Hematology of the keeled box turtle (Cuora mouhotii)

A healthy adult, intact female keeled box turtle (Cuora mouhotii) was found to have a marked heterophilic leukocytosis using normal hematologic parameters established for the eastern box turtle (Terrapene carolina carolina), a related chelonian species. This animal was monitored with serial complete blood counts (CBCs) over the next 15 years despite remaining asymptomatic for an infectious condition. Retrospective CBC data were compiled from 38 presumably healthy keeled box turtles to establish hematologic values for comparison in this species. Using this species-specific data, over the 15-year period, the female keeled box turtle had two times where the white blood cell (WBC) count was greater than 2 standard deviations (SD) above the mean, six times where the WBC count was greater than 1 SD above the mean, six times where the PCV was greater than 2 SD above the mean, and eight times where the PCV was greater than 1 SD above the mean. Infection and inflammation are the most common causes of leukocytosis in reptiles; however, given the clinical presentation of this patient, it was postulated that these clinicopathologic changes could be secondary to a stress response. Establishing reference intervals and understanding how stress impacts CBC parameters are important for evaluating the health status of keeled box turtles kept in captivity and for assessing the effects of environmental changes on the health status of wild populations of this endangered chelonian species.     

Haematological values of post-laying Arrau turtle (Podocnemis expansa) in the Orinoco River, Venezuela

The Arrau turtle (Podocnemis expansa) is an endangered species, as a result of long-lasting, unsustainable exploitation. To obtain reference haematological values from the wild Podocnemis expansa during post-laying, 20 turtles were captured in the Orinoco River. Blood was obtained from the dorsal cervical sinus in lithium heparin tubes. Red blood cells (RBC), white blood cells (WBC), thrombocytes (TC), packed cell volume (PCV), plasmatic protein (PP), haemoglobin (Hgb), mean corpuscular volume (MCV) and differential leukocyte count were determined. Haematological values were: RBC 0.9×10(9)/L, WBC 5.7×10(9)/L, TC 5.4×10(9)/L, PCV 35.6%, PP 4.2g/dL, Hgb 11.8g/dL, MCV 411fL. The differential leukocyte count comprised: 71% heterophils, 23% lymphocytes, 3% eosinophils, 1.6% basophils, and 1% monocytes. The reports of reference haematology values for the wild P. expansa are limited; therefore, the results presented herein contrast with those values obtained in captivity. This study represents a contribution to the referential haematological values of the wild P. expansa.     

Changes in haematological profile of common pheasant (Phasianus colchicus) induced by transit to pheasantry

The aim of this study was to assess haematological changes in hand-reared pheasants (Phasianus colchicus) transported from intensive housing facilities to a pheasantry. Selected haematological parameters were monitored in a group of 100 pheasants (50 males and 50 females) aged of 9 weeks that were transported for 4 hours by a covered lorry in crates, with a total body weight of 12 +/- 0.5 kg per crate (Group C12 - floor space: 290 cm2/kg) and with a total body weight of 18 +/- 0.5 kg per crate (Group C18 - floor space: 195 cm2/kg). Blood samples were taken from 10 randomly selected males and 10 females before transport (CON group) and 20 hours after transport (C12 and C18 groups). Examinations consisted in determining the total erythrocyte and leukocyte counts, haematocrit values, haemoglobin levels and differential leukocyte counts, whereby the proportions of heterophil, basophil and eosinophil granulocytes, lymphocytes and monocytes of the total leukocytes were computed. The changes in the parameters of red blood cell count were manifested by an increase (P < 0.01) in the haemoglobin level, MCH (mean cell haemoglobin) and MCHC (mean cell haemoglobin concentration) values and a decrease (P < 0.01) in the total erythrocyte count and haematocrit level in both C12 and C18 pheasants, when compared with the control group of non-transported pheasants. C18 pheasants exhibited also a significant increase (P < 0.05) in MCV (mean cell value) value. When analyzing differential leukocyte counts, C18 pheasants showed a decrease (P < 0.01) in heterophil counts and H/L ratio, whereas values in C12 pheasants did not differ from the non-transported control group. Individual counts of lymphocytes were decreased (P < 0.05) in C12 pheasants, whereas basophil counts were increased (P < 0.01) in both C12 and C18 pheasants. Total leukocyte count was decreased (P < 0.01) in C12 and C18 pheasants. In conclusion, the specific requirements of pheasants, as primarily wild animals, for the density in crates should be respected during transportation and they should be transported at lower densities than other poultry species, at least 290 cm2/kg live weight should be provided.     

A case-controlled retrospective study of the causes and implications of moderate to severe leukocytosis in dogs in South Africa

BACKGROUND: Previous studies showed that dogs with extreme leukocytosis had specific types of diseases, long hospitalization times, and high mortality rates. Objectives: The aim of this study was to determine whether dogs with moderate to severe leukocytosis are likely to have similar results compared with age-matched control dogs. METHODS: Records at the Onderstepoort Veterinary Academic Hospital, University of Pretoria, were examined retrospectively from dogs with > or =35 x 10(9) WBC/L (Leukocytosis Group) and dogs with < or =30 x 10(9) WBC/L and < or =0.5 x 10(9) band neutrophils/L (Control Group). Hematologic and serum protein data, final diagnosis, and effect of glucocorticoid treatment were compared between groups. RESULTS: One hundred eighty-two dogs were included in the Leukocytosis Group and 179 in the Control Group. Compared with dogs in the Control Group, significantly more dogs in the Leukocytosis Group had infections, babesiosis, immune-mediated hematologic disease, and necrosis. Hospitalization time and neutrophil, lymphocyte, and monocyte counts were significantly higher and HCT, eosinophil count, platelet count, and serum albumin concentration were lower in dogs in the Leukocytosis Group (P<.0001). There was no difference in leukocyte counts between glucocorticoid-treated and non-glucocorticoid-treated dogs. Survival did not differ between Leukocytosis and Control Groups; however, a significant relationship was found between total neutrophil (mature+band) count and survival (P=.01). CONCLUSIONS: Dogs with leukocytosis of > or =35 x 10(9)/L are more likely to have bacterial and fungal infections, complicated babesiosis, immune-mediated hematologic disease, and necrosis. The total neutrophil count has a significant impact on outcome.     

Comparative study of hematologic and plasma biochemical variables in Eastern Atlantic juvenile and adult nesting loggerhead sea turtles (Caretta caretta)

Background: Plasma biochemical and hematologic variables are important in the management of endangered sea turtles, such as loggerheads. However, studies on blood biochemistry and hematology of loggerheads are limited, and different concentrations according to variable criteria have been reported. Objective: The purpose of this study was to establish and compare baseline plasma chemistry and hematology values in Eastern Atlantic juvenile and adult nesting loggerhead sea turtles (Caretta caretta). Methods: Blood samples were collected from 69 healthy juvenile loggerhead sea turtles after their rehabilitation in captivity, and from 34 adult nesting loggerheads after oviposition. Fresh blood was used for leukocyte differential count and PCV determination. Heparinized blood was used for RBC and WBC counts. Plasma biochemical concentrations were measured using an automated biochemical analyzer. For the comparative study, nonparametric statistical analysis was done using the Mann–Whitney U‐test. Results: Minimum, maximum, and median concentrations were obtained for 14 hematologic and 15 plasma chemistry variables. Statistically significant differences between juvenile and adult turtles were found for PCV; RBC, WBC, and leukocyte differential counts; total protein, albumin, globulins, calcium, triglycerides, glucose, total cholesterol and urea concentrations; and lactate dehydrogenase activity. Conclusions: Age, size, and reproductive status cause important variations in the hematologic and plasma biochemical results of loggerheads. The reference values obtained in this study may be used as a standard profile, useful for veterinary surgeons involved in sea turtle conservation.     

Biochemical and hematological reference ranges for Amazon freshwater turtle, Podocnemis expansa (Reptilia: Pelomedusidae), with morphologic assessment of blood cells

Blood parameters are useful to measure physiological disturbances in chelonians. Thus they can provide important information for the diagnosis and prognosis of diseases. The aim of this paper was to determine the reference range of plasma glucose, total protein, triglycerides, cholesterol and urea levels, as well as hematocrit, hemoglobin, red blood cell, thrombocyte and white blood cell counts for healthy Podocnemis expansa (Schweigger, 1812), bred in a turtle farm in the Amazonas State, Brazil. Plasma glucose, hemoglobin, mean corpuscular volume (MCV), lymphocytes and heterophils were the parameters with the smallest variations. Significant (p < 0.001) correlations between red blood cells count and hematocrit and hemoglobin concentration were found. Turtle P. expansa had lymphocytes, azurophils, heterophils, eosinophils and basophils with morphologic features similar to the ones of others turtles species. No investigated blood parameter was influenced by sex. In further studies, the established reference ranges might be useful for the health assessment of this turtle species.     

Hematologic values in normal pygmy goats

Pygmy goat blood was analyzed to determine hematologic base-line values. Male, female, and castrated male goats were used in this investigation. Influence of age on these variables was also studied. Most of the hematologic values of pygmy goats were similar to the reference values reported for several other ruminant species and for man. Values significantly different from reference values in man were as follows: Erythrocyte counts were slightly higher than for human beings. The relative lymphocyte count was comparable to that in infants, but significantly greater than values in adult human beings. Also, although not of statistical significance, hemoglobin levels in goats were slightly lower than in both infants and adult human beings, and the leukocyte count was just slightly higher than the usual values for infants and adult human beings. Thus, of 9 hematologic variables, only 1 was significantly different from those in man, and 2 were slightly different. The fact that pygmy goats have hematologic responses similar to those of man, and the similarity of hematologic values to reference values in man, further enhances the pygmy goat for use as an experimental animal.     

Thrombocytopaenia in canine babesiosis and its clinical usefulness

Canine babesiosis is a common cause of thrombocytopaenia but there are few formal studies that have investigated this haematological finding in dogs. Thrombocyte counts from full blood counts were retrospectively analysed for the years 1996-2002. Thrombocyte counts and mean platelet volumes of dogs with babesiosis were compared with those of dogs, seen over the same period of time, that did not have babesiosis. There were 1162 cases in the Babesiosis group and 10 808 in the Non-babesiosis group. A frequency distribution of the thrombocyte counts showed a trimodal distribution in the Non-babesiosis group compared to a bimodal distribution in the Babesiosis group, with a strong positive skewness. The modes for the frequency distributions were 10, 40, 300 and 10, 35 x 10(9)/l thrombocytes, respectively. The median thrombocyte count in the Babesiosis group was 14 x 10(9)/l and 282 x 10(9)/l in the Non-babesiosis group. There was a statistically significant difference in the median thrombocyte count between the Babesiosis group and the Non-babesiosis group. In the Babesiosis group, 99% of the thrombocyte counts were below the lower reference range value (250 x 10(9)/l) and 62% of thrombocyte counts were below 25 x 10(9)/l. The mean platelet volume (11.1 fl) for the Babesiosis group was greater than the reference range (6-10 fl) and significantly larger than in the Non-babesiosis group (median 9.7 fl). Thrombocyte counts greater than 110 and 250 x 10(9)/l had a predictive value that the dog was not suffering from babesiosis of 99.3% and 99.8%, respectively. There was a statistically significant difference between the thrombocyte counts of dogs with babesiosis when grouped by parasitaemia scores. The mechanisms of the thrombocytopaenia are not fully understood, and multiple mechanisms, including concomitant thrombocytopaenia-inducing diseases such as ehrlichiosis, probably result in this haematological finding. Babesiosis in the South African canine population is associated with thrombocytopaenia in nearly all patients and is severe in the majority of them. In the absence of thrombocytopaenia, babesiosis is an unlikely diagnosis.     

Evaluation of equine hemograms using the ADVIA 120 as compared with an impedance counter and manual differential count

BACKGROUND: The ADVIA 120 is an automated laser cell counter widely used in veterinary medicine. Although specific software for equine samples is available and validated, only a few reports have been published comparing the ADVIA 120 with other methods for equine hemogram evaluation. OBJECTIVES: The purpose of this study was to compare the hematologic values and reference intervals obtained on the ADVIA 120 with those obtained on an impedance cell counter and manual differential counts in healthy horses. METHODS: EDTA-anticoagulated blood samples were obtained from 114 clinically healthy horses of various breeds, both sexes, and 2-6 years of age. Samples were stored for up to 12 hours at 4 degrees C and then analyzed on the ADVIA 120 and the Hemat 8. A 100-cell to 200-cell differential leukocyte count was performed by 3 independent observers on May-Grünwald-Giemsa-stained smears. Intra-assay precision of the ADVIA 120 was determined by analyzing 5 replicates each of 10 of the blood samples. RESULTS: Results from the ADVIA were significantly higher than those from the impedance counter for RBC count, total WBC count, hemoglobin concentration, red cell distribution width, MCH, and MCHC, and significantly lower for HCT and platelet count. Significantly higher neutrophil and basophil counts and significantly lower lymphocyte counts were obtained with the ADVIA 120 compared with manual counts. Based on Passing-Bablok regression analysis, RBC and platelet counts were in good agreement between the 2 analyzers; a constant and proportional bias was present for other values. Coefficients of variation for erythrocyte parameters on the ADVIA were <1%, but were higher for platelet (6%), total WBC (2%), differential WBC (4%-30%), and reticulocyte (75%) counts. CONCLUSIONS: Results obtained with equine samples on the ADVIA 120 were comparable with those obtained on an impedance counter; reference intervals differed statistically but overlapped. The ADVIA had poor precision for reticulocyte and differential leukocyte counts such that the latter should always be verified on smears.     

Comparison between microscopic and automated differential leukocyte counts in the silver fox (Vulpes vulpes) and the blue fox (Alopex lagopus)

Differential leukocyte (WBC) counts in blood from clinically healthy silver foxes (n=32) and blue foxes (n=37) obtained from an automated hematology analyzer (Technicon H*1 Hematology System) with canine software were compared with microscopic differential WBC counts (M-diff). There was good agreement between the automated differential cell count (A-diff) and the M-diff for neutrophil and lymphocyte percentages. The correlation was lower for monocyte percentages and variable for eosinophil percentages. There was no significant difference between the A-diff and M-diff in either fox species. The A-diff counts were very precise, and may be a good alternative to the traditional M-diff for screening populations of clinically healthy foxes or for studies on stress and animal welfare. Intercept values, however, indicated a constant bias that must be taken into account before interpreting results based on different methods of analysis     

Hematology and serum chemistry values of the lesser rhea (Pterocnemia pennata) raised in Patagonian farms (Argentina).

Hematologic and serum chemistry values were determined for chicks, juveniles, and adult farmed lesser rheas (Prerocnemia pennata). Blood samples were taken during March-April 1998 from the brachial veins of lesser rhea, including 64 chicks, 24 juveniles, and 36 adults, raised in Patagonian farms (Argentina). Lesser rhea chicks had significantly lower erythrocyte counts, hematocrit, hemoglobin, red cell indices, and copper than did the juveniles and adult individuals. No significant differences were observed between females and males. Most values were similar to those reported for other ratites. The data obtained provide hematologic and serum chemistry values for lesser rhea from farms.     

Hematology and serum biochemistry parameters of North American river otters (Lontra canadensis).

Blood samples were obtained from 155 North American river otters (Lontra canadensis; 94 adult males, 38 adult females, 10 juvenile males, and 13 juvenile females) to establish baseline hematology and from 50 adult river otters (40 males and 10 females) for baseline serum biochemistry parameters for the species. The otters were livetrapped from eastern North Carolina (USA) during a 4-yr period. Data for 14 routine hematologic parameters and 22 serum chemistry assays showed significant differences in total leukocyte count and absolute neutrophil and monocyte numbers for adults versus juveniles, red blood cell counts and hemoglobin between adult and juvenile males, and calcium and alkaline phosphatase values for adult males between years of the study and an increase in leukocyte counts and absolute neutrophils with increased degree of trap injury sustained.     

Hematologic values in newborn beef calves.

Hematologic values were determined in 35 beef calves at birth, at 24 and 48 hours, and in 22 of these calves at 3 weeks after birth. Thirty calves did not have clinical signs of disease throughout the 3-week period. Variables that changed significantly over time in these healthy calves included hematocrit, RBC count, hemoglobin concentration, mean cell volume, mean cell hemoglobin concentration, WBC count, neutrophil-to-lymphocyte ratio, and plasma total protein and serum immunoglobulin concentrations. Of the 35 calves, 5 had clinical signs of disease at 3 weeks. Comparison of hematologic values from these calves with values for healthy calves revealed significant differences at each sample collection time, although disease was not evident at the 3 early sample times. The band neutrophil count and the neutrophil-to-lymphocyte ratio differed between the 2 groups at birth. At 24 hours, the monocyte count was higher in the 5 ill calves. At 48 hours, total leukocyte, mature neutrophil, and monocyte counts, and neutrophil-to-lymphocyte ratio also were higher in the 5 calves. At 3 weeks when clinical signs of disease were detectable in the 5 calves, the total leukocyte, band neutrophil, and mature neutrophil counts, neutrophil-to-lymphocyte ratio, and plasma total protein and fibrinogen concentrations were higher.(ABSTRACT TRUNCATED AT 250 WORDS)     

Haematological values of young male rusa deer (Cervus timorensis)

OBJECTIVE: To measure haematological values of clinical significance for rusa deer and provide reference data for farmed animals. DESIGN: Blood samples were collected regularly from eight male rusa deer from 14 days to 27 months old. PROCEDURE: Blood samples, collected by venipuncture, were analysed within 6 hours of collection for red cell count, haemoglobin, packed cell volume, plasma glucose, white cell count and differentials. RESULTS: Haemoglobin concentrations appeared to increase with age and ranged from 6.0 to 20.9 g/dL. Packed cell volume and plasma glucose concentration did not appear to vary with age. White cell counts ranged from 6.3 to 7.0 x 10(9)/L and differential counts indicated neutrophils > lymphocytes > monocytes > eosinophils > basophils. In general, the values for packed cell volume, red cell count, mean cell volumes and mean cell haemoglobin concentrations were within ranges previously reported for captive or sedated rusa deer. CONCLUSIONS: Physical restraint and resultant stress was sufficient to generate some of the effects previously reported for physically immobilised or agitated deer. The values reported here do not differ greatly from those previously reported for rusa deer and can be used as reference values for clinically healthy young farmed male rusa deer.     

Hematologic and plasma biochemical values of healthy hybrid tilapia (Oreochromis aureus x Oreochromis nilotica) maintained in a recirculating system.

Tilapia are cultured worldwide and are increasing in popularity among aquaculturalists in the United States; however, data regarding normal health parameters are limited. Few hematologic and plasma biochemical values of clinically normal tilapia have been reported, but these data may be key for identifying and managing disease issues in recirculating systems. Therefore, blood was collected from clinically normal hybrid tilapia (Oreochromis aureus x Oreochromis nilotica) housed in recirculating systems for the purpose of establishing normal hematologic and plasma biochemical reference ranges. Using standard clinical techniques the following hematologic values were determined: packed cell volume, plasma protein, leukocyte counts, leukocyte differentials, and thrombocyte counts. Additionally, the following plasma biochemical values were determined: albumin, total protein, globulins, albumin/globulin ratio, aspartate aminotransferase, alkaline phosphatase, glucose, uric acid, calcium, phosphorus, magnesium, sodium, potassium, chloride, urea nitrogen, and creatinine. The condition of the sample was also noted (lipemic, hemolysis, and icterus). The reference ranges reported in this study can be used in the management of cultured tilapia in recirculating systems.     

Peritoneal fluid values from healthy foals

Peritoneal fluid was analysed from 17 foals, aged 13 to 134 days with a mean age of 68 days. Cytologically, the peritoneal fluid was characterised by a mean total cell count of 0.45 x 10(9)/litre (range 0.06 to 1.42 x 10(9)/litre), rare eosinophils, rare cytophagia and variable percentages of neutrophils and mononuclear cells. These data indicate that peritoneal fluid nucleated cell counts over 1.50 x 10(9)/litre in the foal should be interpreted as elevated. Biochemical evaluation revealed a mean biuret protein level of 12 g/litre, mean refractive index protein level of 16 g/litre and urea nitrogen concentration of 1.96 mmol/litre. There was no correlation between the foals' white blood cell and peritoneal fluid nucleated cell counts. Results of this study indicate that adult horse reference values for evaluation of peritoneal fluid are of questionable validity for foals. Diagnostically, the most important observation was that maximum peritoneal fluid nucleated cell counts in healthy foals were much lower than reported maximal reference values for adult horses (1.5 x 10(9)/litre versus 5.0 x 10(9)/litre or 10.0 x 10(9)/litre).     

Hematologic characteristics of captive western barred bandicoots (Perameles bougainville) from Western Australia

BACKGROUND: The western barred bandicoot (Perameles bougainville) is an Australian marsupial species now considered endangered as a consequence of habitat destruction and predation. A recently discovered papillomatosis syndrome is hindering efforts to repopulate this species. Hematology reference intervals have been lacking for P bougainville, preventing optimal interpretation of hematology results from wart-affected and clinically normal animals. OBJECTIVES: The purpose of this study was to establish hematology reference values and describe morphologic characteristics of blood cells of healthy western barred bandicoots. METHODS: Fifty-nine whole blood samples were collected by jugular venipuncture into EDTA from 47 clinically healthy captive western barred bandicoots at 3 locations on the Western Australian mainland. A CBC was performed using an ADVIA-120 analyzer. Data were compared on the basis of geographic location, sex, age, and lactation status, and reference intervals were calculated. Blood cell morphology was evaluated using light microscopy, and transmission and scanning electron microscopy. RESULTS: Significant differences were found based on sex (RBC indices, fibrinogen), age (% polychromatophilic RBCs), and geographic location (RBC, neutrophil, and lymphocyte counts, MCHC, % polychromatophilic RBCs, fibrinogen). Combined reference intervals were calculated for hemoglobin concentration (122-165 g/L), HCT (0.36-0.49 L/L), and total WBC (2.9-14.9 x 10(9)/L), monocyte (0-0.6 x 10(9)/L), eosinophil (0-0.9 x 10(9)/L), and total plasma protein (47-63 g/L) concentrations. Leukocyte, erythrocyte, and platelet morphology were similar to those of other marsupial peramelid species. Nuclei in neutrophils, monocytes, and eosinophils occasionally had an annular configuration. CONCLUSIONS: Reference intervals and blood cell morphology obtained in this study will be useful for the evaluation of laboratory data from ill animals and assist with population health monitoring of western barred bandicoots.     

Comparison of the effect of dipotassium ethylenediaminetetraacetic acid and lithium heparin on hematologic values in the green iguana (Iguana iguana).

We compared the effects of dipotassium ethylenediaminetetraacetic acid (EDTA) and lithium heparin on hematologic values of green iguanas (Iguana iguana). Thirty-two privately owned sibling iguanas had blood drawn, and the sample was divided into three components: an EDTA tube, a heparin tube, and a nonanticoagulated blood smear. A full reptilian complete blood count was performed on each anticoagulated sample, and white blood cell (WBC) and leukocyte differential counts were performed on the whole-blood smears. Heparin and EDTA samples differed significantly in absolute values of thrombocytes, WBC, heterophils, and monocytes. The EDTA had no significant effect on the packed cell volume or plasma protein values, and the white blood count and differential counts produced with EDTA were more similar to those of the nonanticoagulated blood smear than were the counts produced with heparin.     

CERTAIN HEMATOLOGIC AND BLOOD CHEMICAL VALUES IN JUVENILE RHESUS MONKEYS (MACACA MULATTA)

Certain hematologic and clinical chemistry values of 148 rhesus monkeys were determined. Mean values, standard deviations and minimal-maximal values were determined for hemoglobin concentration and packed cell volume; total erythrocyte, total leukocyte and differential leukocyte count; cellular indices; prothrombin time; serum concentrations of glucose, urea nitrogen, uric acid, bilirubin, creatinine, sodium, potassium and chloride, and serum enzyme activities of alkaline phosphatase, glutamic pyruvic transaminase, glutamic oxalacetic transaminase, and lactic dehydrogenase. Calculations were made separately for males and females and the data were compared to published values for the species.     

Hematologic reference intervals for koi (Cyprinus carpio), including blood cell morphology, cytochemistry, and ultrastructure

BACKGROUND: Hematologic data are used routinely in the health care of humans and domestic mammals. Similar data for fish are largely fragmentary or have not been collected. OBJECTIVES: The primary purpose of this study was to determine hematologic reference intervals for koi, an ornamental strain of the common carp (Cyprinus carpio). Secondarily, the morphology, cytochemical reactions, and ultrastructure of koi blood cells were characterized. METHODS: A CBC was performed manually on heparin-anticoagulated blood specimens using Natt and Herrick's diluent and a Neubauer-ruled hemacytometer. Leukocyte differential counts were done on Wright-Leishman- and Diff-Quik-stained blood smears. Cytochemical reactions of koi leukocytes were determined using commercial kits. Transmission electron microscopy was performed to characterize the ultrastructural features of koi blood cells. RESULTS: Hematologic reference intervals were established for healthy koi for PCV (30-34%), hemoglobin concentration (6.3-7.6 g/dL), RBC count (1.7-1.9 X 10(6)/ microL), WBC count (19.8-28.1 X 10(3)/ microL), RBC indices, and differential leukocyte counts. Lymphocytes were the predominant leukocyte (accounting for up to 80% of all leukocytes), whereas eosinophils were rare. Basophils were positive with PAS staining. Naphthol AS-D chloroacetate esterase activity was observed only in eosinophils. alpha-Naphthyl butyrate esterase and beta-glucuronidase activities were positive in monocytes. Some lymphocytes were reactive for alpha-naphthyl butyrate esterase and acid phosphatase activity. Ultrastructurally, leukocytes, erythrocytes, and thrombocytes were identified on the basis of cytoplasmic organelles and granule appearance. CONCLUSION: Hematologic reference intervals and knowledge of the cytochemical reactions and ultrastructural characteristics of koi leukocytes will help standardize hematologic studies in this species.