Safety of Elevated Dosages of a 0.24% Diflubenzuron Pellet Administered Orally to Horses*

Transcription

1 D. H. Ross, C. Heird, J. W. Byrd, V. Beauchemin, and W. Kiess Safety of Elevated Dosages of a 0.24% Diflubenzuron Pellet Administered Orally to Horses* Douglas H. Ross, PhD a, Charles Heird, PhD b, John W. Byrd, PhD b Vivienne Beauchemin, MS b Wendy Kiess, BS b a Farnam Companies, Inc. b Southwest Bio-Labs, Inc. 301 West Osborn 401 N. 17 th Street, #11 Phoenix, AZ Las Cruces, NM CLINICAL RELEVANCE The safety of a feed-thru pellet formulation containing the insect growth regulator diflubenzuron (0.24%) for control of manure-breeding flies (Musca domestica L. and Stomoxys calcitrans L.) in horses was evaluated. Pellets were administered orally at 0, 1, 3, and 5 times the clinical dosage (0.12 to 0.20 mg/kg) on a daily basis for 31 consecutive days. Variables examined included daily clinical observations, hematology, coagulation, serum chemistry, acetylcholinesterase inhibition, body weight, and physical examination. Horses remained healthy throughout the study, and no adverse reactions or events related to the pellets were observed. Statistically significant differences (P <.10) between dose groups (0, 1, 3, and 5 ) were observed for only four of the 44 serum chemistry and hematologic variables measured, none of which was dose related. Diflubenzuron can be safely administered orally to horses at 0.12 to 0.20 mg/kg for control of manure-breeding flies. INTRODUCTION Several species of nuisance and biting flies (Diptera: Muscidae), including the horn fly (Haematobia irritans L.), face fly (Musca autumnalis De Geer), house fly (Musca domestica L.), and stable fly (Stomoxys calcitrans L.), are significant pests of livestock. Use of oral larvicides to control these flies, which develop in animal manure, has a long history in the livestock and equine industries. Larvicidal formulations contain an active ingredient that is passed in feces when fed to animals; because larvae cannot develop in manure containing larvicides, fewer adult flies emerge. The most *Funding for this study was provided by Farnam Companies, Inc., Phoenix, AZ. Dr. Ross current address is Bayer HealthCare LLC, Animal Health Division, PO Box 390, Shawnee Mission, KS, Dr. Heird s current address is 5492 Clavel, Las Cruces, NM,

2 Veterinary Therapeutics Vol. 8, No. 1, Spring 2007 widely used active ingredient over the past 30 years has been the organophosphate insecticide tetrachlorvinphos (Rabon, KMG-Bernuth, Houston, TX). In recent years, however, organophosphate insecticides have been the object of increased regulatory scrutiny and review, leading to the development and use of other active ingredients in oral larvicide formulations. The insect growth regulator diflubenzuron (Dimilin, Chemtura Corporation USA, Middlebury, CT) is a chitin synthesis inhibitor that interferes with the egg-hatching and larvalmolting processes in insects. 1 The majority of diflubenzuron uses registered by the US Environmental Protection Agency (EPA) are for fruits, field crops, forestry, and ornamentals. 2 However, two diflubenzuron formulations have been registered for feed-thru fly control. One is a sustained-release bolus 3 (Vigilante, Chemtura) for administration to beef and dairy cattle, and the other is a pellet formulation (SimpliFly/Equitrol II; Farnam) for use in horses. The objective of this study was to evaluate the safety of the feed-thru pellet formulation of diflubenzuron administered orally to horses at 0, 1, 3, and 5 times the clinical dose for 31 consecutive days. The study was conducted to satisfy safety requirements of the EPA for a companion animal insecticide. Increased doses at multiples of the recommended level are commonly used as an accelerated safety test to uncover idiosyncratic reactions that might be extrapolated to the general equine population. MATERIALS AND METHODS Test Animals Twenty-four clinically healthy horses (12 geldings and 12 mares; age range: 2 years to adult) were selected for this study based on acceptable physical examinations, clinical chemistries, and hematologic findings assessed on study day 7. No prophylactic medications were administered during the 28-day acclimation period; all horses were current with vaccinations and deworming. Animals were identified with a unique animal ID number printed on a neck collar. Throughout the study, horses were individually housed in outdoor stalls. Each animal was provided with its own feeder and water container. Animals were fed a ration of alfalfa hay in the morning and afternoon and a ration of sweet feed (grain) only in the morning (when dosing occurred). Animals were allowed ad libitum access to well water, which contained no known contaminants. Animal Observations Physical examinations were conducted on all animals on study days 7, 1, and 30. Animal health observations (i.e., eyes, feces, respiration, mentation, locomotion or musculature, and skin condition) were made once daily, starting at receipt (study day 8) and continuing through the end of the in-life portion on study day 30. Animals were also observed approximately 1 to 2 hours after treatment on study days 0 through 30. All animals were individually weighed on study days 7, 1, and 30. The accuracy of the scale was verified before each weighing. Weights from study day 1 were used to determine the correct treatment dose for each animal. Experimental Design The study consisted of four treatment groups in a completely randomized block design, with six horses (three geldings and three females) per dose group using body weight within sex as the blocking factor: Group 1: 0 (control) Group 2: 1 Group 3: 3 Group 4: 5 62

3 D. H. Ross, C. Heird, J. W. Byrd, V. Beauchemin, and W. Kiess TABLE 1. Experimental Design and Allocation of Animals to Treatment Groups Duration No. of Animals Group Treatment Dose (days) Male Female 1 Control F793 insecticide F793 insecticide F793 insecticide TABLE 2. Dose Determination Table for F793 Insecticide Body Weight Test Material Dose (1 ) lb ( kg) 0.4 oz (11.3 g)/horse/day lb ( kg) 0.6 oz (17.0 g)/horse/day lb ( kg) 0.8 oz (22.7 g)/horse/day 900 1,100 lb ( kg) 1.0 oz (28.4 g)/horse/day 1,100 1,300 lb ( kg) 1.2 oz (34.0 g)/horse/day The individual horse was the experimental unit. Animals were weighed on study day 1 and ranked within each sex by weight in descending order, using a computer-generated randomization schedule. The animals with the four highest weights within each sex were assigned to subgroup 1, the next four to subgroup 2, and so forth to create six subgroups. Each treatment group contained one animal from each of the six subgroups, resulting in a randomly assigned equal representation by sex and weight (Table 1). Test Material The test material administered to horses was an extruded pellet formulation (F793 Insecticide; US EPA Registration Number ; Lot Number 5F001) containing 0.24% diflubenzuron [1-(4-chlorophenyl)-3-(2,6-difluorobenzoyl)urea; CAS # ]. The inactive ingredients in the formulation consisted of common feed materials: wheat middlings, dehydrated alfalfa, calcium carbonate, sodium chloride, and calcium propionate. The test material was packaged in a high-density polyethylene pail with a reclosable lid. It was stored in a cool, dry area protected from light in a controlled-access room. Treatment Animals were dosed for 31 consecutive days starting on study day 0 and ending on study day 30. Test material was administered to the horses in the treated groups (Groups 2, 3, and 4) with their daily morning grain ration. Horses not receiving the test material (Group 1) were untreated and dosed only with the daily grain ration. Body weights collected on study day 1 and the product s EPA label dosing directions were used to determine the 1 dose of pellets for each animal (Table 2). The actual 1 dose of diflubenzuron ranged from 0.12 to 0.20 mg/kg, depending on the horse s body weight. The dose was weighed out before treatment. The grain ration was also weighed before dosing and left in a re-sealable zipper storage bag with the study day and animal number written on the bag. Each horse had its own feed bucket labeled with the corresponding ID number on the outside. During each daily dosing, the grain ration and test material were poured into the buckets and offered to the respective animals. The time of dosing was recorded. Control animals were given only grain ration, which was administered in the same fashion. Personnel involved in the recording of observations and clinical pathology analysis were blinded to the identity of treatments. Anyone 63

4 Veterinary Therapeutics Vol. 8, No. 1, Spring 2007 who directly observed the treatment group assignment or treatment administration was excluded from performing daily observations. Blood Collection Blood samples were collected from the jugular vein on study days 7, 1, 7, 15, 22, and 30. Sample tubes were labeled with the study number, date, study day, animal number, and a unique identification number. Collection activities were documented on the blood collection form. Samples for serum chemistry were collected in evacuated (mottled red top) tubes containing a serum separator. Following collection, samples were allowed to clot and then were centrifuged; serum was harvested, frozen, and stored at temperatures below 15 C until analyzed (storage times ranged from 29 to 65 days). Serum samples were assayed for clinical variables (sodium [meq/l], potassium [meq/l], chloride [meq/l], albumin [g/dl], alkaline phosphatase [U/L], alanine aminotransferase [IU/L], amylase [IU/L], aspartate aminotransferase [U/L], direct bilirubin [mg/dl], total bilirubin [mg/dl], blood urea nitrogen [BUN; mg/dl], calcium [mg/dl], cholesterol [mg/dl], creatine kinase [IU/L], carbon dioxide [or bicarbonate; meq/l], creatinine [mg/dl], glucose [mg/dl], γ-glutamyltransferase [IU/L], phosphorus chloride [meq/l], total protein [g/dl], albumin:globulin ratio [calculated]; globulin [g/dl]; sorbitol dehydrogenase [U/L]; lactate dehydrogenase [IU/L], and acetylcholinesterase) using the Abbott Aeroset Chemistry Analyzer (Abbott Diagnostics, Abbot Park, IL). Hematology samples were collected in purple-top tubes containing EDTA. Blood samples were rocked for at least 15 minutes and refrigerated for 1 day at 5 C until analyzed. Blood samples were assayed for hematologic variables (leukocyte count [ 10 3 /µl], erythrocyte count [ 10 6 /µl], hemoglobin [g/dl], hematocrit [%], packed cell volume [i.e., spun hematocrit; %], mean corpuscular volume [fl], mean corpuscular hemoglobin [pg], mean corpuscular hemoglobin concentration [g/dl], erythrocyte distribution width [%], platelet count, mean platelet volume [fl], neutrophils [absolute count and percentage], lymphocytes [absolute count and percentage], monocytes [absolute count and percentage], eosinophils [absolute count and percentage], basophils [absolute count and percentage], and reticulocyte count) using the Abbott Cell Dyn 3500R Hematology Analyzer (Abbott Diagnostics). Manual differentials were not performed. Samples for coagulation were collected in blue-top tubes containing sodium citrate. Collected samples were centrifuged, and plasma was harvested, frozen, and stored at temperatures below 15 C until analysis (storage times ranged from 7 to 66 days). Samples were assayed for coagulation variables (prothrombin time [seconds], activated partial thromboplastin time [seconds], and fibrinogen [mg/dl]) using the Coulter ACL100 Clinical Coagulation Analyzer (Beckman Coulter, Fullerton, CA). Statistical Analyses Data obtained on each quantitative blood variable (serum chemistry, hematology, and coagulation) were statistically evaluated via repeated-measures analyses of covariance, with treatment (i.e., dose group), sex, and time as the main effects, along with all first-order interactions of these main effects; the covariate was the pretreatment (study day 1) value of that blood variable. Least-squares treatment means were computed for each dose group, for each sex, and at each sampling time during the treatment period, as well as for binary combinations of these three fixed effects. The structure of the covariance matrix for the repeated measures was modeled using the first-order auto- 64

5 D. H. Ross, C. Heird, J. W. Byrd, V. Beauchemin, and W. Kiess TABLE 3. Body Weights (lb) of Horses and Weight Gain/Loss during the Study Weight Animal ID Sex Day 7 Day 1 Day 30 Gain/Loss a Group F F F 1,055 1,035 1, M M 1,115 1,110 1, M Average Group F 1,065 1,050 1, F F M M 1,030 1,015 1, M Average Group F 1,020 1, F F M 1,110 1,110 1, M M Average Group F F F 1,045 1,015 1, M M M Average a There was no significant difference between groups in weight gain/loss (P >.05). regressive (AR1) option in SAS s MIXED procedure (SAS Version 9.1, SAS Institute, Cary, NC). In the event that the statistical test for either overall differences among the four dose groups or the time dose interaction during the treatment period was statistically significant (P <.10), the variation in the dose group was further evaluated via application of Fisher s least significant difference (LSD) method to compare the mean of each dose group (i.e., 1, 3, and 5 ) with the mean of the control group (0 ) at the 10% level of significance. In the event of a statistically significant (P <.10) test for the group day interaction, the mean of each of the F793 insecticide groups at each time point following treatment initiation was to be compared with the corresponding control group mean via the LSD procedure at the 10% level of significance. Similarly, in the event of a statistically significant (P <.05) test for the group sex interaction, the mean of each dose group was compared with the control group mean both overall and for each sex via the LSD procedure at the 10% level of significance. SAS s MIXED procedure was used to carry out each of the repeated measures analyses of covariance described above. RESULTS Treatments and Animals Overall, animals were dosed without any problems. During dosing on study day 0, 12 animals were noted as having spilled some of the grain or dose pellets on the ground. An attempt was made to recover the spilled material and return it to each animal s feed bucket. After day 0, no further dosing problems were encountered. All horses readily (text continues on page 68) 65

6 Veterinary Therapeutics Vol. 8, No. 1, Spring 2007 TABLE 4. Summary of Hematology and Coagulation Values a of Horses Treated with Diflubenzuron Pellets (0.24%) at 0, 1, 3, and 5 the EPA Label Dosage Variable F M F M F M F M Leukocytes ( 10 3 /µl) ( ) ( ) ( ) ( ) ( ) ( ) ( ) ( ) Neutrophil count ( 10 3 /µl) ( ) ( ) ( ) ( ) ( ) ( ) ( ) ( ) Neutrophil (%) ( ) ( ) ( ) ( ) ( ) ( ) ( ) ( ) Lymphocyte count ( 10 3 /µl) ( ) ( ) ( ) ( ) ( ) ( ) ( ) ( ) Lymphocyte (%) ( ) ( ) ( ) ( ) ( ) ( ) ( ) ( ) Monocyte count ( 10 3 /µl) ( ) ( ) ( ) ( ) ( ) ( ) ( ) ( ) Monocyte (%) ( ) ( ) ( ) ( ) ( ) ( ) ( ) ( ) Eosinophil count ( 10 3 /µl) ( ) ( ) ( ) ( ) ( ) ( ) ( ) ( ) Eosinophil (%) ( ) ( ) ( ) ( ) ( ) ( ) ( ) ( ) Basophil count ( 10 3 /µl) ( ) ( ) ( ) ( ) ( ) ( ) ( ) ( ) 66

7 D. H. Ross, C. Heird, J. W. Byrd, V. Beauchemin, and W. Kiess TABLE 4. Summary of Hematology and Coagulation Values a of Horses Treated with Diflubenzuron Pellets (0.24%) at 0, 1, 3, and 5 the EPA Label Dosage (cont.) Variable F M F M F M F M Basophil (%) ( ) ( ) ( ) ( ) ( ) ( ) ( ) ( ) Erythrocytes ( 10 6 /µl) ( ) ( ) ( ) ( ) ( ) ( ) ( ) ( ) Hemoglobin (g/dl) ( ) ( ) ( ) ( ) ( ) ( ) ( ) ( ) Hematocrit (%) ( ) ( ) ( ) ( ) ( ) ( ) ( ) ( ) Mean corpuscular volume (fl) ( ) ( ) ( ) ( ) ( ) ( ) ( ) ( ) Mean corpuscular hemoglobin (pg) ( ) ( ) ( ) ( ) ( ) ( ) ( ) ( ) Mean corpuscular hemoglobin ( ) ( ) ( ) ( ) ( ) ( ) ( ) ( ) concentration (g/dl) Platelets ( 10 3 /µl) ( ) (38 220) ( ) (87 334) (33 244) (48 379) (58 257) ( ) Activated partial thromboplastin (64 204) (45 199) (84 174) (91 206) (65 195) (57 213) (86 153) (75 238) time (sec) a Values reported as mean (range). 67

8 Veterinary Therapeutics Vol. 8, No. 1, Spring 2007 TABLE 5. Probability Levels for the Statistical Tests of the Overall Variation among the Four Dose Groups (Group), between Sexes (Sex), and among Measurement Times (Day) and for All First-Order Interactions of These Main Effects Based on the Analysis of Covariance of the Quantitative Variables for Hematology and Coagulation during the Treatment Period a Group Group Sex Variable Covariate b Group Sex Sex Day Day Day Leukocyte count Neutrophil count Neutrophil percent < Lymphocyte count < Lymphocyte percent < Monocyte count Monocyte percent Eosinophil count < < Eosinophil percent < Basophil count < Basophil percent < Erythrocyte count < Hemoglobin < Hematocrit < Mean corpuscular < < volume Mean corpuscular < < hemoglobin Mean corpuscular < hemoglobin concentration Platelet count < < Activated partial < thromboplastin time a Values in boldface indicate statistically significant differences for group (P <.10). b Covariate was the pretreatment value of the same variable at day 1. consumed the test material. Body weights were taken on study days 7, 1, and 30 (Table 3). Throughout the study, body weights ranged from 766 lb (347 kg) to 1,130 lb (513 kg). Eight of the animals showed weight loss during the treatment period, whereas the remaining animals showed a weight gain ranging from 2 to 50 lb (1 to 23 68

9 D. H. Ross, C. Heird, J. W. Byrd, V. Beauchemin, and W. Kiess TABLE 6. Least-Squares Means by F793 Insecticide Dose Group for Each Hematology and Serum Chemistry Variable Found to Have Significant Tests for Group (P <.10) or Group Sex (P <.05) during the Treatment Period a Variable Hematology Lymphocyte count ( 10 3 /µl) Basophil count ( 10 3 /µl) Basophil percent Serum chemistry Glucose (mg/dl) BUN (mg/dl) Alkaline phosphatase (U/L) Amylase (IU/L) a Values in boldface were significantly different (P <.10) than the control (0 ) group mean based on application of Fisher s least significant difference method. kg). One animal showed no weight gain or loss. None of these changes in weight was statistically significant (P >.05). There were no treatment-related changes in weight. Clinical Observations There were no abnormalities recorded during the acclimation period for Groups 2, 3, and 4 (1, 3, and 5, respectively). One horse from Group 1 (0 ) had ocular discharge for 5 days during acclimation. During the treatment period (study days 0 to 30), three horses in Group 1 (0 ) had ocular discharges, with a total of six occurrences. In Group 2 (1 ) animals, there was a total of 25 occurrences of ocular discharge from five horses. Four horses in Group 3 (3 ) had a total of 19 occurrences of ocular discharge. In Group 4 (5 ), three horses showed a total of only seven occurrences of the ocular discharge during the treatment period. The skin condition of all animals during acclimation and treatment was normal, except for one occurrence with one horse in the 1 group. This animal had a small abrasion on the right front ankle on study day 26. Mentation remained normal in all animals throughout the acclimation and treatment periods of the study. There were no abnormalities with the feces throughout the acclimation and treatment periods. No adverse reactions were observed on any treatment day in any of the treatment groups, and no mortalities occurred during the study. Physical Examinations Physical examinations did not detect any treatment-related findings. Overall, animals were healthy and normal throughout the study, as evidenced by the physical examinations. These examinations were consistent with the daily observations. All animals used in the study were graded as either excellent or good in their overall physical condition, with the exception of animals 552, 555, and 577, which were graded as fair. Animal 552 was noted as having yellow-colored discharge from the external genitalia and cracks in both front hoofs. Animal 555 was considered in thin body condition and had slight serous nasal discharge. The left back hoof of animal 577 was noted as being centrally cracked, bottom to top. Horses marked fair on study day 7 were deemed acceptable for study. Environmental Variables The average minimum and maximum ambient temperatures during study days 0 to 30 (text continues on page 72) 69

10 Veterinary Therapeutics Vol. 8, No. 1, Spring 2007 TABLE 7. Summary of Clinical Chemistry Values a of Horses Treated with Diflubenzuron Pellets (0.24%) at 0, 1, 3, and 5 the EPA Label Dosage Variable F M F M F M F M Sodium (meq/l) ( ) ( ) ( ) ( ) (72 140) ( ) ( ) ( ) Potassium (meq/l) ( ) ( ) ( ) ( ) ( ) ( ) ( ) ( ) Chloride (meq/l) (91 103) (93 100) (94 102) (90 104) (55 103) (87 101) (89 105) (88 104) Glucose (mg/dl) (73 104) (83 132) (75 130) (74 116) (63 110) (59 122) (71 109) (76 105) BUN (mg/dl) ( ) ( ) ( ) ( ) ( ) ( ) ( ) ( ) Creatinine (mg/dl) ( ) ( ) ( ) ( ) ( ) ( ) ( ) ( ) Direct bilirubin (mg/dl) ( ) ( ) ( ) ( ) ( ) ( ) ( ) ( ) Total bilirubin (mg/dl) ( ) ( ) ( ) ( ) ( ) ( ) ( ) ( ) Phosphorus (mg/dl) ( ) ( ) ( ) ( ) ( ) ( ) ( ) ( ) Calcium (mg/dl) ( ) ( ) ( ) ( ) ( ) ( ) ( ) ( ) 70

11 D. H. Ross, C. Heird, J. W. Byrd, V. Beauchemin, and W. Kiess TABLE 7. Summary of Clinical Chemistry Values a of Horses Treated with Diflubenzuron Pellets (0.24%) at 0, 1, 3, and 5 the EPA Label Dosage (cont.) Variable F M F M F M F M Alkaline phosphatase (IU/L) ( ) (94 198) ( ) ( ) (82 150) ( ) (93 166) ( ) Albumin (g/dl) ( ) ( ) ( ) ( ) ( ) ( ) ( ) ( ) Total protein (g/dl) ( ) ( ) ( ) ( ) ( ) ( ) ( ) ( ) Alanine aminotransferase ( ) ( ) ( ) ( ) ( ) ( ) ( ) ( ) (U/L) Aspartate aminotransferase ( ) ( ) ( ) ( ) ( ) ( ) ( ) ( ) (U/L) Amylase (IU/L) ( ) ( ) ( ) ( ) ( ) ( ) ( ) ( ) Creatine kinase (IU/L) ( ) (110 1,049) ( ) ( ) ( ) ( ) ( ) ( ) Lactate dehydrogenase ( ) ( ) ( ) ( ) (92 349) ( ) ( ) ( ) (IU/L) Globulin (g/dl) ( ) ( ) ( ) ( ) ( ) ( ) ( ) ( ) Cholinesterase 3, , , , , , , ,386.3 (IU/L) (2,232 5,046) (3,583 4,491) (1,132 4,376) ( ,848) (2,617 3,662) (1,687 5,038) (981 4,293) (1,625 4,568) a Values reported as mean (range). 71

12 Veterinary Therapeutics Vol. 8, No. 1, Spring 2007 TABLE 8. Probability Levels for the Statistical Tests of the Overall Variation among the Four Dose Groups (Group), between Sexes (Sex), and among Measurement Times (Day) and for All First-Order Interactions of These Main Effects Based on the Analysis of Covariance of the Quantitative Variables for Serum Chemistry during the Treatment Period a Group Group Sex Variable Covariate b Group Sex Sex Day Day Day Sodium < Potassium < Chloride < Glucose BUN < Creatinine < < Direct bilirubin Total bilirubin < < Phosphorus chloride < < Calcium Alkaline phosphatase < < Albumin < Total protein < < Alanine aminotransferase < Aspartate aminotransferase < < Amylase Creatine kinase Lactate dehydrogenase < Globulin < < Acetylcholinesterase a Values in boldface indicated statistically significant differences for group (P <.10) and group sex (P <.05). b Covariate was the pretreatment value of the same variable at day 1. were 42 F (±8 F) and 93 F (±9 F), respectively. The average minimum and maximum relative humidity levels were 18% (±4%) and 74% (±23%), respectively. Hematology and Coagulation The dose group means (ranges) for each of the hematology variables and the coagulation variable activated partial thromboplastin time during the treatment period are presented in Table 4. For the 31-day treatment period, the four dose groups were found to vary significantly (P >.10) in the overall mean values of three hematology variables: lymphocyte count, basophil count, and percent basophils (P =.100, P =.020, and P =.064, respectively; Table 5). The variation among the four dose groups was not significantly different among the four measure- 72

13 D. H. Ross, C. Heird, J. W. Byrd, V. Beauchemin, and W. Kiess TABLE 9. Least-Squares Means by F793 Insecticide Dose Group and Sex for Each Serum Chemistry Variable for which the Group Sex Interaction Was Statistically Significant (P <.05) a Variable (unit) Sex Glucose (mg/dl) M F BUN (mg/dl) M F Amylase (IU/L) M F a Values in boldface are significantly different (P <.10) than the control (0 ) group mean based on application of Fisher s least significant difference method. ment days (P >.10; Table 5) or between the two sexes in any of the 18 hematology variables and activated partial thromboplastin time (P >.05; Table 5). The least-squares means for each of the three hematology variables with a statistically significant test for Group are presented by dose group in Table 6. As described previously, statistically significant variation among the four dose groups was further evaluated by application of Fisher s LSD method for pair-wise comparisons of the means for each of the 1, 3, and 5 groups with the control (0 ) group. As shown in Table 6, the 1 group was significantly different (P <.10) from the 0 group in the overall mean for each of the three hematology variables (lymphocyte count, basophil count, and percent basophils), whereas the 3 group differed significantly (P <.10) from the 0 group in the overall mean only for basophils (both count and percent). However, the 5 group did not differ significantly from the 0 group in any of these hematology variables (Table 6). Serum Chemistry and Acetylcholinesterase The dose group means (ranges) for each of the 23 serum chemistry variables and serum acetylcholinesterase are presented in Table 7. During the 31-day treatment period, only one of the serum chemistry variables alkaline phosphatase was found to vary significantly (P =.013; Table 8) overall among the four dose groups. The group day interaction was not statistically significant for any of the serum chemistry variables or for serum acetylcholinesterase (P >.25); however, the group sex interaction was found to be statistically significant (P <.05) for three of the serum chemistry variables: glucose, BUN, and amylase (Table 8). The relevant least-squares means for each of the four serum chemistry variables with a statistically significant test for group or group sex are presented by treatment group in Tables 6 and 9 (overall and for each sex, respectively). For alkaline phosphatase (Table 6), the 1 and 5 groups were significantly different (P <.10) from the 0 control group in the overall mean in serum, whereas mean alkaline phosphatase did not differ significantly between the 3 and 0 groups. The 5 group was also significantly different from the 0 group in the overall mean glucose in serum (P <.10; Table 6). No other overall mean difference between the positive dose groups and the control group was statistically significant for serum glucose, BUN, alkaline phosphatase, and amylase (Table 6). 73

14 Veterinary Therapeutics Vol. 8, No. 1, Spring 2007 TABLE 10. Equine Hematology and Serum Chemistry Reference Ranges Southwest BioLabs a Published Sources b Variable No. of Samples Range Kaneko et al 4 Coles 5 Plumb 6 Aiello 7 Hematology Erythrocytes ( 10 6 /µl) Hemoglobin (g/dl) Hematocrit (%) Mean corpuscular volume (fl) Mean corpuscular hemoglobin (pg) Mean corpuscular hemoglobin concentration (g/dl) Platelets ( 10 3 /µl) Leukocytes ( 10 3 /µl) Neutrophil (%) Neutrophil count ( 10 3 /µl) 3 6 Lymphocyte (%) Lymphocyte count ( 10 3 /µl) Monocyte (%) Monocyte count ( 10 3 /µl) Eosinophil (%) Eosinophil count ( 10 3 /µl) Basophil (%) Basophil count ( 10 3 /µl) Serum chemistry Sodium (mmol/l) Potassium (mmol/l) Chloride (mmol/l) Glucose (mg/dl) BUN (mg/dl) Creatinine (mg/dl) Alkaline phosphatase (IU/L) Total bilirubin (mg/dl) Phosphorus (mg/dl) Calcium (mg/dl) Albumin (g/dl) Total protein (g/dl) Alanine aminotransferase (U/L) 74

15 D. H. Ross, C. Heird, J. W. Byrd, V. Beauchemin, and W. Kiess TABLE 10. Equine Hematology and Serum Chemistry Reference Ranges (cont.) Southwest BioLabs a Published Sources b Variable No. of Samples Range Kaneko et al 4 Coles 5 Plumb 6 Aiello 7 Serum chemistry (cont.) Aspartate aminotransferase (U/L) Direct bilirubin (mg/dl) γ-glutamyl transferase (IU/L) Globulin (g/dl) Albumin:Globulin ratio (%) Sorbitol dehydrogenase (IU/L) Coagulation Prothrombin time (sec) Fibrinogen (mg/dl) Activated partial thromboplastin time (sec) a Blood samples were drawn from adult (>2 yr) horses that had been examined by a veterinarian and determined to be clinically normal as evidenced by a physical examination record obtained within 2 days of the blood draw. Serum chemistry, hematology, and coagulation tests were performed on each sample, and values for each variable were ranked from lowest to highest. The upper 5% and lower 5% of the values were omitted and the means and SDs were calculated from the remaining values. The normal range was determined as the mean ± two SDs. In cases in which the lower limit was calculated to be a negative number, the low value was censored to 0. b See References for full citations. Mean serum glucose (Table 9) for male horses in each of the 1, 3, and 5 groups was significantly greater (P <.10) than that for male horses in the 0 group; however, mean glucose for female horses in each of the 1, 3, and 5 groups was not significantly different than that for the 0 group. For male horses, mean BUN was significantly greater (P <.10) for the 1 group than for the 0 group but did not differ between the 0 group and either the 3 or the 5 group; for female horses, mean BUN for the 0 group did not differ significantly from any of the positive dose groups (Table 9). Mean serum amylase for the 1, 3, and 5 groups was greater than the 0 group mean for male horses and less then the 0 group mean for female horses (Table 9). Whereas the 1 and 0 groups were significantly different in mean serum amylase for both male and female horses (P <.10), the mean difference in serum amylase was statistically significant for male horses between the 3 and 0 groups but not between the 5 and 0 groups and for female horses between the 5 and 0 groups but not between the 3 and 0 groups (Table 9). DISCUSSION The animals used in this study remained healthy, as evidenced by daily observations and physical examinations. All animals appeared to 75

16 Veterinary Therapeutics Vol. 8, No. 1, Spring 2007 tolerate the test material well. After treatment, however, horses in the 1, 3, and 5 groups experienced increased ocular discharge. The 1 group had the most incidences of ocular discharge; however, animal 506 acted as an outlier because of a chronic eye condition that resulted in constant ocular discharge. A dose-dependent reaction did not seem to exist as there was no trend of increased ocular discharge with treatment level. The high frequency of ocular discharge among horses may be related to other factors (e.g., increased fly population in the vicinity, increased dust due to confinement, wind). No adverse reactions were recorded. During the treatment period of this study, only four of the 44 blood variables measured showed a statistically significant difference between groups (P <.10; Tables 5 and 8), two of which were closely interrelated (basophil count and percent basophils). However, further statistical examination of these differences (Table 6) indicated that they did not vary with dose and were therefore not treatment related. For example, the 1 and 5 groups were significantly different (P <.10) from the 0 group in overall mean alkaline phosphatase, whereas this variable did not differ significantly between the 3 and 0 groups (Table 6). For the three serum chemistry variables for which the group sex interaction was statistically significant (i.e., glucose, BUN, and amylase; Table 8), the mean differences between the 0 group and each of the 1, 3, and 5 groups exhibited little, if any, pattern with respect to test material dose for either male or female horses. In fact, for both glucose and amylase, the significance of the group sex interaction appeared to be attributable primarily to a significant (P <.003) mean difference between male and female horses in the control (0 ) group (Table 9). None of the 44 statistical tests for the group day interaction was statistically significant (P >.10; Tables 5 and 8). The means of all the variables for which the group or group sex variation was significant (Tables 6 and 9) fell well within normal equine reference ranges (Table 10). Therefore, these differences were not considered to be biologically significant. CONCLUSION This study showed that diflubenzuron in a pelleted feed supplement can be safely administered orally to horses at 0.12 to 0.20 mg/kg for control of manure breeding flies. ACKNOWLEDGMENTS The authors thank Erin Weich, DVM (Attending Veterinarian at Southwest Bio-Labs), and Myke Browning (Southwest Bio-Labs), who performed or monitored hematology, serum chemistry, and/or coagulation assays, and Dr. Tom Keefe (EnviroStat Associates, Windsor, CO) for the statistical analyses. Drs. Daniel Ciszewski and Jennifer Schofield (Bayer Animal Health) provided helpful comments on an earlier draft of this manuscript. REFERENCES 1. Copping LG, Hewitt HJ: Chemistry and Mode of Action of Crop Protection Agents. Cambridge, UK, The Royal Society of Chemistry, United States Environmental Protection Agency: R.E.D. Facts: Diflubenzuron. Washington, DC, Miller JA, Knapp FW, Miller RW, et al: Diflubenzuron bolus for control of fly larvae. J Agri Entomol 3:48 55, Kaneko JJ, Harvey JW, Bruss ML (eds): Clinical Biochemistry of Domestic Animals, ed 5. San Diego, Academic Press, Coles FH: Veterinary Clinical Pathology. Philadelphia, WB Saunders, Plumb DC: Veterinary Drug Handbook, ed 5. Ames, IA, Blackwell Publishing, Aiello SE (ed): The Merck Veterinary Manual, ed 8. Whitehouse Station, NJ, Merck & Company,