Diagnostic Samples: From the Patient to the Laboratory E-Book

Walter G. Guder

• Herausgeber: Wiley-VCH
• Kategorie: Wissenschaft und neue Technologien
• Sprache: Englisch

Seitenzahl: 229

Veröffentlichungsjahr: 2014

Beschreibung

This fourth, updated edition contains the latest developments in analytical techniques. An international team of authors summarizes the information on biological influences, analytical interferences and on the variables affecting the collection, transport and storage, as well as preparation of samples. In so doing, they cover age, gender, race, pregnancy, diet, exercise and altitude, plus the effects of stimulants and drugs. National and international standards are described for sampling procedures, transport, sample identification and all safety aspects, while quality assurance procedures are shown for total laboratory management.
In addition, this practical book contains a glossary as well as a separate list of analytes containing the available data on reference intervals, biological half-life times, stability and influence and interference factors.
For everyone involved in patient care and using or performing laboratory tests.

Leseprobe

Contents

Preface and Acknowledgements

Foreword to the First Edition

1 Dream and reality

2 Something unavoidable

3 Changing habits

4 May I take coffee, smoke or drink before blood sampling?

5 What if I take herbs?

6 When to test?

7 Sampling during infusion therapy?

8 Sampling in the supine or upright position?

9 What site for sampling blood?

10 Blood from the skin

11 Did the lab mix up my sample?

12 A precious sample

13 A sample that is nearly always available

14 Plasma or serum?

15 Take a lavender tube!

16 Fax me a sample

17 Samples in transit

18 How to keep a sample „fresh“

19 What has to be done on specimen arrival?

20 Continuous or batchwise?

21 Safety aspects during the preanalytical phase

22 What is needed before blood transfusion?

23 Why a separate tube for the coagulation test?

24 Blood cells are sensitive!

25 Everything from a drop of blood?

26 Special tubes for hormones and proteomics?

27 Blood cells can provide important information

28 How to handle genes

29 When gases evaporate

30 The right time for drugs …

31 Bacteria, fungi, parasites and viruses

32 Can turbid samples be used?

33 A difficult case

34 The serum sample looks reddish

35 Does the laboratory have to know all my drugs?

36 Everything under control?

References

Glossary

Index

1st Edition, 1996

2nd Edition, 2001

3rd Edition, 2003

4th Edition, 2009

Front Cover: Fractal image from Mandelbrot’s non linear mathematics. Stephen Johnson; Tony Stone Bilderwelten, Munich.

This book was carefully produced. Nevertheless, authors and publisher do not warrant the information contained therein to be free of errors. Readers are advised to keep in mind that statements, data, illustrations, procedural details or other items may inadvertently be inaccurate.

Library of Congress Card No.:

Applied for.

British Library Cataloguing-in-Publication Data:

A catalogue record for this book is available from the British Library.

Bibliographic information published by the Deutsche Nationalbibliothek

The Deutsche Nationalbibliothek lists this publication in the Deutsche Nationalbibliografie; detailed bibliographic data is available on the Internet at http://dnb.ddb.de.

© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

All rights reserved (including those of translation into other languages). No part of this book may be reproduced in any form by photoprinting, microfilm, or any other means nor transmitted or translated into a machine language without written permission from the publishers. Registered names, trademarks, etc. used in this book, even when not specifically marked as such, are not to be considered unprotected by law.

ISBN 978-3-527-32307-4

Preface and Acknowledgements

Having known each other for many years, the authors decided to summarize their experience of preanalytical variables in 1992 after observing an increasing contribution of these factors on laboratory results. They agreed to summarize their knowledge in a short and understandable form, aiming to increase the awareness of these factors among all professions involved in the preanalytical phase of the laboratory diagnostic process. This idea was generously supported by Becton Dickinson, Europe.

After the style and general contents of the book were agreed upon in a first meeting of the authors together with the publisher, the manuscripts were completed by the authors in a short time with the help of many collaborators and colleagues. The authors would especially like to thank Heidrun Dürr and Edith Rothermel, Heidelberg, Klaus Krischok, Munich, and Ulrich Wurster, Hannover, for providing and designing figures. Thanks also to Ingrid Freina, Ulrike Arnold and Patrick Bernhard, Munich, Carol Pirello, New Jersey, Kerstin Geiger, Marion Wajda and Helga Kallmeyer, Mannheim, Bernd Borstel, and Annelies Frim, Stuttgart for their expert secretarial help. David J. Purnell, Plymouth, Wolfgang Heil, Wuppertal, and James Brawley, Gaiberg/Heidelberg greatly supported our work by critically reading the manuscripts. We would like to thank Alois Jochum for translation support.

The present revised and updated fourth edition includes a new chapter ‘What if I take herbs?’. The Recommendations of the Working Group on Preanalytical Quality, previously supplied as a CD-ROM in the third edition, are now available from the freely available website www.diagnosticsample.com, which is continuously updated and kindly provided by Chronolab AG, Zug, Switzerland. Several figures have been replaced by the newest versions available to represent the latest developments in sample tubes, needles and safety aspects. We would like to thank Mrs Barbon (Becton Dickinson, Europe) for providing new figures. More than 80 references have been added and adapted to more recent literature.

In continuation of a 16 years collaboration with the Publisher GIT we thank A. Sendtko (Wiley-VCH) for his experienced support in editing this new edition in close collaboration with all contributors.

The authors hope that the new edition will help to continuously increase the awareness of preanalytical variables as a possible source of laboratory errors. As with the previous editions, it is devoted to all professions involved in the organization and performance of preanalytical steps. The authors would be pleased if this work helps to improve the quality of patient care by increasing knowledge on preanalytical variables in the laboratory diagnostic process.

Walter G. Guder

Sheshadri Narayanan

Hermann Wisser

May 2009

Foreword to the First Edition

Laboratory tests generally provide a more sensitive indicator of the state of a patient’s health than the patient’s account of how he or she feels. This has prompted an increasing emphasis on laboratory tests in the diagnosis and management of the patient’s disease. Major decisions about the management of a patient are being made on small changes in laboratory data. Thus, a decision to change the dose of a patient’s drug is often made on its plasma concentration.

Laboratories have long been aware that many non-disease factors may affect clinical laboratory test values. These include the potential effect of drugs, either through an effect on the physiological function of various organs or an interference with an analytical method.

Whereas the laboratorian may be aware of the possibility of an analytical interference, clinicians are largely unaware of these effects and the available resources to help them interpret test values correctly. When this information is not given with the result, clinicians may misinterpret test values and take an inappropriate action with their patients.

Clinical decisions based on laboratory test values are correctly made only when the conditions under which blood or other specimens are properly identified and standardized, or when the lack of standardization is recognized and allowances are made for some lack of comparability with previous test values. While laboratorians are aware of the concepts of intra- and interindividual variation as they affect laboratory data, many colleagues are unfamiliar with all but the most obvious causes of differences in test values, such as gender and age.

An understanding of intraindividual variation of test values is important if appropriate clinical decisions are to be made when serial data are being followed. The new concepts of critical differences or reference changes are now important. For proper interpretation of the typically small differences between laboratory data obtained on successive specimens from patients, the variables affecting the test values need to be standardized wherever possible, but first the pertinent variables need to be identified.

These are the issues that prompt the need to revisit all the factors related to preanalytical variables. It is thus particularly timely for this book to be published. The authors hope to reach a broader audience than the laboratorians who are probably quite familiar with many of the factors affecting test results. Since 1956, when Roger Williams published his pioneering studies on the differences between people in a book entitled Biochemical Individuality, physiologists have been concerned with the differences between people. Now that we have a broader understanding of the genetic influence on human physiology and behavior and a greater need to extract more information from small changes in laboratory data, the publication of a new book concerned with preanaytical variables which contribute to intra- and interindividual variability is both timely and welcome. This book is intended not just for laboratorians but also for physicians, nurses and everyone involved in the chain of events from the decision to order a laboratory test to the interpretation of its results.

Proper application of the information contained in this book should lead to less unnecessary testing, reduced costs and a better understanding of the results.

Philadelphia, April 1996

Donald S. Young MD, PhD

1

Dream and reality

The importance of the preanalytical phase

A new patient with diabetes mellitus is encountered

Mrs Haseltine is a 56-year-old lady who lives in a remote area. She consults her nearby practitioner and reports that over the last 2 weeks she has urinated more frequently than usual. Also, her body weight has decreased, although she “drinks more soft drinks than ever before”. The practitioner finds a positive dipstick result for glucose in her urine. Using a glucometer, he measures glucose from fingertip blood obtained by pricking with a fine lancet. The first drop of blood is washed away with a swab of gauze. In the following drop, glucose is measured by the meter, a process that takes about 30 seconds. The result is 280 mg/dL (15.56 mmol/L), far above the upper limit of the normal range. Mrs Haseltine is informed that she may have diabetes mellitus and is referred to a diabetologist the next day

The right sample for the right test at the right time

The diabetologist confirms the result obtained by the practitioner using a capillary blood sample taken 1 hour after breakfast.

Fig. 1-1

Two blood samples are drawn from the patient the following morning (after she has fasted for 12 hours), from the antecubital vein into closed tubes, one, with a lavender-coloured stopper, containing EDTA, the other, with a green cap, containing heparin. Mrs Haseltine is informed that she has type II diabetes mellitus. She is asked to phone the next day to obtain information on her laboratory results and for further advice.

In the meantime, the heparin blood sample has been centrifuged to separate plasma from the cellular elements. Both tubes are sent to the laboratory by courier in a container especially designed to keep samples at constant temperature. The laboratory receives the samples together with the patient’s data and requests for determinations: glycated haemoglobin and blood cell counts from the EDTA blood; potassium and creatinine from the plasma, which has been separated from blood cells, in the closed heparin tube.

The laboratory technician identifies all the samples by comparing the name and bar code number with those on the request sheet. He then enters the request into the lab computer. The samples are put into bar code-reading analyzers for identification and performance of the requested tests. A subsample is taken from the EDTA blood – after slowly mixing it for 3 minutes on a roller mixer – for the determination of haemoglobin A1c by chromatography. The laboratory report, shown in Tab. 1-1, is sent to the diabetologist the next morning.

Mrs Haseltine is taken into the neighbouring county hospital together with a letter from the diabetologist informing the clinician about all tests performed and the results obtained. After 2 weeks of treatment, Mrs Haseltine has learned to control her blood sugar using a small glucometer. No further treatment is needed for the next few years.

Tab. 1-1 Laboratory report Haseltine, Elsa – July 13, 2008, 10 a.m.

This is what might happen in reality

Mrs Haseltine goes to the practitioner with the same symptoms for the same condition. In contrast to the positive urine dipstick result for glucose, the blood sugar is nearly normal (120 mg/dL). The practitioner, to play it safe, again refers the patient to a diabetologist.

One week later, Mrs Haseltine is called in for a glucose tolerance test. The only advice she is given is to fast the night before the test. Mrs Haseltine wakes up late, however, and misses her morning appointment. She arrives at the doctor’s office at noon, having had a snack on the way. She is stressed when the nurse offers her a glucose-containing drink after taking a “fasting” blood specimen.

She feels nauseated while slowly consuming the drink. Whilst waiting for the nurse, she decides not to drink it all and empties the remaining drink down the bathroom sink. Of course, she doesn't report this incident to the nurse when she returns to take a capillary blood sample at 1 and 2 hours after the first sample.

When the results are shown to the doctor (Tab. 1-2), he realizes that the glucose concentrations after the first and second hour are not that much different. The diabetologist, unable to arrive at a diagnosis, asks the patient to report the following day at which time two venous blood samples are collected, one with a lavender-coloured stopper and the other with a green cap. The tubes are sent to a private laboratory by car. Next day, the results shown in Tab. 1-3 are received by E-mail together with the reference values for each test. The glucose value is now normal, potassium elevated and haemoglobin A1c, an indicator of mean blood glucose, elevated to diabetic levels. The diabetologist, concerned by the high potassium level, refers the patient to a clinic. This institution diagnoses that the patient has type II diabetes mellitus, based on their laboratory results.

Fig. 1-2

Tab. 1-2 Results in doctor’s office: glucose tolerance test Haseltine, Elsa – July 12, 2008, 2 p.m.

Fasting

glucose 160 mg/dL

1-Hour

glucose 110 mg/dL

2-Hour

glucose 120 mg/dL

Tab. 1-3 Report from private laboratory Haseltine, Elsa – July 13,2008, 3 p.m.

Fig. 1-3

What happened to Mrs Haseltine’s samples?

Undoubtedly, Mrs. Haseltine was in a diabetic state. Why was the fasting blood sugar nearly normal?

Answer: Fasting may result in nearnormal values in type II diabetics. In this case, the nurse took the first drop of blood from a fingerprick after “milking” the finger to obtain sufficient blood.

Why was the result of the glucose tolerance test inconclusive?

Answer: The first result was related to patient stress, which leads to increased amounts of glucose being released from liver glycogen stores. Moreover, Mrs Haseltine had a snack on her way to the doctor because she was hungry.

She did not report this to the doctor or the nurse, because she wasn't aware of the possible influence of this snack. For the same reason, she did not report not consuming all of the glucose drink, which had led to a decrease rather than an increase of blood glucose after 1 hour. The “increase” at the second hour may have been due either to method variation or to a reactive increase brought about by metabolic reactions in the late afternoon. Normally, a glucose tolerance test is performed in the morning, the reference values being valid only for the morning. It should be carried out under standard conditions, as recommended by national and international expert panels.

Why was potassium elevated and glucose normal in the venous specimen?

Answer: The sample was transported in contact with the cells for over 2 hours in a non-air-conditioned car on a hot day. This caused the blood cells to metabolize glucose and release potassium, the concentration of which is approximately 40 times higher in cells than in plasma. This in vitro influence makes unstabilized blood unsuitable for glucose determination. Potassium can be reliably measured only if plasma is promptly separated from the cells.

All these errors could have been prevented had the preanalytical phase been strictly controlled. Mrs Haseltine would have been diagnosed earlier with less stress and fewer costs would have been incurred.

This book is intended to increase awareness of the importance of all steps of the preanalytical phase, including patient preparation, sampling, transport and storage of patient samples.

In each chapter, possible preanalytical variables are explained with regard to mechanisms, effects and preventive actions intended to prevent misinterpretation of laboratory results. In the respective chapters, warnings are given in red and recommendations in green. Like disease mechanisms, biological influences can change the concentration of measured analytes in vivo, whereas in vitro changes have to be separated into changes undergone by the measured analyte and interference of the method used to measure the analyte. These definitions are important, because only the latter can be avoided by using a more specific method. The interested reader is referred to the References (p. 90) as well as the Glossary which defines all the special terms used in this book (p. 102). Detailed information on preanalytical variables of all analytes together with the recommendations on the choice of anticoagulant, the optimal sample volume and the stability of analytes in sample matrix is referred to the website www.diagnosticsample.com (112). It is to be hoped that the new edition helps to improve sample quality and decrease preanalytical errors (25, 171).

Optimal treatment of the patient and their samples is defined as the gold standard

2

Something unavoidable

Influences of age, gender, race and pregnancy

Intrinsic influences such as race, gender and age may influence target analyte concentrations in clinical chemistry and haematology. These variables are individual features of a subject and hence not subject to change. Quite often, intrinsic and external factors are difficult to distinguish.

Age

Age may affect blood and urine analyte concentrations after birth, during adolescence or in old age (Fig. 2-1). Erythrocyte counts and hence haemoglobin are much higher in neonates compared to adults. Within the first few days following birth, increased arterial oxygen provokes erythrocyte degradation. The resulting increase in haemoglobin leads in turn to enhanced concentrations of bilirubin. Since liver function (here in particular glucuronidation) is not fully established in neonates, increased concentrations of bilirubin are observed.

Fig. 2-1 Age dependence of various substrates and enzyme activity (9, 49). AP was measured at 30 °C (86°F)

Uric acid concentrations in neonates are in a range similar to adults. However, within days after birth, a significant decrease is observed. Other examples of agedependence include alkaline phosphatase activity (AP) in serum (which peaks during the growth phase, mirroring bone osteoblast activity) and total and low-density lipoprotein (LDL)-cholesterol. In addition, agedependent AP activity and LDL- and high-density lipoproteine (HDL)-cholesterol in serum are influenced by gender. These gender differences in turn change as a function of age.

Race

Fig. 2-2 illustrates examples of analytes which are affected by race. Black Americans of both genders have significantly lower white blood cell counts compared to whites. This difference is readily explained by a reduction in the number of granulocytes. In contrast, haemoglobin, haematocrit and lymphocyte counts are almost identical in both groups (148). The monocyte count in whites exceeds that of blacks (13). A significant difference in creatine kinase (CK) activity has been observed for both genders in black and white people. This difference is not due to differences in age, height or body weight (123). The distribution of creatinine is similar to CK. Both quantities are partly determined by the muscle mass. The gender and race differences are more pronounced with increasing age and exist also for the relationship of serum creatinine to glomerular filtration rate (271). Using the same reference range for serum creatinine without consideration of age, gender and race/ethnicity for glomerular filtration rate screening can be misleading. Significant racial differences have been reported for serum concentrations of vitamin B12 (1.35 times higher concentrations in black people) (243) and Lp(a) (2 times higher concentrations in blacks compared to whites). This finding is confirmed by a study in hypercholesterinaemic black students compared with whites with 2.8 (male) and 2.6 (female) higher Lp(a) values of the blacks (127).

Fig. 2-2 Influence of race/ethnicity on CK, creatinine (age 40-59 years) and granulocytes in blood (123, 142, 148)

Gender

Gender-specific hormone patterns, gender differences can likewise be found in clinical chemistry and haematology (Fig. 2-3). The gender difference of serum iron concentrations disappears in patients older than 65 years. Other examples of gender differences are CK and creatinine. The serum activity or concentration depends on muscle mass which is in general more pronouncedinmales.

Pregnancy

When interpreting laboratory results during pregnancy, it is necessary to take into account the gestational week at which each sample was taken.

During a healthy pregnancy, the mean plasma volume rises from about 2.6 to 3.9 L, with probably little change occurring in the first 10 weeks of gestation, and a subsequent progressive rise up to the 35th week, at which time the values level off.

The urine volume may also increase physiologically by up to 25% in the third trimester. There is a 50% physiological increase in the glomerular filtration rate in the last trimester. The well-known changes in hormone production and the plasma concentrations of fertility hormones during pregnancy are accompanied by changes in various analytes, as shown in Fig. 2-4. The concentration changes are caused by different mechanisms, such as increased synthesis of transport proteins, increased metabolic turnover rate or dilution.

Fig. 2-3 Male–female differences related to the mean value of females as given in (49)

Fig. 2-4 Analyte changes in the last trimester of pregnancy (312).

3

Changing habits

Influences that can vary (diet, starvation, exercise, altitude)

Diet

Diet and drinking are major factors influencing a number of analytes in clinical chemistry. From a practical point of view, one should distinguish acute effects from those observed over a longer period. A critical question in daily routine is whether a standard meal affects target analytes. Fig. 3-1 shows the percentage change in different analyte concentrations as a function of food intake (64, 269). Effects of 5% and less may be neglected, since they are clinically irrelevant. Therefore, samples for these analytes do not require strict food deprivation. The extent of food-induced alterations in analytes depends on the composition of the food and the elapsed time between sampling and food intake. The serum concentration of cholesterol and triglycerides are influenced by various factors such as food composition (229), physical activity, smoking, consumption of alcohol and coffee (87). Elevated levels of ammonia, urea and uric acid are observed during a high protein and nucleotide diet. The changes occurring after a standard carbohydrate meal (75 g) are diagnostically helpful in testing glucose tolerance. On the other hand, malnutrition and starvation may alter analyte concentrations in a clinically relevant fashion. Early indicators of low protein diet are reduced serum concentrations of transthyretin (prealbumin) and retinol-binding protein. Some alterations in clinical chemical analytes induced by starvation over 48 hours are summarized in Fig. 3-2. Metabolic acidosis with a decrease of both pH and bicarbonate results from an increase in organic acids, mainly the ketone bodies (acetoacetic acid, 3-hydroxybutyric acid).

Fig. 3-1 Change of the serum concentration of different analytes 2 hours after a standard meal (64,269)

Starvation

Changes in analyte concentrations induced by long-term starvation (4 weeks) are shown in Fig. 3-3 at the end of the starvation period in comparison to the initial values. The concentrations of blood cholesterol, triglycerides and urea are reduced. In contrast, creatinine and uric acid concentrations are elevated. The increase in uric acid concentration during starvation periods even requires treatment. The latter is due to reduced clearance of uric acid as a result of ketonaemia (76). It is readily apparent that long-term starvation is closely associated with reduced energy expenditure. Besides such alterations, urinary excretion of several compounds is likewise affected by long-term starvation. Urinary excretion of ammonia and creatinine is increased, whereas that of urea, calcium and phosphate is reduced (301). Changes in analyte concentrations brought about by long-term starvation are similar to those observed following surgical procedures or in patients with a catabolic status. Not only long-term food restriction, but also short-term restriction induces significant changes of serum concentration of several biochemical and endocrine quantities. After an average weight loss of 1.8 kg during 1 week, the following changes were registered: triglycerides – 25%, free fatty acids +124%, glycerol + 74% by increased lipolysis, ammonia – 17%, urea + 11%, uric acid + 10% by activated protein metabolism, insulin – 42%, adrenocorticotropic hormone (ACTH) + 41%, cortisol – 24% and testosterone – 34% (71).

In measuring quantitative urinary excretion rates, excreted amounts per day are preferable to those per litre in order to eliminate variations in drinking habits and water excretion.

Mechanisms

Changes may be due either to an increase in reabsorption of the measured analyte (triglycerides, glucose, amino acids), intestinal or liver metabolism of reabsorbed metabolites (very-LDL (VLDL), urea, ammonia) or regulatory changes due to food intake or deprivation (uric acid, γ-glutamyltransferase, cholinesterase, thyroxine, retinol-binding protein, ketone bodies).

Recommendation

In order to avoid misinterpretation of laboratory results, sampling after 12 hours fasting and reduced activity is recommended as a standard procedure.

Exercise

Before considering the influence of exercise on target analytes in clinical chemistry, two types of exercise have to be distinguished. First, static or isometric exercise of brief duration and high intensity which utilizes the energy (ATP and creatine phosphate) already stored in muscle and, second, dynamic or isotonic exercise of lower intensity and longer duration (e.g. running, swimming, cycling) which utilizes ATP produced by aerobic or anaerobic pathways. In addition, the effect of physical training and muscle mass should be mentioned. Acute changes of analytes during exercise are due to volume shifts between the intravasal and interstitial compartments, volume loss by sweating, and changes in hormone concentrations (e.g. increase in the concentrations of epinephrine, norepinephrine, glucagon, somatotropin, cortisol, ACTH and decreased concentrations of insulin) (5, 232). These changes in hormone levels may in turn alter the leukocyte count to more than 25 G/L as well as increasing glucose concentrations. Fig. 3-4 shows changes in analyte concentrations induced by marathon running (258, 266). The extent of change depends on a variety of individual and/or environmental factors (e.g. training status, air temperature, and intake of electrolyte- and carbohydrate-containing liquids during the actual run).

Fig. 3-2 Variation of several analytes after 40–48 h starvation (164). *Starting point after 14 hours starvation

Fig. 3-3 Change (%) of clinical chemical analytes after 4 weeks starvation and a daily supply of 33 g protein, vitamins and electrolytes (76, 302)

Fig. 3-4 Effects of marathon running on biochemical and haematological parameters. Blood was drawn 1–3 days before the race and within 1 hour of finishing (258, 264, 266)

The changes observed (e.g. increased albumin) can in part be attributed to the above-mentioned volume shift from intravasal to the interstitium or to loss of volume by sweating, but the small increase in plasma volume indicates only a minor haemodilution in most runners (258). The increased uric acid concentration in serum is a consequence of reduced urinary excretion due to increased lactate concentrations. Hypoxia-mediated CK increase depends on the training status and hence shows a high degree of individual variability. The less physically fit an individual is, the more pronounced the increase in CK. Training increases both the number and the size of mitochondria which is associated with increased capacity of the oxidative enzyme system. This effect in turn increases the capacity of the muscle to metabolize glucose, fatty acids and ketone bodies in aerobic pathways. As a consequence, mitochondrial CK-MB increases to more than 8% of the total CK activity without evidence of altered myocardial function. Well-trained individuals have a higher percentage of total activity in terms of the CK-MB of skeletal muscle compared to untrained persons. In the study of Smith et al.