The Association of Nonalcoholic Steatohepatitis and Hepatocellular Carcinoma

Abstract

Background:

Current guidelines recommend surveillance for hepatocellular carcinoma (HCC) in high risk patients. This high risk is defined by the presence of cirrhosis. However, HCC due to underlying nonalcoholic steatohepatitis (NASH), even without progressing to cirrhosis, is a rising concern. Hence, we aimed to determine the association of HCC with NASH using a large national database.

Methods:

Cross-sectional study was performed using the 2012 National Inpatient Sample (NIS). The study group was all adult patients’ age 18–90 years who have a diagnosis of NASH which was identified using the International Classification of Diseases 9^th version (ICD-9) codes. The control group included the rest of adult individuals without discharge records of NASH. We identified the diagnosis of hepatocellular carcinoma (HCC) in both study and control groups using the ICD-9 codes. We calculated the association between NASH and HCC using univariable and multivariate logistic regression.

Results:

30,712,524 hospitalizations were included in our study. This cohort included 218,950 patients with NASH (study group) and 30,493,574 patients without NASH (control group). The study group patients aged 57.3±0.10 years (59.4% females) comparing to 54.5±0.11 years (57.1% female) in the control group. HCC prevalence in subjects with NASH was 0.50% (95% CI: 0.41, 0.59) compared to 0.21% (0.20, 0.23) in subjects without NASH (p<0.001). After adjusting for age, gender, smoking, alcohol use, obesity, HCV, HBV, hemochromatosis, HIV, cirrhosis and the modified comorbidity index, subjects with NASH were 60% more likely to have HCC than those without NASH (adjusted OR :1.6, 95% CI: 1.4- 1.9 , p<0.001).

Conclusions:

Our study showed that NASH patients are 60% more likely to develop HCC compared to patients without NASH. Close monitoring and even periodical surveillance might be needed.

INTRODUCTION:

Nonalcoholic fatty liver disease (NAFLD) is the most common cause of chronic liver diseases in the western world¹. Nonalcoholic fatty liver (NAFL) refers to the fat accumulation in the liver parenchyma in the absence of alcohol abuse². On the other hand, nonalcoholic steatohepatitis (NASH) is an aggressive subtype of NAFLD which is marked by the presence of lobular inflammation and evidence of liver cell injury³. NAFLD affects approximately 30% of the general population compared to NASH prevalence of 5–7% in the United States (US)⁴. Even though NAFLD is generally a benign diagnosis, NASH imposes a significant impact on advanced liver diseases including fibrosis, cirrhosis and hepatic cell carcinoma (HCC)⁵.

HCC is considered the fifth most common malignancy and the third most deadly cancer in the US⁶. Nonetheless, It is the fastest growing cause of death since 1980s with an annual incidence of 6/100,000 in the US⁷. Multiple infectious and metabolic disorders have been correlated with HCC development. For instance, hepatitis C virus (HCV) and hepatitis B virus (HBV) were listed among the most important risk factors for developing HCC⁸. Furthermore, morbid obesity and diabetes are known metabolic risk factors for developing NASH, they are also reported as independent risk factors for HCC development^9,10. It is noted that HCC incidence has doubled in the last two decades^11,12.

The correlation between NASH and HCC has been controversial in the last decade. Although emerging data correlated the development of HCC with NASH ^13–16, a recent meta-analysis found no statistically significant increased risk for HCC with NASH¹⁷. Since it remains unclear if NASH is an independent risk factor for HCC, we aimed to determine the association of HCC with NASH patients using a large national database.

PATIENTS AND METHODS

Data Source:

This cross sectional study was conducted using the 2012 National Inpatient Sample (NIS). The NIS is the largest all-payer inpatient database in the US and contains a sample of over 7 million inpatient admissions, which represent an approximately 20 percent sample of discharges from all community hospitals participating in the Healthcare Cost and Utilization Project (HCUP). It excludes rehabilitation and long-term acute care hospitals. The data includes primary and secondary diagnoses up to 25 and primary and secondary procedures up to 15. In addition, it contains the patient demographics, discharge status, length of stay, disease severity and comorbidity measures.

The principal diagnosis in these patients is defined, according to HCUPnet, as “the condition established after study to be chiefly responsible for occasioning the admission of the patient to the hospital for care. The principal diagnosis is always the reason for admission”. Hence the diagnosis of NASH and HCC should be reliable in this study.

Study Population, Inclusion and Exclusion Criteria

Data on hospital admissions of all adult patients 18 to 90 years old was extracted from the 2012 NIS. We divided discharges into two groups. The study group was identified using the International Classification of Diseases 9^th version (ICD-9) code for NASH. The rest of patients with no discharge diagnosis of NASH were considered the control group. ICD-9 codes were also used to find all patients with a diagnosis of HCC in both control and study groups. Comorbidities of interest were defined by querying all diagnostic and procedural fields for the corresponding ICD-9 codes. Table 1 shows all ICD-9 diagnostic codes used for patient selection.

Table 1:

ICD-9 Codes Used

Description	ICD-9
NASH	571.8
HCC	155.0-155.2
OSA	327.23
Smoking	305.1, V158.2
Alcohol	303.00-303.93
Cirrhosis	571.2, 571.5, 571.6
HCV	070.44, 070.54, 070.70, 070.71
HBV	070.22, 070.23, 070.32, 070.33
Hemochromatosis	275.03
Obesity	278.00-278.03, 278.1-278.8
Diabetes	250.00-250.93
HTN	401.0-401.9, 405.01-405.99
Dyslipidemia	272.4
Metabolic Syndrome	277.7
HIV	042, V08

Study Variables

The HCUP Comorbidity Software was used to generate Elixhauser comorbidities from ICD-9 CM diagnosis codes and calculate the Elixhauser comorbidity index (ECI)¹⁸. The Deyo modification of the Charlson Comorbidity Index (CCI) was used to define severity of co-morbid conditions¹⁹. This score was modified to exclude HCC. Comorbid conditions were recorded if they were listed among the diagnoses for the hospitalization. All available demographic data, patients’ age, gender, race, Elixhauser’s comorbidities and the rest of comorbidities were extracted (Table 2). Of the included variables, race was the factor with most missing data (5%) as certain states do not document race in the discharge information; we included this as a separate category under said variable. In addition, gender was missing for 0.007% of subjects; these observations were dropped from the final multivariable analysis if said variables were included.

Table 2.

Patients Characteristics

Factor	Overall N=6,142,504 Wgt N = 30,712,524	No NASH N=6,098,714 Wgt N = 30,493,574	NASH N=43,790 Wgt N = 218,950	p-value
Age (years), mean ± se	57.3 ± 0.10	57.3 ± 0.10	54.5 ± 0.11	<0.001a
Age (years), % ± se				<0.001c
18-44	29.3 ± 0.21	29.3 ± 0.21	26.1 ± 0.29
45-64	29.3 ± 0.12	29.2 ± 0.12	47.2 ± 0.25
65+	41.4 ± 0.21	41.5 ± 0.21	26.8 ± 0.31
Gender, % ± se				<0.001c
Male	40.7 ± 0.13	40.6 ± 0.13	42.9 ± 0.29
Female	59.3 ± 0.13	59.4 ± 0.13	57.1 ± 0.29
Race, % ± se				<0.001c
White	65.1 ± 0.55	65.1 ± 0.55	67.7 ± 0.70
Black	13.9 ± 0.32	13.9 ± 0.32	9.1 ± 0.28
Hispanic	9.7 ± 0.31	9.7 ± 0.31	12.9 ± 0.47
Other	6.1 ± 0.24	6.1 ± 0.24	5.9 ± 0.33
Unknown	5.2 ± 0.39	5.2 ± 0.39	4.4 ± 0.49
Obesity, % ± se	11.8 ± 0.09	11.6 ± 0.09	36.5 ± 0.45	<0.001c
Smoking, % ± se	24.7 ± 0.20	24.7 ± 0.20	29.3 ± 0.32	<0.001c
Alcohol use, % ± se	3.2 ± 0.05	3.2 ± 0.05	4.8 ± 0.12	<0.001c
Diabetes, % ± se	24.4 ± 0.11	24.3 ± 0.11	42.6 ± 0.28	<0.001c
Hypertension, % ± se	38.8 ± 0.13	38.7 ± 0.13	49.7 ± 0.29	<0.001c
Dyslipidemia, % ± se	24.0 ± 0.16	24.0 ± 0.16	30.9 ± 0.30	<0.001c
Metabolic syndrome, % ± se	0.18 ± 0.01	0.17 ± 0.01	1.3 ± 0.10	<0.001c
HIV, % ± se	0.70 ± 0.02	0.70 ± 0.02	0.84 ± 0.05	0.002c
Cirrhosis, % ± se	1.9 ± 0.02	1.8 ± 0.02	8.2 ± 0.23	<0.001c
HCC, % ± se	0.22 ± 0.01	0.21 ± 0.01	0.50 ± 0.04	<0.001c
OSA, % ± se	4.9 ± 0.05	4.8 ± 0.05	13.6 ± 0.30	<0.001c
HCV, % ± se	2.0 ± 0.03	1.9 ± 0.03	2.5 ± 0.09	<0.001c
HBV, % ± se	0.11 ± 0.004	0.11 ± 0.004	0.16 ± 0.02	0.002c
Hemochromatosis, % ± se	0.06 ± 0.001	0.06 ± 0.001	0.24 ± 0.02	<0.001c
Num. of Elixhauser’s comorbidities, % ± se				<0.001c
0	19.6 ± 0.16	19.7 ± 0.16	0.38 ± 0.03
1	17.9 ± 0.07	18.0 ± 0.07	6.6 ± 0.15
2	19.2 ± 0.05	19.2 ± 0.05	14.1 ± 0.20
3+	43.3 ± 0.19	43.1 ± 0.19	78.9 ± 0.28
Modified Comorbidity Index1, mean ± se	1.5 ± 0.01	1.5 ± 0.01	1.7 ± 0.01	<0.001a
Modified Comorbidity Index1, % ± se				<0.001c
0	42.7 ± 0.19	42.8 ± 0.19	31.7 ± 0.29
1	21.5 ± 0.06	21.4 ± 0.06	27.0 ± 0.24
2	14.1 ± 0.06	14.1 ± 0.06	16.3 ± 0.19
3+	21.7 ± 0.13	21.6 ± 0.13	24.9 ± 0.33

Data presented as Weighted Frequency (%) unless otherwise stated.

N are the unweighted frequency and Wgt N are the weighted frequencies

^*Data not available for all subjects. Gender = 429.

P-values: a=linear regression; c=Rao-Scott chi-square test.

SAS Survey Procedures used for all analyses.

Statistical Analysis

Data are presented as mean ± standard error for continuous variables or weighted frequency (percent) for categorical factors. A univariable analysis was performed to assess differences between subjects with and without NASH; continuous variables were compared using t-tests and categorical variables were compared using Rao-Scott chi-square tests. In addition, univariable and multivariable logistic regression analysis was performed to assess the association between NASH and HCC; HCC was modeled as the outcome with NASH as the independent variable. The following variables were also included as independent predictors in the multivariable model: age, gender, smoking, alcohol use, obesity, HCV, HBV, hemochromatosis, HIV, cirrhosis and the modified comorbidity index. All analyses were performed using SAS survey procedures (version 9.4, The SAS Institute, Cary, NC), which account for the complex sampling design of NIS and appropriately weight participants in statistical models. A p < 0.05 was considered statistically significant.

Results

We analyzed 30,712,524 hospitalizations from the 2012 NIS database. The study group included 218,950 patients with a history of NASH, and the control group included 30,493,574 with no history of NASH.

Univariable Comparison

As shown in Table 2, the NASH group patients were significantly younger (54.5±0.11years) and were more likely to be female (57.1% females) as compared to the control group patients who were older (57.3±0.10 years) and also more likely to be female (59.4% females), P<0.001. The prevalence of obesity and the metabolic syndrome were also higher in the NASH group compared to the control group (P<0.001); as 36.5% of the patients in the NASH group were obese and 1.3% had a diagnosis of metabolic syndrome compared to 11.6% obese patients and 0.17% patients with metabolic syndrome in the control group. The NASH group had higher prevalence of diabetes and hypertension compared to the control group [42.6% versus 24.3%, P<0.001] and [49.7% versus 38.7%, P<0.001] respectively. Dyslipidemia was present in 30.9% of NASH patients compared to 24.0% of control patients (P<0.001). HCC prevalence in subjects with NASH was 0.50% (95% CI: 0.41, 0.59) compared to 0.21% (0.20, 0.23) in subjects without NASH (p<0.001) (Figure 1). The NASH group had higher prevalence of cirrhosis compared to the control group [8.2% versus 1.8%, P <0.001]. NASH patients had a significant higher prevalence of HCC compared to the control group [OR: 2.3, (95% CI: 2 – 2.7), P<0.001]. Table 1 presents a summary of patient characteristics.

Figure 1.

Prevalence of HCC in adult subjects with and without NASH

Multivariable Comparison

After adjusting for age, gender, smoking, alcohol use, obesity, HCV, HBV, hemochromatosis, HIV, cirrhosis and modified comorbidity index, subjects with NASH were 60% more likely to have HCC than those without NASH [adjusted OR: 1.6, (95% CI: 1.4 – 1.9), P<0.001] (Figure 2).

Figure 2.

Association between NASH and HCC in adult subjects

*Adjusted for age, gender, smoking, alcohol use, obesity, HCV, HBV, hemochromatosis, HIV, cirrhosis and the modified comorbidity index.

DISCUSSION

This large database study proved that NASH patients have significant increased risk of developing HCC. After controlling for the confounding factors which increase the risk of HCC, patients with a history of NASH were found 60% more likely to develop HCC.

It is well recognized that the incidence of HCC in the US has significantly increased in the last three decades^11,12. This increased in incidence of HCC is mirrored by the increased in obesity and diabetes in the US, which both are linked to increased risk of HCC^9,10,20. It is estimated that 3–15% of patients with NAFLD and NASH can progress into cirrhosis and liver failure²¹. Eighty to ninety percent of patients with HCC have underlying liver cirrhosis^7,8. Furthermore, 15% of HCC cases are labeled as cryptogenic in origin. It is highly suggested that NASH is responsible for a substantial ratio of the cryptogenic HCC⁵, that suggestion is consistent with our study results.

Despite the great advances, the exact mechanism of HCC development due to NASH is still not quite clear. Some studies reported that insulin resistance and subsequently hyperinsulinemia which is highly associated with NASH play a major role in HCC development by up-regulating multiple growth factors ^22,23.

Another study linked HCC with cellular DNA damage which is followed by factors promoting malignant cell development²⁴. Adding to that, increased release of fatty acids and adipokines secreted by adipose tissue was also noted in NASH patient, and this affects the release of inflammatory and inhibitory cytokines such as tumor necrosis factor-a, nuclear factor k –B, and interleukin-6^25,26. Furthermore, oxidative stress is thought to promote carcinogenesis in multiple mechanisms including mutations in P53 gene^3,27. Oxidative stress is also thought to suppresses the expression of transcription factor Nrf1, an activator of hepatic antioxidant response factor that controls gene transcription encoding enzymatic antioxidants³. The administrative nature of our study database did not allow us to prove any of these previously reported possible mechanisms. Further studies are needed to fully understand and prevent the possible pathogenetic pathway which links HCC with NASH.

One of the strengths in this study is the large database which represents a large segment of the US population. Our study has some limitations, NIS cannot specify if the diagnosis of NASH was biopsy proven or based on imaging studies. NIS cannot provide clinical information about specific medication use or laboratory results for each individual. In addition, NIS relies on the accuracy of clinical data and validity of medical diagnoses, which might be different between individuals and facilities. In addition, the outpatients’ exclusion and inpatients’ inclusion could lead to more sick individuals in the data, which might affect the generalizability of the results. On the other hand, exclusion of academic hospitals by the database could potentially exclude patients with more complex disease.

In conclusion, this large nationwide database study strongly suggests that NASH is an independent risk factor for HCC development with a 60% higher prevalence of HCC in patients with NASH. Further studies are needed to explore the possible etiologies and determine the roles of screening NASH patients for HCC periodically.