Potential role of diabetes mellitus in the progression of cirrhosis to hepatocellular carcinoma: a cross-sectional case-control study from Chinese patients with HBV infection

Beijing, China

Potential role of diabetes mellitus in the progression of cirrhosis to hepatocellular carcinoma: a cross-sectional case-control study from Chinese patients with HBV infection

Chun Gao, Long Fang, Hong-Chuan Zhao, Jing-Tao Li and Shu-Kun Yao

Beijing, China

BACKGROUND:Diabetes mellitus (DM) is regarded as a new risk factor for hepatocellular carcinoma (HCC), but few studies have focused on the potential role of DM in the progression of cirrhosis to HCC as well as in patients with simple HBV infection.

METHODS:A cohort of 1028 patients, treated at our hospital and with a hospital discharge diagnosis of HCC and/or cirrhosis, was screened. Among them, 558 were diagnosed with chronic HBV infection and 370 were analyzed statistically according to the diagnostic, inclusion and exclusion criteria. The demographic, clinical, metabolic, virological, biochemical, radiological and pathological features were analyzed and the multivariate logistic regression model was used to determine the potential role of DM.

RESULTS:In 248 cirrhotic patients, 76 were diabetic and their mean duration of DM was 4.6 years. In 122 HCC patients with cirrhosis, 25 were diabetic and their mean duration of DM was 4.4 years. Univariate analysis showed that compared with cirrhotic patients, the HCC patients had a higher percentage in males (P=0.001), a lower percentage in DM patients (P=0.039), a higher percentage in cigarette smokers (P=0.005), a higher percentage in patients with AFP＞400 ng/mL (P＜0.001), higher values of white blood cells (P＜0.001), hemoglobin (P＜0.001) and platelet (P＜0.001), increased levels of ALT (P＜0.001) and GGT (P＜0.001), higher total bilirubin (P=0.018) and albumin levels (P＜0.001), and a lower international normalized ratio (P＜0.001). Multivariate logistic regression analysis showed that DM was anindependent associated factor for HCC [odds ratio (OR)=0.376; 95% CI, 0.175-0.807;P=0.012]. Even after the HCC patients were restricted to those with decompensated cirrhosis and compared with decompensated cirrhotic patients, the similar result was observed (OR=0.192; 95% CI, 0.054-0.679;P=0.010).CONCLUSIONS:DM is an independent factor in the progression of cirrhosis to HCC, but the role may be contrary to our current viewpoint. To clarify the causal relationship of DM and HCC, prospective and experimental studies are required.

(Hepatobiliary Pancreat Dis Int 2013;12:385-393)

diabetes mellitus;hepatocellular carcinoma; cirrhosis; chronic hepatitis B

Introduction

Hepatocellular carcinoma (HCC) is a common malignancy worldwide, and the incidence rate is increasing over the past two decades in China as well as in the United States, Japan and other countries.[1-3]Hepatitis virus, especially HCV, or alcoholic liver disease, has been confirmed to play important roles in more than 50% of this increase;[2,4]however, the reason has not yet been explained clearly. In 15%-50% of HCC patients no specific risk factor or reason has been found,[4-6]although some risk factors have been identified, including HBV, HCV, liver cirrhosis, heavy alcohol consumption, increasing age, male gender, non-alcoholic steatohepatitis (NASH), alfatoxin exposure, and positive family history.

Diabetes mellitus (DM) has been regarded as a potentially new risk factor for HCC[4,5,7-10]although a few earlier epidemiologic studies found no relationship between HCC and DM.[11,12]In the Swedish cohort study,[9]patients with DM were at increased risk of developing primary liver cancer, which was supported by the two subsequent cohort studies conducted in Taiwan province of China and the USA.[5,10]

The impact of DM on the development of HCC in HCV-or alcohol-related cirrhosis have been conducted.[13-16]They concluded DM is an independent risk factor for HCC in cirrhosis,[14,16]while no association was demonstrated in another two studies.[13,15]No consensus has been reached about the role of DM in these studies. In addition, the four studies were conducted in the Netherlands, Italy and Japan, which could not been regarded as the representatives of the general population. Moreover, no information was available in patients with HBV-related cirrhosis. Our study was designed to determine the potential role of DM in the progression of cirrhosis to HCC in Chinese patients with simple HBV infection.

Methods

Study population

A cohort of 1028 patients (including 482 HCC and 546 cirrhotic patients) who were treated at our hospital from January 2003 to June 2009, and with a hospital discharge diagnosis of HCC and/or cirrhosis, were screened. Chronic HBV infection was defined as serum HBsAgpositive for at least six months or at diagnosis of HCC and/or cirrhosis. Patients who followed these criteria would be excluded: 1) those who had been treated by any method at inclusion or with confirmed diagnosis of HCC for more than 15 days; 2) those who had a confirmed HCV, HDV or HIV infection; 3)those who had heavy alcohol consumption ＞80 g/d in male or ＞40 g/d in female for more than 10 years; 4) those who had a confirmed diagnosis of drug- or poison-induced liver damage, including a confirmed exposure toaspergillus flavus; 5) those who had a presence of other malignancies, including leukemia and lymphoma; 6) those who were non-Chinese; and 7) those who had a presence of autoimmune hepatitis, schistosomiasis, primary biliary cirrhosis, Budd-Chiari syndrome, primary sclerosing cholangitis, hemachromatosis, Wilson's disease, rheumatic diseases or allergic disorder. The study was approved by the Human Research Ethics Committee of the hospital and it was in accordance with the principles of theDeclaration of Helsinki.

Subject determinations

Liver cirrhosis was histologically diagnosed by needle biopsy or surgical specimens, or based on typical radiological features shown by at least two image examinations including ultrasound (US), contrastenhanced dynamic computed tomography (CT) and magnetic resonance imaging (MRI), or by a single image technique associated with typical manifestations of decompensated liver function and portal hypertension.[6]HCC diagnosis was based on the histological findings of needle biopsy/surgery, or typical radiological features shown by at least two image examinations including US, CT, MRI and hepatic angiography or by a single positive imaging with a serum AFP level ＞400 ng/mL.[17]DM was characterized by fasting plasma glucose of 126 mg/dL or greater on at least two occasions, plasma glucose of 200 mg/dL or greater at 2-hour oral glucose tolerance test, or the need for oral hypoglycemic drug or insulin to control glucose.

Clinical and laboratory parameters

The demographic, metabolic, virological, biochemical, radiological and pathological features of the patients with HCC were recorded. The data were obtained at the diagnosis of HCC, but excluded those obtained on 15 days before or after the diagnosis. For the cirrhotic patients in our hospital, the first detected or measured value was regarded as the recorded value. Patients who have missing values which may affect statistical results would be excluded from the final analysis. According to the Asian and Chinese criteria, overweight was defined as BMI≥23 kg/m2and obesity BMI≥25 kg/m2. The diagnosis of hypertension was defined as systolic blood pressure (SBP) ≥140 mmHg and/or diastolic blood pressure (DBP) ≥90 mmHg, and mean artery pressure (MAP) was computed as 1/3 SBP plus 2/3 DBP. Total bilirubin, serum albumin, international normalized ratio, ascites and hepatic encephalopathy were used to calculate the Child-Turcotte-Pugh (CTP) score.

The findings of physical examinations and image techniques including US, CT, MRI and hepatic angiography were re-assessed carefully by at least two authors independently for clinical classification, clinical stage and TNM stage of HCC. We classified tumor stages according to the criteria recommended by the Anticancer Committee of China and the International Union Against Cancer. In clinical classification, massive HCC was defined with a diameter of ≥5 cm, nodular HCC with a diameter of ＜5 cm, and small HCC with a diameter of ＜3 cm for single or two nodules.

Statistical analysis

Statistical analysis was performed using SPSS version 14.0 for Windows (SPSS, Chicago, IL., USA). Chi-square test, Student'sttest and Mann-WhitneyUtest were used to compare the differences between the HCC andthe cirrhotic groups. The association of HCC with any factor, especially DM, was analyzed using multiple factor unconditional logistic regression. According to the results of univariate analysis, together with our current knowledge and the basic principles of logistic analysis, some variables with statistical differences in univariate analysis were included in the multivariate analysis. For example, serum albumin would be excluded because of the independence of variables if CTP score had been included. Stepwise multiple regression analysis (backward: Wald; entry: 0.05, removal: 0.10) was used. We expressed results as odds ratios (ORs) and 95% confidence intervals (CIs).P＜0.05 was considered statistically significant, and allPvalues quoted were two-sided.

Results

Study population and baseline characteristics

A total of 1028 patients with a hospital discharge diagnosis of HCC and/or cirrhosis were admitted in our hospital, including 482 patients with HCC and 546 with cirrhosis. Among them, 558 were diagnosed with chronic HBV infection and 370 were included in the statistical analysis. In the 482 HCC patients, 360 were excluded (Fig.) because of HCC without cirrhosis (102), HCV infection (38), heavy alcohol consumption (15), cryptogenic conditions (non-HBV and non-HCV) (120), confirmed diagnosis of HCC for more than 15 days or being treated at inclusion (57), non-Chinese (6), confirmed drug- or poison-induced liver damage (3), lack of some data (15), and other reasons (4). Table 1 shows the demographic, clinical, laboratory, metabolic and instrumental features of the left 122 HCC patients with HBV infection and cirrhosis. Of these patients, 25 (20.5%) were diabetic, 67 (54.9%) had decompensated cirrhosis, and 103 (84.4%) had a history of HBV infection. Their mean age was 54.7±10.0 years and 106 patients (86.9%) were male. Fifty-four patients (60.7%, 54/89) were overweight or obese, and 33 (27.0%) hadhypertension. The clinical classification, clinical stage and TNM stage of the patients are shown in Table 2.

Fig.Study population and patient selection. *: Sixteen patients (foreigners): 3 with HBV infection, 4 with HCV infection, 2 with HBV/ HCV infection, and no specific reasons (cryptogenic) for the remaining 7. #: Fifteen patients with presence of malignancies: 6 with gastric cancer (3 with HCV infection, 1 with heavy alcohol intake and 2 with unknown reasons); 2 with esophageal cancer (1 with heavy alcohol intake and 1 with unknown reason); 2 with renal carcinoma (unknown reasons); 1 with colonic cancer (unknown reason); 1 with duodenal cancer (unknown reason); 1 with pancreatic cancer (unknown reason); 1 with lung cancer (unknown reason); and 1 with acute leukemia (HCV infection). **: Reasons included: 2 patients with serious chronic renal failure and 2 with severe chronic heart failure. ▲: Two patients had heavy alcohol intake among the 34 patients with HCV infection. ★: Six foreigners: 3 with HCV infection, 2 with HBV infection, and 1 with no specific reason (cryptogenic). ▼: Reasons included: Bechterew's spondylitis, autoimmune hepatitis and primary biliary cirrhosis for the three patients.

Table 1.Baseline characteristics of the study population and results of univariate analysis for potential role of DM in HCC

Data were available ina268 (group A1, 89; group A2, 49; group B1, 179; group B2, 99),b343 (A1, 109; A2, 63; B1, 234; B2, 137),c156 (A1, 41; A2, 27; B1, 115; B2, 68),d317 (A1, 116; A2, 63; B1, 201; B2, 130),e277 (A1, 87; A2, 51; B1, 190; B2, 113) andf245 (A1, 79; A2, 46; B1, 166; B2, 101) patients. The numbers before the brackets indicate the total available cases in the two groups. *: Plus-minus value indicates mean±SD; ?: Median (interquartile range, Q1-Q3); #: MELD=3.78×loge(bilirubin in mg/dL)+11.2×loge(INR)+9.57×loge(creatinine in mg/dL)+6.43.

In the 546 cirrhotic patients, 298 were excluded (Fig.) because of HCV infection (54), heavy alcohol consumption (25), cryptogenic conditions (non-HBV and non-HCV) (138), non-Chinese (16), confirmed drugor poison-induced liver damage (7), presence of other malignancies (15), autoimmune hepatitis (7), primary biliary cirrhosis (9), lack of some data which influenced the statistical analysis (8), and other reasons (19). Table 1 shows the baseline characteristics of 248 cirrhotic patients. Of these patients, 76 (30.6%) were diabetic, 147 (59.3%) had decompensated cirrhosis, and 202 (81.5%) had a history of HBV infection. Their mean age was 52.8±11.3 years and 178 patients (71.8%) were male. One hundred and two patients (57.0%, 102/179) were overweight or obese, and 58 patients (23.4%) had hypertension.

Role of DM in HCC in the progression of cirrhosis to HCC

Of the 122 HCC patients with cirrhosis, 25 were diabetic and their mean duration of DM was 4.4±4.3 years. The mean duration of cirrhosis was 2.2±4.2 years and 43 patients had been diagnosed for more than oneyear before the diagnosis of HCC. Of the 248 cirrhotic patients, 76 were diabetic and their mean duration of DM was 4.6±6.2 years. The mean duration of cirrhosis was 2.4±4.1 years and 107 patients (43.1%) had been diagnosed for one year before entry. Univariate analysis (Table 1) showed that there were more diabetics in the cirrhotic group than in the HCC group (30.6% vs 20.5%,P=0.039), but no significant differences were observed in the duration of DM and cirrhosis.

Table 2.Clinical classification, clinical stage and TNM stage of HCC patients

Univariate analysis (Table 1) showed that compared with the cirrhotic patients, the HCC patients had a higher percentage of males (P=0.001), a lower percentage of DM (P=0.039), a higher mean artery pressure level (P=0.007), a higher percentage of smokers (P=0.005), a lower mean blood glucose/GLU level (P=0.045), a higher percentage of patients with AFP＞400 ng/mL (P＜0.001), higher values of white blood cells (P＜0.001), hemoglobin (P＜0.001) and platelets (P＜0.001), increased levels of ALT (P＜0.001) and GGT (P＜0.001), higher levels of total bilirubin (P=0.018) and albumin (P＜0.001), and a lower international normalized ratio (INR) (P＜0.001).

Based on the results of univariate analysis, sixteen variables were included in the multivariate logistic regression, including male sex, DM, MAP, smoking, white blood cells, AFP, hemoglobin, platelet, ALT, GGT, total bilirubin, albumin, INR, serum sodium, portal vein diameter, and presence of ascites (Table 3). The results of multivariate analysis showed that DM was an independent factor for HCC and the value of OR was below the cut-off point of 1 (OR=0.376; 95% CI, 0.175-0.807;P=0.012). In addition, eight other variables were significantly different (Table 3).

Similar results from HCC patients who were restricted to those with decompensated cirrhosis and compared with decompensated cirrhotics

Because of the lack of pathologic diagnosis in most patients and selection bias, our study population was restricted to HCC patients with decompensated cirrhosis and compared with decompensated cirrhotics. In these patients, biopsy for diagnosis of cirrhosis was unnecessary and the selection bias should not be further considered. Thus 214 patients were analyzed, including 67 HCC patients and 147 cirrhotics. Among them, 58 patients were diagnosed with DM: 13 in the HCC group and 45 in the cirrhotic group (19.4% vs 30.6%,P=0.087). Univariate analysis (Table 1) revealed that compared with decompensated cirrhotic patients, the HCC patients had a higher percentage of males, a higher percentage of smokers and alcohol intakers, a higher percentage of patients with AFP＞400 ng/mL, a lower blood glucose level, elevated white blood cells, hemoglobin and platelets, increased levels of ALT and GGT, elevated levels of albumin and total bilirubin, lowered INR, elevated total cholesterol, lowered sodium, wider portal vein diameter, and a higher percentage of patients with ascites. According to the results of univariate analysis, fifteen variables were included in the multivariate logistic regression (Table 3). The similar results were shown by multivariate analysis (DM, OR=0.192; 95% CI, 0.054-0.679;P=0.010).

Table 3.Multivariate analysis for potential role of DM in HCC

Association between diabetes duration/treatment and HCC risk

Considering the potentially different effects of antidiabetic agents in the process of hepatocarcinogenesis, we studied the association between DM duration/ treatment and HCC risk. To ensure that diabetes was not induced by HCC, this association was restricted to those who were diagnosed with DM more than one year before HCC diagnosis or before control recruitment, including 16 patients in the HCC group and 43 in the cirrhotic group (Table 4). Among the 59 diabetic patients, 29 received oral anti-diabetic regimens (9/16=56.3% vs 20/43=46.5%,P=0.506), 25 received insulin treatment (4/16=25.0% vs 21/43=48.8%,P=0.100), and 11 were dependent on diet alone to control serum glucose level (5/16=31.3% vs 6/43=14.0%,P=0.254). No significantdifference was observed between DM duration/treatment and HCC risk, even after adjustment for age, gender, smoking and alcohol drinking after unconditional multivariable logistic regression analysis (Table 4).

Table 4.Association between diabetes duration/treatment and HCC risk

Comparison of the role of DM with AFP

Although the role of AFP was questioned for earlystage or small HCC by recent studies, AFP has been used as one of the accepted diagnostic criteria in clinical practice. We compared the role of DM with AFP in the progression of cirrhosis to HCC. When the HCC patients with cirrhosis were compared with cirrhotics, contrary to diabetes (Table 3), a completely opposite role of AFP was observed (OR=22.382; 95% CI, 8.400-59.638;P＜0.001). Even after the study population was restricted to those HCC patients with decompensated cirrhosis and compared with decompensated cirrhotics, a similar result was demonstrated for the opposite role of diabetes and AFP (OR=28.934; 95% CI, 6.970-120.109;P＜0.001).

Discussion

DM was an independent factor for HCC in the progression of cirrhosis to HCC in our patients with HBV infection. The OR of 0.376 was below the cutoff value of 1, indicating that DM may be a potentially protective factor for HCC in this process. Even after the population was restricted to those HCC patients with decompensated cirrhosis and compared with decompensated cirrhotic patients, a similar result was observed by multivariate analysis while considering the lack of pathological diagnosis in most patients and selection bias. In addition, an opposite role of AFP was observed in the aforementioned analyses, which may provide supporting evidence for our results and conclusion.

For the first time, we found that DM may be a potentially protective factor for HCC in the progression of cirrhosis to HCC, at least in our patients with HBV infection. Thus conclusion could be anticipated because many studies have shown that the percentage of diabetics in cirrhosis patients was higher than that in HCC patients.[4-8,10,17]This phenomenon was contrary to the deduction that the risk of HCC would be increased among patients with both DM and HBV-related cirrhosis if diabetes was a risk factor in this progression. To clarify this question, more prospective cohort studies could be performed and the conclusion could be validated in more patients, more centers and more countries.

The major limitation of the present study is the observational cross-sectional case-control design.In our study, selection bias was the major concern. However, after the study population was restricted to those with decompensated liver function, similar result was observed. For these patients, biopsy for diagnosis of cirrhosis was completely unnecessary. Another limitation is that most of the HCC and cirrhotic patients in our study were diagnosed clinically rather than by biopsy, and the diagnosis of most diabetics was dependent on their self-reported history or fasting serum glucose, not on oral glucose tolerance test.[18]However, we followed strictly the diagnostic criteria recommended by the authorized institutes and used them widely in clinical practice. We believe that our results are more likely applicable in clinical practice.

Our conclusion may be contrary to the generally recognized viewpoint that DM is a new risk factor for HCC.[2,4,7,8,19,20]But negative conclusion has been drawn from some studies,[21-25]one of which showed that neither DM nor overweight was a risk factor for HCC.[25]Three studies[13,15,16]were conducted in patients with HCV infection, and concluded that DM increased the risk of developing HCC, but no information was available on HBV infection. Since cirrhotic patients with HCV infection are more likely to suffer from type-2 DM than those with HBV infection,[26]our study was designed to determine the effect of DM on the development of HCC in HBV-related cirrhosis patients.

Our conclusion was also supported by some studies,[27-29]one of which determined the possible effect of metformin on esophageal, gastric, colorectal cancers as well as HCC and pancreatic cancer.[27]It was a prospective cohort study of 800 000 individuals. The results of the study showed that: 1) In patients with diabetes but no anti-hyperglycemic medication, the HCC incidence increased at least two times; 2) In patients on metformin, the HCC incidence decreased to near non-diabetic level; and 3) Adjustment for some covariates made the benefit of metformin more evident (hazard ratio 0.06; 95% CI, 0.02-0.16).[27]The authors of the study concluded that metformin can reduce the HCC incidence in treated diabetics. Similar results were obtained from studies on pancreatic cancer and colorectal cancer, showing that metformin use was associated with reduced risk, and insulin or insulin secretagogue was associated with increased risk in diabetic patients.[28-30]

Reports[27-29]showed the protective role of DM in the progression of cirrhosis to HCC was related to different treatment strategies with oral agents, human insulin and insulin analogues, which was not supported by our results due to the relatively limited number of diabetic patients subjected to biguanide (metformin) treatment. Moreover, the protective role of DM may be related to the interaction of hepatitis virus, cirrhosis and diabetes.[31]DM is a risk factor for HCC in hepatitis patients, but it is not a risk factor for HCC in cirrhotic patients. Exogenous insulin or sulphonylurea treatment was associated with an increased incidence of HCC in patients with HCV infection, but not in those with HCV-related cirrhosis.[31]Apart from the above limitations, some possible factors could not be adjusted, for example non-alcoholic fatty liver disease,[32,33]including non-alcoholic steatohepatitis because of the design of the retrospective cross-sectional case-control study.

In conclusion, DM is an independent factor for HCC in the progression of cirrhosis to HCC. To clarify the causal relationship between DM and HCC, prospective and experimental studies are required.

Contributors:GC conceived the study and wrote the first draft. GC, FL and LJT collected and analyzed the data. ZHC and YSK made critical revision of manuscript. All authors contributed to the design and interpretation of the study and to further drafts. GC is the guarantor.

Funding:The study was supported by grants from the National Natural Science Foundation of China (No. 30772859) and the Research Fund of the China-Japan Friendship Hospital, Ministry of Health (No. 2010-QN-01).

Ethical approval:The study was approved by the Human Research Ethics Committee of the hospital.

Competing interest:No benefits in any form have been received or will be received from a commercial party related directly or indirectly to the subject of this article.

1 Yuen MF, Hou JL, Chutaputti A; Asia Pacific Working Party on Prevention of Hepatocellular Carcinoma. Hepatocellular carcinoma in the Asia pacific region. J Gastroenterol Hepatol 2009;24:346-353.

2 Gao C, Yao SK. Diabetes mellitus: a "true" independent risk factor for hepatocellular carcinoma? Hepatobiliary Pancreat Dis Int 2009;8:465-473.

3 El-Serag HB, Davila JA, Petersen NJ, McGlynn KA. The continuing increase in the incidence of hepatocellular carcinoma in the United States: an update. Ann Intern Med 2003;139:817-823.

4 Davila JA, Morgan RO, Shaib Y, McGlynn KA, El-Serag HB. Diabetes increases the risk of hepatocellular carcinoma in the United States: a population based case control study. Gut 2005;54:533-539.

5 El-Serag HB, Tran T, Everhart JE. Diabetes increases the risk of chronic liver disease and hepatocellular carcinoma. Gastroenterology 2004;126:460-468.

6 Schuppan D, Afdhal NH. Liver cirrhosis. Lancet 2008;371: 838-851.

7 Hassan MM, Hwang LY, Hatten CJ, Swaim M, Li D, Abbruzzese JL, et al. Risk factors for hepatocellular carcinoma: synergism of alcohol with viral hepatitis anddiabetes mellitus. Hepatology 2002;36:1206-1213.

8 Yuan JM, Govindarajan S, Arakawa K, Yu MC. Synergism of alcohol, diabetes, and viral hepatitis on the risk of hepatocellular carcinoma in blacks and whites in the U.S. Cancer 2004;101:1009-1017.

9 Adami HO, Chow WH, Nyrén O, Berne C, Linet MS, Ekbom A, et al. Excess risk of primary liver cancer in patients with diabetes mellitus. J Natl Cancer Inst 1996;88:1472-1477.

10 Lai MS, Hsieh MS, Chiu YH, Chen TH. Type 2 diabetes and hepatocellular carcinoma: A cohort study in high prevalence area of hepatitis virus infection. Hepatology 2006;43:1295-1302. 11 Kessler II. Cancer mortality among diabetics. J Natl Cancer Inst 1970;44:673-686.

12 Lu SN, Lin TM, Chen CJ, Chen JS, Liaw YF, Chang WY, et al. A case-control study of primary hepatocellular carcinoma in Taiwan. Cancer 1988;62:2051-2055.

13 Cimino L, Oriani G, D'Arienzo A, Manguso F, Loguercio C, Ascione A, et al. Interactions between metabolic disorders (diabetes, gallstones, and dyslipidaemia) and the progression of chronic hepatitis C virus infection to cirrhosis and hepatocellular carcinoma. A cross-sectional multicentre survey. Dig Liver Dis 2001;33:240-246.

14 Torisu Y, Ikeda K, Kobayashi M, Hosaka T, Sezaki H, Akuta N, et al. Diabetes mellitus increases the risk of hepatocarcinogenesis in patients with alcoholic cirrhosis: A preliminary report. Hepatol Res 2007;37:517-523.

15 Di Costanzo GG, De Luca M, Tritto G, Lampasi F, Addario L, Lanza AG, et al. Effect of alcohol, cigarette smoking, and diabetes on occurrence of hepatocellular carcinoma in patients with transfusion-acquired hepatitis C virus infection who develop cirrhosis. Eur J Gastroenterol Hepatol 2008;20:674-679.

16 Veldt BJ, Chen W, Heathcote EJ, Wedemeyer H, Reichen J, Hofmann WP, et al. Increased risk of hepatocellular carcinoma among patients with hepatitis C cirrhosis and diabetes mellitus. Hepatology 2008;47:1856-1862.

17 Huo TI, Wu JC, Lui WY, Huang YH, Lee PC, Chiang JH, et al. Differential mechanism and prognostic impact of diabetes mellitus on patients with hepatocellular carcinoma undergoing surgical and nonsurgical treatment. Am J Gastroenterol 2004;99:1479-1487.

18 Yang W, Lu J, Weng J, Jia W, Ji L, Xiao J, et al. Prevalence of diabetes among men and women in China. N Engl J Med 2010;362:1090-1101.

19 Huo TI, Lui WY, Huang YH, Chau GY, Wu JC, Lee PC, et al. Diabetes mellitus is a risk factor for hepatic decompensation in patients with hepatocellular carcinoma undergoing resection: a longitudinal study. Am J Gastroenterol 2003;98: 2293-2298.

20 Gao C, Zhao HC, Li JT, Yao SK. Diabetes mellitus and hepatocellular carcinoma: comparison of Chinese patients with and without HBV-related cirrhosis. World J Gastroenterol 2010;16:4467-4475.

21 El-Serag HB, Richardson PA, Everhart JE. The role of diabetes in hepatocellular carcinoma: a case-control study among United States Veterans. Am J Gastroenterol 2001;96: 2462-2467.

22 Yu L, Sloane DA, Guo C, Howell CD. Risk factors for primary hepatocellular carcinoma in black and white Americans in 2000. Clin Gastroenterol Hepatol 2006;4:355-360.

23 Toyoda H, Kumada T, Nakano S, Takeda I, Sugiyama K, Kiriyama S, et al. Impact of diabetes mellitus on the prognosis of patients with hepatocellular carcinoma. Cancer 2001;91:957-963.

24 Poon RT, Fan ST, Wong J. Does diabetes mellitus influence the perioperative outcome or long term prognosis after resection of hepatocellular carcinoma? Am J Gastroenterol 2002;97:1480-1488.

25 Tung HD, Wang JH, Tseng PL, Hung CH, Kee KM, Chen CH, et al. Neither diabetes mellitus nor overweight is a risk factor for hepatocellular carcinoma in a dual HBV and HCV endemic area: community cross-sectional and case-control studies. Am J Gastroenterol 2010;105:624-631.

26 Kawaguchi T, Yoshida T, Harada M, Hisamoto T, Nagao Y, Ide T, et al. Hepatitis C virus down-regulates insulin receptor substrates 1 and 2 through up-regulation of suppressor of cytokine signaling 3. Am J Pathol 2004;165:1499-1508.

27 Lee MS, Hsu CC, Wahlqvist ML, Tsai HN, Chang YH, Huang YC. Type 2 diabetes increases and metformin reduces total, colorectal, liver and pancreatic cancer incidences in Taiwanese: a representative population prospective cohort study of 800,000 individuals. BMC Cancer 2011;11:20.

28 Li D, Yeung SC, Hassan MM, Konopleva M, Abbruzzese JL. Antidiabetic therapies affect risk of pancreatic cancer. Gastroenterology 2009;137:482-488.

29 Yang YX, Hennessy S, Lewis JD. Insulin therapy and colorectal cancer risk among type 2 diabetes mellitus patients. Gastroenterology 2004;127:1044-1050.

30 Kawaguchi T, Izumi N, Charlton MR, Sata M. Branchedchain amino acids as pharmacological nutrients in chronic liver disease. Hepatology 2011;54:1063-1070.

31 Kawaguchi T, Taniguchi E, Morita Y, Shirachi M, Tateishi I, Nagata E, et al. Association of exogenous insulin or sulphonylurea treatment with an increased incidence of hepatoma in patients with hepatitis C virus infection. Liver Int 2010;30:479-486.

32 Tarantino G, Colicchio P, Conca P, Finelli C, Di Minno MN, Tarantino M, et al. Young adult obese subjects with and without insulin resistance: what is the role of chronic inflammation and how to weigh it non-invasively? J Inflamm (Lond) 2009;6:6.

33 Tarantino G, Conca P, Pasanisi F, Ariello M, Mastrolia M, Arena A, et al. Could inflammatory markers help diagnose nonalcoholic steatohepatitis? Eur J Gastroenterol Hepatol 2009;21:504-511.

December 12, 2012

Accepted after revision May 4, 2013

Author Affiliations: Department of Gastroenterology, China-Japan Friendship Hospital, Ministry of Health, Beijing 100029, China (Gao C, Fang L, Zhao HC, Li JT and Yao SK)

Shu-Kun Yao, MD, Department of Gastroenterology, China-Japan Friendship Hospital, Ministry of Health, No. 2 Yinghua East Road, Beijing 100029, China (Tel: 86-10-84206160; Fax: 86-10-64222978; Email: yaosk@zryhyy.com.cn)

? 2013, Hepatobiliary Pancreat Dis Int. All rights reserved.

10.1016/S1499-3872(13)60060-0