Milk thistle is an herbal product that is widely recommended and used by veterinarians. Like glucosamine, it is a supplement which has leapt over the gap between alternative and conventional medicine. Unfortunately, like glucosamine, this acceptance has come about on the basis of pretty weak evidence.
The active ingredient is a cluster of compounds called silymarin. There has been extensive in vitro research on silymarin, and it has a wide range of potentially useful effects. It appears to interfere with pro-inflammatory chemicals, functions as an anti-oxidant, and may interfere with the metabolism of some chemicals into toxic compounds in the liver. It also has some activity which could be potentially harmful, including interfering with the metabolism of a number of drugs and stimulating the effects of hormones like estrogen. As usual, these laboratory findings indicate the possibility of useful clinical effects, but most compounds that have these sorts of potentially useful effects in test tubes don’t work out to be good medicines.
The primary uses of silymarin in humans are to protect against or treat liver damage from toxins and infectious diseases, to improve the condition of diabetics, and to protect the kidneys from toxins. Some of these uses are based on traditional folklore, but as usual there are many traditional uses no longer recommended and for which there is not yet scientific support, including disease of the spleen, uterine disease, malaria treatment, appetite stimulation, stimulation of lactation, and others. In dogs and cats the primary use of for non-specific “support” of the liver regardless of the specific disease.
In humans, clinical trial evidence is mixed. A couple of studies have suggested it reduces insulin resistance in diabetic and may lower blood lipid levels. A Cochrane review of 13 studies including 915 people “could not demonstrate significant effects of milk thistle on mortality or complications of liver disease in patients with alcoholic and/or hepatitis B or C liver disease.” High quality trials were negative, and low quality trials suggested a benefit.
Very little research exists in dogs and cats. A small study of 20 cats given acetaminophen, a known liver toxin, found that those given a single oral dose of silymarin did not show the elevation of liver enzyme levels seen in those not given the compound. A similar study of dogs found some differences in elevations of kidney values between those that got silymarin and those that didn’t following exposure to a kidney toxin, though there was not a completely consistent pattern.
A study done in 1978 showed that dogs given a toxic mushroom compound orally and then given silymarin intravenously did not show the increase in liver values that was seen in control dogs. Another in 1984 found that 30% of the control dogs died whereas none of the dogs given IV silymarin along with the mushroom toxin died, and the livers from the treated dogs did not appear damaged by the toxin.
As far as risks, there appear to be few. Nausea, diarrhea, and other gastrointestinal effects are sometimes seen, and allergic reactions have been reported in humans.
So overall, the in vitro and laboratory animal evidence indicates it is plausible that milk thistle extract might have beneficial effects, though harmful effects in some situations could be expected as well. In humans, the clinical trials show weak evidence for benefit in diabetics and inconsistent but generally negative evidence for benefit in alcoholic or hepatitis-associated liver disease. Very little experimental, and apparently no high quality or controlled clinical research exists in dogs and cats. What there is suggests a benefit is possible. But we must bear in mind that preliminary, low-quality trials of milk thistle in humans looked promising but were not supported by subsequent better quality trials.
A clinical trial comparing animals with naturally occurring liver disease treated identically except for receiving either milk thistle or a placebo would be quite useful. In the meantime, use of the compound is not unreasonable given the suggestive low-level evidence, but it is not much more than a hopeful shot in the dark at this point.