NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.

National Collaborating Centre for Acute Care (UK). Preoperative Tests: The Use of Routine Preoperative Tests for Elective Surgery. London: National Collaborating Centre for Acute Care (UK); 2003 Jun. (NICE Clinical Guidelines, No. 3.)

  • This publication is provided for historical reference only and the information may be out of date.

This publication is provided for historical reference only and the information may be out of date.

Cover of Preoperative Tests

Preoperative Tests: The Use of Routine Preoperative Tests for Elective Surgery.

Show details

6Guideline for Preoperative Investigations in Patients Undergoing Elective Surgery

6.1. Types of recommendation

For each preoperative test, three types of recommendation are given that take into consideration the different reasons for preoperative testing and the likely risk/benefit to patients. The recommendations are based on the findings of the consensus process. The three types of recommendation are:

Image ch6fu1

All three types of recommendation are important. We have summarised the consensus opinion in ‘lookup’ tables rather than as text statements, to reduce ambiguity. Readers might otherwise conclude from text statements that tests should not be carried out unless there is consensus that they should be, or they may conclude that tests should be carried out unless there is consensus that they should not be.

Areas in which consensus was not obtained are particularly complex to interpret, since a lack of consensus could arise for a variety of reasons. We recommend that, in such circumstances, a clinician should carry out the test if there is a specific clinical reason for doing so. Conversely, a clinician should not carry out the test if there is no specific clinical reason for doing so. When a decision is made to carry out the test, we recommend that the clinician responsible for a patient’s care should document the reason for carrying out the test.

6.2. Level of evidence and grading for recommendations

The evidence for the recommendations in the look-up tables comes entirely from the consensus process, since none of the literature addressed the question of the value (clinical effectiveness/cost-effectiveness) of preoperative testing. This evidence is Level IV (expert opinion), using the internationally accepted framework for grading evidence.1 The level of evidence for the recommendations does not vary, so we have not labelled each recommendation in the look-up tables as based on Level IV evidence. Since all recommendations are based on Level IV evidence, they are all graded D according to the accepted grading scheme.1

Case series studies considered in the literature review (also Level IV evidence) sometimes indicated that the probability of obtaining an abnormal test result, requiring a change in clinical management because of an abnormal test result or experiencing a complication, increased with age or ASA grade. However, since none of the literature addressed the value (effectiveness) of preoperative testing, none of the recommendations are based on the literature.

Panellists who took part in the consensus development process were asked to represent their personal opinion about ‘best practice’, ie what is in the best interests of the patient, rather than necessarily to describe what happens in their own hospital or practice.

Nevertheless, it should be borne in mind that consensus opinion may not reflect the likelihood of reducing the risk of a complication of surgery by preoperative testing. Instead it may reflect panellists’ perceptions about the risk of a complication of surgery and how this risk changes in different circumstances.

It is important to stress that the lack of high quality evidence about the effectiveness of preoperative testing does not imply that preoperative testing is unnecessary and should not be carried out. Indirect methods for estimating the effectiveness of preoperative testing, such as the crude cost-effectiveness models described in Appendix 5 (CD ROM), may, in principle, help to inform decisions about whether to test or not. However, at present, there are no reliable data from which to estimate parameters even for the simple models described, and we are not confident that these parameters adequately characterise the complexity of clinical practice.

6.3. How to use this guideline

These recommendations are to help guide the appropriate generic use of preoperative tests for patients before elective surgery. The guideline is described in look-up tables that provide guidance on when to test and when not to test, based on weighing up the potential harms and benefits of testing for the patient.

The look up tables are structured so that the appropriate information can be accessed for each patient according to their age (age categories are listed along the top), ASA grade and grade of surgery. The recommendations suggest which tests should be done, these are colour coded like traffic lights: YES recommended (in green); may be considered (yellow); or NO not recommended (in red) (see key in Section 6.3.1).

There are two sets of these look-up tables, the first shows the tables organised by ASA grade and the second by grade of surgery. We show the same information in two ways to improve accessibility of the information. The look-up tables were piloted amongst NHS clinicians who suggested that information organised by grade of surgery is useful when the ASA grade of a patient is uncertain.

ASA grade: Each table represents consensus opinion for one ASA grade and for one grade of surgery. ASA grades considered include ASA grade 1, and ASA grades 2 and 3 specific to three common comorbidities (cardiovascular disease, respiratory disease and renal disease). These recommendations are designed for general guidance only and will require modification when used with patients with other specific comorbidities. For a patient with more than one comorbidity, follow the recommendations in all relevant tables. Definitions of ASA are provided in the Glossary and definitions of comorbidities are provided in Chapter 2, Table 2.3.

Grade of surgery: There is no widely accepted and validated system for classifying the physiological stressfulness of operative procedures, and the surgical grades used in the tables are a simple graded scale, from Grade 1 (minor surgery) to Grade 4 (major + surgery). Examples of the kinds of procedures that fall into each grade are shown in the Glossary; a more extensive list is provided in Appendix 2 (CD ROM).

Preoperative tests considered: In each Table (square) for each grade of surgery and ASA grade, the preoperative tests considered are listed on the left:

  • Full blood count
  • Haemostasis
  • Renal function
  • Random glucose
  • Urine analysis
  • Blood gases (For ASA grade 2 and 3 patients only)
  • Lung function (For ASA grade 2 and 3 patients only)

Within each table, recommendations about the appropriateness of the tests considered for different age groups are tabulated. Footnotes describe exceptions, additional information or insights arising from the consensus panel or the Guideline Development Group (GDG).

To help the user find the right square, both sets of look-up tables are preceded by a flow chart. The flow chart identifies the corresponding look-up table for a patient’s ASA grade, age, the severity grade of the surgery planned and specific common comorbidities.

In the first set of look-up tables (Sections 6.4 to 6.11), each page represents a patient’s ASA grade, and tables are provided in the corresponding sections:


In the second set of tables (Sections 6.12 to 6.17), each page represents a patients’ surgical grade and tables are provided in the corresponding sections:


Recommendations for sickle cell testing and pregnancy testing are also provided:


An example of using the look-up tables

Consider a patient classed as ASA grade 3 with comorbidity from respiratory disease and cardiovascular (CVD), aged 60, undergoing grade 3 (major) surgery. The recommendations for each of the comorbidities for this patient are described in look-up tables A19 and A27 or S20 and S24.

The recommendations suggest that an ECG, full blood count and renal function tests are appropriate. Haemostasis and a random glucose test are not recommended. Other tests, where the benefit is uncertain, may be considered, such as chest x-ray, urine analysis and blood gas tests.

6.3.1. Key to look-up tables

Image ch6fu2
Flowchart Icon

Look-up tables by ASA grade (PDF, 37K)

Flowchart Icon

Look-up tables by surgical grade (PDF, 47K)

6.18. Tests for the sickle cell gene in adults and children

The following recommendations and observations are in addition to those shown in Table 37:

TABLE 37. Testing for the sickle cell gene.


Testing for the sickle cell gene.

  • The sickle cell gene is found in many nationalities including families that come from Africa, the Caribbean, the Eastern Mediterranean, Middle East and Asia. It has also been detected in Cypriot people and a few other white ethnic groups.
  • It is important to offer to test all people considered to be at risk before an anaesthetic, both at hospital and dental clinics.
  • This is especially important for people of ethnic groups considered to be at risk, who have a family history of homozygous sickle cell anaemia or sickle cell trait and do not have a surgical history where it may have been detected previously.
  • People should be offered screening, with genetic counselling before and after screening.
  • Appropriate counselling for this test is important so that patients are able to give their informed consent, as there may be implications for patients who discover they are carriers of the sickle cell gene. The results of testing, even when negative, should be reported to families, with the patient’s consent, and documented in the patient’s medical record to avoid unnecessary repeat testing.

6.19. Pregnancy testing

The following recommendations and observations are in addition to those shown in Table 38:

TABLE 38. Pregnancy testing.


Pregnancy testing.

  • The need to test for pregnancy depends on the risk presented to the fetus by the anaesthetic and surgery. All women of child-bearing age should be asked sensitively whether or not there is any chance that they may be pregnant.
  • Women must be made aware of the risks of surgery to the fetus.
  • A pregnancy test should be carried out with the woman’s consent if there is any doubt about whether she may be pregnant.
  • Before having a chest x-ray, all women of child-bearing age should be asked sensitively whether they may be pregnant.

6.20. Good practice recommendations

During the development of this guideline, the GDG agreed certain principles of good practice. Although the aspects of preoperative testing to which they relate were not strictly within the scope of the guideline, it is important to describe them because the guideline was developed with the assumption that these principles were in place.

6.20.1. Ensuring clinical competence

It is important to ensure that staff undertaking clinical preoperative assessments receive appropriate education and training to allow them to apply the guideline correctly.

6.20.2. Preoperative assessment

It is crucial to ensure that a thorough medical history is taken from the patient to inform the recommendations about which preoperative tests to carry out. Taking a thorough medical history requires someone with the appropriate training.

6.20.3. Timing and setting of tests

The consensus process did not cover the issue of who should carry out the preoperative tests. However, it is clear that preoperative tests are often ordered or carried out by nurses in preoperative assessment clinics. The timing of tests should be appropriate for the tests concerned. It may be appropriate both from the doctor’s and patient’s perspective to test for certain conditions at the earliest stage possible, after a patient has been placed on the waiting list for an operation, so that there is time for the patient to be treated and for their condition to stabilise, ensuring patients are in the best possible state when they have surgery.

Some tests could be carried out in the primary care setting by the patient’s GP or practice nurse. For example, when a patient is listed for a particular operation it may be appropriate for the consultant in charge of the patient’s care to consider the possible tests that may be required for the patient and, after discussion with the patient, to inform their GP. Excellent communication between primary and secondary care, to ensure that test results are shared, would be essential if such changes in the responsibility and timing of testing were to be implemented.

Whoever carries out the tests, protocols for testing should be followed. This is particularly important for tests like urine analysis (dipstick), where not following the recommended protocol may render the result of the test meaningless.

6.20.4. Patient information and consent

Staff undertaking clinical preoperative assessments should discuss with patients which tests are recommended (or required), what they involve and why they are being carried out.

Decisions about whether or not to test should follow discussion between the patient and the doctor or nurse, especially where there is uncertainty about whether a test should be recommended or not. For some tests, a positive result carries a far greater significance for the patient than others, such as testing for previously undetected diabetes, the sickle cell gene and pregnancy.

Patients should have access to information about the tests and the possible implications of a positive result so that they can give their informed consent. Doctors or nurses carrying out or ordering tests should write in the patient’s notes that they have discussed the recommended tests and their implications with the patient.

Patients should be informed of the results of tests and about the implications for treatment, and any longer term implications for their health, if the results are abnormal.

For further guidance, clinicians should refer to the Good Practice in Consent2 guidance on issues of consent in the NHS (available from: 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 access to sufficient information about risks, benefits and alternatives to be able to make an informed decision about whether to consent.


The evidence underpinning the guideline is weak. It is therefore not justified to enforce compliance with the recommendations about when to test and when not to test. However, as with other guidance issued by NICE, we believe that the collection of the information described below (the ‘minimum dataset’) should be mandatory when a person is ordering tests in the NHS in contravention to the guideline or where the guideline is uncertain. We recognise that such a policy will be difficult to enforce, especially with paper-based systems for test ordering. However, Trust directors with responsibility for clinical governance need to recognise that auditing compliance with the guideline (see 6.22, below) will be much more difficult if the minimum dataset is not collected at the time of ordering.

Minimum dataset
  1. ASA grade of patient (potentially available from other sources since it is proposed that this item of information will become part of the Hospital Episode Statistics minimum dataset).
  2. Main comorbidity (eg renal, respiratory and cardiovascular; main categories could be precoded on the test order form).
  3. Grade of surgery.
  4. Reasons for ordering.

Ideally, given the weak evidence for the guideline’s recommendations, the minimum dataset should be collected when any preoperative test is ordered, ie even when ordering the test is consistent with the guideline’s recommendations. High quality datasets collected in this way would provide evidence about why tests are ordered for different types of patients undergoing different operations, which would be very valuable when this guideline is revised. Clearly, collection of these data also has implications for data entry for Trusts without electronic test ordering systems.

6.21. Implementation in the NHS

NHS Trusts carrying out elective surgery have a responsibility to implement this guideline. We recommend that implementation is audited (in addition to auditing compliance with the guideline) and that the methods for auditing implementation are maintained to provide a mechanism for regular review, ensuring that a revised guideline or relevant new evidence is disseminated promptly as it becomes available and new recommendations are incorporated into local guidance.

Trusts may want to consider the following steps in deciding how best to implement the guideline:

  • review existing practice against the recommendations of the guideline;
  • where available, review relevant local clinical guidelines and protocols in the light of this guideline and revise them, if appropriate;
  • if no local clinical guidelines exist, disseminate this guideline or write local guidance customising this guideline to take account of local circumstances;
  • customise the guideline for local settings to describe specific common clinical indications and exceptions and to reflect the kinds of patients assessed in particular preoperative assessment clinics; and
  • ensure that this guideline (or other local guidance) is available and effectively displayed in locations where preoperative tests are ordered.

This guideline should be used in conjunction with the guidance from the NHS Modernisation Agency on preoperative assessment for inpatients and day surgery3, 4 which is available from

6.22. Audit criteria

In addition to auditing the implementation of the guideline, Trusts should also put in place methods for auditing compliance with the recommendations of the guideline. The sophistication of the methods that Trusts can establish is likely to depend crucially on the flexibility of their IT systems. Trusts with state-of-the-art systems providing electronic ordering of tests and access to test results at the point-of-care should be able to implement several of the suggestions described below.

In Trusts with inflexible or ageing IT systems, periodic random sampling of case notes (over time, to cover the range of surgical specialities) should be carried out to review compliance with the guideline. These audits would be similar to existing periodic audits of the quality of documentation in medical records (and could be ‘piggy-backed’ on to medical record audits).

We strongly recommend that the following summary statistics are derived:

  • the percentage of patients who are not tested, in compliance with the guideline;
  • the percentage of patients who are tested, in compliance with the guideline;
  • the percentage of patients who are not tested, against the recommendations of the guideline;
  • the percentage of patients who are tested, against the recommendations of the guideline;
  • the percentage of patients who are tested and for whom one or more reasons for testing are documented; and
  • the percentage of patients for whom the minimum dataset (see above) is available.

The consistency of information also needs to be audited, eg check whether patients coded as having ‘grade 2’ surgery actually have grade 2 operations or that comorbidities are consistently described. Doctors and nurses responsible for ordering tests may change their criteria for classifying patients in order to justify continuing to order preoperative tests.

For Trusts with more up-to-date and flexible IT systems, it should be possible to collect the minimum dataset at the time of ordering (eg by selecting from drop-down menus), allowing the above audit statistics to be produced for all elective operations by a standard database query run periodically. Checks on the consistency of information may be unnecessary if classification of patients is implemented in the ordering software or can be carried out much more easily.

With electronic ordering and state-of-the-art IT systems, it may also be possible to implement some of the following suggestions:

  • provide interactive feedback about the guideline criteria for testing, as the person ordering the test enters information about the patient;
  • implement mapping of operation codes to scale of severity of surgery (see Appendix 2, CD ROM);
  • audit whether test results are ‘opened’, and by whom, before the operation takes place;
  • audit the proportion of test results that are abnormal for different categories of patient; and
  • trigger e-mail queries to the person who ordered a test or who opened the test results, asking whether the test results altered the clinical management and, if yes, how.

6.23. Costs and cost-effectiveness

The economic aspects of preoperative testing are discussed in full in Appendix 5 (CD ROM) and are summarised by the following key points.

  • Preoperative testing represents a substantial drain on the resources of the NHS in England and Wales.
  • Published evidence, mainly from the USA, suggests that substantial cost savings can be achieved by eliminating ‘unnecessary’ preoperative testing (see Appendix 5, CD ROM).
  • Such cost savings may not be achievable in England and Wales, if:
    • – the prevalence of testing is lower; or
    • – there are subsequent cost-savings attributable to testing.
  • Our cost impact analysis suggests that testing costs could potentially be reduced. However, in any Trust, the costs may be either increased or decreased depending on current testing practices (see Appendix 5, CD ROM). Any cost savings would be offset by implementation costs.
  • Tests that add to NHS costs are justified if they are accompanied by substantial improvements in patient outcomes (ie if they are cost-effective).
  • The level of cost-effectiveness of each preoperative test has not been established for any population subgroup. Estimating cost-effectiveness would require carefully collected empirical evidence on:
    • – the number of cases detected;
    • – the health outcomes associated with detecting a case; and
    • – resources used (and their cost) as a consequence of detecting a case.

The context of testing may have important resource implications. The literature suggests that, wherever possible, tests should be conducted in advance of the day of surgery to avoid last-minute cancellations and to ensure optimal use of operating theatres. More research is needed in this area.


Eccles M, Mason J. How to develop cost-conscious guidelines. Health Technol Assess. 2001;5(16):1–69. [PubMed: 11427188]
Department of Health. Good practice in Consent Implementation Guide: Consent to Examination or Treatment. 2002.
1. NHS Modernisation Agency’s Operating Theatre and Pre-operative Assessment Programme. National Good Practice Guidance on Pre-operative Assessment for Inpatients. Department of Health; 2003.
2. NHS Modernisation Agency’s Operating Theatre and Pre-operative Assessment Programme. National Good Practice Guidance on Pre-operative Assessment for Day Surgery. Department of Health; 2002.
Copyright © 2003, National Collaborating Centre for Acute Care.
Bookshelf ID: NBK48482


  • PubReader
  • Print View
  • Cite this Page
  • PDF version of this title (1.1M)
  • Disable Glossary Links

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...