims were to compare the sensitivities and specificities of distinct autoantibodies for AIH diagnosis and to create a valuable hypothesis that may inform future updates to clinical decision algorithms for the management of patients with positive autoantibodies in suspected AIH.

MATERIALS AND METHODS

Study design and population

A retrospective cohort study was conducted in a university-affiliated hospital, including consecutive adult patients with liver dysfunction of unknown etiology and positive autoantibodies typically present in hepatic autoimmune disorders, who underwent liver biopsy, between June 2016 and August 2023. Patients with evidence of other autoimmune liver diseases (primary biliary cholangitis, primary sclerosing cholangitis, IgG4-associated cholangiopathy); chronic hepatitis B or C virus infection; Wilson’s disease; alcoholic liver disease; haemochromatosis; and/or contraindication for percutaneous liver biopsy (coagulopathy, pregnancy or ascites) were excluded. Patients who underwent liver biopsy and had a poor specimen quality (specimen measuring less than 15 mm or with less than 11 portal tracts on histological assessment) were also excluded. A flow diagram illustrating the exclusion process is represented in Figure 1. The study protocol conforms to the ethical guidelines of the 1975 Declaration of Helsinki as reflected in a priori approval by the Institutional Review Board (Ref: 116/2023).

Figure 1 Flow diagram illustrating the patients' inclusion and exclusion process.

Study definitions and procedures

AIH diagnosis relied on the presence of compatible histological findings supported by elevation of serum aminotransferases, positivity of characteristic autoantibodies and elevated IgG levels, as well as the exclusion of other liver diseases, according to the established criteria[1,2]. The histological features were categorized as typical, compatible or atypical of AIH[1,12]. The presence of interface hepatitis, lymphocytic or lymphoplasmacytic infiltrates in portal tracts and extending into the lobule, emperipolesis and hepatic rosette formation were regarded as typical[12]. Features of chronic hepatitis with lymphocytic infiltration without the criteria considered typical were considered compatible findings, and the presence of signs of another diagnosis such as steatohepatitis were considered atypical[12]. The non-organ-specific autoantibodies, namely ANA and SMA, and the organ-specific autoantibodies, such as anti-LKM, anti-LC-1 and anti-SLA/LP, were considered the serological markers of AIH[1]. Autoantibodies were considered positive when titers were above 1:160.

MASLD diagnosis was based on evidence of hepatic steatosis identified by imaging or liver biopsy in adult patients, associated with any cardiometabolic criteria and in the absence of alcohol consumption superior to 20 g per day in women or superior to 30 g per day in men[3]. The cardiometabolic criteria included: Overweight or obesity [body mass index (BMI) ≥ 25 kg/m² and/or a waist circumference ≥ 94 cm in men and ≥ 80 cm in women]; prediabetes (HbA1c 5.7%-6.4% or fasting plasma glucose 100-125 mg/dL), T2DM (HbA1c ≥ 6.5% or fasting plasma glucose ≥ 126 mg/dL) or treatment for T2DM; plasma triglycerides ≥ 150 mg/dL or lipid-lowering treatment; high-density lipoprotein-cholesterol ≤ 39 mg/dL in men or ≤ 50 mg/dL in women or lipid-lowering treatment; blood pressure ≥ 130/85 mmHg or treatment for hypertension[3].

Transient elastography (FibroScan^® Compact 530, Echosens, Paris, France) was performed after a minimum fasting period of 2 hours[13]. Liver stiffness measurement (LSM) and controlled attenuation parameter (CAP) were recorded and reported in kilopascals and decibels per square meter (db/m²), respectively. The procedure was conducted with the patient in the dorsal decubitus position, with the right arm in maximal abduction. The probe, covered with ultrasound gel, was placed over the right hepatic lobe through the 9^th to 11^th intercostal spaces along the mid-axillary line. The LSM was considered valid if at least 10 valid measurements were obtained and the interquartile range (IQR) to median (M) ratio was below 30%[13]. The M probe was used initially for all patients, except for those with a skin-to-liver capsule distance greater than 25 mm which is an independent predictor of M probe failure[14]. In cases of M probe failure, measurements were repeated using the XL probe. Patients with conditions known to impair LSM reliability, such as extrahepatic cholestasis, serum aminotransferase levels equal or superior to five times the upper limit of normality, right heart failure or other causes of hepatic congestion, were excluded from assessment. The LSM and CAP values considered were the ones closest to the liver biopsy, with a maximum interval of three months between the TE and the liver biopsy.

Liver biopsy was performed via a percutaneous transthoracic approach, with the patient in the supine position. Biopsy location was determined using a prior abdominal ultrasound, with the patient holding their breath in deep expiration[11]. Local anaesthesia was administered at the selected intercostal space along the mid-axillary line, midway between the dome and the tip of the liver, and a small skin incision was made[11]. A 16G Menghini needle was then inserted through the intercostal space, just above the lower rib, and advanced toward the xiphisternum, parallel to the bed surface[11]. Following the procedure, patients were positioned on their right side for 2 hours, and vital signs, including pulse and blood pressure, were monitored regularly to detect any adverse events[11]. All included patients underwent liver biopsy with a high-quality specimen, measuring at least 15 mm and showing a minimum of 11 portal tracts on histological assessment. Histological assessment was performed by two pathologists from our university hospital, highly experienced in liver specimen analysis.

Data collection

Demographic, clinical, biochemical, imaging and histological data were collected and registered anonymously in a database. Patients’ demographics included gender and age. The patients’ cardiometabolic comorbidities were reported, namely, arterial hypertension, T2DM, dyslipidemia, obesity and previous history of cardiovascular events. Obesity was regarded as a BMI equal or superior to 30.0 kg/m². Additionally, the presence of other autoimmune disorders was reported. Biochemical data included total bilirubin, aspartate aminotransferase (AST), alanine aminotransferase (ALT), gamma-glutamyl transferase (GGT) and alkaline phosphatase (ALP), as well as the IgG levels and the autoantibodies ANA, SMA, anti-LKM, anti-LC-1 and anti-SLA/LP. The LSM and the CAP values were evaluated by transient elastography, and the histological features present in liver biopsies were also reported.

Statistical analysis

Categorical variables were described as frequencies and percentages, and continuous variables as mean and SD if normally distributed, or as median and IQR if not normally distributed. Comparison of categorical variables was performed using the χ² or Fisher’s exact test, which were also used to calculate odds ratio (OR) and 95%CI. Means and medians of continuous variables were compared using independent group t tests or Mann–Whitney U test, respectively. Multivariable analysis was performed using a binary logistic regression including significant variables on univariable analysis and using the Wald method, to identify independent predictive factors associated with non-confirmation of AIH diagnosis in patients with liver dysfunction of unknown etiology and positive autoantibodies. The sensitivity and specificity of each antibody in confirming AIH diagnosis was also calculated. All reported P values are two-tailed, and a P value less than 0.05 was considered statistically significant. Statistical analysis software IBM SPSS version 27.0 (IBM Corp., Armonk, NY, United States) was used for all tests performed.

RESULTS

Baseline characteristics of the study population

A total of 81 patients were included, most were female (70.4%), with a mean age of 55 ± 15 years. The most commonly detected autoantibody was ANA in 67 patients (82.7%), followed by anti-SLA/LP in 13 (16.0%) and SMA in 10 patients (12.3%). A total of 19 patients (23.5%) tested positive for ANA concurrently with at least one other autoantibody, such as SMA, anti-LKM, anti-LC-1 or anti-SLA/LP.

Liver biopsy was performed in all patients, with 55 patients (67.9%) having lymphocyte infiltrate, 39 (48.1%) hepatic steatosis and 5 patients (6.2%) interface hepatitis. AIH was diagnosed in 43 patients (53.1%), and the remaining 38 patients (46.9%) were diagnosed with MASLD.

One patient (1.2%) experienced a self-limited hemorrhage following liver biopsy, which led to hospitalization due to hemoperitoneum. The patients’ baseline demographic and clinical characteristics, presence of autoantibodies and histological features are presented in Table 1.

Table 1 Patients’ baseline characteristics, n (%)/mean ± SD.

Variable	Patients’ characteristics
Gender
Female	57 (70.4)
Male	24 (29.6)
Age (years)	55 ± 15
Cardiometabolic factors
Arterial hypertension	31 (38.3)
T2DM	18 (22.2)
Dyslipidaemia	32 (39.5)
Obesity	26 (32.1)
Previous cardiovascular event	4 (4.9)
Presence of other autoimmune disorders	6 (7.4)
Positivity of autoantibodies
ANA	67 (82.7)
SMA	10 (12.3)
Anti-LKM	5 (6.3)
Anti-LC-1	6 (9.0)
Anti-SLA/LP	13 (16.0)
Elevated IgG levels	21 (29.6)
Histological findings
Hepatic steatosis	39 (48.1)
Lymphocyte infiltrate	55 (67.9)
Interface hepatitis	5 (6.2)
Emperipolesis	0 (0.0)
Hepatocyte rosettes	0 (0.0)
Histological criteria for AIH
Typical	5 (6.2)
Compatible	38 (46.9)
Atypical	38 (46.9)

T2DM: Type 2 diabetes mellitus; ANA: Anti-nuclear antibodies; SMA: Smooth muscle antibodies; Anti-LKM: Anti-liver kidney microsomal antibodies; Anti-LC-1: Anti-liver cytosol type 1; Anti-SLA/LP: Antibodies against soluble liver antigen/Liver pancreas autoantigen; AIH: Autoimmune hepatitis.

Autoantibodies’ sensitivity and specificity in the diagnosis of AIH

ANA had a high sensitivity (83.7%) in diagnosing AIH but the lowest specificity (18.4%) of all autoantibodies. The other evaluated autoantibodies had high specificity and low sensitivity in diagnosing AIH. The sensitivity and specificity of each autoantibody is presented in Table 2.

Table 2 Sensitivity and specificity of each autoantibody in diagnosing autoimmune hepatitis.

Autoantibody	Sensitivity	Specificity
ANA	83.7%	18.4%
SMA	18.6%	94.7%
Anti-LKM	7.0%	94.7%
Anti-LC-1	11.4%	93.8%
Anti-SLA/LP	14.0%	81.6%

ANA: Anti-nuclear antibodies; SMA: Smooth muscle antibodies; Anti-LKM: Anti-liver kidney microsomal antibodies; Anti-LC-1: Anti-liver cytosol type 1; Anti-SLA/LP: Antibodies against soluble liver antigen/Liver pancreas autoantigen.

The presence of elevated levels of IgG had a sensitivity of 38.5% and a specificity of 81.3% in diagnosing AIH.

Factors associated with AIH diagnosis in patients with ANA positivity

A total of 67 patients (82.7%) had ANA positivity. In these patients, AIH diagnosis was confirmed in 36 patients (53.7%) and the remaining 31 patients (46.3%) were diagnosed with MASLD. The ANA titers showed no correlation with the diagnosis of AIH (P = 0.274), as well as positivity of two or more autoantibodies (P = 0.126).

Male patients were 4 times more likely to have a non-confirmed diagnosis of AIH (OR = 3.61, 95%CI: 1.17-11.24, P = 0.022). Diabetics and obese patients were 3 times more likely to have a non-confirmed AIH diagnosis (OR = 3.42, 95%CI: 1.03-11.24, P = 0.039 and OR = 2.88, 95%CI: 1.00-8.26, P = 0.046, respectively). A higher CAP value was associated with non-confirmed diagnosis of AIH (288 ± 56 vs 244 ± 60, P = 0.012).

Patients with elevated IgG were 3 times more likely to have a confirmed diagnosis of AIH (OR = 3.27, 95%CI: 0.984-10.841, P = 0.048). Patients with confirmed AIH diagnosis had a higher median value of ALT [85 (154) vs 50 (54), P = 0.021].

Factors associated with AIH diagnosis in patients with ANA positivity are detailed in Table 3. In multivariate analysis, presented in Table 4, male gender and higher CAP values were independent predictive factors associated with non-confirmed diagnosis of AIH (P = 0.011 and P = 0.034, respectively).

Table 3 Factors associated with autoimmune hepatitis diagnosis in patients with anti-nuclear antibodies positivity, n (%)/mean ± SD.

Variable	AIH diagnosis confirmed (n = 36)	AIH diagnosis not confirmed (n = 31)	P value
Gender
Female	30 (83.3)	18 (58.1)	0.022
Male	6 (16.7)	13 (41.9)
Age (years)	55 ± 16	58 ± 13	0.333
Arterial hypertension	13 (36.1)	14 (45.2)	0.451
T2DM	5 (13.9)	11 (35.5)	0.039
Dyslipidaemia	13 (36.1)	16 (51.6)	0.202
Obesity	8 (22.2)	14 (45.2)	0.046
Previous cardiovascular event	2 (5.6)	2 (6.5)	1.000
Other autoimmune disorders	4 (11.1)	1 (3.2)	0.363
Positivity of autoantibodies
SMA	7 (19.4)	1 (3.2)	0.060
Anti-LKM	2 (5.6)	1 (3.2)	1.000
Anti-LC-1	4 (11.1)	1 (3.2)	0.353
Elevated IgG levels	14 (38.9)	5 (16.1)	0.048
Total bilirubin (mg/dL), median (IQR)	0.67 (0.52)	0.68 (0.60)	0.905
AST (UI/L), median (IQR)	66 (105)	36 (40)	0.092
ALT (UI/L), median (IQR)	85 (154)	50 (54)	0.021
GGT (UI/L), median (IQR)	99 (150)	52 (94)	0.065
ALP (UI/L), median (IQR)	106 (76)	91 (48)	0.258
CAP (dB/m2)	244 ± 60	288 ± 56	0.012
LSM (kPa), median (IQR)	5.4 (3.1)	6.7 (14.9)	0.205

AIH: Autoimmune hepatitis; T2DM: Type 2 diabetes mellitus; SMA: Smooth muscle antibodies; Anti-LKM: Anti-liver kidney microsomal antibodies; Anti-LC-1: Anti-liver cytosol type 1; Anti-SLA/LP: Antibodies against soluble liver antigen/Liver pancreas autoantigen; IQR: Interquartile range; AST: Aspartate aminotransferase; ALT: Alanine aminotransferase; GGT: Gamma-glutamyl transferase; ALP: Alkaline phosphatase; CAP: Controlled attenuation parameter; LSM: Liver stiffness measurement; IQR: Interquartile range.

Table 4 Multivariate analysis of predictive factors associated with non-confirmed diagnosis of autoimmune hepatitis in patients with anti-nuclear antibodies positivity.

Variable	Exp (B)	95%CI	P value
Male gender	6.883	1.549-30.578	0.011
T2DM	3.944	0.791-19.661	0.094
CAP (dB/m2)	1.014	1.001-1.027	0.034

T2DM: Type 2 diabetes mellitus; CAP: Controlled attenuation parameter.

DISCUSSION

Patients with significant liver steatosis, such as those with NAFLD, and liver dysfunction often present with positive serum autoantibodies, a feature also commonly seen in AIH[10]. Liver biopsy is frequently required to establish an accurate etiologic diagnosis, since the presence of autoantibodies alone does not confirm AIH[1,2,11]. However, liver biopsy is an invasive procedure associated with potential complications, such as bleeding[11]. Therefore, identifying which patients are more likely to truly have AIH is essential to determine those who would benefit the most from undergoing liver biopsy.

The most commonly detected autoantibody was ANA in 83% of patients. In fact, ANA had a high sensitivity in diagnosing AIH but the lowest specificity of all autoantibodies, which has been previously described[15,16]. Previous studies have identified an association between ANA positivity and the presence of NAFLD and the severity of non-alcoholic steatohepatitis (NASH)[17,18]. In fact, a recent systematic review and meta-analysis revealed that ANA was present in approximately one-quarter of patients with biopsy-proven NAFLD[10]. The other evaluated autoantibodies, namely SMA, anti-LKM, anti-LC-1 and anti-SLA/LP, had high specificity and low sensitivity in diagnosing AIH, thus the positivity of these antibodies, although not present in most patients, is associated with a more likely diagnosis of AIH. However, in our cohort, the presence of two or more positive autoantibody was not associated with a confirmed diagnosis of AIH, even though previous studies showed improved sensitivity in this scenario. A recent multicenter and longitudinal study has identified that ANA and ASMA were commonly found in patients with MASLD and the positivity in these antibodies was associated with a higher risk of liver-related outcomes and reduced survival[19]. The authors hypothesized that ANA and ASMA may serve as markers of subclinical inflammation and can help identify patients at higher risk for disease progression and poorer prognosis[19]. Though, the presence of these autoantibodies in patients with MASLD did not reflect typical histological features of AIH[19]. There is limited data available regarding other autoantibodies, and given the rarity of their positivity, it would be of interest to develop multicenter and large population-based studies to identify potential predictors of MASLD in patients testing positive for these antibodies. Nevertheless, considering their high specificity, it is unlikely that the presence of these antibodies alone would obviate the need for liver biopsy in clinical practice.

This retrospective study, which included adult patients with liver dysfunction and positive autoantibodies who underwent liver biopsy, aimed to identify predictive factors for the non-confirmation of AIH diagnosis in patients with liver dysfunction of unknown etiology and positive autoantibodies. In our cohort, AIH was diagnosed in only 53% of the patients, with the remaining 47% being diagnosed with MASLD. A recent meta-analysis revealed that the overall prevalence of ANA positivity in patients with NAFLD was 23%[10]. Other studies have reported the presence of positive autoantibodies in up to one third of patients with NAFLD without concurrent AIH, both in children and in adults[17,18,20-22]. Therefore, the approach to these patients should be reconsidered. Liver biopsy could be postponed in patients with non-specific autoantibody positivity and higher risk of non-confirmation of AIH diagnosis, since it is an invasive diagnostic procedure, with potential adverse events.

Our study also evaluated the presence of elevated levels of IgG, which had a low sensitivity but a high specificity of more than 80% in diagnosing AIH. Additionally, in patients with ANA positivity, those with elevated IgG were more likely to have a confirmed diagnosis of AIH. This parameter is also important in the diagnosis of AIH and has been previously demonstrated to be typical of AIH, although not necessarily present[1,2,23]. In a study which evaluated the prevalence of non-specific autoantibody positivity in overweight and obese children with NAFLD, a normal level of IgG was the most significant negative predictor of AIH[24].

Since ANA was the most commonly detected autoantibody and revealed the highest sensitivity with the lowest specificity, we particularly evaluated patients with ANA positivity to identify possible predictive factors for the non-confirmation of AIH diagnosis in these patients. Notably, we found that male patients were more likely to exhibit positive ANA without a subsequent diagnosis of AIH, suggesting that ANA positivity in men may not be indicative of underlying autoimmune liver disease. In fact, MASLD has been established as a pro-inflammatory condition[25], which can suggest that, in our male population, this inflammatory environment was likely the underlying mechanism leading to the development of non-specific autoantibodies, rather than the classical autoimmune pathways. Furthermore, this observation aligns with existing literature reporting a strong female predominance in most autoimmune conditions, including AIH[10]. The reasons for this sex-related discrepancy are not fully understood, but hormonal differences likely play a key role[26,27]. Estrogen and prolactin, both elevated in women, have pro-inflammatory properties that enhance T-cell and B-cell activation, promoting autoantibody production[26]. Additional contributing factors may include genetic predispositions and gender-related environmental exposures, such as differing patterns of cosmetic, hair dye, or toxin exposure, which can be involved in stimulating autoantibody production[27]. In contrast, androgens, which are higher in men, tend to exert immunosuppressive effects[26]. Thus, although female sex is a recognized risk factor for autoimmune conditions, the increased rate of ANA positivity in males without AIH could be representative of a higher prevalence of non-specific autoantibody positivity or false-positive results in this population.

In our cohort, patients with T2DM and obesity, which are typical features of cardiometabolic syndrome, were more likely to have a non-confirmed diagnosis of AIH. More importantly, our analysis also revealed that patients with liver dysfunction and ANA positivity and with higher CAP values were more likely to have non-confirmation of AIH diagnosis. In fact, CAP is a validated, non-invasive parameter that correlates with the degree of hepatic steatosis[13,28,29]. According to standardized CAP scoring systems, higher CAP values indicate more extensive fat infiltration in the liver: CAP values 150-248 dB/m² (S0) indicate a 0%-10% fatty change in the liver; 248-260 dB/m² (S1) indicate a 11%-33% fatty change; 260-280 dB/m² (S2) indicate a 34-66% fatty change in the liver; and CAP values of more than 280 dB/m² (S3) indicate there is a more than 67% fatty change in the liver[13,28,30]. A systematic review and meta-analysis which evaluated the diagnostic accuracy of CAP as a non-invasive test for steatosis in suspected NAFLD revealed that the mean area under the receiver operating characteristic value for CAP in the diagnosis of mild, moderate, and severe steatosis was 0.96, 0.82 and 0.70, respectively[29]. Therefore, our findings suggest that greater hepatic fat infiltration and more advanced steatosis is associated with an increased likelihood of non-specific autoantibody positivity that does not reflect underlying AIH. This observation aligns with previous reports indicating a high prevalence of ANA positivity among patients with NAFLD, in the absence of AIH[7-10,31,32]. Low-grade systemic inflammation is a frequently cited pathogenic mechanism in NAFLD and is characterized by elevated levels of several inflammatory markers[25]. The presence of autoantibodies, produced by B cells, contribute to liver damage and fibrosis progression in NASH, with a potential link between oxidative stress-induced antigens and the production of IgG[33]. This is the first study to quantitatively demonstrate that higher CAP values are significantly associated with non-confirmation of AIH diagnosis in patients with liver dysfunction of unknown etiology and positive autoantibodies.

In multivariate analysis, male gender and higher CAP values were independent predictive factors associated with non-confirmed diagnosis of AIH. We created a hypothesis that could be used in prospective studies and may inform future updates to clinical decision algorithms for the management of patients with positive autoantibodies in suspected AIH, represented in the Supplementary material. Since liver biopsy is invasive and carries potential risks, we suggest an expectant, non-invasive approach in male patients with advanced steatosis (S3), particularly in the absence of progressive or severe liver dysfunction. In such cases, prioritizing the management of metabolic and cardiovascular risk factors alongside close monitoring of liver tests could help avoid unnecessary biopsies while maintaining surveillance. However, it is important to note our findings need external validation, particularly in large multi-center studies in order to support the reliability and applicability of the algorithm. Our study has some limitations, including its retrospective design and relatively small sample size. Nonetheless, it is the first study to identify predictive factors for the non-confirmation of AIH diagnosis in patients with liver dysfunction of unknown etiology and positive autoantibodies.

CONCLUSION

AIH was not confirmed in nearly half of the patients with liver dysfunction and positive autoantibodies. Among the autoantibodies typically associated with AIH, ANA showed the lowest specificity despite its high sensitivity. In this context, male gender and elevated CAP values emerged as independent predictors of a non-confirmed AIH diagnosis. These findings offer important insights for updating clinical decision algorithms, especially given the rising prevalence of MASLD. In male patients with advanced steatosis (CAP values above 280 dB/m²), we advocate for a more conservative approach, focusing on metabolic risk factor control and weight reduction, before proceeding with invasive diagnostics such as liver biopsy.