Reuters Health Information
Jul 25, 2013
By Rob Goodier
NEW YORK (Reuters Health) Jul 25 - Marijuana smoking doesn't appear to accelerate liver disease progression in HIV patients with hepatitis C coinfection, a new prospective study from Canada has found.
"Based on previous studies, physicians have counseled patients that marijuana is harmful to their livers," Dr. Marina Klein, of McGill University Health Centre in Montreal, told Reuters Health by email.
"The results of our study suggest that physicians can reassure their patients that marijuana use, while it may have other deleterious effects, likely will not make their liver disease progress more rapidly," said Dr. Klein, who led the study, published online June 28 in Clinical Infectious Diseases.
Patients in Canada can apply for permission to use marijuana for medicinal purposes. The United States has conflicting marijuana laws. The federal government outlaws it, but 18 states allow for medicinal use, including two, Colorado and Washington, that also allow for recreational use.
The researchers followed up with 690 patients at 17 Canadian HIV clinics for an average of 2.7 years. The patients had HIV-hepatitis C coinfection at baseline, but did not have significant liver fibrosis (aspartate aminotransferase-to-platelet ratio (APRI) < 1.5).
At baseline, more than half of the patients said they had smoked marijuana in the past six months (median, seven joints per week) and 40% of those did so daily. A similar proportion said they used the drug for symptom relief.
In the course of the study, 19% of the patients reached an APRI score of 1.5, and 15% percent progressed to a score of 2. Eight patients, representing 1.2% of the study group, developed cirrhosis and 11, or 1.6%, developed end-stage liver disease (ESLD).
On multivariate analysis, there was no link between marijuana smoking and progression to liver disease based on the APRI score.
The researchers did find an association between marijuana use and progression to a clinical diagnosis of cirrhosis, with a hazard ratio of 1.33 per 10 additional joints per week. But when they reevaluated the findings using a lagged model of marijuana exposure, the association became statistically insignificant.
The lagged model measured marijuana use six to 12 months before the period during which liver disease developed, while the main model measured marijuana use during the same time that liver disease was diagnosed.
"The idea, really, was to make sure that smoking started before the participants got a liver problem and using the exposure and outcome at the same interval doesn't allow that (e.g. patients may have simply increased their use because they had symptoms of a disease that was already present). When we did this, it no longer looks as though marijuana is associated with any of the outcomes we investigated," Dr. Klein said.
That bias may explain some of the associations found between marijuana use and liver disease in past studies, she added.
Clin Infect Dis 2013.