Chapter 1
ANALYTICAL AND DIAGNOSTIC CONCEPTS: GETTING IT RIGHT

Kristen R. Friedrichs, Michael A. Scott, and Steven L. Stockham

• Clinical Pathology
• Samples
  1. I. Whole Blood, Plasma, and Serum
  2. II. Urine Samples
  3. III. Other Body Fluid Samples
  4. IV. Nonfluid Samples for Cytologic or Histologic Evaluation
• Common Types of Clinical Pathology Laboratory Assays
• Significant Digits (Significant Figures)
• Units
• Analytical and Biological Variation
• Reference Intervals
• Quality of Laboratory Results
  1. I. Sources of Error
  2. II. Analytical Properties of Assays
  3. III. Quality Assurance
• Variations in Results Across Assays and Laboratories
• Which Laboratory Should One Use?
• Evaluating and Validating Laboratory Methods
  1. I. Reasons for Evaluating Laboratory Methods
  2. II. What are Sources of Analytical Error?
  3. III. Acceptable Analytical Performance
  4. IV. Analytical Validation Process
  5. V. Comparison of Assays
• Diagnostic Properties and Predictive Value of Laboratory Assays
• Receiver Operating Characteristic (ROC) Curves
• Herd-based Testing for Cattle

Table 1.1. Abbreviations and symbols in this chapter

[fT4]ed Free thyroxine concentration by equilibrium dialysis [x] x concentration (x = analyte) BHB ß-Hydroxybutyrate Ca2+ Calcium CBC Complete blood count CLSI Clinical Laboratory Standards Institute CV Coefficient of variation CVA Analytical variation CVG Inter-individual (between-subject) biological variation CVI Intra-individual (within-subject) biological variation EDTA Ethylenediaminetetraacetic acid fCa2+ Free ionized calcium FN False negative FP False positive fT4 Free thyroxine II Index of individuality K2EDTA Dipotassium salt of ethylenediaminetetraacetic acid K3EDTA Tripotassium salt of ethylenediaminetetraacetic acid LRL Lower reference limit Na2EDTA Disodium salt of ethylenediaminetetraacetic acid NIST National Institute of Standards and Technology NPV Negative predictive value (predictive value of a negative test) PPV Positive predictive value (predictive value of a positive test) QA Quality assurance QC Quality control QCM Quality control material RBC Red blood cell, erythrocyte RCV Reference change value RI Reference interval ROC Receiver operating characteristic SD Standard deviation SI Système international d'unités (International System of Units) tCa2+ Total calcium TEa Total error allowable (allowable total error) TEo Total error observed (observed total error) TN True negative TP True positive tT4 Total thyroxine U International unit URL Upper reference limit USD Usual standard deviation WBC White blood cell, leukocyte WRI Within reference interval

Note: See Table 1.4 for abbreviations of units of measurement and the figure legends for abbreviations unique to figures.

CLINICAL PATHOLOGY

1. What is clinical pathology?
   1. Definitions
      1. Pathology is the study of disease. It deals with understanding the structural and functional changes causing or caused by disease.
      2. Clinical pathology, a subtype of pathology, is the study of disease in the clinical environment by use of laboratory assays. It deals with the use of laboratory testing (clinical chemistry, hematology, .) for the diagnosis and management of disease.
   2. As currently certified by the American College of Veterinary Pathologists, veterinary clinical pathologists are specialists in the disciplines of basic pathology, hematology (study of blood and bone marrow, including hemostasis), clinical chemistry (study of physiologic and biochemical reactions in blood or other body fluids), cytologic pathology (study of cells in tissues or body fluids), surgical pathology (study of disease via microscopic analysis of tissue samples obtained during surgery), and principles of laboratory medicine (equipment and methods, test properties, QA).
   3. Veterinary clinical pathologists, medical laboratory scientists (medical technologists, clinical laboratory scientists), medical laboratory technicians, and veterinary technicians often work in a clinical laboratory that limits its assay "menu" (offered tests) to hematologic assays including hemostasis tests, clinical chemical assays, urinalysis, and cytologic examinations. Other assays or diagnostic laboratory procedures may be offered but are often provided by other specific laboratories (e.g., microbiology, histopathology, and toxicology) that are commonly supervised by microbiologists, histopathologists, and toxicologists, respectively.
2. Laboratory tests should be used purposefully, judiciously, and along with other diagnostic procedures. Before laboratory tests are used to pursue a possible diagnosis, two diagnostic procedures are imperative: (i) obtain a complete history and (ii) perform a complete physical examination. With knowledge gained from these two basic procedures, a diagnostician can select appropriate diagnostic procedures to clarify or classify identified problems prior to developing or refining a differential diagnosis and pursuing further (or definitive) testing. Veterinarians frequently use laboratory assays in conjunction with other diagnostic methods to identify or classify pathologic states that develop in domestic mammals. Some body systems (e.g., integument, nervous, skeletal, and cardiovascular) are relatively easily evaluated via visual or imaging methods (physical examination, radiography, and ultrasonography), whereas other body systems (e.g., hemic, immune, urinary, and endocrine) are better evaluated by laboratory tests. Consideration of the following questions and associated material in this chapter and elsewhere1 will help optimize selection of cost-effective tests that benefit patient management:
   1. Why is the test being ordered? Is it needed? How will the test results influence patient management or outcome?
   2. How well does the test discriminate between health and disease? How well does it discriminate among various disorders?
   3. How are the test results interpreted? Are appropriate reference values available?
3. What are the major reasons for analyzing patient samples via laboratory procedures?
   1. To detect an unidentified pathologic state
   2. To define, classify, or confirm a pathophysiologic disorder or disease state
   3. To eliminate a possible cause of the animal's illness
   4. To assess changes in a pathologic state either due to natural progression of the disease or because of medical or surgical intervention

SAMPLES

1. Whole blood, plasma, and serum
   1. Most clinical laboratory assays are designed to detect or quantify substances or cells in blood samples; the substance or cell of interest is called the analyte while the quantity that is measured is called the measurand. Obtaining useful results for the analyte requires appropriate samples. Whenever there is doubt about the appropriate sample for a particular test at a particular laboratory, the laboratory should be contacted prior to sample collection. (See Quality of Laboratory Results, Section I.B.)
   2. Whole blood...