All patients with chronic HBV infection who underwent liver biopsy at The Third Hospital of Zhenjiang Affiliated Jiangsu University from January 2005 and September 2020 were selected. Chronic HBV infection was defined as continuous positive serum hepatitis B surface antigen (HBsAg) or HBV DNA results for more than 6 mo. The inclusion criterion was a follow-up time greater than 6 mo. The exclusion criteria were a history of excessive alcohol consumption (defined as alcohol intake ≥ 20 g/day for men and ≥ 10 g/day for women)[11], history of schistosomiasis of the liver, autoimmune hepatitis, primary biliary cirrhosis, malignancy, immunodeficiency virus infection, viral hepatitis C or D, long-term use of drugs that can cause hepatic steatosis (amiodarone, sodium valproate, tamoxifen, dexamethasone, or methotrexate), and incomplete clinical data. This study was approved by the ethics committee of The Third People’s Hospital Affiliated to Zhenjiang, Jiangsu University. It was registered in the Chinese Clinical Trial Registry (No: chictr2200060304).

The demographic data of patients (sex, age, height, and weight) were collected along with their clinical history (diabetes mellitus, hypertension, drug use history, drinking history); antiviral treatment; blood routine, biochemistry, and serological examination of hepatitis B pathogen. Other data collected were tumor index results during liver biopsy, including total bilirubin, albumin, prealbumin (PB), alanine aminotransferase (ALT), aspartate aminotransferase, γ-glutamyl transferase, alkaline phosphatase (ALP), fasting blood glucose (GLU), total cholesterol (TC), triglycerides (TG), low-density lipoprotein (LDL), high-density lipoprotein, platelets (PLT), HBsAg, HBeAg, HBV DNA level, alpha fetoprotein (AFP), other types of viral hepatitis indicators, human immunodeficiency virus antibody, and autoantibody test results. Body mass index (BMI) was calculated using each patient’s height and weight with the following formula: BMI = weight (kg)/height (m)².

The calculation results were graded according to the Asian standard[12], in which overweight and obesity were defined as BMI ≥ 23 kg/m² and ≥ 25 kg/m², respectively. The detection limit of HBV-DNA was 500 IU/mL.

All liver specimens were evaluated by experienced pathologists and scored according to the nonalcoholic steatohepatitis clinical research network[13] for hepatic steatosis (0-3), lobular inflammation (0-2), portal inflammation (0-3), and ballooning degeneration (0-1). The degree of fibrosis was divided into F0-4 stages according to the METAVIR evaluation system, and the F4 stage was defined as cirrhosis[14]. NAFLD occurs when more than 5% of hepatocytes with steatosis are present in a specimen. The activity score (NAS) of nonalcoholic fatty liver disease was calculated according to the scores of steatosis, lobular inflammation, and ballooning degeneration. NAS ≥ 5 indicates the presence of NASH[13].

The start time of follow-up was the date when the patient underwent liver biopsy. The follow-up endpoint was the date of the last follow-up or the date of the occurrence of clinical outcomes (cirrhosis, HCC, or death). Follow-up of the entire cohort ended in August 2021. The electronic medical record was consulted to obtain the date of the last follow-up, test results (blood routine, HBV pathogen serology, liver function, blood lipids, AFP), and imaging results. The treatment of patients with an interval of more than 6 mo between the last follow-up date and the research deadline (telephone follow-up, regular follow-up) is recommended, and the clinical outcome of those who do not wish to follow up is determined according to the last follow-up record. If there was an out-of-hospital follow-up, the out-of-hospital examination results were obtained, and if there was a death, the time and cause of death were obtained. Those who did not have contact information or could not be contacted were regarded as being lost to follow-up.

The diagnosis of cirrhosis was made by experienced sonographers according to the ultrasound diagnostic criteria of cirrhosis, and computed tomography and magnetic resonance imaging were applied when necessary.

The diagnosis of HCC was based on histological or imaging findings, and the latter was positive lesions detected by at least two imaging techniques (ultrasound, computed tomography, or magnetic resonance imaging), or the use of imaging technology combined with AFP > 400 ng/mL.

The data were analyzed using SPSS version 22.0 (SPSS Inc., Chicago, IL, United States). Continuous variables are expressed as mean ± SD or median (interquartile range), and categorical variables are expressed as percentages. T-test was applied when continuous variables were normally distributed, and the Mann–Whitney U test was used when they were non-normally distributed. Categorical variables were analyzed using the chi-squared test. Kaplan-Meier (K-M) survival analysis was used to evaluate clinical events, and Cox proportional hazards regression was applied for univariate and multivariate analyses. This study also used 1:1 propensity score matching (PSM) to match the NAFLD and non-NAFLD groups, and the caliper value was set to 0.01. All tests were two-tailed, and statistical significance was set at P < 0.05.

A total of 981 patients with chronic HBV infection underwent liver biopsy at The Third People’s Hospital of Zhenjiang between January 2005 and September 2020. After screening based on the inclusion and exclusion criteria, 456 patients were included in the final study, 67 of whom had histologically confirmed cirrhosis at baseline. Figure 1 shows the specific process.

Open in New Tab   Full Size Figure   Download Figure

Figure 1 Flow chart of patient screening and grouping. CHB: Chronic hepatitis B; NAFLD: Nonalcoholic fatty liver disease; HIV: Human immunodeficiency virus; AIH: Autoimmune hepatitis; PBC: Primary biliary cholangitis.

Basic information of the general population: The total number of study patients was 456; of these patients, 152 (33.3%) had histologically confirmed NAFLD. The median follow-up time of the entire cohort was 70.5 mo. The average age of the population was 41 years and 45.1% were female. Regarding BMI, 43% had a normal BMI, 27.9% were overweight (BMI ≥ 23 kg/m², < 25 kg/m²), and 29.2% were obese (BMI > 25 kg/m²). There were 42 patients with diabetes and 38 with hypertension, accounting for 9.2% and 8.3% of all patients, respectively. Most of the patients (72.4%) received antiviral therapy. At baseline, 358 patients were HBV DNA positive, with a median detection value of 3.63 × 10⁴ (10^4.56) IU/mL, while in 98 (21.5%) patients the HBV DNA level was below the detection limit. Among all HBV-infected patients, 66% were HBeAg-negative. Table 1 shows the basic information of the total population.

Comparison of baseline data of chronic HBV infection with and without NAFLD: There were 152 patients with chronic HBV infection complicated by NAFLD and 304 patients without NAFLD. Table 2 shows the demographic and main clinical indicators of the two groups. The proportion of female patients in the NAFLD group was higher than that in the NAFLD group (P < 0.05), and the median follow-up time was longer than that in the NAFLD group (73 vs 63 mo, P < 0.05). Compared with the non-NAFLD group, the NAFLD group had a higher prevalence of diabetes and higher BMI (P < 0.001), and its LDL, TG, PB, ALT levels were also significantly higher (P < 0.05). However, there were no differences in age, prevalence of hypertension, proportion of liver cirrhosis, HBV DNA, and other indicators (P > 0.05).

Comparison of clinical characteristics of patients with chronic HBV infection without cirrhosis at baseline with NAFLD and those without NAFLD: Taking patients without cirrhosis at the time of liver biopsy as the research object, the demographic and main clinical indicators of the NAFLD and non-NAFLD groups were compared (Table 3). There were significant differences in sex, BMI, prevalence of diabetes, follow-up duration, HBV DNA, ALT, TC, TG, LDL, and other indicators between the two groups (all P < 0.05). However, there was no difference in the antiviral status and other indicators (P > 0.05). The prevalence of diabetes, follow-up duration, and ALT levels in the two groups were 1:1 PSM. There were 109 patients in the NAFLD and non-NAFLD groups. After PSM, there were no significant differences in sex, diabetes prevalence, follow-up duration, HBV DNA, ALT, and other indicators between the two groups (P > 0.05); however, there were differences in BMI, PA, and TG (P < 0.05). Moreover, the NAFLD group was divided into the NASH (21 cases) and non-NASH (88 cases) groups.

Comparison of the pathological characteristics of chronic HBV infection with and without NAFLD: Among patients with NAFLD, 58.6% had mild hepatic steatosis, 31.6% had moderate hepatic steatosis, and 9.9% had severe hepatic steatosis. Differences were observed between the NAFLD group and non-NAFLD group in the degree of liver fibrosis, portal inflammation, and NAS score (P < 0.05). There was no difference in lobular inflammation or ballooning deformation (BD) between the two groups (P > 0.05) (Table 4).

There was no significant difference in the proportion of patients receiving antiviral treatment, types of antiviral drugs, and duration of antiviral treatment between the NAFLD group and non-NAFLD group (P > 0.05). At the last follow-up, 302 (91.2%) patients who received antiviral therapy were HBV DNA-negative, including 102 (78.9%) in the NAFLD group and 200 (93%) in the non-NAFLD group. There was no significant difference in the proportion of HBV DNA negativity between the two groups (P > 0.05). A total of 275 (83.1%) patients had normal ALT levels: 90 (77.6%) in the NAFLD group and 185 (56%) in the NAFLD group. There was no significant difference in the proportion of normal ALT levels between the two groups (P = 0.05) (Table 5).

Occurrence of cirrhosis: Patients without cirrhosis at the time of liver biopsy were selected as research participants, and the risk of progression to cirrhosis was observed. During the follow-up period, 34 patients developed liver cirrhosis diagnosed by ultrasound, with a median follow-up time of 72 (30-134) mo, including 10 (7.8%) in the NAFLD group and 24 (9.2%) in the NAFLD group. This study was conducted during the follow-up period.

K-M survival analysis of NAFLD and the risk of cirrhosis: The results of the K-M survival analysis showed that there was no significant increase in the risk of liver cirrhosis diagnosed using ultrasound in the combined NAFLD group before and after PSM (log-rank, P = 0.69). The results of the K-M survival analysis after PSM are shown in Figure 2A. F0, 1, 2 was regarded as mild fibrosis, while F3, 4 was regarded as severe fibrosis. Patients with fibrosis after PSM had an increased risk of cirrhosis diagnosed using ultrasound (log rank, P < 0.05). The results of K-M survival analysis after matching are shown in Figure 2B. Figure before PSM. In the NAFLD group, three out of 21 cases in the NASH group and seven out of 88 patients in the non-NASH group developed cirrhosis. The results of K-M survival analysis after PSM are shown in Figure 2C. There was no statistical difference between the two groups (P = 0.17).

Open in New Tab   Full Size Figure   Download Figure

Figure 2 Kaplan-Meier survival analysis of nonalcoholic fatty liver disease and the risk of cirrhosis. A: Results of the Kaplan-Meier (K-M) survival analysis of the risk of cirrhosis after propensity score matching (PSM) in nonalcoholic fatty liver disease (NAFLD) and non NAFLD with chronic hepatitis B; B: Results of K-M survival analysis of the risk of cirrhosis after PSM in patients with mild fibrosis and in those with severe fibrosis; C: Nonalcoholic steatohepatitis (NASH) and non-NASH stratification of NAFLD and K-M survival analysis of the risk of cirrhosis. NAFLD: Nonalcoholic fatty liver disease; NASH: Nonalcoholic steatohepatitis; PSM: Propensity score matching.

Univariate and multivariate Cox regression analysis of cirrhosis: Cox regression univariate analysis showed that age, antiviral duration, lobular inflammation, BD, liver fibrosis, ALP, LDL, and PLT were related to cirrhosis (P < 0.05). Multivariate analysis showed that age, ballooning degeneration, liver fibrosis, and PLT were independent risk factors for cirrhosis diagnosed using ultrasound (all P < 0.05) (Table 6).

Univariate and multivariate Cox regression analysis of cirrhosis after PSM: Cox regression univariate analysis showed that age, hypertension, diabetes mellitus, lobular inflammation, portal inflammation, liver fibrosis, ALT, and PLT were related to cirrhosis diagnosed using ultrasound (P < 0.1). Multivariate analysis showed that diabetes mellitus, BD, and PLT were independent risk factors for liver cirrhosis diagnosed using ultrasound (P < 0.05) (Table 7).

Characteristics of patients with HCC: During the follow-up period, 10 patients (2.2%) developed HCC (Table 7), and two died (one died due to HCC and the cause of death in the other case was not related to liver disease). The median time interval between liver biopsy and HCC diagnosis was 100.5 mo (68.5-128.0). HCC occurred in six patients (3.3%) with NAFLD and four patients (1.6%) without NAFLD. Seven patients developed cirrhosis at the time of liver biopsy, and one of the other three developed cirrhosis before HCC.

K-M survival analysis of HCC risk: Patients with baseline cirrhosis were included in the K-M survival analysis of HCC risk. It was found that the cumulative risk of HCC in the NAFLD group was significantly higher than that in the non-NAFLD group (log rank, P = 0.02) (Figure 3A). At the same time, the risk of HCC in patients with severe liver fibrosis (F3-4) was also significantly increased (log rank, P = 0.005) (Figure 3B). When the NAFLD group was divided into the NASH group with 32 patients (3 HCCs) and the non-NASH group with 120 patients (3 HCCs), the risk of HCC in the NASH group was increased (log rank, P = 0.03) (Figure 3C), and the risk of HCC in patients with hepatic ballooning was significantly increased (log rank, P = 0.01) (Figure 3D). There was no significant difference in the risk of HCC among patients with steatosis, lobulitis, and portal inflammation (log rank, P > 0.05) (figures ignored).

Open in New Tab   Full Size Figure   Download Figure

Figure 3 Kaplan-Meier survival analysis of hepatocellular carcinoma risk. A: Results of the Kaplan-Meier (K-M) survival analysis of the risk of hepatocellular carcinoma (HCC) in patients with chronic hepatitis B and nonalcoholic fatty liver disease (NAFLD) and those without NAFLD; B: Results of K-M survival analysis of the risk of HCC in patients with mild fibrosis and those with severe fibrosis; C: Nonalcoholic steatohepatitis (NASH) and non-NASH stratification of NAFLD and K-M survival analysis of the risk of HCC; D: Ballooning and non-ballooning stratification of NAFLD and K-M survival analysis of the risk of HCC. NAFLD: Nonalcoholic fatty liver disease; NASH: Nonalcoholic steatohepatitis; HCC: Hepatocellular carcinoma.

Univariate and multivariate Cox regression analysis of HCC: Cox regression univariate analysis showed that diabetes mellitus, NAFLD, NASH, lobular inflammation, BD, liver fibrosis, GLU, TC, TG, and PLT were correlated with HCC (all P < 0.05). Multivariate analysis of factors with P < 0.10 in the Cox regression univariate analysis showed that ballooning, liver fibrosis, and diabetes were independent risk factors for HCC (Table 8).

Chronic hepatitis B (CHB) virus infection and nonalcoholic fatty liver disease (NAFLD) are important causes of liver-related complications and death. With the increasing prevalence of NAFLD, the number of patients with combined NAFLD and hepatitis B virus (HBV) infection is also on the increase.

This study aimed to further explore the impact of NAFLD and the pathological changes confirmed by liver pathology in patients with chronic HBV infection.

To study the effect of NAFLD confirmed using liver pathology on the outcomes of long-term serious adverse events in patients with CHB virus infection.

Among 456 cases of chronic HBV infection, 152 were confirmed by liver histology to have NAFLD, and 304 were simple chronic HBV infection. The incidence of serious clinical events at the follow-up endpoint was compared by Kaplan-Meier (K-M) survival analysis at baseline using propensity score matching balance parameters.

After a median follow-up of 70.5 mo, there were 34 cases of ultrasound-diagnosed cirrhosis and 10 cases of HCC. K-M survival analysis showed no significant difference in the occurrence of CHB complicated with NAFLD cirrhosis, and the cumulative incidence of HCC in the NAFLD group was higher than that in the non-NAFLD group (log-rank test, P < 0.05). Hepatocyte ballooning and severe liver fibrosis were also associated with an increased risk of HCC (log-rank test, all P < 0.05).

Baseline hepatocyte ballooning is a risk factor for adverse events in patients with CHB complicated with NAFLD.

Larger samples and more rigorous prospective studies are needed for further verification. The study cohort needs to be observed for a longer time, and the conclusion may change if this is done, in the study.