As described in the June 2011 Journal of Hepatology, Charlotte Costentin and colleagues in France set out to evaluate the association between caffeine intake and histological liver injury revealed by biopsies in people with untreated chronic hepatitis C virus (HCV) infection.

The researchers enrolled 238 patients with chronic hepatitis C at Groupe Hospitalier Henri Mondor-Albert Chenevier between 2004 and 2006. All participants were treatment-naive and had previously received or agreed to a liver biopsy. Exclusion criteria included coinfection with hepatitis B virus or HIV.

Participants ranged in age from 34 to 56 years; 65% were men and 35% were women. Injection drug use was the main route of HCV infection. HCV genotype 1 accounted for 62% of infections, followed by genotype 3 at 17%.

Participants completed a questionnaire about their use of caffeine (coffee, caffeinated tea, and caffeinated soda), alcohol, tobacco, and cannabis during the previous 6 months. Participants were asked about the quantity and frequency of their daily caffeine consumption over the previous 6 months. Serum ALT levels and fasting blood glucose were measured at the same time the liver biopsy was performed and the questionnaire administered.

In analyzing the participants' biopsy results, the researchers used the Metavir scoring system to determine the degree of fibrosis and histological activity or liver inflammation. The Metavir system assigns 2 standardized numbers: one describes the degree of fibrosis, while the other measures the degree of inflammation. Fibrosis is scored on a scale from F0 to F4 and histological activity is scored on a scale from A0 to A3:

Fibrosis score:
	F0 = absent fibrosis
	F1 = mild fibrosis
	F2 = moderate fibrosis
	F3 = advanced fibrosis
	F4 = cirrhosis

Histological activity score:
	A0 = no inflammation
	A1 = mild inflammation
	A2 = moderate inflammation
	A3 = severe inflammation

Participants were classified into 4 groups reflecting their daily caffeine consumption:

	Group 1 = < 225 mg/day (less than 1.5 cup/day)
	Group 2 = 225-407 mg/day (more than 1.5 but less than 3 cups/day)
	Group 3 = 408-678 mg/day (at least 3 but less than 5 cups/day)
	Group 4 = > 678 mg/day (at least 5 cups/day)

The study compared participants' caffeine consumption and its correlation with histological activity, degree of fibrosis, and serum ALT levels.

Results

	Histological activity scores > A2 declined gradually with increasing amounts of caffeine consumed:
	Among participants in Group 1 (less than 1.5 cups/day), 78% had a Metavir activity score of A2 or greater. Of the participants in Group 4 (at least 5 cups/day), 48% had an inflammation score of A2 or greater. The number of participants with a Metavir activity score of A3 was 10, too small to draw conclusions, and scores of A4, if any, were not reported.
	Other factors related to a score > A2 included age > 40 years, BMI > 25, moderate or marked steatosis (fatty liver), fibrosis >F2, and elevated ALT levels.
	No significant relationships were observed between inflammation score and sex, route of HCV transmission, hyperglycemia, diabetes, alcohol use, tobacco use, cannabis use, or HCV genotype.
	A Metavir activity score > A2 was related to fibrosis stage F2-F4, moderate to severe steatosis, and elevated ALT levels.
	Fibrosis >F2 was significantly correlated with male sex, age > 40 years, alcohol abuse, tobacco use (more than 15 cigarettes/day), daily cannabis use, HCV genotype 3, BMI > 25, moderate to severe steatosis, and elevated ALT levels.
	There was no association, however, between fibrosis severity and caffeine consumption.
	Similarly, there was no association between caffeine consumption and ALT levels.
	Tobacco use was not independently associated with Metavir activity levels; however, the proportion of heavy smokers rose with increasing caffeine consumption.

The researchers noted 2 possible limitations of their study. "Socio-economic status, quality of life, and comorbidities, which may be significantly associated with coffee intake, have not been examined in our cohort," they wrote. The absence of this information limits the scope of the study since other factors associated with caffeine consumption could explain the study results. Additionally, there was a low prevalence of alcohol use among study participants, which may account for the weak impact of alcohol consumption on necroinflammation.

"[O]ur study uncovers an inverse relationship between caffeine consumption and Metavir activity grade in patients with chronic hepatitis C, suggesting that caffeine intake may lower necroinflammatory injury by an as yet undetermined mechanism," the authors wrote.

"Given the strong relationship between activity grade and fibrosis progression, these results support the hypothesis that caffeine intake may also reduce the progression of liver fibrosis," they concluded. "Additional studies are required to assess whether our findings also apply to other inflammatory chronic liver diseases."

Appreciating that hepatitis C is influenced by many co-factors and co-morbidities that affect disease progression and long-term outcomes, the researchers did not make any clinical recommendation regarding caffeine consumption and liver disease. However, this study adds to the increasing evidence that caffeine consumption does not appear to be harmful to people with liver disease and may, in fact, be beneficial.

Investigator affiliations:
Service d'Hépatologie et de Gastroentérologie, Groupe Hospitalier Henri Mondor-Albert Chenevier, Créteil 94000, France; INSERM, U955, Créteil 94000, France; Université Paris-Est, Faculté de Médecine, UMR-S955, Créteil 94000, France; Service de Santé publique, Groupe Hospitalier Henri Mondor-Albert Chenevier, Créteil 94000, France; Service d'Anatomo-pathologie, Groupe Hospitalier Henri Mondor-Albert Chenevier, Créteil 94000, France; Service de Virologie, Groupe Hospitalier Henri Mondor-Albert Chenevier, Créteil 94000, France.

5/31/11

Reference
CE Costentin, F Roudot-Thorava, ES Zafrani, et al. Association of caffeine intake and histological features of chronic hepatitis C. Journal of Hepatology 54(6): 1123-1129 (abstract). June 2011.