Can I drink alcohol when I take methotrexate? This is one of the most often asked questions I’m asked in clinic when starting methotrexate. I also suspect it is one of the reasons people sometime decline treatment with methotrexate. This is a shame, because methotrexate the cornerstone medication for rheumatoid arthritis.
You would think the answer to this question is obvious. Alcohol can cause liver failure and cirrhosis. Methotrexate can also cause liver inflammation and fibrosis/cirrhosis. So combining the two doesn’t sound like such a hot idea. It turns out the answer to the question isn’t so black and white.
Yes, alcohol can cause liver failure, however, not everyone who drinks alcohol gets cirrhosis. It tends to happen when someone drinks excessively for years. The same goes for methotrexate. Not everyone who takes methotrexate gets liver inflammation.
The real question is, how MUCH alcohol is safe to take with methotrexate? Shockingly there was little if any data addressing this question until recently. How much alcohol is too much? How much alcohol is likely safe?
Science to the rescue! A recent study from the UK specifically addressed this question¹. But before I get into that, let’s quickly review what a standard unit of alcohol actually is. “I drink alcohol socially” simply isn’t going to cut it. “Social” is highly variable!
What is a standard unit of alcohol?
This depends on the country you’re talking about. Every country defines a unit of alcohol differently. Since I’m writing based from the US, I define a unit of alcohol as follows:
“NIH standard drink comparison” by the National Institutes of Health is in the public domain in the United States. This file has been identified as being free of known restrictions under copyright law, including all related and neighboring rights.
Moreover, each country has their own definition for the maximum amount of alcohol per week. Some countries are more permissive than others. In the US, defines low risk drinking for women as no more than 3 drinks on a single day and no more than 7 drinks per week. Men can drink no more than 4 drinks on a single day and no more than 14 drinks per week².
Quantifying the hepatotoxic risk of alcohol consumption in patients with rheumatoid arthritis taking methotrexate
Researchers in the UK recently looked into this matter. As I alluded to, our current guidelines actually offer no specific guidance about alcohol and methotrexate. In the UK, “patients taking methotrexate should limit their alcohol intake to well within the UK national recommendations”… whatever that means. The American College of Rheumatology offers similar non-specific guidance. When I was training, some mentors would have a strict “no alcohol” policy and some would say, a glass of wine or two per week should be okay.
The aim of the recent study was to quantify the risk of alcohol consumption on hepatotoxicity (liver damage) in a contemporary group of methotrexate users with rheumatoid arthritis, in a large national primary care database.
The researchers recruited people from the Clinical Practice Research Datalink (CPRD) which is a large database of information gathered by primary care providers. They looked at all patients identified as having rheumatoid arthritis and starting methotrexate after 1987 up until Feb 2016. People were included if their liver enzymes were measured at least six times per year.
The researchers want to measure any episode of liver inflammation (transaminitis) defined as a level that was three times the upper limit of normal. This was called the primary definition of transaminitis. Since we know that persistently elevated liver inflammation can lead to liver fibrosis, the researchers were also interested in identifying people that that have three back-to-back levels that were above the upper limit of normal. This was called the secondary definition of transaminitis.
Alcohol consumption was measured first by seeing whether the person drinks alcohol, yes or no. Then the person’s alcohol consumption was categorized:
- Mild: 1-7 units per week
- Moderate: 8-14 units per week
- Moderate-high: 15-21 units per week
- High: > 21 units per week
The researchers identified 44 586 people but only included 11 839 people in the study. The people that were excluded typically were a little younger, were female, and tended to drink no alcohol or very little alcohol.
Using the primary definition of transaminitis, there were 530 first episodes in 47 090 person-years. That’s about 11.26 per 1000 person-years. This rate was similar between drinkers of alcohol and non-drinkers of alcohol. There was no increased risk in the occurrence of transaminitis in drinkers compared with non-drinkers. But this is pulling all the data together. What about people who drank mildly vs moderately vs high?
After analyzing the data ever further, the researchers found, unsurprising, that the rate of transaminitis increased with increasing levels of alcohol consumption. Drinking mild to moderate amounts of alcohol was not associated with transaminitis. Drinking more than 21 units of alcohol per week tended to be associated with transaminitis.
Alcohol consumption below 14 units per week was associated with a very low probability (0.93%) of having clinically important risk of transaminitis. Alcohol consumption in excess of 14 units per week was associated with increasing risk of transaminitis. More specifically, the risk was 33% with moderate-high consumption (15 – 21 units weekly) and 81% with high (>21 units weekly) consumption.
When the researchers used the secondary definition of transaminitis, i.e., 3 or more consecutive episodes of increased liver enzymes above the upper limit of normal, they found similar data: Mild = 0.01%, moderate-high = 8%, and high 17%.
No study is perfect.
First, the data came from a primary care database. This means, it was the PCP or rather the general practitioner that was responsible for coding the diagnosis appropriately, not a rheumatologist. That being said, some misclassification may have occurred inadvertently.
Another issue was that the amount of amount of alcohol consumption was self-reported. Most people tend to under report their alcohol consumption, so we can kind of assume that the levels of alcohol reported by the people in the study were actually a lot more than stated.
Another limitation included the fact that many people were excluded because their liver enzymes were measured less than 6 times per year. Things get a little muddled up here. People that have their blood tested less often statistically have a decreased chance of having an abnormal test simply because they get tested less. Then again, having your blood tested 6 times a year is a bit much, unless that person has risk factors that put them at increased risk of developing liver problems.
An important limitation is the fact that the dose of methotrexate was not included. There could be an increased risk of hepatotoxicity if someone were to take 25 mg of methotrexate weekly versus someone who takes 10 mg.
It’s also unclear whether these results are generalizable to other autoimmune diseases. Methotrexate is also used to treat psoriasis and psoriatic arthritis. The study only included people with rheumatoid arthritis. People with rheumatoid arthritis and psoriasis are not the same. People with psoriasis tend to have more liver problems in general. For example, they tend to develop fatty liver. So the risk of hepatotoxicity with alcohol could be different.
Lastly, some people say that measuring liver function tests (AST and ALT) may be insufficient to actually assess long-term damage from methotrexate because some people develop liver fibrosis without having transaminitis. The problem with the studies that look at methotrexate and liver fibrosis are for the most part, dated and most looked at people with psoriasis. As I mentioned, people with psoriasis tend to have more liver problems than people with rheumatoid arthritis. It’s also important to note that since the 80’s, we’ve changed the way we prescribe methotrexate as well as changed the way we check labs. Measuring liver function tests, NOT performing serial liver biopsies to monitor methotrexate toxicity, remains current best practice.
On a side note, liver function tests is kind of misnomer because the AST and ALT actually don’t measure liver function. They measure liver inflammation.
According to this study modest amounts of alcohol consumption when taking methotrexate may not be as harmful as once though. I would like to remind you that the information provided today does not constitute medical advice. Please talk to your doctor. Everyone’s health is unique. Please leave comments below!
- Humphreys JH, Warner A, Costello R, Lunt M, Verstappen SM, Dixon WG. Quantifying the hepatotoxic risk of alcohol consumption in patients with rheumatoid arthritis taking methotrexate. Ann Rheum Dis.2017 Mar 23. pii: annrheumdis-2016-210629. doi: 10.1136/annrheumdis-2016-210629. [Epub ahead of print]