Introduction

Canine distemper virus (CDV) is a Morbillivirus in the family of paramyxoviruses, with dogs serving as the main reservoir of the virus, although some wildlife species may also serve as a reservoir for outbreaks in dogs.^1–3 Disease susceptibility and expression are variable and dependent on the immunological status and response of the dog at the time of infection and, possibly, the viral strain.¹ Infections may be asymptomatic but can present with respiratory, gastrointestinal, and/or neurological signs.^1,2

Canine distemper virus infection of the nervous system can manifest concurrently with other systemic signs but may also develop in the weeks following apparent recovery from the respiratory or gastrointestinal forms.^1,2 Histological lesions, which can include neuronal degeneration or necrosis and demyelination, can affect the white or gray matter (or both) and be focal, multifocal, or widespread throughout the central nervous system.^4–7 The most common histological form of CDV infection of the nervous system is demyelinating leukoencephalomyelitis (CDV-DL).² The neurological signs described for CDV infection include seizures and myoclonus, which are common, and, variably, vestibular signs (head tilt and nystagmus), cerebellar signs (hypermetria, intention tremors, and truncal ataxia), blindness, cranial nerve deficits, paresis or paralysis of one or more limbs, and spinal hyperesthesia.^1–7

Although vaccination against CDV infection remains the most effective means of preventing the disease, the combination of asymptomatic carriers and unvaccinated dogs in animal shelters creates the ideal situation for an outbreak.⁸ In one study, 7.4% (37 of 503) of asymptomatic dogs presented to animal shelters were positive for CDV by polymerase chain reaction (PCR) assay on an ocular or oronasal swab.⁹ Although this number may seem relatively low, the number of susceptible dogs is apparently high, as another study showed that up to 64.5% of dogs entering shelters lacked protective immunity to CDV.¹⁰ Unfortunately, for dogs that become infected with CDV and develop neurological signs, there are no therapies documented to clear the virus and resolve clinical signs.

An unorthodox approach to the treatment of CDV infection and CDV-DL using the modified live Newcastle disease virus vaccine (NDV-MLV) was first introduced in the early 1970s by Alson Sears, DVM, and is currently promoted at www.kindheartsinaction.com, claiming a success rate of 86% for all forms of CDV infection.¹¹ The NDV-MLV is used in one of two ways: the vaccine is injected IV into healthy dogs and then blood is collected 24 hr later to create a “serum” that is subsequently injected subcutaneously into dogs with systemic CDV infection; or by direct intrathecal injection of the NDV-MLV in dogs with CDV-DL. In May 2012, a 2 yr old, 18 kg, unvaccinated, female border collie with CDV-DL presented to the Kansas State University Veterinary Health Center (KSUVHC) with a request for treatment with the intrathecal NDV-MLV. The dog survived and completely recovered, as documented by the owner on YouTube^a.

The reason for the apparent response in this dog, or any dog, to intrathecal administration of NDV-MLV is not known. Although Newcastle disease virus (NDV) is also in the paramyxovirus family, inoculation with various NDV strains was not demonstrated to provide protective immunity against CDV infection in dogs.¹² The intrathecal administration of an attenuated vaccine is not a new concept and was used successfully with rabies in dogs; however, that success was attributed to the development of protective antibodies.¹³ CDV paradoxically results in systemic immunosuppression while at the same time inducing a strong localized immune response within the brain leading to neuronal necrosis and severe demyelination.² Previous studies identified high levels of proinflammatory cytokines in the brain and cerebrospinal fluid (CSF) of dogs with CDV-DL, with the most consistently upregulated cytokines being interleukin (IL)-6, tumor necrosis factor, transforming growth factor-β, and IL-12.^14–17 If an immune response to the intrathecal administration of NDVMLV modulated the expression of cytokines within the CSF, this could potentially explain the efficacy of this novel therapeutic intervention. Recombinant NDV vaccines have been used in oncology for their oncolytic properties and act through either a direct cytotoxic effect or the stimulation of the immune system.¹⁸

The purpose of this study was to evaluate the clinical efficacy of intrathecal administration of NDV-MLV in dogs with neurological signs and the treatment effect on CSF cytokines. We hypothesized that if therapy was effective, a significant difference would be noted between proinflammatory and anti-inflammatory cytokines between baseline and post-treatment.

Materials and Methods

Results

Eleven dogs were of mixed breed. There was one vizsla and one Catahoula leopard dog. There were eight male dogs (five neutered) and five spayed female dogs. Ages ranged from 3 to 76 mo, with a median of 10 mo. Weights ranged from 3.9 to 28.5 kg, with a median of 10.0 kg. The duration of neurological signs ranged from 10 days to 5 mo (median, 4 wk). On general physical examination, four dogs were febrile on presentation (39.33°C to 39.94°C), two dogs were assessed to be dehydrated, two dogs were emaciated, two dogs had ocular discharge, and one had severe trauma to the tongue attributed to myoclonus. Myoclonus was the most common finding on neurological examination (n = 11 dogs [84.6%]). The findings of the neurological examinations and results of PCR testing are summarized in Table 1. Two dogs that previously tested positive by PCR for CDV were negative at KSUVHC but were still enrolled based on prior positive testing and the presence of myoclonus. One dog positive for CDV by PCR on whole blood also tested positive on whole blood 4 wk earlier.

TABLE 1 Neurological Examination Findings and PCR Results in 13 Dogs with Canine Distemper

View Fullscreen

The results of the CBC, serum biochemistry profile, and urinalysis were within normal limits in five dogs. Abnormalities in the remaining dogs included anemia (n = 3 dogs; hematocrit 0.34 in all three [reference interval 0.37–0.55]); eosinophilia (n = 2; 1.8 and 3.0 × 10⁹/L [reference interval 0–0.75 × 10⁹/L]); inflammatory leukogram (n = 2; band neutrophils 0.2 and 1.1 × 10⁹/L [reference interval 0–0.1 × 10⁹/L]); and lymphopenia (n = 1; 0.2 × 10⁹/L [reference interval 0.8–4.3 × 10⁹/L]). Fecal flotation was positive (n = 6 dogs) for Cystoisospora spp. (n = 3 dogs), Toxocara canis (n = 1), Trichuris vulpis (n = 1), and Physaloptera spp. and Ancylostoma caninum (n = 1).

The CSF analysis for all dogs had a median protein of 41.5 mg/dL (range 13–128 mg/dL) and cell count of 37 cells/μL (range 1–153 cells/μL). The CSF analysis was normal in 3 dogs (30.7%), even though one dog had a positive PCR for CDV on CSF. Pleocytosis was described as lymphocytic (n= 7 dogs), monocytic (n= 2), and mixed mononuclear and lymphocytic (n= 1). Antibodies to CDV were assayed in serum on 10 dogs (inadvertently not performed on 3 dogs) and CSF on 13 dogs (Table 2). The results of cytokine analysis are presented in Table 3. The only statistically significant differences were found comparing dogs with a normal or abnormal CSF analysis in the following cytokines: IFN-γ (P = .006), IL-8 (P = .03), IL-10 (P = .02), IL-15 (P = .002), KC-like (P = .02), and MCP-1 (P = .03).

TABLE 2 Serum and CSF Neutralizing Antibody Titers to Canine Distemper Virus from 13 Dogs with CDV-DL

View Fullscreen

TABLE 3 CSF Cytokines at Baseline Evaluation (Abnormal CSF, Normal CSF, Nonsurvivors Pre, Survivors Pre) and at Follow-Up (Survivors Post)

View Fullscreen

Seven dogs survived a minimum of 3–4 months; however, only six were returned for follow-up. These six dogs (dogs 1 and 3–7) ranged in age from 4 to 13 mo (median 6.5 mo) with a duration of clinical signs from 2 to 4 wk (median 3 wk). The dog with cerebellar ataxia and intention tremors was neurologically normal at follow-up. Three dogs showed no change in severity or distribution of their myoclonus at follow-up. In dog 5, facial myoclonus resolved but pelvic limb myoclonus persisted unchanged. In dog 7, myoclonus progressed from facial and body to include all limbs; this dog was euthanized shortly after the follow-up, but necropsy was not performed. All six dogs initially had evidence of inflammation on the CSF analysis, and all had a normal CSF analysis at follow-up. At follow-up, only three dogs were positive for CDV by PCR (conjunctival scrape only in dogs 3, 5, and 7). The NDV PCR was negative in all dogs at follow-up. One dog (dog 6) that survived to follow-up was lost to long-term follow-up. Long-term follow-up was available for four dogs (dogs 1 and 3–5), all of whom were still alive at the time of this writing (July 9, 2019), with survival times ranging from 3.7 to 4.5 yr. This group included the dog with cerebellar ataxia and three dogs with myoclonus, which was no longer present in any of the dogs. One dog was known to have survived a minimum of 3 mo but did not return for the second evaluation and was subsequently lost to follow-up.

Six dogs did not survive to follow-up, with a median survival of 11 days (range 1–23 days). All but one dog died, and that dog was euthanized 23 days after the procedure (dog 11). Necropsy was performed on four dogs (dogs 8–11), and the histological lesions were consistent with CDV-DL in three (dogs 9–11). Dog 11 had CSF sampled at the time of euthanasia and was still positive by PCR for CDV. These three dogs also had pneumonia consistent with CDV infection, although respiratory signs were not evident at the time of initial evaluation. Dog 10 also had aberrant parasite migration in the kidneys and tongue (this dog had an eosinophilia of 3.1 K/μL and Cystoisospora found on fecal flotation). The fourth dog (dog 8), a 3.5 mo old, neutered male Catahoula leopard dog with severe myoclonus and pelvic limb paresis, died 12 days after the procedure and did not have lesions consistent with CDV infection identified in any organ, although the diagnosis of CDV was established by PCR at least 6 wk before presentation to KSUVHC. The significant findings included severe, bilateral necrotizing and eosinophilic adrenalitis, moderate granulomatous and eosinophilic pneumonia, and mild necrotizing and histiocytic hepatitis; however, an etiology was not identified. This dog did not have any findings on CBC or serum biochemistry profile consistent with hypoadrenocorticism on initial evaluation, and a serum cortisol (performed after the dog died on serum collected from the initial visit) was 91.3 nmol/L (reference interval 28–110 nmol/L). At initial evaluation, there were no abnormalities on the CSF analysis, reciprocal titers against CDV of CSF (32) were higher than serum (8), and PCR for CDV was negative (all samples) despite previously testing positive by nasal swab. This was the only dog to exhibit postprocedure neck pain (∼3–4 days after the procedure), prompting the use of tramadol, but there was no evidence of bacterial meningitis or any other explanation for the neck pain identified at necropsy.

Discussion

Four dogs treated with an intrathecal injection of the NDV-MLV survived long term (more than 3.5 yr) with eventual complete resolution of all clinical signs for which they initially presented. However, this should not be interpreted as a successful response to therapy. This study was undertaken without including placebo-treated controls because we did not believe that dog owners recruited through www.kindheartsinaction.com would agree to possible randomization into a placebo group based on the bias that dogs with CDV-DL would not survive long term or at least have a survival rate less than 10%. In that context, a 28.6% long-term survival seems encouraging; however, this survival rate is difficult to evaluate in comparison with limited historical controls.

In a review of the pathogenesis of CDV infection, it was stated that dogs with neurological manifestations “usually die,” with the caveat that “some recover.”² Although a study of 192 puppies aged 2–4 mo with naturally acquired CDV infection documented a 61.5% mortality rate, mortality was not correlated to the various clinical presentations.¹⁹ In an investigative study of mesenchymal stem cell therapy for naturally acquired CDV infection, the mortality rate was 80% (4 of 5 dogs) in the control group and 67% (2 of 3 dogs) in the treatment group (at least 6 dogs had neurological signs).²⁰ Additionally, in a study comparing the use of xenogeneic anti-CDV antibodies in the treatment of puppies with naturally occurring CDV infection manifesting initially only with severe respiratory signs, the mortality rate for all puppies in the control group (supportive care only) was 68.7% (11 of 16 puppies), with a mortality rate of 91.7% (11 of 12 puppies) for puppies that developed neurological signs.²¹ In that same study, the mortality rate for puppies that developed neurological signs in the treatment group was 50% (4 of 8 puppies).²¹ In comparison, survival was reported to be 100% at 6 mo in 11 shelter dogs in Switzerland, none of whom developed neurological complications, including 4 with concurrent vector-borne infections.²²

The 4 long-term survivors had an abnormal CSF analysis on initial evaluation but a normal CSF analysis at follow-up. This was initially interpreted as a successful response to intrathecal injection of NDV-MLV, as all dogs evaluated up to that point initially had an abnormal CSF analysis. However, subsequently, 3 dogs were seen that initially did not have an abnormal CSF analysis, including one in whom the CSF was positive on PCR for CDV. These 3 dogs had clinical signs present for 5–12 wk at the time they were enrolled into the study, and it is possible that they would have had an abnormal CSF analysis if seen earlier in the course of the disease. Resolution of CSF pleocytosis may represent the routine course of the disease, regardless of the eventual outcome. In one study, a CSF pleocytosis was absent in 11 of 19 dogs confirmed to have CDV-DL at necropsy, including 7 of 13 dogs with myoclonus, suggesting that the intrathecal administration of NDV-MLV likely played no role in the resolution of CSF abnormalities in the 6 dogs of the current study.⁴ In further support of the lack of effect of the treatment, CSF cytokines were no different between dogs with a normal CSF analysis at baseline and dogs with a normal CSF analysis at follow-up. Likewise, long-term survival with eventual resolution of all clinical signs, as was seen in 4 dogs in this study, may be independent of any intervention and more dependent on the innate immune response of the dog and supportive nursing care.

The only significant differences in CSF cytokines were found between dogs with an abnormal versus normal CSF analysis at baseline, which was not unexpected. The number of survivors evaluated was too low to reach significance; however, changes in IFN-γ, IL-8, IL-10, and IL-15 showed reductions similar in comparison with abnormal versus normal dogs at baseline. Elevations in IL-8 and IL-10 have been previously documented in the CSF of dogs with CDV-DL and did not appear to be predictive of outcome in the current study.^14,23 IL-8, IL-15, KC-like, and MCP-1 are inflammatory cytokines likely produced by resident glial cells or monocytes in response to viral infection.^17,24–26 Although IL-10 is recognized as an anti-inflammatory and neuroprotective cytokine, its expression is also likely stimulated by the presence of CDV.^17,27 To that point, it was significantly elevated in dogs with an abnormal CSF analysis, did not predict survival, and was highest in a nonsurvivor that was positive for CDV by PCR on CSF. In contrast to the study by Frisk et al., the current study did not find IL-6 elevations in any group that differed significantly, which may have reflected that these dogs were sampled in a later stage of their disease.¹⁴ Likewise, tumor necrosis factor-α, which is likely elevated early in the course of disease, was undetectable in the CSF in 9 of 14 dogs initially.^14,17 Additionally, IFN-γ is reported to be elevated later in the course of disease and was significantly elevated in the dogs with an abnormal CSF in the current study.^14,17 The lack of any significant difference in CSF cytokines between dogs with a normal CSF at baseline and those at follow-up suggests that changes noted in the cytokines in survivors were associated with the absence of inflammation in the natural course of the disease and not as a response to the intrathecal injection of NDV-MLV.

An inclusion criterion for this study was the confirmation of CDV infection by PCR and not reliance on CDV antibody testing, which has been shown to increase the misdiagnosis of CDE.²⁸ Eleven dogs had a positive PCR for CDV from nasal, pharyngeal, or conjunctival swabs before arrival, but 2 of those dogs subsequently tested negative at KSUVHC. These discordant test results may have represented false-positive tests initially but also may reflect the natural course of CDV infection in dogs with a progressive reduction in detectable virus associated with duration of the disease. Similarly, 3 of the 6 dogs that survived to follow-up had no detectable CDV by PCR. Although the conjunctival scrape yielded the most frequently positive results on PCR (11 out of 14 dogs were positive), another limitation in the current study was the failure to perform PCR on urine, which was shown to be more sensitive than CSF, blood leukocytes, or serum (no comparison with conjunctival scrape was performed in that study).²⁹

Serum and CSF titer interpretation in these dogs was confounded by the fact that all had received variable numbers of vaccinations before evaluation. All dogs had a serum-to-CSF ratio that would be suspicious of CDV-DL.²⁸ Neither the serum or CSF antibody titers nor the ratio of serum to CSF titers was predictive of outcome: nonsurvivors had the lowest and highest reciprocal serum titers (8 and 1024), the lowest and second-highest CSF titers (<2 and 32), and the lowest and highest ratios (0.25 and 64). Had the presence of CSF antibodies been used as a diagnostic criteria, however, two of the dogs with a positive PCR on CSF for CDV would have been excluded from the study.

Conclusion

The results of this study do not support the intrathecal injection of NDV-MLV for the treatment of neurological symptoms caused by CDV infection in dogs. Although there were no apparent detrimental consequences of the procedure to any of the dogs in this study, the potential for such exists. The long-term survival of four dogs appeared to be consistent with the natural course of disease and not therapeutic intervention.