This article was published in the December 2016 edition of Medical Observer, under the title, “Does milk thistle do any good?” (pp. 64-65). (PDF)
Thanh, a 35-year-old pharmacist with chronic hepatitis B, consulted with me recently. He mentioned that he had been treating himself with a milk thistle (Silybum marianum) product and wondered what my thoughts about it were. I knew that milk thistle is commonly marketed for “liver health” and wondered what the evidence was.
What is the effect of milk thistle extracts, on liver disease severity, in people living with chronic hepatitis B?
What does the research evidence say?
Step 1: The Cochrane Library
The Cochrane Library has a systematic review, updated to 2005, on milk thistle for alcoholic and/or hepatitis B or C virus liver diseases .
Step 2: TripDatabase
I conducted a search using the TripDatabase PICO search tool (Participant: “hepatitis B”, Intervention: “milk thistle”, Comparator: “placebo”, Outcomes: blank). There did not appear to be a more appropriate or newer paper. TripDatabase did identify two newer studies in patients with viral hepatitis – however, one was in patients with chronic hepatitis C , and the other was terminated with no results available.
Although it isn’t ideal, let’s look at the Cochrane systematic review by Rambaldi et al. (2007) in detail .
I will use the systematic reviews critical appraisal sheet from the Centre for Evidence Based Medicine .
What PICO question does the systematic review ask?
In people living with alcoholic liver disease, and/or viral induced liver diseases (hepatitis B and/or C) (Participants); what is the effect of milk thistle or any milk thistle constituent at any dose or duration (Intervention); compared to placebo or no intervention (Comparator); on several outcome measures including, (i) death, (ii) development of hepatitis clinical symptoms and complications (e.g., ascites, variceal bleeding), (iii) liver biochemistry, (iv) liver biopsy findings, and (v) adverse events (Outcome).
Is it clearly stated?
Is it unlikely that important studies were missed?
Unclear. Although the authors searched multiple electronic databases, and approached the principal authors of the identified trials and the manufacturers of milk thistle products to enquire about any relevant unidentified or unpublished randomised trials, this systematic review is close to 12 years out-of-date.
Were the criteria used to select articles for inclusion appropriate?
Yes. The authors only included randomised trials that enrolled patients with the aforementioned types of liver disease. Other types of liver disease, and trials of liver disease prevention, were excluded.
Were the included studies sufficiently valid for the question asked?
Unclear, possibly not. The authors formally assessed the risk of bias of the included studies using a clearly described process (pp 4-5) . Of the 13 included trials, only 1 provided a sample size calculation (see Stat Facts), 6 had an adequate method to generate the allocation sequence, only 3 described adequate allocation concealment, with only 1 described as double-blinded.
Were the results similar between studies?
Mostly. There was little heterogeneity in the results as presented, though given the small number of studies, there were not many direct comparisons.
What were the results?
The results are limited. Of the included studies, only one was specifically in people with hepatitis B, and there were fewer than 30 participants in that study. Participants who received milk thistle compared to placebo :
- There was no difference in mortality (analysis 1.1, p 39) or liver-related mortality (analysis 2.1, p 53) (note: there were no deaths in the studies of participants with viral hepatitis).
- There were no differences in the risk of any liver complications (analysis 3.4, p 59) (note: results for people with HBV not available)
- There was no effect on fibrosis score (analysis 3.13, p 67) (note: single study in alcoholic liver disease)
Discussion and conclusion
There is limited evidence on the effect of milk thistle on liver diseases generally, and on chronic hepatitis B specifically. Of the evidence that exists, the quality appears to be low, with most of the randomised trials having very small numbers of participants, and methodological issues.
When all the trials are taken together, there was a statistically significant beneficial effect on liver-related mortality, but this was mostly in the alcohol-liver disease participants, and the benefit was no longer statistically significant when the analysis was restricted to high-quality studies .
My interpretation of the evidence is that it is insufficient to provide an estimate of the effect of milk thistle on chronic hepatitis B. Of the evidence that does exist, there does not appear to be any dramatic beneficial effects, and the possibility that there may be no meaningful therapeutic effects seems likely. The precautionary principle should apply and milk thistle products cannot be recommended for chronic hepatitis B.
On further discussion with Thanh, he had not experienced any subjective benefit from milk thistle and was happy to stop using it.
The sample size of a clinical trial – the number of participants in each arm – should be determined in the study design. Importantly, this depends on the expected magnitude of the effect (more participants are required to identify a smaller effect). Studies with too few participants have low statistical power and thus a lower chance of detecting a true effect. Less commonly appreciated is that low power also increases the risk that a positive result is false, and that the detected effect size is an over-estimate .
- Rambaldi A, Jacobs BP, Gluud C. Milk thistle for alcoholic and/or hepatitis B or C virus liver diseases. Cochrane database Syst Rev 2007 Oct 17(4):CD003620.
- Fried MW, Navarro VJ, Afdhal N, et al. Effect of silymarin (milk thistle) on liver disease in patients with chronic hepatitis C unsuccessfully treated with interferon therapy: a randomized controlled trial. JAMA 2012 Jul 18;308(3):274-82.
- Centre for Evidence Based Medicine. Systematic Review: Are the results of the review valid? Oxford: University of Oxford, 2005.
- Button KS, Ioannidis JP, Mokrysz C, et al. Power failure: why small sample size undermines the reliability of neuroscience. Nat Rev Neurosci 2013 May;14(5):365-76.