Critique 013: Alcohol and liver cirrhosis 27 July 2010
Rehm J, Taylor B, Patra SM, Irving H, Baliunas D, Patra J, Roerecke M. Alcohol as a risk factor for liver cirrhosis: A systematic review and meta-analysis. Drug and Alcohol Review 2010,29,437–445. DOI: 10.1111/j.1465-3362.2009.00153.x 45
Introduction and Aims. Alcohol is an established risk factor for liver cirrhosis. It remains unclear, however, whether this relationship follows a continuous dose–response pattern or has a threshold. Also, the influences of sex and end-point (i.e.,mortality vs. morbidity) on the association are not known. To address these questions and to provide a quantitative assessment of the association between alcohol intake and risk of liver cirrhosis, we conducted a systematic review and meta-analysis of cohort and case–control studies.
Design and Methods. Studies were identified by a literature search of Ovid MeDLINE,EMBASE,Web of Science, CINAHL, PsychINFO, ETOH and Google Scholar from January 1980 to January 2008 and by searching the references of retrieved articles. Studies were included if quantifiable information on risk and related confidence intervals with respect to at least three different levels of average alcohol intake were reported. Both categorical and continuous meta-analytic techniques were used to model the dose–response relationship.
Results. Seventeen studies met the inclusion criteria. We found some indications for threshold effects. Alcohol consumption had a significantly larger impact on mortality of liver cirrhosis compared with morbidity. Also, the same amount of average consumption was related to a higher risk of liver cirrhosis in women than in men.
Discussion and Conclusions. Overall, end-point was an important source of heterogeneity among study results. This result has important implications not only for studies in which the burden of disease attributable to alcohol consumption is estimated,but also for prevention.
Background: There is no question that heavy alcohol consumption is a key factor in the development of Laennec’s cirrhosis, and that almost all studies show that women may be at greater risk than men for a specified level of intake. The key question for public health officials is whether or not there is a threshold level of drinking associated with increased risk of cirrhosis.
Based on physiologic studies, it has generally been stated that regular alcohol intake above a certain level (often, 40 to 50 grams/day) is necessary to produce cirrhosis. In epidemiologic studies, however, there is sometimes reported to be increased risk of cirrhosis at any level of drinking. Some have suggested that the purported increase at very low levels of alcohol consumption is due to underestimation of amount by heavier drinkers. Also, it is clear that heavy drinkers tend to have periods of no intake or light intake, and such persons may have been included in the very-light drinking category. Others have suggested that while a threshold of consumption may be true for people with normal livers, those with liver dysfunction from other causes may be unusually susceptible to even small amounts of alcohol.
Results reported in paper: This paper sought to determine if a dose-response curve was seen starting at very low levels of alcohol. Of particular interest were the noteworthy differences in the results for alcohol on a continuous scale for both morbidity and mortality (Figure 2 in the paper, copied below) and those based on categories of alcohol intake, displayed in Table 2.
The study reported that for morbidity from cirrhosis, both men and women consuming up to one drink per day had a lower risk than that of lifetime abstainers. Women had a significantly increased risk only above 24 grams/day and men above 36 grams/day. This suggests that there may a threshold effect of alcohol on the risk of cirrhosis. Further, there may be biological support for a lower risk for light drinkers in that some studies have shown that light drinking may decrease inflammation in the liver.1
Comments on study design and analysis: One Forum member commented: “The authors combined data from retrospective case-control studies with prospective cohort studies and with previous analyses, a cocktail very far from the Tom Chalmers proposal for meta-analysis (reworking previous published analyses could be indicated as ‘ruminant epidemiology”). Another Forum member also stated: “I find it of interest that this group have really reworked data from an earlier meta-analysis, then added some newer studies. I would have thought it would have been more sound to base their new analysis on new studies.”
It is not possible for the reader to assess the validity of the current conclusions given that no raw data were presented. It would have been useful to have had some effect estimates of OR or RR of liver cirrhosis for selected alcohol-consumption categories. The authors did not perform critical evaluation on whether the results varied by study design, by time of publication, by confounder adjustment, etc. Furthermore, while the curves in the figure are impressive, the confidence intervals are not provided.
Underreporting of alcohol: One Forum member stated: “Concerning the limits of moderate consumption and risk of liver cirrhosis, the authors have not included the issue of underreporting in their discussion. Drinkers in general may lie or grossly underestimate the reported quantity and frequency of their alcohol consumption. Comparison studies in the alcohol literature have shown that self-reported alcohol consumption accounts for only 40-60% of alcoholic beverages sold as measured by sales and tax data.2 It also appears that heavier drinkers underreport their drinking more than light drinkers. Hilton found some underreporting of the number of occasions on which larger numbers of drinks were consumed and a compensatory over-reporting of occasions on which only 1 or 2 drinks were consumed.3 Under-representation of heavier amounts was also reported by Perrine et al comparing daily telephone report of consumption with retrospective quantity-frequency self-report.4
Is there a threshold amount of alcohol intake for an increase in risk of cirrhosis? One Forum member stated: “As a lifelong liver physician, it has always seemed to me that one of the most important issues around the ‘threshold vs continuous’ argument concerns the fact that until around 1990 no cognizance was taken of the presence of HCV in populations under consideration. HCV was not discovered until 1988, and not regularly and widely tested for until about 1992. Furthermore, the increasing importance of the metabolic syndrome as a cause of cirrhosis morbidity and mortality did not really begin to be appreciated until the late 1990s. While we now know that there is a likely additive effect of heavy alcohol consumption and HCV, I am not aware of good epidemiological studies in which multivariate analysis of relative risks of HCV, obesity, and alcohol have been taken into account, particularly at the lower end of alcohol consumption.” He also added: “It seems biologically implausible that, while there is a threshold — actually quite a high threshold — for cirrhosis-related morbidity from alcohol among men in this analysis, the same does not appear to be the case for mortality.” It is true that some people with abnormal LFTs are often told they have “cirrhosis” – effectively to scare them off of drinking. “Furthermore, because there is a good deal of ignorance about non alcoholic steatohepatitis among family physicians, these doctors often assume that anyone who says that they drink, for example, 30-40 Gm ethanol/day, and who has persistently abnormal liver function tests, probably drinks more, and has alcoholic liver disease.” Diagnostic criteria for “alcoholic cirrhosis” are very important, and may have been lacking for the current paper.
On the other hand, if an individual has biopsy-proven alcoholic cirrhosis, and this is well compensated, then many studies show that stopping drinking enormously increases life expectancy. This can allow an older person to live long enough to die of something either unrelated, or slightly related, but not a “liver death”. Hepatologists do not see patients with alcoholic cirrhosis who drank less than 40gm ethanol/day. But many studies suggest that 20+ gm ethanol/day may well be additive in leading more quickly to cirrhosis from HCV. So it may be sensible to advise individuals with any significant liver disease to abstain from alcohol.
Advising against any alcohol intake for everyone with potential liver dysfunction, however, could lead to other problems. The most common liver disease in the US population is non-alcoholic fatty liver disease (NAFLD). Studies have shown that it is less frequent among moderate wine drinkers than among abstainers or consumers of other beverages.5 Also, Alkerwi et al6 performed a meta-analysis of relevant publications on the relation between alcohol consumption and the prevalence of metabolic syndrome (that can be associated with a fatty liver); these authors demonstrated that moderate alcohol consumption significantly reduces the prevalence of metabolic syndrome.
Lack of beverage-specific effects: The authors state that their data were insufficient to allow them to determine the effects on risk of cirrhosis from the pattern of drinking or according to the type of beverage consumed. This is unfortunate as there is considerable data suggesting differences in the effects of wine, beer, and spirits on liver disease.7 The literature search by Rehm et al gave a reference relating wine to cirrhosis (Gronbaek et al, 2004) which is a secondary report in a thematic issue of Biological Research. The original reference is from Becker in 2002,8 in which these authors present data from follow up of more that 30,000 subjects in the Copenhagen area. In that study, compared with individuals who drank no wine (relative risk set at 1.0), individuals drinking 16% to 30% of their total alcohol intake from wine had a relative risk of 0.4 (95% confidence limits, 0.3-0.6) and those drinking 51% or more of their alcohol from wine had a relative risk of 0.3 (95% confidence limits, 0.2-0.5) for developing cirrhosis. Their results suggest that a high intake of all three types of alcohol conveys an increased risk for cirrhosis, but wine drinkers are at a lower risk than beer and spirits drinkers for similar intakes.
In a small case-control study in France that enrolled 42 cirrhotic and 60 non-cirrhotic patients from four alcohol treatment units located in the Languedoc-Roussillon area, wine consumption was not associated with a decreased risk of alcoholic cirrhosis in heavy drinkers.9 The observed protective effect of wine drinking in Denmark may merely reflect a lower risk for later excessive drinking or a lower risk for being exposed to the unknown additional risk factors that initiate liver damage in subjects conditioned by heavy drinking. In addition to constituents in wine itself, unknown factors affecting risk of cirrhosis could be lifestyle factors such as nutrition quality, smoking, obesity, drinking patterns etc. Unknown factors might also be at play in the Languedoc-Roussillon, where the mean total lifetime alcohol consumption was higher (non-significant) in non-cirrhotic controls (2306 ± 1942 kg, median = 1646 kg) than in cirrhotic patients (1748 ± 1114 kg, median = 1564 kg).
Forum Conclusions: Despite some concerns about the approach used, the present meta-analysis supports a strong role of heavy alcohol drinking in the development of cirrhosis. The differences found by the investigators between the effects of moderate alcohol intake on cirrhosis morbidity and mortality are difficult to explain, but may relate to misclassification of exposure in observational studies of the effects of alcohol. The results related to morbidity suggest that small amounts of alcohol are not associated with an increased risk of cirrhosis, and may be associated with lower risk of disease. Thus, this study provides evidence for a “threshold” of alcohol intake for the development of cirrhosis. Limited previous data are available suggesting that wine consumption may be associated with lower risk of cirrhosis than the intake of other beverages, but heavy drinking of any type of beverage increases the risk.
References from Forum review
1. Szabo G. Moderate drinking, inflammation, and liver disease. Ann Epidemiol 2007;17:S49–S54.
2. Midanik L. The validity of self-reported alcohol consumption and alcohol problems: a literature review. Br J Addict 1982;77:357-382.
3. Hilton ME. A comparison of a prospective diary and two summary recall techniques for recording alcohol consumption. Br J Addict 1989;84:1085-1092.
4. Perrine MW, Mundt JC, Searles JS, Lester LS. Validation of daily self-reported alcohol consumption using interactive voice response technology. J Stud Alcohol 1995;56:487-490.
5. Dunn W, Xu R, Schwimmer JB. Modest wine drinking and decreased prevalence of suspected nonalcoholic fatty liver disease. Hepatology 2008;47:1947-1954.
6. Alkerwi A, Boutsen M, Vaillant M, Barre J, Lair M-L, Albert A, Guillaume M, Dramaix M. Alcohol consumption and the prevalence of metabolic syndrome: A meta-analysis of observational studies. Atherosclerosis 2009;204:624–635.
7. Gronbaek M, Jensen MK, Johansen D, Sorensen TI, Becker U. Intake of beer, wine and spirits and risk of heavy drinking and alcoholic cirrhosis. Biol Res 2004;37:195-200.
8. Becker U, Gronbaek M, Johansen D, Sorensen TI. Lower risk for alcohol-induced cirrhosis in wine drinkers. Hepatology 2002;35:868-875.
9. Pelletier S, Vaucher E, Aider R, Martin S, Perney P, Balmes JL, Nalpas B. Wine consumption is not associated with a decreased risk of alcoholic cirrhosis in heavy drinkers. Alcohol Alcohol 2002;37:618-621.
* * *
Contributions to these comments from the International Scientific Forum on Alcohol Research were from the following:
R. Curtis Ellison, MD, Section of Preventive Medicine & Epidemiology, Boston University School of Medicine, Boston, MA, USA
Harvey Finkel, MD, Hematology/Oncology, Boston University Medical Center, Boston, MA, USA
Oliver James, MD, Head of Medicine, University of Newcastle, UK
Francesco Orlandi, MD, Dept. of Gastroenterology, Università degli Studi di Ancona. Italy
Erik Skovenborg, MD, Scandinavian Medical Alcohol Board, Practitioner, Aarhus, Denmark
Yuqing Zhang, MD, DSc, Epidemiology, Boston University School of Medicine, Boston, MA, USA