With the increasing costs of healthcare worldwide, there is a focus on the value a service provides to our patients. As anaesthetists, that value depends upon our ability to allow the safe delivery of surgical care through preoperative preparation, intraoperative management and postoperative care. Traditionally, the focus has been on reducing morbidity and mortality directly related to anaesthesia. More recently, it has been recognised that interventions within the control or influence of anaesthetists can also lead to reduced complications from both the surgical stress and the underlying disease. A key question is how do we use the best evidence appropriately to aid in our decision-making and lead to improved patient outcomes.
One example of the evolving role of anaesthetists lies in the area of surgical site infections.1 Recent studies have suggested that activities potentially within the control of anaesthetists, for instance, maintenance of normothermia and appropriate timing of antibiotics, can influence overall patient outcomes. Additionally, complications traditionally thought to be a reflection of underlying disease, such as cardiac or pulmonary disorders, can be reduced by treatment initiated or performed by the anaesthetist. One example might be continuation of β-blockers.2 As perioperative physicians, we should look for all opportunities to improve perioperative care and apply best evidence to our perioperative decision-making.
An underlying assumption in medical care is that treatment should be individualised for a given patient if optimal outcomes are desired. In the perioperative period, anaesthetists have multiple options with regard to the choice of anaesthesia, monitors and drugs used to achieve the desired outcome. The concept of harvesting the available information to establish a baseline database for the assessment of risk of disease and of developing complications can be framed within a decision paradigm. Using our understanding of clinical findings, prior experience and the application of published evidence, we can assess the probability of a clinical problem within a certain confidence interval. Within this paradigm, the clinician may decide that the probability of the problem reaches some threshold for action. For patients with cardiovascular disease who are undergoing noncardiac surgery, the actions triggered might include changes in medical management such as initiation or continuation of β-blockers and statins, treatment for unstable coronary symptoms, coronary revascularisation and other possible interventions.3
There are multiple examples of the above process in which risk assessment has been formally collated with clinical and laboratory information to dictate subsequent care. A recent example is the use of the Revised Cardiac Risk Index (RCRI).4 The RCRI has been employed in multiple studies to define a population at risk that was subsequently randomised to two different treatment regimens. This can be seen in the Coronary Artery Revascularization Project (CARP) which used a modification of the RCRI to determine who should undergo coronary angiography. This group was then considered for potential randomisation to medical versus revascularisation therapy.5 Similarly, the RCRI has been linked with administrative data in several studies to determine the potential benefit or risk of perioperative β-blocker or statin therapy.6,7 This Bayesian approach to decision-making is the basis for the American College of Cardiology/American Heart Association (ACC/AHA) Guidelines on Perioperative Cardiovascular Evaluation and Management for Noncardiac Surgery.3
Integrating clinical evaluation and knowledge in this way unfortunately does not always lead to a clear decision for action, and it is important this is understood. A key question is the optimal action when the probability and extent of disease is less precise and the threshold for any action lies within the confidence interval of the risk assessment. The application of a test can potentially have value in such situations, as a negative test would lower the probability and obviate any need for action, whereas a positive test would raise the probability above threshold and, therefore, lead to the action.
One source of valued evidence is randomised controlled trials (RCTs). They frequently form the basis for determining the evidence supporting an action, and over several decades there has been a marked increase in their number and quality in the perioperative period. RCTs are designed to determine whether an intervention works under ideal conditions: its efficacy. In this context, we use efficacy to reflect the results of an intervention when applied in a well controlled experiment. In contrast, we can use effectiveness to describe how the intervention works when applied to normal working conditions. For example, a randomised trial may show that a drug works, but when used in a different population or in different conditions it may not work so well. Another example is epidural analgesia which may work in a randomised trial, but if used at a hospital with a poor pain service it will not work well (lack effectiveness). RCTs have very clearly defined patient inclusion/exclusion criteria and usually have strict protocols of care. Although internally, within the terms of the study, the validity of these trials is very high, their external validity and clinical application might be similarly so or it might be low. Accordingly, the intervention may behave in an identical manner or it might be different from the RCTs.
Examples of RCTs involving patients with cardiovascular disease include the study of β-blockers, statins, coronary revascularisation and thermal management. The recent series of studies of perioperative β-blockade in noncardiac surgery help illustrate both the importance of how the baseline risks of the patient and how study protocol can influence effectiveness.8 For example, a perioperative β-blocker protocol has been shown in several trials to be associated with improved outcome.2,9 However, other studies, with different protocols, have questioned this finding and its application in the real world: its effectiveness. The importance of study protocol and sample size with respect to the efficacy of a drug are demonstrated by the Perioperative Ischemic Evaluation Study trial in which the large study size provided sufficient power to detect less common events such as perioperative stroke.10
In this regard, from an evidence perspective, cohort studies are also important. Because they can include large numbers, they can achieve a sufficient sample size to determine the presence of rare associations. It is becoming increasingly common for institutions to examine their large anaesthesia information system in an attempt to answer important questions. For example, investigators at the University of Michigan have used their database to evaluate predictors of early postoperative tracheal intubation.11 One statistical method of obtaining more robust information from large databases or cohorts is through the use of propensity analysis. This approach was an important determinant in the studies demonstrating the potential harm identified with the use of aprotinin.
Another approach to the study of effectiveness is the analysis of administrative datasets (e.g. Medicare claims data) to evaluate the effect of an intervention in clinical practice. For example, Lindenauer et al.7 demonstrated that the relationship between perioperative β-blocker treatment and the risk of death varied directly with cardiac risk.
If an intervention is found to be effective or advocated in a guideline, then should it always be implemented in every patient? Clearly, the answer is no, as the individual of interest may be dissimilar to the original population studied, or the processes of care and their method of implementation may differ from those at the site originally studied. Patients have different levels of risk and additionally some have a complex characteristic of diseases for which the best practice may actually be in conflict. Applying best practices (which may be defined in guidelines), therefore, requires an understanding of the applicability of the results and the risks and benefits in the individual patient. This is a further application of efficacy versus effectiveness and illustrates the importance of incorporating information that can be applied to the individual into guidelines, and of having defined triggers for re-evaluating the evidence. For example, institution of new β-blocker therapy on the morning of surgery or in low-risk patients may actually cause more harm than good. Importantly, continuation of β-blocker therapy or initiating therapy in the appropriate population with a well defined protocol has been shown to benefit patients.2
Patient preferences are an important component of the application of evidence to the individual patient. Specifically, different patients value or assign different weights to different outcomes and, therefore, the optimal decision for any given patient is sensitive to these values. Patient preferences can be assessed using willingness-to-pay, a questionnaire-based methodology which asks patients to put a price on complication-free care, which has been applied to issues such as nausea and vomiting and choice of inpatient or outpatient care.12
A critical area in which patient preferences influence decision-making is in deciding whether to perform preoperative cardiovascular testing in patients undergoing intermediate risk surgery with one to two risk factors. As outlined in the ACC/AHA Guidelines, proceeding to the operating room without testing and with control of heart rate by β-blockade was assigned a class IIa recommendation, whereas preoperative testing was assigned a class IIb recommendation.3 In trying to determine the correct decision for the individual patient, a patient's preference for surgery versus alternative management can drive the decision. For example, if patients are determined to be high risk then they may choose to undergo a less invasive procedure or no surgery at all.
In summary, application of evidence-based practices in the perioperative period by anaesthetists should result in improvement in patient outcomes. In order to achieve this goal, the anaesthetist must understand the actual evidence with respect to the appropriate population and protocols, and should take into account patient preferences.
This commissioned Editorial is based on a refresher course lecture, held at the Euroanaesthesia meeting, June 2012 in Paris, France and is adapted in part from Fleisher LA. Improving perioperative outcomes: my journey into risk, patient preferences, guidelines, and performance measures – Ninth Honorary FAER Research Lecture. Anesthesiology 2010; 112:794–801.
The Editorial has been checked and accepted by the editors, but was not sent for external peer review.
1. Griffin FA. Reducing surgical complications. Jt Comm J Qual Patient Saf
2. Wallace AW, Au S, Cason BA. Association of the pattern of use of perioperative beta-blockade and postoperative mortality. Anesthesiology
3. Fleisher LA, Beckman JA, Brown KA, et al. 2009 ACCF/AHA focused update on perioperative beta blockade incorporated into the ACC/AHA 2007 guidelines on perioperative cardiovascular evaluation and care for noncardiac surgery: a report of the American college of cardiology foundation/American heart association task force on practice guidelines. Circulation
4. Lee TH, Marcantonio ER, Mangione CM, et al. Derivation and prospective validation of a simple index for prediction of cardiac risk of major noncardiac surgery. Circulation
5. McFalls EO, Ward HB, Moritz TE, et al. Coronary-artery revascularization before elective major vascular surgery. N Engl J Med
6. Lindenauer PK, Pekow P, Wang K, et al. Lipid-lowering therapy and in-hospital mortality following major noncardiac surgery. JAMA
7. Lindenauer PK, Pekow P, Wang K, et al. Perioperative beta-blocker therapy and mortality after major noncardiac surgery. N Engl J Med
8. Bangalore S, Wetterslev J, Pranesh S, et al. Perioperative beta blockers in patients having non-cardiac surgery: a meta-analysis. Lancet
9. Brooke BS, Dominici F, Makary MA, Pronovost PJ. Use of beta-blockers during aortic aneurysm repair: bridging the gap between evidence and effective practice. Health Aff (Millwood)
10. Devereaux PJ, Yang H, Yusuf S, et al. Effects of extended-release metoprolol succinate in patients undergoing noncardiac surgery (POISE trial): a randomised controlled trial. Lancet
11. Ramachandran SK, Nafiu OO, Ghaferi A, et al. Independent predictors and outcomes of unanticipated early postoperative tracheal intubation after nonemergent, noncardiac surgery. Anesthesiology
12. Macario A, Fleisher LA. Is there value in obtaining a patient's willingness to pay for a particular anesthetic intervention? Anesthesiology