Reduced ecacies of diclazuril and toltrazuril against Eimeria ovinoidalis and Eimeria crandallis in two French sheep-meat farms

Background: Lamb coccidiosis, caused by intestinal parasites of the Eimeria genus, has pronounced health and economic impacts throughout the world. Eimeria ovinoidalis and Eimeria crandallis are the most pathogenic species in sheep. Control of these parasites requires the use of anticoccidial drugs such as sulfonamides, diclazuril, and toltrazuril. In this study, resistance to anticoccidial drugs was suspected in two farms as clinical signs (diarrhea) persisted after treatment. Method: On each farm, 4.5-month-old rams were divided into three groups so that they were either (i) left untreated (Control group), (ii) treated with diclazuril (1 mg/kg body weight), or (iii) treated with toltrazuril (20 mg/kg body weight). Animals were treated at day 0 (D0) and fecal samples were collected at D0 and eight days later (D8) to evaluate the reduction in intensity of Eimeria oocyst excretion. Oocyst morphology and morphometry were used to identify Eimeria species at both sampling dates. Results: Reduction of oocyst shedding was incomplete in both farms (92.44% and 93.58%) after diclazuril treatment. More specically, the ecacy was reduced in both farms for E. ovinoidalis/Eimeria marsica (92.59% and 91.87%) and E. crandallis/Eimeria weybridgensis (75.34% and 80.10%). The general ecacy of toltrazuril was high in both farms (97.6% and 97.96%). However, a slightly reduced ecacy was noted in farm 1 for E. crandallis/E. weybridgensis (93.26%) while this ecacy was high in farm 2 (98.88%). Conclusions: We suggest a simple protocol to investigate the ecacy of anticoccidial treatments in sheep and to rapidly identify potentially resistant species. In these two farms, treating animals with diclazuril will select pathogenic species, and toltrazuril could favor resistant E. crandallis/E. obtain the number of Eimeria spp. oocysts per of feces.

infected animals, where a wide diversity of species is usually found. Moreover, molecular identi cation is still underdeveloped for ovine Eimeria species, even though partial genomic sequencing of the ITS-1 region is now available [23].
Thus, this study aimed to i) propose a simple protocol to evaluate the e cacy of toltrazuril and diclazuril in the treatment of lamb coccidiosis in eld conditions and ii) propose simple keys for morphological identi cation of oocysts of sheep Eimeria species. In this study, we investigated two farms with coccidiosis management issues.

Methods
Breeding selection, farming context, and necessity to control coccidiosis In France, the sheep-meat selection scheme is based on a breeding program supervised by a breeders' association. It is based on the use of a core group of farms selected in the breed population. Each year, within a breed, crossbreeding is carried out between dams of sires and sires of sires, and the best male lambs resulting from these crossbreedings are collected after weaning in a common breeding center for evaluation called an "individual control station" (ICS). In an ICS, lambs are tested on criteria of interest for sheepmeat production such as average daily weight gain and fat and muscle thickness.
During 2021, two of these ICS were investigated in Berrichon du Cher (BC) and in Rouge de l'Ouest (RO) breeds, both related to the breeders' association GEODE. Lambs were collected in the ICS at three months of age. On arrival, they received a diclazuril treatment (Vecoxan ND, MSD Santé Animale, 1 mg/kg body weight (BW) and an ivermectin treatment (Baymec ND, Norbrook Laboratories, 200 µg/kg BW) to control coccidiosis and Strongyloides papillosus infections, respectively. The lambs came from different farms of origin: in 2021, we enumerated 9 different origins for a total of 55 BC lambs entering the ICS and 12 different origins for the 52 RO lambs.
Throughout the evaluation period in the ICS, the lambs were kept exclusively indoors and fed ad libitum. During their growth phase, they were regularly weighed as part of the protocol to evaluate their growth abilities, and when signs of diarrhea appeared on several lambs, an anticoccidial treatment was carried out on all rams with diclazuril (Vecoxan ND, MSD Santé Animale, 1mg/kg BW) or sulfonamides (Sulfadimerazine 33% ND, Huvepharma, 90 mg/kg BW over three successive days).
In this very particular breeding context, where coccidiosis must be controlled as much as possible to allow the lambs to express their full genetic background, BC and RO breeders noticed the persistence of diarrhea after the anticoccidial treatment and observed that it was necessary to treat more frequently than in previous years. These observations have been recurrent for several years, despite implementation of a sanitary vacuum and disinfection of the breeding center between two years of male lamb evaluation.
A protocol to evaluate the e cacy of anticoccidial treatments was then implemented in both ICS.

Sampling and analysis
The Fecal Oocyst Count Reduction Test (FOCRT) started 43 days after the animals were recruited in the ICS. The protocol to test the e cacy of anticoccidial drugs was based largely on the protocol proposed by Odden et al. [14], with some changes to adapt it to our breeding context. The animals were divided into three groups of 11 to 16 lambs, with one group for each treatment that was tested. The groups were constituted so as to be homogeneous in terms of weight and farm of origin. The treatments under evaluation were: diclazuril (Vecoxan ND, MSD Santé Animale, 1 mg/kg BW orally) and toltrazuril (Baycox ND, Elanco, 20 mg/kg BW or Toltranil ND, KRKA, 20 mg/kg BW, both orally). An untreated group was added in both breeds. Individual fecal samples were taken from each lamb on the day of treatment (D0) and eight days after the treatment (D8).
Individual microscopic analyses were performed on all series of fecal samples. Eimeria spp. oocysts were counted using a modi ed Raynaud's method [24]. Brie y, three grams of feces were diluted in a saturated sodium chloride solution with a density of 1.2 g/mL before being ltered three times through a tea strainer. The ltrate was then analyzed on a McMaster slide. The oocysts in both grids were counted, summed, and multiplied by 50 to obtain the number of Eimeria spp. oocysts per gram of feces. The detection limit of this technique was 50 oocysts per gram (OPG) of fecal material.
In parallel, identi cation of the Eimeria species was performed per treatment group and per date (D0 and D8). To do so, all individual ltrates of a group were mixed in a glass ask. After homogenization, an aliquot of the total suspension was placed in a test tube, lled to the brim, and covered with a coverslip. The otation time lasted 15 minutes before examination of the slide. The Eimeria species were then identi ed by the morphological and morphometric criteria of their oocysts as de ned by Eckert (Eckert, 1995) at 400x magni cation. The oocyst measurements were performed with Zeiss image processing software Zen (Zen 2.6 blue edition, Carl Zeiss Microscopy GmbH, 2018).
Species identi cation was performed on simple and relevant criteria on non-sporulated oocysts. The identi cation pathway is presented in Figure 1. First, we checked the presence/absence of a polar cap on the oocyst. When a polar cap was present on the oocyst, the total length of the oocyst was measured to distinguish two sub-categories of Eimeria: species with large oocysts (> 30 µm) and species with oocysts of medium size (< 30 µm). Among the Eimeria species that have large oocysts with a polar cap, two species were readily distinguishable: Eimeria intricata, with a brown, thick wall, and Eimeria ahsata, with a very prominent polar cap. When oocysts had a medium size and a polar cap, the total length of the oocysts ( Figure 2) could differentiate the two following clusters: Eimeria crandallis/Eimeria weybridgensis with smaller oocysts (total length ranging from 17 to 31 µm for a mean size of 22-24 µm) versus Eimeria granulosa/Eimeria bakuensis (from 22 to 37 µm for a mean size of 30-32 µm).
When the oocyte did not have a polar cap, the presence/absence of a micropyle was checked. Oocysts of Eimeria pallida and Eimeria parva species were readily identi ed as their oocysts lacked both a polar cap and a micropyle, had a round shape, and were small in size (12 to 22 µm) compared to all the other species. When a micropyle was present, the shape of the oocyst was of interest: an oval shape indicated the Eimeria ovinoidalis/Eimeria marsica cluster whereas a poultry egg shape was characteristic of Eimeria faurei.
In each group, we identi ed one hundred oocysts before and after treatment in order to have an accurate proportion of the species composition. However, in some groups after treatment, a smaller number of oocysts were identi ed due to the low number excreted after treatment. In this case, we identi ed as many oocysts as possible.

Calculation of the e cacy of anticoccidial treatments
The percentages reduction of the intensities of Eimeria oocyst excretion were estimated by the Kochapakdee formula [25]: E cacy = (1-arithmetic mean OPG of the treated group at D8 / arithmetic mean OPG of the treated group at D0) * 100 First, this formula was used on the total number of oocysts counted on McMaster slides at D0 and D8 to obtain the overall e cacy of the drug. According to the WAAVP guidelines for anthelminthic resistance, treatment resistance occurs when the percentage of reduction is less than 95% and the lower 95% con dence interval (CI) is less than 90%. To calculate the con dence intervals, we used the following formula: CI (95%) = (mean of e cacy in treated group ± 1.96 (Standard Variation of e cacy in treated group / √number of animals in treated group) Secondly, we evaluated the e cacy of each drug against individual Eimeria species or clusters. Following the species identi cations, the proportion of each species was calculated in each treatment group. This proportion was plotted against the total number of oocysts counted to obtain an estimate of the number of oocysts of each species in each group. Then, the same formula [25] was used for these estimated numbers of oocysts at D0 and D8 to obtain the e cacy of a given drug against each Eimeria species.

Results
The overall e cacies of the different drugs are listed in Table 1. At D0, all individuals excreted Eimeria oocysts, but the intensities of the oocyst excretions exhibited high individual variabilities in each group (see Supplementary Data). The mean intensities of the oocyst excretions at D0 were very similar in the control and treated groups for each ISC (between 4 518 OPG and 7 514 OPG for BC lambs and 15 018 and 16 850 OPG for RO lambs). At D8, the average intensities of the oocyst excretions in the control groups were equal or slightly higher than at D0 (5 477 OPG at D8 compared to 4 518 OPG at D0 in BC rams and 25 453 OPG compared to 16 075 OPG in RO rams). All groups treated with the various anticoccidial drugs exhibited a reduction in the intensity of excretions at D8, including some lambs that were no longer excreting Eimeria oocysts. In BC lambs, the e cacy of toltrazuril was high, at 97.  Table 2. These proportions were then used to obtain an estimated number of oocysts of each species in the group. Within the same ISC, the same species were identi ed in the different groups at D0. In the BC groups, the majority of the identi ed species were the clusters Eimeria pallida/Eimeria parva, Eimeria ovinoidalis/Eimeria marsica, Eimeria granulosa/Eimeria bakuensis and Eimeria ahsata. In the RO groups, the dominant clusters or species were E. ovinoidalis/E. marsica, E. granulosa/E. bakuensis and E. ahsata. A smaller proportion of oocysts identi ed as Eimeria crandallis/Eimeria weybridgensis was also found in all groups of both ISC. In contrast, Eimeria faurei and Eimeria intricata were only present in small proportions or even absent from some groups at D0, such as in the diclazuril BC group.

Discussion
There are currently no guidelines available to evaluate the e cacy of anticoccidial drugs in sheep coccidiosis in the eld. Some general recommendations have been made by the WAAVP [26], and protocols to evaluate the e cacy of anticoccidian drugs have been proposed [8, 14,16,19,20] with substantial variations in terms of group constitution, age of the animals at the time of testing, and natural infection levels.
In our context, the need for effective anticoccidian treatment is important not only to allow the lambs to express their growth potential, but also because, after three months together in the ICS, they are sold to different breeders. If previous treatments have not been effective, then they may contaminate their new farm with resistant and potentially pathogenic Eimeria species.
In our study, male lambs from different origins were corralled in the breeding center at three months of age and were treated at the time of their arrival with diclazuril. Therefore, it was not possible to perform the FOCRT immediately upon their entry. Odden et al. [14] proposed the use of twin lambs, one in a treated group and the other one in the control group. In our study, this was not possible. However, other studies did not take this into account [8,16,19,20] Nevertheless, although evaluation of the overall e cacy of a drug is easy, estimation of the e cacy against each Eimeria species is nonetheless an onerous undertaking and the pathogen must be present at the beginning of the test and at a su cient level in the treated groups. Indeed, progressive immunity of the animal is established for the majority of Eimeria species [2]. It is then possible that FOCRT cannot be performed on all species because they may be absent at the beginning of the test or present at low levels due to the age of the animals and possible immunity.
Taking into account all these considerations, we propose a simple protocol to evaluate the e cacy of anticoccidial drugs in farms, as well as an easy-to-use key for the identi cation of Eimeria species by microscopy.
At the beginning of our study, the intensities of excretion were relatively close between groups within the same farm and high enough to determine treatment e cacy. The untreated groups ensured the natural dynamics of Eimeria oocyst excretion as well as the natural change in the proportion of species in the two breeding centers during the duration of the test. In our case, the intensities of Eimeria oocyst excretions were equal or even slightly higher in the two control groups of the ISC from D0 to D8. Odden et al. (2018a) proposed in their protocol that the best period to evaluate the e cacy of anticoccidial treatments is when the intensity of oocyst excretion increases signi cantly from the beginning of the trial to the end. However, in their case, the animals were younger ("more than 14 days old") and had never received anticoccidial drugs before the test. In our study, we performed the test on older animals that had undergone anticoccidial treatment 43 days before the beginning of the test. In our case, it seems di cult to obtain an excretion intensity that increases signi cantly between the beginning and the end of the FOCRT, but the maintenance of this excretion intensity between the two dates seems like an acceptable compromise that still allows interpretation of the FOCRT.
Our results show that in both ISC, the toltrazuril treatment was effective at reducing the overall intensity of Eimeria oocyst excretion. These results are very similar to those obtained in previous studies [8,16,[20][21][22]27]. The percentages of reduction of oocyst excretions were 92.44% and 93.58% with diclazuril, which is lower than the minimum threshold of e cacy (95%) required for anthelmintic treatments [28].
In studies where the e cacy of diclazuril was not investigated, this drug exhibited an e cacy higher than 97% at four to seven days after treatment [8,18,27,29]. In our case, the e cacy was slightly lower.
Regarding the e cacies of the drugs against each species, we noticed that although the general e cacy of toltrazuril was higher than 95%, some FOCRT were lower for some species. This was the case in the BC ICS, where the reduction was below 95% for the species E. crandallis/E. weybridgensis and Eimeria faurei. In this group, only a few oocysts of E. faurei were identi ed at D0 and D8, which likely led to inaccuracy in the calculation of the FOCRT, and the result should hence be interpreted with a degree of caution. By contrast, the number of oocysts identi ed as E. crandallis/E. weybridgensis was substantial at the time of the toltrazuril treatment, and we can assume that there was a real lack of reduction for this cluster of species. In the RO ICS, toltrazuril exhibited a loss of e cacy toward E. parva/E. pallida but remained effective toward the pathogenic species.
In the BC ICS, reductions of oocyst excretions below 95% were noted for the species E. ovinoidalis/E. marsica, E. crandallis/E. weybridgensis, and E. granulosa/E. bakuensis after diclazuril treatment. At D8, only a few oocysts (29 oocysts) were identi ed in this group because the treatment drastically reduced the total intensity of excretion. The proportions were, therefore, less accurate than if they had been estimated with 100 oocysts. Nevertheless, the majority of identi ed oocysts at D8 belonged to the pathogenic clusters in the diclazuril group. In the RO ICS, diclazuril no longer appeared to be fully effective against E. parva/E. pallida as well as against the pathogenic species E. ovinoidalis/E. marsica and E. crandallis/E. weybridgensis. In these situations, it is di cult to conclude that one of the pathogenic species is resistant due to the di culty distinguishing it from a less pathogenic morphologically related species. The development of molecular tools could ll this gap. For the time being, in sheep, molecular techniques, especially realtime Polymerase Chain Reaction, are not as advanced as in poultry [30]. Although some genomic sequences are available [23,31], to our knowledge, no routine technique has been developed to date.
Thus, we cannot presently draw rm conclusions regarding the actual resistance of these species of sheep Eimeria to diclazuril or toltrazuril, although it is clear that there is a lack of e cacy in these two breeding centers. Indeed, resistance is normally con rmed by experimental infections in naive lambs under coccidia-free conditions, with oocysts recovered post-treatment and submitted to a new experimental FOCRT [15]. Unfortunately, we do not have the resources to validate the resistance by this method. Nevertheless, the maintenance of clinical signs after treatment observed by farmers (diarrhea and slowing of weight and mass gain) seems to indicate a signi cant persistence of pathogenic species after treatment.

Conclusion
In this study, we evaluated the e cacy of diclazuril and toltrazuril on two sheep-meat farms in France. Using an identi cation protocol for the ovine Eimeria species, we estimated the overall e cacy of the treatments based on reduction of the total intensity of oocyst excretion, as well as by species. The results of this study show that in the two investigated farms, there was an overall lack of e cacy of diclazuril and more particularly in the control of the two pathogenic species E. ovinoidalis and E. crandallis. In addition, in one of the farms, the overall e cacy of toltrazuril was high but seemed to be slightly reduced in regard to the pathogenic species E. crandallis. However, to de nitively attribute these reduced e cacies to resistance of Eimeria species to these drugs, additional experiments are required. In addition, molecular identi cation of oocyst species after treatment would further assist with reaching de nitive conclusions regarding e cacies. Nevertheless, the approach developed in this study could allow rapid and simple testing of the overall e cacy of anticoccidial drugs and of the various sheep Eimeria species, in particular E. ovinoidalis and E. crandallis in eld conditions.

Declarations
In bold, the nal results of the Fecal Oocyst Count Reduction Test (FOCRT), which is conclusive as to whether the anticoccidian treatment tested was effective or not. ISC = Individual Station Control; BC = Berrichon du Cher breed; RO = Rouge de l'Ouest breed  Figure 1 Identi cation keys for oocysts of ovine Eimeria species according to the morphological criteria of Eckert [3].

Figure 2
Size scale of oocysts of sheep Eimeria species. The grey line below indicates the sizes in micrometers; * denotes the presence of a micropyle in the species; ° presence of a polar cap in the species; °° presence of a prominent polar cap in the species; black bar: average size in micrometers for each species.

Supplementary Files
This is a list of supplementary les associated with this preprint. Click to download.