Vet Clin Pathol
· 2026 Jun · PMID 42281407
·
Publisher ↗
BACKGROUND: Serial measurements of acute-phase proteins help monitor the development of and recovery from sepsis in people and dogs. To date, the kinetics of regenerating protein (REG) 3E, a novel biomarker for inflammat...BACKGROUND: Serial measurements of acute-phase proteins help monitor the development of and recovery from sepsis in people and dogs. To date, the kinetics of regenerating protein (REG) 3E, a novel biomarker for inflammation in dogs, are unknown. OBJECTIVES: The goal of our study was to assess changes in the concentrations of REG3E and compare these to C-reactive protein (CRP) concentrations over 48-72 h in different cohorts of dogs. METHODS: Plasma REG3E concentrations were measured using an ELISA on 3-4 consecutive days from 43 dogs, including 12 healthy controls, 17 undergoing tibial plateau leveling osteotomy (TPLO), and 14 with sepsis. Canine CRP was measured using a turbidimetric immunoassay. RESULTS: REG3E concentrations were low in control and TPLO dogs throughout all time points. In the TPLO group, there was a mild but significant increase in CRP over time. REG3E and CRP concentrations were higher in septic dogs at admission compared to the other groups and progressively decreased after initiation of treatment to 14% and 28% of initial levels, respectively, within 72 h. REG3E, but not CRP concentrations, stayed high in two dogs, one of which developed recurrent sepsis. There was a significant moderate correlation between both acute-phase proteins among all groups and within the sepsis group, but not within the TPLO group. CONCLUSIONS: In conclusion, based on this cohort of dogs, CRP appears to be more sensitive to detect minor postsurgical inflammation, but the potential of REG3E to predict developing or recurrent sepsis deserves further investigation.
Whetstone K, Heinz J, Grantham LE
… +4 more, Stavropoulos M, Hall C, Dröes FC, Edwards TH
Vet Clin Pathol
· 2026 May · PMID 42157618
·
Publisher ↗
BACKGROUND: A thromboelastogram viscoelastic monitor (the TEG 6s) has recently become available for use in veterinary patients. It is therefore necessary to establish reference intervals for the new device in each specie...BACKGROUND: A thromboelastogram viscoelastic monitor (the TEG 6s) has recently become available for use in veterinary patients. It is therefore necessary to establish reference intervals for the new device in each species where it is employed. This study aimed to establish accurate and acceptable reference intervals for the TEG 6s in dogs. OBJECTIVE: The goal of this study was to establish dog reference intervals (RIs) for the TEG 6s by sampling healthy adult dogs from a variety of breeds. METHODS: One hundred and thirty dogs (130) aged 1-9 years old were enrolled in this study. Inclusion criteria included healthy adult dogs with no previously reported coagulopathies, no medications that are known to alter coagulation, and normal CBC, biochemistry, and physical exam. TEG 6s measurements were performed using a Haemonetics Global Hemostasis cartridge and citrated blood sample. RESULTS: One hundred and twenty-three (123) dogs were included in the TEG 6s RIs. The determined RIs for the citrated kaolin activated were reaction time (R) 2.1-5.6 min, kinetic time (K) 1.1-2.5 min, alpha angle 64.6°-74.9°, and maximum amplitude 47.3-62.0 mm. The determined RIs for citrated Rapid TEG (CRT) maximum amplitude and reaction for the kaolin-heparinase activated were 40.0-59.3 mm and 2.1-5.6 min, respectively. CONCLUSIONS: RIs were established in a large group of normal dogs, which can assist small animal veterinarians with the interpretation of a newer viscoelastic testing method.
Lizarraga I, Reid P, Wieliczko K
… +2 more, Beeler-Marfisi J, Castillo-Alcala F
Vet Clin Pathol
· 2026 May · PMID 42144375
·
Publisher ↗
BACKGROUND: Biological samples are stored for delayed or add-on analyses. However, storage conditions, including time and temperature, can affect the stability of measurands in the samples. OBJECTIVES: To assess the stab...BACKGROUND: Biological samples are stored for delayed or add-on analyses. However, storage conditions, including time and temperature, can affect the stability of measurands in the samples. OBJECTIVES: To assess the stability of blood urea nitrogen (BUN) in cattle serum samples. METHODS: Blood samples collected from 72 heifers were immediately analyzed for BUN by a commercial laboratory and our in-house laboratory (optimal conditions, t). In-house, samples were also analyzed after storage at -20°C and -80°C for 3 and 6 months (test samples, t). Agreement between laboratories was assessed using Passing-Bablok and Bland-Altman analyses. The maximum permissible instability (MPI = 4%) was calculated as the sum of in-house intra- and inter-assay analytical variation. Percentage deviation (PD% = [(t-t)/t] × 100) was estimated for each frozen sample. Stability was assessed by comparing individual PD% against the MPI. RESULTS: Our in-house analytical methodology performed within the ASVCP recommended guidelines. BUN (mmol/L) ranged from 1.9 to 5.4 in the commercial laboratory and 1.6 to 6.1 in our in-house laboratory. There was poor agreement with proportional bias in which the higher the BUN concentration, the greater the difference between laboratories. Storage at -20°C and -80°C yielded PD% between -18.6% and 27.1% after 3 months and -17.5% and 71.6% after 6 months. Regardless of freezing temperature, 54% and 80% of PD% values were greater than the MPI after 3- and 6-month's storage, respectively. CONCLUSIONS: Although cattle serum sample degradation begins by three months storage time, further studies are needed to establish the actual stability limit for BUN.
Vet Clin Pathol
· 2026 May · PMID 42125911
·
Publisher ↗
BACKGROUND: Current veterinary guidelines for establishing reference intervals (RIs) from small samples typically recommend the robust method (RM) or, alternatively, parametric or nonparametric methods (PNPM) based on da...BACKGROUND: Current veterinary guidelines for establishing reference intervals (RIs) from small samples typically recommend the robust method (RM) or, alternatively, parametric or nonparametric methods (PNPM) based on data distribution (Gaussian or not). A web application, HARISS, refines this approach by using visual inspection of distribution histograms to more accurately select the statistical method for small samples, based on a 3-class classification of data distribution (Gaussian, lognormal, left-skewed). Prior simulation studies have shown HARISS's decision rules lead to more accurate RIs, assuming Gaussian, lognormal, or left-skewed population distributions. OBJECTIVES: This study aimed to compare the accuracy of HARISS, RM, and PNPM in estimating RIs when small samples are drawn from populations with diverse distributions. METHODS: We randomly selected samples of n = 40, 50, or 60 values (50 replicates per size) from seven simulated populations (5000 values each): Gaussian, Student's t, lognormal, right-skewed, left-skewed, bimodal, and irregular. No outlier was included. RIs were constructed using HARISS, RM, and PNPM. Their accuracy was compared using repeated-measures ANOVA. RESULTS: Overall, HARISS significantly improved the accuracy of both RI lower and upper limit estimations (p < 0.001), irrespective of population distribution or sample size. However, the methods' accuracy varied depending on the specific population distribution. Specifically, HARISS significantly outperformed RM for RIs derived from Gaussian and Student's t population samples and was significantly more accurate than PNPM for RIs from skewed population samples. CONCLUSIONS: Overall, HARISS provided the most accurate RI limit estimations from small samples across the seven population distributions simulated.
Vet Clin Pathol
· 2026 May · PMID 42095434
·
Publisher ↗
A 3-year-old female-intact German Shepherd presented for the evaluation of progressive shifting leg lameness, carpal swelling, and weight loss with a poor appetite since adoption about three months prior. Routine blood s...A 3-year-old female-intact German Shepherd presented for the evaluation of progressive shifting leg lameness, carpal swelling, and weight loss with a poor appetite since adoption about three months prior. Routine blood smear evaluation, lymph node, and synovial fluid cytologies all revealed rare neutrophils with cytoplasmic inclusions consistent with "Lupus Erythematosus (LE) cells" and "ragocytes". The combination of clinical signs, cytologic findings, and high antinuclear antibody (ANA) titer in the absence of identifiable triggers were consistent with a diagnosis of systemic lupus erythematosus (SLE).
Vet Clin Pathol
· 2026 May · PMID 42083290
·
Publisher ↗
A 1-year-old female-intact arctic fox (Vulpes lagopus) presented for diarrhea, lethargy, severe thrombocytopenia, and hyperbilirubinemia. Blood film evaluation revealed intracytoplasmic coccoid bacteria within moderate n...A 1-year-old female-intact arctic fox (Vulpes lagopus) presented for diarrhea, lethargy, severe thrombocytopenia, and hyperbilirubinemia. Blood film evaluation revealed intracytoplasmic coccoid bacteria within moderate numbers of leukocytes consistent with infection by a rickettsial organism. Similar organisms were observed within many leukocytes from fine-needle aspirates of an enlarged peripheral lymph node. The lymph node also contained moderate necrosis and an expansion of granular lymphocytes. Parasitic ovum consistent with the trematode Nanophyetus salmincola was found on fecal floatation. The patient responded to treatment with doxycycline, enrofloxacin, ampicillin/sulbactam, and other supportive therapies. Polymerase chain reaction testing confirmed the presence of Neorickettsia sp., and DNA sequencing identified the organism as Stellantchasmus falcatus agent. To the authors' knowledge, this is the first report describing this organism microscopically in peripheral blood, the first reported case of infection in this species of canid, and describes the various clinicopathologic changes with this infection.
Neal SV, Sommer SL, Golconda S
… +9 more, Hostnik ET, Sakacs R, Park J, Potts A, da Costa RC, Premanandan C, Corps K, Kim S, Wellman ML
Vet Clin Pathol
· 2026 May · PMID 42076787
·
Publisher ↗
A 7-year-old, male, castrated, mixed-breed dog presented with right pelvic limb lameness and pain on posturing to defecate that began 1.5 years prior to presentation. Radiographs taken 8 months prior to presentation show...A 7-year-old, male, castrated, mixed-breed dog presented with right pelvic limb lameness and pain on posturing to defecate that began 1.5 years prior to presentation. Radiographs taken 8 months prior to presentation showed a small lytic lesion of the L6 vertebral body, and CBCs showed a persistently increased hematocrit (HCT) for at least 2 years. Lumbar magnetic resonance imaging (MRI) revealed a right-sided L5-L7 intradural-extramedullary mass. Cytologic evaluation of a fine-needle aspirate of the mass was consistent with a diagnosis of malignant neoplasia, with a primary differential of paraganglioma. Additional testing revealed an increased normetanephrine-to-creatinine ratio and low serum erythropoietin (EPO). Histologic evaluation, immunohistochemistry, and electron microscopy were consistent with a diagnosis of spinal sclerosing paraganglioma. No mutations were found on DNA sequencing that are associated with polycythemia vera in humans, a dog with primary erythrocytosis, or in humans with polycythemia-paraganglioma syndrome (PPS). The patient was treated with palliative radiation, after which the neurologic deficits markedly improved and the erythrocytosis resolved. The patient continued to do well with an HCT within the reference interval. Two years later, the patient's neurologic signs recurred, and MRI revealed the paraganglioma had progressed in size. The patient continued to have an HCT within the reference interval and underwent repeat palliative radiation with no complications. The concurrent erythrocytosis and paraganglioma in this patient resemble reports of PPS in humans.
Vet Clin Pathol
· 2026 May · PMID 42068092
·
Publisher ↗
BACKGROUND: Pancreatic lipase assays are commonly utilized in the diagnostic approach to suspected pancreatitis in dogs. While a previously available point-of-care assay for the quantification of canine pancreatic lipase...BACKGROUND: Pancreatic lipase assays are commonly utilized in the diagnostic approach to suspected pancreatitis in dogs. While a previously available point-of-care assay for the quantification of canine pancreatic lipase immunoreactivity (Vcheck cPL 1.0) performed sub-optimally, the updated version (Vcheck cPL 2.0) lacks independent validation. OBJECTIVE: To evaluate the analytic validity of the Vcheck cPL 2.0 assay. METHODS: Linearity was assessed via dilutional parallelism. Intra- and inter-assay variability were assessed by calculating coefficients of variation across replicates on the same day and across consecutive days, respectively. Interference studies were performed to assess the effects of bilirubin, hemoglobin, and lipid presence in test samples on assay performance. RESULTS: Linearity appeared within target parameters at pancreatic lipase concentrations up to ~800 md/dL (mean O/E ratio: 0.968 ± 0.104) but was unacceptable at pancreatic lipase concentrations > 2000 mg/dL (mean O/E ratio: 1.432 ± 0.178). Intra-assay precision (median coefficient of variation 7.2%) was acceptable; however, inter-assay precision was sub-optimal (coefficient of variation > 10%) in 50% of samples analyzed. Variable precision may result in different diagnostic interpretations. Icterus and lipemia did not impact assay results. However, hemolysis, when severe (375-500 mg/dL), may impact test results (p < 0.005). CONCLUSIONS: The Vcheck cPL 2.0 shows improved precision compared to previous versions. Linearity was unacceptable at higher pancreatic lipase concentrations and hemolysis may impact Vcheck cPL 2.0 assay results.
BACKGROUND: Harmonization enables direct comparison of patient results from different analyzers and/or measurement techniques and the use of common reference intervals and decision thresholds. Point-of-care analyzers (PO...BACKGROUND: Harmonization enables direct comparison of patient results from different analyzers and/or measurement techniques and the use of common reference intervals and decision thresholds. Point-of-care analyzers (POC) cannot be calibrated by the end user, so regression offers an opportunity for calculated harmonization. This approach has been used to harmonize (POC) with commercial laboratory analyzers (CL) with > 90 paired feline plasma samples. OBJECTIVE: Our aims were to use 20 pairs of results to derive regression equations to harmonize the POC results with those of the CL. Harmonized results were assessed by recognized quality goals. METHODS: Systematic sampling was used to choose the 20 samples used to calculate regression parameters (with three techniques). These were applied to the remainder of the original dataset to harmonize the POC results to those from the CL. POC results were assessed for bias and the number of results reaching quality goals before and after adjustment. RESULTS: Regression from 20 paired samples was able to harmonize POC analyzer results to a commercial laboratory analyzer for 16 feline plasma biochemistry measurands. CONCLUSIONS: Most analytes could be harmonized by this approach. Reasons for harmonization failure were inadequate concentration range of patient results (AST, lipase, total bilirubin, GGT), potential non-linear relationship (lipase), and no clinical performance goal (globulins). Additional studies using samples with clinically important values may overcome these limitations, and similar studies of other species and other types of point-of-care analyzers would be of interest.
INTRODUCTION: Codocytes, or target cells, are a morphologic variation of erythrocytes characterized by increased membrane surface area relative to volume. In dogs, codocytosis is frequently noted on blood smear evaluatio...INTRODUCTION: Codocytes, or target cells, are a morphologic variation of erythrocytes characterized by increased membrane surface area relative to volume. In dogs, codocytosis is frequently noted on blood smear evaluation, but its clinical significance remains poorly understood. OBJECTIVES: To characterize the clinical conditions associated with codocytosis in dogs and assess relationships between codocyte severity and hematologic, biochemical, and diagnostic findings. METHODS: A retrospective descriptive study was conducted on 345 dogs from a veterinary teaching hospital from September 2020 through October 2022. Dogs were included if their CBC report described codocytosis graded as ≥ 2+. Data collected included CBC results, serum biochemistry, clinical diagnoses, and current medications. Codocytosis was graded semi-quantitatively (2+ to 4+). Statistical comparisons were made using chi-square tests. RESULTS: Codocytosis ≥ 2+ was identified in 2.5% of 14 010 canine CBCs during the study. The majority (81.4%) had 2+ codocytes. There were no significant differences in anemia, macrocytosis, or regeneration across grades. Dogs with 3+ or 4+ codocytosis had significantly higher alanine transferase (ALT) activity and were more frequently diagnosed with hepatobiliary (p = 0.006) and gastrointestinal (p = 0.009) diseases. Endocrine disease, particularly diabetes mellitus was also frequently observed. Antibiotic use was significantly associated with higher codocyte grades (p < 0.001), though no causation was identified. CONCLUSIONS: Codocytosis in dogs is not primarily associated with anemia or regeneration, but rather appears more frequently in diseases affecting hepatobiliary, gastrointestinal, and endocrine systems. These findings suggest a multifactorial pathogenesis likely involving altered lipid metabolism and membrane remodeling. Recognition of codocytosis may provide insight into systemic disease processes in canine.
Schaefer DMW, Smith JS, Mulon PY
… +1 more, Hampton CE
Vet Clin Pathol
· 2026 Mar · PMID 41906331
·
Publisher ↗
BACKGROUND: Published descriptions of expected findings for laboratory analysis of peritoneal fluid (PTF) from companion pigs are lacking. OBJECTIVES: To describe routine PTF analysis results from clinically healthy and...BACKGROUND: Published descriptions of expected findings for laboratory analysis of peritoneal fluid (PTF) from companion pigs are lacking. OBJECTIVES: To describe routine PTF analysis results from clinically healthy and ill companion pigs. METHODS: Intraoperative PTF was collected prospectively from 32 healthy, sexually mature companion sows during ovariohysterectomy, and PTF analyses from 10 clinically ill companion pigs, with fluid collected transabdominally, were identified retrospectively. Clinically ill pigs had peritonitis (n = 4), uterine neoplasia (n = 4), end stage renal disease (n = 1) or disseminated intravascular coagulation (DIC, n = 1). Analysis included total nucleated cell concentration (TNCC), total protein by refractometry (TP-ref), and differentials of nucleated cells. RESULTS: In PTF from healthy pigs, TNCC was commonly < 10 000/μL (range 190-14 670/μL) and TP-ref was < 6.0 g/dL, with percentage differentials predominated by mononuclear cells (macrophages, lymphocytes, and individualized mesothelial cells) with < 15% neutrophils and < 10% eosinophils. Pigs with peritonitis or uterine neoplasia had TNCC similar to that of healthy pigs, but with higher percentages of neutrophils, often > 50%, and lower percentages of large mononuclear cells, often < 25%. Most pigs with neoplasia had tumor-associated inflammation, likely contributing to higher neutrophil percentages. Two patients with peritonitis and one with DIC had higher TP-ref than the healthy pigs. CONCLUSIONS: PTF fluid from companion pigs with peritonitis or uterine neoplasia, particularly with concurrent inflammation, may have similar TNCC but increased proportions of neutrophils compared to healthy sows. Therefore, in contrast to some other large animal species, high percentages of neutrophils in PTF may warrant evaluation for peritonitis or abdominal neoplasia in pigs, regardless of the TNCC.
Dröes FC, Jeffery U, Rutter CR
… +3 more, Suchodolski JS, Steiner JM, Lidbury JA
Vet Clin Pathol
· 2026 Mar · PMID 41797002
·
Publisher ↗
BACKGROUND: Hemostasis has traditionally been assessed by clotting times; however, point-of-care viscoelastic coagulometers have added to the diagnostic options for assessing hemostasis. Biological variation of diagnosti...BACKGROUND: Hemostasis has traditionally been assessed by clotting times; however, point-of-care viscoelastic coagulometers have added to the diagnostic options for assessing hemostasis. Biological variation of diagnostic tests determines whether population- or subject-based reference intervals are used for result interpretation. OBJECTIVE: To assess the biological variation of hemostatic parameters in healthy dogs when using a point-of-care viscoelastic coagulometer, VCM Vet. MATERIALS AND METHODS: Nine healthy dogs. Four VCM Vet tests were simultaneously performed on four different machines immediately after sample collection for each dog. Repeated VCM Vet tests were performed on each dog a total of four times every 28 days. A nested ANOVA was used to assess inter-individual (CVg), intra-individual (CVi), and analytical variation (CVa). Indexes of individuality (IoI) and reference change values (RCV) were calculated. RESULTS: Two dogs were suspected to have artifactual clot retraction during ten separate tests. These tests were excluded from further analysis. CVg, CVi, CVa, IoI, and RCV for each of the hemostatic parameters were: clotting time (4.4, 11.5, 8.2, 3.2, 39.2%); clot formation time (10.0, 10.8, 8.9, 1.4, 38.7%), alpha (4.5, 4.8, 4.2, 1.4, 17.6%), maximum clot formation (5.3, 6.4, 6.0, 1.7, 24.4%), amplitude at 10 min (6.6, 8.8, 6.6, 1.7, 30.5%), and amplitude at 20 min (6.1, 7.5, 6.1, 1.6, 26.8%). CONCLUSIONS: Intra-individual variation was higher than inter-individual variation for parameters measured with this point-of-care viscoelastic coagulometer. Based on an IOI above > 1.4 for all parameters assessed, population-based reference intervals are appropriate for this point-of-care viscoelastic coagulometer.
Phillips EM, Blois SL, Monteith G
… +3 more, Abrams-Ogg ACG, Wood RD, Cuq B
Vet Clin Pathol
· 2026 Mar · PMID 41794405
·
Full text
BACKGROUND: Thrombin generation assessment using calibrated automated thrombography (CAT) requires a standard volume of plasma (80 μL) and reagent (20 μL) run in triplicate. A CAT method using lower plasma and reagent vo...BACKGROUND: Thrombin generation assessment using calibrated automated thrombography (CAT) requires a standard volume of plasma (80 μL) and reagent (20 μL) run in triplicate. A CAT method using lower plasma and reagent volumes could benefit veterinary patients as it lowers the sampling burden in patients. OBJECTIVES: To compare standard CAT methodology to a low plasma and low reagent volume CAT method in dogs. METHODS: Platelet-poor plasma samples were obtained by direct jugular venipuncture in dogs with low (n = 10), high (n = 10), and normal (n = 10) thrombin generation potential, recruited from a tertiary referral hospital. Method comparison was performed between standard CAT (80 μL plasma, 20 μL reagent- method 1) and low volume CAT (40 μL plasma, 10 μL reagent- method 2). Lag time (lag time); endogenous thrombin potential (ETP); peak (peak); and time to peak (ttpeak) were assessed on the thrombin generation curves. RESULTS: There was excellent agreement between methods 1 and 2 for all parameters. Lin's concordance coefficients were 0.97, 0.94, 0.96, and 0.96 for lag time, ETP, ttpeak, and peak, respectively. There was a small bias for all parameters (p < 0.05), resulting in a significant change for lag time only. There was a significant predictive linear equation that, when modeled, allowed the conversion of method 2 to method 1 values. The methods had similar variation when measuring TG variables. CONCLUSIONS: Based on the results in this small group of dogs, we conclude that low-volume CAT appears to be an alternative to the standard testing method in dogs.
Leineweber C, Geisler G, Öfner S
… +1 more, Marschang RE
Vet Clin Pathol
· 2026 Mar · PMID 41787686
·
Publisher ↗
BACKGROUND: Vitamins are important for the health and metabolism of living organisms, but little is known about the physiologic blood vitamin levels of tortoises. OBJECTIVES: The aim of this study was therefore to measur...BACKGROUND: Vitamins are important for the health and metabolism of living organisms, but little is known about the physiologic blood vitamin levels of tortoises. OBJECTIVES: The aim of this study was therefore to measure vitamins A, B, B, B, B, B, and E in blood samples of Hermann's tortoises (Testudo hermanni). METHODS: Vitamin concentrations were measured in heparinized whole blood and plasma of 522 Hermann's tortoises from May to September 2022 using high-performance liquid chromatography (HPLC) and spectrophotometry to evaluate possible seasonal and sex-specific influences on the blood vitamin concentrations and to establish reference intervals for these vitamins in Hermann's tortoises. RESULTS: The measured median concentrations were: vitamin A 275.00 μg/L, B 42.40 μg/L, B 483.00 μg/L, B 4.26 μg/L, B 13.24 ng/mL, B 933.20 pg/mL, and E 2.50 mg/L. Significant (p ≤ 0.05) seasonal variations were found for vitamins A, B, B, B, B, B, and E, and sex-specific variations for vitamins A, B, B, and E. CONCLUSIONS: The established reference intervals serve as a basis for health monitoring and for further studies on possible deficiencies or oversupply and their clinical significance. In addition, it was shown that the season of sample collection and the sex should be considered when interpreting blood vitamin levels in tortoises.
LaDouceur EEB, Barrett AE, Armien AG
… +2 more, Lore D, Stacy NI
Vet Clin Pathol
· 2026 Mar · PMID 41772860
·
Publisher ↗
OBJECTIVE: In teleosts, it is often difficult to differentiate between cutaneous neoplasms originating from peripheral nerves and fibrous tissues, and chromatophoromas, and ultrastructural examination is frequently requi...OBJECTIVE: In teleosts, it is often difficult to differentiate between cutaneous neoplasms originating from peripheral nerves and fibrous tissues, and chromatophoromas, and ultrastructural examination is frequently required in addition to light microscopy for definitive diagnosis. METHOD: A 10-year-old koi (Cyprinus rubrofuscus) had a 3-week history of a 3-4 cm, cutaneous, ulcerated, multinodular mass on the dorsum. RESULTS: Fine needle aspiration revealed spindle cells with abundant, cytoplasmic, variably granular magenta pigment and moderate cellular and nuclear atypia. Biopsy was performed several weeks later and revealed neoplastic spindle cells forming intersecting streams separated by collagenous stroma. Ultrastructural examination showed neoplastic cells had cytoplasmic extensions that were continuous with collagen fibrils, consistent with fibrosarcoma; neoplastic cells lacked pigment and refracting organelles, ruling out chromatophoroma. The nature of the intracytoplasmic pigmented material observed on cytology is uncertain. Neither ultrastructural nor histologic correlates for this cytologically evident material were identified, and differentials for the material are discussed. DISCUSSION: In domestic animals, the presence or absence of concurrent inflammation can often be used to differentiate spindle cell neoplasia from fibroplasia on cytology; this feature is less useful in teleosts, however, as sarcomas are often heavily inflamed and chronic granulomas can be composed of a nearly uniform population of spindle cells. Additionally, there is morphologic overlap between poorly pigmented chromatophoromas and variants of soft tissue sarcoma (STS). For these reasons, definitive diagnosis of STS, chromatophoromas, and many other neoplasms in teleosts often requires a combination of light microscopy and transmission electron microscopy.