Prevalence of elevated alanine aminotransferase levels stratified by the levels of cigarette smoking (smoking) and alcohol consumption (drinking) among persons who were seropositive for anti–hepatitis C virus antibody. PPD indicates packs of cigarettes per day. Levels of alcohol consumption are described in the "Subjects and Methods" section.
Customize your JAMA Network experience by selecting one or more topics from the list below.
Wang C, Wang S, Chang T, Yao W, Chou P. Smoking and Alanine Aminotransferase Levels in Hepatitis C Virus Infection: Implications for Prevention of Hepatitis C Virus Progression. Arch Intern Med. 2002;162(7):811–815. doi:10.1001/archinte.162.7.811
Alcohol consumption is a well-known risk factor for elevated ALT levels, but the role of cigarette smoking is unclear.
We collected a cross-sectional sample of 6095 inhabitants 35 years or older in a community with hyperendemic hepatitis B and C virus infections. We assayed levels of serum alanine aminotransferase (ALT), hepatitis B surface antigen (HBsAg), and anti–hepatitis C virus antibody (anti-HCV). Multivariate logistic regression was performed to determine the factors for elevated ALT levels (≥40 U/L) among people with different hepatitis infection statuses.
Prevalence of elevated ALT levels in individuals who were seronegative for both infections or seropositive for HBsAg or anti-HCV was 3.9%, 11.1%, and 30.8%, respectively. Subjects with elevated ALT levels were more likely to be seropositive for anti-HCV, male, and seropositive for HBsAg; to drink alcohol; to smoke; and to have undergone blood transfusion (P<.05). An association was found between elevated ALT levels and the consumption of cigarettes and alcohol among anti-HCV–seropositive subjects. In multivariate logistic analyses, alcohol consumption (odds ratio [OR], 2.2; 95% confidence interval [CI], 1.2-4.1) and smoking (OR, 1.8; 95% CI, 1.1-2.7) were significantly associated with elevated ALT levels among anti-HCV–seropositive subjects, but no such association was found among HBsAg-seropositive subjects. The odds of elevated ALT levels were 7 times higher (95% CI, 2.7-18.8) for the anti-HCV–seropositive patients who smoked 1 or more packs of cigarettes per day and frequently drank alcohol than for those who did not.
Smoking and alcohol consumption are independently associated with elevated ALT levels among anti-HCV–seropositive individuals but not among HBsAg-seropositive individuals. Patients who are seropositive for anti-HCV are strongly advised not to smoke and drink alcohol to reduce the possible risk for aggravating the liver dysfunction.
SERUM ALANINE aminotransferase (ALT) is the most suitable and useful protein enzyme used in the evaluation of hepatocellular damage. It is also a surrogate marker for disease severity or index of hepatic activity.1,2 Elevated ALT levels are associated with obesity, sex, alcohol consumption, use of medication, and viral hepatitis infection.2-4 Its association with viral hepatitis infection makes the ALT level a useful indicator of drug efficacy in the treatment of hepatitis B (HBV) and C virus (HCV).5,6
Alcohol consumption is a well-known risk factor for alcoholic liver diseases, cirrhosis, and hepatoma, independent of HBV or HCV infection.7,8 An additive or possibly synergistic effect exists between alcohol consumption and HCV on elevated ALT levels.9 However, the effect of smoking on the activity of serum ALT among patients with HBV or HCV infection has not been widely reported. Whether the effect is additive to that of alcohol consumption or depends on hepatitis infection status remains unclear.
The aim of this study was to determine whether an association exists between alcohol consumption and ALT level, and between cigarette smoking and ALT level, and whether the relationships vary according to hepatitis infection status. These questions are important because, if these risk behaviors have different effects on ALT levels for different hepatitis infection statuses, they could provide evidence of different viral activity under the influence of alcohol consumption and/or cigarette smoking that could be helpful for secondary prevention. Previous studies of elevated ALT levels were mostly hospital based or conducted in regions with low prevalences of seropositivity for hepatitis B surface antigen (HBsAg) or anti-HCV antibody. To our knowledge, no large-scale population study on elevated ALT levels has been performed in a community with a high prevalence of HBsAg and anti-HCV seropositivity.10 We surveyed a community with hyperendemic HBV and HCV infection to determine the risk factors for elevated ALT levels.
In A-Lein, a township of about 30 000 in southern Taiwan, patients with serious liver diseases abound, and mortality due to hepatoma is more than double the average in Taiwan,11 which exceeds 30 cases per 100 000 per year. This rate is at least 6 times higher than that of the countries in low-risk areas such as northern Europe and the United States.12 In a community-wide hepatitis screening of the 8800 adults 35 years or older in A-Lein, the prevalence of anti-HCV seropositivity was found to be approximately 17% and of HBsAg seropositivity to be approximately 13%.10 The study lasted from January 1, 1996, through December 31, 1997. Intensive health promotion activities for hepatitis screening were held before blood samples and questionnaires were collected. Villagers came voluntarily during this period. We have found no differences in the age, sex, health behaviors, or prevalence of HBV and HCV between the responders and nonresponders. A total of 6095 persons participated in this hepatitis screening program, a response rate of 69.3%.
The questionnaire asked about demographic data, average daily consumption of cigarettes, the habit of alcohol consumption, and blood transfusions. The habit of drinking alcohol was subclassified as frequent, occasional, or rare. Cigarette consumption was defined as (1) equal to or greater than 1 pack per day (PPD), (2) less than 1 PPD, (3) abstained (stopped ≥6 months before this study), or (4) never smoked. The town had no reported cases of human immunodeficiency virus infection and was believed to have no or a negligible number of intravenous drug abusers among the target population. Very few people who were seropositive for HBsAg or anti-HCV had received interferon therapy.
Serum samples were sent to the Tainan Blood Center of the Chinese Blood Service Foundation, Tainan, Taiwan, to test for HBsAg and anti-HCV markers and ALT levels. Seropositivity for HBsAg was determined by means of an enzyme immunoassay method (Version I; Murex, London, England). Seropositivity for anti-HCV was tested by means of the third-generation Murex anti-HCV enzyme immunoassay method, which contains the antigen from the HCV core and nonstructural 3, 4, and 5 regions. Levels of ALT were measured by means of a kinetic UV test for clinical chemistry analyzers (Olympus System Reagent; Olympus Diagnostica GmbH, Mills, Ireland). The cutoff value for diagnosis of elevated ALT level was set at 40 U/L according to the suggestion of the Blood Center of the Chinese Blood Service Foundation; this standard was also adopted by the National Health Insurance Bureau for eligibility of claims about liver disorders. The same cutoff value has been used for identifying communities with a high prevalence of HCV infection.13 In this study, HBsAg seropositivity also indicates anti-HCV seronegativity; anti-HCV seropositivity also indicates HBsAg seronegativity; and seronegativity describes the negative status for both infections.
Univariate analysis and multiple logistic regression were used to determine the association of risk factors for elevated ALT levels (<40 vs ≥40 U/L). Multivariate logistic analysis was used to determine the risk factors for elevated ALT levels among seronegative, HBsAg-seropositive, and anti-HCV–seropositive individuals. The factors included in the multivariate logistic regression analysis were sex, stratified age (<40, 40-49, 50-59, and ≥60 years), alcohol and cigarette consumption, and blood transfusion. In multivariate logistic regression analysis, individuals with coinfection (HBsAg and anti-HCV seropositivity) were not included. We analyzed all results using Stata software.14
Of the 6095 community members who participated in this study, 9.2% had elevated ALT levels (≥40 U/L). The prevalences of elevated ALT levels in individuals who were seronegative, seropositive for HBsAg, and seropositive for anti-HCV and coinfected were 3.9%, 11.1%, 30.8%, and 23.1%, respectively. Individuals who were seropositive for HBsAg were 3 times more likely to have elevated ALT levels than seronegative individuals (odds ratio [OR], 3.1; 95% confidence interval [CI], 2.3-4.1). Individuals who were seropositive for anti-HCV were 11 times more likely to have elevated ALT levels than seronegative individuals (OR, 11.0; 95% CI, 8.9-13.5). In addition, individuals who were male, habitual (frequent) drinkers of alcohol, or habitual smokers of cigarettes (≥1 PPD) or who had undergone a blood transfusion were more likely to have elevated serum ALT levels (P<.05) (Table 1). The mean and median levels of ALT in the various comparison groups are also shown in Table 1. Among these variables, anti-HCV–seropositive and coinfected individuals and those with frequent alcohol consumption had higher mean and median ALT levels.
In multivariate logistic regression analysis of the risk factors for elevated ALT levels stratified by hepatitis infection status, no significant factor affected HBsAg-seropositive subjects, whereas alcohol consumption (OR, 2.2; 95% CI, 1.2-4.1) cigarette smoking (OR, 1.8; 95% CI, 1.1-2.7), and being aged 50 to 59 years (OR, 1.4; 95% CI, 1.0-2.0; P = .045) were significant factors associated with elevated ALT levels among anti-HCV–seropositive subjects (Table 2). Among seronegative subjects, being male (OR, 1.5; 95% CI, 1.0-2.2), drinking alcohol (OR, 2.7; 95% CI, 1.6-4.5), being aged 40 to 49 years (OR, 1.7; 95% CI, 1.2-2.5), and having undergone a blood transfusion (OR, 1.6; 95% CI, 1.0-2.4) were significant factors for elevated ALT levels, but those in older stratified age groups were less likely to have elevated ALT levels (OR, 0.98; 95% CI, 0.97-0.99).
The additive effect of cigarette and alcohol consumption on elevated ALT levels was found among anti-HCV–seropositive subjects. The risk (70.0%) was highest among those who habitually smoked 1 PPD or more and who frequently drank alcohol (Figure 1). The odds of elevated ALT levels were 7 times higher (95% CI, 2.7-18.8) for that group than for those who did neither. An association between the prevalence of elevated ALT levels and the levels of alcohol drunk and cigarettes smoked was found in the anti-HCV–seropositive group but not in the HBsAg-seropositive group. Figure 1 also shows that the ALT level is more elevated for heavy smokers (≥1 PPD) than for frequent drinkers who did not smoke (50.0% vs 44.4%).
This study demonstrates that consumption of cigarettes and alcohol are associated with elevated ALT levels among anti-HCV–seropositive subjects but not among HBsAg-seropositive subjects. Furthermore, cigarette smoking and alcohol consumption had an additive effect on the prevalence of elevated ALT levels in these patients. These results indicate that cigarette smoking and alcohol consumption are closely associated with the prevalence of elevated ALT levels in anti-HCV–seropositive individuals. The prevalence of elevated ALT levels for those who only smoked heavily (≥1 PPD) is not less than the prevalence for those who only drank frequently (50.0% vs 44.4%), which implies that smoking, like alcohol consumption, is an independent promoting factor for hepatic necroinflammation. A plausible explanation for the significant effect of cigarette smoking on hepatic necroinflammation is that the liver is a target organ for the chemicals in tobacco and alcohol. Patients who were seropositive for anti-HCV with elevated ALT levels require repeated hospital visits for biochemical tests, ultrasonography, and medical treatment, and hepatocellular carcinoma is more likely to develop than in those with normal ALT levels.15,16 Therefore, abstinence from smoking and alcohol consumption by anti-HCV–seropositive individuals could probably help to slow their progression into advanced liver diseases and minimize their need for health care services. At present, treatment for chronic HCV infection is recommended for patients with persistently elevated ALT levels,17 and no rationale exists for treating anti-HCV–seropositive patients with normal ALT levels.18 Patients who are seropositive for anti-HCV and who have active, heavy alcohol intake are not recommended for treatment, because alcohol increases viremia and interferes with the response to treatment.9 It is also rational to advise all anti-HCV–seropositive patients, particularly those with elevated ALT levels, to stop smoking to eliminate the additive effect that smoking has on HCV antibodies. Whether this will cause elevated ALT levels to return to normal in some patients needs further study. However, to suggest that they abstain from smoking before initiating expensive and unpleasant anti-HCV treatment might be a cost-effective measure and might increase treatment response.
A high prevalence of HCV infection and rate of viral replication, rapid changes in histological progression of liver lesions, and increased carcinogenesis have been noted in patients with alcoholic liver diseases.8,9,19 People who habitually smoked cigarettes showed a higher prevalence of HCV infection than those who did not, and multivariate logistic regression showed smoking to be a significant risk factor for acquiring HCV infection.10 Cigarette smoking is also highly associated with the development of hepatocellular carcinoma.15,20 Nevertheless, the effect of smoking on the necroinflammatory activity of anti-HCV–seropositive patients is not fully understood. Smoking has been shown to influence all aspects of the immune systems, including alterations of humoral and cellular immunity.21 Further study is needed to determine whether cigarette smoking has effects similar to those of alcohol consumption on the increase of viral replication, the worsening of histological progression, and the immunosuppression within anti-HCV–seropositive individuals.
In this study, the prevalence of elevated ALT levels was about 3 times higher for those who habitually smoked 1 PPD or more and frequently drank alcohol than for those who did neither. In addition, 75.2% of those anti-HCV–seropositive subjects who did not smoke cigarettes or drink alcohol had ALT levels within the normal range. The reason for a greater prevalence of normal ALT levels among these anti-HCV–seropositive individuals could be that more of them were seronegative for HCV RNA and they had less change in histological findings in the liver. As a result, they might be less liable to development of chronic disease than those who do consume cigarettes or alcohol.1,17,22 In addition, they might be more likely to recover from an acute HCV infection and become seronegative for HCV RNA because of better host immunity. A clear correlation exists between alcohol consumption and liver HCV RNA levels.9 Whether a correlation also exists between smoking levels and liver HCV RNA levels requires further study. The prevalence of cigarette smoking is much higher than that of alcohol consumption (27.2% vs 5.5%) in this study, but the effect of smoking on anti-HCV–seropositive people is almost overlooked in practice. It will be helpful for secondary prevention to suggest that anti-HCV–seropositive patients abstain from cigarette smoking as well as alcohol consumption.
Confounding factors might exist in this study. First, because smokers might use health care services more often than nonsmokers,23 they might be given more medications or medical injections that could result in different genotypes of HCV infections or that could produce hepatic necroinflammation and thereby cause more elevated ALT levels.2 Second, smokers were more likely to have a lower socioeconomic status than nonsmokers and therefore were more likely to work in more hazardous environments that cause more elevated ALT levels.2
Some limitations exist in this community-wide population study. First, because the amounts of cigarette and alcohol consumption were self-reported, a bias toward underestimation might exist due to social stigmata associated with the reporting of these adverse lifestyle practices. Misclassification of alcohol consumption might have occurred, because information about the amount of alcohol consumed was not obtained. Second, the duration of cigarette and alcohol consumption was not recorded, but duration is associated with liver damage that can cause elevated ALT levels. Third, polymerase chain reaction analysis for HCV RNA levels was not performed to elucidate the relationship between the actual viral status or viral titers and past infections. Finally, this cross-sectional study may blur etiologic associations; hence, a long-term prospective study is needed to see the effects of cigarette smoking, the combined effects of smoking and alcohol consumption, and the effects of abstinence on HCV infection.
Alcohol and cigarette consumption are independently associated with elevated ALT levels among anti-HCV–seropositive individuals but not among HBsAg-seropositive individuals. This finding will be useful for secondary prevention among anti-HCV–seropositive patients to slow further progression into advanced liver diseases and to decrease excessive use of health care services. Abstention from cigarettes and alcohol by anti-HCV–seropositive patients might be a cost-effective measure before initiation of treatment for anti-HCV seropositivity. The detailed biological effect of smoking on the necroinflammation, viral loads, histological progression, and hepatocarcinogenesis among anti-HCV–seropositive patients needs further evaluation.
Accepted for publication July 31, 2001.
This study was supported by the C. T. Hsu Cancer Research Foundation, Taipei City, Taiwan.
We thank the A-Lein Community Health Promotion Committee, Kaohsiung County, Taiwan; the staff and volunteers of the A-Lein Community Health Center, Kaohsiung County, for their assistance in data collection; and the Tainan Blood Center of the Chinese Blood Service Foundation, Tainan City, Taiwan, for assistance with laboratory analysis.
Corresponding author and reprints: Pesus Chou, DrPH, National Yang-Ming University, Institute of Public Health, Shih-Pai, Taipei 112, Taiwan, Republic of China (e-mail: email@example.com or firstname.lastname@example.org).