Assessment of Fluralaner Treatment in Eliminating Dermanyssus Gallinae Infestation From Commercial Layer Farms, and the Potential for Derived Benets of Improved Bird Welfare and Productivity.

Background. Infestations with the poultry red mite (PRM), Dermanyssus gallinae, can result in anaemia, stress-related behaviours linked to reduced laying hen welfare, and impaired productivity. A study was conducted to investigate the potential welfare benets of PRM elimination following uralaner treatment. Methods A single poultry house was selected on each of two layer farms, one free-range, one aviary, containing 5,400 Lohmann LSL hens and 42,400 Lohmann Brown hens, respectively. Fluralaner (Exzolt®; 0.5 mg/kg body weight) was administered twice, seven days apart (Weeks 0 and 1). Mite populations were regularly monitored by traps left in place for one to two days. Infra-red cameras monitored night-time hen behaviours weekly, beginning ve weeks pre-treatment, continuing for ve and six weeks post-treatment on the free-range and aviary farms, respectively. Weekly daytime behaviours were recorded on the free-range farm. On the aviary farm, blood samples were collected from 50 randomly selected hens during Weeks -3, -1 and 5. On the free-range farm, eggs were randomly collected for corticosterone analysis during Weeks -6, 0 and 6, and on the aviary farm during Weeks -3, -1 and 5. Production parameters were assessed using farm records.

wing shaking and gentle feather pecking. On the free-range farm there were signi cant daytime reductions in head scratching, head shaking and preening. Post-treatment increases in egg and plasma corticosterone were suggestive of stress factors extra to mite infestation. Red blood cell counts and haematocrit increased following treatment. On the free range farm, relative to standard production parameters for LSL hens, the decline in laying rate with increasing bird age was less than expected, and the increase in egg weight greater than expected.

Conclusion
Fluralaner treatment eliminated mite challenge, leading to improved hen welfare and health, based on reductions in stress-related behaviours, and restoration of the anaemia-inducing effects of mite blood feeding.

Background
Infestations with Dermanyssus gallinae, the poultry red mite (PRM), an obligatory blood feeder, have a negative effect on all types of poultry production, ranging from backyard and organic farms to intensive, enriched cage or barn systems [1]. The average prevalence of PRM infestation in in European ocks has been estimated to be as high as 83% [2]. Anaemia can be a consequence of heavy infestations, and the PRM has been shown to be vector of important viral and bacterial pathogens, including Salmonella enteritidis, and avian in uenza virus [3][4][5]. Infestations have been shown to cause signi cant reductions in ock productivity, including reduced laying rate, increased percentage of downgraded eggs, increased food consumption and higher mortality rates [1,6].
Laboratory infestations of poultry have been found to be a cause of stress and impaired welfare, manifested by behavioural changes such as an increase in grooming and head scratching [7], increased levels of plasma corticosterone, adrenaline and β-globulins, and reductions in γ-globulins in infested hens compared to uninfested controls [8,9]. Thus, the ndings of moderate somatic and high psychogenic stress (corticosterone secretion and adrenaline secretion, respectively) in infested birds indicates that PRM infestations are a cause of impaired welfare, which in turn can lead to impaired productivity.
Fluralaner is an isoxazoline compound that is approved for the treatment of poultry red mite infestations by two drinking-water administrations with a 7-day interval. The mite-killing activity of uralaner is evident within 4 hours of the rst administration and is maintained for at least the next 14 days [10]. Adult mites are quickly killed, while any egg or non-feeding larvae present at the time of treatment are subsequently killed as they mature to nymphs and take a blood meal, with mite populations reported to decline by as much as 99% within 3 days of the rst administration and up to 100% within 2 days of the second administration [11]. Elimination of the mites in this way may reduce or remove the welfare and productivity effects of infestations (6,11,12]. A study was conducted in Europe to generate further evidence of the potential bene ts of elimination of PRM infestations.

Methods
The study was conducted on two commercial, egg production layer sites, one free-range farm in Germany housing LSL hens, the other an aviary farm in France with Lohmann Brown hens. The objective was to assess the effects of mite elimination by an acaricidal treatment with uralaner on behavioural welfare parameters in chickens naturally infested with D. gallinae. The potential bene ts of mite elimination on bird health and productivity were monitored by comparing pre-and post-treatment parameters and by assessing production parameters for each ock against the published standards for the age and breed of hen on each farm. All procedures were in alignment with the principles of Good Clinical Practice VICH GL9 (GCP) [13]. An informed consent was completed by the farm owners prior to any enrolment and initiation of treatment.
To qualify for the study, a farm had to have a known history and current presence of PRM infestation, along with a facility in which a PRM challenge could be monitored, to have suitable equipment for accurate delivery of uralaner via drinking water (dosing pump or medication tank), and to accept the installation of cameras to monitor bird behaviours. On each farm, a single house was used for the study. To avoid cross-contamination to the study house, the farmer was required to accept uralaner administrations of all other farm houses concurrently with study bird administrations, and to avoid use of any acaricidal product, other than the scheduled uralaner treatment, during the two weeks prior to starting and throughout the study period.
The study was initiated in July and August of 2018 in the free-range and aviary farms, respectively, and concluded in September and October of the same year. The study house on the free-range farm was naturally ventilated, with both natural and arti cial light from 04:00 until 20:00, and contained 5,400 chickens, with a mean body weight of 1.8 (± 0.1 kg). Birds were allowed to roam outside during daytime. On the aviary farm, the study house contained 42,400 birds, 66 weeks old with a mean body weight of 1.8 ± 0.2 kg at Week -5. For both farms, feed was provided according to standard procedure and water was provided ad libitum. Formal health assessments of study birds were completed by a veterinarian at weekly intervals, and general observations were made by farm staff throughout the study, with any abnormal observations to be noti ed to the veterinary investigator.

Treatment
At the time of initial treatment, in August on the free-range farm and September on the aviary farm, birds were 62 and 71 weeks of age, respectively. Before treatment it was ensured that the drinking water system worked, was free of leaks and there was no other drinking water-source available. The uralaner solution (Exzolt ® , MSD Animal Health, Germany) was administered according to label directions to achieve a dose rate of 0.5 mg/kg body weight, twice with a 7-day interval (Weeks 0 and 1) ( Table 1) [14]. The required volume of product was calculated using the body weights of 20 representative birds to estimate the total body weight of the group of chickens to be treated. That volume was added to one day's water consumption, based on farm records of the previous day's consumption. On the aviary farm a review of records indicated that an overestimation of the number of hens in the study house resulted in the labelled dose rate being exceeded by11%. The duration of treatment administration ranged from 7h45 to 10h.

Assessments of challenge with Dermanyssus gallinae
To determine the level of mite challenge, PRM traps (Avivet) were placed according to manufacturer's guidance [15]. On the free-range farm, 20 traps were placed near the lowest pole of the perches, above the manure pits, in Weeks -6, -1, 0 and 6. On the aviary farm, 24 traps were placed under perches at weekly intervals from Week -5 to Week 6. After one day (free-range farm) or two days (aviary farm), traps were collected and individually sealed in small plastic bags, each of which was then placed in a large plastic bag and stored at −18 to −20°C or colder before being sent to a laboratory with experience in identifying mites by species, and in counting and differentiating mite stages. The mites in each trap and its plastic bag were poured into a Petri dish, and mites or eggs remaining on the cardboard of the trap or in the plastic bag were thoroughly collected and added to the mites in the dish. Mite eggs, larvae and nymphs (both stages together) and adults were differentiated and counted separately. In traps with up to 250 mg of mites (total weight of eggs and mobile stages), all D. gallinae were differentiated and counted. If the weight of collected mites exceeded 250 mg, a subsample of approximately 100 mg was used.

Behavioural observations
On each farm, two cameras were used to monitor hen behaviours once per week. On the free-range farm, one camera recorded daytime behaviours, the other (infra-red) night-time behaviours. Two infra-red cameras were used on the aviary farm, where only night-time behaviours were monitored. Selection of the observation eld for each camera was based on the expected number of birds present during the observation times, approximately 50 birds were observed at the aviary farm and approximately 30 birds at the free range farm. Each recording began three hours after the onset of darkness, when mites are most active, at the same time on the same day in each week, beginning at Week -6 (free-range farm) or -5 ve (aviary farm) and continuing to Weeks 5 and 6, respectively. All camera recordings were evaluated by a single trained observer. Hens were scored, higher scores indicating greater stress, for the behaviour categories of body shaking, vertical wing shaking, gentle feather pecking, severe feather pecking, aggression, head shaking, head scratching, and preening [12]. Behavioural categories were mutually exclusive. Observations included continuous recordings of each targeted behaviour every 2 minutes for 1 minute over a 60-minute period, and scan samplings of hen resting time. The percentage of resting hens at each observation point was expressed in proportion to the total number of observed hens. The number of events for a behaviour was expressed as the number of bouts (number of times a behavioural element was observed) per hen within 15 minutes in proportion to the total number of hens observed. The number of resting and active hens was assessed by scan sampling every two minutes. Each category was measured as the number of events or animals performing the behaviour.

Haematology and stress indicators
Blood samples were collected during Weeks -3, -1 and 5 from 50 randomly selected hens on the aviary farm for complete blood counts and blood chemistry measurements. On the free-range farm, 40 eggs were randomly collected during Weeks -6, 0 and 6, and on the aviary farm in Weeks -3, -1 and 5. Corticosterone concentrations in blood samples were estimated from hens on the aviary farm, and in egg samples from both study farms. The red blood cells, haematocrit, haemoglobin concentration, mean corpuscular volume, mean corpuscular haemoglobin and mean corpuscular haemoglobin concentrations were measured using an automated hematology analyzer (ADVIA 120 hematology system, Siemens Healthineers, Spain). Stained blood smears (Diff-Quik) were examined under 100X light microscopy [16] to manually count 60 white blood cells and determine heterophil to lymphocyte (H:L) ratio. Heparin tubes were centrifuged for 10 minutes at 3000 g to obtain plasma. Corticosterone levels of egg extracted albumin and yolk and also of the plasma samples were measured by a high sensitivity EIA kit (Corticosterone HS (High Sensitivity) EIA, IDS® Immunodiagnostic Systems, Boldon, UK) following manufacturer instructions. Extraction procedure for egg albumin and yolk corticosterone were following a procedure previously described by Cook et al. [17] and results expressed as nanograms per gram [ng/g] of the freeze-dried sample taken for extraction. All tests were performed at the Interdisciplinary Laboratory of Clinical Analysis (Interlab-UMU, University of Murcia, Spain).

Production parameters
Farm production records were accessed to determine the hen weekly mortality rate (%), mean (%) laying rate, mean egg weights for each study week.

Statistical analysis
Statistical units were: the trap for the assessment of PRM infestation, the house for general health observations and performance evaluation, the observational point for behavioural evaluation, the individual animal for assessment of body weight and blood sampling and health status evaluation, and the egg for egg collection and corticosterone determination.
The antiparasitic e cacy was calculated for each post-treatment time point using the formula: where X pre is the arithmetic mean pre-treatment mite (mobile stages: larvae, nymphs, adults) count per trap, and X post is the mean post-treatment mite count per trap. Product e cacy was claimed if the percentage e cacy exceeded 90% and if the post-treatment mite counts were signi cantly less than pre-treatment counts (two sample t-test).
Changes in production data (weekly mortality rate, mean laying rate, and mean egg weight) were analysed for signi cant changes over time using a linear regression model. The slope of the regression line was compared to zero, a positive slope indicating an increase with time, a negative slope a decrease with time. Laying rates and egg weights were assessed in relation to published standard production data for the breed of bird used on each farm [18,19]. Pre-and post-treatment behavioural observations were compared using a mixed linear model. The number of bouts (or % activity of hens) was the dependent variable, study phase (pre-or post-treatment) and, for the free-range farm, day and night observation point interactions were the main effects to be investigated, study week was the repeated factor and observation point the random factor. The physiological parameters were analysed by comparing each post-treatment value with the pretreatment value using a two-sided two-sample ttest. The statistical tests were conducted using a threshold of α=0.05.

Results
There were no adverse events reported from either farm following treatment.

Counts ofDermanyssus gallinae
On both farms mite counts dropped sharply following the rst administration of uralaner, and remained very low or were 0 until the end of the study, with e cacy exceeding 99% from Week 0 on the free-range farm (P = 0.0014, Fig 1). E cacy was 100% by Week 1 on the aviary farm (P < 0.0001) and remained at that level at each posttreatment assessment. The antiparasitic e cacy was therefore achieved and the post-treatment challenge to the birds is considered to have been eliminated.

Behavioural observations
On both farms, uralaner treatment and mite elimination was followed by signi cant reductions in night-time activity (i.e., increase in the percentage of resting hens, Fig 2; (free range farm, P = 0.0175; aviary farm, P < 0.0001, where there was no signi cant interaction of study phase and observation point, P = 0.6859) and in nighttime preening, and head shaking (Table 2; Figs 3,4). Body shaking, vertical wing shaking, head scratching and gentle feather pecking at night-time were also signi cantly reduced on the aviary farm. Daytime observations from the free-range farm found signi cant reductions in head scratching, head shaking and preening. Following uralaner treatment of birds on the aviary farm, the percentage of resting hens increased quickly, from 86.7% using observations from Camera 1 and 79.5% from Camera 2 at Week -2, to >95% from Week 1 in both cameras through the remainder of the study.

Physiological and haematological analyses
Corticosterone levels in egg yolk samples from the free-range farm were 9.3 (± 6.5) ng/mL prior to treatment (Week -6), 17.7 (± 6.7) during the week of treatment (Week 0) and 17.6 (± 6.7) in samples collected in Week 6. The difference in corticosterone levels between Days 1 and 43 was not signi cant (P = 0.9321). On the aviary farm corticosterone in egg yolk and albumen and plasma increased at each assessment, from pre-treatment through the week of treatment (Week -1) to reach their highest levels at the nal assessment (Week 5) ( Table 3).

Production assessments
The weekly average mortality rate did not signi cantly change with time on either the free-range farm (0.04% to 0.24%) (P = 0.3583) or the aviary farm (0.21% to 0.44%) (P = 0.6509). Consistent with the industry standard, the laying rate of the LSL hens on the free-range farm showed a signi cant decline as the birds aged (P = 0.0054), but this decline occurred at a lower rate (0.27% per week) than the published standard for LSL hens (0.42% per week) of the same age and time in production. Egg weights on this farm on Weeks -6, 0 and 6 were 63.3, 65.9 and 66.6g, an increase that was signi cant with time (P = 0.0008), a mean increase of 0.28g per week, while according to the breed standard, an increase of 0.10g per week would be expected. The rate of decline of the laying rate of the Lohmann Brown hens on the aviary farm (0.56% per week; P = 0.0288) was similar to the mean decrease of the standard laying rate expected for birds of the same age and breed. Although the egg weights of study hens on that farm (range 65.1 to 65.6g) did not change signi cantly as the birds aged (P = 0.4065), throughout the study egg weights were approximately 3g below the breed standard weight for hens of the same age.

Discussion
This study provides further con rmation of laboratory and eld studies showing that the rapid onset and high e cacy of uralaner allows rapid control of the PRM in infested poultry ocks. On the free-range farm, the mean mite count at the end of the study was one. No mites were detected at any time after Week 1 on the aviary farm, on which a calculation error resulted in administration of a dose rate 10% higher than recommended. As uralaner e cacy on the free-range farm was equivalent to that on this farm, and as uralaner e cacy at the recommended dose rate has been consistently shown to approximate 100% on other layer farms, it can be concluded that this increase in dose rate did not affect the results, and there were no treatment-related adverse events [6, 10 -12]. Nonetheless, the error emphasizes the need for careful calculation of the appropriate product volume to ensure accurate delivery of the correct dose rate.
In the current study, the nding on both farms that mite elimination was followed by an improvement in nighttime hen resting behaviour, reductions in preening, and head shaking and, on the aviary farm, head scratching, align with similar ndings from a recent study on an enriched cage farm in Spain [12]. Similarly, the current study and the Spanish study both align with earlier work showing that PRM infestation results in increases in those stress-related behaviours and provide further con rmation of the effects of PRM infestation in reducing bird welfare [7,8]. As with the daytime ndings in our study, the report from Spain also described daytime reductions in head scratching, head shaking and preening following mite elimination.
Physiologically, those behaviours have been linked to markers of somatic stress, including increases in plasma corticosterone with the potential to lower humoral immunity, and increases in the heterophil:lymphocyte ratio [11,20]. However, in the current study blood and egg levels of corticosterone increased from pre-treatment to the week of treatment, and increased further at the nal assessments, as did the H:L ratio on the aviary farm. In the study of layer hens in Spain, post-treatment reductions in stress-related behaviours were accompanied by signi cant reductions in blood corticosterone levels and in the H:L ratio [11]. In that study, birds had been trained to accommodate handling, and the same birds were sampled on each occasion. In contrast, in the current study, birds received no such training and samples were collected from different birds on each occasion. As described previously, the stress of restraint and blood collection from untrained birds may account for the increases in physiological stress markers seen in the current study [21]. There may also have been other unrecognized stress factors affecting study birds. Thus, removing the irritant factor of an infestation may have reduced external stress behaviours, but other factors may have driven the increases in the stress-physiology ndings.
Whether clinical or subclinical in appearance, anaemia is a well-established consequence of infestation with the PRM. A laying hen may lose 3% of its blood volume every night due to exposure to heavy infestations, while subacute anaemia up to death through severe anaemia has been reported [3,22,23]. The signi cant increase in red blood cell counts and haematocrit from samples on the aviary farm indicate that removal of the PRM challenge by the uralaner administrations resulted in a restoration of those blood values, providing further con rmation of the blood-draining impact of infestation.
The effects of PRM infestations in interfering with layer productivity have been documented in a number of studies. Affected parameters have included increases in hen mortality percentage and in the proportion of downgraded eggs, and reductions in egg weight and laying rate [2,3,11,12]. There was no noticeable change in mortality rates between pre-and post-treatment periods on either farm. On the free-range farm, based on the industry standard for LSL hens, the expected decline in laying rate over time was lower than expected, and the rate of increase in egg weights was greater than expected. This was less the case on the aviary farm where, throughout the observation period, the laying rate paralleled the industry standard for Lohmann Brown hens, while the rate of increase in egg weights remained below the breed standard, both before and after mite elimination. Overall, the results suggest that any production-impairing effects of the PRM infestation were removed by treatment. That conclusion is consistent with a report in which egg laying rates on seven of eight layer farms improved by up to 12.6% following PRM elimination by treatment with uralaner, compared with the laying rates observed in infested control birds that were either untreated or that received rescue treatments [11].
The primary objective of demonstrating the bene ts of PRM elimination in reducing the frequency or duration of stress-related bird behaviours was achieved. Although production improvements were a secondary objective in this study, positive bene ts were suggested by the results from the free-range farm, if not so clearly seen in results from the aviary farm. Importantly, the results of the current study provide further con rmation of the bene cial effects of PRM elimination on bird welfare.

Conclusion
On two commercial poultry farms affected with heavy PRM infestations, two drinking water administrations of uralaner, 7 days apart, eliminated mite challenge and reduced the incidence of stress-related layer hen behaviours. For the assessment of corticosterone a different surrogate than blood is recommended in untrained birds. Further work is needed to determine the productivity bene ts suggested by the production data of treated The study was funded by MSD Animal Health.
Availability of data and materials The datasets generated and analysed during the current study are available on reasonable request to MSD Animal Health Innovation, Schwabenheim Germany.

Ethics approval and consent to participate
The study was announced to the respective local governmental bodies and was in alignment with local, national and global guidelines. The farmers were fully advised of all study procedures prior to signing informed consent and study start.

Consent for publication
Not applicable.
Competing interests IP, EZ and ET are employees of MSD Animal Health.