Survey-based pilot study into the chosen therapy and prophylaxis used by UK primary 1 care veterinary surgeons against canine angiostrongylosis 2

22 Canine Angiostrongylosis (CA), a gastropod-borne parasitic infection caused by the 23 metastrongyloid nematode Angiostrongylus vasorum , is an important cause of significant 24 morbidity to domestic dogs across the UK as well as in other European countries. This study 25 aimed to ascertain the frequency at which particular drugs were used by primary care 26 practitioners in the UK for therapy against and prophylaxis for CA. Primary care veterinary 27 clinicians were surveyed using an online questionnaire and face-to-face or telephone 28 interviews. Eighty-six veterinary surgeons responded. The majority of practices (n = 52) 29 included lungworm in their standard anthelmintic protocols; moxidectin was the most 30 common drug used for prophylaxis (n = 71). Fenbendazole was the most frequently selected 31 drug, by 45% of vets, for treatment of confirmed cases of CA despite it being unlicensed for 32 this purpose in the UK and the absence of a clear treatment protocol. The results of this pilot 33 study provide an initial insight into the approach taken by primary care practitioners in their 34 approach to CA. This provides an important starting point for future studies investigating the 35 decision-making for CA amongst UK veterinary surgeons, particularly to clarify whether in a 36 larger cohort an unlicensed drug remains the treatment of choice. The absence of a clear 37 protocol for fenbendazole means that treatment of dogs affected by CA may be suboptimal, 38 increasing the risk of morbidity and mortality. 39 40


51
Canine angiostrongylosis (CA) is a parasitic disease caused by the cardiorespiratory 52 nematode Angiostrongylus vasorum (Superfamily: Metastrongyloidea). Dogs become 53 infected after ingesting the third larval stage (L3), usually within an intermediate gastropod 54 host or a paratenic host, such as the common frog (Rana temporaria) (Bolt et al., 1993) and severe pathology. This is particularly the case for dogs presenting with coagulopathies or 79 neurological signs in the absence of obvious respiratory signs. A major concern with CA is 80 that it may remain largely asymptomatic and can manifest as sudden death (Bourque et al.,81 2002; Brennan et al., 2004).

82
In the UK, a number of macrocyclic lactone-based products are licensed to treat 83 angiostrongylosis, containing moxidectin in a topical form (along with imidacloprid) or 84 milbemycin oxime in tablet form. The treatment efficacy for a single dose of the formulation 85 of imidacloprid/moxidectin was found to be 85.2%. There was no significant difference 86 between this and the 91.3% efficacy of fenbendazole administered daily for 20 days. The 87 former of these treatment being licensed for use against AV in dogs, whereas the latter is not.

88
Of those dogs still shedding larvae which received a further dose of imidacloprid/moxidectin, 89 all were then found to be Baermann negative (Willesen et al., 2007).   Manning, 2007), in the absence of licensing, optimal dose, frequency and duration 103 remain unclear. The risks of using unlicensed products are highlighted by levamisole, which 104 was historically advocated for the treatment of CA due to its potency and rapid onset of 105 action. However despite efficacy against CA, this is no longer recommended due to the 106 occurrence of significant side-effects including anaphylaxis in levamisole-treated patients, 107 thought to result at least in part from the rapid increase in circulating worm antigen (Søland 108 and Bolt, 1996).

109
The efficacies of many of these drugs have been studied experimentally and in a number  Data were collected for this study using a combination of an online questionnaire (Table 1) (Table   186 2). One case was excluded from this analysis due to the fact CA was only discovered at   There were a variety of treatment protocols for CA reported by the responding clinicians.

216
The most consistent was monthly moxidectin, with 19/22 (86%) of respondents advising this 217 was given to treat CA. Four respondents using milbemycin oxime to treat cases of CA 218 advised that this should be given weekly for 4 weeks, one respondent prescribed this weekly 219 for 2 weeks and one recommended a single dose to be sufficient for treatment. The remaining 220 9 respondents selecting milbemycin oxime to treat CA did not specify the frequency or 221 duration of treatment. The dosing schedule reported by 43% of the respondents for 222 fenbendazole was once daily. However, there was considerable discrepancy in duration of 223 therapy. One clinician reported 3 days, one recommended 5 days, three recommended 7 days, 224 11 two recommended 10 days and two recommended 14 days. Unfortunately, 12 clinicians using 225 fenbendazole therapy failed to provide a therapeutic frequency or duration. We asked the 19 226 clinicians reporting confirmed cases of CA to describe response to therapy. Interestingly, 227 only 4 clinicians reported that they re-tested dogs after therapeutic intervention, 228 corresponding to 6 of the 41 (15%) infected dogs. Of these six re-tested dogs, all produced 229 negative results using the in-house serological assay (Angio Detect™).  Those respondents that included lungworm prevention in their standard protocol were then 238 asked which product was chosen to a achieve this (more than one product could be selected).  which represents a limitation. This is likely to reflect the short period of data collection, 250 which was necessitated due to the timescale over which this pilot study was conducted. There 251 was a predominance of responses from those vets with fewer years in small animal practice.

252
It is difficult to know whether this reflects that newer graduates are more likely to respond to    This study found that, moxidectin and milbemycin oxime were the most frequently 311 prescribed preventative drugs, reflecting their general popularity in primary practice.

312
Alternatively, this may represent an attempt to include lungworm prevention in the standard 313 de-worming protocol of a practice, without the need to use an additional drug. In this small 314 survey, moxidectin was more frequently prescribed than milbemycin oxime. Whether this is 315 owing to its higher efficacy or simply its existing popularity in veterinary practice is unclear.

316
The majority of clinicians responded yes when asked if their practice's standard de-worming 317 protocol included lungworm prophylaxis. Whilst important that 65% of the surveyed 318 practices are including prophylaxis for CA, this means that 35% of practices do not. This may 319 relate to a perceived lower prevalence in their area, however it is now widely accepted that A. 320 vasorum is spreading throughout the UK implying that no area can be guaranteed as test. Whilst the specific reasons for this were not surveyed, it is perhaps due to its simplicity, 346 reliability and rapidity of results (Liu et al., 2017). Despite a specificity of 100%, it must be 347 remembered that with a sensitivity of 84.6% (lower than the commercially available CA 348 ELISA 94.9%) it is less suited as a screening test (Schnyder et al., 2014). Therefore given the 349 potential for a false negative result with Angio Detect™, it is important that additional 350 diagnostics are employed to effectively rule out CA when it is a major differential. The authors would like to thank Medivet Group Ltd for assistance in distributing the 366 questionnaire and to all clinicians who took part in the study. We are also grateful to Dr.         Table 1 539 Questions completed by clinicians who took part in the online survey.