Should we be Vaccinating Dogs at Acupuncture Points?

A couple of recent articles by acupuncturists have suggested that giving vaccines at acupuncture points may be more effective than giving them in other locations. This is clearly an attempt to demonstrate the serious, scientific legitimacy of acupuncture. However, there are a number of serious problems with these claims and with the studies used to support them. 

Study 1
Jin H, Xu Y, Shi F, Hu S. Vaccination at different anatomic sites induces different levels of the immune responses.Res Vet Sci. 2019;122:50-55. doi:10.1016/j.rvsc.2018.11.005

The Study
This study vaccinated rats with a live attenuated combination vaccine for dogs containing antigens for canine distemper, canine parvovirus, canine parainfluenza, canine adenovirus, and rabies. The rats were injected at several different anatomic locations (see Figure 1). A group of 10 female Rottweiler puppies between 32-36 days old were also vaccinated twice, three weeks apart, first with a modified live distemper/parvo vaccine and then with a modified-live vaccines for distemper, parvovirus, adenovirus, and parainfluenza. These vaccines were given at different locations in different dogs (see Figure 1). 

Figure 1

The study measured blood antibody levels as a marker of vaccine response. The rats were vaccinated twice, 2 weeks apart, and blood was collected at 2, 4, and 6 weeks after the second vaccination. The dogs were vaccinated twice, with different vaccines, three weeks apart. Blood was collected from the dogs 2, 4, and 6 weeks after the second vaccination. The levels of antibodies detected at each time period following vaccination at the different locations in the rats and dogs are shown in Figures 2 and 3.

Figure 2
Figure 3

Analysis of the Study
Methodologically, this study lacks a number of important features to control for error and bias. There was no randomization of subjects, no blinding of investigators or caregivers to the different treatments, and only a small number of dogs of one sex and breed were included. Such controls are especially important given that the study was funded by a veterinary acupuncture group with a clear interest in the outcome. It is also unclear what relevance the immune response of rats to injection with canine vaccines might have to the effect of those vaccines in protecting dogs from infectious diseases.

The dogs in the study were initially vaccinated at about 4 weeks of age, which is earlier than recommended because most puppies will have maternal antibodies at this age that interfere with the effect of vaccination. The second vaccine they received was at about 7 weeks of age, which is still slightly earlier than typically recommended. It is unclear whether these puppies had nursed or had maternal antibodies to begin with, but generally vaccination is done repeatedly between 8 and 16-20 weeks to ensure an adequate immune response. 

It is also unclear whether the differences in antibody levels have any real-world significance. With only a couple of exceptions, the levels were above or below the cutoff for all dogs in all groups at each time point. All dogs achieved antibody levels above the cutoff by about 11 weeks of age, well before the end of the usual vaccine series, so there is little reason to think the differences measured have any implications for susceptibility to disease, which is the import thing in any immunization program. There were also no adverse events for any of the dogs, so there is no reason to think the location of vaccination influences safety or effectiveness of immunization. 

The fact that antibody levels different with the site of vaccination is interesting. There has been some research in humans suggesting that the anatomic location of vaccination can influence the strength of the response.1-2This is suspected to be due to differences in blood supply, fat density, presence of immune surveillance cells, and other such factors. However, what, if anything, this has to do with acupuncture is a different question. For one thing, as I have discussed in detail in the past, there is little compelling evidence that acupuncture points exist at all as distinctive anatomic or functional locations. There is also significant variation in where acupuncturists locate such points, suggesting the concept is more a metaphor than a biological reality. 

In this study, the highest antibody levels were detected at the huohaiacupuncture point (also known as GV 1). This is a sensitive spot between the dorsal edge of the anus and the ventral tail base (see Figure 1). There likely are differences in the activity of the immune system between such a protected spot located close to an obvious source of bacteria and a location like the nape of the neck, a spot fortified against attack with thick skin and subcutaneous fat, little blood supply, and far from any excretory organs. There are plenty of distinctions between such locations that are far more plausible than the mystical notions of energy channels, Ch’i, Yin and Yang, and the other ideas that form the theoretical structure of Traditional Chinese Medicine. Any difference that might be found between the immune response at these spots does not require us to accept the reality of such folk mythology as the only explanation.

Study 2
Perdrizet JA, Shiau D-S, Xie H. The serological response in dogs inoculated with canine distemper virus vaccine at the acupuncture point governing vessel-14: A randomized controlled trialVaccine. 2019;37(13):1889-1896. doi:10.1016/j.vaccine.2018.10.076

The Study
100 client-owned dogs were randomly assigned (though by an odd method with some potential for umasking) to be vaccinated with a modified-live vaccine containing antigens for canine distemper, adenovirus, parvovirus, and parainfluenza at either the side of the neck or at the GV-14 acupuncture point (dorsal midline cranial to the dorsal spinous process of T1). Antibody titers for canine distemper were measured at the time of vaccination and 2 weeks later for all dogs.

Analysis of the Study
The dogs in both groups were vaccinated about a year after previous vaccination, which is earlier than the 3-5 years typically recommended, and both groups had both measurable antibody levels before vaccination and an increase in these levels after vaccination. There was tremendous individual variation in antibody titers and response to vaccination (Figures 4 and 5). Whether there is a consistent and meaningful effect of location of immune response is difficult to tease out of this background variation. The authors did some transformations and various statistical manipulation of the data to identify a statistically significant difference. Whether this is a true difference or a function of the particular study and the methods used is unclear.

Figure 4. Antibody responses to vaccination in different sites
Figure 5. Change in antibody titers after vaccination in different sites.

It is also unclear if there is any real-world difference in the protection against disease between these groups, a fact the authors acknowledge. Even if the location of vaccination predictably influences the magnitude of immune response, most dogs will be protected either way and a small proportion will not be protected either way, so is there any meaningful difference in the chances of dogs becoming ill? That is not a question this study can answer, but given how effective current vaccination is for core infectious diseases, it seems unlikely that changing the location in which we vaccinate is going to protect more dogs.

The authors imply that one reason to determine if vaccination at supposed acupuncture points might be worthwhile is that we could reduce the risks of vaccination if we generated a stronger or more durable response and, presumably, could then vaccinate less. This is implausible. For one thing, the disease they suggest result from vaccination, such as immune-mediated destruction of platelets or red blood cells, have not been consistently or convincingly associated with vaccines.And if we were able to trigger a stronger immune response by vaccinating in a particular location, there is no reason to believe this would not also increase any risks associated with the vaccination response. No adverse events were reported for either group in this study (not surprisingly, since such reactions are uncommon, though it is also possible that dogs with a history of vaccine reactions simply were not included in the study since they may no longer be receiving vaccines). 

General Issues 
These studies do not make a convincing case for vaccinating dogs in supposed acupuncture points. The reasons for this include the following:

  1. The principles of Traditional Chinese Medicineare unscientific folk mythology, and there is no reason to believe any practice guided by these ideas is going to be equal to or superior to practices based on scientific principles.

  2. Despite thousands of years of use, and decades of research, it has not been convincingly shown that so-called acupuncture points exist at all, except as metaphorical concepts. Acupuncturists in different sects identify very different points, and studies show acupuncturists are highly variable and imprecise in their localization of points. No consistent evidence indicates that acupuncture points in general, or the specific points in these studies, have distinctive and relevant anatomic or functional differences from other locations not designated as “points” by the TCM mythology.

  3. The variability in antibody levels and immune response may well be influenced by the location of vaccination. However, any such effect is more likely to be explained by objective and conventional anatomic and functional features of these locations, not by their status as special locations in terms of folk metaphors.

  4. There is great individual variation in antibody levels and immune response to vaccination, and it is unclear whether the statistical differences identified in these studies represent real differences, study artefacts, or normal variability. There is a growing recognition that statistical significance testing is misused in medical research and readily creates the impression of meaningful biological difference where none exists. This awareness has led to calls from statisticians and researchers to abandon the practice and emphasize more meaningful measures of effect in research studies.

  5. Even if it were established that there were true differences between the groups in these studies, from whatever cause, it has yet to be established that these have any real-world significance in terms of protection from disease or safety. Current vaccination practice are incredibly safe and effective, and while improvement is always desirable, the small differences seen in these studies are unlikely to have any implications for the actual health of patients. Certainly, nothing in this research suggests the locations chosen by the authors are safer or more effective at presenting disease than current locations typically used.

  6. From the practical perspective of a clinician actually treating dogs, I can tell you that vaccinating routinely at GV-1 (just above the anus) is going to be far more painful and difficult than current locations. Without a large, proven benefit to doing so, it makes no sense to subject patients or staff to this practice. 

Interestingly, the least antibody response seen in the Jin study was at a location very close to the GV-14 point used in the Perdrizet study. While I’m sure the authors would argue that these points are not comparable because of the precise localization of acupuncture point, again the research indicates that such points are in fact quite variable and imprecise, so to a small extent these studies actually contradict one another in terms of the effect of vaccination over the shoulders.

Bottom Line
There is no plausible reason to think that vaccinating dogs at supposed acupuncture points, if these even exist, will improve the protection from disease conveyed by vaccination or will reduce the risks of vaccinating.


  1. Jin H, Xu Y, Shi F, Hu S. Vaccination at different anatomic sites induces different levels of the immune responses. Res Vet Sci. 2019;122:50-55. doi:10.1016/j.rvsc.2018.11.005

  2. Shaw FE, Guess HA, Roets JM, et al. Effect of anatomic injection site, age and smoking on the immune response to hepatitis B vaccination. Vaccine. 1989;7(5):425-430. Accessed April 8, 2019.
This entry was posted in Acupuncture. Bookmark the permalink.

6 Responses to Should we be Vaccinating Dogs at Acupuncture Points?

  1. v.t. says:

    Yeah, let’s subject pets to one of the most sensitive areas on the body (not to mention under the jaw as well) and use a needle injection there (gasp!)- I propose a study where acupuncturists are the subjects in the same study, let’s see how that works out.

    Fear not, acupuncturists (and other woo-meisters) will never stop with the crazy.

  2. Mireille says:

    Thank you for this site. I searched on this blog but didn’t find any specific posts on injection-site sarcoma in cats. I believe (albeit I haven’t done a Google Scholar search on peer-reviewed articles) that there’s some evidence for reducing the risk of injection site sarcoma when using adjuvant free vaccines (such as PureVax), I’m wondering if you can do a post on this? Mostly because PureVax is not readily available in all vet clinics (esp. outside of the US), and can be more expensive than regular vaccines that do contain adjuvants, so I’m wondering if the evidence is strong enough to warrant the extra $ and effort to ensure my cats get PureVax. Thanks!

  3. skeptvet says:

    There are no direct comparisons of cats vaccinated only with adjuvanted vaccines and those vaccinated only w/ non-adjuvanted vaccines comparing the incidence of injection-site sarcomas. Given how rare these tumors are, the timeframe for their development, and the many other potential triggers, it is unlikely that we will see such a study or have a definitive answer any time soon, if ever.

    Some epidemiological survey research has been done, and it is mixed. So far it doesn’t seem to show any change in the incidence of these tumors with changes in vaccination practices in Canada or the US, though the location of the tumors does seem to be changing, reflecting the change in the anatomic location of vaccine administration (here’s a good review). However, one survey in Switzerland did show a decreased incidence following introduction of non-adjuvanted vaccines.

    The hypothesis of a causal role for aluminum and inflammation is plausible and has some support, but it is by no means ironclad (Prevention of Feline Injection-Site Sarcomas: Is There a Scientific Foundation for Vaccine Recommendations at This Time?). It is clear that the Purevax vaccines induce less inflammation, but there is no data I am aware of on whether this translates into a meaningful decrease in the relative or absolute risk of injection-site sarcomas. Overall, it is likely but not clearly proven that using non-adjuvanted vaccines will reduce the risk of injection-site sarcomas.

    Set against this is there risk of rabies, which depends on lifestyle, geographic location, and other factors. In CA, where I practice, there is a mean of 1.1 cases of rabies in domestic animals per year (cats make up about half of those). Pretty rare given given the many millions of cats living in the state, though of course still an important public health risk.

    So I don’t think there is a simple “right” answer based on the existing evidence. I use Purevax 3-yr vaccine almost exclusively because it is demonstrably as effective as adjuvanted and may reduce the FSA risk some. I also don’t always boost the vaccine for indoor only cats if the client asserts they don’t have any escapes. A bat could, of course, come in the house, but that’s a rare event. I think the evidence is good enough to recommend it, but I don’t think we can definitively say it will reduce cancer risk significantly yet.

  4. Jennifer Robinson says:

    The background variation is interesting, however.
    My immigration to New Zealand had to be delayed by 3 months because one of my dogs rabies titer came in a bit under the cut off point. She had been vaccinated 9 months before, so this came as a surprise. It’s only marginally significant to this discussion, as it’s doubtful that there was any health or safety risk, and the issue is more duration of immune response than initial response. Still, it suggests that understanding of responses to vaccination isn’t great… and that there may be regulatory as well as health and safety concerns.

  5. skeptvet says:

    The background variation could be random error, or it could indicate differences in response to antigen presentation at different sites. Unfortunately, none of that has anything to do with the mythology behind the location and identification of “acupuncture points.” Such variation, especially if you can torture a significant p-value out of it, is all too often the foundation for specious, implausible medical claims.

  6. Jazzlet says:

    Vaccinating at GV-1 seems like a great way to induce fear of routine vetinary visits as compared to say the scruff of the neck which (anecdote) my dogs don’t seem to notice at all.

    Thanks for going through this, you provide useful evidence to counter ridiculous claims.

Leave a Reply

Your email address will not be published. Required fields are marked *