New Book: Abnormal Laboratory Results, Geoffrey Kellerman


Published on

Abnormal Laboratory Results is a comprehensive collection of common abnormal test results which examine the dilemmas of abnormal tests and outlines how to approach them. Each test is analysed and interpreted in a structured, logical way and assessed in terms of its abnormal and normal reference range. The results are then presented in a variety of clinical contexts so an accurate prognosis and recommendations on management can be made. Each chapter was originally published as an article in the series titled 'Abnormal Laboratory Results' by the journal Australian Prescriber

More info:

  • Be the first to comment

No Downloads
Total views
On SlideShare
From Embeds
Number of Embeds
Embeds 0
No embeds

No notes for slide

New Book: Abnormal Laboratory Results, Geoffrey Kellerman

  1. 1. AbnormalLaboratoryResults CK ICK LI FManual QUThird Edition ly on s geEdited by paGeoffrey Kellerman e Australian Prescriber pl m sa
  2. 2. AbnormalLaboratory ResultsEdited by Geoffrey Kellerman Australian Prescriber ly on Third Edition s ge pa e pl m sa
  3. 3. NoticeMedicine is an ever-changing science. As new research and clinical experience broaden ourknowledge, changes in treatment and drug therapy are required. The editors and the publisher ofthis work have checked with sources believed to be reliable in their efforts to provide informationthat is complete and generally in accord with the standards accepted at the time of publication.However, in view of the possibility of human error or changes in medical sciences, neither theeditors, nor the publisher, nor any other party who has been involved in the preparation orpublication of this work warrants that the information contained herein is in every respect accurateor complete. Readers are encouraged to confirm the information contained herein with othersources. For example, and in particular, readers are advised to check the product information sheetincluded in the package of each drug they plan to administer to be certain that the informationcontained in this book is accurate and that changes have not been made in the recommended doseor in the contraindications for administration. This recommendation is of particular importance inconnection with new or infrequently used drugs.This third edition published 2011First edition published 2001Second edition published 2006Text © 2011 Commonwealth Government of Australia and National Prescribing ServiceIllustrations and design © 2011 McGraw-Hill Australia Pty LimitedAdditional owners of copyright are acknowledged on the Acknowledgments pageEvery effort has been made to trace and acknowledge copyrighted material. The authors andpublishers tender their apologies should any infringement have occurred.Reproduction and communication for educational purposesThe Australian Copyright Act 1968 (the Act) allows a maximum of one chapter or 10% of the pagesof this work, whichever is the greater, to be reproduced and/or communicated by any educationalinstitution for its educational purposes provided that the institution (or the body that administersit) has sent a Statutory Educational notice to Copyright Agency Limited (CAL) and been granteda licence. For details of statutory educational and other copyright licences contact: CopyrightAgency Limited, Level 15, 233 Castlereagh Street, Sydney NSW 2000. Telephone: (02) 9394 7600.Website: and communication for other purposesApart from any fair dealing for the purposes of study, research, criticism or review, as permittedunder the Act, no part of this publication may be reproduced, distributed or transmitted in anyform or by any means, or stored in a database or retrieval system, without the written permissionof McGraw-Hill Australia including, but not limited to, any network or other electronic storage.Enquiries should be made to the publisher via or marked for the attentionof the permissions editor at the address below.National Library of Australia Cataloguing-in-Publication DataTitle: Abnormal laboratory results / edited by Geoffrey Kellerman.Edition: 3rd ed.ISBN: 9780070998421 (pbk.)Notes: Includes index. lySubjects: Diagnostic errors. Diagnosis, Laboratory—Evaluation. on Clinical chemistry.Other Authors/Contributors: Kellerman, Geoffrey.Dewey Number: 616.0756 sPublished in Australia byMcGraw-Hill Australia Pty Ltd geLevel 2, 82 Waterloo Road, North Ryde NSW 2113Associate editor: Fiona RichardsonArt director: Astred Hicks paInternal design: Peta NugentProduction editor: Claire LinsdellCopyeditor: Kathy KramerIllustrator: Alan Laver, Shelly Communications eProofreader: Vicki Deakin plIndexer: Barbara BessantTypeset in Zapf Humanist 8/10.5pt by Midland Typesetters, AustraliaPrinted in China on 80gsm matt art by IBook Printing Limited m sa
  4. 4. ForewordA bewildering array of laboratory tests is available to today’s healthprofessionals, and there is more to these tests than ticking a box on a requestform and looking at the normal range of results. To provide more detail aboutcommonly ordered tests, Australian Prescriber runs a series called ‘AbnormalLaboratory Results’. Although this may seem an odd topic for a journal mainlyconcerned with drugs, laboratory tests have an important role in therapeutics. Australian Prescriber first published a booklet on laboratory tests in 1987.Over the ensuing years more articles appeared in the series, so there wasa need to update the booklet. Compiling and revising all the material wasbeyond the resources of Australian Prescriber but McGraw-Hill was able tohelp and gave Dr Robert Dunstan the responsibility of updating all of thearticles published since 1978. The result was the first edition of AbnormalLaboratory Results, which appeared in 2001. By 2006, Australian Prescriber had published enough articles on newtopics to warrant a second edition of the book. This was edited by ProfessorGeoffrey Kellerman, who not only added the new chapters but also ensuredthat the earlier chapters were brought up to date. Australian Prescriber has published more articles in the AbnormalLaboratory Results series over the past four years. I am pleased that ProfessorKellerman has again agreed to prepare these articles for the third edition ofthe book. This 2011 edition is timely because NPS, the publisher of AustralianPrescriber, has recently established a Quality Diagnostic Referrals program. lyThis has the aim of improving the use of laboratory tests. Abnormal onLaboratory Results will be a perfect complement to these educationalactivities. sDr John S. Dowden geEditor, Australian PrescriberCanberra eAustralian Prescriber is published by NPS, an independent, not-for-profit organisation plfunded by the Australian Government Department of Health and Ageing. m sa
  5. 5. ContentsForeword iiiPreface to the third edition viiContributors viiiAcknowledgments xviiiPart 1 General Issues 1 1 What to do about abnormal laboratory results 2 2 Abnormal laboratory results 6 3 What does the reference range of a biochemical screen test mean? 10 4 Pitfalls in interpreting laboratory results 14 5 Moving beyond sensitivity and specificity: using likelihood ratios to help interpret diagnostic tests 23 6 Point-of-care testing comes of age in Australia 30 7 Urine testing 40Part 2 Biochemistry Tests 47 8 Plasma sodium 48 9 Serum potassium 5610 Serum urea 60 ly11 Assessing renal function 65 on12 Interpreting arterial blood gases 7813 Calcium and vitamin D 90 s14 Magnesium: the forgotten electrolyte 102 ge15 Hyperuricaemia 111 pa16 Liver function tests 11517 Interpretation and significance of a high blood cholesterol 121 e pl18 Managing hyperlipidaemia: criteria for investigating lipids 127 m sa
  6. 6. 19 New cardiac markers 13620 B-type natriuretic peptide: a new diagnostic tool for congestive heart failure 14221 Thyroid function tests 14722 Evaluating adrenocortical function in adults 15523 The glucose tolerance test 16524 Diabetes monitoring: use of glycated haemoglobin and glycated protein assays 17225 Fertility testing 17926 Biochemical tests in pregnancy 18427 Biochemical tests for abnormalities in pregnancy 19228 Interpreting paediatric biochemistry results 20029 Drug screens 20730 Therapeutic drug monitoring: which drugs, why, when and how to do it 213Part 3 Haematology Tests 21931 The red cells 22032 Interpreting biochemical tests for iron deficiency: diagnostic difficulties imposed by the limitations ly of individual tests 22533 Appropriate use of tests for folate and vitamin B12 on deficiency 23134 Screening for thalassaemia and haemoglobin variants 238 s35 Investigations for thrombotic tendencies 247 ge36 Tests of haemostasis: detecting the patient at risk pa of bleeding 25637 Abnormal haematology results in children 264 e pl m sa
  7. 7. ContentsPart 4 Microbiology Tests 26938 Hepatitis B: laboratory diagnosis and vaccination 27039 Hepatitis C: laboratory diagnosis and monitoring 28140 HIV testing in Australia 29341 Testing for sexually transmitted infections 30142 Testing for Helicobacter pylori 314Part 5 Immunology Tests 31943 Screening for multiple myeloma 32044 Testing cell-mediated immunity 32745 Cell markers 33546 Skin prick testing and in vitro assays for allergic sensitivity 34047 Antinuclear antibodies (ANA) 34548 Rheumatoid factor and anti-citrullinated peptide antibody 350Part 6 Genetic Tests 35549 BRCA testing for familial breast cancer 356 ly onIndex 362 s ge pa e pl m sa
  8. 8. Preface to thethird editionSince the publication of the second edition of this work in 2006 there havebeen 13 further articles in Australian Prescriber in the ‘Abnormal LaboratoryResults’ series. These articles, and the 39 from the second edition ofMcGraw-Hill’s Abnormal Laboratory Results, have been collected in thisnew edition, and almost every one has been updated. These updates havebeen made by the original author, or authors, with the exception of the few,identified at the beginning of the chapter, where the original author wasunable to perform the task and another expert did so. In most articles the changes are still relatively minor, so that the bookcan maintain its role as a reissue of articles originally published in AustralianPrescriber. However, research gradually adds to or changes acceptedknowledge and practice and this is especially noticeable in the areasof hyperlipidaemia, calcium and bone metabolism, cardiac troponins,vitamin B12 metabolism, rheumatoid arthritis and viral diseases such asHepatitis B and C and HIV, where extensive rewriting has occurred.Chapters from new authors have replaced two of the original articles, and asingle chapter combines the original two on serum creatinine and creatinineclearance. Most articles include some key references so that interestedreaders can extend their study and understanding of areas of special interest. I am grateful to the authors for their cooperation and to the publishersfor their help and advice during the writing and production periods. It is ourhope that this volume will prove to be of ongoing value to all readers who areinvolved, or preparing to become involved, in patient care in any aspect of lythe health care system. on While it includes nearly 50 chapters, this book is not yet a complete coverof all laboratory testing procedures; however, further articles are plannedfor Australian Prescriber to cover important aspects that have not yet been saddressed. There is a real lack of a suitable entry-level textbook for students geand new graduates in the area of test ordering and interpretation. This gapcould be filled by a fourth edition of this publication if the requisite articles pacan be identified and selected over the next four to five years. We welcome suggestions from readers for priority areas for future topics. e plGeoffrey M. Kellerman m sa
  9. 9. ContributorsT. I. Robertson Chapter 1Visiting PhysicianWestmead Hospital, Sydney, AustraliaJ. Fowler Chapter 1Retired Consulting Physician, Newcastle, AustraliaC. G. Fraser Chapter 2Retired Clinical LeaderDepartment of Biochemical MedicineNHS Tayside, Ninewells Hospital and Medical School,Dundee, ScotlandG. M. Kellerman Chapter 3Emeritus Professor of Medical BiochemistryUniversity of NewcastleVMO in Clinical ChemistryHunter Area Pathology Service, Newcastle, AustraliaP Phillips . Chapter 4Senior DirectorEndocrinology UnitQueen Elizabeth Hospital, Adelaide, AustraliaJ. Attia Chapter 5Academic Director, General MedicineJohn Hunter Hospital, Newcastle, Australia lyM. Shephard Chapter 6 onAssociate ProfessorDirector and Senior Research FellowCommunity Point of Care Services sFlinders University Rural Clinical School geBedford Park, SA, Australia paJ. F. Mahony Chapter 7Former Renal PhysicianDepartment of Renal Medicine eRoyal North Shore Hospital, Sydney, Australia pl mT. H. Mathew Chapter 8Medical Director saKidney Health Australia, Adelaide, Australia
  10. 10. G. S. Stokes Chapter 9Emeritus ProfessorHypertension Unit, Department of CardiologyRoyal North Shore Hospital, Sydney, AustraliaE. P MacCarthy . Chapter 9Former HeadHypertension UnitUniversity of Cincinnati Medical Centre, Cincinnati,Ohio, United States of AmericaT. H. Mathew Chapter 10Medical DirectorKidney Health Australia, Adelaide, AustraliaB. J. Nankivell Chapter 11Department of Renal MedicineWestmead Hospital and University of Sydney, Sydney, AustraliaA. K. Verma Chapter 12Resident, General MedicineGosford Hospital, Gosford, AustraliaP Roach . Chapter 12Consultant PhysicianRespiratory and Sleep MedicineGosford Hospital, Gosford, Australia lyJ. D. Wark Chapter 13 onProfessor of MedicineUniversity of MelbourneHead, Bone and Mineral Service sRoyal Melbourne Hospital, Melbourne, Australia geC. J. Yates Chapter 13 paResearch FellowDepartment of Diabetes and EndocrinologyRoyal Melbourne Hospital, Melbourne, Australia e plJ. Wu Chapter 14 mRegistrar, Chemical PathologyQueensland Health Pathology Service saBrisbane, Australia
  11. 11. ContributorsA. Carter Chapter 14Director Chemical PathologyQueensland Health Pathology ServiceBrisbane, AustraliaB. T. Emmerson Chapter 15Emeritus ProfessorDepartment of Medicine, University of Queenslandand Honorary Research ConsultantPrincess Alexandra Hospital, Woolloongabba, AustraliaL. W. Powell Chapter 16Director of Clinical ResearchRoyal Brisbane and Women’s Hospital, Brisbane, AustraliaM. L. Bassett Chapter 16Australian National University Medical SchoolGastroenterology and Hepatology UnitCanberra Hospital, Canberra, AustraliaW. G. E. Cooksley Chapter 16Professorial Research FellowDiscipline of MedicineUniversity of Queensland, Royal Brisbane and Women’s HospitalBrisbane, AustraliaA. M. Dart Chapter 17 lyProfessor, Clinical Program DirectorCardiorespiratory/Intensive Care onAlfred Hospital, Melbourne, AustraliaC. M. Reid Chapter 17 sPrincipal Research Fellow and Head geClinical Informatics and Data Management UnitSchool of Public Health and Medicine paMonash University, Melbourne, Australia e pl m sa
  12. 12. G. L. Jennings Chapter 17Professor and DirectorBaker IDI Heart and Diabetes InstituteHead, WHO Collaborating Centre for Research andTraining in Cardiovascular DiseaseProfessor (Honorary)Faculty of Medicine Nursing & Health SciencesMonash University, Melbourne, AustraliaR. A. J. Conyers Chapter 17Professor, Department of ImmunologyMonash University, Melbourne, AustraliaE. M. Nicholls Chapter 17The Alfred Hospital and Baker Medical Research InstituteMelbourne, AustraliaH-G. Schneider Chapter 17Director of PathologyAlfred Hospital, Melbourne, AustraliaP Nestel . Chapter 18Professor, Senior Faculty, Baker IDI Heart and Diabetes InstituteMelbourne, AustraliaP E. Hickman . Chapter 19Director of Chemical PathologyACT Pathology lyAssociate Professor onAustralian National University Medical SchoolCanberra, Australia sJ. M. Potter Chapter 19 geProfessor of PathologyAustralian National University Medical SchoolExecutive Director paACT PathologyCanberra, Australia e plB. Ewald Chapter 20General Practitioner and mSenior LecturerCentre for Clinical Epidemiology and Biostatistics saUniversity of Newcastle, Australia
  13. 13. ContributorsR. H. Mortimer Chapter 21Department of EndocrinologyRoyal Brisbane and Women’s HospitalProfessor, University of Queensland, Brisbane, AustraliaJ. T. Ho Chapter 22Consultant EndocrinologistRoyal Adelaide Hospital, Adelaide, AustraliaD. J. Torpy Chapter 22Associate ProfessorSenior Consultant EndocrinologistRoyal Adelaide Hospital, Adelaide, AustraliaS. K. Gan Chapter 23Associate Professor in MedicineRoyal Perth Hospital and University of Western Australia,Perth, AustraliaD. J. Chisholm Chapter 23Emeritus Professor and Head Clinical Diabetes ResearchDiabetes and Obesity Research Program, Garvan Institute ofMedical Research and St Vincent’s Hospital, Sydney, AustraliaG. Jones Chapter 24Head, Department of Chemical PathologySt Vincent’s Hospital, Sydney, Australia lyD. J. Chisholm Chapter 24 onEmeritus Professor and Head Clinical Diabetes ResearchDiabetes and Obesity Research Program, Garvan Institute ofMedical Research and St Vincent’s Hospital, Sydney, Australia s geH-G. Schneider Chapter 24Director of Pathology paAlfred Hospital, Melbourne, AustraliaR. J. Norman Chapter 25 eDirector, Research Centre for Reproductive Health, plAcademic Head, Repromed Pty Ltd mProfessor, Department of Obstetrics and GynaecologyUniversity of Adelaide, Adelaide, Australia sa
  14. 14. H. A. Tran Chapter 26Head and Associate ProfessorDivision of Clinical Chemistry, Hunter Area Pathology ServiceJohn Hunter Hospital and University of Newcastle, Newcastle, AustraliaH. A. Tran Chapter 27Head and Associate ProfessorDivision of Clinical Chemistry, Hunter Area Pathology ServiceJohn Hunter Hospital and University of Newcastle, Newcastle, AustraliaP Verras . Chapter 28Complex Biochemistry DepartmentDivision of Laboratory ServicesRoyal Children’s Hospital, Melbourne, AustraliaR. Greaves Chapter 28Senior Lecturer, Clinical BiochemistrySchool of Medical SciencesRoyal Melbourne Institute of TechnologyMelbourne, AustraliaN. A. Buckley Chapter 29Professor of MedicineUniversity of NSWPrince of Wales Hospital, Sydney, AustraliaR. A. Ghiculescu Chapter 30Former Registrar lyDepartment of Clinical Pharmacology onPrincess Alexandra Hospital, Brisbane, AustraliaP Pillans . Chapter 30 sProfessor and Director, Clinical Pharmacology gePrincess Alexandra Hospital, Brisbane, AustraliaW. R. Pitney Chapter 31 paLate Emeritus ProfessorDepartment of Medicine eUniversity of NSW, Sydney, Australia plM. Seldon Chapter 31 mSenior HaematologistHunter Haematology Unit, Mater Hospital and saHunter New England Area Health Service, Newcastle, Australia
  15. 15. ContributorsF. Firkin Chapter 32Associate Professor of MedicineSenior HaematologistSt Vincents HospitalMelbourne, AustraliaB. Rush Chapter 32Former Director of Laboratory HaematologySt Vincent’s Hospital, Melbourne, AustraliaJ. Metz Chapter 33HaematologistDorevitch Pathology, Fairfield, AustraliaD. K. Bowden Chapter 34Associate ProfessorThalassaemia Service, Monash Medical Centre,Melbourne, AustraliaR. Baker Chapter 35Professor, Centre for Thrombosis and HaemophiliaRoyal Perth HospitalMurdoch University, Perth, AustraliaJ. McPherson Chapter 36Consultant and Honorary Professor of Medical EducationUniversity of Papua New Guinea lyFormerly Consultant Haematologist, Newcastle Mater Hospital,Newcastle, Australia onA. Street Chapter 36Head, Haemostasis and Thrombosis Units sThe Alfred Hospital, Melbourne, Australia geA. L. Greenway Chapter 37 paPaediatric HaematologistRoyal Children’s Hospital, Melbourne, Australia eP Monagle . Chapter 37 plStevenson Professor of Paediatrics mHead Department of HaematologyRoyal Children’s Hospital, Melbourne, Australia sa
  16. 16. D. S. Bowden Chapter 38Head, Molecular MicrobiologyVictorian Infectious Diseases Reference LaboratoryMelbourne, AustraliaS. A. Locarnini Chapter 38Head, Research and Molecular DevelopmentVictorian Infectious Diseases Reference Library andDirector, WHO Collaborating Centre for Virus Referenceand Research Melbourne, AustraliaS. Nicholson Chapter 38Victorian Infectious Diseases Reference LaboratoryMelbourne, AustraliaI. Gust Chapter 38ProfessorDepartment of Microbiology and ImmunologyUniversity of Melbourne, Melbourne, AustraliaD. Siebert Chapter 38Director of Clinical VirologyQueensland Health Pathology ServiceRoyal Brisbane Hospital, Brisbane, AustraliaA. J. Thompson Chapter 38Victorian Infectious Diseases Reference Laboratory lyMelbourne, Australia onD. S. Bowden Chapter 39Head, Molecular MicrobiologyVictorian Infectious Diseases Reference Laboratory sMelbourne, Australia geA. M. Breschkin Chapter 40 paSenior ScientistVictorian Infectious Diseases Reference LaboratoryMelbourne, Australia e plM. G. Catton Chapter 40 mMedical Director and HeadVictorian Infectious Diseases Reference Laboratory saMelbourne, Australia
  17. 17. ContributorsC. J. Birch Chapter 40Senior ScientistVictorian HIV Reference Laboratory, VIDRLMelbourne, AustraliaC. Ooi Chapter 41Director, Sexual Health ServiceConjoint LecturerFaculty of HealthUniversity of Newcastle, AustraliaD. Badov Chapter 42Head of Gastroenterology and Consultant GastroenterologistDepartment of GastroenterologyFrankston Hospital, Melbourne, AustraliaF. Firkin Chapter 43Associate Professor of MedicineSenior HaematologistSt Vincents Hospital, Melbourne, AustraliaS. Limaye Chapter 44ImmunologistConcord Hospital, Sydney, AustraliaK. Cartwright Chapter 45Clinical Associate Professor lySenior Consultant HaematologistWollongong Hospital, Wollongong, Australia onR. M. O’Brien Chapter 46Associate Professor of Medicine, Immunology and Rheumatology sAustin Hospital, Melbourne, Australia geS. M. Chatfield Chapter 47 paRheumatologistRoyal Melbourne Hospital, Melbourne, Australia eS. M. Chatfield Chapter 48 plRheumatologist mRoyal Melbourne Hospital, Melbourne, Australia sa
  18. 18. C. Lau Chapter 49Registrar, Genetics and Molecular PathologyWomen’s and Children’s Hospital, Adelaide, AustraliaG. Suthers Chapter 49Head, Familial Cancer UnitWomen’s and Children’s Hospital, Adelaide, Australia ly on s ge pa e pl m sa
  19. 19. AcknowledgmentsThe editor would like to thank the following people for their assistance in thework which led to the production of this book:• Each chapter’s contributors who, despite their busy schedules, met the deadlines.• J. S. Dowden and staff from Australian Prescriber, for their advice and encouragement.• Staff from McGraw-Hill Australia, for their cheerful guidance.• Kathy Kramer, for her copyediting. ly on s ge pa e pl m sa
  20. 20. PartGeneral Issues 1 ly on s ge pa e pl m sa
  21. 21. 1 What to do about abnormal laboratory results The context A note from the Editor In the mosaic that represents the totality of the patient’s problem and its management, laboratory results often play a minor part, sometimes are moderately important and occasionally are major contributors. These proportions depend on whether, for example, the patient is in general practice or a teaching hospital environment. Laboratory test results represent information and information serves to provide material for thought by the doctor: it is not a substitute for thought. Only in science fiction do machine tests make a magic assessment without human intervention. Dr Robertson (deceased) wrote initially from the point of view of the physician receiving results of laboratory tests. Dr Fowler has written recently from the point of view of the physician who is considering which tests to order. I believe that these two versions of the same introductory message are both worth reading. T. I. Robertson lyA biochemical profile in a symptomless male patient of 50 years shows ona modestly elevated serum uric acid. Anti-gout medication lowers thelevel but causes a generalised skin rash that lasts for 3 months. A patient,55 years old, in hospital with a myocardial infarct, is found on a routine sblood count to have a haemoglobin level of 110 g/L. The stained film gesuggests iron deficiency. Follow up investigation uncovers a symptomlesscarcinoma of the caecum, which is successfully removed. These two paexamples, each initiated by a routine test, show on the one hand anannoying disability resulting from treatment that was probably unnecessary eand on the other, a successful conclusion to the pursuit of an apparently plminor abnormality. m We live in a maze of biochemical profiles, screening tests and routineinvestigations, a lot of them uncalled for and some presented to us by our sa
  22. 22. CHAPTER 1: WHAT TO DO ABOUT ABNORMAL LABORATORY RESULTS 3patients themselves. A new category has been created: the symptomlesspatient with an investigational abnormality. These people are at special K CK FLICrisk. Some of them have incipient, developing or subclinical disease andsome have no disease at all. But all will be affected by the medical advice 1 UIthey receive. Neglect of an apparent triviality may be lethal, yet a clearly Qabnormal result may, in the patient’s total interest, be best set aside. Theadvent of routine screening procedures has produced fresh problems andresponsibilities and has tended to complicate rather than simplify medicalmanagement. When should an abnormal result be considered undebatably abnormal?When should it be acted upon? How vigorously should it be pursued? Whenshould it be ignored? The magnitude of the problem can be reduced at thesource by discrimination in the investigations requested in the first place.The fewer tests ordered in isolation—that is, without proper integrationwith history, physical examination and general consideration of the clinicalproblem—the less difficult will be their interpretation. If a patient requiresinvestigation of any type, he or she deserves the courtesy of proper historyand examination first. It is unrealistic to deny the ease, advantages and extra informationprovided by the automated blood count and multiple biochemical analysis.An investigation should, however, be mounted by way of a working diagnosisto be proved or disproved and the results of all tests viewed in that light.It should not be an undisciplined fishing expedition. If it is practical for onlyselective tests to be requested, to advance the working hypothesis, onlythese should be ordered. Secondary tests may be needed after analysisof the primary ones but this is the preferred approach. The practitioneradvances his or her capabilities by being careful, even parsimonious, withinvestigations and by understanding them and their limitations. This appliesto all levels of clinical practice, not least in the teaching hospitals. Nevertheless the problem of interpreting more or less isolated abnormal lyresults will always remain. Some of the difficulties in relation to particular ontests are addressed by experts in the chapters that follow. s J. Fowler geThe clinical assessment, done with diligence, directs the investigationsto support or refute a provisional diagnosis. A poor clinical history and paphysical examination (a time consuming process to do properly) does notpermit you to ask the correct questions and poorly directed laboratory etests will not compensate for the omission. The solution of more difficult pldiagnostic dilemmas is dependent on your clinical assessment (that is, mthe pretest probability) as modified by the sensitivity and specificity oftest results. When these test results are abnormal but do not support the sa
  23. 23. 4 PART 1: GENERAL ISSUESprovisional diagnosis, they do not of themselves become the problem. Theclinical scenario is reconsidered by a further meticulous history and physicalexamination, which will often widen the differential diagnosis, encompassingthe tests already done and perhaps suggesting further tests to enhance theprobability of a diagnosis. The isolated discordant laboratory test(s) will thenbe incorporated into the final diagnosis or will need to be considered as aspecial pathological problem. Patients and physicians want immediate answers but many diseasesare subtle and at initial presentation are often incompletely developed.While there are clinical circumstances which demand immediate action onincomplete evidence (e.g. suspected bacterial meningitis), it is better wherepossible to allow time and observation to assist in solving the problem.The quick, dogmatic incorrect diagnosis closes off the thinking process.Honesty requires that we admit difficulty and doubt so that over time carefulobservation, with judicious laboratory tests, can give a more accurate finalconclusion than the apparently brilliant guess. Advances in the accuracy and range of laboratory investigations haveenabled earlier and more exact definition of many clinical problems. Indeedmodern medicine is dominated by X-ray reports, pathological profiles andtests, followed by more tests. The tests are interpreted by highly trainedspecialists, often unfortunately with no available clinical information, sothe clinician should never believe these interpretations unless they can beintegrated into the clinical assessment. Screening tests not linked to symptomsor a specific hypothesis pose a special problem. How can the clinicianinterpret and integrate such abnormal results? A good history, exploring suchthings as genetic predisposition and lifestyle risk factors, will often permitthis integration. Laboratory test results purely as numbers are prone to error. For example,a ‘fasting glucose’ taken after a fruit juice (not considered as food by manypatients and therefore not reported by them) can give an incorrect diagnosis lyof diabetes mellitus. A urine sample to assess an albumin leak will give a onspuriously high result if collected after heavy exercise or during menstruation.A normal glycated haemoglobin (HbA1c) does not mean good diabetic scontrol if the red cell life span is shortened by, for example, haemolytic geanaemia or slow blood loss. The clinician must know the limitations of testsordered and the common causes of inaccurate results and must gather pasupporting evidence before acting on the result. When there is discordance adiscussion with the laboratory expert can often be beneficial. e Patients commonly present with non-specific tiredness, exhaustion and pllethargy, and the differential diagnosis is wide. A series of tests is orderedand the thyroid stimulating hormone (TSH) is found to be slightly elevated mat, say, 6 (range 0.4–4). A possible answer is apparent—but no, the test sa
  24. 24. CHAPTER 1: WHAT TO DO ABOUT ABNORMAL LABORATORY RESULTS 5result must answer the question and a TSH of 6 does not equate with majorsymptoms—so the clinician must think again. K CK FLIC The experts in the following chapters will discuss the range, specificityand sensitivity of their laboratory tests, which is the other half of the 1 UIequation. Read on and enjoy the challenges. Q ly on s ge pa e pl m sa
  25. 25. 4 Pitfalls in interpreting laboratory results P. Phillips Synopsis The results of laboratory tests are affected by the collection and handling of the specimen, the particular laboratory and the method of analysis. They are also affected by variability within the individual and within the laboratory. Interpretation at one point in time should consider the position of the measurement within the laboratory reference range appropriate for the sample and the person being tested. Interpreting results over time should consider the likely variability of the measurement and the level of certainty required to identify a true change or absence of change. The more variable the measurement and the higher the required level of certainty, the larger the change between measurements needs to be before it can be considered clinically significant. IntroductionHealth professionals may find it hard to get clinically useful information fromthe barrage of figures, ranges, asterisks and comments in laboratory results. lySome knowledge about the accuracy of laboratory results can help to sort onout important clinical signals from the background noise. The laboratorydoes not know all the patient’s details. Clinicians should consider test resultsin the context of the clinical presentation and not rely completely on the slaboratory’s interpretation. ge Reference ranges paQuoted reference ranges depend on the method used in the laboratory and ethe population from which the reference range was derived. The results plfrom one method may be systematically different from those of anotherand therefore the reference ranges will be different. m sa
  26. 26. CHAPTER 4: PITFALLS IN INTERPRETING LABORATORY RESULTS 15 Some laboratories give the range quoted by the manufacturer of the testor derived from an easily accessible population such as blood donors. Othersgive ranges in terms of age, sex or biological phase. For example, the rangesquoted for female sex hormones are related to pre- and post-menopausalstatus and the phase of the menstrual cycle. Some important biologicalinfluences, such as seasonal effects on 25-hydroxyvitamin D, are often notincluded in the reference ranges. Perhaps this is because users would find itharder to interpret results if the reference ranges were changing all the timeand because of the logistics and laboratory workload needed to derive suchspecific reference ranges. The ideal reference range would relate to the individual being testedwhile healthy, at the same age and in the same biological phase and season.Clearly this is not possible but sometimes one gets insights from looking backthrough previous results (ideally reported by the same laboratory using the Ksame method). CK FLIC By tradition, laboratories quote a reference range that includes 95% 4of the reference population. If the results are normally distributed, this UI Qincludes results within approximately two standard deviations aboveand two standard deviations below the mean value. The reference rangetherefore covers four standard deviations. Some results vary so much withinthe population that the laboratory may quote a reference range that includesa smaller proportion of the population. For example, the reference rangecommonly quoted for serum insulin may include only results within onestandard deviation above and one standard deviation below the mean value.This covers 68% of the reference population. In this case, 16% of normalpeople will have ‘abnormal’ high insulin and 16% will have ‘abnormal’ lowinsulin according to the quoted reference range. Serum insulin is thereforenot a useful test for assessing ‘insulin resistance’. Results have to be interpreted in terms of the particular laboratory’sreference range. When monitoring results over time, clinicians also need to lybe aware that different laboratories will have different reference ranges. on As reference ranges are population based, a patient might have a resultnear the top or bottom of the normal range. Clinically significant changes scould then occur without the results moving out of the population reference gerange. For example, if an elderly patient’s plasma creatinine concentrationis usually near the bottom of the reference range but then rises to the upper paend of that range, the patient may have had a significant deterioration inrenal function. Similar considerations apply to a haemoglobin concentration efalling from a high normal to a low normal value. pl m sa
  27. 27. 16 PART 1: GENERAL ISSUES Specimen collection and handlingLaboratory results can be affected by the procedures for specimen collectionand handling (Table 4.1).1 If a result is a surprise, check the patient’s nameand date of birth on the result report. You can also contact the laboratory andask if the specimen looked normal and consider repeating the test.Table 4.1 Abnormal laboratory results caused by incorrect collection andhandling† MeasurementStep Mechanism Result affectedSample Incorrect sample Incorrect results For example, random spot urine calcium:creatinine ratio instead of first voidedVenepuncture Prolonged Plasma filtration Protein venostasis and concentration concentrations— globulins, albumins and lipoproteins and measurements affected by them (e.g. calcium) Difficult Haemolysis Red cell leakage venepuncture with high potassium, phosphate and lactate dehydrogenaseSpecimen tube Incorrect collection Assay affected • If potassium tube EDTA used ly for chemistry— potassium; on calcium and enzymes (calcium binding s and enzyme ge inhibition) • Lithium heparin pa anticoagulant— lithium assay e pl m sa
  28. 28. CHAPTER 4: PITFALLS IN INTERPRETING LABORATORY RESULTS 17 MeasurementStep Mechanism Result affectedSpecimen handling Delay in transport Red cell use Blood glucose of glucose and (if fluoride tube not leakage of contents used) Potassium, bicarbonate, phosphate, lactate dehydrogenase Specimen Incorrect results Virtually everything mislabellingLaboratory Machine Incorrect results Virtually everything malfunction Transcription error K CK FLIC†Derived from reference 1 4 UI Why normal people often have abnormal results QA multiple biochemical analysis can be performed by one machine andproduce 20 results. Assuming these results were all independent of eachother (which they are not) and that results from the reference population arenormally distributed (which they may not be), only 36% of normal peoplewill have all 20 results in the reference range and 64% will have at least oneabnormal result (Table 4.2). However, the more abnormal the result and themore related tests are abnormal, the more likely the abnormality is clinicallysignificant.Table 4.2 Normal results in normal people lyIf the reference range covers 95% of results for a normal population, the chance of ahealthy individual having a certain number of normal tests is: on• Two out of two tests: 0.95 0.95 = 90%• All 20 of 20 tests: 0.9520 = 36% s geIf you consider the 99% reference range (approximately ± 2.6 standarddeviations) and the 99.9% reference range (approximately ± 3.3 standard padeviations), 82% and 98% of people will have all 20 tests within thereference range (0.9920 and 0.99920, respectively). These facts can be useful ewhen interpreting an isolated abnormal result. pl For example, the reference range of alkaline phosphatase is 30–110 U/L.This covers two standard deviations below the mean and two above the mmean. One standard deviation is therefore 20 U/L [(110 − 30) ÷ 4]. A result saof 150 U/L is two standard deviations above the upper limit of the reference
  29. 29. 18 PART 1: GENERAL ISSUESrange and therefore four standard deviations above the mean. This is veryunlikely to occur in a normal individual. However, the result may be normalif the quoted reference range is inappropriate. For example, in pregnancyand growing children alkaline phosphatase is produced by the placenta andbone. These are good examples of why it is important to consider whetherthe population reference range is appropriate for the individual being tested. When deciding if a result is abnormal, look at related tests. Alkalinephosphatase is one of the liver function tests (others are bilirubin, gammaglutamyl transferase, alanine aminotransferase, aspartate aminotransferaseand lactate dehydrogenase). Abnormalities in the other tests would suggestthat the abnormal alkaline phosphatase could be the result of liver disease.An elevated alkaline phosphatase in isolation may indicate another problem,such as bone disease. Laboratory accuracyWe often know the within-laboratory, within-method variability as this isusually quoted by the laboratory. Modern laboratories provide remarkablyconsistent results for many analytes—typical coefficients of variation (seeTable 4.3) are 1–6% for the components of multiple biochemical analysis,electrolytes, calcium and phosphorus and renal and liver function tests.Table 4.3 Coefficient of variationThe coefficient of variation (CV) is calculated as:CV = standard deviation of the measured value ÷ mean value of the measurements 100Variability is different at different absolute values of the measurement and is usuallyquoted at a specific clinically relevant value. For example:CV for plasma sodium: 0.8% at 139 mmol/L lyCV for plasma bilirubin: 6.1% at 10 micromol/L onThe coefficient of variation is one way of expressing the variability of biologicalmeasurements. Laboratories sometimes also refer to the imprecision of ameasurement. s geNational quality control programs monitor the accuracy and imprecision ofdifferent methods used in different laboratories. One result has been that the padifferences between laboratories for individual methods are now usually asmall component of the overall variability of measurements. e pl m sa
  30. 30. CHAPTER 4: PITFALLS IN INTERPRETING LABORATORY RESULTS 19 Why values vary within one individualIn addition to the variations caused by specimen collection and handling andthe differences within and between laboratories and their methods, thereis intra-individual variation. Assuming specimen collection and processingerrors do not occur, the largest source of variability is within the individual.Values vary by age and sex and within the menstrual, diurnal and seasonalcycles. Intra-individual biological variability for different analytes can rangefrom very large to moderate, for example, 8% for total cholesterol2 versus40% for microalbuminuria assessed by the albumin:creatinine ratio3.In addition, the longer the interval between tests, the greater the totalintra-individual variability of the measure. It is much more difficult for laboratories to provide information onthe total intra-individual variability than for the within-laboratory, within-method variability, which is automatically generated by their quality control K CK FLICprograms. However, it is the total variability within an individual which isimportant when interpreting results. 4 UI Q Are changes in results caused by intra-individual variability or the effects of treatment?One trap is the phenomenon of regression to the mean.4 Results within anapparently homogeneous group of patients are likely to lie within the 95%reference range for that measurement. If the same patients are re-tested ata different time, the pattern of the overall results will look much the same.In a normal distribution, values are bunched around the group mean andprogressively ‘thin out’ further from the mean. However, individual resultsare likely to have changed, particularly those at the extremes. The initial results at the extremes are the result of extreme randomvariability in one direction or the other. The same amount and directionof variability is unlikely to occur on the second measurement in the lysame individual. Subsequent measurements will therefore move closer to onthe middle (or regress to the mean). Results from other individuals whoinitially were closer to the mean may now lie closer to the extremes of sthe distribution. ge This phenomenon can be exploited intentionally or unintentionally intrials that select and treat individuals with high values of a measurement pato demonstrate that a treatment is effective. Regression to the mean is onereason why randomised placebo controlled prospective trials are the gold estandard for assessing treatments. pl A large difference between two measurements is more likely to bea signal of a true change than the result of the background noise of mmeasurement variability. Similarly, the smaller the total intra-individual sa
  31. 31. 20 PART 1: GENERAL ISSUESvariability, the more likely a specific absolute change is a signal. The less likelythe observed change is caused by variability, the surer one can be that thechange is real. These three elements are brought together in the concept of the leastsignificant change. To be 80% confident the observed change is real, thechange should exceed approximately twice the intra-individual coefficientof variation (CVi) (Table 4.4). For example:• A total cholesterol that decreases from 7.0 to 5.6 mmol/L after starting a statin is a 20% fall from the initial value. The CVi for total cholesterol is 8% so the least significant change is approximately 16% (2CVi). You can be 80% sure that the 20% change is real rather than apparent.• A decrease in microalbuminuria from an albumin:creatinine ratio of 5.0 mg/mmol to 2.0 mg/mmol after starting an angiotensin converting enzyme (ACE) inhibitor is a 60% fall. The total CVi of the albumin:creatinine ratio is 40% so the least significant change is approximately 80% (2CVi). It is likely that this 60% change is apparent rather than real.Table 4.4 When is a change significant?Least significant change1. The overall variability of the difference between two measurements is greater than the variability of the individual measurements: √2 CVi.2. The more confident one wishes to be that the change in a measurement is a signal rather than noise, the greater the change needs to be relative to this: √2 CVi z. The z value is used to refer to normally distributed values and describes the distance of a particular value from the mean in numbers of standard deviations (SD). The greater the distance from the mean (the z value) the less likely a result has occurred by chance. z varies from 1.28 for 80% confidence to 2.6 for 99% confidence.3. Generally an 80% confidence is used (z = 1.28): ly Least significant change = √2 CVi 1.28 = 1.8 CVi. This approximates to 2CVi. onIntra-individual coefficient of variation (CVi)For a list of intra-individual CVi for different analytes, see reference 5. s geVariability of the difference between two measurementsCVi1 = intra-individual coefficient of variation for the first measurement paCVi2 = intra-individual coefficient of variation for the second measurementThe variability of the difference between two measurements is √CVi12 + CVi22.If CVi1 = CVi2 (as measuring the same variable), then CVi12 + CVi22 = 2CVi12, so the evariability of difference is √2CVi12 = √2 √CVi12 = √2CVi1. pl m sa
  32. 32. CHAPTER 4: PITFALLS IN INTERPRETING LABORATORY RESULTS 21 The effects of treatment on measurements may be delayedLaboratory results may take a long time to change after starting treatment.This may reflect pharmacokinetics, biology or a combination of the two. The half-life of thyroxine in the body is approximately seven days. Testingafter one week will only show half the expected total effect. (This maysometimes still be useful information.) By six weeks (six half-lives in this case)98.4% of the effect will have occurred [1 – (½)6]. When starting a thiazolidinedione (glitazone) the full effect on bloodglucose requires a steady state of the drug (pharmacokinetic effect) butalso requires the shift in fat metabolism that in turn causes the reduction inglucose (biologic effect). Finally, the glycated haemoglobin (HbA1c) reflectsthe average blood glucose over the preceding 4–6 weeks because of theslow turnover of the red cells (biologic and pharmacokinetic effects).6 The K CK FLICcombination of these factors means that testing after 1 week of treatmentmay show little change in the HbA1c and that 2–3 months may be needed 4 UIto show the full effect of treatment. Q Another glycated protein (albumin, which becomes fructosamine) has amuch faster turnover. It therefore reflects the average glucose over a shorterperiod (2–3 weeks). One can reduce the variability of the measurement change by reducingthe variability of the baseline and final measurements (e.g. using the meanof two measurements for each). If both initial and final measurements wererepeated the variability of the change would be reduced to CVi (not √2CVi). Using the microalbuminuria example, with two measurements beforeand after the intervention, the least significant change would be 51%(1.28 40%). You could then be 80% sure that the 60% observed changewas real and not apparent. ly Summary onWhen interpreting laboratory results it is important to know that the samplewas collected and handled correctly. The appropriate reference range for the stest should be used. Different laboratories may report different results on the gesame specimen. When comparing results over time, use the same laboratory and pamethod for testing. Consider the variability of results within the individualand the least significant change, that is, the amount of difference between emeasurements that is likely to be a real biological signal instead of resulting plfrom the noise of biological variability within the individual and within theend measurement variability of the laboratory. As a rough rule, the least msignificant change is twice the intra-individual coefficient of variation (2CVi). sa
  33. 33. 22 PART 1: GENERAL ISSUES If an important clinical decision depends on whether a change occurswith a particular treatment, consider making two (or more) measurementsbefore and after starting treatment. This reduces the variability and thepossibility of misinterpreting the regression to the mean of an initial highor low value. Monitoring trends with time involves more measurementsand gives a more reliable indication of change than a single comparison attwo points. Remember, the more tests you do the more likely you are to get at leastone false positive outside the laboratory reference range. Aim to limit thenumber of tests to those that are relevant to the clinical situation rather thanrequesting a screening battery. When assessing the effects of treatment, consider how long the treatmentwill take before the therapeutic effect reaches a steady state (e.g. 4–6 half-lives of a drug) and how long the biological response will take before themeasurement you make reaches a steady state. Trying to assess therapeuticeffects before the response to treatment has reached a steady state canseriously underestimate the therapeutic effect. References1. Phillips P and Beng C. Electrolytes—‘fun with fluids’. Check (Continuous . Home Evaluation of Clinical Knowledge) program of self assessment. No. 323. South Melbourne: Royal Australian College of General Practitioners, 1999.2. Cooper G.R., Myers G.L., Smith S.J. and Schlant R.C. Blood lipid measurements. Variations and practical utility. JAMA 1992; 267: 1652–60.3. Phillipou G. and Phillips P Variability of urinary albumin excretion in .J. patients with microalbuminuria. Diabetes Care 1994; 17: 425–7.4. Irwig L., Glasziou P Wilson A. and Macaskill P Estimating an individual’s ., . true cholesterol level and response to intervention. JAMA 1991; 266: ly 1678–85. on5. Desirable specifications for total error, imprecision and bias, derived from intra- and inter-individual biologic variation. Updated 2010 Feb 15, s [cited Mar 3, 2010]. ge6. Phillipov G. and Phillips P Components of total measurement error for .J. haemoglobin A1c determination. Clin Chem 2001; 47: 1851–3. pa e pl m sa