The SkeptVet Blog

The following is a post I contributed to the Science-Based Medicine blog.

On this site there have been several thoughtful posts (e.g. by Dr. Atwood and by Dr. Novella), and subsequently much heated commentary, on the distinction between Evidence-Based Medicine (EBM) and Science-Based Medicine (SBM). I agree wholeheartedly with the position that the two are not mutually exclusive, and that SBM is essentially EBM as it should be practiced, with a comprehensive consideration of all relevant evidence, including the subject of plausibility. As a practicing veterinarian, and an officer of the Evidence-Based Veterinary Medicine Association (EBVMA), I am keenly interested in bringing to my profession a greater reliance on high quality research evidence and sound scientific judgment, and reducing the reliance on individual practitioner intuition and experience in making clinical decisions. However, those of us in veterinary medicine face some special challenges which make the subtle but important distinction between EBM and SBM especially salient. 

Where’s the Evidence?

The first of these challenges is the paucity of high quality clinical research evidence. As an example, in his 2007 book Snake Oil Science, R. Barker Bausell examined the research evidence concerning the use of glucosamine as a treatment for osteoarthritis in humans. He was able to analyze the strengths and weaknesses of a Cochrane Review which included 20 studies with 2570 patients (the most recent revision of this review includes 25 studies with 4963 patients), a NEJM study with 1583 patients, and an Annals of Internal Medicine study with 222 patients treated for two years. His conclusion was that the intervention was not more effective than placebo.

I recently did a targeted search of the PubMed literature database for a brief evidence-based medicine feature on the subject of glucosamine and chondroitin as treatment for osteoarthritis in dogs, currently in press at the Journal of the American Veterinary Medical Association. A search of the terms “glucosamine,” “arthritis,” and “dog” yielded eight references, of which three were relevant (a more comprehensive search strategy yielded sixteen references, but only the same three were relevant to the clinical question). The three useful references included two clinical trials involving a total of 113 dogs and each lasting about 2 months, and a systematic review of treatments for canine osteoarthritis which evaluated one of these two clinical studies. Predictably, the larger, better designed trial with objective measurement criteria showed no benefit of glucosamine, while the smaller, less well-controlled trial with only subjective criteria and a 23% dropout rate in the glucosamine group showed some benefit at some assessment points.

Where’s the Money?

Glucosamine is an extremely popular, and profitable, supplement routinely recommended by veterinarians and administered by owners to their geriatric dogs. Yet the clinical trial evidence concerning its effects is nearly non-existent. The depth of the evidence is no better for many, many routine clinical interventions in veterinary medicine. The primary reason for this is simple: money. 

Obviously, the health of companion animals is not as high a societal priority as human health. Many countries have little or no formal companion animal medicine at all, of course, much less high quality, evidence-based pet medicine. And even in the developed world, the absolute size of the veterinary medical profession and associated industries is dwarfed by that of the human medical industry. 

In the United States, surveys show that most dog and cat owners have come to consider their pets to be members of their family, and their willingness to pay for veterinary care has increased along with this shift in attitude. The same appears to be the case in Europe and other developed nations. This has allowed the quality and technological sophistication of veterinary care to increase. 

Pharmaceutical companies have followed this trend, increasing their financial investment in their own internal research activities, as well as funding the lion’s share of companion animal health research generally (with all the ethical and practical problems that creates). Pfizer, the largest fish in the “Big Pharma” pond, claims to spend $300 million annually on veterinary research globally, for both companion and agricultural animals. However, the company is expected to spend $9-$9.6 billion this year on its human research and development. The same pattern is true of government research spending. Veterinary medicine will always be the poor stepchild of medicine, and we cannot expect to have anything close to the quantity or quality of research evidence available to MDs trying to practice evidence-based medicine. 

A Pack of Lone Wolves?

Another barrier to effective utilization of research evidence in veterinary medicine may be demographic and cultural. In the United States, the average veterinary practice has fewer than three veterinarians, and between one-third and one-half of veterinarians are self-employed practice owners. And most companion animal veterinarians are general practitioners, only about 10-15% of practicing vets being board-certified, with the extended academic training and, hopefully, greater awareness of and respect for research evidence that might be expected to come from this training.

As a profession, we veterinarians tend to be entrepreneurial, self-reliant, and independent. This contributes to a reluctance to let anyone tell us what to do, which may be how veterinarians perceive the position of evidence-based medicine. There is no solid data on the subject (though I am involved in a survey study which will hopefully provide some soon), but in discussions with colleagues I have sensed a great deal of anxiety about the notion of “cookbook medicine” which disdains the hard-won wisdom and experience of the individual clinician. Veterinarians are reluctant to accept the idea that there may be broadly applicable standards of care they ought to adhere to, even if their personal judgment conflicts with the evidence for these.

Undoubtedly, our colleagues in human medicine share a similar temperament and similar sorts of anxieties about “cookbook medicine”. However, these may be tempered to some extent by more widespread advanced training, more structured and supervised practice environments, and greater assessment and monitoring of outcomes, which may partially explain the greater acceptance of EBM in the human medical profession. And though the case of Dr. Rolando Arafiles, Jr. illustrates the weaknesses in the systems for monitoring physician behavior, I think it is clear that the influence of government regulation, and the threat of litigation, give the concept of adhering to a recognized medical standard of care far greater teeth in the field of human medicine than it has in veterinary medicine.

SBM, EBM, or OBM?

So how does this relate to the difference between SBM and EBM? Well, traditionally the scarcity of clinical trial evidence has led veterinarians to practice primarily opinion-based medicine. Personal experience and intuition and the opinions of individual experts or mentors are the predominant foundations for clinical decision-making. There is little or no outcome assessment, so veterinarians must rely on their own clinical experience to judge whether their practices are effective. 

The negative consequences of these strategies are many. There is dramatic inconsistency in the diagnosis and treatment of even common diseases. I routinely have to explain to my clients that if they ask ten vets a question, they are likely to get seven or eight different answers. You can imagine how frustrating this is for them, and how little confidence it inspires in our expertise.

OBM Leads Kids to the Hard Stuff, FBM!

As most readers of this blog likely already know, there are many reasons why individual judgment is an unreliable guide to the true efficacy of a medical intervention, and why we should be reticent to entirely trust our own intuitions and experience. But opinion-based medicine is also a “gateway drug” to faith-based medicine, otherwise known as complementary or alternative medicine. If you are accustomed to judging the safety and efficacy of interventions on the basis of the cases you have seen personally or the opinions of “experts,” you are more likely to be persuaded by the individual experiences of clinicians promoting and alternative practice, and more likely to think that giving it a try yourself is the most reliable way to know if it really works or not. 

The Internet abounds with holistic veterinarians who claim they started their careers as scientific, skeptical doctors but that their frustration with the limitations of mainstream medicine and the problems they could not solve led them to experiment with, and ultimately become promoters of, faith-based miracle therapies of every kind that share no theoretical or practical features in common other than being validated primarily by testimonial and not consistent with scientific knowledge or evidence. 

Tooth Fairy Science exists in veterinary medicine, but it is less of a problem than the simple lack of research evidence and the consequent reliance on even less trustworthy forms of evidence. So veterinary medicine needs a science-based approach even more desperately than human medicine because we have so little clinical trial evidence to rely on, and so few resources to generate more and better evidence. The tragedy of money and talent wasted on studying therapies that have vitalist theoretical foundations inconsistent with established scientific knowledge is even more poignant in the relatively impoverished world of companion animal medical research. Plausibility must play an important role in deciding how we allocate the scarce resources we have in order to maximize the useful information we can generate, and the subsequent clinical benefits for our patients.

Towards a One Health Approach

Veterinarians must also take advantage of the evidence that our colleagues in human medicine have generated for us. There are serious dangers in extrapolating research evidence across species, of course, but we cannot afford to entirely ignore the wealth of human medical research that is relevant to our patients. Examined cautiously and judiciously, this data can help us target our own research efforts more efficiently. Just as animal models have an important role, despite their limitations, in human health research, so human clinical research can inform veterinary medicine. As clinicians, we can make more science-based decisions, even when relevant veterinary research is lacking, if we are aware of the research in humans that already exists on the conditions and interventions we are considering. 

If glucosamine is shown to be no more than a placebo after years of research in thousands of people, how much money and effort should we invest in studying its effects in dogs? And how strongly should we promote it to our clients, the vast majority of whom must pay for their pet’s care out of pocket, without insurance coverage, and who commonly must eschew needed care or even euthanize their companions for want of money to pay medical costs?

A Worthy Goal

There has been a steady growth in the quality and sophistication of care available to companion animals in the last several decades, and I am hopeful that this will continue. But I believe the interests of our patients and clients will be best served if the care we provide is as soundly science-based as possible. And while I think evidence-based medicine can become the standard in the veterinary field, with beneficial effects on the quality of the care we provide, we need the additional features of the science-based medicine approach even more than our MD colleagues: a respect for the importance of plausibility in allocating research resources and an understanding of the need to integrate all relevant evidence when making clinical decisions about interventions in the face of a scarcity of high quality clinical trial research. 

Despite all the histrionic accusations of some alternative medicine advocates about mainstream veterinarians being tools of the pharmaceutical industry or reluctant to accept unconventional approaches only out of closed-minded prejudice or a fear for our income, the reality is that we care deeply for our patients and want to provide them with the best care we can. I truly believe, and I hope the profession as a whole will come to accept, that science-based medicine is far more likely to help us do so than the opinion-based medicine we have traditionally relied on.

There is an excellent article in the current issue of the Journal of the American Veterinary Medical Association which I wanted to draw attention to, both for its conttent and the design of the study.

Brown,D.C.; Boston,R.C.; Farrar,J.T. Use of an activity monitor to detect response to treatment in dogs with osteoarthritis. J.Am.Vet.Med.Assoc. 2010, 237, 1, 66-70.

A perennial problem in the evaluation of any therapy for arthritis in animals is the difficulty in assessing effect. Most studies use subjective evaluations by investigators or owners, which are subject to many kinds of bias and almost always show a benefit of every therapy for at least some of the multiple measures of function evaluated. A few studies use force plate analysis, where the dog walks over a device that measures how much weight they place on an arthritic limb. But this is a complex and often difficult measurement to take. The recent article describes a study using a small accelerometer to measure activity at home over a prolonged period. This is potentially a very useful tool in evaluating the efficacy of arthritis treatments in dogs.

The study was very well designed, and the authors clearly recognized the importance of controlling for investigator and owner bias even when using an objective measurement tool. The patients were randomized to treatment either with an agent already clearly established as effective (the NSAID Rimadyl) or an identical placebo, and both investigators and owners were blinded to the group assignment. Confounding factors were controlled for by selecting a study population newly diagnosed and not already treated, and by using statistical techniques to account for factors such as age, weight, and so on.

The results showed an expected increase in activity for the patients given Rimadyl (about 20% greater activity than before treatment) and no change from baseline for the placebo group. Interestingly, some of the dogs in the placebo group did show an increase in activity (20% of the dogs in this group had an increase of 10-30% in activity from the baseline period). As the authors explained, this sort of change likely represents the phenomenon of regression to the mean. For many chronic diseases, including arthritis, the symptoms wax and wane, getting worse and better alternately around an average, or mean, degree of severity. People tend to seek medical treatment, for themselves and their pets, when their symptoms are especially bad, and so the natural course of the disease tends to lead to an improvement in symptoms following an intervention regardless of whether the intervention is actually doing anything. Add to this the tendency for people to behave differently when involved in a medical study, usually exhibiting better health habits than they usually do, and it is routine for studies to show improvement with placebo therapies that aren’t actually doing anything. This is a significant contributor to the so-called “placebo effect.” It also explains why adequate controls are critical to showing real treatment effects, and why poorly controlled studies tend to find effects where they don’t really exist.

I am hopeful that the monitoring device discussed in this article will make it possible to more easily and inexpensively evaluate the effectiveness of the various purported osteoarthritis treatments currently in use as well as new treatments. I am less hopeful that such future studies will be as carefully designed and controlled as this one. Objective measures and all the statistics in the world don’t make experimental results real or useful unless human bias is adequately compensated for through techniques such as random allocation of subjects, blinding of investigators and owners, and placebo controls, or ideally a combination of active and placebo controls as well as no treatment at all. This study is a nice example of how such techniques should be used, as well as a demonstration of why they are needed.

Behavioral problems, especially fear and aggression in dogs and elimination in the house in both dogs and cats, are a major reason for people to give up or euthanize their pets, so they represent a serious medical condition. There are many methods of treating such disorders with behavioral modification, and medications can sometimes be helpful, though our understand of which medications might help for which problems in which patients is very poor. However, the sad truth is that we have very limited success in alleviating many of these problems.

In the face of serious medical problems for which there are not strikingly effective scientific therapies, more questionable methods can become widely used. Some, like Bach flower essences, such as Rescue Remedy, or homeopathic treatments are clearly useless nonsense. Others are plausible scientifically, but not really shown to work in definitive ways. One such treatment is the use of pheromones.

Pheromones are chemicals animals produce that can affect the behavior of other members of the same species. They are believed to be common and important in coordinating social behavior in animals, especially mammals. The role, if any, that they play in human behavior is less clear, though they may be involved in the synchronizing of menstrual cycles in women living together. In any case, it is clearly reasonable that such substances might have an impact on the behavior of dogs and cats and so might be useful in managing behavior problems. Notice all the “mights” in that statement? The devil, of course, is in the details

It has become quite common for veterinarians to recommend use of synthetic pheromone analogs, that is chemicals made to be structurally  natural pheromones, to help treat behavioral problems. The most common products are Feliway, and analogue of the facial pheromone cats leave behind on furniture and people the rub their faces against, and DAP, short for Dog Appeasing Pheromones, a pheromone nursing mothers release which is believed to calm puppies. These are sold with dramatic claims of efficacy for a wide range of conditions, but of course the claims of folks selling the product have to be viewed as perhaps less objective than other forms of evidence.

In the most recent issue of the Journal of the American Veterinary Medical Association, a systematic review was published which evaluated the research evidence for the use of these products.

Frank,D. Beauchamp,G. Palestrini,C. Systematic review of the use of pheromones for treatment of undesirable behavior in cats and dogs. J.Am.Vet.Med.Assoc., 2010, 236, 12, 1308-1316.

The purpose of a systematic review is to evaluate all the published data on the basis of quality, and then evaluate the results of studies that meet a reasonable minimum quality. In this review, studies were excluded that did not meet minimum quality standards or that were conducted by researchers working for the company selling the product under study. This left a total of 14 studies, 7 in cats and 7 in dogs.

In general, as is too often the case in veterinary medicine, the methodological quality of the studies overall was lower than expected in human medicine, with no studies reaching the highest standards. But as the authors wisely note, imperfect information is better than no information, so we must make our judgments based on the best available evidence even if it is prone to errors that better studies would avoid.  I will spare you the interesting but complex details of the various study designs and their strengths and weaknesses, though reading the full review is recommended for anyone interested. Having read through the full report, I am convinced that unlike others I have reviewed, the conclusions of the authors of this systematic review are supported by the methods and results they detail elsewhere in the paper.

Of the 7 cat studies, none provided convincing evidence of a benefit. Some decrease in urine spraying occurred in some cats, but the significance of this was muddied by failure to follow up on cats which dropped out of studies, which were probably cats who didn’t improve with the treatment, and other methodological flaws in the studies reviewed. Pheromone therapy also was not clearly of benefit in cats with interstitial cystitis (a condition in which irritation in the bladder causes symptoms much like those of a bladder infection). The pheromone also did not seem beneficial in calming cats in the hospital or facilitating the stressful process of placing IV catheters in hospitalized cats.

In dogs, one study found some evidence that pheromone therapy might reduce anxiety in puppies during training. Other than that, no convincing evidence of benefit was found for anxiety associated with veterinary care, anxiety in shelter dogs, or barking and elimination indoors by recently adopted dogs.

The limitations in the quality of the evidence mean we cannot definitively declare that pheromone therapy doesn’t work. What we can say is that based on the best evidence to date, it does not appear to have a benefit. Further study is certainly reasonable, but as always I question the wisdom and the ethics of widespread sale and use of products which, despite years of testing, don’t seem to have much evidence that they work. These products appear harmless, and if clients wish to spend money rolling the dice on a treatment that is not well supported by the limited clinical research available that is certainly up to them. But when veterinarians recommend such products, it does tend to convey the impression that they are legitimate, validated therapies, and I think we do a disservice to our clients if we make such recommendations without a clear statement of the limitations in the evidence. Thanks to this paper, it is now easier to make such a statement.

 Apparent Lack of Association between Primary Immune-Mediated Thrombocytopenia and Recent Vaccination in Dogs
A.A. Huang; J. Coe; G.E. Moore; J.C. Scott-Moncrieff

I’ve written about the proposed association of vaccination and immune-mediated diseases, in particular immune-mediated hemolytic anemia (destruction of red blood cells). For that disease, the evidence is contradictory, and in general it seems likely that very rare cases of immune-mediated disease are triggered by vaccination, but this is not a common or widespread problem, and the vast majority of cases occur from natural exposure to triggers in the environment, probably in dogs with a genetic susceptibility since there are breed predilections.

This study looked for an association between vaccination and immune-mediated thromboctopenia (ITP, the destruction of platelets, necessary for blood clotting). Looking back through medical records at dogs diagnosed with ITP and then comparing them to dogs of similar age, sex, and breed with other diseases, the authors did not find that vaccination increased the odds of developing ITP.

Again, all the usual caveats about small, retrospective, case-controlled studies apply, so these results cannot be viewed as the final word on the question. But even this qualified evidence is more reliable than the wild guesses, hunches, and clinical impressions of those who confidently blame vaccines for diseases like ITP.

Systemic Effects of a Commercial Preparation of Chondrotin Sulfate, Hyaluronic Acid and N-Acetyl-D-Glucosamine When Administered Parenterally to Healthy Cats

M.R. Lappin; J.K. Veir; C.B. Webb

This poster investigated an injectable nutraceutical containing glucosamine, chondroitin, and hyaluronic acid. I’ve written about glucosamine/chondroitin before, and despite the fact that oral products are almost universally recommended and used in both human and veterinary medicine, there is very little reason to think they are helpful. There is slightly better hope for injectable products, thought the evidence is still pretty lackluster given how long and widely used these products are. Still, with the lack of safe and effective therapies for arthritis in cats, I understand the desire to keep trying these things in the hopes that one of them will eventually be clearly shown to be of some benefit. The cynic in me also recognizes that they are a lucrative area in the pet care market, and that any evidence supporting a new product could translate into better sales.

In this study, the investigators injected the product into 8 healthy cats 5 times over 4 months and measured a number of indices of inflammation, oxidative stress, and anti-oxidation ability (a total of 14 measures alltogether) as well as routine bloodwork and urinalysis at 4 different measurement points.

No change was observed in routine bloodwork or urinalysis. The measure of oxidative stress decreased at 1/4 measurement points. The measure of blood anti-oxidant ability did not change significantly. Of the 12 measures of inflammatory cells and signaling molecules, 4 showed changes. Two types of inflammatory cell counts decreased, one at 1/4 time points, the other at 4/4 time points. Two inflammatory signaling molecules increased,  both at 2/4 time points.

Overall, the study suggests some systemic response to the injection, though only a few of the variables measured exhibited any change. The nature of the change was inconsistent, since some markers of inflammation decreased and others increased. Likewise, the anti-oxidant effects were not robust, with one marker unchanged and the other changed only at 1/4 time points. There was, of course, no control group given that this was a small pilot project, but this makes it impossible to know whether the changes that did occur were strictly due to the product or just the injection process. Also, these clinical laboratory tests are only markers of activity within the body’s inflammatory and anti-oxidant systems, and they do not necessarily indicate any positive or negative clinical effect.

While it is important to conduct such small scale trials to look for safety concerns and to justify further, more clinically relevant research, the data generated here does not strongly suggest the product is likely to have a big impact on inflammatory disease. The authors conclude that, “Some of the findings (decreased oxidative stress, increased anti-oxidant capability, and increased IL10 concentrations) could be of potential benefit to cats with inflammatory diseases and suggest that controlled studies of clinically affected cats are indicated.” This is appropriately qualified (“could be of potential benefit”), but such equivocal findings hardly seem a reason for great optimism, especially given the existing literature on glucosamine/chondroitin products, which is extensive and generally not impressive in terms of supporting a clinical benefit. There’s nothing wrong with further research, of course, and it is true that we lack good therapies for inflammatory urinary tract and joint diseases in cats, but I can’t help but feel this isn’t the most fruitful use of limited research resources, especially given the number of times I’ve had to hear lecturers today say, “At this time, there is no clinical research to support….” with regard to equally or even more pressing problems.

I’m currently attending the American College of Veterinary Internal Medicine (ACVIM) continuing education forum, followed by the Evidence-Based Veterinary Medicine Association (EBVMA) symposium, and between lectures, labs, and other events I’m going to try and post tidbits of interest I come across.

This morning I looked at the poster presentations. Posters at scientific meetings are like mini journal articles. They usually report on small studies, often with significant methodological limitations, and they are considered a lower level of evidence than peer-reviewed papers in the journals. This is not a criticism of them as they are a very important starting point for deciding whether to pursue new ideas. It is just important to recognize their limitations and to take a cautious approach to both positive and negative findings presented as posters.

Some researchers are looking at a possibility that leftover bits of feline kidney proteins in vaccines manufactured using cat kidney cells might be a risk factor for immune-mediated diseases which are relatively common in the cat, especially some types of kidney disease and inflammatory liver and intestinal diseases. A couple of papers (1, 2, 3) have shown that cats do form antibodies to some of these proteins after receiving vaccines, but no conclusive evidence has linked this to clinical disease. The concern is a significant one. If it is true, then changes need to be made in vaccine manufacture and use, and there is the potential to reduce significantly an important set of cat diseases. However, if it is not true, then suggestive but inconclusive research on the possibility will only serve to fuel anti-vaccine fears unnecessarily.

The same research group behind the previous papers presented a poster at the ACVIM forum looking for associations between bloodwork values and the presence of antibodies to several feline kidney proteins of concern.

Association Between Feline Antibody Responses to Crandell Rees Feline Kidney (CRFK) Cell Lysates, Alpha-Enolase, and Annexin A2 and Biochemical Abnormalities in 1,477 Privately-owned Cats. J.C. Whittemore; J.R. Hawley; S.V. Radecki; M.R. Lappin

In this project, blood submitted to a commercial laboratory from 1,477 cats was analyzed. No information about age, sex, vaccination or medical history, or health status was available for specific patients. 16 biochemical measures were compared with the presence of antibodies against 3 feline kidney proteins known to be present in some vaccinated cats. The results were fairly inconsistent and not supportive of the hypothesis that antibodies against feline kidney antigens in vaccines is a cause of disease. For one antigen (CRFK) there was a positive association with one of the 16 measured values, bilirubin (a measure of possible liver disease, among other things). There was also a significant negative association with 2 of the 16 values, creatinine (a measure of kidney function) and alkaline phsophatase (a measure of possible liver disease). The other two antibodies examined both had negative associations with blood sugar, a common blood protein and, for one of them, alkaline phosphatase.

In general, then, these results don’t suggest a strong association between chemistry abnormalities in the blood and the production of antibodies against feline kidney proteins found in vaccines. There was no association with most of the values measured, and most of the associations that were found were negative, which if they were clinically significant might suggest the antibodies somehow protected against liver and kidney diseases, which is highly unlikely.  Given all the limitations of the study, especially the lack of any information about the cats the blood came from, this by no means rules out the possibility that the underlying theory is correct, it simply provides a tiny bit of low-level evidence against it. Undoubtedly, this group and others will continue investigating the problem and will hopefully eventually amass sufficient information for a strong conclusion one way or the other. In the meantime, though, it is appropriate to reassure cat owners that the hypothesis is a real but so far unproven concern, and it is not yet appropriate to alter vaccination practices on the basis of it.

Allergies are an incredibly common, and frustrating, problem in pet dogs. Unlike humans, respiratory manifestations of environmental allergies are not usually very noticeable. Dogs with allergies get itchy skin, and they scratch like crazy as a result. The underlying inflammation and the scratching both lead to secondary infections, especially of the ears and skin.

There are several categories of skin allergies, including allergy reactions to fleas, to foods, and to general environmental allergens such as pollen, dust mites, and so on. There are, not surprisingly, a lot of myths about allergies, and a lot of CAM treatments recommended, as is always the case with chronic disease science does not yet have definitive cures for.  In the future I hope to address some of these, but right now I want to discuss some evidence-based recommendations for treatment of atopic dermatitis (AD: environmental allergies, as distinguished from flea or food allergies).

The veterinary dermatology community seems to have adopted not only the rhetoric but the actual practice of evidence-based medicine more effectively than the profession as a whole. There are a number of published evidence-based reviews of therapies for various skin diseases, and a new review recently appeared titled Treatment of canine atopic dermatitis: 2010 clinical practice guidelines from the International Task Force on Canine Atopic Dermatitis. (Olivry T, Deboer DJ, Favrot C, Jackson HA, Mueller RS, et al; for the International Task Force on Canine Atopic Dermatitis. Treatment of canine atopic dermatitis: 2010 clinical practice guidelines from the International Task Force on Canine Atopic Dermatitis. Vet Dermatol. 2010 Apr 23 [Epub ahead of print].) After discussing what is known about the cause and mechanisms of AD, the report reviews various treatments for acute and chronic atopy and grades the evidence and strength of the recommendations for them. I have reproduced the grading scales for evidence and recommendations and then collated the findings in the chart below.

Table 1. Categories of evidence and strengths of recommendation

Category of evidence

Ia. Evidence from meta-analysis or systematic reviews
Ib. Evidence from at least one randomized controlled trial
IIa. Evidence from at least one controlled study without randomization
IIb. Evidence from at least one other type of quasi-experimental study
III. Evidence from non-experimental descriptive studies, such as comparative studies, correlation studies, and case-control studies
IV. Evidence from expert committee reports or opinions or clinical experience of respected authorities or both
LB. Evidence from laboratory-based studies

Strength of recommendations

A. Directly based on category I evidence
B. Directly based on category II evidence or extrapolated from category I evidence
C. Directly based on category III evidence or extrapolated from category II evidence
D. Directly based on category IV evidence or extrapolated from category III evidence
E. Directly based on category LB evidence
F. Based on consensus from Specialty Task Forces
Modified from: Leung DYM et al. Ann Allergy, Asthma, Immunol 2004; 93:S1–21.

Obviously, the text of the review discusses the details of these interventions and the supporting evidence (or lack thereof). The evidence is often of low quality and quantity, but the Task Force has effectively followed the principles of evidence-based medicine in basing recommendations on the best evidence that is available and acknowledging the shortcomings of this evidence where appropriate. Obviously, the strength of their recommendations, and the confidence we as practitioners place in them, should be proportional to the quality of the supporting evidence.

The therapies with the best-documented efficacy, glucocorticoids and cyclosporin, are of course those with the best understood side effects. There is no free lunch in physiology! However, it is also notable that the supposedly safer therapy almost universally recommended in general practice, antihistamines, have little to no evidence of meaningful benefit. It seems likely we give these out mostly in response to the pressure, from the owner and from ourselves, simply to do something. The evidence seems pretty clear that we need to give up this unsupported practice.

Essential fatty acid supplements (EFA, usually fish oils) do have some demonstrable efficacy, but their effects are relatively small and best used to reduce the need for other therapies, such as glucocorticoids, not as primary therapeutic agents. One herbal preparation seemed to have a steroid sparing effect based on a single well-designed clinical trial (Phytopica), but the Task Force was careful to state clearly “Whether or not similar observations would be made with other nutritional supplements has not been established, and care must be taken to not extrapolate these findings to other untested products.”

Such a review is very helpful for the ordinary vet in practice, and while the quality of the evidence is not what we would desire, having it reviewed and explicitly graded in this way is a fine example of how evidence-based medicine can and should be applied in the veterinary field even with the limitations we face.

I’ve written before on the subject of probiotics, bacteria or yeast fed to people or animals with the intent of affecting health in some way. I consider them to be in a bit of a gray zone between mainstream medicine and CAM. There is some plausibility to the underlying idea, and there is some clinical evidence that probiotics can be helpful for GI problems such as antibiotic-associated diarrhea and community acquired diarrhea. However, the wild claims that are sometimes made about probiotics of “boosting the immune system” and treating virtually any disease whether related to the GI tract or not fall into the realm of CAM.

A colleague recently went to a continuing education conference and reported that one of the speakers suggested a particular probiotic product, Fortiflora, can help with symptoms of Herpesvirus rhinitis. Feline herpesvirus 1 (FHV-1) is a ubiquitous viral infection that in some cats can cause chronic, recurrent rhinitis, with sneezing, nasal discharge, and often concommitant conjunctivitis (red, watery eyes). These symptoms come and go, especially with various kinds of stress, and there is no cure and limited benefit from vaccination and oral medications.

I was curious about the evidence for the claim that Fortiflora might help cats with herpesvirus rhinitis since the disease is a common and frustrating one and since the idea that bacteria (specifically Enterococcus faecium) fed to an animal to affect the makeup of the normal GI microflora should help nasal and eye symptoms doesn’t seem to make sense at first glance. My colleague was kind enough to ask the speaker at the conference to send the research paper he had written on the topic, which I reviewed(Lappin MR, Veir, JK, Satyaraj E, Czarnecki-Maulden G. Pilot study to evaluate the effect of oral supplementation of Enterococcus faecium SF68 on cats with latent feline herpesvirus 1. Journal of Feline Medicine and Surgery 2009;11(8):650-4). The experiment was only a small pilot study, and more research on the subject may be warranted, however the existing data do not provide much support for the notion that this product has a role in the treatment of herpesevirus rhinitis.

The study involved 12 cats which had previously been documented to have FHV-1 infection and symptoms of conjunctivitis. The subjects were randomly allocated to a group supplemented with Fortiflora and a control group given an appropriate placebo. They were subjected to stresses such as housing changes and spaying or neutering, and the frequency of clinical symptoms was monitored, as well as measures of GI microflora diversity and PCR to look for shedding of active FHV-1.

Prior to the onset of the study, the cats in the treatment group had a lower rate of conjunctivitis than the control group (13% vs 21%), though this was not statistically significant and might have been due to random chance. There was no significant change in conjunctivitis symptoms for the treatment group during the study (up to 16%), but the placebo group appeared to worsen (up to 29%) which created a significant difference between the groups in symptoms during the treatment period. Of the 6 cats in the supplementation group, conjunctivitis got significantly better in one and significantly worse in another during the treatment period. Of the control cats, one’s symptoms got significantly worse. There was not significant change in symptoms for the remaining 9 of the cats. Respiratory signs were rare in both groups and no differences were measured. With such a small number of cats in the study, it is difficult to view such mixed results as definitively showing or disproving an effect of the treatment on symptoms.

Microbial diversity in the feces (based on PCR) decreased significantly in the placebo group, though there were no significant differences between the level of diversity when the placebo and control groups were compared at any point. Despite some claims of relevance for overall immune function, there is no sound evidence that such measures of diversity represent clinically significant effects on susceptibility to disease. This is typical of studies of probiotics which tend to show that their use does something but which doesn’t tell us what if any real-world significance these effects might have.

Some intermittent shedding of active FHV-1 virus was detected in some cats, but there were no differences between individuals or groups in this measure. There were also no detectable differences in FHV-1 antibody levels or other measures of immune function between groups.

So overall, the study provides little support for the thesis that Fortiflora boosts general immune function in a way that would be clinically useful for cats with FHV-1 rhinitis. Certainly, the additional research the authors suggest is needed may provide greater evidence for this hypothesis, or it may show no meaningful effect. The authors are clearly good scientists, and they are careful in their paper not to make dramatic claims not supported by the data. They tend, as we all do, to interpret the results of their efforts in the most positive light, which is why independent replication of new findings is so important in science. However, the problem with traditional “opinion-based medicine” is to that the theories and opinions of smart and experienced scientists such as these authors are given great weight by those of us in general practice, and there is a tendency to make changes in clinical practices not justified by the quality of the data. This paper is interesting, and the line of inquiry may lead somewhere eventually, but it does not justify the widespread use of Fortiflora for FHV-1 rhintis.

The argument that probiotics “boost the immune system” in general is not a very convincing one, as illustrated in a critique of the concept by Dr.Mark Crislip at Science-Based Medicine.  There is no clinical trial evidence that shows resistance to infectious disease can be enhanced in a healthy, normal person. Certainly, white blood cells and inflammatory mediators can react in a variety of ways to all kinds of stimuli, in test tubes and in living organisms. But the connection between these reactions and a meaningful improvement in resistance to disease has yet to be made. Probiotics undoubtedly have some benefits, however the GI microflora and its role in immune function is a dazzlingly complex subject which science is only beginning to understand. Most of the normal organisms present in healthy people and animals have not been identified, and the ecological niche or function of these organisms isn’t known. The effect of tinkering with this complex system without understanding it is ultimately unpredictable, and the assumption that such tinkering must be beneficial is unwarranted. As our basic understanding of the topic improves, I hope targeted probiotic therapies for specific problems will continue to emerge. But at this point, there is little scientific justification for the widespread use of such products for nearly any condition with the assumption of safety and efficacy.