Current Guidelines for Preoperative Management Leave Me Wanting More

As an academic hospitalist who often teaches preoperative medicine, I was asked to lecture on the revisions to the 2007 version of the American College of Cardiology/American Heart Association (ACC/AHA) guidelines for noncardiac surgery.1 In my effort to brush up on the new recommendations, I came across two articles from the Controversies in Cardiovascular Medicine series in Circulation.2^,3 These articles focus on the opinions of two thought leaders: one in cardiology and one in internal medicine. Dr. Gabriel Gregoratos is a clinical professor of medicine emeritus in the Division of Cardiology at the University of California San Francisco Medical Center, and Dr. Allan Brett is a professor of medicine in the Division of General Internal Medicine at the University of South Carolina School of Medicine. (Dr. Brett also is the Editor-in-Chief of Journal Watch General Medicine, but the fact that he is a Journal Watch colleague has not influenced my opinion on this subject.)

The controversy focuses on whether the 2007 ACC/AHA guidelines for perioperative assessment and management of cardiac risk are flawed. This is an area of intense interest among practicing hospitalists, because we often are asked to risk-stratify inpatients before noncardiac surgery.

Compared with the 2002 guidelines, the 2008 guidelines contain substantial changes. Chief among these are the diminished roles of preoperative stress testing and coronary revascularization that are performed strictly for evaluating and lowering perioperative risk. Dr. Gregoratos describes the five-step algorithm of the 2007 guidelines as more straightforward than the cumbersome three-part, eight-step version in the 2002 guidelines.4 He admits that the new algorithm is ambiguous in some areas, but he maintains that it serves as a useful framework for clinicians who are charged with performing preoperative risk assessment and perioperative management.

Dr. Brett agrees that the 2007 guidelines are an improvement and that they result in fewer preoperative interventions; however, he stops short of praise and argues that the guidelines remained trapped in an algorithmic structure that, when followed, results in many procedures that do not improve perioperative outcomes. So, who is right?

To weigh the evidence, let’s first review changes in the new guidelines:

(1) Major clinical risk factors now are termed active cardiac conditions.

• Acute coronary syndromes

• Decompensated heart failure

• Clinically significant arrhythmias

• Severe valvular disease

(2) Previously defined intermediate-risk predictors have been replaced by five of the six variables contained in the revised cardiac-risk index (type of surgery is addressed separately).

• History of ischemic heart disease

• History of compensated or prior heart failure

• History of cerebrovascular disease

• Diabetes mellitus that requires insulin therapy

• Renal insufficiency (serum creatinine level, >2 mg/dL)

As in the 2002 guidelines, exercise capacity remains an important determinant of perioperative risk, elective major vascular surgery remains the surgery type with the highest associated risk, and low-risk surgery types (e.g., endoscopic procedures, superficial procedures, cataract surgery, breast surgery) carry such low adverse cardiac-event risk that even patients at high cardiac risk should proceed directly to surgery, provided that they do not have any active cardiac conditions.

(3) Recommendation classes for noninvasive preoperative stress testing now are:

• Class I — Benefit greatly exceeds risk: Patients with active cardiac conditions in whom noncardiac surgery is planned should be evaluated and treated in accordance with other relevant ACC/AHA guidelines before undergoing noncardiac surgery.

• Class IIa — Benefit exceeds risk: Patients with three or more clinical risk factors and poor functional capacity (<4 metabolic equivalents [METs]) and who require major vascular surgery can reasonably be tested if results could change management.

• Class IIb — Benefit is equal to, and might exceed, risk: The most controversial aspect of the new guidelines is the recommendation for patients with lower (but not negligible) risk, for whom the 2002 guidelines would have recommended noninvasive stress testing. The 2007 version advises physicians to proceed with planned surgery with heart-rate control but still offers noninvasive testing as a viable option if results could change management.

• Class III — Risk exceeds benefit: Testing still is not recommended for patients without cardiac risk factors and for patients who undergo low-risk surgery.

These recommendations leave a considerable amount of leeway for managing Class IIb patients. Should so much flexibility be part of the revised guidelines? As I discuss below, evidence since 2002 suggests that noninvasive cardiac testing and coronary revascularization do not improve perioperative outcomes in such patients who undergo noncardiac surgery.

In the Dutch Echocardiographic Cardiac Risk Evaluation Applying Stress Echo-II (DECREASE-II) study, researchers evaluated 770 intermediate-risk patients who were scheduled for vascular surgery and randomized them to noninvasive testing or no testing. At 30 days after noncardiac surgery, both groups had similar rates of cardiac death and myocardial infarction (1.8% and 2.3%; odds ratio, 0.78; 95% confidence interval, 0.28–2.1; P=0.62).5

The Coronary Artery Revascularization Prophylaxis (CARP) randomized trial involved 510 patients who were scheduled for major vascular surgery and who had at least one major coronary artery with stenosis >70%. Neither 30-day nor 2-year outcomes were better in the revascularization group than in the no-revascularization group.6 A smaller randomized controlled trial (DECREASE-V pilot study) of 101 higher-risk patients also failed to show benefit from coronary revascularization.7

Further support for optimal medical therapy in patients with stable coronary disease comes from the Clinical Outcomes Utilizing Revascularization and Aggressive Drug Evaluation (COURAGE) trial, published in 2007. Rates of short-term mortality and myocardial infarction were nearly identical in the optimal-medical-therapy group and in the percutaneous coronary intervention (PCI)-plus-medical-therapy group (18.5% and 19.0%; hazard ratio for PCI group, 1.05; 95% CI, 0.87–1.27; P=0.62).8

Finally, studies of PCI in patients who received bare-metal or drug-eluting stents have shown a higher rate of stent thrombosis in patients who do not receive dual antiplatelet therapy (with aspirin and clopidogrel) for the recommended duration after their procedures.9^,10^,11^,12 Stent placement might delay elective surgery substantially or lead to risk for early stent thrombosis if antiplatelet agents are discontinued.

Given these data, we can conclude that coronary revascularization offers no benefit to patients who do not have indications for revascularization independent of their noncardiac surgery. Additionally, scheduling PCI can delay surgery — a real issue for semi-elective procedures, such as cancer surgery or abdominal aortic aneurysm repair. Yet, the new guidelines do not steer clinicians strongly in this minimalist direction. The guidelines often recommend proceeding with surgery directly but also state that noninvasive testing might be warranted “if it changes management.” I find this statement to be self-evident, as, prior to ordering any test, clinicians should always ask themselves, Will the results change what I do? Dr. Gregoratos defends this position by stating that clinicians should use the algorithm as a “guide” and not as a “protocol” and that the treating physician needs to consider individual patient circumstances and use clinical judgment to decide when noninvasive cardiac testing is indicated.

Although I agree that sound clinical judgment is the cornerstone of good medical practice, clinicians should not be left to fend for themselves when evaluating stable patients for noncardiac surgery. The data clearly favor less cardiac testing. To have a large portion of the algorithm remain ambivalent is akin to telling clinicians, Do what you want. In that case, what purpose does the algorithm serve? Although these revised guidelines move us in the right direction, they definitely leave this preoperative consultant wanting more.