Should routine pre‐operative testing be abandoned?

More preoperative assessment by physicians and less by laboratory tests [1]. In January 2010, the Association of Anaesthetists of Great Britain and Ireland published its safety guideline Pre-operative Assessment and Patient Preparation – The Role of the Anaesthetist 2 [2]. The opening line of the section on tests and investigations states that ‘Routine pre-operative investigations are expensive, labour intensive and of questionable value, especially as they may contribute to morbidity or cause additional delays due to spurious results’. Citing Clinical Guidelines 3, published in 2004 by the National Institute for Clinical Excellence (NICE) [3], the authors go on to recommend that routine tests should indeed be undertaken in a number of specific patient groups before most types of surgery, including many procedures that are typically performed on a day-case basis. In the prevailing economic climate, can continued expenditure on tests of questionable value be justified, and is there sufficient evidence to consider abandoning routine pre-operative testing altogether? The laws of probability dictate that multiple testing of patients will yield abnormal results. For continuous variables, such as haemoglobin concentration, the ‘normal’ or reference range quoted by clinical laboratories is typically the mean ± 1.96 standard deviations. By definition, therefore, 5% of normal individuals will have a test result that lies outside the reference range (2.5% below and 2.5% above). Given the probability that a test result will lie within the reference range is 0.95, the likelihood of obtaining twenty normal test results is only 36% (i.e. 0.9520). In other words, the more tests that are performed, the more likely it is that an abnormal result will be detected, even in healthy individuals. The important question is the significance of the abnormal result in an asymptomatic patient, and how this will alter his or her management. As the term suggests, routine pre-operative testing is ubiquitous, having become a firmly embedded component of innumerable integrated care pathways (ICPs) that both describe and proscribe the ‘patient journey’. Despite counselling our trainees only to order tests that are clinically indicated, the standard answer to a question on pre-operative assessment nearly always begins with the phrase ‘I would take a history, examine the patient and order routine tests’. Is it any wonder that surgical colleagues, allied healthcare professionals and even patients themselves attach so much importance to routine testing? Perhaps a universal approach based more on clinical findings would lead to a more reasoned testing regimen, and this could be encouraged during teaching sessions. This, in turn, might eventually lead to a change in the automatic response given by anaesthetists in training. Most clinicians advocate routine testing of patients before minor surgery if significant comorbid disease is present. Not infrequently, these patients are under regular assessment by their general practitioner or hospital specialist. A careful history may reveal that tests have already been recently undertaken, and will not need repeating, and the results should be relatively easy to track down. Routine testing before major surgery (such as cardiac or vascular) is usually justified on the grounds that it establishes a pre-operative baseline, with which to compare postoperative test results. These tests are often performed immediately before surgery as part of a protocol or ICP. In many cases, these tests have already been completed as part of the diagnostic workup, and in a stable patient, it is unlikely that the results will have changed sufficiently to justify repeating them. Accessing these data has been considerably simplified by the introduction of digital information and communication systems. Unfortunately, in many instances it is more expedient to repeat a test needlessly than to spend time searching for the results of recent tests. Despite the publication of clinical guidelines, fundamental misunderstandings persist about the utility of routine pre-operative testing [4]. Erring on the side of caution, tests are often ordered in the belief that they are an effective screen for disease that cannot be predicted on the basis of history and examination. Moreover, it is widely believed (rightly or wrongly) that ordering a large battery of tests will somehow ‘appease’ the anaesthetist – allowing him or her to rectify unexpected pre-operative problems, thereby reducing the risk of a surgical procedure’s being delayed or cancelled. One could be forgiven for thinking that routine testing has acquired its own therapeutic value and, at least in part, replaced the need for taking a clinical history and performing a competent physical examination. Ordering tests simply to pacify the anaesthetist is not without risk – especially if the results are unavailable at the scheduled start of surgery. In this situation it could be reasonably argued that anaesthesia should be delayed on the grounds that doing otherwise theoretically exposes the anaesthetist to medicolegal risk. That said, how many anaesthetists could truthfully say they have never proceeded to induction of anaesthesia in the absence of a pre-operative test result? Furthermore, how often has an unexpected, abnormal routine test result actually resulted in a significant alteration to the anaesthetic or surgical technique? Ignoring the absence of a pre-operative test result devalues the time and effort expended in performing the test and sends a bad message to the patient. Regardless of whether a pre-operative test was indicated or not, the finding of an abnormal result carries with it the obligation to determine its clinical relevance and decide if further investigation is required. This burden of clinical responsibility may not be immediately obvious to well-intending non-medical staff who order tests that are not clinically indicated. Abnormal test results are often not documented in the medical records, and remarkably, in up to 60% of cases, an abnormal test result discovered before routine surgery does not lead to further investigation [5]. Paradoxically, this might actually be in the patient’s best interests, because acting upon a clinically insignificant abnormal test result may lead to postponement of planned surgery and expose the patient to further investigations and the risks of unnecessary treatment or procedures. The positive predictive value of individual tests, when used to screen for disease, is low. A routine pre-operative full blood count may demonstrate a haemoglobin concentration < 9 g.dl−1 in 0.8% of patients, but it is the presence of symptoms rather than the absolute result that determines whether a change in management is appropriate [6]. Abnormalities on pre-operative ECGs are common but are poor predictors of postoperative cardiac complications [7]. Similarly, abnormal chest radiograph findings are common, but fewer than 5% of abnormal results actually alter management [4]. Does routine pre-operative testing improve patient outcomes, increase efficiency and benefit healthcare organisations? In a landmark study, Schein et al. [8] demonstrated that routine testing in elderly patients before cataract surgery produced no significant outcome benefit. Cost savings in North American centres embracing this doctrine have been estimated to be up to £20 (US$30) per patient. More recently, a prospective randomised study in 1026 ambulatory surgical patients, undertaken in Canada, demonstrated that routine pre-operative testing did not reduce the incidence of adverse events during and up to 30 days after surgery [9]. Cost savings were reported to be similar to those reported by Schein et al. The superiority of history and examination over routine pre-operative testing was demonstrated in a prospective study of 17 638 day-case surgical patients [10]. Clinical risk factors such as hypertension, obesity, smoking, asthma and gastro-oesophageal reflux were much better predictors of peri-operative adverse events than abnormal pre-operative test results. The number of patients undergoing day-case (ambulatory) surgery in the UK is likely to rise. The NHS Modernisation Agency’s High Impact Changes for Service Improvement states that day surgery (rather than inpatient surgery) should be the norm for elective surgery, and that at least 80% of patients could be treated as day cases (< 24 h hospital stay) [11]. Such an increase makes an examination of the efficacy and feasibility of routine pre-operative testing even more important, since so many patients are now admitted a very short time before their surgery. Whilst the widespread adoption of pre-admission clinics has undoubtedly streamlined the preparation of patients for surgery, much of this process has now been delegated to nurses and non-medical staff. Following procedure-specific pathways that require little independent thought, patients undergo a pre-ordained battery of tests regardless of the findings of the history or clinical examination. In fact, the investigations are often performed or ordered before the patient is even seen. The price paid for efficient and ‘demedicalised’ pre-operative assessment is the loss of medical intelligence. In monetary terms, the cost is substantial – in the US, this has been estimated at $18 billion per year [1]. Existing guidelines recommend routine pre-operative testing on the basis of patients’ age, pre-existing disease, the type and complexity of surgery and American Society of Anesthesiologists (ASA) physical status classification. Many guidelines are complex and confusing, and as a consequence may lead to unnecessary over-investigation. Using one flow chart and no fewer than 38 tables, the 2004 NICE guidelines cover the application of 11 pre-operative tests in the setting of four grades of surgical complexity and three organ system comorbidities [3]. Not surprisingly, there have been calls for simplification, but their rationale and presentation have been robustly defended by the authors [12]. Eliciting a comprehensive clinical history and performing a competent and comprehensive physical examination remain the most efficient and accurate way of detecting significant morbidity. Perhaps the time has come to abandon all routine pre-operative testing before day-case surgery, and indeed consider such a policy before all elective surgery. No external funding and no competing interests declared. doi:10.1111/j.1365-2044.2010.06503.x