Investigation of suspected moxidectin resistance on a boarding farm in North Carolina.
D. Little and R.D. Metcalf.
Introduction
The cyathostomins are widely recognized as the gastrointestinal parasite of the horse that have the greatest pathogenic potential (Love et al 1999), but control can be successfully achieved by anthelmintic treatment of horses. Unfortunately, anthelmintic resistance in the cyathostomins is becoming more widespread. Resistance to the benzimidazoles has been recognized around the world for over 20 years, and resistance to the pyrantel salts is becoming increasingly widespread, and is highly prevalent in the southeastern United States (Kaplan et al, 2004). Resistance to the third group of drugs effective against the cyathostomins, the macrocyclic lactones has not yet been reported, despite concern that increased use of moxidectin would enhance development of resistance to ivermectin in cyathostomins (Sangster 1999). The aim of this report is to describe the results of an investigation into suspected failure of moxidectin to adequately control cyathostomins in horses resident on a boarding farm.
Materials and Methods
Study farm and history
A mixed breeding and boarding facility was identified in January 2004 after multiple faecal worm egg counts (FWEC) using a modified McMaster technique with a lower level of detection of 25 eggs per gram (epg) by one of the authors (RM) on a 2-year old filly boarded at the farm demonstrated 275-1000 eggs per gram (epg), approximately 6-8 weeks after moxidectin (Quest) treatment 1.
Parasite control on the farm from approximately 1990 until early 2000 had involved the use of a rotation of pyrantel pamoate and ivermectin at 6-8 week treatment intervals in mature horses. In early 2000, all mature horses were moved to a treatment program of moxidectin every 8 weeks. Foals born on the farm were treated with 10mg/kg fenbendazole at three months of age. Prior to 2000, foals were then treated with a rotation of pyrantel pamoate and ivermectin every 6-8 weeks after the age of three months. After 2000 they were given 5mg/kg fenbendazole once monthly until the age of 12 months at which time they entered the mature horse program of moxidectin treatment every 8 weeks. No faecal worm egg counts (FWEC) were performed on the farm until January 2004, and weight-estimation of the horses was not performed prior to treatment.
At the beginning of the study, there were 46 horses on the farm. There were three 6-7 year old resident stallions. Thirteen horses on the farm were less than four years of age. The remaining 30 horses on the farm had a mean age 11.4 (range 5-22 years). The horses were of a variety of breeds, but predominantly Paint, Quarter horses or crosses of these breeds. Five mature horses left the farm during the study.
Grazing on the farm was divided into six pastures; Pastures 1-3 each housed one of the three resident stallions. Pasture 4 was reserved for foals and weanlings from mixed sources, some were homebred and others were boarding horses. Horses from this pasture are moved to grazing used by mature horses when they become yearlings, if they remained on the farm. Approximately 22 horses of mixed age and sex grazed Pasture 5; these horses were mainly boarding horses, and grazed approximately 12 hours each day. The farm held weekly rodeos and monthly shows through the summer. During the shows, horses from pasture 5 were stalled, and this pasture was used for the show, to which visiting horses had access. Pasture 6 housed 14 resident boarding horses, but during the rodeos, rodeo horses were kept in this pasture, and in one of the stallion pastures. There was no attempt made on the farm to rest or rotate grazing, and faeces were not removed from pastures; instead faeces from stalled horses was spread directly onto pasture 5.
Farm investigation
At the start of the study, owners were contacted in writing and were advised of the potential implications regarding the apparent treatment failure of moxidectin. They were also advised that if the FWEC in any individual horse was found to be unacceptably high they would be informed, the individual horse removed from the study and treated as deemed appropriate. The regular farm veterinarian was also kept informed and cooperated fully with the investigation. One horse was kept on daily pyrantel tartrate (2.6mg/kg PO) at the request of the owner. FWEC were initially evaluated on 9 horses. Weight was estimated on all horses treated using a weigh-band with an extra 10% added to minimize the risk of under-dosing from under-estimation of weight using the weigh-band. All horses on the farm were treated with the recommended dose of moxidectin the same day of the initial FWEC. Eight horses had FWEC 14 days after moxidectin administration in order to obtain data for initial FWEC reduction tests (FWECRT)
These horses were also subjected to egg reappearance period (ERP) testing, where the time to reappearance of eggs in faeces was monitored by FWEC every 2 weeks. All other horses were randomly allocated to one of four treatment groups once their FWEC was ≥ 100epg for horses < 4 years old, or when ≥ 200epg for horses ≥ 4 years old in order to test the efficacy of other anthelmintics on the farm. Treatment groups were ivermectin, pyrantel pamoate or fenbendazole at their recommended doses, or fenbendazole 10mg/kg once daily for 5 days followed by ivermectin (0.2mg/kg PO) on day 6. FWECRT were done approximately 14 days after last drug administration for all groups. Egg reappearance periods (ERP) were also recorded for these drugs when possible. Duration of the study was 8 months.
Analysis
A FWEC was identified as positive when FWEC ≥ 100epg for horses < 4 years old, and when ≥ 200epg for horses ≥ 4 years old. These thresholds were used as points of entry into a FWECRT or for targeted drug treatment. Results of FWECRT for an individual horse were calculated as a percentage using the formula:
(FWEC pre-treatment – FWEC post-treatment) x 100
FWEC pre-treatment
A FWECRT of <95% was used to determine whether resistance was present, because in a previous study Craven et al (1998) recommended that a 95% reduction provided the earliest indication of field resistance. Furthermore, all data were calculated as arithmetic mean in order to give a better estimate of the overall faecal worm egg output (Craven et al 1998). In order to increase sensitivity of ERP, the authors elected to use the definition of ERP as the time at which any eggs were detected by FWEC after anthelmintic treatment. The classification system of the National Animal Health Monitoring System (Wineland 2000) was used in order to assess whether horses were shedding cyathostomin eggs onto pasture at a high (≥600epg), moderate (300-590 epg) or low level (10-290 epg) at any point through the study period.
Results
Of horses that were less than 4 years of age, 80% shed high levels (>600epg) of strongyle eggs onto pasture. One horse (1/15, 6.6%) shed moderate levels (300-590 epg) onto pasture, and 2/15 (13.4%) shed low levels (10-290 epg) onto pasture. Of horses 4 years of age or greater, 13/30 (43.3%) shed high levels, 3/30 (10%) shed moderate levels, 8/30 (26.7%) shed low levels, and 6/30 (20%) did not shed eggs onto pasture.
Drug resistance to fenbendazole and pyrantel pamoate was identified on the farm by FWECRT. Drug resistance to moxidectin or ivermectin, or to five days treatment with fenbendazole (10mg/kg) and ivermectin (0.2mg/kg) treatment on day 6 was not identified by FWECRT (Table 1). ERP (Figure 1) for ivermectin were less than 8 weeks in all horses less then 4 years of age, and on 78.6% occasions when used in horses 4 years old or greater. ERP for moxidectin were less than 12 weeks (84 days) on 75% occasions when moxidectin was used in horses less than 4 years of age, and on 55.8% occasions when used in horses 4 years old or greater. Combination larvicidal drug treatment with fenbendazole and ivermectin failed to suppress faecal egg output for more than 8 weeks in any age group. ERP were monitored in five horses after 2 treatments with moxidectin, and in 6 different horses after 2 treatments with ivermectin, spread over the study period. Anoplocephala perfoliata ova were identified during FWEC in 1/12 (8.3%) horses <4 years old and in 12/33 (36.4%) horses ≥ 4 years old.
Discussion
Compared with the results of a previous study of 8516 faecal samples from horses on 1178 operations in the United States in 1998 (Wineland 1998), 5 times as many mature horses on this farm shed high levels of strongyle eggs onto pasture. This suggested that the level of pasture contamination was very high as a result of poor pasture hygiene, or that the drugs used to control strongyles on this farm were ineffective, or both.
Pasture hygiene on this farm was considered to be inadequate. Mature horses were grazing with immature horses and therefore acting as a source of infection to immature horses with a lower level of immunity and a greater propensity to encyst large numbers of larvae. The pasture that was used predominantly for foals, weanlings and yearlings was not rested sufficiently between foal crops. Large numbers of horses visited the farm on a weekly basis during the summer to participate in rodeos, or monthly for horse shows, and it was interesting to note that the stallion with the highest FWEC was the stallion whose pasture was also used by rodeo horses. Manure from stalls was spread directly onto pastures without composting. Fields were not periodically rested, or harrowed and rested, or cross-grazed with other species. Other problems with the management identified on this farm that are known to predispose to development of resistance in other species were failure to use a weigh-band to estimate body-weight, potentially leading to under-dosing, and the frequent use of moxidectin in all horses greater than 1 year of age at 8-week intervals instead of the recommended 12-week (84 days) treatment interval, increasing selection pressure for development of resistance.
Faecal worm egg count reduction tests (FWECRT) are most commonly used to assess anthelmintic treatment efficacy in horses, but are insensitive, because they do not reliably detect the presence of resistance until the proportion of resistant worms is greater than 25% (Coles et al 1992). However, FWECRT did detect both fenbendazole and pyrantel pamoate resistance in cyathostomin populations on this farm. This is not surprising given the recent study by Kaplan et al (2004), where resistance to fenbendazole and pyrantel was found to be highly prevalent through the southeastern USA.
Avermectin/milbemycin resistance in cyathostomins of the horse has not yet been reported. However, several species of cyathostomins have recently been found to carry at least two of the adenosine triphosphate-binding transporter P-glycoprotein genes that have been found to be involved in development of ivermectin resistance in Haemonchus contortus (Drogemuller et al 2004), suggesting that development of resistance to ivermectin is possible in the cyathostomins. This early evidence that multiple genes are likely to be involved in ivermectin resistance may explain why ivermectin resistance has not previously been reported in cyathostomins despite many years of drug use. Genetic resistance is not the only mechanism of surviving drug treatment available to a population of cyathostomins, but it is currently almost impossible to distinguish between true drug resistance and strategies such as anthelmintic avoidance, where different species of cyathostomins are not exposed to a given drug treatment because they are encysted in mucosal larval stages against which the drug is ineffective, or because they are in faecal piles on pasture. Repeated, frequent use of ivermectin and failure to detect resistance at an early stage have been hallmarks of development of ivermectin resistance in other species (Shoop 1993). Sangster (1999) suggested that shortening ERP after anthelmintic treatment would be detected in drug resistant populations of cyathostomins prior to detection of resistance by FWECRT. Typical ERP after ivermectin treatment are 6 - 8 weeks (Tarigo-Martinie et al 2001). However, ERP of as little as 3 weeks after ivermectin treatment have been identified previously by Little et al (2003) in a herd of horses where cyathostomins were also found to be resistant to both fenbendazole and pyrantel pamoate. In another study by Tarigo-Martinie et al (2001) FWEC had returned to 20% of pre-treatment values by 6 weeks after ivermectin treatment on 30% of farms evaluated. Moxidectin should suppress faecal worm egg counts (FWEC) for 12 weeks (84 days) (Di Pietro et al 1996). In this study there was clear evidence of shortened ERP after either ivermectin or moxidectin treatment in both immature (< 4 years of age) and mature horses (≥ 4 years of age). The shortened ERP following ivermectin treatment did not appear to be ameliorated using a combination of fenbendazole and ivermectin treatment. This was disappointing, because combination drug treatment strategies have been proposed as a mechanism by which the deleterious effects on health of a population of resistant parasites may be reduced, particularly if the dose or treatment duration of the drugs is increased when safe to do so. However, it is not surprising that additional control was not achieved in this case, given the high level of resistance to 5mg/kg fenbendazole already present on the farm (Table 1).
In order to distinguish between drug avoidance and true genetic resistance in a population of cyathostomins, molecular techniques must be employed, and while they are becoming more widely used to study the mechanisms of resistance and population dynamics of different cyathostomins populations, they are not yet applicable to field investigations of suspected resistance problems. ERP may be shortened by rapid re-infection from heavily contaminated pasture, in which species of cyathostomins with a short pre-patent period (time from ingestion of egg to production of eggs by mature adult parasite) completes its life-cycle within the treatment interval of the drug. However, in studies evaluating development of drug resistance by computer modelling, development of genetic resistance was more rapid on heavily contaminated pasture (Barnes et al 1995). Alternatively, emergence and maturation of encysted larvae after drug treatment but within the treatment interval could contribute to a shortened ERP. If this is the case, then the ERP for moxidectin would be expected to be longer than ivermectin because moxidectin has greater efficacy against encysted larval stages than ivermectin (Sangster 1999, Reinemeyer 1998). However, in this study, the ERP for ivermectin and moxidectin were comparable at 3-8 weeks post-treatment.
We cannot eliminate factors other than anthelmintic resistance causing shortened ERP after both moxidectin and ivermectin treatment, but in light of the devastating health, welfare and epidemiological implications of ignoring shortened ERP, and waiting until resistance to ivermectin and moxidectin is diagnosed by FWECRT, the authors speculate that a low level of resistance to ivermectin and moxidectin is present on this farm, currently is at a level below the detection limit of the FWECRT. We recommend that urgent changes in pasture management be made to slow the spread of this problem through the herd, the use of targeted anthelmintic treatments, and the intensive use of FWECRT and ERP as a means of monitoring for increasing levels of resistance.
Acknowledgements
The authors would like to thank Fort Dodge Animal Health (FDAH) and Dr John H. Tuttle of FDAH for their cooperation with this study, and also for funding this study. In addition, the authors would like acknowledge the Central Carolina Equine Practice, Dr Ghiloni for their assistance with the study.
Manufacturers’ addresses
- Moxidectin 2% gel, Quest, Fort Dodge Animal Health, Overland Park, Kansas, USA
References
Barnes, E.H., Dobson, R.J., Barger, I.A. (1993) Worm control and anthelmintic resistance: Adventures with a model. Parasitology Today 11, 56-63.
Coles, G.C., Bauer, C., Borgsteede, F.H.M., Geerts, S., Klei, T.R., Taylor, M.A., and Waller, P.J. (1992) World Association for the Advancement of Veterinary Parasitology (WAAVP) methods for the detection of anthelmintic resistance in nematodes of veterinary importance. Vet Parasitol. 44, 35-44.
Craven, J., Bjorn, H., Barnes, E.H., Henriksen, S.A. and Nansen, P. (1999) A comparison of in vitro tests and a faecal egg count reduction test in detecting anthelmintic resistance in horse strongyles. Vet Parasitol. 85, 49-59.
DiPietro, J.A., Hutchens, D.E., Lock, T.F., Walker, K., Paul, A.J., Shipley, C., Rulli, D. (1996) Effect of moxidectin oral gel on post treatment strongyle egg count suppression in horses. Proceedings of the American Association of Veterinary Parasitologists 41st Annual Meeting, Louisville, Kentucky, p32.
Drogemuller, M., Schnieder, T,. von Samson-Himmerstjerna, G. (2004) Evidence of p-glycoprotein sequence diversity in cyathostomins. J Parasitol. 90, 998-1003.
Kaplan, R.M., Klei, T.R., Lyons, E.T., Lester, G., Courtney, C.H., French, D.D., Tolliver, S.C., Vidyashankar, A.N., and Zhao, Y. (2004) Prevalence of anthelmintic resistant cyathostomes on horse farms. J Am Vet Med Assoc 225, 903-910.
Love, S., Murphy, D., and Mellor, D. (1999) Pathogenicity of cyathostome infection. Vet Parasitol 85,123-133.
Reinemeyer, C.R. (1998) Practical and theoretical consequences of larvicidal therapy. Equine Practice 20, 10-13.
Sangster N.C. (1999) Pharmacology of anthelmintic resistance in cyathostomins: will it occur with the avermectin/milbemycins? Vet Parasitol 85, 189-204.
Shoop, W.L. (1993) Ivermectin resistance Parasitology Today 9, 154-159.
Wineland, N.P.L. (1998) National Animal Health Monitoring System. Internal Parasites & U.S. horses: Strongyles. U.S. Department of Agriculture, Washington DC, USA.
Table 1: FWECRT results for horses less than 4 years of age (A) or 4 years of age and greater (B). - = decrease in FWEC post-treatment compared to pre-treatment; + = increase in FWEC post-treatment compared to pre-treatment.
A
|
Anthelmintic
|
Number of FWECRT
|
Number horses tested
|
Mean FWECR
(%)
|
Range (%)
|
|
Moxidectin
|
6
|
6
|
-100
|
All -100
|
|
Ivermectin
|
9
|
6
|
-100
|
All -100
|
|
Pyrantel Pamoate
|
4
|
4
|
-64
|
+70 – [-90.2]
|
|
Fenbendazole
|
6
|
6
|
+158.5
|
+566.6 – [-18.1]
|
|
Fenbendazole & ivermectin
|
6
|
6
|
-100
|
All-100
|
B
|
Anthelmintic
|
Number of FWECRT
|
Number of horses tested
|
Mean FWECR (%)
|
Range (%)
|
|
Moxidectin
|
14
|
12
|
-100
|
All -100
|
|
Ivermectin
|
10
|
8
|
-100
|
All -100
|
|
Pyrantel Pamoate
|
4
|
4
|
-66.1
|
-3.2 – [-95.4]
|
|
Fenbendazole
|
3
|
3
|
+146.9
|
+200 – [+73.9]
|
|
Fenbendazole & ivermectin
|
1
|
1
|
-100
|
-100
|
Figure 1a: Egg reappearance period as assessed by time (weeks) after treatment with ivermectin, moxidectin or 5 days fenbendazole (10mg/kg) and ivermectin (0.2mg/kg) on day 5 in horses <4 years old.
|