Safer non-cardiac surgery for patients with coronary artery disease

How best to manage patients with coronary artery disease who undergo major non-cardiac surgery is an increasingly important issue as the population ages. Such patients, particularly those with easily induced ischaemia, are at increased risk of perioperative cardiac complications and death.¹ Various pre-emptive interventions have been considered to minimise this risk, but often their precise role is poorly defined.

Coronary artery bypass grafting is effective but carries its own risks, and overall survival benefit is seen only in patients who warrant bypass surgery independently of their major non-cardiac operation.² These patients, although few, are a well defined³ population who should be offered prophylactic coronary revascularisation. The role of percutaneous transluminal coronary angioplasty is less well defined because, even in the wider population of patients with coronary artery disease, no prospective randomised trial has shown a prognostic benefit for angioplasty over medical treatment. Use of preoperative angioplasty should therefore be restricted to patients with readily inducible ischaemia, in whom a single coronary stenosis subtends a large area of viable myocardium.⁴

Most patients with coronary artery disease presenting for elective major non-cardiac surgery do not have disease severe enough to justify the risks of prophylactic cardiac catheterisation or coronary revascularisation, and for these perioperative intensification of medical treatment should be more widely considered. Various options are available, including β blockers, aspirin, calcium antagonists, nitrates, α₂ agonists, heparin, and newer agents such as the potassium channel activators. Unfortunately there is only limited direct randomised evidence on the efficacy of these interventions in the perioperative period, although inferences can be made from studies in other fields.

The most compelling direct evidence supports the perioperative use of β blockers. Administered perioperatively they reduce the amount of silent myocardial ischaemia detected by S-T segment analysis of electrocardiograms,⁵ and recently atenolol has been shown to reduce mortality and improve event free survival for up to two years after major non-cardiac surgery.⁶ In a non-operative setting β blockers reduce the size of, and mortality from, myocardial infarction as well as increasing event free survival in patients with chronic stable angina.⁷ β Blockers probably help by obtunding the inotropic and chronotropic effects of excess sympathetic stimulation, thereby reducing myocardial oxygen requirements and, importantly, increasing diastolic coronary perfusion time. Although many patients with coronary artery disease undergoing major surgery are already taking β blockers, the dose is often suboptimal because of concerns over bradycardia during daily living and therefore there will be an opportunity to intensify treatment. Alternatively α₂ agonists could be considered. They modify perioperative ischaemia via their centrally mediated reduction in sympathetic outflow.⁸

Aspirin has not been properly investigated as means of reducing perioperative cardiac complications, but strong indirect evidence supports its use. Aspirin undoubtedly has a major role in the primary and secondary prevention of myocardial infarction.⁹ It also reduces the severity of silent myocardial ischaemia in both stable¹⁰ and unstable angina. These effects are mediated through its antiplatelet actions, and because the perioperative period is associated with increased platelet reactivity aspirin may well be particularly useful at this time. Paradoxically antithrombotic treatment is often withdrawn before major surgery because of a perceived increased risk of bleeding. Although aspirin interferes with platelet aggregation induced by thromboxane A₂, it has no effect on that induced by either thrombin or high concentrations of collagen,¹¹,¹² and therefore clinically significant bleeding should not be made worse by perioperative aspirin.¹³

Calcium antagonists, nitrates, and the potassium channel activator nicorandil all delay the onset of ischaemia during exercise testing or reduce the amount of silent myocardial ischaemia recorded during ambulatory monitoring¹⁴ and therefore may be useful perioperatively. However, the evidence for these is less convincing than that for β blockers or aspirin. Also, because they have vasodilating or negative inotropic effects, which may be associated with a reflex tachycardia, these drugs can further compromise coronary perfusion and reduce the patient’s ability to cope with major fluid shifts or haemorrhage.

Intravenous therapeutic doses of heparin undoubtedly improve outcome in patients with unstable angina. Heparin’s use perioperatively is, of course, complicated by the risk of haemorrhage, but its use postoperatively should not be instantly dismissed, especially in very high risk patients (who have the most to gain from aggressive intervention). Low dose subcutaneous heparin is already widely used to prevent development of perioperative deep vein thrombosis and pulmonary embolism and may also modify the severity of myocardial ischaemia.

Thus with its low cost, ease of administration, and relatively low intrinsic risk, optimising medical treatment is probably the best option for most patients with coronary artery disease undergoing major surgery. The strongest direct evidence supports the perioperative use of β blockers, and these should be considered in all patients thought to be at substantial risk of cardiac complications. The use of other drugs—particularly aspirin—needs to be studied in randomised controlled trials large enough to document reliably the balance of benefit and risks.

Finally, stable haemodynamics throughout the whole perioperative period are essential to ensure a successful outcome: anti-ischaemic medication is only one facet of this. The most important determinants of cardiovascular stability are an anaesthetic technique that obtunds the pathophysiological responses induced by surgery, combined with excellent control of pain, fluid balance, and oxygenation postoperatively. All these processes must be monitored properly. The best environment for this is a high dependency unit, which also provides the ideal setting from which to assess the effects of prophylactic medical interventions.