Small Animal Clinical Diagnosis by Laboratory Methods, 5th

Edition

Visit the link below to download the full version of this book:

https://medidownload.com/product/small-animal-clinical-diagnosis-by-laboratory-m
ethods-5th-edition/

Click Download Now

3251 Riverport Lane
St. Louis, Missouri 63043

SMALL ANIMAL CLINICAL DIAGNOSIS BY LABORATORY METHODS,

FIFTH EDITION ISBN: 978-1437706574
Copyright © 2012, 2004, 1999, 1994, 1989 by Saunders, an imprint of Elsevier Inc. All rights
reserved.

No part of this publication may be reproduced or transmitted in any form or by any means, electronic
or mechanical, including photocopying, recording, or any information storage and retrieval system,
without permission in writing from the publisher. Details on how to seek permission, further
information about the Publisher’s permissions policies and our arrangements with organizations such
as the Copyright Clearance Center and the Copyright Licensing Agency, can be found at our website:
www.elsevier.com/permissions.

This book and the individual contributions contained in it are protected under copyright by the
Publisher (other than as may be noted herein).

Notices

Knowledge and best practice in this field are constantly changing. As new research and experience
broaden our understanding, changes in research methods, professional practices, or medical
treatment may become necessary.
Practitioners and researchers must always rely on their own experience and knowledge in
evaluating and using any information, methods, compounds, or experiments described herein. In
using such information or methods they should be mindful of their own safety and the safety of
others, including parties for whom they have a professional responsibility.
With respect to any drug or pharmaceutical products identified, readers are advised to check
the most current information provided (i) on procedures featured or (ii) by the manufacturer of
each product to be administered, to verify the recommended dose or formula, the method and
duration of administration, and contraindications. It is the responsibility of practitioners, relying
on their own experience and knowledge of their patients, to make diagnoses, to determine dosages
and the best treatment for each individual patient, and to take all appropriate safety precautions.
To the fullest extent of the law, neither the Publisher nor the authors, contributors, or editors,
assume any liability for any injury and/or damage to persons or property as a matter of products
liability, negligence or otherwise, or from any use or operation of any methods, products,
instructions, or ideas contained in the material herein.

Previous editions copyrighted

ISBN: 978-1437706574

Vice President: Linda Duncan

Content Strategy Director: Penny Rudolph
Content Development Specialist: Brandi Graham
Publishing Services Manager: Julie Eddy
Project Manager: Jan Waters
Design: Teresa McBryan

Working together to grow

libraries in developing countries
Printed in the United States of America www.elsevier.com | www.bookaid.org | www.sabre.org

Last digit is the print number: 9 8 7 6 5 4 3 2 1

Contributors
Jeanne A. Barsanti, DVM, MS, DACVIM Mark C. Johnson, DVM, DACVP (Clinical
Josiah Meigs Distinguished Teaching Professor, Emeritas Pathology)
Small Animal Internist, Emeritas Clinical Associate Professor
Department of Small Animal Medicine and Surgery Department of Veterinary Pathobiology
The University of Georgia Teaching Hospital College of Veterinary Medicine & Biomedical Sciences
The University of Georgia Texas A&M University
Athens, Georgia College Station, Texas

Dawn Merton Boothe, DVM, MS, PhD, DACVIM, Ned F. Kuehn, DVM, MS, DACVIM
DACVP Chief of Internal Medicine Services
Professor Michigan Veterinary Specialists
Department of Physiology and Pharmacology Southfield, Michigan
College of Veterinary Medicine
Director, Clinical Pharmacology Laboratory Michael R. Lappin, DVM, PhD, DACVIM
Auburn University Professor
Auburn, Alabama Department of Veterinary Clinical Sciences
College of Veterinary Medicine and Biomedical Sciences
Sharon A. Center, DVM, DACVIM Colorado State University
Professor Fort Collins, Colorado
Department of Veterinary Clinical Sciences
College of Veterinary Medicine Gwendolyn J. Levine, DVM
Cornell University Veterinary Clinical Associate
Ithaca, New York Department of Pathology
College of Veterinary Medicine & Biomedical Sciences
Stephen P. DiBartola, DVM, DACVIM Texas A&M University
Associate Dean for Administration and Curriculum College Station, Texas
Professor of Medicine
Department of Veterinary Clinical Sciences Jonathan M. Levine, DVM, DACVIM (Neurology)
College of Veterinary Medicine Assistant Professor of Neurology/Neurosurgery
The Ohio State University Department of Small Animal Clinical Sciences
Columbus, Ohio College of Veterinary Medicine & Biomedical Sciences
Texas A&M University
Sonya G. Gordon, DVM, DVSc, DACVIM College Station, Texas
Associate Professor
Department of Small Animal Medicine and Surgery Richard W. Nelson, DVM, DACVIM
College of Veterinary Medicine & Biomedical Sciences Professor of Internal Medicine
Texas A&M University Department of Medicine and Epidemiology
College Station, Texas School of Veterinary Medicine
University of California
Paula Martin Imerman, BS, MS, PhD Davis, California
Clinician and Associate Scientist
Department of Veterinary Diagnostic and Production Gary Osweiler, DVM, MS, PhD, DABVT
Animal Medicine Professor
College of Veterinary Medicine Department of Veterinary Diagnostic and Production
Iowa State University Animal Medicine
Ames, Iowa Veterinary Toxicologist
Veterinary Diagnostic Laboratory
Cheri A. Johnson, DVM, MS, DACVIM College of Veterinary Medicine
Professor Iowa State University
Department of Small Animal Clinical Sciences Ames, Iowa
College of Veterinary Medicine
Michigan State University
East Lansing, Michigan

v
vi Contributors

Rose E. Raskin, DVM, PhD, DACVP David C. Twedt, DVM, DACVIM

Professor of Veterinary Clinical Pathology Professor
Department of Comparative Pathobiology Department of Veterinary Clinical Sciences
School of Veterinary Medicine College of Veterinary Medicine and Biomedical Sciences
Purdue University Colorado State University
West Lafayette, Indiana Fort Collins, Colorado

Jennifer S. Thomas, DVM, PhD, DACVP Douglas J. Weiss, DVM, PhD

Associate Professor Professor Emeritus
Department of Pathobiology and Diagnostic Department of Veterinary Biomedical Sciences
Investigation College of Veterinary Medicine
College of Veterinary Medicine University of Minnesota
Michigan State University St. Paul, Minnesota
East Lansing, Michigan
Preface
Our intent has been to help veterinarians and veterinary students appropriately
select and accurately interpret laboratory tests in as simple, practical, and rapid
a manner as possible. The popularity of the first and then subsequent editions
has surprised and pleased us. Simple is good. With age one forgets the small
details and remembers only the major principles that get one through the day.

As with the first four editions, this fifth edition of Small Animal Clinical Diagnosis
by Laboratory Methods is intended to present organized methods of answering
commonly asked questions that reflect the problems frequently encountered
with interpretation of laboratory test results. Most authors have updated their
chapters as opposed to writing new ones with the idea that they could focus on
making them as timely as possible. New authors have been added in some areas.
We have attempted to make this text as current as possible, but this goal borders
on impossible in a profession that is continually advancing. References and
pathophysiology have been kept to a minimum because this text is designed to
be user-friendly to both the busy clinician in the middle of a hectic day, as well
as the student who is learning the art/science of problem solving.

Michael D. Willard and Harold Tvedten

vii
 Dedicated to my wife, Gladys,
without whom I could not have done this,
nor would I have cared to do it.

Michael D. Willard

Dedicated to the thousands of veterinary students

and veterinarians, hundreds of medical technicians,
and tens of clinical pathology residents with whom I have
shared tens of thousands of discussions on diagnosis
with laboratory methods over the last 40 years. It is for them
and with them that it is fun to come to work each day.
Speakers at veterinary meetings need not fear my questions,
for they are only questions I ask myself each day.

Harold Tvedten
General Laboratory Concepts

Harold Tvedten and Jennifer S. Thomas

1
TEST SELECTION AND ASKING THE context of the history, physical examination, and other
RIGHT QUESTION diagnostic findings in a patient. Unexpected results are
common and should stimulate the clinician to reevaluate
Veterinarians have many choices regarding laboratory the provisional diagnosis and look for additional diseases
testing. Important factors include availability of reference or consider possible causes for erroneous laboratory
laboratory testing, reliability and ease of in-clinic testing, results. Trends over several days are often more informa-
cost-effectiveness, accuracy, and turnaround time. One tive than test results on a single day. Typically, not all test
must determine what tests to perform in-clinic and what results that “should be” abnormal in a disease situation
tests to send out to a veterinary reference laboratory or to are abnormal in each affected patient.
a local human laboratory. Recent improvements in the
automation and ease of use of analyzers designed for NOTE: A reasonable level of skepticism about laboratory
in-clinic use are changing what is acceptable. Correct results should be maintained.
choices vary with the needs and patient population of
each veterinary clinic. No one answer fits all situations.
To get a specific and meaningful answer from labora- When interpreting laboratory tests, it is important to
tory testing, the diagnostician must ask a specific and keep in mind that reference intervals include the results
meaningful question and understand whether a particular expected in 95% of normal animals. Thus 5% of results
laboratory test is likely to yield a useful answer. As an in normal animals (i.e., 1 of 20) are expected to be
example, compare the likely outcome of asking the fol- outside the reference intervals. If a profile of 20 tests is
lowing questions: “Is the animal anemic?” “What is wrong performed, only 36% of normal animals would have all
with the animal?” A microhematocrit procedure (in addi- 20 results within the 95% confidence interval reference
tion to knowledge of the animal’s hydration status) will values. Diagnosticians must expect some false-positive
usually answer the first specific question, but a serum and false-negative test results. No tests are 100% sensitive
chemistry profile, complete blood count (CBC), urinaly- and 100% specific for a disease.
sis, and fecal examination may or may not answer the
second vague, nonspecific question. A clinician should NOTE: Only slightly more than one third of normal animals
ask, “What will a high, low, or normal test result specifi- are likely to have “normal” results in all tests of a 20-test
cally mean in terms of making a correct diagnosis, pro­ profile. The clinician should not over interpret small changes
viding accurate prognostic information, or choosing an from reference intervals.
appropriate therapeutic plan?” If the answer is meaningful
(i.e., it will change some action taken by the clinician), the Abnormal results in normal animals are often only
test is worth the cost. Normal laboratory results may elimi- slightly above or below the reference interval. The mag-
nate certain diseases (i.e., have high negative predictive nitude of a change helps determine one’s confidence
value [NPV]) and can be as valuable as abnormal results. that a disease is present. Large alterations usually allow
greater confidence that the animal is abnormal, because
NOTE: To choose the appropriate test that will provide they are less likely the result of statistical chance. With
a specific diagnostic answer, a very specific question must many tests, increasing magnitude of deviation from
be asked. normal also reflects a more severe disease and worsening
prognosis.
Laboratory methods vary in their ability to provide
SIMPLE STATISTICS AND PRACTICAL the same result when a sample is repeatedly analyzed (i.e.,
INTERPRETATIONS analytical precision). The coefficient of variation (CV) is
often used to indicate the precision of an assay. Assays
A reasonable level of skepticism about laboratory results with a low CV have a high degree of precision; small
should be maintained. Clinicians should not believe all changes in results can be attributed to changes in the
numbers. All laboratory data should be interpreted in the patient and not random variation in the assay itself.

1
2   SMALL ANIMAL CLINICAL DIAGNOSIS BY LABORATORY METHODS

Assays with a high CV have poorer precision; small

changes in results may be due to variation in the assay 1
and not indicative of disease in the patient. For example,
0,9
because of the great imprecision of a manual leukocyte,
platelet, or erythrocyte count, results can vary 10% to 20% C
0,8
only because of technique; therefore mild changes from U

Sensitivity (true positives)

one day to the next may reflect only imprecision in the 0,7
procedure rather than actual changes in the patient.
Evaluating populations of apparently healthy animals 0,6 U/C
with screening tests is much different from testing indi-
0,5
vidual sick animals. The predictive value of a test is
strongly affected by the prevalence of disease in a popula- 0,4
tion.3 For example, if a disease occurs in 1 of 1000 animals
and a test is 95% specific and sensitive for the disease, 0,3
what is the chance that an animal with a positive test result
actually has the disease (i.e., positive predictive value [PPV])? 0,2
Most students, residents, and clinicians answered this
question incorrectly; the average response was 56% with 0,1
a range of 0.095% to 99%. If the test is 95% sensitive, 0
95% of all animals with the disease should be detected. 0 0,1 0,2 0,3 0,4 0,5 0,6 0,7 0,8 0,9 1
Therefore the one animal in 1000 that has the disease
should be positive. If the test has a specificity of 95%, 1- Specificity (false positives)
then 5% of the 999 animals in 1000 that do not have the
disease, or about 50, will have a false-positive test result. FIGURE 1-1. A ROC curve is a way to show the effectiveness
of a test. Increased serum urea concentration, creatinine con-
The PPV (i.e., the number of true-positive tests/total
centration, and urea/creatinine ratios were compared in diag-
number of positive test results) of this test is only about nosis of 417 dogs with renal failure, 1463 normal dogs, and
2%, because only 1 of those 51 animals with a positive 2418 sick dogs without renal disease.2 The area under the ROC
test result will have the disease. There are mainly false- curves show that serum creatinine and urea concentrations
positive results to interpret and explain to the animal were very similar in diagnostic accuracy but the urea/creatinine
owners. ratio was obviously worse than either of them.
Screening tests with a high sensitivity are often useful
to rule out a disease. In the example above with a test that
has a sensitivity of 95%, 5% of animals with the disease
will not be detected and will be false negatives. Using the
above situation, if 1 in 1000 animals has the disease, then indicated by the diagnostic sensitivity of an assay) against
0.05 animals will have a false-negative result. If the speci- the false-positive rate (1 − the diagnostic specificity of an
ficity is 95%, then 949 of 999 animals in 1000 that do assay) calculated at various concentrations over the range
not have the disease are true negatives. Thus the NPV of the test’s results. A good test has a great increase in the
(number of true-negative tests/total number of negative true-positive rate along the y axis for a minimal increase
test results; 949/949.05) is greater than 99.9%. in the false-positive rate along the x axis. The 45-degree
line in Figure 1-1 would indicate an ineffective test, which
would have an equal increase in false positives and in true
NOTE: Evaluating test results from populations of appar- positives. Whether a positive result on such a test was a
ently healthy animals is much different from evaluating true positive or a false positive would be random chance,
results in individual sick animals. like tossing a coin. Figure 1-1 illustrates that serum creati-
nine and urea (measured as blood urea nitrogen [BUN])
are very good tests of renal failure in dogs and very similar
If a test is performed only when the disease is likely in effectiveness.2 The urea/creatinine ratio is noticeably
instead of screening all animals (including those with no worse than either creatinine or urea, as illustrated by
clinical signs) for a disease, then the frequency of diseased being closer to the 45-degree angle line (and having less
animals in the test population is higher. Testing for area under the curve).
disease in sick patients is exemplified by heartworm ROC curves are also useful in selecting upper and
testing. Consider an example in which a test for heart- lower decision thresholds that can be used to decide
worm disease is 99% sensitive and 90% specific and is when a diagnosis can be ruled in or ruled out. Note that
used in 100 outside dogs in a heartworm-endemic area.5 decision (or diagnostic) thresholds are different than ref-
If the incidence of disease is 50%, then one should iden- erence intervals. Animals with a test result below the
tify 49.5 of the 50 ill dogs and obtain 5 false-positive lower decision threshold limit are unlikely to have the
results in the 50 dogs without heartworms. Thus the PPV disease being tested for; animals with a test result above
in this situation is 49.5/54.5 or 91%. There are still false the higher decision threshold limit are likely to have the
positives to interpret, but greatly fewer. disease.
Receiver operating characteristic (ROC) curves (Figure Diagnostic thresholds for renal failure are suggested
1-1) are used to determine the effectiveness of a test in where the creatinine or urea ROC curves in Figure 1-1
diagnosis. ROC curves plot the true-positive rate (as rapidly change their upward angle and begin to turn and
 Chapter 1: General Laboratory Concepts   3

plateau to the right. Lower to the left along the curve is a dog versus calm dog), or changes in instrument software
higher concentration threshold with greater specificity may change the sensitivity of detection of reticulocytes.
and lower sensitivity. More to the upper right is a lower Mean values are used to detect trends that depart
threshold with greater sensitivity and lower specificity. At from the norm. For example, if a dog has a low carbon
the bend in the curve, the test has optimal sensitivity dioxide partial pressure (PCO2) (indicating a respiratory
(increase in true positives) with minimal loss of specific- alkalosis trend) and a low bicarbonate (HCO3) concen-
ity (increase in false positives). tration (indicating a metabolic acidosis trend), the diag-
nostician uses pH to indicate which is the disease change
and which is likely compensation. A pH that is within the
REFERENCE VALUES reference range can still be discriminating by how it devi-
ates from the mean for pH. For example, a low-normal
Reference values (i.e., reference ranges, reference intervals, pH indicates an acidifying tendency and that the disease
“normal” ranges) are used to determine if a test result process is more likely metabolic acidosis with respiratory
appears normal or abnormal. A laboratory result is mean- compensation, rather than respiratory alkalosis with met-
ingless without knowing what values normal animals in abolic compensation. Means are used with increases in
that situation should have. It is not unusual for a veteri- enzyme activity that should be reported as an x-fold
narian to request that a test be performed in a species for increase (e.g., a tenfold increase over the mean). It is more
which the laboratory has no reference values, nor is it common to use the fold increase over the upper reference
unusual to find that the laboratory has not validated the value, because mean values are often not available.
test for accuracy in disease diagnosis in species not com- Reference values are often suboptimal. New reference
monly tested. Reference intervals may be presented as a intervals should, theoretically, be established whenever a
range or a mean (or median) plus and minus 2 standard laboratory changes instruments, methods, or even types
deviations. Reference intervals should optimally also have of reagents. The expense is considerable and often pro-
95% confidence intervals around the upper and lower hibitory considering the number of species involved; the
values to help show that the limits can be “fuzzy.” Too variety of breeds; the effect of age, sex, and other factors;
often veterinarians use an upper or lower value as an exact and the number of “normal” animals for a reference
breakpoint between normal and abnormal. For example, population optimally needed for each category. An ideal
if a serum sodium reference interval is 146 to 156 mmol/L, reference population should include 120 individuals for
a common error is to consider 146 mmol/L normal but parametric and 200 individuals for nonparametric dis-
145 mmol/L as indicating hyponatremia despite the fact tributed values. A robust method for determining refer-
that imprecision in the method or rounding off of values ence intervals is recommended when only 20 to 40
may mean that these values are essentially the same. individuals are available.9
Another example is that less than 60,000 reticulocytes/µl Unfortunately, use of readily available animals for
is often incorrectly given as a breakpoint between regen- reference populations is often is found to be inappropri-
erative and nonregenerative anemia. The 60,000 should ate because of later findings of subclinical disease or
be considered an approximate, rule-of-thumb, mean ref- deviations from “the typical adult dog or cat” because of
erence value. factors such as breed, age, sex. Results from any reference
One uses the mean or range of reference values in population should be evaluated for animals with values
different situations. Upper and lower reference values are that deviate from the main group to see if those animals
best for individual patients without a previous evaluation. came from one kennel (e.g., breed-related deviations such
The best reference values are a patient’s own values (if as those in greyhounds, nutritional or toxic disorder in
available) before an illness, because individual animals the kennel population) or for any other explanation for
or members of special groups (e.g., sight hounds, puppies) why those animals should be removed from the reference
may have unique characteristics. When comparing groups population.
of animals (e.g., a research project), one should use a An alternative method to generate reference intervals
mean or median value for the groups for interpretation when new techniques, reagents, or instruments are added
of changes (e.g., packed cell volume [PCV] 45%) instead to laboratories is to perform at least 20 to 60 duplicate
of published reference values (e.g., PCV 37% to 54%). analyses with both the new and the previous or “stan-
Specific reference values should be used for different dard” procedure. Regression analysis is used to predict the
methods and instruments. One laboratory’s reference new procedure’s reference values from the previous refer-
values for canine reticulocytes for the ADVIA 120 instru- ence values, assuming the previous values were properly
ment (Siemens Healthcare Diagnostics) is 11,000 to established from an appropriate reference population. It
111,000/µl (see Appendix II). The XT-2000iV analyzer is very important to include a wide range of low and high
(Sysmex Corporation) reports higher numbers of reticulo- results in the group of duplicate samples.
cytes than the ADVIA 2120 and should have a different set Reference values in hematology or clinical chemistry
of canine reference values (19,400–150,100/µl). A current books and articles will vary from a clinic’s own instru-
problem with available automated reticulocyte results on ments and methods but are useful to identify factors that
most hematologic samples is that many nonanemic dogs typically cause deviations from “the typical adult dog or
appear to have a regenerative erythropoietic response cat” due to breed, age, sex, and the like. The number of
because they have reticulocyte counts higher than cur- tests analyzed in most clinics for some species (e.g., pet
rently available reference values. This may occur because, birds, wildlife, zoo animals) might be too low to justify
even with properly established reference values, there may establishing reference values. Literature values are often
be variations in how some samples were collected (excited used for many tests if a laboratory does not have its own.
4   SMALL ANIMAL CLINICAL DIAGNOSIS BY LABORATORY METHODS

One example is the International Species Inventory

BOX 1-1. COMMON CAUSES FOR
System (ISIS) Physiologic Data Reference Values for zoo
PREANALYTICAL LABORATORY ERRORS
animals. Many of the species are uncommon, and only a
few may be present in a state or country. The ISIS values
• Incomplete sample labeling
were derived from normal animals at 65 institutions so
• Improper venipuncture techniques
that a reasonably sized database could be established.
Selected or “groomed” hospital patient data may • Sample contamination when collecting via catheter
be used to reevaluate reference values for one or more • Wrong anticoagulant
parameters that come under question. For example, a • Delayed mixing of blood with anticoagulant
reagent company may change the formulation of reagents • Inadequate mixing of blood just before aspiration into
for a test (e.g., calcium) and suddenly many patients the instrument
appear to have abnormally high or low calcium concen- • Delayed removal of serum or plasma from cells
trations. Patient values are not from animals proven to be • Delayed sample analysis
normal but are a readily available source of recently • Improper sample storage
obtained, inexpensive, and locally produced data. These • Inadequate warming of refrigerated samples before
data represent the laboratory’s current patient popula- analysis
tion. In the above example, if values for calcium concen- • Inadequate patient preparation (e.g., not fasted)
tration in the laboratory’s current patient population • Interfering substances in sample
(minus patients having a disease affecting calcium) are
compared to previous reference values, then new refer-
ence values derived from patient results can be a tempo-
rary adjustment. One would expect to find a shift in the
quality control (QC) results that chronologically matches
obtaining new reagents. cause, artifacts may make it impossible to determine the
real concentration or activity of an analyte. Anytime one
spurious result is found in a panel of tests, all results
INTERNATIONAL SYSTEM OF UNITS should be closely evaluated to determine whether they
have also been affected.
The International System of Units (Système Internatio-
nale d’Unités [SI units]) has standardized the reporting
of data for improved comparison of results throughout
Preanalytical Errors
most the world, with the exception of the United States, When an artifact is suspected, it is useful to determine if
Brazil, and a few other countries. Units used for serum the spurious findings resulted from a preanalytical or
enzyme activity were particularly inconsistent in the past, analytical error.12 Preanalytical problems occur before the
when many enzyme procedures had results reported in laboratory analyzes the sample and are the most common
units named after the author of the procedure. Now cause for laboratory errors.1,11 Common types of preana-
enzyme activity is reported as international units per liter lytical errors are listed in Box 1-1. Most preanalytical
(IU/L) in the United States or ukat/L in many other coun- errors are the result of sample collection or handling
tries. Note that IU/L for enzyme activity is not an SI unit! problems that can be avoided. Whenever possible, new
The SI unit for enzyme activity is ukat/L. U.S. laboratories samples should be collected if these types of preanalytical
still use “traditional” units such as mg/dl. Appendix II errors are suspected. Treatment with drugs often causes
includes common conversion factors to convert a result artifacts in laboratory testing. For example, chloride con-
from one unit of measure (e.g., mg/dl) to another unit of centration usually cannot be accurately measured in
measure (e.g., mmol/L). Unfortunately, laboratories in patients receiving potassium bromide because the most
the same hospital may report results using different units commonly available assays cannot distinguish bromide
of measure, causing confusion for clinicians when inter- from chloride. Certain drugs are insoluble in urine (e.g.,
preting those results. sulfa drugs), causing crystalluria.
Some sources of preanalytical errors (e.g., hyperbili-
rubinemia, lipemia, or in vivo hemolysis) are physiologic
SOURCES OF LABORATORY ERROR or pathologic in the patient and may not be easily con-
trolled. The severity of effect of these interferants depends
Laboratory error is common and needs to be detected on the analyte measured, the species involved, and the
early to avoid misdiagnosis. Laboratories should be asked analytical method used.
to recheck results that do not make sense in the context
of the animal’s history, physical examination, or other
diagnostic findings, such as marked hyperkalemia or
Hemolysis
hypoglycemia in a clinically normal animal. A variety of Hemolysis is recognized by reddish discoloration of
artifacts may cause the measured concentration or activity plasma or serum. It can cause significant artifacts on a
of an analyte or multiple analytes in a panel to be falsely CBC, hemostasis profile, and chemistry panel. Hemolysis
increased or decreased. Spurious results make it difficult occasionally occurs in vivo; it more commonly results
to accurately interpret laboratory results; artifacts may from in vitro erythrocyte damage associated with improper
cause abnormal results in a healthy animal or mask sample handling or collection. Samples that are hemo-
abnormal results in a sick animal. Depending on the lyzed due to handling or collection errors are best
 Chapter 1: General Laboratory Concepts   5

discarded. In vitro hemolysis can be minimized with laboratories; however, ultracentrifugation is not available
the following: (1) use of sharp needles to collect blood; in most practices. When using clearing polymers, it is
(2) employment of proper venipuncture technique, important to first determine whether the procedure will
including clean entry into a vessel, limiting vessel stasis, falsely alter the concentration or activity of measured
and avoidance of excess negative pressure during blood analytes for the analyzer system used.
collection; (3) gentle mixing and handling of tubes
promptly after collection; (4) proper centrifugation tech-
niques to separate plasma or serum from cells; (5) prompt
Analytical Errors
removal of serum or plasma from cells; and (6) preven- Analytical problems occur during actual performance
tion of overheating or freezing. of a laboratory assay. Operator errors include improper
Hemolysis may falsely increase the measured serum sample aliquoting, incorrect analyzer use, improper
concentration of substances that are present in higher reagent handling, and unauthorized modification of a
concentrations in the erythrocyte cytoplasm than in procedure. Reagent problems include using out-of-date or
plasma (e.g., lactate dehydrogenase [LDH], aspartate ami- improperly stored reagents. Analyzer malfunctions may
notransferase [AST], creatine kinase [CK]).1,14 These result from improper maintenance, aging of instruments,
changes vary according to species or breed and are inde- or purchase of a lower quality instrument. In general,
pendent of the analyzer or methodology used. Hemolysis analytical errors are minimized by proper training of tech-
releases free hemoglobin that may interfere with spectro- nicians and strict adherence to a quality assurance
photometric assays that measure substances at wave- program, including documented standard operating pro-
lengths similar to the absorbance range of hemoglobin. cedure for testing protocols, equipment maintenance,
Finally, hemolysis releases erythrocyte contents that may and QC.
positively or negatively interfere with the chemical reac- If a laboratory error is suspected, the first step is
tions used to measure analytes. The errors associated with usually to repeat the analysis using the same sample. If
these last two hemolysis effects vary significantly, depend- the suspect results do not repeat, operator error or a
ing on the specific analyzer and methodology used. random error is likely. If the suspect results repeat on the
Therefore referral laboratories or analyzer manufacturers second analysis, a new blood sample should be collected
should provide information about the effects of hemoly- and the analysis repeated. If the suspect results do not
sis on their specific analytical system. Veterinarians should repeat on the new sample, then a preanalytical error is
not choose laboratories that report out results on hemo- likely. If the suspect results repeat, then either there is an
lyzed, lipemic, or icteric samples over laboratories that analytical problem or the results accurately reflect the
refuse to do so. Reporting out a result may not indicate patient’s status. Test reagents and analyzer function
that a laboratory has better methods unaffected by hemo- should be evaluated. If they are functioning properly,
lysis, lipemia, or icterus, but merely that they report ques- then the patient should be reassessed for an alternative
tionable results with a disclaimer at the bottom of the diagnosis. If laboratory error is still suspected, then the
report. sample should be sent to a referral laboratory for
analysis.
Lipemia
Lipemia causes serum or plasma to appear milky white
and turbid when triglyceride concentrations exceed 300 SAMPLE COLLECTION:
to 400 mg/dl. Lipemia alters the light-scattering property RIGHT AND WRONG
of blood,1 causing alterations of a variety of hematologic
and chemical results, depending on the methodology and Proper sample collection and handling techniques are
analyzer used. Fat in lipemic serum displaces an equal required to avoid preanalytical laboratory errors and
volume of the aqueous part of serum. Because analytes obtain reliable results. Tubes should be labeled to assure
such as sodium are dissolved in the water portion of that results are reported for the correct patient. Clean
serum, and not in the lipid portion, this may cause a false venipuncture is needed to avoid hemolysis, tissue con-
decrease in sodium (pseudohyponatremia) and other tamination, or inappropriate clotting. Samples should be
substances but not a physiologic hyponatremia, because collected from a blood vessel that can rapidly provide the
it is the concentration of electrolytes in the aqueous required amount of blood without causing vessel col-
plasma in contact with cell membranes that have a bio- lapse. If blood is collected from a catheter, it is important
logic effect. Whether this volume-related problem occurs to discard an appropriate amount of fluid from the cath-
depends on how an instrument aspirates and dilutes a eter to assure that the patient sample is not contaminated.
sample or on the method of analysis. Lipemia increases Collection tubes containing anticoagulant should be
erythrocyte fragility and often causes hemolysis. filled first and promptly mixed (see Chapter 5 for sample
Recent feeding is the most common cause of lipemia. collection for platelet counting and hemostasis testing).
Fasting 12 hours before blood collection can usually Improper filling (usually incomplete) of the tubes alters
prevent postprandial lipemia. If lipemic samples are the ratio of anticoagulant to blood and significantly
unavoidable (e.g., patient with a lipid metabolism disor- affects hemostasis testing and even some CBC, cytologic,
der), then the serum or plasma can be cleared by ultra- or chemical assays. Collection of blood into the wrong
centrifugation or by the addition of a polymer that binds anticoagulant may interfere with some chemical assays8
lipids for removal by centrifugation (e.g., LipoClear). (see Box 1-2 for recommendations for collection tubes).
Ultracentrifugation is the standard by which other lipid- Serum tubes with clot activator can damage cell morphol-
clearing methods are judged and is used by many referral ogy in cytologic samples. Delayed analysis of blood or
6   SMALL ANIMAL CLINICAL DIAGNOSIS BY LABORATORY METHODS

handling and submission and routine charges should be

BOX 1-2. SELECTION OF COLLECTION TUBES
directed to the medical technologist, clerk, or secretary
who answers them daily. The pathologist’s time should
No Anticoagulant: Red-Top Tube
Serum for chemistry testing be used for questions that are interpretive, diagnostic, or
of a policy nature.
Fluid for cytologic examination

Heparin: Green-Top Tube

Plasma for chemistry testing
PROFILES VERSUS INDIVIDUAL
TEST SELECTION
EDTA* (Sodium or Potassium): Purple-Top Tube
Binds divalent cations: ↓ Ca and Mg concentrations
Some criticize profiles for being a “shotgun approach” or
Preferred anticoagulant for CBC a nonspecific “fishing trip.” Profiles instead should be
Fluid for cytologic examination considered cost-effective screens for a large number of
Interferes with chemistry testing: methodology common problems. It usually costs less for most labora-
dependent tories to perform a standard profile of 20 chemistry tests
↑ Na or K concentration than to perform 3 or 4 individual tests that vary for each
individual patient. This is due in part to using the auto-
Sodium Citrate: Light Blue–Top Tube mation of the instruments to perform profiles. Also, labor
Binds divalent cations: ↓ Ca and Mg concentrations
costs required to enter patient information into the com-
Preferred anticoagulant for hemostasis testing puter system, label tubes, and prepare the instrument are
Interferes with chemistry testing: methodology the same if 20 tests are performed per sample or only 1
dependent to 3 tests. A profile of many tests is recommended for
those clinicians who have access to a larger laboratory not
Fluoride (Sodium): Gray-Top Tube
only for cost-effectiveness but for the reliability and accu-
Use for glucose or lactate determinations only
racy of labs using QC and experienced laboratory profes-
Inhibits glycolytic enzymes
sionals (see also later discussions on stat and point-of-care
testing).
*EDTA, Ethylenediaminetetraacetic acid; CBC, complete blood
count. Serum chemistry profiles have variable numbers
and types of tests included, although the “basic profile”
includes the same or similar tests in most laboratories.
A well-designed profile should, for a minimal fee and
minimal redundancy, have a high probability of detecting
the common diseases for the particular situation. A CBC
other body fluids causes loss of cells and may make it is actually a profile that includes many tests to screen for
impossible to identify cell types or evaluate morphology anemia, inflammatory disease, stress, thrombocytopenia,
on a smear. Whole blood or body fluids should be kept and various other problems. The urinalysis is a profile
refrigerated until analyzed. Direct smears of fresh blood of chemical, morphologic, microbiological, and physical
and fluid cytology samples should be made promptly tests designed not only to reveal hemorrhagic, inflamma-
following collection and should accompany tubes of tory, or functional deficits in the urogenital tract but also
blood or fluids. Smears should be air-dried and not be to detect systemic disorders such as diabetes mellitus,
refrigerated. Exposure to excessive heat or cold can cause hepatic disease, massive acute muscle injury, or intra­
lysis of cells, leakage of intracellular constituents, or loss vascular hemolysis. A hemostatic profile for evaluating
of temperature-sensitive analytes. Delayed separation of bleeding disorders is discussed in Chapter 5.
plasma or serum from blood cells causes leakage of intra- Individual test selection becomes important if the
cellular constituents and consumption of glucose.12 To method or instrument locally available requires that
avoid breakdown or loss of enzymes or other substances, each test be individually performed or if certain tests are
it is best to keep serum or plasma refrigerated or frozen not offered in an inexpensive profile. Some tests may
if it will not be analyzed promptly after collection. not be cost-effective enough to include in an initial
A key to effective sample collection is good commu- screening profile. Tests that infrequently have abnormal
nication with referral laboratories. Laboratory personnel results or that are expensive are requested only when
will explain how to submit samples properly. When an patient information indicates a problem that justifies
unsatisfactory sample is received by a laboratory or when their use. After a specific disease has been diagnosed
instructions or sample labeling are incomplete, someone (e.g., diabetes mellitus), only one test may be needed to
must either try to contact the busy veterinarian for further monitor the treatment process and may be most
instructions or arbitrarily decide what to do with the cost-effective.
specimen. This may delay results or may lead to unor-
dered assays being performed. It is important not to waste
the effort used in obtaining the sample or lose a diagnos- STAT TESTS
tic opportunity that may not be available later. Many
laboratories have prepared written information on sub- Very short turnaround time between sample collection
mission procedures pertaining to their laboratories on and the availability of results is important in some “stat”
their websites (see Appendix I, Listing of Selected Referral situations (e.g., designing fluid therapy for acid-base and
and Commercial Laboratories). Questions about sample electrolyte abnormalities). Instruments that can analyze
 Chapter 1: General Laboratory Concepts   7

plasma or heparinized whole blood save the time (e.g., consistent staining characteristics of blood smears, bac-
30 minutes) required for blood to fully clot before har- teria and fungi, and cytologic smears. New methylene
vesting serum. Unlike cost-efficient batches, stat tests are blue (NMB) is easy to use for urine sediments, cytology
usually analyzed individually. Stat tests require laboratory smears, and reticulocytes in blood (see stain and micro-
personnel to interrupt their efficient work routines to scope discussions in Chapter 16). Air-dried smears of
perform single tests. cytology, blood, or urine sediment should be sent to a
clinical pathologist whenever there is any doubt about
the diagnosis. Manual erythrocyte and leukocyte count-
SEND IT OUT OR DO IT YOURSELF? ing techniques are discussed in Chapter 2, and platelet
counting is discussed in Chapter 5. A refractometer is
In-Clinic Laboratory Testing needed to determine urine specific gravity, plasma total
Veterinarians must decide what testing to do “in clinic” protein concentration, and total protein concentration
and what to send out. Major advantages to in-clinic in various body fluids. With the refractometer, only a
testing are rapid turnaround time and potentially drop of urine is needed to determine specific gravity,
improved client satisfaction by eliminating the need to which is essential for evaluating renal function. Simi-
delay treatment or surgery until results are received from larly, only the volume of plasma in a microhematocrit
a referral laboratory. Other advantages include minimiz- tube is needed for an accurate estimate of plasma
ing preanalytical errors associated with transportation of protein.
samples, providing an additional source of revenue for Most laboratories need at least two good-quality cen-
the clinic, and increased incorporation of biochemical trifuges: a single-purpose centrifuge for microhematocrit
profiles into presurgical screens, geriatric screens, and tubes to ensure consistent PCV determinations and a
routine health maintenance examinations.13 Disadvan- basic centrifuge for separating serum or plasma. Addi-
tages to in-clinic testing include the costs required to tional centrifuges to consider include a high-speed micro-
purchase equipment and maintain adequate stock of centrifuge for small (Eppendorf-type) tubes that allows
reagents, the need to perform QC testing, and the need more effective separation of serum or plasma from small
for additional staff time to properly perform assays in the samples. A refrigerated centrifuge is needed for some
laboratory. (See later discussion of calculation of the cost labile substances. If the clinic has a blood bank, a blood
to perform a test based on test volume, etc.) A laboratory banking centrifuge is needed to separate fresh frozen
can be a revenue-producing unit, but this is seldom true plasma from concentrated red blood cells (RBCs) or
for low-volume, inefficient testing situations. platelet-rich plasma. A cytocentrifuge is needed for prepa-
The quality of in-clinic laboratories varies greatly. ration of cytologic smears with good morphology of cere-
Accurate and reliable laboratory results require invest- brospinal fluid (CSF) or bronchoalveolar lavage (BAL)
ment in experienced medical technologists or their equiv- samples. Serum or plasma may need to be stored in a
alent, and a willingness to provide them adequate time refrigerator (4°C) or freezer (−20°C) before it is mailed
to set up and maintain a system of good laboratory to a referral laboratory. Freezers included in refrigerators
testing. Many clinics have nonlaboratory personnel per- and many home freezers may not maintain a consistent
forming their testing who lack the experience and training temperature below −20°C.
typical for technical staff in referral laboratories. Techni-
cian time must be allotted for training, inventory and
maintenance of reagents, equipment maintenance, trou-
Veterinary Referral Laboratories
bleshooting problems, and performing repeated measure- Large referral laboratories offer more cost-effective testing
ment on samples with suspected artifacts or results that (large profiles usually cost less than two to four individual
are questionable. tests), top-end instruments, more experienced personnel
The clinic is responsible for assuring the laboratory’s specializing in laboratory testing, better quality assurance,
results are reliable. One person should be designated to better detection of laboratory errors, and a wide range of
be in charge of a quality assurance program to ensure that tests. Many veterinarians purchase in-office analyzers only
results are accurate, repeatable, and properly reported. QC to find that, without frequent calibration and consistent
testing must be performed, recorded, and monitored to use of controls, the results are untrustworthy. Over the
prove validity of the answers sold to the client. Practices years, different instrument manufacturers have promoted
with low test volume should analyze control samples various in-clinic systems that disappeared (along with
with each patient’s sample and for each type of test per- reagents) a few years later.
formed. Because this is a significant expense in time and The larger, more sophisticated instruments not only
money, the frequency and extent of QC testing is often analyze tests more quickly and cheaply but also better
suboptimal and trust in the reliability of test results fades. detect sample and laboratory errors. Small, simplified
Standard operating procedures should be in place to instruments often report only numbers, with limited
assure that testing is properly and consistently performed additional information to detect errors. In contrast, larger,
no matter who performs the test. automated hematology cell counters have graphic dis-
Microscopic analysis of cytology and hematology plays to illustrate errors in what is being counted. Analy-
smears rapidly provides useful diagnostic information. ses may be performed by more than one method (e.g.,
Therefore someone in a practice should learn and two types of total white blood cell [WBC] count), and if
practice these techniques (see Chapters 2 to 5 and 16). results for the same parameter do not match, a flag signals
A good-quality microscope is necessary. “Quick” stains the operator to check that value. These more sophisticated
(e.g., Diff-Quik or Hemocolor) are easy and provide instruments usually also provide a much broader range
8   SMALL ANIMAL CLINICAL DIAGNOSIS BY LABORATORY METHODS

of data that are more precise, accurate, and rapid than are Reference values for heparinized whole blood may differ
obtained by small in-clinic instruments (see Chapters 2 from routine serum values.
to 5). The ideal analyzer should be easy to use, provide fast
Referral laboratories employ well-trained, experi- and accurate results, require minimal operator time and
enced personnel and often board-certified veterinary training, be easy to maintain, have stable reagents, rarely
clinical pathologists. These are specialists focusing only break down, and be easy to repair or be immediately
on laboratory testing, analogous to specialists in other replaced by the manufacturer as needed. Published objec-
fields (e.g., cardiology, ophthalmology, orthopedics). tive evaluation is available for relatively few commercially
Referral laboratories offer a wider variety of tests, whereas available analyzers used for veterinary testing.10,15,16,18
the variety of tests a veterinary clinic may offer is limited Manufacturers of the analyzers should be able to provide
by test volume. Offering a wider variety of tests requires data showing the precision of their assays and how well
more reagent storage and increases the likelihood that the results compare with those obtained using a reference
reagents will be outdated before they are used in a veteri- method for each species of interest.
nary practice. Outdated reagents should never be used, and The simplest analyzer measures the concentration of
loss of unused reagents adds significant cost to testing. only one analyte, such as blood glucose. Advantages of
Some referral laboratories have courier systems that pick blood glucose analyzers include rapid results, require-
up samples and return results (e.g., by fax) within hours, ment for small quantities of blood (minimum require-
fulfilling the need for rapid answers. ment of 3 to 5 µl whole blood), affordability, and
portability. Comparison of results obtained using a
variety of portable blood glucose meters (AlphaTrak
[Abbott Laboratories], Precision Xtra [Abbott Laborato-
NOTE: It is recommended to use larger veterinary labora- ries], Ascensia Elite XL [Bayer Diagnostics, Inc.], Ascensia
tories with experienced laboratory personnel, better quality Contour [Bayer Diagnostics, Inc.], Accu-Chek Advantage
assurance, better instrumentation, and lower cost per test [Roche Diagnostics, Inc.], and OneTouch Ultra2 [Life­
as much as possible and practical. Scan, Inc.]) with a reference method in dogs revealed
differences in the accuracy of the different analyzers. In
some samples, this leads to misclassification of the glyce-
mic state of the patient.4
Many in-clinic chemistry analyzers have the capacity
Use of Human Laboratories to measure multiple analytes (see Table 1-1). Those that
Many chemistry tests for animals, such as electrolytes, allow the clinic to measure a single analyte or to custom-
glucose, urea, and creatinine, are performed accurately in ize a biochemical panel are the most flexible but tend to
large human laboratories. However, veterinary referral be more labor intensive and require more technician
laboratories avoid methods that fail to work with animal attention when performing multiple tests. Analyzers
samples. Veterinary-specific testing is needed with various offering a limited number of biochemical panels that are
protein, hemostasis, immunologic, and hematologic predetermined by the manufacturer are less flexible but
analyses. Occasionally uncommon tests, such as methe- generally cost less per test. They also tend to require less
moglobin and drug testing, may be only be available at a technician time and expertise.18
local human laboratory, because they are too infrequently Different chemistry analyzers often use different
requested to offer at veterinary laboratories. Human labo- reagents or methods; therefore each analyzer system
ratories do not produce their own laboratory-derived ref- requires its own species-specific reference intervals. Two
erence intervals for animal species and cannot help with or more different instruments analyzing the same sub-
diagnosis of animal diseases. stance may be used in the same clinic, which causes con-
fusion in comparing results to a reference interval or
results on the same patient from different instruments.
Point-of-Care Chemistry Analyzers Reference intervals from referral laboratories should not
For tests that are truly “stat” (e.g., measuring blood be used for in-clinic chemistry testing unless the same
glucose concentration for suspected hypoglycemic sei- analyzer system and reagents are used. If a clinic is unable
zures), point-of-care testing is necessary. Point-of-care to generate its own reference intervals, it is important to
testing is performed in close proximity to the patient; it determine if appropriate reference intervals are available
includes analyzers used at the patient’s side and bench- from the manufacturer for the species of interest. Manu-
top instruments found nearby (e.g., in the ward). The facturers and laboratories should provide information
test available must match the clinic’s needs and the about how their reference intervals were calculated (e.g.,
species being tested. Sample size requirement is impor- the number, age, breed, and sex of animals used) upon
tant with regard to how little can be collected from dehy- request. It is also important to know if in-clinic analyzers
drated cats or puppies (Table 1-1). Analyzers that use were validated to perform well with ill patients, because
small volumes of heparinized whole blood may elimi- the data provided by manufacturers are often established
nate some artifacts associated with plasma or serum sep- in healthy animals.
aration. Use of analyzers that require plasma or serum Cost and time required for QC procedures must
means collection of 2 to 3 times more blood than be considered. Manufacturers should provide recommen-
needed for an instrument using heparinized whole dations for a QC program. If a QC program from the
blood, because of “dead space” and the fluid-cell inter- manufacturer is not recommended, is unavailable, or is
face involved in separation of serum from blood cells. inadequate, then that analytical system should be suspect.
TABLE 1-1. COMPARISON OF SOME COMMERCIALLY AVAILABLE CHEMISTRY ANALYZERS

ANALYZER DISTRIBUTOR/ REFLOVET/ SPOTCHEM DRI-CHEM 4000 VETSCAN VS2 VETTEST IDEXX
MANUFACTURER I-STAT1 ABAXIS PLUS SCIL EZ SCIL HESKA ABAXIS LABORATORIES

Number of available chemistry and 9 plus acid-base, blood 15 plus hemoglobin 21 22 23 plus canine 26
electrolyte tests gas, coagulation, heartworm, T4
troponin assays
Panels or individual tests 6 predetermined panels + Individual tests only 1 predetermined 5 predetermined 9 predetermined 7 predetermined
individual tests panel + panels + panels panels plus
individual tests individual tests individual tests
Species with available reference NA* Dog, cat, horse, cow, Dog, cat, horse Dog, cat, horse, Dog, cat, horse 18 species
intervals rabbit, guinea pig, mouse, rabbit, supported
hamster ferret, cow
Sample type Whole blood Whole blood, serum, Whole blood, Whole blood, Whole blood, serum, Serum or plasma
or plasma serum, or serum or or plasma
plasma plasma
Sample size <95 µl 32 µl 250 µl (blood); 1.5 ml (blood); 100 µl 40 µl for 1 test plus
100 µl (serum/ 0.5 ml (serum/ 10 µl for each
plasma) plasma) additional test
Recommended quality control Automatic testing of Liquid controls Liquid controls Liquid controls Check with Monthly liquid
procedures analyzer and sensors recommended available upon (daily to manufacturer controls
on cartridges; liquid request weekly
controls available upon depending on
request test volume)
Reagent type Cartridges Dry test strips Dry test strips Slides Plastic rotors Dry slides
Assay time <2 min 2–3 min per test 8–14 min <13 min for <12 min for panel 6 min for panel
complete
panel
Technical support available 24 hours Yes Yes Yes Yes Yes Yes
per day, 7 days per week†
Results transmissible to hospital Yes Yes Yes Yes Yes Yes
computer system‡

T4, thyroxine.
NA*Reference intervals for dog, cat, and horse are anticipated to be available in 2011.
†
Current availability as of December 2010; the reader should check with manufacturer for any changes and for specific information about costs associated with technical support.
‡
Check with manufacturer regarding specific information about compatible practice management software.
Chapter 1: General Laboratory Concepts   9