Just one more nail in the lid that’s never coming off the coffin of glucosamine and chondroitin. Despite the ubiquity of these supplements and the innumerable testimonials for their benefits in people and in pets, the research continues to accumulate that they are nothing more than a placebo when it comes to treating arthritis pain. A new meta-analysis in the British Medical Journal looking at 10 trials with over 3800 patients followed for up to 2 years found no clinically significant affect of glucosamine or chondroitin supplementation on pain scores of joint width (a measure on cartilage degradation). Not surprisingly, industry-funded trials tended to have more positive results than independently-funded trials.

Of course, people will argue that this is not directly relevant to the use of these supplements in dogs and cats, and we should not discount their potential value in these species without similar high-quality studies. There is some truth to this, but given the limitations in resources and technical difficulties of veterinary research, the absence of any plausible reason to think that these products might perform better in animals than in humans, and the uninspiring results of the non-industry funded veterinary research on glucosamine and chondroitin so far, it would seem to be placing our bets on the long shot to put much more of our limited resources into studying these therapies.

Then again, it’s hard to find a dog, cat, or horse that isn’t already on glucosamine and chondroitin, so a cynic might ask whether the evidence really matters or not anyway. Luckily, I am an inexhaustible well of optimism about human nature, so I would never be so cynical. 🙂

In January, a couple of research studies were published in the Journal of the American Veterinary Medical Association (JAVMA) investigating the possible usefulness of dietary fish oil supplementation for treating the symptoms of arthritis in dogs.[1-2] I have previously discussed these papers in detail, but overall they were well-designed studies with mostly negative findings. However, the few statistically significant measures seen were generously interpreted as evidence for a beneficial effect of the treatment. A more accurate interpretation would be that no clear, consistent, meaningful effect was observed. Still, the few effects that were documented would justify additional research. Ideally, this research would be conducted by different researchers since the group naturally has a bias in favor of the intervention they are studying, and a number of the authors of the JAVMA studies have institutional affiliations with the Hill’s Pet Food company which manufactures the diet with supplemental fish oil tested in the study, and this raises some possibility of a conflict of interest. 

However, the same research group has now published another study looking at fish oils as a treatment for canine arthritis, this time in the Journal of Veterinary Internal Medicine.[3] This study was specifically designed to look for effects of varying levels of fish oil supplementation on subjective measures of arthritis.

A Bad Beginning
My first concern about this study is the background statement and introduction, which begin with the claim that “Food supplemented with fish oil improves clinical signs and weight bearing in dogs with osteoarthritis (OA).” The evidentiary support for this claim is twofold. First, it is supported by reference to findings in humans that fish oils may improve the symptoms of rheumatoid arthritis. There is some limited evidence that this may be the case (c.f. literature summaries available from the Agency for Healthcare Research and Quality, and Medline), however even if this were established, the relevance is questionable since this is a different disease in a different species. The authors acknowledge that “the pathophysiology of rheumatoid arthritis and OA differ,” but they let stand the implication that since both are inflammatory conditions, if fish oil helps one it might help the other. Plausible but very weak reasoning.

The second major pillar of support for the a priori claim that fish oils are proven effective as a treatment for canine arthritis is, you guessed it, the two previous papers published by the same group in JAVMA. Obviously, given my review of these papers, I don’t consider that claim to be justified by those studies. So this study begins with the assumption the treatment works, and the hypothesis is then examined that since some helps more might help even more.  Shaky ground to start from.

The Study
As with the previous work, the study was soundly designed. 212 dogs were randomly assigned to eat one of three diets with different levels of fish oils but relatively comparable otherwise. The diet with the lowest level of fish oil (Diet A) was the same as used in the previous studies, approximately 0.8% EPA+DHA on a dry matter basis. Diet B had ~ 2% of the fish oils, and Diet C about 2.9%. There was no negative control (diet without supplemental fish oils).

Subjects were allowed to use concurrent therapies, including NSAIDs and glucosamine, but no overall statistically significant difference in the rate of such usage was found between the groups of patients on the different diets. Dropout rates were moderate and higher for Food A than the other two foods (Food A-22%, Food B-13%, Food C-17%). The authors state that the inclusion evaluation data from dogs that did not complete the study did not affect the results. Owners and participating veterinarians were blinded to the treatment individual dogs were assigned to.

The assessment measures included a series of subjective physical examination parameters assessed by participating veterinarians similar to those used by the participating veterinarians in the previous studies. These were performed on Days 0, 21, 45, and 90 of the study. Two new variables were included, which were general questions the participating vets were asked at the conclusion of the study:

A. “Based on the pet owner’s input and your own clinical assessment, how would you describe the impact the study food has had on this dog’s arthritic condition?”
1. significantly slowed progression of the arthritis
2. slightly slowed progression of the arthritis
3. no effect
4. slightly accelerated progression of the arthritis
5. significantly progression of the arthritis

B. “Based on the pet owner’s input and your own clinical assessment, how has the arthritic condition of this dog changed since eating the study food?”
1. extreme improvement
2. moderate improvement
3. slight improvement
4. no change
5. slight deterioration
6. moderate deterioration
7. extreme deterioration

The authors also measured the blood levels of a variety of omega-3 and omega-6 fatty acids for all subjects.

The Results
Serum levels of omega 3 fatty acids increased significantly more for Foods B and C than for Food A, showing that greater levels in the food correlates with greater levels in the actual animal. Some omega 6 fatty acid levels were significantly lower in subjects eating Foods B and C, others were not.

In terms of the 5 clinical evaluations done by participating veterinarians, all improved significantly over time for all diets except that the score for weight bearing did not improve for Food A. Of the 5 measures, 2 improved more for Food C than for Food A, and there was no difference between Food B and Food A.

For the two subjective questions asked of the participating veterinarians, the answers showed a statistically greater improvement for Food C than for Food A, and again no difference between Foods A and B.

Sixteen adverse events were recorded, and of these 2 cases of vomiting and/or diarrhea were attributed to the intervention (one for Food A and one for Food B). What the other events were and why they were not considered related to the diets was not discussed. 

Discussion
So, do fish oils effectively treat the symptoms of canine osteoarthritis? Maybe. Improvements were seen for all foods. The measures were subjective, and the two summary questions asked of veterinarians in the study were structured in a redundant and somewhat leading way, so there are some limitations in the reliability of the conclusions based on these. Interestingly, the authors claim these results are consistent with their previous studies, but this is not entirely true. In one of the earlier studies, no veterinarian clinical examination measures improved, and only 3 out of 39 possible data points for owner evaluation improved significantly. In the other study, none of the owner evaluations changed significantly, but 3 of 5 measures evaluated by clinicians showed improvement. An objective measure, force plate gait analysis, was used in this second study, and this showed no change from baseline for any of 6 variables but a significant mean improvement for 1 of the 6 measures.

By contrast, in the current study only subjective veterinarian evaluations were used as measures of outcome, and these showed improvements in all variables for all foods. The authors point out that these improvements were “relatively modest” (aka small), but rather than attributing this to the treatment having only a modest effect, they attribute it to the insensitivity of their subjective measurement criteria. This is a bit odd since in medical research generally, subjective criteria are much more likely to show positive changes than objective criteria, even when there aren’t any real differences.

The inconsistency in the results of the various trials is a cause for concern, as it suggests the criteria for measuring any possible treatment effect may not be reliable. The trials designed to see if the treatment worked found few and inconsistent positive changes in subjective measures and almost no change in objective measures. The trial designed to see if more worked better than less found that everything worked a little, but while the difference between the diet with 0.8% fatty acid content and that with 2% was not significant for any measure, the difference between 0.8% and 2.9% was significant for 2 of 5 clinical measures and both global assessment questions asked of the participating vets. And all of this is ignoring the potential biases introduced by having all trials done by the same research group and the association with a company manufacturing a commercial food used in all of the trials.

This does not create a solid, consistent pattern of evidence to support the use of fish oils to treat canine arthritis. At best, it can be said that: the serum fatty acid level data indicates the fish oil changes the fatty acid profile of dogs in ways that would be desirable if the fish oil actually has clinically meaningful effects;  there is weak evidence to suggest there may be some such effects, though small and inconsistent; and there is little evidence of short-term harm.

All of this is useful data, of course, and generally the studies were well-conducted. However, the conclusions tend to be much stronger than is justified by the data, which is a reflection of the bias of the investigators in favor of the intervention they are researching. This is to be expected, but if a small group of researchers become the only source of clinical data on the subject, and if most vets read only the conclusions and don’t critically examine the studies and the results, it is likely to have an inappropriately large impact on clinical practices. I wouldn’t be surprised if fish oil for arthritis becomes the new glucosamine, and while I hope it turn out to work better than glucosamine does, that product had much the same kind of weak positive data from limited sources at the beginning as well, and now it has become a ubiquitous product despite subsequently stronger results suggesting it has no benefit.

References
1. Roush JK, Dodd CE, Fritsch DA, Allen TA, Jewell DE, et al. Multicenter veterinary practice assessment of the effects of omega-3 fatty acids on osteoarthritis in dogs. J Am Vet Med Assoc. 2010 Jan 1;236(1):59-66.

2. Roush JK, Cross AR, Renberg WC, Dodd CE, Sixby KA,, et al. Evaluation of the effects of dietary supplementation with fish oil omega-3 fatty acids on weight bearing in dogs with osteoarthritis. J Am Vet Med Assoc. 2010 Jan 1;236(1):63-73.

3. Fritsch D, Allen TA, Dodd CE, Jewell DE, Sixby PS, et al. Dose-titration effects of fish oil in osteoarthritis dogs. J Vet Int Med 2010;24(5):1020-26.

Few purported medical interventions have such wide appeal or such an entrenched reputation for being benign and beneficial as vitamin supplements. Vitamins are chemicals (yes, they are, though you can call them “compounds” or even “essential nutrients” if it sounds nicer) that are required for normal health in minute quantities and cannot be made internally but must be obtained from the diet. Different species have different vitamin requirements, of course, so the defining characteristic is dependent on the organism, not the chemical. Vitamin C, for example, is a dietary requirement for primates and guinea pigs, but all other mammals can make enough on their own from other chemicals that they don’t require it in the diet, so it is not truly a vitamin for them.

Part of the reason vitamins are so widely believed to be good for us is the vague memory (at least in developed nations, though it is still ordinary reality in many parts of the world) of a time when vitamin deficiencies were common due to inadequate diets. Correcting such deficiencies has the kind of tangible, dramatic impact on health that antibiotics or polio vaccine can have, so it is easy to see such things as miraculous. And in America, where more is always better, the idea that it is a good idea to give vitamin supplements even to people with an adequate diet and no obvious signs of deficiencies is a naturally appealing one.

Unfortunately, the grim naysayers of science such as myself are often stuck with the thankless job of dispelling even appealing misconceptions, and recent editorial in the American Journal of Epidemiology has done that for some the proponents of vitamin supplements most hoped to be of benefit in preventing one of the major classes of illness that has replaced nutritional deficiencies and infectious diseases, cancer.

Byers T. Anticancer Vitamins du Jour—The ABCED’s So Far. American Journal of Epidemiology 2010;172:1–3.

The editorial surveys the history of anti-cancer vitamins from early reason for hope seen in observational and animal model studies through disappointing and often frightening clinical trials. In alphabetical order, then (citations are omitted and can be found in the full text of the article):

Vitamin A

Animal experimental models led us to the notion that cancer risk might be ‘‘materially’’ reduced by supplementation with beta-carotene, a retinol precursor. Although that idea was seductive, we were all disappointed when 2 large randomized controlled trials that began in 1985 in Finland and the United States reported an 18% increased risk of lung cancer caused by high-dose beta-carotene supplementation and a 28% increased lung cancer risk caused by a combination of beta-carotene and retinol. The vitamin A era was over. 

Vitamin B

Again, based on animal experimental evidence and supported by epidemiologic evidence of connections between diets low in B vitamins and increased cancer risk, a large randomized controlled trial was begun in 1985 in central China, where micronutrient deficiency was common and where rates of cancers of the stomach and esophagus were extraordinarily high. Nonetheless, several years of supplementation with a combination of riboflavin (vitamin B2) and niacin (vitamin B3) had no effect on incidence of upper gastrointestinal cancers. Interest in folic acid (vitamin B9) persisted, though, in part because of its striking effect on neural tube birth defects, coupled with speculation about possible benefits of food fortification for diseases such as colorectal cancer that were inversely associated with diets rich in folate-containing foods and supplements. However, a 7-year randomized controlled trial found that high-dose folic acid supplements actually increased risk of colorectal adenomas. The vitamin B era was over.

Vitamin C

Next came vitamin C, a popular charge led by none other than Linus Pauling, the brilliant and charismatic 2-time Nobel laureate. Of all the cancers thought to be related to vitamin C deficiency, gastric cancer led the way, and of all the places on Earth where a vitamin C deficiency correction trial might yield benefits for gastric cancer, Linxian, China, would be the best. Indeed, vitamin C was tested in the Linxian trial, but just as for the B vitamins, vitamin C produced no change in gastric cancer rates .

Vitamin D

Over 2 decades of searching for an anticancer vitamin, we had seemed to skip over vitamin D in its proper alphabetical sequence…the International Agency for Research on Cancer conducted a comprehensive review of the evidence for vitamin D and cancer prevention, concluding that vitamin D may play a protective role in colorectal cancer, but not for prostate cancer, and that the evidence is weak for breast cancer. The conclusion by the International Agency for Research on Cancer about the weakness of the evidence for breast cancer has been a source of controversy among vitamin D protagonists, but subsequent nested cohort studies have found no relation between breast cancer risk and circulating levels of vitamin D. Nonetheless, vitamin D remains the cancer-preventing vitamin du jour.

…An outstanding set of papers in this issue of the American Journal of Epidemiology reports on findings about the relation between circulating levels of vitamin D and subsequent cancer risk…These studies found no suggestion of an inverse association between vitamin D levels in the circulation and later incidence of 6 types of cancers (upper gastrointestinal, ovary, endometrial, pancreatic, kidney, and non-Hodgkin lymphoma)… The only association observed in this set of 6 analyses was a troubling one: that risk of pancreatic cancer was doubled for those in the highest quintile of circulating vitamin D levels.

…many ongoing randomized controlled trials are now using quite high doses of vitamin D. As we await clearer evidence of benefits from those trials, we will also need to be prepared to be vigilant about their individual and collective power to assess any potential harms.

Vitamin E

In 1993, we launched headlong into a love affair with vitamin E fueled by compelling observations that those who chose to take vitamin E supplements were at lower risk of heart disease. Vitamin E supplementation became the rage as several large, randomized controlled trials were mounted. When those results finally came in, the findings were again disappointing: vitamin E supplementation offered no benefit for heart disease, and it slightly increased overall mortality. In the meantime, though, because of a secondary observation that prostate cancer incidence was lower in the vitamin E arm of the same Finnish trial that tested beta-carotene (vitamin E had also been included as a factor), a large factorial trial of vitamin E (and selenium) was carried out for reducing prostate cancer incidence. Disappointment again: there was no effect of either selenium or vitamin E on incidence of prostate cancer. The vitamin E era ended in a whimper.

Of course, alternative medicine proponents often dispute this evidence, and it is easy to find claims of efficacy for all of these vitamins in preventing or treating cancer: Vitamin A, Vitamin B, Vitamin C, Vitamin D, Vitamin E.

Given the ease with which the evidence can be ignored, cherry-picked, or distorted, the underlying conclusion of this editorial is even more salient and important than the specifics of the failed promise of vitamin supplementation to prevent:

It is timely for us to now reflect on the history of the past 25 years of our alphabetical approach to studying single vitamin deficiency states as causal factors for cancer. We have learned some hard lessons along the alphabetical way. We now know that supernutritional levels of vitamins taken as supplements do not emulate the apparent benefits of diets high in foods that contain those vitamins, and we now know that taking vitamins in supernutritional doses can cause serious harm. In short, we have found that the reality of human biology is far more complex than is suggested by our simple ideas.

Let me emphasize again these key points:

1. Supernutritional levels of vitamins taken as supplements do not emulate the apparent benefits of diets high in foods that contain those vitamins.

 2. Taking vitamins in supernutritional doses can cause serious harm.

 3. The reality of human biology is far more complex than is suggested by our simple ideas.

Point 3 is one of the most important principles of science-based medicine. The understandable search for simple answers and the inevitability and challenges posed by complexity and uncertainty naturally drive us to accept simple, clear explanations and recommendations and make us shy away from the messiness and limitations of reality. But unfortunately we must accept that reality is more complex than our ideas, and that as a consequence we will often be wrong and our understanding and ability to control our health will always be limited and incomplete. The best chance we have of actually doing good is to strive for this acceptance and to deal with reality as it is rather than as we would wish it to be.

Alternative medicine often relies on the idea that belief is enough and that reality can be forced into our simple models of one cause/one cure for disease. Science can, and should do better, and though the process is slow, cumbersome, and often not psychologically comforting, the truth is that it just works better. 25 years of scientific research has shown us what supplementation of some vitamins can and cannot do for some diseases, and this information is worth the wait and the effort, and it will serve us better than the false hope and comfort of simplistic theories and wishful thinking.