Vaccination and Associated Adverse Events in Dogs Previously Treated for Primary Immune-Mediated Hemolytic Anemia
ABSTRACT
This study described the rate of vaccine reactions in a population of dogs receiving vaccines after diagnosis of primary immune-mediated hemolytic anemia (IMHA). A secondary objective was to describe the time elapsed between vaccination and initial diagnosis of IMHA. A medical record search identified cases meeting criteria for primary IMHA. Owners and referring veterinarians were surveyed regarding vaccination of the dog following diagnosis. Referring veterinarians were surveyed regarding vaccination prior to diagnosis of IMHA. A completed survey was returned in 44 cases. Twenty-two dogs received vaccinations after diagnosis, and 22 dogs did not. The median time elapsed between vaccination and initial diagnosis was 280 days. No dog was vaccinated within 30 days of diagnosis. Two of the following possible reactions were noted out of 22 dogs vaccinated: vomiting and urticarial eruption in a dog administered a rabies and canine distemper vaccine, and recurrent anemia in a dog administered a rabies vaccine. The rate of vaccine reactions was higher than previously reported, although the time period evaluated was longer than in previous studies. The relationship between initial vaccination and development of IMHA, and between vaccination and vaccine reaction, in this population is uncertain and may reflect coincidence or differences in susceptibility.
Introduction
Much discussion exists over altering vaccination schedules for dogs suspected to be at risk for vaccine reaction, including those with immune-mediated disease.1 Currently, it is common practice to recommend that dogs with a history of immune-mediated disease not receive vaccinations for fear of triggering a relapse; however, published support for this recommendation is lacking. A similar debate exists in human medicine; as in veterinary medicine, evidence for a causal relationship between vaccination and onset or exacerbation of immune-mediated disease is scant.2 Although reports exist suggesting an association between vaccination and immune-mediated disease, a large body of research has documented no relationship between vaccination and disease activity in humans with immune-mediated inflammatory disease.2–6
The link between vaccination and immune-mediated disease in dogs is far from clear. Some report the impression of an increased incidence of autoimmune disease in dogs since the introduction of modified live virus vaccines.1,7 This impression may, however, simply reflect increased recognition of such diseases as a result of the same increased attention to canine medical care that brought about modern vaccination practices. In this case, increased vaccination may simply correlate with an increased incidence of autoimmune disease without being causally related. Data investigating the frequency of immune-mediated diseases before and after the advent of modern vaccination practices would be necessary to confirm or refute a causative mechanism, and this data does not exist.
The mechanism by which vaccines might induce autoimmune disease is unknown. Plausible proposed mechanisms include acceleration of an underlying immune process by vaccinal viral epitopes, mimicry by viral antigens triggering attack on host tissues, cross-reaction of tissue proteins with proteins present in vaccines, induction of an aberrant immune response by viral-induced cytokines, or reaction to vaccine adjuvant.8 Causation is inferred when a close temporal relationship exists between vaccination and onset of immune-mediated disease; however, establishing a causative link is difficult because of the low overall rate of immune-mediated disease in the canine population and the existence of dogs developing immune-mediated disease without any recent history of vaccination.7,9
A few studies have sought to document a temporal link between vaccination and diagnosis of immune-mediated disease. One study found that 26% of dogs presenting for immune-mediated hemolytic anemia (IMHA) to a veterinary teaching hospital over a 27 mo period had been vaccinated in the 30 days prior to presentation.10 No association was found between timing of last vaccine and date of presentation for non-IMHA causes in the same population.10 This study provides evidence suggestive of a temporal relationship between vaccination and onset of IMHA; however, a subsequent study of a similar teaching hospital population found no difference in timing of last vaccination and hospital presentation between dogs presenting for IMHA and dogs presenting for other reasons.11 Two additional studies found rates of recent vaccination in dogs presented for IMHA that were much lower than that found in the original study, although the number of vaccinated dogs in these studies was not compared with a background population.12,13 A small number of additional reports exist of immune-mediated disease in dogs following vaccination, including thrombocytopenia associated with the modified live distemper vaccine, polyarthropathy following administration with a multivalent vaccine, and cutaneous vasculitis induced by rabies vaccination.14–18 One study failed to find an association between the development of immune-mediated thrombocytopenia (IMT) and recent vaccination.19 An additional study of dogs treated for IMT found no increased risk of relapse when dogs were vaccinated following cessation of immunosuppressive therapy after achieving remission of IMT.20
In humans, there is a small number of immune-mediated sequelae of vaccination for which strong evidence exists. These include neurologic reactions, arthropathies, and vasculitis.8 Additionally, a causal link has been suggested between the measles-mumps-rubella (MMR) vaccine and immune thrombocytopenic purpura (ITP).8,21 Other than these few examples, however, strong evidence for widespread implication of vaccination in immune-mediated disease is lacking. A comprehensive review of the literature surrounding vaccination and immune-mediated disease in humans concluded that although hemolytic or aplastic anemias have been documented in close temporal relationship with a variety of viral vaccinations, they are extremely rare.8 In fact, patients with immune-mediated inflammatory disease may be at increased risk for certain vaccine-preventable diseases, and inactivated vaccines are considered generally safe and effective in this population.2
Development of immune-mediated disease in response to vaccination is considered an adverse vaccine reaction. Few studies have investigated the rate of adverse vaccine reactions in the general canine population. Information from postmarketing surveys of vaccine safety report adverse events in terms of number of vaccines sold, which may not reflect the true number of vaccines given. Additionally, reliance on voluntary reporting of adverse events inevitably leads to underreporting.7,9,22 One study used a centralized database encompassing the medical records of many primary care veterinary hospitals to evaluate vaccine-associated adverse events identified within 3 days of vaccination of dogs. This study reported a low overall adverse event rate of 0.38%, most of which were mild, transient allergic reactions.23 The short follow-up time, however, although likely to detect most immediate-type hypersensitivity reactions, may preclude detection of many cases of nonimmediate-type immune-mediated disease.
IMHA is a disease of erythrocyte destruction secondary to the presence of anti-erythrocyte antibodies on the erythrocyte membrane. Primary IMHA is diagnosed when other potential factors causing the development of anti-erythrocyte antibodies have been ruled out.24 This disease was chosen for this study because relapses are often catastrophic, leading to euthanasia of the patient either as a result of refractory disease or on account of the large emotional and financial burden to the owner. As a result, owners of these dogs tend to be cautioned more strongly to avoid vaccination than are owners of pets with other immune-mediated diseases.
Data from human medicine does not support that viral vaccines cause progression or exacerbation of pre-existing immune-mediated disease except for ITP, especially if the ITP was caused by vaccination in the first place. Vaccination of these patients is considered on an individual basis.8 One study found no risk of relapse in a group of children experiencing MMR-associated thrombocytopenia after subsequent MMR or other vaccinations, which may have implications for the common recommendation to avoid future vaccination in veterinary patients with immune-mediated disease.21
The purpose of the current study was to investigate the rate of vaccine reactions (including but not limited to relapse of IMHA) after subsequent vaccination in a population of dogs previously diagnosed with primary IMHA and to describe the characteristics of subsequent vaccination of dogs treated for primary IMHA at veterinary teaching hospitals. A secondary objective was to describe the time elapsed between vaccination and initial onset of IMHA in this population. Because of the probable interaction of patient-specific characteristics with exogenous triggers in the pathogenesis of immune-mediated disease, it is plausible that a dog with IMHA could have characteristics that predispose it to immune dysregulation, which might manifest as other types of vaccine reactions as well. Therefore, the occurrence of other vaccine reactions in addition to relapse of IMHA was included in the survey.
Materials and Methods
Case Selection
The electronic medical records databases of two veterinary medical teaching hospitals (Colorado State University [CSU] and Oregon State University [OSU]) were searched for cases of primary IMHA diagnosed between the dates of August 1, 2009, and February 17, 2016. A list of potential cases was generated using the search terms “immune-mediated hemolytic anemia” or “autoimmune hemolytic anemia.” Potential cases were evaluated to ensure that they met the case definition of primary IMHA. A case was defined as a dog with a packed cell volume of 35% or lower, with the concurrent finding of one or more of spherocytes, autoagglutination, a positive Coombs test, or a bone marrow sample indicating destruction of red blood cell precursors or changes consistent with peripheral destruction of red blood cells. Criteria for diagnosis of IMHA were based on previously published criteria, but using a packed cell volume cutoff of 35% (compared with the published 37%) to adjust for institutional reference intervals.25,26
Dogs were excluded if IMHA secondary to neoplasia or infectious disease could not be ruled out based on the available information. Dogs were excluded if radiographs or computed tomography of the thorax and ultrasound or computed tomography of the abdomen were not performed, or if findings were diagnostic for or highly consistent with neoplasia. Additionally, cases were excluded if at least the minimum amount of infectious disease testing considered necessary in the regiona was not performed or was positive for one or more infectious agents. Although zinc toxicosis, a possible cause of hemolysis with spherocytosis, was not specifically ruled out, it is considered unlikely in this group of dogs because of a lack of reported radiopaque foreign bodies or history of exposure. Dogs were excluded if they were euthanized or died prior to hospital discharge, as these dogs could not have received vaccinations between the time of IMHA diagnosis and death, although these dogs were included in the analysis of initial vaccination compared with initial onset of IMHA. Finally, dogs were excluded if there was inconsistency between owner-reported and veterinarian-reported survey responses.
The Institutional Review Boards at CSU and OSU determined that no review of this study was necessary.
Survey Administration
Owners of dogs identified as cases were administered surveys via postal mail. The survey asked whether the dog currently receives routine vaccination or, if the dog was deceased, whether he received routine vaccinations between the time of its IMHA diagnosis and the time of death or euthanasia, and which vaccines he received. The survey additionally asked whether the owner was aware of their dog experiencing any adverse vaccine reaction, and if so, what type of reaction (relapse of IMHA, vomiting or diarrhea, anaphylaxis, facial swelling, hives, onset of a different immune-mediated disease, or other). Primary veterinarians of the same dogs were administered surveys asking whether they gave vaccines to the dog currently or in the time between diagnosis and death or euthanasia, and if so, which vaccines; whether they vaccinated the dog more than once following diagnosis; and whether they had documented a vaccine reaction and, if so, the type of vaccine reaction. Primary veterinarians were contacted by phone to determine the date of last vaccination of the patient prior to diagnosis of the initial episode of IMHA.
Data Compilation
Data was compiled using a commercially available spreadsheet softwareb. Given the small number of reported adverse events, no statistical comparison was investigated of the frequency of relapse or adverse events in the population that was vaccinated posttreatment as compared with the population from which vaccination was withheld. Confidence intervals for the days elapsed between vaccination and onset of IMHA were calculated using a commercially available statistical softwarec.
Results
Review of the medical records initially identified 243 potential cases (CSU: 195; OSU: 48). Fourteen dogs were excluded for having IMHA as a differential diagnosis or historical problem only, with no available blood work to confirm the diagnosis (CSU: 11; OSU: 3). Seven dogs were excluded for diagnosis prior to the study period (CSU: seven; OSU: zero). Thirty-nine dogs were excluded for not meeting blood work criteria (CSU: 21; OSU: 18). Ninety-one dogs were excluded for lack of necessary additional diagnostics (CSU: 74; OSU: 17). Nine dogs were excluded for neoplasia confirmed or strongly suspected (CSU: eight; OSU: one). Two dogs were excluded for testing positive for one or more vector-borne diseases (CSU: two; OSU: zero).
Eighty-one cases (CSU: 72; OSU: 9) of confirmed primary IMHA were identified based on inclusion and exclusion criteria. Of these confirmed cases, 67 survived to hospital discharge (CSU: 58; OSU: 9). Owners and primary veterinarians of these cases received surveys via postal mail.
At least one completed survey was received in 44 cases (CSU: 37; OSU: 7). Thirty owner-completed surveys and 32 primary veterinarian–completed surveys were returned. Neither survey was returned in 23 cases. One case was excluded for inconsistency of information between owner and primary veterinarian surveys. Vaccination after diagnosis occurred in 22 of 44 dogs, with rabies being the most common vaccine administered (19 of 22 dogs) (Table 1). Twenty-two of 44 dogs received no vaccination after diagnosis.
Two possible vaccine reactions were noted in 2 out of 22 dogs receiving vaccines (9%). These were vomiting and urticarial eruption in a dog administered a rabies and canine distemper vaccine and anemia in a dog administered the rabies vaccine. This was suspected to be a relapse of IMHA based on the assessment of the primary veterinarian in light of the dog’s history of IMHA. The date of most recent vaccination prior to these relapses was not able to be obtained.
A date of last vaccination prior to diagnosis was available for 46 dogs. The median number of days between vaccination and initial diagnosis of IMHA was 280 days (95% confidence interval, 248–386). Of these 46 cases, none had a history of vaccination within 30 days prior to onset of clinical signs, although 3 dogs were vaccinated at 31, 33, and 35 days prior. All three of these dogs survived to discharge after diagnosis of IMHA. These three dogs were not included as the dogs identified as having a possible vaccine reaction following vaccination after diagnosis.
Discussion
Multiple concerns exist regarding limiting vaccination in dogs. The possibility of disease exposure in unprotected dogs is paramount. Although the low prevalence of vaccine-preventable diseases in populations in which most dogs are vaccinated makes a single dog less likely to develop disease if not vaccinated, it is still possible, and the effects on the individual dog can be severe. Additionally, re-emergence of uncommon diseases is possible if fewer animals are vaccinated.27 Although the prevalence of IMHA is low enough that excluding dogs with a history of IMHA from vaccination is unlikely to influence “herd immunity,” the concern over whether to vaccinate these dogs is a small part of a larger issue regarding perception of vaccine safety by the pet-owning population. The necessity of remaining in compliance with municipal regulations is also a concern when recommending not to vaccinate. The preponderance of rabies vaccination in dogs who were vaccinated in this study likely reflects the legal requirement for rabies vaccination in most locations. Quality of life issues for owner and dog also exist; dogs who are not vaccinated may be excluded from boarding, day cares, domestic and international travel, and acceptance by rental agencies and homeowner associations.
Genetic susceptibility to allergy has been documented in humans and dogs.28,29 Additionally, genetic susceptibility to immune-mediated disease has been described in dogs; examples include the association of specific dog leukocyte antigen haplotypes with chronic hepatitis in Doberman pinschers, and with a disease resembling systemic lupus erythematosus in Nova Scotia duck tolling retrievers.30,31 It is possible that a genetically susceptible population of dogs exists for whom vaccinations are more likely to result in adverse reactions, including onset of immune-mediated disease, but at this time, there is no way to identify these patients in a way that is meaningful to the development of vaccine protocols.
There are multiple limitations to the current study. Because the data came from a survey of a variety of practices, the study was not able to be designed to include a control population. Therefore, no comparison of the frequency of vaccine-related adverse events in case dogs and dogs in the general population, or with illnesses other than IMHA, was possible. This comparison would ideally be performed using a large database of dogs containing information on complete vaccination status in addition to disease diagnosis. This may not be feasible, however, as many dogs with IMHA are diagnosed at referral hospitals, where vaccination history may not be recorded. Because of this, no cause-and-effect relationship between vaccination and vaccine reaction can be inferred from this study. Likewise, it is impossible to say conclusively based on this study that no relationship exists between vaccination and vaccine reaction in this population.
The retrospective nature of the study does not allow us to say with absolute certainty that all cases of IMHA included in the study were primary. However, the inclusion and exclusion criteria applied to each case is likely to have excluded secondary cases to the greatest degree possible.
The data collected may be subject to an owner’s memory or record keeping of which vaccinations their dog received. Because of the use of postal mail, there is a possible bias toward more recent cases, as owners who have not relocated are more likely to be easily reachable by mail. Additionally, records are more likely to be available for review in recent cases. Finally, there is the possibility of misinterpretation of the surveys given to referring veterinarians; the question asking whether they had vaccinated dogs treated for IMHA between diagnosis and death could have been interpreted as asking whether they had ever given the dog vaccinations in its lifetime, including before diagnosis. Although attempts were made to exclude surveys with inconsistency of information, this may have resulted in a falsely elevated number of dogs reported to receive vaccinations after diagnosis.
The proportion of possible vaccine reactions in vaccinated dogs in this study (9% of vaccinated dogs) was higher than that reported in the study that most closely estimates the general population (0.38%); however, the observation period of 3 days in the previous study is much shorter than the observation period of up to several years in the present study.23 Previous studies of dogs with IMHA have reported proportions of patients with relapses of disease. Direct comparison is difficult, both between the previous studies and with the present study, because of differences in timing of follow-up and how a relapse was defined. For example, one case series described relapse in 13% of cases overall (24% of dogs surviving to discharge).13 Another case series described relapse in 12% of cases overall (19% of dogs surviving to discharge who were re-evaluated at least once at the original institution).32 An additional case series reported relapse in 15% of cases surviving longer than 60 days.33 It seems worth noting, however, that the 9% of dogs in this study experiencing a suspected vaccine reaction is not higher than the proportion of dogs experiencing relapses in previous studies. Comparison with a control group in which vaccinations were withheld would be necessary to draw conclusions regarding relative risk of relapse in this group, although this would be logistically challenging for the reasons described previously.
No dog was vaccinated within 30 days of initial onset of IMHA, which may imply a difference from the results of a previous study in which a temporal relationship was noted.10 Without a comparison of vaccination dates of a control group presented for non-IMHA complaints taken from the same population, it is not possible to challenge the results of the previous study; however, subsequent studies exist in which the relationship between vaccination and onset of IMHA was less clear. A study by Carr et al. found no difference in the time from vaccination to onset of IMHA compared with vaccination and presentation for non-IMHA-associated causes.11 Burgess et al. reported only 13% of dogs vaccinated within the 30 days prior to presentation for IMHA, whereas Reimer et al. reported only 4.2% of dogs vaccinated within 14 days of presentation.12,13 Although not the focus of these studies, these percentages are lower than the 26% of dogs vaccinated within 30 days of onset of IMHA in the original study.10 These findings, in addition to the findings of the present study, may provide evidence against a consistent temporal association between vaccination and initial onset of IMHA.
Conclusion
No dog was vaccinated within 30 days of initial diagnosis of IMHA. Two vaccine reactions were reported in 22 dogs previously diagnosed with IMHA receiving vaccinations. Although the proportion of vaccine reactions in this study was higher than that in previous studies, the relationship between initial vaccination and development of IMHA, as well as between vaccination and relapse in this population, is uncertain. Possible explanations for these findings may include coincidence or differences in genetic susceptibility to adverse immunologic consequences following vaccination.
Contributor Notes
