Routine preoperative testing - Second consultation, Full ... - NICE

DRAFT FOR CONSULTATION

National Collaborating Centre for Acute Care at The Royal College of Surgeons of England, 34-43 Lincoln’s Inn Fields, London WC2A 3PE

T: 020 7869 6630 F: 020 7869 6639

Draft clinical guideline

Routine preoperative testing

Preoperative testing (second consultation)

Page 1 of 203

DRAFT FOR CONSULTATION Abbreviations and glossary of terms Guideline development Group Membership and acknowledgements Guideline Development Group Membership and Acknowledgements

10

1 Introduction

14

1.1 Background ..............................................................................................................14 1.2 Objectives of the guideline ......................................................................................16 1.3 The purposes of these guidelines .............................................................................17 1.4 Who has developed the guideline? ..........................................................................17 1.5 Scope of the guideline..............................................................................................18 2 Methods

22

2.1 Outline of methods used ..........................................................................................22 2.2 Assembly of an advisory group ...............................................................................24 2.3 Aims and scope of the guideline..............................................................................24 2.4 Systematic review methods......................................................................................27 2.5 Economic review of the cost of tests included in the guideline...............................32 2.6 Choice of development method by formal consensus .............................................32 2.7 Testing the assumptions: informal consensus interviews with surgical trainees and consultant anaesthetists ...........................................................................................33 2.8 Consensus: selection of expert panel .......................................................................34 2.9 Development of the consensus questionnaire ..........................................................35 2.10 Consensus: Round 1 – initial rating .........................................................................37 2.11 Consensus: Round 2 – meeting................................................................................37 2.12 Drafting the guideline (Chapter 8)...........................................................................39 2.13 Comments sought from a wider audience................................................................39 2.14 Adjustments made in light of comments .................................................................39 3 Findings from the systematic review of the literature


41

Page 2 of 203

DRAFT FOR CONSULTATION 3.1 Papers identified for the review ...............................................................................41 3.2 Preoperative chest radiographs ................................................................................46 3.3 Preoperative electrocardiograms..............................................................................48 3.4 Preoperative haemoglobin, haematocrit and full blood count tests .........................50 3.5 Preoperative haemostasis tests.................................................................................52 3.6 Preoperative biochemistry tests ...............................................................................54 3.7 Preoperative urine tests ............................................................................................56 3.8 Preoperative pregnancy tests....................................................................................58 3.9 Preoperative sickle cell disease/trait tests ................................................................59 3.10 Preoperative lung function tests...............................................................................59 3.11 Preoperative blood gas testing .................................................................................61 4 Interview results: testing the assumptions

63

4.1 Interview part 1: Reasons for testing .......................................................................63 4.2 Interview part 2: Physiological models....................................................................69 4.3 Interview part 3: Sickle cell testing .........................................................................73 4.4 Interview part 4: Pregnancy testing .........................................................................74 5 Consensus results

76

5.1 Identification and make up of consensus panels......................................................76 5.2 Development of the consensus questionnaire ..........................................................76 5.3 First round of consensus ..........................................................................................76 5.4 Second round: Consensus meetings.........................................................................77 5.5 Presentation of results ..............................................................................................77 5.6 General comments and assumptions made by the panellists ...................................78 5.7 Table of definitions of consensus and key to the following results tables...............79 5.8 Chest x-ray (radiograph) ..........................................................................................79


Page 3 of 203

DRAFT FOR CONSULTATION 5.9 Resting electrocardiogram (ECG)............................................................................84 5.10 Full blood counts......................................................................................................87 5.11 Haemostasis tests .....................................................................................................91 5.12 Renal function tests..................................................................................................95 5.13 Random blood glucose tests.....................................................................................98 5.14 Urine analysis for ASA 1 patients .........................................................................102 5.15 Testing of blood gases (ASA grade 2 and 3 only) .................................................106 5.16 Lung function tests.................................................................................................108 5.17 Adults and children: tests for sickle cell disease/trait............................................110 5.18 Pregnancy testing ...................................................................................................111 6 Patient/carer perspective

112

6.1 Selection of panel members ...................................................................................112 6.2 How patient/carer representatives were involved ..................................................112 6.3 Meeting to elicit comments and observations from the patient perspective..........113 6.4 Other involvement of patient representatives ........................................................113 6.5 Overall views of patient representatives ................................................................113 6.6 Issues......................................................................................................................114 6.7 Patient representation.............................................................................................117 7 Economics of routine preoperative testing

118

7.1 Introduction............................................................................................................118 7.2 Background ............................................................................................................118 7.3 Methods..................................................................................................................122 7.4 Results....................................................................................................................129 7.5 Discussion ..............................................................................................................137 Annex: Unit costs - Data collection form .......................................................................164


Page 4 of 203

DRAFT FOR CONSULTATION 8 Guidelines for preoperative investigations in patients undergoing elective surgery171 8.1 Types of recommendation......................................................................................171 8.2 Level of evidence and grading for recommendations............................................172 8.3 How to use this guideline.......................................................................................173 8.4 ASA Grade 1 Adults (age ≥ 16 years) ...................................................................179 8.5 ASA Grade 1 Children (age < 16 years)................................................................180 8.6 ASA Grade 2 Patients with cardiovascular disease comorbidity undergoing elective surgery ...................................................................................................................181 8.7 ASA Grade 3 Patients with cardiovascular disease comorbidity undergoing elective surgery ...................................................................................................................182 8.8 ASA grade 2 Patients with comorbidity from respiratory disease.........................183 8.9 ASA grade 3 Patients with comorbidity from respiratory disease.........................184 8.10 ASA grade 2 Patients with comorbidity from renal disease ..................................185 8.11 ASA grade 3 Patients with comorbidity from renal disease ..................................186 8.12 Tests for sickle cell disease/trait adults and children.............................................187 8.13 Pregnancy testing ...................................................................................................187 8.14 Good practice recommendations............................................................................188 8.15 Implementation in the NHS ...................................................................................190 8.16 Audit criteria ..........................................................................................................190 9 Discussion and research recommendations

193

9.2 Factors not considered in the development of the guideline..................................198 9.3 Research recommendations ...................................................................................201


Page 5 of 203


Appendices Appendix 1

Search strategy

Appendix 2

Search results

Appendix 3

Data extraction form

Appendix 4

Quality of case series form

Appendix 5

Interview format

Appendix 6

Phase I consensus questionnaire (results)

Appendix 7

Phase II consensus questionnaire

Appendix 8

Phase I consensus statements

Appendix 9

Phase I consensus table

Appendix 10

Phase I consensus charts

Appendix 11

Phase II consensus statements

Appendix 12

Phase II consensus table

Appendix 13

Phase II consensus charts

Appendix 14

Laboratory costs

Also available

Text of systematic review findings


Page 6 of 203


Abbreviations and Glossary of Terms Glossary of Terms Term

Definition

Generic (routine) testing

Testing carried out for all patients who satisfy certain criteria and that is not directly related to the planned procedure. For example, carrying out ECGs in all patients with minor co-morbidity (i.e. ASA grade 2) over the age of 75 years would constitute generic testing. However, carrying out ECGs in patients with minor co-morbidity undergoing cardiac surgery would not constitute generic testing, because the test is related to the planned surgery. (Note: we have adopted the word ‘generic’ because of the ambiguity of the term ‘routine’ but we acknowledge that this definition may become difficult to apply for patients with comorbidity, when it may be difficult to distinguish whether a test is ‘indicated’ because of the comorbidity or because of the planned procedure. The guidelines are primarily interested in tests that are appropriate and not appropriate given a patients comorbidity, severity of comorbidity, and planned surgical procedure.

ASA grades

ASA stands for American Society of Anesthesiologists. ASA grades are a simple scale describing fitness to undergo an anaesthetic. The American Society of Anesthesiologists clearly states that it does not endorse any elaboration of these definitions. However, anaesthetists in the UK often qualify (or interpret) these grades as relating to functional capacity, i.e. co-morbidity that does (ASA grade 3) or that does not (ASA grade 2) limit a patient’s activity. Table 2.3 in Chapter Two attempts to characterise signs and symptoms associated with grades 2 and 3, with no intention of opting for a particular definition. Panellists were asked to use their own interpretation of the ASA grades if they used them habitually and to refer to the table for illustration of the ‘kinds of patients’ likely to be classified in each category ASA grade 1 “normal healthy patient” i.e. without any clinically important co-morbidity and without a clinically significant past/present medical history. ASA grade 2 “A patient with mild systemic disease” ASA grade 3 “A patient with severe systemic disease” but not a constant threat to life. (Severe systemic disease that is a constant threat to life moves a patient from ASA grade 3 to ASA grade 4.)

Elective

Scheduled procedure, i.e. NOT an urgent or emergency procedure.


Page 7 of 203

DRAFT FOR CONSULTATION Severity of surgery grades

An operation represents a physiological stress. The magnitude of the physiological stress increases with the ‘invasiveness’ of the procedure. We have not been able to identify any widely accepted and validated system for classifying the stressfulness of operative procedures. We have therefore adopted a simple graded scale, which we have illustrated with examples Grade 1 (minor)

Excision of lesion of skin; drainage of breast abscess

Grade 2 (intermediate) Primary repair of inguinal hernia; excision of varicose vein(s) of leg; tonsillectomy/ adenotonsillectomy; knee arthroscopy Grade 3 (major) Grade 4 (major+)

Total abdominal hysterectomy; endoscopic resection of prostate; lumbar discectomy; thyroidectomy Total joint replacement; artery reconstruction; Colonic resection; radical neck dissection; neuro-surgery; cardiac surgery

(in Phase I, for ASA grade 1 patients, panellists were asked to consider neurosurgery and cardiac surgery separately, these categories of surgery were combined in grade 4 for Phase 2) A more extensive list is shown in Chapter 7 Type of anaesthetic

We have assumed that the type of anaesthetic used for a procedure does not influence decisions about whether generic pre-operative tests should be carried out. The reason for this assumption is that an anaesthetist always has to be ready to use a general anaesthetic, in the event of a complication when using a regional anaesthetic.

Costs of tests

The decision to implement generic pre-operative testing may be influenced by the cost or cost-effectiveness of tests, e.g. the cost per major complication or death avoided by pre-operative testing. We do not have high quality estimates of the number of serious events avoided by different tests. We have, however, been able to calculate the approximate costs of the tests (lower, mid and upper estimates) and these are reported on the next page and in the document alongside questions about each test.

Nonnorthern European ethnicity

The consensus questionnaire includes questions about the appropriateness of sickle cell testing in different ethnic populations. The descriptions chosen are intended to represent all of the main ethnic groups. Separate questions address the issue of how to determine ethnicity and what factors, in practice, prompt testing for sickle cell disease / trait.

Phase I and Phase II

The population of interest was split into two phases: Phase I: all uncomplicated, ASA grade 1 Children or adults, undertaking elective surgery. Phase II: all adult patients with comorbidity (cardiovascular, respiratory or renal) ASA grade 2 and 3, undertaking elective surgery.


Page 8 of 203

DRAFT FOR CONSULTATION Abbreviations GDG

Guideline Development Group

NICE

National Institute for Clinical Excellence

Chest x-ray

Chest radiograph

APTT

Activated Partial Thromboplastin Time

PT

Prothrombin Time

COAD

Chronic Obstructive Airways Disease

COPD

Chronic Obstructive Pulmonary Disease

CVD

Cardiovascular disease

HTA

Health Technology Assessment


Page 9 of 203


Guideline Development Group Membership and Acknowledgements Guideline Development Group Academic staff (National Collaborating Centre for Acute Care and London School Hygiene and Tropical Medicine) Chair

Dr Barnaby Reeves , Senior Lecturer in Epidemiology

Project Manager

Miss Julia Langham, Research Fellow in Epidemiology

Systematic Reviewer

Miss Nirree Phillips, Research Fellow

Information Scientist

Mr Carlos Sharpin, Information Scientist

Health Economist

Mr David Wonderling, Health Economist

Clinical Specialists Royal College Surgeons

Mr Mark Emberton, Senior Lecturer in Oncological Urology/Honorary Consultant in Urology, Institute of Urology, Royal Free and University College of London, London.

Royal College Anaesthetists

Dr Gavin Thoms, Consultant Anaesthetist, Director, Evaluation and Audit Unit, Manchester Royal Infirmary, Manchester.

Royal College Radiologists

Dr Paul Taylor, Clinical Director of Radiology, Manchester Royal Infirmary, Manchester.

Royal College Pathologists

Dr Danielle Freedman, Consultant Chemical Pathologist, Luton and Dunstable Hospital, Luton.


Dr Mike Galloway, Consultant Haematologist, Sunderland Royal Hospital, Sunderland.

Additional Expert Advisory members to the GDG representing professional stakeholder groups Trainee surgeon

Mr Sanjaya Wijeyekoon, Surgical Research Fellow, Royal College of Surgeons, London

Royal College of Ophthalmologists

Mr Hamish Towler, Consultant Opthalmologist, Whipps Cross Hospital, London

Consultant nurse /clinical nurse Ms Dorothy Weeden, General Manager - Surgery, The North practitioner Middlesex Hospital, London. Consultant Anaesthetist

Dr John Carlise, Consultant Anaesthetist, Torbay

Cardiologist representative

Dr Bruce E Keogh, Consultant in Cardiothoracic Surgery, Queen


Page 10 of 203

DRAFT FOR CONSULTATION Elizabeth Medical Centre, Birmingham (retired from group) Royal College of Obstetricians and Gynaecologists

Dr Jane Thomas, Director Clinical Effectiveness Support Unit, Royal College of Obstetricians and Gynaecologists, London (retired from group)

Four representatives from the patient liaison groups of the following Royal Collages:

Royal College Anaesthetists

Mrs Ann Seymour

Royal College Radiologists

Mrs Barbara Greggains


Mr Alan Wright

Royal College Surgeons

Mrs Daphne McKenzie (retired from group due to ill health)

NICE, Guidelines Commissioning Manager Observer

Ms Christine Sealey, Guidelines Commissioning Manager, National Institute for Clinical Excellence, London

Consensus panel members Dr Alison Bliss, Consultant in Paesiatric Anaesthesia, St James's University Hospital, Leeds Dr Andy Hartland, Consultant Chemical Pathologist, NW London NHS Trust. Dr Eric Watts, Consultant Haematologist, Basildon Hospital, Basildon Dr Gwyn McCreanor, Consultant Clinical Biochemist, Kettering General Hospital, Northants Dr Isabeau Walker, Consultant Paediatric Anaesthetist, Great Ormond Street Hsopital, London. Dr Natasha Robinson, Consultant Anaesthetist, Northampton General Hospital, Northampton Dr Ndu Okonkwo, Consultant Anaesthetist, Nottingham City Hospital, Nottingham. Dr Nicola Barham, Consultant anaesthetist, James Cook University Hospital, Middlesbrough Dr Bob Haden, Consultant anaesthetist, The Alexandra Hospital, Redditch Dr RJS Chinn, Consultant Radiologist, Chelsea & Westminster Hospital, London Dr Sam Kaddoura, Consultant in Cardiology, Chelsea & Westminster Hospital, London Dr Sophie Jayamaha, Consultant anaesthetist, Grantham Hospital, Grantham Dr Su Underwood, Consultant anaesthetist, Bristol Royal Infirmary, Bristol Dr Zor Maung, Consultant Haematologist, University Hospital of North Tees Miss Pari Mohanna, Surgical Research Fellow, London Mr Charles S W Wright, Consultant in Obstetrics & Gynaecology, Hillingdon Hospital, Middlesex


Page 11 of 203

DRAFT FOR CONSULTATION Mr Christian Brown, Surgical Research Fellow, Clinical Effectiveness Unit, Royal College of Surgeons, England Mr Indie Singh, Surgical Research Fellow, St. Thomas' Hospital, London Mr Sigi Jospeph, General Surgeon, London School Hygiene and Tropical Medicine Mrs Celia Ingham Clark, Consultant in General Surgery, Whittington Hospital, London Ms Jackie Sambrook, Specialist Nurse, Whittington Hospital trust, London Ms Penny Roberts, Gynaecology Nurse Specialist, Whittington hospital trust, London

Interviewees (testing the assumptions) Surgical Trainees: Mr Christian Brown, Surgical Research Fellow, Clinical Effectiveness Unit, Royal College of Surgeons, England Mr Imran Saeed, Surgical Research Fellow, Clinical Effectiveness Unit, Royal College of Surgeons, England Mr Joel Desmond, Surgical Research Fellow, Clinical Effectiveness Unit, Royal College of Surgeons, England Miss Katie Grace, Surgical Research Fellow, Mr Paris Tekkis, Surgical Research Fellow, Hunterian Professor and RCS Research Fellow, Kings College Hospital, London Miss Pari Mohanna, Surgical Research Fellow, London Mr Ganesh Kuhan, Surgical Research Fellow, Mr Frederick Banks, Surgical Research Fellow, Mr Jeremy Cundall, MRCS, Colorectal Research Fellow, South Yorkshire Mr Henry Lazarowicz, Surgical Research Fellow, Mr Santbir Mehta, Surgical Research Fellow, RCSEng Research Fellow, Royal Hallamshire Hospital, Sheffield Mr Marcus Bisson, Surgical Research Fellow, The RAFT Institute, Middlesex Consultant Anaesthetists: Dr Bruce Ferguson, Consultant Anaesthetist, Bro Morgannwg NHS Trust Dr Paul Harvey, Consultant Anaesthetist, Derriford Hospital, Plymouth Dr Andy Dennis, Consultant Anaesthetist, Northern General Hospital, Sheffield Dr Paul Upton, Consultant Anaesthetist, Royal Cornwall Hospital, Truro


Page 12 of 203

DRAFT FOR CONSULTATION Dr Andy Black, Consultant Anaesthetist, Royal Bristol Infirmary, Bristol. Dr Andy Kilner, Consultant Anaesthetist, The Freeman Hospital,Newcastle upon Tyne. Dr Peter Sanderson, Consultant Anaesthetist, ICU, Gloucestershire Royal Hospital Dr John Tomlinson, Consultant Anaesthetist, New Cross Hospital, Wolverhampton. Prof Chris Dodd, Consultant Anaesthetist, James Cook University Hospital, Middlesbrough Dr Kathy Wilkinson, Consultant Paediatric Anaesthetist, Norfolk & Norwich Hospital, Norwich.

Acknowledgements Many thanks to •

Professor Nick Black from the London School Hygiene & Tropical Medicine for advice about consensus methods and for facilitating all four consensus group meetings.

•

The authors of the HTA review 1 Mr James Munro, Mr Andrew Booth and Mr Jon Nicholl, Medical Care Research Unit, ScHARR, The University of Sheffield, who helped us get started on the systematic review by providing their search strategy, copies of the papers they had identified, and have commented various drafts.

•

all those who helped in the preparation of this guideline, in addition to those listed above, thanks to the three anaesthetists who kindly agreed to review these guidelines, Dr Vicki Webster, Dr John Curran, Dr Oliver Dearlove.

•

Dr Anthony Gray, Consultant in Anaesthesia, Norfolk and Norwich NHS Trust, for helping identify the Consultant anaesthetists who participated in our interviews.

•

Patient Liaison Groups who will help to produce the patient information guideline


Page 13 of 203


1 Introduction 1.1

Background Annually, over three million operations are performed in NHS hospitals. For many years it has been routine clinical practice to test apparently healthy patients preoperatively for unsuspected conditions that might affect their treatment. The main purpose of these investigations is to supplement clinical history and provide additional diagnostic and prognostic information with the aim of reducing the risk or increasing the benefit to patients by altering their management. Other benefits of preoperative testing also include, for example: 1. Provision of baseline information to confirm that the current course of clinical management is correct; 2. Prediction of postoperative complications; 3. Establishment of baseline measurement for later reference; 4. Opportunistic screening that is unrelated to the surgery. The clinical ‘value’ of testing apparently healthy individuals before their operation is, however, uncertain. The possible benefits of routine preoperative investigations include identification of unsuspected conditions that may require treatment before surgery, or a change in surgical or anaesthetic management. However, preoperative investigations may also cause harm, for example, from false positive findings and leading to unnecessary, costly and maybe even harmful treatments or further investigations and delays in surgery. 1 Also, some evidence suggests that clinicians may not change the management of their patients even in the light of true positive abnormal preoperative tests in healthy individuals.2 A Health Technology Assessment (HTA) systematic review was published in 1997 2

which assessed the evidence for the use of preoperative testing in healthy patients

admitted for elective surgery. The review covered preoperative chest radiographs, electrocardiography, haemoglobin and full blood counts, haemostasis tests, biochemistry tests and urine testing. The authors found a wide range of estimates of the proportion of patients with abnormal results in the literature they reviewed, even


Page 14 of 203

DRAFT FOR CONSULTATION in apparently health individuals. The clinical importance of these abnormal results was uncertain because overall, abnormal preoperative test results led only to a small proportion of changes in clinical management, and for some tests there were virtually never any changes in clinical management reported. The review concluded that a policy of preoperative testing in apparently healthy individuals is likely to lead to little benefit if any. This conclusion is supported by two earlier reviews, one from Sweden 3and the other from the Basque country.4 However, it is crucially important to have a definition of what a change in management is, to fully understand the potential value of preoperative investigations to the patient and the clinician. This is discussed in more detail in Chapter 2, Methods and in Chapter 9 (Discussion and limitations). Chest radiographs were the first preoperative investigation for which clinical guidance was issued in the UK. The guidance was published as part of a broader guideline (non-evidence based) document, “Making the best use of a Department of Clinical Radiology”, by the Royal College of Radiologists which is currently in the fourth edition 5. The guidelines recommend that preoperative chest radiographs are not indicated routinely and list a number of specific clinical indications. The American Society of Anesthesiologists (ASA) published a statement on routine preoperative laboratory and diagnostic screening6 which is consistent with the conclusion of the HTA review: “No routine laboratory or diagnostic screening is necessary for the preanaesthetic evaluation of patients.” Although the ASA recommends that departments should develop local guidelines for selected populations, ultimate responsibility is left to the clinical judgement of individual clinicians: “Individual anesthesiologists should order test(s) when, in their judgement, the results may influence decisions regarding risks and management of the anesthesia and surgery.” More recently the American Society of Anesthesiologists have published guidelines on pre-anaesthesia evaluation. The guideline concludes:


Page 15 of 203

DRAFT FOR CONSULTATION “routine preoperative tests (i.e., tests intended to discover a disease or disorder in an asymptomatic patient) do not make an important contribution to the process of perioperative assessment and management of the patient by the anaesthesiologist” 7 It states that selective pre operative testing , based on indications in the history and examination of patients, or type and invasiveness of the planned procedure and anaesthesia, may help patient management. However guidelines could not be unequivocally determined due to the lack of evidence available. In the context of clinical guidance for preoperative investigation for the majority of patients, these reviews demonstrate a consensus that preoperative testing is not necessary when there are no clinical indications. However, there remains the opportunity for widespread variation with respect to what individual anaesthetists and surgeons regard as relevant clinical indications. Moreover, given that preoperative testing is widely carried out, it is important to distinguish between a lack of evidence of important benefits of preoperative investigations and convincing evidence that preoperative investigations have no important benefits.

1.2

Objectives of the guideline Clinical guidelines are ‘systematically developed statements which assist clinicians and patients in making decisions about appropriate treatment for specific conditions’8. The recommendations included in this document were arrived at after careful consideration of the available evidence and consensus opinion of an expert panel and should be considered as guidelines only. Healthcare professionals involved in pre-, peri- and postoperative care must use their professional knowledge and judgement when applying the recommendations to the management of individual patients.


Page 16 of 203


1.3

The purposes of these guidelines •

To evaluate the evidence relating to the ‘value’ of routine preoperative testing in elective surgical patients and in selected groups of patients with comorbid conditions by carrying out a systematic review of the literature.

•

To develop guidance to clinicians on the use of preoperative investigations in ASA grade 1 patients and in ASA grade 2 and 3 patients with selected comorbidity

•

To produce an illustrative economic model investigating plausible rates of abnormal results, rates of changes in management, rates of postoperative complications avoided by preoperative investigations and the subsequent costs of preoperative investigations and of adverse events and their sequelae.

1.4

Who has developed the guideline? The development of the guideline was supported by funding from the National Institute for Clinical Excellence (NICE). The guideline was developed by a multi-professional and lay working group (the Guideline Development Group, GDG) and other experts who represented the main stakeholders in this guideline, convened by the National Collaborating Centre for Acute Care. Members included representatives from: •

Royal College of Surgeons

•

Royal College of Pathologists

•

Royal College of Anaesthetists

•

Royal College of Radiologists

•

Royal College of Obstetricians and Gynaecologists

•

The British Cardiac Society

•

Department of Public Health Policy, London School of Hygiene and Tropical Medicine

•

The Whittington Hospital, London

•

University Hospital, Cardiothoracic Surgical Unit, Birmingham

•

Patient representatives from the Royal Colleges.


Page 17 of 203

DRAFT FOR CONSULTATION Staff from the NICE National Collaborating Centre for Acute Care, the Clinical Effectiveness Unit at the Royal College of Surgeons and the Department of Public Health Policy at the London School of Hygiene and Tropical Medicine provided support and guidance with the guideline development process, undertook the systematic searches, retrieval and appraisal of the evidence and wrote the document.

1.5

Scope of the guideline

1.5.1

What the guideline covers Inclusions: the guideline is aimed mainly at secondary care, but may have relevance to some tests carried out or ordered in primary care. The preoperative tests considered are for the preassessment of patients ASA grade 1 (adults and children) and ASA grade 2 and 3 (adults only) undergoing elective surgery. The preoperative tests to be considered were agreed at the scoping meetings with the Guideline Development Group and are listed below in Table 1.1. Certain investigations, for example computed tomography (CT scan) of the thorax and Muga scans were excluded from the guidelines as they did not fulfil the criteria for a ’generic’ preoperative investigation. More information about included and excluded tests is included in Chapter 2 and the full version of the systematic review.


Page 18 of 203

DRAFT FOR CONSULTATION Table 1.1

The preoperative tests considered in this guideline

Test

Description

Chest x-ray

Plain chest x-ray

Resting ECG

Resting electrocardiography

Full blood count

Full blood count includes haemoglobin measurement, white blood cell count and platelet count. Using current laboratory automated analysers it is not possible to obtain only a haemoglobin measurement. It is possible to measure haemoglobin only with point of care testing but we have excluded this form of testing from the development of the guideline.

Haemostasis tests

Haemostasis tests include prothrombin time (PT), activated partial thromboplastin time (APTT) and international normalised ratio (INR; derived from the patient’s PT and normative data)

Renal function tests

Renal function tests include measurement of potassium, sodium and creatinine levels.

Blood glucose test

Blood glucose test is a measurement of glucose from a blood sample, as opposed to measurement in the urine using a urine ‘dipstick’ test.

Urine ‘dipstick’ test

Urine dipstick tests are manufactured to test for different conditions separately and together. Urine dipsticks test for pH, protein, glucose, ketones, blood/haemoglobin.

Sickle cell test

Testing for sickle cell disease/trait usually takes place in two stages. First a test used for rapid detection of sickle cell disease/trait (for example the Sickledex test), followed by more detailed (and more expensive) testing for other haemoglobinopathies if the first test is positive

Pregnancy test

Biochemical testing for pregnancy. In most cases this would be a urine test.

Blood gases (Phase II only)

Arterial/venous blood gas analysis (measure the pH (acidity), oxygen content, and carbon dioxide content of the blood) to evaluate respiratory diseases and determine the effectiveness of oxygen therapy. The acid-base component of the test also gives information on renal function.

Pulmonary function tests (Phase II only)

Pulmonary function tests include measurements of peak expiratory flow rate, Forced vital capacity, forced expiratory volume and maximum ventilatory volume


Page 19 of 203

DRAFT FOR CONSULTATION 1.5.2

What the guideline does not cover The GDG concluded that it was beyond the scope of this guideline to address: •

the appropriateness of innovative technologies that have yet to be established in NHS practice;

•

definitive advice on the assessment and wider clinical management of patients before surgery, for example, it is beyond the scope of the guidelines to address follow-up and treatment of conditions identified by preoperative tests;

•

the specific needs of patients with rare conditions, or those with complex or unusual comorbidities;

•

the diagnosis and management of specific diseases or conditions, follow up and treatment of conditions identified by the preoperative investigations, or specific needs of patients with rare conditions;

•

how tests should be carried out, frequency of testing or interpretation of test results (e.g., specific parameters and criteria for defining abnormalities);

• 1.5.3

other aspects of pre operative assessment such as history and examination.

Consent The issue of consent to undergo preoperative tests is addressed briefly in relation to specific tests in chapters 4-9 of the main guideline. For further guidance clinicians are advised to refer to the ‘Good Practice in Consent’ range of guidance on issues of consent in the NHS9 www.doh.gov.uk/consent. This guideline supports the advice given in that publication, that it is “a general legal and ethical principle that valid consent must be obtained before starting treatment or physical examination, or providing personal care, for a patient” and that patients should have sufficient information in order to make an informed decision about whether to consent (i.e., risks, benefits and alternatives). Reference List 1. López-Argumedo M., Asua J. Preoperative Evaluation in Elective Surgery. (INAHTA Synthesis Report). 1999. Osteba, Vitoria-Gasteiz. Dpt. of Health Basque Government. Basque Office for Health Technology Assessment.


Page 20 of 203

DRAFT FOR CONSULTATION 2. Munro J, Booth A, Nicholl J. Routine preoperative testing: a systematic review of the evidence. Health Technol.Assess 1997;1:i-iv. 3. Swedish Council on Technology Assessment in Health Care. Preoperative routines. Int J Technol Assess Health Care 1991;7:95-100. 4. Osteba: Health Department. Pre operative evaluation in healthy/asymtomatic patients. 1995. Osteba, Health Department. 5. Royal College of Radiologists. Making the best use of a department of clinical radiology. 1998. London, Royal College of Radiologists. 6. American Society of Anesthesiologists. Statement on routine preoperative laboratory and diagnostic screening. http://www.asahq.org/Standards/28.html . 1993. American Society of Anesthesiologists. 1-3-2002. 7. American Society of Anesthesiologists. Practice advisory for preanesthesia evaluation: a report by the American Society of Anesthesiologists Task Force on Preanesthesia Evaluation. Anesthesiology 2002;96:485-96. 8. NHS Executive. Clinical guidelines using clinical guidelines to improve patient care within the NHS. 1996. London, NHS Executive. 9. Department of Health. Good practice in Consent implementation guide: consent to examination or treatment. 2002.


Page 21 of 203


2 Methods 2.1

Outline of methods used There were several steps in the development of these guidelines: •

Assembly of the Guideline Development Group - (See Section 2.2);

•

Aims and scope of the guideline. Defined at beginning of the project in collaboration with the Guideline Development Group and other experts (see Section 2.3);

•

Systematic review of the literature to update previous reviews, including a review of recommendations in textbook and other sources of ‘received wisdom’. Critical appraisal and synthesis of literature (see Section 2.4 and Chapter 3);

•

Economic review of the cost of tests included in the guidelines (see Section 2.5 and Chapter 6);

•

Choice of development method by formal consensus process based upon lack of scientific evidence from literature review (see Section 2.6);

•

Interviews to test assumptions of the guideline development group with trainee surgeons and consultant anaesthetists (see Section 2.7 and Chapter 4);

•

Selection of expert panel for consensus, reflecting a range of users and providers of the tests. Patient representatives were present at each meeting (see Section 2.8 and Chapter 5);

•

Development of consensus questionnaire (see Section 2.9, Chapter 5 and Appendices 6 and 7);

•

Consensus round 1: consensus questionnaire sent to and rated by expert panels (see Section 2.10, Chapter 5);

•

Consensus round 2: Meeting of expert panels: statements discussed in turn and re-rated (See Section 2.11, Chapter 5);

•

Drafting of guideline (see Chapter 7);

•

Comments sought from a wider audience (i.e. contributors to the guideline and other interested parties not formally registered as stakeholders).

•

Adjustments made in light of comments.


Page 22 of 203

DRAFT FOR CONSULTATION Assembly of Guideline Development Group

Systematic Review of the Literature

Aims and scope of the guideline

Choice of Consensus Development Method

Economic Review of the literature and modelling

Interviews to test assumptions about preoperative testing

Selection of Experts for panel

Development of Consensus Questionnaire

Statements rated appropriateness of testing (two rounds)

Development of Guidelines (Chapter 4)

Figure 1: Outline of Methods


Page 23 of 203


2.2

Assembly of an advisory group The Clinical Guideline Development Group (GDG) was formed to make executive decisions about the project and was responsible for the day-to-day overseeing and management of the project. The GDG had representatives from all the key professional bodies. In addition to the GDG a further group of experts was assembled with representatives from other health care professions and patient/carer groups, to ensure the GDG had access to wider expert opinion. (See Acknowledgements)

2.3

Aims and scope of the guideline From the first meeting, the GDG established that, in reviewing evidence for the guideline, it would not be sufficient simply to “update” the earlier HTA review1There was concern that the review did not appear to have consulted with the users of preoperative test information (e.g. anaesthetists and surgeons), and that some of the complexities of preoperative testing had not been explored. In finalising the detailed scope for the guideline, the GDG debated a number of issues in detail. First, it was necessary to interpret the term ‘routine’ testing. At least three possible interpretations were suggested: •

preoperative testing carried out in ‘uncomplicated’ patients (requiring further elaboration of uncomplicated);

•

preoperative testing carried out in the ‘majority’ of or ‘typical’ patients (requiring further elaboration of majority or typical);

•

preoperative testing carried out ‘habitually’.

The GDG chose the second interpretation. Members considered the first interpretation to be too limited (see below, with respect to comorbidity) and the third interpretation to be too difficult to define. Because of the difficulty of interpreting the term routine, we have chosen instead to use the word generic throughout the Guideline (see Glossary for a definition of generic testing).


Page 24 of 203

DRAFT FOR CONSULTATION A second issue concerned ‘legitimate’ reasons for testing. The HTA review [Munro et al.] appeared to require that abnormal test results should bring about changes in clinical management, with the presumed aim of preventing immediate operative complications. The GDG were not satisfied that the value of preoperative testing should be limited in this way, for two reasons. First, most studies have adopted a narrow definition of what constitutes a change in clinical management (see below). Second, anaesthetists in the GDG could envisage other legitimate reasons for preoperative testing that would not necessarily result in any observable change in management. For example, an anaesthetist might want to document a biochemical variable in advance of an operation in order to interpret better a measure of the same physiological variable after operation; that is, in order to manage the postoperative care of a patient optimally, an anaesthetist or surgeon might need to know whether the variable was abnormal in advance of the operation, or whether it became abnormal during the course of the operation. The definition of a ‘change in clinical management’ mentioned above was a third issue of concern. The GDG was not satisfied that the frequency of changes in clinical management should be quantified simply by the frequency with which abnormal test results led to operations being postponed or cancelled. It seemed to the GDG that these particular changes in clinical management had often been selected because they were clearly observable or were likely to have been documented (in retrospective studies). However, anaesthetists in the GDG could envisage other, less observable changes in clinical management that might be very important in reducing the risk of a complication, for example a change in the choice of anaesthetic protocol, or the selection a more experienced anaesthetist to administer the anaesthetic. A fourth issue was the cost, and cost-effectiveness, of preoperative testing, which the GDG was explicitly instructed to address. While acknowledging the intrinsic importance of the costs, and cost-effectiveness of tests, there was concern that this was not the way that resource issues are considered by hospital clinicians. It was felt that while clinicians do consider cost implications of tests, they also have to consider availability of tests and weigh up the resource implications to the NHS and


Page 25 of 203

DRAFT FOR CONSULTATION potential risks to the patient, of going ahead without a test result compared with postponing an operation in order to get a test result. The GDG was unanimous in the opinion that users of the guideline would be disappointed if the guideline only addressed uncomplicated patients, assumed to imply patients classified as ASA grade 1; hence, their rejection of the first interpretation of routine preoperative testing. Interpreting ‘routine’ to mean preoperative testing in common or typical patients led the GDG to include patients with comorbidity within the scope of the review. Members chose to use the framework of ASA grades in order to do so. The GDG acknowledged the difficulty of tackling the use of preoperative tests in patients with comorbidity undergoing elective surgery. However, members were keen to make some progress in this area since it is increasingly common for elective surgical patients to have comorbidity. Also, it was felt that attempting to develop a guideline for these patients would give developers in the future some foundation to work on when updating the guideline. Members of the GDG were aware throughout that they may not have been considering all of the relevant factors. Some additional factors emerged in the GDG meetings long after formal scoping had finished, e.g. timing of preoperative testing, the optimal setting for preoperative testing, and issues of consent. Where such factors have not been formally considered in the guideline, we have tried to summarise the views and concerns of the GDG in the Discussion (see Chapter 9). 2.3.1

Steps in developing the guideline The GDG identified the following steps in developing the guideline: 1. The first step was to select the tests to be covered by the guideline. Selection took place in two stages, first in the context of patients classified as ASA grade 1 and then in the context of patients classified as ASA grade 2 and 3. The choice of tests was guided by clinical expertise in the GDG about (a) tests that were commonly ordered and (b) tests that might be valuable. The chosen definition of generic testing, that is testing carried out for all patients who satisfy certain criteria (including the presence of one of the comorbidities addressed for patients classified as ASA grade 2 or 3) and which is not directly related to the planned procedure, limited the number of tests considered.


Page 26 of 203

DRAFT FOR CONSULTATION 2. The second step was to carry out a systematic review of the literature. The earlier HTA review [Munro et al.] provided the basis for this review. However, there was a need to search for more recent evidence for tests covered by the HTA review, to search for evidence on other tests included in the guideline and to appraise the quality of the evidence. 3. The GDG outlined a third step, namely the need to survey opinion about some of the areas which were considered to be highly relevant, on which there appeared to be little or no evidence and which were not addressed by the HTA review [Munro et al.]. 4. The fourth step was to carry out a formal consensus method to supplement the available evidence. Based on the earlier HTA review [Munro et al.], the GDG did not expect the systematic review of the literature to identify much evidence that directly addressed the value of preoperative testing (e.g. because all of the studies in the HTA review were case series). 5. A fifth step was to identify and review evidence about the costs of tests (consulting primary sources if necessary). The GDG had been instructed to considered costs explicitly in the formal consensus process. 6. A sixth step was to consider the role of economic modelling in assisting recommendations about the appropriateness of preoperative testing.

2.4

Systematic review methods Our aim was to carry out a systematic review of the value of generic preoperative investigations in elective surgical patients. A generic pre operative test was defined (by the GDG) as an investigation carried out preoperatively on all patients, that was not directly indicated by either the surgical procedure being carried out or by comorbidity. (See Glossary and the limitations in Chapter 8 for more details about the difficulty in applying this definition.) The starting point for this review was to update the HTA report 1to include additional evidence published between 1996 and February 2002 and review all evidence (1966-2002) for additional tests covered i.e., pregnancy tests, lung function tests, and blood gasses. (For all tests included see Chapter 1, Section 1.4).


Page 27 of 203

DRAFT FOR CONSULTATION For a full transcript of the review, please refer to the CD ROM or the NCACC website. 2.4.1

Search strategy A comprehensive search strategy was developed in collaboration with the GDG with the aim to identify all papers relating to generic preoperative testing. Both MeSH headings and free text searches were used to capture all publications.

2.4.2

Types of intervention Generic investigations carried out before surgery, required to determine the fitness for anaesthesia and surgery (see Glossary) Inclusions: The investigations considered in this review are listed in Table 1.1 in Chapter 1. These include all investigations reviewed in the HTA report, and three additional ones (pregnancy testing, blood gases and lung function tests). Exclusions: Four investigations initially considered in the scope of the guidelines were excluded from the review. These were computed tomography (CT scan) of the thorax, Muga scans, haemoglobin electrophoresis and blood cross matching largely because test results are used primarily to assist in the staging of malignant disease, to inform prognosis and post operative management of the patient rather than to inform the anaesthetic and surgical management of the patient. Results of the CT scan, for example, may influence the perioperative management of the patient, because it is usually carried out preoperatively in patients with malignancy or for reasons directly related to the disease being treated or the operation being carried out, e.g., as a guide to the anatomy it is not considered a ‘generic’ test. Muga scanning was excluded because it is normally only ordered for patients with cardiac failure undergoing cardiac related surgery. The test does not qualify as a ‘generic’ test. In addition, at the time of this review, it was not widely available. Haemoglobin electrophoresis was excluded as it was considered to be a test for differential diagnosis of haemoglobinopathies, used, for example, following a positive sickle cell test, rather than for risk assessment. Blood cross matching was excluded, it is used to prepare a number of units of red blood cells for transfusion, stored for a limited time for a specific patient. The GDG agreed that its function is procedural rather than diagnostic and it does not change the management of the


Page 28 of 203

DRAFT FOR CONSULTATION patient as there are no ‘abnormal’ results to consider. Also the need to do cross matching or a ‘group and save’ is dependent on the severity of surgery and the likelihood of blood loss. 2.4.3

Types of participants The population of interest was split into two phases. Phase I:

all uncomplicated, ASA grade 1 children or adults, undertaking elective surgery.

Phase II:

all adult patients with comorbidity (cardiovascular, respiratory or renal) ASA grade 2 and 3, undertaking elective surgery.

2.4.4

Types of outcome measures Estimates of the frequency, risk difference or relative risk of an abnormal result and a change in management (see Table 2.1).

2.4.5

Types of studies All study designs (full publications) that might provide information about the ’value’ of preoperative investigations were considered for this review. In theory the literature could have yielded three types of evidence with relevance to this review (see Table 2.1): •

Comparisons of effectiveness between groups of patients who had and had not undergone pre operative investigations;

•

Estimates of diagnostic accuracy of preoperative investigations for predicting complications, adverse events and changes in management;

•

Estimates of the ‘yield’ for carrying out investigations (i.e., the proportion of patients undergoing preoperative investigations in whom an abnormal result was observed and whose clinical management changed as a result, or who went on to have an adverse event or perioperative complication).

Based on experiences of Munro et al 1we expected to find few studies that provided strong good quality evidence. We therefore considered all quantitative study designs for the review.


Page 29 of 203

DRAFT FOR CONSULTATION Exclusion criteria Papers were excluded from the review if:

2.4.6

•

They did not report primary outcome data relevant to the review;

•

They reported clinically indicated tests;

•

They were not written in the English language;

•

They included fewer than 10 patients

Search strategy for identification of studies Databases were searched from 1995 to 2002. The search strategy deliberately included one year overlapping with the HTA review (1995) to ensure that no articles were missed because of a time lag in indexing some articles for bibliographic databases. In addition for tests not included in the HTA review (pregnancy testing, sickle cell tests and random glucose testing) MEDLINE was searched from 1966 to 2002 and EMBASE from 1989 to 2002. a) Electronic searching i)

The Cochrane Library 2001 issue 4 (Databases used: CDSR, DARE, CCTR, HTA, NHS Economic Evaluations Database)

ii) MEDLINE iii) EMBASE iv) Science Citation Index v) HealthSTAR (up to the end of the year 2000, the database no longer exists after this year). b)

Manual searching reference lists in identified studies, reviews

c)

Professional contacts

The search strategies used for MEDLINE and EMBASE are detailed in Appendix 1. These search strategies were adapted for searching other databases. 2.4.7

Selecting studies Studies were identified and excluded as follows.


Page 30 of 203

DRAFT FOR CONSULTATION (1)

From reading the titles and abstracts alone papers were excluded if they were definitely irrelevant

(2)

Full publications of the remaining articles were obtained, and again papers were excluded if they were definitely irrelevant A 10% sample of papers was reviewed by a second reviewer to ensure inter-observer consistency. We used any disagreements to inform the selection. However no analysis of agreement was performed.

(3) 2.4.8

Data were extracted from all eligible papers

Data extraction Data regarding the patient population, interventions and outcomes of the paper were extracted (Appendix 3). Again, 10% of papers was performed independently by two reviewers . Studies were categorised as descriptive, diagnostic or as addressing effectiveness questions. Characteristics of included studies are shown in the full systematic review (CD ROM).

2.4.9

Methodological quality The methodological quality of all studies was assessed. Three data forms were created in order to record aspects of quality specific to the three types of evidence (Appendix 4). Papers were assessed by one reviewer; a 10% sample was assessed by a second reviewer and disagreements were discussed with a third reviewer if necessary. No formal analysis of inter-rater reliability was performed.

Table 2.1. Different types of evidence relevant to the review Type of evidence / study type

Description

Outcome measure

Estimates of the yield of tests Prospective or retrospective case series

Studies describing the proportion of surgical patients who have an abnormal preoperative test result, who are subject to a change in clinical or who experience an adverse event / complication. Diagnostic accuracy of preoperative test, i.e. ability to predict complications, adverse events or changes in management Analytical comparison of outcomes (adverse events, changes in management, treatment or health outcomes) for patients who have

Estimates of the frequency of abnormal results, changes in clinical management, adverse events or complications

Diagnostic accuracy of tests

Effectiveness of testing Randomised or non-randomised controlled trials, cohort studies, case control studies.


Sensitivity and specificity, negative and positive predictive values, likelihood ratios Ratio or difference measures of effect size

Page 31 of 203

DRAFT FOR CONSULTATION undergone preoperative testing with those who have not

2.5

Economic review of the cost of tests included in the guideline Methods and results for this aspect of the guidelines are presented in Chapter 7.

2.6

Choice of development method by formal consensus The evidence base available for these guidelines was both anticipated to be and found to be extremely poor. Consequently the GDG decided to use formal consensus to identify areas of consensus, on which the guidance could be based, when the evidence-base from the literature could not be relied upon. The modified nominal group technique based on the RAND/UCLA (University College of Los Angeles) consensus panel method for assessing appropriateness of procedures2; 3; 4 was the consensus method chosen. This method has previously been identified as the method most commonly used for the development of consensus in health care.5 There were two stages of consensus: Informal and formal. The GDG was aware that in scoping the area to be covered by the consensus, a number of assumptions had been made on which the formal consensus would be based. The GDG was concerned that these assumptions might not be acceptable to practising clinicians, and hence, before the development of the consensus questionnaire the GDG though it important to test these assumptions by informal consensus. This was carried out by interviewing a selection of surgical trainees and consultant anaesthetists (clinicians that order the tests and clinicians that use them) to check these key assumptions. The scope of the formal consensus was limited by time and resources to focus on key dimensions of preoperative testing. The GDG also used these interviews to explore areas that would not be dealt with in the formal consensus


Page 32 of 203


2.7

Testing the assumptions: informal consensus interviews with surgical trainees and consultant anaesthetists The questionnaire was developed by members of the GDG (Appendix 5).

2.7.1

The key areas addressed by the questionnaire Assumptions •

To confirm that a patient graded as ASA 1 would not require any preoperative tests as a consequence of smoking or obesity ( i.e. these factors are unimportant unless they [or other factors] have already ‘shifted’ the patient into category ASA2).

•

Severity of surgery affects the degree of physiological stress imposed on a patient by having an operation. If yes, is it reasonable to classify the severity of surgery and how this might best be done.

•

Do we need to consider preoperative testing for patients undergoing local anaesthesia, bearing in mind that there is always the chance the procedure may need to be converted to general anaesthesia if the local fails?

To inform the formal consensus process •

Do we need to test for pregnancy?

•

In what ethnic groups do we need to test for sickle cell disease and does the need to test for sickle cell disease interact with any other dimensions of interest e.g. age, type of surgery etc.

Areas not covered by the consensus •

Reasons why preoperative tests are carried out?

•

If the severity of surgery affects the degree of physiological stress imposed on a patient by having an operation is it reasonable to classify the severity of surgery and how best might this be done

• 2.7.2

For how long do test results remain ‘valid’

Identification of interviewees The aim was to interview 10 consultant anaesthetists and 10 surgical trainees. Consultant anaesthetists were initially identified through various professional


Page 33 of 203

DRAFT FOR CONSULTATION contacts, and subsequently through lists of past and potential advisors to NCEPOD. Surgical trainees were identified through lists of research, provided by the Research Board of the Royal College of Surgeons 2.7.3

Conducting the interviews Potential participants were given information about the guideline and the aims and objectives of the interviews were explained either by e-mail or telephone. Those who agreed to participate were sent the questionnaire to answer in their own time and a time was arranged for one of two members of the development team to telephone the participant to discuss their responses to the questions. Interviews lasted approximately 1 hour. Two surgical trainees were interviewed face-to-face to pilot the questionnaire.

2.8

Consensus: selection of expert panel The consensus process was split into two halves for manageability as previously described (in Chapter 1, Scope of the review). I)

Phase I: appropriateness of preoperative testing of normal healthy patients (including children) varying by age and by grade of surgery

II) Phase II: appropriateness of preoperative testing of adult patients with three common comorbidities (cardiovascular disease, respiratory disease and renal disease). The consensus process was further complicated by the need to include representatives from all key health care professions on the panel including subspecialities of surgery; and anaesthetists from different specialities (including paediatrics). In addition, the maximum recommended size of the consensus panels was 11 panellists. Therefore, for each Phase, it was decided that, to promote the applicability of the guidelines, two consensus panels would be run in parallel (each dealing with the same material). For Phase II, the composition of the parallel groups was to be the same as in Phase I, if panellists agreed to participate in a second round of rating and meeting. Because the nominal group was given the task of reaching consensus about the appropriateness of preoperative tests, a task that requires considerable clinical


Page 34 of 203

DRAFT FOR CONSULTATION expertise, members of the panel were exclusively in clinical practice. As patient/carer representatives could not contribute to these opinions, they did not form part of the panel but instead were invited to observe the meetings in order to listen to the discussion between panellists and use this information to inform their comments on the draft(s) of the guideline. Table 2.2

Composition of the parallel consensus panels used in each of the two

phases. Group 1 (n=11)

Group 2 (n=11)

2 1 1 1 1 1 1 1 1 1

2 1 1 1 1 1 1 1 1 1

Anaesthetists Paediatric anaesthetist Radiologist Pathologist – Haematologist Pathologist – Biochemist Cardiologist Surgeon – Trainee Surgeon – General Surgeon – Obstetrics and Gynaecology or Orthopaedic Nurse Practitioner

2.9

Development of the consensus questionnaire The consensus questionnaires (Phase I, Appendix 6, Phase II Appendix 7) contained statements about the appropriateness of a particular test for particular patients (varying in ASA grade and age) for different types of surgery (defined in the Glossary). The first phase of the consensus process aimed to explore two key dimensions, namely age (splitting children 100 µmol/L and < 200 µmol/L); some dietary restrictions

documented poor renal function (creatinine > 200 µmol/L); regular dialysis programme, (peritoneal or haemodialysis)

ASA definition:

Cardiovascular (CVD):

Respiratory:

Renal disease:


Page 36 of 203


Evidence about the costs of tests was provided as a list to members of the consensus panel, including plausible upper and lower limits for the costs (see Chapter 4). This information was also included in each of the scenarios used for consensus development. However, cost information was not considered explicitly as an independent variable during consensus development.

2.10 Consensus: Round 1 – initial rating Before a meeting, panellists were sent the consensus questionnaire, a short summary of the evidence produced by the GDG and a covering letter which gave instructions and definitions. Participants were asked to rate their agreement with the statements taking into account the research evidence and their clinical expertise. We used a scale of 1 to 9 scale, where 1 represented least agreement (i.e. indicating testing was not appropriate) and 9 most agreement (i.e. meaning testing was appropriate). Participants were asked to represent their personal opinion about ‘best practice’, i.e. what is in the best interests of the patient, rather than to describe what happens in their own hospital (or some similar reason for practising in a particular way), unless this practice is the same as their opinion about best practice. If they did not have an opinion or if it was outside their field of expertise, they were asked to ring the midpoint on the scale.

2.11 Consensus: Round 2 – meeting Each group of panellists then met for a full day in order to discuss and to re-rate. The first two meetings for Phase I were held in March 2002 and the second two meetings for Phase II of the consensus were held in April 2002. The aim of this process was to explore possible reasons for disagreements, with the aim of obtaining an improved consensus. At each meeting, the distribution of responses to each statement was presented to group members, together with each member's response to that statement. This enabled participants to see the spread of views and how their own response related to this


Page 37 of 203

DRAFT FOR CONSULTATION At the nominal group meeting each statement was discussed, paying particular attention to questions for which the distribution of responses indicates a lack of consensus and then re-rated privately by each participant. 2.11.1 Definition of agreement Only the ratings from the second round were used to develop the guidelines. (The rating for each of the four meetings are shown in Appendix 6 and Appendix 7.) The median (measurement of central tendency or average) and inter-quartile range (25th and 75th percentile as a measure of distribution) were calculated for each statement from the ratings of the second round 2. Table 2.4

Definition of agreement within each consensus panel

Agreement 100% consensus Less than 100 % but greater than 75% consensus

No agreement/no consensus

Table 2.5

Definition Ratings of all panel members fall within a single 3 point region, i.e., 1 -3 (inappropriate test) 4-6 (equivocal) or 7- 9 (appropriate test) For consensus groups with 11 participants: The ratings of at least 8 panellists must lie within the three point region of consensus (1-3 or 7-9) For consensus groups with 8 to 10 panellists: The ratings of at least 6 panellists must lie within the three point region of consensus (1-3 or 7-9) Anything outside this agreement was taken as no consensus

Definition of agreement between each consensus panel (i.e. between

group 1 and group 2) Agreement

Definition

100% agreement Inappropriate to test 75% 60 years Adults Adults and Children Children Not stated Outcome: Abnormal test result Change in clinical management

Number of studies (%) Chest ECG 38 29

Hb and blood 36

Haemostasis 27

Biochemistry 10

Urinalysis 16

Pregnancy 6

All studies 106

20 (53.8) 14 (35.9) 4 (0.3)

18 (62.1) 8 (27.6) 3 (10.3)

14 (38.9) 7 (19.4) 15 (41.7)

16 (59.3) 4 (14.8) 7 (25.9)

4 (40.0) 3 (30.0) 3 (30.0)

10 (62.5) 3 (18.8) 3 (18.8)

6 (100.0) 0 0

65 (61.3) 20 (18.9) 21 (19.8)

2 (5.1) 1 (2.6) 0 1 (2.6) 0 0 0 2 (5.1) 2(5.1) 33 (84.6)

3 (10.3) 3 (10.3) 1 (3.4) 2 (6.9) 0 1 (3.4) 0 1 (3.4) 1 (3.4) 19 (65.5)

5 (13.9) 1 (2.8) 0 1 (2.8) 1 (2.8) 0 0 1 (2.8) 1 (2.8) 28 (77.8)

1 (3.7) 1 (3.7) 0 1 (3.7) 1 (3.7) 0 0 0 0 24 (88.9)

4 (40.0) 0 0 0 0 0 0 0 0 6 (60.0)

2 (12.5) 1 (6.3) 0 0 0 0 0 0 0 13 (81.3)

0 0 0 0 0 0 0 0 0 0

7 (6.6) 4 (3.8) 1 (0.9) 2 (1.8) 1 (0.9) 1 (0.9) 0 2 (2.1) 2 (2.1) 80 (75.5)

14 (35.9) 25 (64.1)

14 (48.3) 15 (51.7)

13 (36.1) 23 (63.9)

10 (37.0) 17 (73.0)

3 (30.0) 7 (70.0)

14 (35.9) 25 (64.1)

5 (83.3) 1 (16.7)

36 (37.1) 61 (62.9)

19 (48.7) 20 (51.3)

18 (62.1) 11 (37.9)

11 (30.6) 25 (69.4)

18 (66.7) 9 (33.3)

3 (30.0) 7 (70.0)

19 (48.7) 20 (51.3)

1 (16.7) 5 (83.3)

54 (55.7) 43 (44.3)

4 (10.3) 16 (41.0) 10 (25.6) 4 (10.3) 5 (12.8)

4 (12.8) 18 (62.1) 5 (17.2) 1 (3.4) 1 (3.4)

2 (5.6) 10 (27.8) 13 (36.1) 4 (11.2) 7 (19.4)

0 9 (33.3) 4 (14.8) 7 (25.9) 7 (25.9)

1(10.0) 4 (40.0) 1 (10.0) 2 (20.0) 2 (20.0)

1 (6.3) 9 (56.3) 1 (6.3) 4 (25.0) 1 (6.3)

0 1 (16.7.) 3 (50.0.) 1 (16.7) 1 (16.7)

7 (6.8) 40 (38.8) 18 (17.5) 19 (18.4) 20 (19.4)

37 (94.9) 24 (61.5)

29 (100.0) 16 (55.2)

36 (100.0) 22 (61.1)

27 (100.0) 25 (92.6)

10 (100.0) 9 (90.0)

16 (100.0) 14 (87.5)

6 (100.0) 6 (100.0)

106 (100) 65 (61.3)


Page 45 of 203