Which NSAID is the Best for Osteoarthritis?

A new study might have found the answer

Running can help ease the symptoms of hip and knee osteoarthritis.
Running can help ease the symptoms of hip and knee osteoarthritis. Rick Gomez/Getty Images

            Non-steroidal anti-inflammatory drugs (NSAIDs) are one of the mainstays of treatment for osteoarthritis of the hips and knees. NSAIDs work by blocking a molecule, called cyclooxygenase that in turn blocks the production of a number of signaling molecules called prostaglandins. Prostaglandins are responsible for a number of functions in our body, one of which is inflammation. By blocking the synthesis of prostaglandins NSAIDs decrease inflammation in our body, and with that comes a significant decrease in pain.

Prostaglandins are also responsible for a few other functions in the body such as up-keep of the stomach lining, and thus some of the negative effects of NSAIDs come from blocking those functions as well; specifically NSAIDs can be hard on the stomach, and prolonged use of NSAIDs can cause stomach or duodenal (small intestine) ulcers. These drugs have been well studied and are commonly prescribed for people with osteoarthritis. The use of NSAIDs in this setting is one of the few strong positive recommendations that the American Academy of Orthopaedic Surgeons (AAOS) make in their guidelines of the treatment of knee osteoarthritis.

Physicians commonly suggest NSAIDs as the first line treatment for OA. There are a number of drugs under this drug class the most popular of which are aspirin, ibuprofen (Advil), and naproxen (Aleve). Acetaminophen also known as paracetamol or Tylenol is a drug that is very close to the NSAID class, is often used for similar purposes, but does not have strong anti-inflammatory properties and is thus usually not considered strictly part of the NSAID class.

While we have a lot of data that support the use of NSADs in the setting of OA, we have very little data to tell us, which NSAID is more effective than the others. That is until a recent review that came in the Lancet this March. Before we go further, lets briefly talk about the source of this research.

The Lancet is a medical journal based in the United Kingdom that has been around since the 1820s, is one of the most respected journals, and has been the platform for some of the highest impact most ground breaking articles in the history of medicine.

As an example of some of the ground breaking articles that have been published first in the Lancet are Lister’s publication in 1870 on the effect of sterile technique on surgical outcomes which was the very first publication that led surgeons to sterilizing their hands and all tools before surgery. The lancet also Ronald Ross publishes his discovery that mosquitos transmit malaria, McBride publishes that thalidomide (a once popular antinausea medication used in pregnancy) causes serious birth defects.

            The Lancet publishes high quality articles and in their more recent history have undertaken the task of publishing high quality reviews where they integrate the data from multiple studies in order to try to answer a clinically important question.

In their last issue, the journal attempted to answer the question “Which NSAID is the best at treating pain related to osteoarthritis of the knee and hip?”

            Let’s stop here for one second and make a very important announcement. This study DID NOT EVALUATE SAFETY. So none of this data pertains to side effect profiles or how harmful these drugs are, they only looked at the effectiveness of pain relief.

            The study found 8,973 reports, of which 74 were randomized controlled studies comparing seven different NSAIDs and Tylenol. This is by far the highest quality evidence to be collected on the topic. Between all these trials there were almost 60,000 people treated with various NSAIDs and various doses for pain. They then performed a meta-analysis, which pools all of these patients into one group to see which of the NSAIDs and dosages used was most effective in improving pain. While this is an over simplification of their methods, the details of the methodology is beyond the scope of this article.

            Out of the 22 drug-dose combinations examined, 5 did not perform any better than placebo (a sugar-pill). Those 5 were Tylenol at less than 2 grams and less than 3 grams daily dosing, diclofenac (Voltaren) at 70mg a day, Naproxen at 750mg per day, and ibuprofen at 1200mg a day. It is important to note that these are dose specific outcomes, and higher doses of these medications in some cases proved effective. Six interventions stood out as most effective: diclofenac (aka Voltaren) 150 mg/day, etoricoxib (aka Arcoxia) 30 mg/day, 60 mg/day, and 90 mg/day, and rofecoxib  (Vioxx) 25 mg/day and 50 mg/day.

            More sophisticated statistical testing showed that diclofenac 150 mg/day and etoricoxib 60 mg/day performed the best in terms of pain relief when comparing to all other drugs and dosages. When comparing the different NSAIDs in terms of their effect on improving function, it looked like diclofenac (Voltaren) 150mg/day and rofecoxib (Vioxx) 25mg / day.

            Before we proceed further we should have a brief focus on Vioxx (rofecoxib). While in this trial it was shown to be an effective drug for pain relief and improvements in function, Vioxx is not a safe drug. The FDA has removed it from the market over concerns for increased risk of heart attacks and stroke associated with long term use. It’s effective as a pain reliever is irrelevant due to its dangerous side effects. It should not be considered as a viable option.

            So what does it all mean? This is a whole lot of data to sort through and interpret. In my opinion there are a couple of take home points from this very high quality meta-analysis. First, paracetamol (also known as acetaminophen or Tylenol) may be less effective at controlling pain than we thought. Now it’s important to note that this study is specific to osteoarthritis of the hip and knee. So maybe Tylenol is great at controlling pain in other settings, like headaches, this study doesn’t address that. As far as OA of the hips and knees goes, it looks like Tylenol is a poor choice. Diclofenac at 150mg/day was the most effective at controlling pain and improving function.

            This looked at outcomes over short and mid-term time frames of about 3 months on average. Would these drugs performed differently over the longer term such as a number of years? Possibly, we would need more long-term studies to say for sure. The hardest thing about interpreting this data is that it does not address the safety profile of these drugs. NSAIDs have been shown to increase the risk of gastrointestinal bleeding (stomach and small intestine), as well as increase the risk of heart attacks. And the degree to which each NSAID, and each NSAID dosing, raises those risks is potentially different. The decision to start a new medication for the treatment of OA of the hip and knee must be carefully discussed with your physician, so that the two of you can weight the pros and cons of each drug individually and choose the solution that is right for you.  

Continue Reading