THE study by Hoeks et al. 1in this issue of Anesthesiology indicates that only an astonishingly low 21% of patients for whom the American College of Cardiology (ACC)/American Heart Association (AHA) guidelines recommend preoperative noninvasive cardiac testing were actually tested. Anesthesiologists (in The Netherlands) thus do not strictly follow the ACC/AHA guidelines on preoperative cardiac testing in patients undergoing vascular surgery. Is this substandard care, or are there probably valid reasons for this approach?
A considerable percentage of patients scheduled to undergo vascular surgery are high-risk patients because of concomitant coronary artery disease. Postoperative myocardial ischemia and infarction are serious adverse events accounting for up to 40% of postoperative fatalities and increased duration and costs of hospital stay, and significantly contribute to long-term mortality.2–5Detailed preoperative assessment and risk-reduction strategies have been proposed to ultimately improve outcome.6,7However, there is uncertainty about whether such algorithms indeed improve outcome.8–10
A first question of course is whether in daily practice physicians actually adhere to such algorithms. The study by Hoeks et al. 1in this issue of the Journal gives some interesting insights. In patients enrolled in the Euro Heart Survey Program (Sophia Antipolis, France), only 21% of patients for whom the ACC/AHA guidelines recommend preoperative noninvasive cardiac testing were actually tested. Conversely, 89% of those for whom no testing was recommended by the ACC/AHA guidelines were indeed not tested. Therefore, in the Euro Heart Survey Program, the majority of patients scheduled for vascular surgery did not undergo preoperative cardiac testing. Interestingly, long-term outcome was nearly identical in patients preoperatively tested or not, irrespective of the recommendations by the ACC/AHA guidelines.
This study asks many interesting questions: Does this mean that such guidelines are of limited value? Are they simply too complex to be followed? Do anesthesiologists consider patients treated with β-blockers, statins, and antiplatelet drugs as already maximally protected perioperatively, so that any further testing would not change the medical treatment? Is the cardiac testing capacity actually available for additional preoperative testing in due time? Is additional cardiac testing affordable in times of budget restriction? Do anesthesiologists fear that any further testing might result in a cardiologic intervention transforming a stable coronary plaque into an unstable coronary plaque with the need for dual antiplatelet treatment of largely unknown duration while the patient is at an increased risk of coronary (stent) thrombosis perioperatively?
These are important topics of great clinical interest, and they are currently vividly debated. The study by Hoeks et al. 1adds new information but has significant limitations. First, the number of patients enrolled in this retrospective observation is rather small. Moreover, the precise definition of cardiovascular outcome and its assessment remains unclear from the article. Patients as young as 18 yr were compared with elderly at-risk patients bearing a significant burden of arteriosclerosis. Arbitrary definitions with respect to “advanced age” and grading of risk within the vascular surgical patients (“low,”“intermediate,” and “high”) were applied by the authors in their analysis, and no information about possible confounding variables, such as the perioperative use of cyclooxygenase-2 inhibitors, sulfonylurea drugs, or α2agonists, was detailed. Most importantly, the authors do not report on whether β-blockers, statins, and platelet inhibitors were given in the context of a long-term chronic treatment or whether these drugs were newly introduced perioperatively and whether the heart rate of β-blocked patients indeed was in the protective target range. Also, no information is provided about contraindications for β-blockers or about serious side effects because of preventive medical treatment. Finally, it is unclear whether the AHA guidelines are similarly applicable to patients undergoing semiinvasive endovascular stenting procedures during sedation as opposed to patients having open surgical procedures during anesthesia.
The cited guidelines may advise the anesthesiologist to consider clinical predictors of increased perioperative cardiovascular risk, exercise capacity of the patient, and surgery-specific risks.6,7A combination of clinical predictors and low exercise capacity (usually ≤4 metabolic equivalents [METs]) ask for noninvasive cardiac testing, and the further steps are determined by the results of such testing. A valid question now is whether such noninvasive testing is indeed necessary or whether one could not simply opt to apply the maximum medical protective strategy without testing as if testing would have recommended such treatment.10,11In fact, the therapeutic options are few: Chronic β-blockade, statin, and antiplatelet therapy should be continued. Preoperative acute β-blocker treatment may be beneficial in these patients as well,12,13but there are well-designed perioperative β-blocker studies questioning the efficacy of this therapy.14–16Preoperative initiation of statin therapy is even more controversial, albeit likely to be efficacious as well.17–19Unfortunately, it is not reported in the article by Hoeks et al. 1whether β-blocker and statin therapies were continued chronically or newly introduced preoperatively.
Do we miss some high-risk patients by skipping noninvasive testing, and would this result in an adverse outcome when we simply treat (in the absence of contraindications) all high-risk patients with perioperative β-blockers and statins and continuing antiplatelet drugs? One might argue that Boersma et al. 20had identified in 2001 a small subgroup (2% of a selected population) in which perioperative β-blockade did not reduce postoperative cardiac complications and thus myocardial revascularization was proposed. Interestingly, the same group of researchers recently tested this proposal in the very subgroup of high-risk patients with extensive ischemia in preoperative testing in a prospective randomized trial and found no benefit of preoperative revascularization as opposed to optimized medical treatment, either at 30 days or at 1 yr.21This is in keeping with previous studies in lower-risk patients where coronary artery bypass grafting before vascular surgery or percutaneous coronary interventions also did not improve long-term outcome.22,23
Today, percutaneous coronary interventions involve placement of stents, many of which are drug-eluting stents in a majority of cases.21,24During the reendothelialization period, these stents are highly thrombogenic, patients must be treated by a dual antiplatelet regimen of at least 12 months, and all elective surgery should be postponed for at least 1 yr according to current recommendations.25Even later stopping of clopidogrel may be associated with an increase of stent thrombosis and major adverse events such as myocardial infarction and death.26–28Because any coronary intervention per se renders plaques unstable, these procedures should, whenever possible, be avoided preoperatively. Therefore, not testing high-risk patients preoperatively but treating them medically with β-blockers and statins and continuing antiplatelet therapy may not be a hazardous but more likely a beneficial regimen. In addition, one third to one half of postoperative myocardial infarction are linked to unstable plaque rupture,29,30and it can be expected that preoperative stress testing will miss a significant proportion of patients at high risk of postoperative myocardial infarction; unstable plaques cause only moderate coronary stenoses and are usually silent during stress tests.31On the other hand, cardioprotective drugs are potentially dangerous and may harbor significant side effects.32Despite the fact that studies assessing the outcome of patients with or at risk of coronary artery disease have not proved a real benefit of preoperative testing,8–10larger studies may be necessary before generally adopting such a regimen.
The story may be even more complicated. According to the current guidelines of the ACC/AHA, the patient’s perioperative cardiovascular risk can be estimated from (1) the medical history, (2) the type of surgery, and (3) the physical performance, which ultimately determines preoperative evaluation and the choice of prophylactic therapy. However, information from the patient’s genetic background is not part of these guidelines but may be of paramount importance and help to individualize perioperative medicine to improve outcome.16,33So-called “average” beneficial and detrimental effects observed in clinical trials may result from subgroups of patients with a particular genetic background. Accordingly, not all patients may profit from β-blockers to the same degree because of patient-specific genetic polymorphisms related to pharmacodynamic properties of receptors and/or pharmacokinetic properties of drug metabolism. In the β-Blocker Evaluation of Survival Trial study,34Arg389 homozygotes of the β1-adrenergic receptor treated with bucindolol had the largest benefit in mortality reduction (−38%), whereas Gly389 carriers had virtually no clinical benefit from bucindolol therapy compared with placebo.35Equally important and maybe more pertinent to this discussion, based on their genetic background some patients may be more reliably detected as being at high risk in dobutamine stress echocardiography than other patients.36Therefore, preoperative genomic testing should be integrated in future guidelines to tailor testing and treatment to the genetic background of patients. Because patient genotyping will be soon available to clinicians, updated guidelines should take into account these new perspectives.
From a theoretical point of view, to ultimately achieve our goal of improving patient outcome, we may have to test more thoroughly, including preoperative genomic profiling, to treat efficaciously based on genetic background. However, for the time being, we need a pragmatic approach. Hence, putting our resources into perioperative medical treatment and high-quality intraoperative and postoperative anesthesia care may be more efficacious than elaborative but incomplete preoperative testing.
*Department of Anesthesiology, University Hospital Zurich, Zurich, Switzerland. donat.spahn@usz.ch†Department of Anesthesiology, University Hospital Lausanne (CHUV), Lausanne, Switzerland.