Veterinary · Clinical · also MSVE VLD-411, Clinical Pathology
Using the laboratory well: haematology, biochemistry panels, organ function tests, urinalysis, cytology, sampling for microbiology, reference intervals and quality control.
In the exams: NAVLE data gathering and interpretation 35%; BCSE diagnostics 22-25 items; PAVE QSE diagnostics 23% (to Sep 2026)
Veterinary clinical pathology, which includes clinical biochemistry and laboratory diagnosis, is the use of blood, urine, body fluids and cells to find out what is wrong with an animal, how badly, and whether treatment is working. It joins physiology and pathology to bedside medicine. A practitioner orders a complete blood count, a biochemistry panel, a urinalysis or a fine-needle aspirate every day, in clinics from Ludhiana to Davis. Most errors in the process happen not in the analyser but in choosing, collecting and handling the sample, and in reading the result without regard to species.
This manual is written for the student preparing for licensing and postgraduate examinations and for the working veterinarian. It covers sample collection and handling, haematology (complete blood count interpretation, anaemia classification, leukogram patterns by species and haemostasis), biochemistry panels (proteins, electrolytes, minerals, muscle, pancreas, glucose, liver and kidney), urinalysis, cytology, sampling for microbiology, endocrine testing principles, reference intervals, test performance statistics, point-of-care testing and quality control.
Two warnings run through every chapter. First, reference intervals depend on species, breed, age, physiological state, analyser and method; this manual quotes numbers only where they are standard teaching values or published staging cut-offs, and you should always use the interval supplied by the laboratory that ran the test. Second, a laboratory result is a piece of evidence whose value depends on the probability of disease before the test; the chapter on test performance shows how to calculate that.
Globally, the subject is examined as data interpretation: NAVLE data gathering and interpretation (about 35%), BCSE and PAVE QSE diagnostics, the Australasian Veterinary Examination pathology items, and India's ICAR postgraduate and PhD papers under pathology and medicine.
Laboratory error is usually divided into three phases. Pre-analytical error happens before the sample reaches the analyser: wrong patient, wrong tube, wrong fill volume, delay, heat, haemolysis or a stressed animal. Analytical error happens inside the instrument. Post-analytical error is a transcription, unit or interpretation mistake. Quality studies in human and veterinary laboratories consistently find that most errors are pre-analytical, which means the clinician and nurse, not the laboratory, control most of the accuracy of a result.
Collect blood with a needle of adequate bore, a clean single venepuncture and minimal suction, because a traumatic draw activates platelets (clumps, falsely low platelet count) and ruptures red cells (haemolysis). Jugular samples are preferred in dogs and cats for haematology and coagulation; cephalic or saphenous samples are acceptable for small volumes. In cattle the coccygeal (tail) vein and jugular vein are used; in horses the jugular. Record the time of the last meal and any drugs given, because both change results. A fasting sample (commonly about 12 hours in dogs and cats; not applicable to ruminants) reduces post-prandial lipaemia and post-prandial rises in bile acids, urea and glucose.
Each anticoagulant serves a purpose. EDTA (ethylenediaminetetraacetic acid, lavender top) chelates calcium and preserves cell morphology best, so it is the tube for the complete blood count (CBC) and blood smear. EDTA destroys the red cells of some bird species and many reptiles, notably chelonians, so lithium heparin is used for their haematology. Sodium citrate (light blue top) at a fixed ratio of nine parts blood to one part citrate is the tube for coagulation tests; under-filling dilutes the plasma and prolongs clotting times falsely. Lithium heparin (green top) gives plasma for biochemistry quickly, without waiting for a clot. Plain or clot-activator tubes (red top, or gold with gel) give serum; serum is required for some immunoassays and protein electrophoresis. Sodium fluoride with potassium oxalate (grey top) inhibits glycolysis and is used when glucose or lactate cannot be separated promptly.
Fill additive tubes in a fixed order so that additive does not carry over: blood-culture bottles, then citrate, then serum, then heparin, then EDTA, then fluoride. EDTA carried into a serum or heparin tube is a classic cause of a result showing marked hyperkalaemia with severe hypocalcaemia in a clinically well animal, because the potassium salt of EDTA adds potassium and the EDTA chelates calcium.
Haemolysis releases intracellular contents and interferes optically. It raises potassium only in species whose red cells are rich in potassium, such as horses, pigs and dogs of some East Asian breeds such as the Akita; in most dogs, cats, cattle and sheep the red cell potassium is low. Haemolysis also raises phosphorus, lactate dehydrogenase (LDH) and aspartate aminotransferase (AST) and can interfere with bilirubin and many photometric assays. Lipaemia scatters light, can falsely change many analytes, causes pseudohyponatraemia with indirect ion-selective electrodes, and haemolyses cells in the tube. Icterus interferes with some creatinine, cholesterol and protein methods.
If whole blood stands unseparated at room temperature, red and white cells keep consuming glucose, so glucose falls by several percent per hour; separate serum or use fluoride. Cell swelling during storage raises the mean cell volume (MCV) and packed cell volume (PCV) in EDTA blood, and leukocyte morphology degrades within hours, so make a blood smear within an hour of collection and send it unstained and unrefrigerated with the tube. Refrigerate (do not freeze) EDTA blood for the CBC; freezing lyses cells. Separate serum within about an hour for biochemistry. Ammonia and adrenocorticotropic hormone (ACTH) are unstable and need special handling described in later chapters.
Patient factors matter as well. Excitement releases adrenaline, which mobilises marginated neutrophils and (in cats) lymphocytes and causes splenic contraction and hyperglycaemia. Cats in particular may show stress hyperglycaemia high enough to be mistaken for diabetes mellitus.
The CBC reports red cell count (RBC), haemoglobin (Hb), PCV or haematocrit (HCT), and red cell indices. PCV is measured by spinning a microhaematocrit tube; HCT is calculated by an analyser from RBC and MCV. The spun tube also gives the plasma colour (clear, icteric, haemolysed, lipaemic), the buffy coat height, and plasma for total protein by refractometer, which together are the most informative five-minute test in practice.
The indices are derived. MCV (fL) = PCV (%) × 10 ÷ RBC (× 10^12/L). Mean cell haemoglobin concentration, MCHC (g/dL) = Hb (g/dL) × 100 ÷ PCV (%). Mean cell haemoglobin, MCH (pg) = Hb (g/dL) × 10 ÷ RBC (× 10^12/L). Red cell distribution width (RDW) measures size variation. Species differ greatly: goats and sheep have small red cells, dogs larger ones; camelid red cells are elliptical; healthy cats and horses often show rouleaux formation and small numbers of Howell-Jolly bodies; canine red cells have obvious central pallor while those of cats and horses have little. A falsely high MCHC usually means haemolysis or lipaemia (Hb measured includes free or turbid material), not a true increase, because red cells cannot be over-filled with haemoglobin.
Anaemia is a reduced PCV, Hb or RBC. Classify it first by the marrow response: regenerative or non-regenerative. In dogs and cats regeneration is judged by the absolute reticulocyte count (reticulocyte percentage × RBC), not by the percentage alone, because the percentage rises artefactually as the RBC falls. Cats release two forms; count aggregate reticulocytes, which reflect recent production, and ignore the long-lived punctate forms. It takes about three to five days after blood loss or haemolysis for a reticulocyte response to appear, so an anaemia less than a few days old looks pre-regenerative. Horses do not release reticulocytes into blood even when the marrow is responding; regeneration in horses is judged by serial PCV, a rising MCV or RDW, or a bone marrow aspirate. Ruminants mount a modest reticulocyte response and may show basophilic stippling when regenerating.
A corrected reticulocyte percentage = observed % × (patient PCV ÷ normal PCV). Some clinicians also divide by a maturation time to give a reticulocyte production index; this was adapted from human medicine and is an approximation.
Next, use morphology and plasma protein to find the mechanism. Blood loss: total protein falls with the PCV in acute external haemorrhage; chronic external loss causes iron deficiency with microcytic, hypochromic cells, keratocytes and often thrombocytosis. Haemolysis: protein normal, plasma may be pink (intravascular) and bilirubin rises; spherocytes and agglutination suggest immune-mediated haemolytic anaemia (IMHA) in dogs; Heinz bodies and eccentrocytes suggest oxidant injury (onions and garlic in dogs and cats, paracetamol in cats, red maple leaves in horses, chronic copper accumulation in sheep, brassicas in ruminants); organisms such as Babesia, Theileria, Anaplasma and haemotropic mycoplasmas are seen on the smear or detected by PCR. Reduced production: normocytic, normochromic, non-regenerative anaemia from inflammation (the commonest), chronic kidney disease (erythropoietin deficit), endocrine disease, marrow disease or drugs.
A high PCV is relative or absolute. Relative polycythaemia comes from dehydration (protein also high) or splenic contraction (excitement, especially in horses and some dog breeds). Absolute polycythaemia is primary (a myeloproliferative disorder, rare) or secondary to hypoxia (cardiopulmonary disease, altitude) or inappropriate erythropoietin (renal tumours). Sighthounds such as greyhounds normally have higher PCV than other dogs; this is a reason to use breed-specific intervals where they exist.
Practical checks: always look at the smear, because analysers miss agglutination, Heinz bodies, parasites and platelet clumps. For suspected agglutination, mix a drop of blood with saline on a slide; true agglutination persists, rouleaux disperses. Neonates have different values: foals, calves and puppies typically have lower PCV and different MCV in the first weeks of life. Finally, read the red cell results together with the white cells, platelets and total protein, because a pattern across the whole CBC, not a single number, carries the diagnosis.
Always interpret absolute counts (total white cell count × differential percentage), never percentages alone. Recognise four core patterns. A physiological (adrenaline) leukocytosis is caused by fear or exercise: mature neutrophilia from the marginal pool and, especially in cats and young animals, lymphocytosis; it lasts minutes to an hour. A stress (corticosteroid) leukogram follows endogenous cortisol or glucocorticoid treatment: mature neutrophilia, lymphopenia, eosinopenia, and in dogs often monocytosis. An inflammatory leukogram shows neutrophilia with a left shift (increased band neutrophils), often with toxic change, and sometimes monocytosis. A degenerative left shift, where immature neutrophils outnumber mature ones or a left shift occurs with a normal or low total neutrophil count, means tissue demand exceeds marrow supply and is a poor prognostic sign.
Toxic change describes cytoplasmic features of accelerated production: Döhle bodies, diffuse basophilia, foamy vacuolation and, when severe, toxic granulation. A few Döhle bodies are common in cats with mild disease. Toxic change suggests significant inflammation, often bacterial, but is not specific.
An absence of a stress leukogram in a sick, stressed dog (normal or high lymphocyte and eosinophil counts) should raise suspicion of hypoadrenocorticism. Eosinophilia suggests parasitism, hypersensitivity, eosinophilic diseases or some tumours, notably mast cell tumours. Persistent marked lymphocytosis suggests lymphoid leukaemia, or in cattle, persistent lymphocytosis associated with bovine leukaemia virus.
Dogs, cats and horses normally have more neutrophils than lymphocytes. Cattle and other ruminants normally have more lymphocytes than neutrophils, a neutrophil to lymphocyte ratio around 1:2 in adults. Ruminants also have a small marrow storage pool of neutrophils. In acute severe bovine inflammation such as acute coliform mastitis, metritis or peritonitis, the first one to two days typically show neutropenia with a left shift, and leukocytosis is often modest even later. A neutrophil count greater than the lymphocyte count in an adult cow (a reversed ratio) is itself a clue to inflammation or stress. Plasma fibrinogen, measured by heat precipitation, is a sensitive inflammation marker in cattle and horses and may rise when the leukogram is unremarkable.
Horses show strong physiological responses, and in horses a serum amyloid A rise is a rapid, sensitive marker of inflammation. Camelids have high normal white cell counts. Birds and reptiles have heterophils instead of neutrophils and nucleated red cells and thrombocytes, so automated counters that work by impedance cannot give a valid differential for them; manual counts are used. Nucleated red blood cells in a mammalian blood film are counted as white cells by many counters, so the white cell count is corrected: corrected WBC = observed WBC × 100 ÷ (100 + nRBC per 100 WBC).
Thrombocytopenia arises from decreased production, increased consumption (disseminated intravascular coagulation, DIC, or haemorrhage), destruction (immune-mediated thrombocytopenia) or sequestration (splenomegaly). Before acting, exclude pseudothrombocytopenia: platelet clumps from a difficult draw, very common in cats, whose large platelets may also be counted as red cells by impedance analysers. Cavalier King Charles Spaniels can inherit a macrothrombocytopenia that causes no bleeding, and greyhounds normally have lower platelet counts than other dogs. On a smear, each platelet seen per oil-immersion field roughly reflects a count, and a count below about 30-50 × 10^9/L is the range at which spontaneous bleeding becomes a concern; exact thresholds differ by source.
Secondary haemostasis is assessed by the prothrombin time (PT; extrinsic and common pathways, including factor VII), the activated partial thromboplastin time (aPTT; intrinsic and common pathways) and the activated clotting time (ACT), a crude in-clinic intrinsic test. Primary haemostasis (platelet function and von Willebrand factor) is assessed by buccal mucosal bleeding time when the platelet count is adequate. Patterns: anticoagulant rodenticide toxicosis prolongs PT first, because factor VII has the shortest half-life of the vitamin K-dependent factors, then both PT and aPTT; haemophilia A (factor VIII deficiency) prolongs aPTT with a normal PT; von Willebrand disease prolongs bleeding time with a normal platelet count; DIC typically shows thrombocytopenia, prolonged PT and aPTT, increased fibrin degradation products and D-dimers, low antithrombin and schistocytes.
Total protein is measured in serum or plasma by refractometer or biuret method; plasma includes fibrinogen and so reads slightly higher. Albumin is measured by dye binding; globulin is calculated as total protein minus albumin. Dehydration raises both. Hypoalbuminaemia arises from decreased production (chronic liver failure, severe malnutrition), loss (protein-losing nephropathy, protein-losing enteropathy, haemorrhage, exudative skin lesions or burns) or sequestration in effusions; it is also a negative acute phase response in inflammation. A useful rule: protein-losing enteropathy and haemorrhage usually cause panhypoproteinaemia (both fractions low), whereas protein-losing nephropathy and liver failure usually lower albumin selectively.
Hyperglobulinaemia is polyclonal (a broad gamma-region rise on electrophoresis) in chronic antigenic stimulation such as chronic infection or feline infectious peritonitis, or monoclonal (a narrow spike) in multiple myeloma and other B-cell tumours, and occasionally in infections such as canine ehrlichiosis and leishmaniasis. Acute phase proteins are produced by the liver in response to inflammatory cytokines. The major ones differ by species: C-reactive protein in dogs and pigs, serum amyloid A in cats and horses, haptoglobin and serum amyloid A in cattle, and fibrinogen in cattle and horses.
Sodium and water go together, so interpret sodium as a statement about water balance. Hyponatraemia with hyperkalaemia, and a sodium to potassium ratio below about 27:1, is classically associated with hypoadrenocorticism in dogs; but a low ratio also occurs with renal failure, urinary obstruction or uroabdomen, gastrointestinal disease including whipworm infection, and repeated drainage of effusions, so it is a screening clue, not a diagnosis. Hyperkalaemia also arises from urethral obstruction (classically blocked male cats), uroabdomen, acute kidney injury, severe tissue damage and pseudohyperkalaemia (haemolysis in high-potassium species, marked thrombocytosis, EDTA contamination). Hypokalaemia is common in anorexic cattle and in cats with chronic kidney disease.
The anion gap is (sodium + potassium) minus (chloride + bicarbonate). An increased gap with low bicarbonate indicates metabolic acidosis from unmeasured anions: lactate, ketones, uraemic acids or toxic metabolites such as those of ethylene glycol. A normal-gap (hyperchloraemic) acidosis indicates bicarbonate loss, for example in diarrhoea of calves.
In cattle, abomasal displacement, abomasal volvulus or any obstruction to abomasal outflow traps hydrochloric acid in the abomasum and causes a hypochloraemic, hypokalaemic metabolic alkalosis. In severe cases urine becomes acidic despite systemic alkalosis (paradoxical aciduria), because the kidney conserves sodium and potassium at the expense of hydrogen ions. Calculated serum osmolality approximates 2(Na + K) + glucose + urea when all are in mmol/L; a measured osmolality much higher than calculated (an increased osmolal gap) indicates an unmeasured osmole, classically ethylene glycol in the first hours after ingestion.
Total calcium includes protein-bound, complexed and ionised fractions; only ionised calcium is biologically active. Hypoalbuminaemia lowers total calcium without necessarily lowering ionised calcium, and correction formulas for total calcium have proved unreliable, so measure ionised calcium when it matters. Hypercalcaemia in dogs is most often due to malignancy, especially lymphoma and apocrine gland adenocarcinoma of the anal sac, which secrete parathyroid hormone-related protein; other causes include primary hyperparathyroidism, hypoadrenocorticism, vitamin D toxicosis (including calcipotriol and some rodenticides), granulomatous disease and kidney disease; idiopathic hypercalcaemia is common in cats. Hypocalcaemia occurs in parturient paresis (milk fever) of dairy cows, puerperal tetany (eclampsia) of bitches, ethylene glycol toxicosis, acute pancreatitis and hypoparathyroidism.
Phosphorus is normally higher in young growing animals, rises with reduced glomerular filtration and with haemolysis, and is low in post-parturient haemoglobinuria of cattle and in refeeding syndrome. Hypomagnesaemia underlies grass tetany in grazing ruminants, often with concurrent hypocalcaemia; because blood magnesium is unreliable after death while vitreous humour magnesium stays relatively stable for a time, vitreous humour (or aqueous humour) from a fresh carcass is used for post-mortem support of the diagnosis.
Creatine kinase (CK) is concentrated in skeletal and cardiac muscle and is very sensitive to muscle injury. It rises within hours, peaks within roughly a day and, because its half-life is short, falls quickly once damage stops. AST is present in muscle and liver and has a longer half-life, so it remains raised for days. Reading the two together dates the injury: high CK with modest AST is recent; falling CK with persistently high AST is resolving; persistently high CK means ongoing damage. Intramuscular injections, recumbency, seizures, a difficult restraint and even a traumatic venepuncture can raise CK mildly. LDH is widespread and has little specificity.
Clinically important myopathies include exertional rhabdomyolysis in horses, nutritional myodegeneration (white muscle disease) from vitamin E or selenium deficiency in calves, lambs, kids and foals, capture myopathy in wildlife, and the muscle damage of prolonged recumbency in downer cows. Severe muscle injury releases myoglobin, which is filtered by the glomerulus and can damage tubules.
Serum amylase and total lipase activities are inexpensive but have limited specificity for pancreatitis in dogs and cats: both are produced by other tissues, both are cleared in part by the kidney, so azotaemia raises them, and both may be normal in genuine pancreatitis. Species-specific pancreatic lipase immunoreactivity (PLI) assays, both laboratory and in-clinic lateral-flow versions, are more sensitive and specific and are preferred; even so, results must be combined with clinical signs and ultrasonography.
For exocrine pancreatic insufficiency (EPI), the test of choice in dogs and cats is fasting serum trypsin-like immunoreactivity (TLI): a markedly low result is diagnostic, and the disease is classically seen as pancreatic acinar atrophy in young German Shepherd dogs. Serum cobalamin (vitamin B12) and folate help localise intestinal disease: cobalamin falls with distal small intestinal disease and with EPI, while folate may rise with small intestinal dysbiosis. Hypocobalaminaemia is worth detecting because it can itself impair recovery and is corrected by supplementation. In ruminants and horses these pancreatic tests are not used routinely.
Hyperglycaemia arises from diabetes mellitus, stress (marked in cats), glucocorticoid excess, drugs such as alpha-2 agonists, and post-prandial sampling. Hypoglycaemia arises from delayed separation of the sample (an artefact), sepsis, insulinoma, liver failure, hypoadrenocorticism, xylitol in dogs, neonatal starvation and pregnancy toxaemia in ewes. Fructosamine is glycated serum protein and reflects average glycaemia over roughly the previous one to three weeks; it is not affected by momentary stress, so it helps separate stress hyperglycaemia from diabetes in cats and is used in monitoring. Hypoproteinaemia lowers fructosamine, and hyperthyroidism can lower it by speeding protein turnover.
In dairy cows, negative energy balance after calving is monitored with non-esterified fatty acids (NEFA, reflecting fat mobilisation, usually measured before calving) and beta-hydroxybutyrate (BHB, the main circulating ketone, measured after calving). Thresholds for subclinical ketosis in the literature are commonly in the range 1.2-1.4 mmol/L BHB in blood; herd-level interpretation uses the proportion of sampled cows above the threshold. Hand-held meters that measure blood BHB are widely used on farms. Ketosis also occurs in pregnancy toxaemia of ewes and does and in diabetic ketoacidosis in dogs and cats.
Cholesterol rises in hypothyroidism, diabetes mellitus, hyperadrenocorticism, protein-losing nephropathy and cholestasis, and with post-prandial sampling. Hypertriglyceridaemia occurs post-prandially and in primary hyperlipidaemia of Miniature Schnauzers. In ponies, donkeys and miniature horses, negative energy balance can trigger hyperlipaemia, with grossly lipaemic plasma and high triglycerides, a serious condition.
Liver tests fall into two groups. Enzyme activities show injury or cholestasis but not function. Function tests show what the liver can still do.
Leakage enzymes are released from damaged hepatocytes. Alanine aminotransferase (ALT) is liver-specific and the main hepatocellular marker in dogs, cats and primates, but liver ALT content is too low in horses and ruminants to be useful. AST leaks from hepatocytes and muscle, so an AST rise is interpreted with CK. Sorbitol dehydrogenase (SDH) and glutamate dehydrogenase (GLDH) are liver-specific in horses and ruminants; SDH is unstable in vitro and must be measured quickly, while GLDH is more stable. The magnitude of a leakage enzyme rise reflects the number of injured cells, not the reversibility of the injury or the remaining function.
Induced (cholestatic) enzymes are produced in larger amounts when bile flow is impaired or when drugs induce them. Alkaline phosphatase (ALP) has several isoenzymes: bone ALP is high in young growing animals; in dogs only, glucocorticoids (endogenous or given) induce a separate isoenzyme, and some anticonvulsants also raise ALP. Cats have little hepatic ALP and its half-life is short, so any ALP rise in a cat is significant; in feline hepatic lipidosis ALP is often markedly increased while gamma-glutamyl transferase (GGT) is only mildly increased, whereas cholangitis and extrahepatic bile duct obstruction tend to raise GGT more. In horses and ruminants, ALP reference intervals are very wide, and GGT is the preferred cholestatic marker. Colostrum is rich in GGT, so calves, lambs and kids that have suckled colostrum have high serum GGT in their first days, which has been used as a rough indicator of passive transfer.
Bilirubin is classified by origin. Pre-hepatic hyperbilirubinaemia follows haemolysis (anaemia present); hepatic follows reduced uptake or conjugation; post-hepatic follows bile duct obstruction or rupture. In horses, simple anorexia raises unconjugated bilirubin markedly (fasting hyperbilirubinaemia), so icterus in a horse that has stopped eating is not by itself evidence of liver disease. Dogs, especially males, have a low renal threshold for conjugated bilirubin, so small amounts of bilirubin in concentrated canine urine can be normal; in cats any bilirubinuria is abnormal.
Serum bile acids are the most practical function test in dogs and cats. Paired fasting and two-hour post-prandial samples assess hepatic uptake and the portal circulation; post-prandial values are usually more sensitive. Bile acids are markedly increased with congenital or acquired portosystemic shunts and with reduced functional mass. Horses have no gallbladder, so a single random sample is used. Bile acids add nothing in an animal already icteric from a hepatic or post-hepatic cause, because the same uptake failure that causes icterus raises bile acids.
Blood ammonia rises when the liver cannot convert ammonia to urea, for example with shunts or end-stage liver failure. Collect it into a chilled heparin tube, keep it on ice and analyse within about 30 minutes, because ammonia is generated in the tube. Other markers of reduced function are low urea, low albumin, low glucose, low cholesterol (in shunts) and prolonged clotting times from reduced factor synthesis. A classic portosystemic shunt profile in a young dog includes microcytosis, low urea, low albumin, ammonium urate crystals in urine and high bile acids.
Urea (reported in the United States as blood urea nitrogen, BUN) and creatinine are cleared mainly by glomerular filtration, so they rise when the glomerular filtration rate (GFR) falls. Both are insensitive: creatinine typically rises above the reference interval only after about three-quarters of nephron function has been lost. Urea also rises with a high-protein meal, gastrointestinal haemorrhage and catabolism, and falls with liver failure, low-protein diets and polyuria. In ruminants urea is recycled to the rumen, so creatinine is the more reliable renal marker. Creatinine depends on muscle mass: thin or cachectic animals can have a reduced GFR with a creatinine still within the interval, and well-muscled breeds such as greyhounds normally have higher values.
Symmetric dimethylarginine (SDMA) is filtered by the glomerulus and is less affected by muscle mass. It rises earlier, at roughly 25-40% loss of function, and is used with creatinine in dogs and cats. Unit conversions recur in examinations: urea in mmol/L × 2.8 = BUN in mg/dL (BUN mg/dL × 0.357 = urea mmol/L); creatinine in mg/dL × 88.4 = µmol/L.
Azotaemia means increased urea and/or creatinine. Classify it with a urine specific gravity (USG) taken before fluid therapy. Pre-renal azotaemia, from reduced renal perfusion (dehydration, shock, heart failure), is accompanied by concentrated urine. Renal azotaemia is accompanied by inadequately concentrated urine, often isosthenuric (1.008-1.012, the same as plasma filtrate). Commonly cited adequate concentration is above about 1.030 in dogs, 1.035 in cats and 1.025 in horses and cattle; texts vary slightly. Post-renal azotaemia arises from urethral obstruction or a ruptured bladder; USG is variable and the history, palpation and imaging usually reveal it.
Traps: any condition that impairs concentrating ability without renal failure (hypoadrenocorticism with medullary washout, diuretics, intravenous fluids, glucocorticoids, hypercalcaemia, pyometra with endotoxin, liver failure) produces azotaemia with dilute urine that mimics renal azotaemia. Cats sometimes retain partial concentrating ability despite chronic kidney disease. In uroabdomen, abdominal fluid creatinine above about twice serum creatinine in dogs strongly supports the diagnosis; potassium ratios are also used. Foals with ruptured bladders develop hyponatraemia, hypochloraemia, hyperkalaemia and azotaemia.
Fractional excretion of an electrolyte compares its clearance to that of creatinine: FE(x) % = (urine x ÷ plasma x) ÷ (urine creatinine ÷ plasma creatinine) × 100. In healthy animals fractional excretion of sodium is low (generally below 1%); it rises in tubular injury. It is used mainly in horses and ruminants.
The International Renal Interest Society (IRIS) stages canine and feline chronic kidney disease, once diagnosed in a stable, hydrated patient, by fasting creatinine and SDMA, and substages by proteinuria (urine protein to creatinine ratio, UPC) and systolic blood pressure. Creatinine cut-offs used by IRIS (check the current guideline, which is periodically revised): dogs, stage 1 below 125 µmol/L (1.4 mg/dL), stage 2 125-250 µmol/L (1.4-2.8 mg/dL), stage 3 251-440 µmol/L (2.9-5.0 mg/dL), stage 4 above 440 µmol/L (5.0 mg/dL); cats, stage 1 below 140 µmol/L (1.6 mg/dL), stage 2 140-250 µmol/L (1.6-2.8 mg/dL), and stages 3 and 4 as for dogs. UPC substaging: non-proteinuric below 0.2, borderline 0.2-0.5 in dogs and 0.2-0.4 in cats, proteinuric above those. Persistent renal proteinuria is confirmed only after excluding pre-renal causes (haemoglobin, myoglobin) and post-renal causes (urinary tract inflammation or haemorrhage, which shows in the sediment).
Other typical findings in chronic kidney disease are hyperphosphataemia, metabolic acidosis, non-regenerative anaemia, hypokalaemia in cats and systemic hypertension. In oliguric acute kidney injury hyperkalaemia is the immediate threat.
Cystocentesis gives a sample free of urethral and genital contamination and is the method of choice for bacterial culture in dogs and cats; it may add a few red cells. It is avoided when bladder neoplasia is suspected, because of the risk of seeding tumour cells along the needle tract. Catheterisation and free-catch (voided) samples are acceptable for the chemical examination but contaminated for culture. Free-catch mid-stream samples are the norm in large animals; in cows, stimulating the perineum below the vulva often induces urination. Analyse within about 30-60 minutes; otherwise refrigerate and bring back to room temperature before testing. Standing urine at room temperature allows bacterial overgrowth, a rise in pH, casts and cells to degrade, and crystals to form or dissolve; refrigeration can itself precipitate crystals, so crystals in a stored sample may be artefacts. Record the collection method on the request form, because the interpretation of red cells, protein and bacterial counts depends on it, and use sterile containers whenever culture may be needed.
USG must be measured with a refractometer; the dipstick specific gravity pad was designed for human urine and is unreliable in animals. Carnivore urine is usually acidic and herbivore urine alkaline; horse urine is normally turbid and viscous from calcium carbonate crystals and mucus.
On the dipstick, the protein pad responds mainly to albumin and is falsely positive in highly alkaline urine; interpret protein against USG and the sediment, and quantify persistent proteinuria with the UPC. Glucosuria appears when blood glucose exceeds the renal threshold (approximately 10-12 mmol/L in dogs and higher, around 14-16 mmol/L, in cats; values differ between sources) or with proximal tubular disease, as in Fanconi syndrome in Basenjis and some toxicoses. The ketone pad uses nitroprusside, which detects acetoacetate and acetone but not beta-hydroxybutyrate, so it can underestimate ketosis. The blood pad reacts with intact red cells, haemoglobin and myoglobin. The leukocyte pad gives false positives in cats and misses many cases of pyuria in dogs; the nitrite pad is unreliable in animals; and the urobilinogen pad has little value. Use the sediment instead.
To separate red urine causes, centrifuge: a red sediment with clear supernatant is haematuria. If the supernatant stays red, look at the plasma: pink plasma with red urine indicates haemoglobinuria (intravascular haemolysis), while clear plasma with red-brown urine suggests myoglobinuria, because myoglobin is small, filtered quickly and does not discolour plasma; a high CK supports it.
Examine the sediment of a standard volume of centrifuged urine under low and high power. Up to a few red and white cells per high-power field are normal in most texts (more after cystocentesis or catheterisation). Pyuria with bacteria suggests urinary tract infection but must be confirmed by culture; bacteria in an unstained sediment are easily confused with amorphous debris.
Casts are cylindrical protein moulds of the tubules. A few hyaline or granular casts can be seen in health; increased granular or cellular casts indicate active tubular injury, which may precede azotaemia. Crystals have clinical value only in context. Struvite (magnesium ammonium phosphate, coffin-lid shape) is common in alkaline urine of dogs and cats. Calcium oxalate dihydrate (envelope shape) is often incidental, but calcium oxalate monohydrate (picket-fence or dumb-bell forms) in an acutely ill animal suggests ethylene glycol toxicosis. Ammonium urate crystals suggest portosystemic shunt or liver failure, except in Dalmatians and some English Bulldogs, which have a genetic defect in urate handling. Cystine crystals indicate cystinuria, an inherited tubular defect. Bilirubin crystals can occur in concentrated canine urine. In ruminants, urolithiasis (often silica or phosphate-based, depending on diet) mainly obstructs castrated males.
Cytology examines cells rather than tissue architecture. It is quick, cheap and minimally invasive, and is often enough to separate inflammation from neoplasia, or to identify organisms. Fine-needle sampling is done either without suction (the needle is moved in and out several times and cells enter by capillary action; this reduces blood contamination) or with suction using an attached syringe, releasing the suction before the needle is withdrawn. Other methods are impression smears of cut or ulcerated surfaces, scrapings, swabs (ear, vagina) and fluid collection.
Expel the material onto a slide and spread it gently, for example with the squash technique, because rough spreading ruptures cells. Air-dry quickly. Most practices stain with a rapid Romanowsky-type stain; note that the granules of some mast cells stain poorly with some aqueous rapid stains, which can lead to underdiagnosis. Never send or store unstained slides near formalin: formalin fumes alter staining irreversibly. Send slides separately from biopsy containers, with a full history.
Begin at low power to assess cellularity, smear quality and areas to examine, then classify the population. Inflammation is classified by the dominant cell: neutrophilic (suppurative), macrophagic (granulomatous), mixed, or eosinophilic. Degenerate neutrophils, with swollen pale nuclei, suggest a toxic environment and bacterial infection; look for intracellular bacteria, which confirm sepsis rather than contamination. Fungi, protozoa such as Leishmania, and mycobacteria (which do not stain and appear as negative-image rods) may be seen.
Neoplastic populations are classified by cell shape and arrangement. Epithelial cells exfoliate readily in cohesive clusters or sheets. Mesenchymal cells exfoliate poorly and appear singly as spindle or stellate cells. Round (discrete) cells appear individually and include lymphoma, mast cell tumour, histiocytoma, plasma cell tumour and transmissible venereal tumour; melanocytic tumours are sometimes treated as a fourth group.
Criteria of malignancy are mostly nuclear: anisokaryosis, a high and variable nuclear to cytoplasmic ratio, large or multiple or angular nucleoli, nuclear moulding, multinucleation, coarse chromatin and abnormal mitoses. A common teaching rule is that several (often three or more) nuclear criteria should be present in many cells before calling a population malignant, and that inflammation can induce dysplasia that mimics malignancy. In a lymph node aspirate, a population dominated by large lymphocytes (classically more than half) supports high-grade lymphoma, whereas a mix of small lymphocytes with some large cells and plasma cells suggests a reactive node. Cytology rarely grades tumours reliably, and histopathology is needed for grade and margins.
Body cavity effusions are classified by total protein and nucleated cell count into transudates (low protein, low cellularity), modified transudates and exudates (high protein, high cellularity); the exact cut-offs differ between textbooks, and newer schemes classify by mechanism instead. Specific tests then refine the cause. A septic peritoneal exudate shows degenerate neutrophils with intracellular bacteria; in dogs, an effusion glucose well below blood glucose (a difference above about 1.1 mmol/L, 20 mg/dL) supports septic peritonitis. A chylous effusion has a triglyceride concentration higher than serum. A haemorrhagic effusion that has been present for some time shows erythrophagocytosis and lacks platelets. Uroabdomen is confirmed by the creatinine ratio. Effusion from feline infectious peritonitis is typically yellow, viscous, high in protein and globulin with a low albumin to globulin ratio, and contains non-degenerate neutrophils and macrophages; it supports but does not prove the diagnosis.
Synovial fluid is examined for viscosity, cell count and type: neutrophilic inflammation suggests infection or immune-mediated polyarthritis; mononuclear-dominated fluid suggests degenerative joint disease. Cerebrospinal fluid needs analysis within about an hour because cells degrade rapidly in its low-protein environment.
A culture result is only as good as the sample. Sample before antimicrobials are given, or note the treatment if they have been. Sample the lesion itself, aseptically, avoiding the surface flora: tissue or fluid is better than a swab, and a swab must be in transport medium (such as Amies or Stuart medium) to stop it drying. Anaerobic culture needs tissue or fluid in an anaerobic transport container with air excluded. Refrigerate most bacterial samples at about 4 °C and deliver them within 24-48 hours; specific organisms and laboratories have their own requirements, so ask the laboratory. Samples for virus isolation go in viral transport medium, chilled; samples for polymerase chain reaction (PCR) avoid formalin, which degrades nucleic acid, and must avoid cross-contamination between animals.
Urine for culture is collected by cystocentesis in dogs and cats and cultured quantitatively. Growth of about 10^3 colony-forming units (CFU) per mL or more from a cystocentesis sample is generally considered significant, while catheter and voided samples need progressively higher counts (commonly 10^4-10^5 CFU/mL) because of contamination; exact thresholds vary by source and species. For blood culture in suspected bacteraemia, collect several samples aseptically from separate sites, at intervals if possible, with an adequate volume of blood per bottle, because bacteraemia may be intermittent and low-grade.
Milk for mastitis culture should be collected aseptically: clean and dry the teats, discard the first strips, scrub each teat end with alcohol starting with the teats furthest away, then collect into a sterile tube starting with the nearest teats, so that the arm does not pass over a cleaned teat. Keep the sample cold; it can usually be frozen for common mastitis pathogens, but check with the laboratory. Faeces should be fresh, collected from the rectum where possible, and kept cool. For dermatophytes, plucked hairs from the lesion edge or a sterile toothbrush brushing are cultured. For abortion investigation, the fetus and placenta together, plus a maternal serum sample, give the best yield.
When anthrax is a possibility (sudden death in grazing ruminants with dark unclotted blood from body openings), do not open the carcass: exposure to air allows sporulation and long-term soil contamination. Collect blood from a peripheral vein, such as an ear vein, for a smear stained with polychrome methylene blue (the McFadyean reaction shows square-ended rods with a pink capsule), and notify the competent authority according to national law.
Diagnostic specimens travel in triple packaging: a leak-proof primary receptacle, a leak-proof secondary packaging with enough absorbent material for the whole contents, and a rigid outer packaging. Under United Nations transport rules, most diagnostic specimens are 'Biological Substance, Category B' (UN3373); infectious substances capable of causing permanent disability or fatal disease are Category A, UN2814 when they affect humans and UN2900 when they affect animals only. Postal and courier rules differ by country; follow the laboratory's instructions. For samples that may become legal evidence, keep a documented chain of custody.
Antimicrobial susceptibility tests report a minimum inhibitory concentration (MIC) or a zone diameter, interpreted as susceptible, intermediate or resistant using breakpoints. Breakpoints are specific to the bacterial species, the animal species, the drug and the site of infection, and veterinary-specific breakpoints exist only for some combinations (for example, those published by the Clinical and Laboratory Standards Institute); where a human breakpoint is used instead, interpretation is less certain. A culture of a contaminant, or of commensal flora from a poorly taken swab, can lead to unnecessary or wrong antimicrobial use.
Endocrine tests are most reliable when the clinical suspicion is already reasonably high. Many hormones fluctuate, are affected by non-endocrine illness and drugs, and have overlapping values between health and disease. For that reason endocrine diagnosis often uses dynamic tests, in which the gland is stimulated or suppressed and the response measured, and screening tests that are good at ruling a disease out before a more specific test rules it in. Sample handling is critical: some hormones, such as ACTH, degrade quickly and need chilled EDTA samples in plastic tubes, separated promptly and kept cold or frozen. Use the reference intervals and cut-offs of the laboratory that performs the assay, because immunoassays are species- and method-dependent. Record the time of sampling and any drug treatment on the request form. Finally, remember that treating a laboratory number rather than a patient is a common error in endocrinology: a low thyroxine in a sick dog is usually a consequence of illness, not its cause.
In dogs, hypothyroidism is over-diagnosed. Total thyroxine (T4) is a good screening test because a normal value makes hypothyroidism unlikely, but it is lowered by non-thyroidal illness (the euthyroid sick syndrome) and by drugs such as glucocorticoids, phenobarbital and sulfonamides, and sighthounds normally have lower values. Free T4 measured by equilibrium dialysis is less affected by illness, and combining it with canine thyroid-stimulating hormone (TSH) improves specificity. In cats, hyperthyroidism is diagnosed by a high total T4; in early disease or with concurrent illness, total T4 may be within the upper reference interval, and repeat testing or free T4 helps, remembering that free T4 alone can be falsely high in sick euthyroid cats.
For hyperadrenocorticism (Cushing's syndrome) in dogs, the low-dose dexamethasone suppression test is more sensitive than the ACTH stimulation test, while the ACTH stimulation test is more specific and is required to monitor treatment and to detect iatrogenic disease. The urine cortisol to creatinine ratio is highly sensitive but poorly specific, so a normal result helps exclude the disease, and a high result does not confirm it; collect the urine at home to avoid stress. For hypoadrenocorticism (Addison's disease), the ACTH stimulation test is the confirmatory test: a low cortisol both before and after ACTH confirms adrenal insufficiency. A basal cortisol above about 55 nmol/L (2 µg/dL) makes hypoadrenocorticism very unlikely, while a lower value needs an ACTH stimulation test.
Diabetes mellitus is confirmed by persistent fasting hyperglycaemia with glucosuria, supported by a high fructosamine. Insulinoma is suggested by an inappropriately normal or high insulin at the time of hypoglycaemia.
In horses, pituitary pars intermedia dysfunction (PPID) is tested by basal plasma ACTH, which must be interpreted against season-specific reference intervals because healthy horses show a natural autumn rise, or by the thyrotropin-releasing hormone stimulation test. Equine metabolic syndrome is assessed by insulin dysregulation tests such as basal insulin and the oral sugar test; follow current consensus guidance, as cut-offs depend on the assay.
Serum progesterone is used to time ovulation and breeding in bitches and to confirm luteal activity in other species. Results from in-clinic semi-quantitative kits differ from laboratory assays, so serial measurements on the same method are interpreted as a trend. In all these tests the same discipline applies: choose the test for the question, handle the sample as the assay requires, and use the cut-offs of the laboratory that ran it.
A reference interval is usually the central 95% of values from a healthy reference population, between the 2.5th and 97.5th percentiles. By definition, 5% of healthy animals fall outside it. If several independent tests are run, the chance that a healthy animal has at least one result outside the interval grows: with n independent tests it is 1 - 0.95^n, about 64% for 20 tests. Results just outside the interval in a well animal are therefore expected, and repeat testing or clinical correlation is appropriate. A value inside the interval is not proof of health, either: an animal's normal may sit at one end of the interval, so a change within the interval can matter. For analytes with high individuality, comparison with the same animal's earlier results is more sensitive than comparison with the population interval; the reference change value expresses how large a change must be to exceed analytical and biological variation.
Guidelines from the American Society for Veterinary Clinical Pathology (ASVCP) recommend at least 120 reference individuals for nonparametric intervals, with robust methods allowing smaller samples, and partitioning by species, and where needed by breed, age, sex and physiological state (for example foals, puppies, lactating cows), with methods matched to the analyser. Laboratories may transfer and validate an interval from another source with a smaller sample. Reference intervals are species-, breed-, age-, method- and laboratory-specific; always read a result against the interval supplied by the laboratory or analyser that produced it.
Sensitivity is the proportion of diseased animals that test positive; specificity is the proportion of non-diseased animals that test negative. Both are properties of the test at a given cut-off. Predictive values depend also on prevalence: positive predictive value (PPV) is true positives divided by all positives; negative predictive value (NPV) is true negatives divided by all negatives. When prevalence is low, even a good test yields many false positives, so PPV falls.
Worked example: a test with 90% sensitivity and 95% specificity is used on 1,000 animals with a prevalence of 10%. There are 100 diseased animals: 90 test positive (true positives) and 10 negative (false negatives). There are 900 healthy animals: 855 test negative (true negatives) and 45 positive (false positives). PPV = 90 ÷ 135 = 66.7%. NPV = 855 ÷ 865 = 98.8%. Use the same test where prevalence is 1%: 9 true positives and 49.5 false positives per 1,000, so PPV = 9 ÷ 58.5 = 15.4%.
The positive likelihood ratio is sensitivity ÷ (1 - specificity); here 0.90 ÷ 0.05 = 18. Post-test odds = pre-test odds × likelihood ratio. With a pre-test probability of 20% (odds 0.25), a positive result gives post-test odds of 4.5, a probability of 4.5 ÷ 5.5 = 81.8%. Moving a cut-off to raise sensitivity lowers specificity and vice versa; a receiver operating characteristic curve shows this trade-off. Two tests used in parallel (positive if either is positive) increase sensitivity: 1 - (1 - Se1)(1 - Se2). Used in series (positive only if both are positive) they increase specificity: 1 - (1 - Sp1)(1 - Sp2). A highly sensitive test that is negative helps rule a disease out; a highly specific test that is positive helps rule it in.
Point-of-care (in-clinic) testing includes haematology analysers (impedance or laser flow cytometry), dry-slide or rotor chemistry analysers, blood gas and electrolyte analysers, glucose and ketone meters, refractometers, microhaematocrit centrifuges and lateral-flow immunoassays (for example, feline leukaemia virus antigen and feline immunodeficiency virus antibody, canine heartworm antigen, canine parvovirus antigen, pancreatic lipase). Advantages are speed and immediate treatment decisions. Limitations are the lack of a trained technologist, less rigorous quality control, analyser-specific reference intervals, and species limitations: impedance counters may not separate feline platelets from red cells, and portable glucose meters designed for human blood can read differently from laboratory methods in animals, so use meters validated for the species. Always pair an analyser CBC with a smear review. Lateral-flow test results must be interpreted against prevalence and vaccination history; for example, feline immunodeficiency virus antibody tests can be positive in kittens through maternal antibody and in vaccinated cats where vaccines have been used.
Internal quality control means running control materials of known value, usually at two concentrations, each day or each run, and plotting them on a Levey-Jennings chart (mean ± 1, 2 and 3 standard deviations). Westgard rules decide whether to accept a run: a single control beyond 2 SD is a warning (1-2s); beyond 3 SD (1-3s) or two consecutive controls beyond the same 2 SD limit (2-2s) indicates rejection; a range of more than 4 SD between two controls (R-4s) suggests random error; four consecutive beyond 1 SD on one side (4-1s) or ten consecutive on one side of the mean (10x) indicate systematic error (bias or drift).
Imprecision is the coefficient of variation, CV = SD ÷ mean × 100. Bias is the difference between the analyser mean and a reference or target value. The ASVCP publishes total allowable error (TEa) goals for many analytes. Performance can be summarised by the sigma metric, (TEa - |bias|) ÷ CV, all in percent; a higher sigma means fewer rules and controls are needed, and a value below about 3 indicates poor performance. External quality assessment (proficiency testing) compares results with other laboratories on the same unknown samples and detects bias that internal controls cannot. Calibrate according to the manufacturer, keep maintenance records, and train staff; most errors still arise before the sample reaches the analyser.
Learn to read every result in a fixed order: is the sample valid (haemolysis, lipaemia, delay, wrong tube), is the species interval right, then which pattern does the result fit. Practise calculating red cell indices, absolute counts, anion gap, fractional excretion, UPC and predictive values by hand until they are automatic, because examinations test them and analysers do not explain them. Make and stain your own blood smears and urine sediments; most of the important findings (agglutination, Heinz bodies, platelet clumps, toxic change, crystals, organisms) are missed by machines and found only by someone who looks.
Each day you decide which tests to run and whether to act on them. Choose the tube and handling for the test; send a smear with every CBC; take a USG before giving fluids so that azotaemia can be classified; repeat a mildly abnormal result in a well animal rather than chase it; use species-specific PLI, fructosamine and dynamic endocrine tests correctly. Run daily controls on in-clinic analysers and keep a Levey-Jennings chart. Know when cytology is enough and when to submit histopathology, and know the packaging and notification rules for samples from notifiable disease suspects.
A senior clinician or practice owner sets the laboratory system: which analysers to buy, what QC and external quality assessment to run, how to validate or transfer reference intervals for the patient population, and which tests the practice sends out. Senior work also includes interpreting complex cases where results conflict, mentoring juniors away from over-testing and over-treating minor deviations, auditing diagnostic accuracy, and in herd practice designing metabolic profiles and sampling schemes (for example BHB and NEFA testing of transition cows) that give herd-level answers with a defensible number of animals.
A researcher in clinical pathology designs and validates assays and reference intervals following ASVCP and CLSI guidance, studies biological variation and reference change values for veterinary analytes, evaluates diagnostic accuracy with appropriate gold standards, ROC analysis and latent class models when no gold standard exists, and investigates new biomarkers (for example SDMA, acute phase proteins, cardiac troponins) for sensitivity, specificity and clinical utility. They also contribute to total allowable error goals and to harmonising results across laboratories and countries.
This manual gives principles and standard teaching values, not a full set of reference intervals; those depend on species, analyser and laboratory and are deliberately not tabulated. Coverage of exotic, avian, reptile, fish and laboratory-animal clinical pathology is limited to a few handling points. Blood gas interpretation is introduced only through the anion gap and the abomasal alkalosis pattern; full acid-base physiology (Henderson-Hasselbalch and strong ion approaches) belongs in physiology and emergency care. Bone marrow cytology, flow cytometry, PCR for clonality, immunocytochemistry, therapeutic drug monitoring and detailed equine metabolic syndrome cut-offs are not covered. Molecular microbiology methods are mentioned only as sampling requirements. Urine culture thresholds, effusion classification cut-offs and ketosis thresholds are given as ranges because sources differ.
Original practice questions written to the syllabus and pattern of the exams named in each question's examTags. They are not past papers and do not reproduce any board's questions.
These notes are free to read here, with no account.
What the Laisha app adds:
Free for ten days, then a subscription. We would rather write that here than have you find it later.