This was the last session I attended and concerns a topic of great interest to me, so I’m glad I was able to attend. I expect that this is going to turn into a long post, because it concerns some issues which are close to my heart and that’s I’d like to explore in a little detail.
At issue is whether glucosamine and chondroitin are effective in treating osteoarthritis (OA), especially OA of the knee which is the most common site that OA hits. There are two separate issues here: 1) whether they help pain; and 2) whether they help slow or stop the progression of the disease. The rheumatology community is remarkably divided on this issue. Many think glucosamine/chondroitin are probably quite effective and recommend them to their patients, while others think that they are not at all effective. Hence the goal of this session was to present the evidence for both sides and to try to let the rheumatology community decide for themselves. I’m not sure if it was effective in this. It would be interesting to see whether any practitioners actually changed their minds about G/CS utility, or whether those who were unsure finally came down on one side or the other. (Hmm … that might be an interesting study …)
Here’s my disclosure: I think glucosamine and chondroitin are placebos. I think the clinical evidence that they work is weak, and we’ll discuss this later. However, I acknowledge that many people feel that they work well for them, and since they appear to be entirely safe, I have no difficultly if my patients take them, as long as they have the money to spend on placebo tablets. I have no financial interest in any company that makes G/CS or in any of their competitors.
So this session was presented in 4 parts: the 1st was a review of the basic science of glucosamine by Roland Moskowitz from Case-Western Reserve University. The 2nd part was a review of the GAIT trial (I can’t remember what the acronym stands for) – the major NIH-funded study of the two agents that was reported in the New England Journal of Medicine last year. It was presented by Daniel Clegg, the study’s primary author. The 3rd was the argument for G/CS’s efficacy and was given by Jason Theodosakis, author of The Arthritis Cure and a physician who also has a degree in public health, which means he should have training in epidemiology. Finally, the 4th part was the argument against G/CS and was given by David Felson, a prominent researcher in osteoarthritis, and a leading figure in clinical epidemiology.
Moskowitz tried to give us an overview of some of the basic science evidence for and against the mechanism of G/CS’s actions. He pointed out that there are pharmacokinetic studies of glucosamine that demonstrate that it is well absorbed from the GI tract, that it is measurable in the blood after absorption, and that it makes it into the joint. I should point out that glucosamine and chondroitin are elements of the underlying molecules and tissues that make up cartilage and joint fluid, and that the original thought was that they supplemented these elements, stimulated the production of other elements essential for joint strcture and function (proteoglycanas and type II collagen), and therefore helped to slow the loss of cartilage and stimulated its repair. I’m not sure why this would lead to less pain but that’s the idea.
Anyway, it makes it into the joint and into the cartilage. He did not mention, however, that glucosamine is also made by the human body and in much higher levels than are found in the oral forms. I think there is good reason to believe (and some evidence to support) that the amount of glucosamine in a pill that makes it into the blood is negligible compared to what the body makes every day anyway.
Moskowitz also pointed out that there is another theoretical method of action for both these substances. Some studies have suggested that they act as anti-inflammatory drugs, reducing IL-1 production (remember? It’s a mediator of inflammation) and inhibiting prostaglandin production (another inflammatory mediator – the major substance that non-steroidal drugs like ibuprofen inhibit). To my mind, this explanation appeared when the first explanation of G/CS’s action started to be less tenable, which really makes me wonder whether it’s not an example of finding another horse to ride when the first one looks like it’s about to collapse.
He looked at the formulations of these substances. Glucosamine comes in two major forms: the hydrochloride and sulfate ‘salts’ (I guess they’re not really salts in the strict sense of the term). He noted that most glucosamine sulfate is completely unstable which makes many formulations of it of doubtful efficacy – if it breaks down as soon as it gets into the GI tract it’s unlikely to do much good. However, there are other forms which have been ‘stabilized’ and therefore may be effective. The hydrochloride form is apparently more stable, but many authors are now suggesting that the sulfate moiety is essential for the proper function of glucosamine, and that therefore this hydrochloride form is inferior. (I think this is another convenient post-hoc explanation for the fact that it’s been demonstrated not to work in several trials, but more about that later.)
Chondroitin comes in two forms too: they all are sulfated (not salts, but through covalent bonds), but one is sulfated at the 4th carbon, while the other is sulfated at the 6th carbon. I’m not aware of any data showing that one is more effective than the other, but I don’t know the literature about that.
He then went quickly through some trial evidence showing there efficacy but did not dwell on this as it was to be discussed in more detail later in the session, but he did emphasize that the safety profile of this drug appears to be excellent. I agree with this.
Clegg then gave us an overview on the status of ‘nutraceuticals’ in the US, and how the GAIT investigators came to include the formulations that they did in their trial. Mea clupa on this one: I failed to take notes, not realizing that he didn’t provide copies of his slides in the syllabus, so I don’t remember everything he talked about and the points he made. But I do remember a few.
For starters, he pointed out that glucosamine and chondroitin are not classified as drugs in the United States, but as dietary supplements. Thus they are not regulated by the FDA, who requires manufacturers to show efficacy and safety of a drug before they are allowed to market it. Therefore, trials of ‘nutraceuticals’ have not generally involved any FDA oversight, unlike almost every drug that is brought to market.
Second, this means that manufacturers of these products are not subject to the strict regulations that govern drug manufacturers. Dietary supplements are not required to be ‘pure’ and manufacturers are not required to show that formulations contain a specified amount of product. Thus, most (though not all) of these products are of unknown quality and purity.
For the GAIT study, the NIH required that the investigators ensure that all products have demonstrable purity and contained what they were supposed to contain. This was the major reason they chose to use the products that they finally used in the trial (glucosamine hydrochloride and the specific chondroitin product).
Finally, you need to know the results of the GAIT trial. GAIT found that neither glucosamine, chondroitin, nor the combination of the two produced results that were significantly different from placebo. They didn’t work. However, it didn’t end debate on these substances for two big reasons (or at least these are the two reasons I’ve heard most frequently).
First, there was a very high placebo response. 60% of patients who were randomized to placebo had what was defined as a clinically meaning response. This has perplexed many people who feel that this response was much higher than what was reasonable and make them doubt the results of the trial. Related to this, the persons randomized to a drug that most certainly should have been effective for pain, celecoxib (Celebrex), responded 70% of the time. To many this appears to be a very marginal response to a drug with known efficacy. They expected a much higher rate of response compared to placebo.
Second, the combination of glucosamine and chondroitin was ‘shown’ to be effective in one subgroup of patients when compared to placebo: those with more severe pain. This has been widely touted in the lay press and by some supporters of glucosamine. The reasoning is that even though each alone may be marginally effective, they are ‘synergistic’ with each other and have a greater effect when taken together. This effect is more noticeable in those with severe pain than in those with mild pain. We’ll come back to this argument in part 4.
But that’s the basic gist of Dr. Clegg’s presentation. This post has turned out to be quite long, so parts 3 and 4, where the real debate is, will be presented separately in 2 other posts.