The choice of cardiac functional imaging will depend on the patient’s clinical findings and on local resources and expertise.
Stress echocardiography (SE), myocardial perfusion imaging (MPI) and stress cardiac magnetic resonance (CMR) provide information regarding the cardiac ischaemic burden under stress conditions.
Computed Tomography Coronary Angiography (CTCA) creates an anatomical image of the heart. It demonstrates plaques in the coronary artery walls and an indication of the degree of luminal stenosis.
Currently we have been equipped with clear and concise guidelines in identifying increased cardiovascular risk, preoperative assessment, perioperative therapy and management (Fleisher et al., 2014; Kristensen et al., 2014). It is still a daunting task to choose the right test for a patient. In this chapter, we will try to provide brief guidance on the pros and cons of available imaging techniques.
In many instances, local expertise and resources dictate the choice of cardiovascular imaging (Table 5.E.1). All functional tests, echocardiography, myocardial perfusion imaging (MPI) and cardiac magnetic resonance (CMR) can provide information on cardiac function at rest and ischaemic burden under stress. Both stress echocardiography (SE) and MPI have been well established with the recent useful addition of stress perfusion CMR. There are currently no randomised trials comparing SE, MPI and stress CMR perioperatively. A retrospective, meta-analysis comparing MPI (thallium imaging) and SE in patients scheduled for elective non-cardiac surgery showed that a moderate-to-large defect detected by either method predicted post-operative cardiac events (Beattie et al., 2006), with SE having a minor advantage. Invasive and non-invasive imaging to determine coronary anatomy has not received the endorsement of European or American guidelines (Fleisher et al., 2014; Kristensen et al., 2014).
Legend: MPI = myocardial perfusion imaging, CMR = cardiac magnetic resonance, CT Angiography = computed tomography angiography, MRI = magnetic resonance imaging, COPD = chronic obstructive airways disease, MI = myocardial infarction, LBBB = left bundle branch block, ESRF = end-stage renal failure, min = minute, + minor, ++ medium, +++ major
SE can be performed utilising both pharmacological stimulation and exercise. First of all, it provides clinicians with valuable data on resting echocardiography. It is well known that left ventricular systolic dysfunction at rest or severe valvular heart disease (Flu et al., 2010) are associated with a worse outcome. Identification of LV geometry prone to development of dynamic gradients (Cha et al., 2014) is advantageous in adequate fluid management perioperatively. Also, our understanding of diastolic heart failure has improved vastly and prognostic data are emerging in some reports (Matyal et al., 2009). In the presence of tricuspid regurgitation, it is easy to pick up significant pre-existing pulmonary hypertension, which is associated with high mortality, morbidity and complication rates secondary to cardiac and/or respiratory failure (Kaw et al., 2011). It is important to specify in the request form that the full data on the resting parameters are required if the aforementioned information is not known.
DSE is safe and feasible as part of a preoperative assessment: a positive test result was reported in the range of 5 per cent to 50 per cent. In these studies, with event rates of 0 per cent to 15 per cent, the ability of a positive test result to predict an event (nonfatal MI or death) ranged from 0 per cent to 37 per cent. Perhaps such a wide range was determined by the absence of a uniform definition of an abnormal test. In some studies, it was restricted to the presence of new wall motion abnormalities with stress as a marker of myocardial ischaemia, but in other studies, the presence of a fixed myocardial defect was included. The latter might have altered statistics.
It has become clear on follow-up studies that the presence of an old MI identified on rest imaging is of little predictive value for perioperative MI or cardiac death. Furthermore, the normal stress protocol for non-cardiac surgery screening does not differ from the ones used for patients who present with chest pain and performed off beta-blockers, which could potentially be a source of compliance bias. Overall, the strength of DSE lies with the negative predictive value, which is invariably high, typically in the range of 90 per cent to 100 per cent.
Exercise SE (ESE) is not widely used in preoperative assessment for non-cardiac surgery. Potentially the combination of physiological stress and information on exercise capacity, blood pressure and heart rate response can provide a more powerful positive predictive value. However, the acquisition of echo signals during exercise is technically more demanding and requires a certain level of patient mobility as the majority of centres use a treadmill rather than a supine veloergometer.
Overall SE is a comprehensive, low-cost, non-invasive risk-assessment tool. It does not require additional equipment apart from echo machine and monitors for vital observations; the maintenance cost is minimal. Echo windows could be difficult in patients with COPD, obesity or extreme chest deformity, but this is largely improved with the use of echo contrast agents. Dobutamine should be avoided in patients with serious ventricular arrhythmias or recent MI. The technique is highly operator dependant. Echocardiographers must remain high-volume operators to maintain diagnostic skills (100 studies/year) (Popesku et al., 2009). Usually, the studies are performed and reported by a cardiologist, which adds the value of clinical consultation on top of diagnostic findings.
Myocardial Perfusion Imaging
In general, an abnormal MPI test is associated with very high sensitivity for detecting patients at risk for perioperative cardiac events. The negative predictive value of a normal MPI study is also high for MI or cardiac death, although post-operative cardiac events do occur in this population (11). This might be related to cases of severe CAD where balanced ischaemia with even distribution of the radionuclide tracer is not calibrated against resting images. Most studies have shown that a fixed perfusion defect, which reflects infarcted myocardium, has a low positive predictive value for perioperative cardiac events but is predictive of long-term cardiac events.
In patients with left bundle branch block, perfusion defect in the septum secondary to ventricular dyssynchrony could reduce the specificity of the test. Intravenous dipyridamole and adenosine should be avoided in patients with significant heart block, bronchospasm, critical carotid occlusive disease or a condition that prevents theophylline preparations being withdrawn.
MPI is a highly sensitive technique in the assessment of perioperative risk, but could require long-day or two-day imaging in positive cases. Severe ischaemia could be missed if other indirect signs are overlooked: transient ischaemic LV dilatation, reduction in LVEF with stress; be aware of breast attenuation artefacts. There is radiation exposure, although relatively low per test the cumulative value could go up with multiple tests. Overall study cost as well as maintenance costs remain high. There is no problem with operator dependency, but reporting requires highly skilled specialists. Outside the routine list of contraindications for pharmacological stressors there are no limiting factors for this test.
Stress Cardiac Magnetic Resonance
CMR is an actively evolving area in diagnostic cardiac imaging which has the advantage in distinguishing ischaemic from non-ischaemic cardiomyopathies (Pfeiffer and Biederman, 2015); it is essential for diagnosing hypertrophic cardiomyopathy and infiltrative heart conditions. Stress perfusion CMR has been reported to be superior to MPI (Greenwood et al., 2012; Schwitter et al., 2008) and to have excellent negative predictive value (96–99%) (Freed et al., 2013). There are currently insufficient data to support the use of stress CMR in preoperative risk assessment, but this is a rapidly evolving field.
No radiation exposure is involved. CMR is generally safe even in the presence of coronary or peripheral artery stents, sternal wires and modern prosthetic cardiac valves. It is unsafe with ocular metal shavings and non-MRI conditional pacemakers/defibrillators. CMR remains expensive and time-consuming, and not widely available. Nevertheless, it appears to be more cost-effective than MPI. Image quality easily degenerates in the presence of atrial fibrillation, frequent ventricular ectopy and dyspnoea. In patients with end-stage renal failure, contrast-related complication (systemic nephrogenic fibrosis) has been reported and is generally not advised when the estimated glomerular filtration rate is less than 30 mL/min (Grobner and Prischl, 2007).
Presently stress CMR can be used for more detailed evaluation of important cardiac structure and functional parameters, particularly in patients with poor echo windows.
Computed Tomography Coronary Angiography
CTCA has found a steady footing in diagnostic routine for chest pain patients with low to intermediate pre-test probability due to its high sensitivity and negative predictive value for excluding significant CAD (Montalescot et al., 2013; Wolk et al., 2014). Unlike the stress imaging techniques which provide information about myocardial ischaemia, CTCA is an anatomical technique with the ability to demonstrate the plaques in the wall of the coronary arteries and to assess the degree of luminal stenosis. Three recent studies have attempted to demonstrate the value of CTCA in assessment of risk for non-cardiac surgery (Hwang et al., 2015; Sheth et al., 2015).
One of these (CTA VISION trial) was a large, prospective multi-centre trial in 955 patients undergoing non-cardiac surgery which demonstrated that ≥50% stenosis in two or more coronary arteries on CTCA provides independent prognostic information with respect to 30-day post-operative mortality and non-fatal MI (Sheth et al., 2015). This study has also demonstrated that 28 per cent of events occurred in patients without obstructive CAD, demonstrating that the mechanism of post-operative cardiac events could be multi-factorial (54). It is therefore not surprising that risk prediction improved when obstructive disease on CTCA was combined with revised clinical risk score (54, 52).
Summary and Recommendations
Functional imaging remains the most tested and validated technique for risk stratification of patients prior to non-cardiac surgery. The choice of imaging is largely defined by local expertise and resources. When multiple imaging modalities are available, the limitations of the certain techniques mentioned previously and patient specifics are important factors to consider rather than the advantage of one test over the other (Fig. 5.E.1). Authors suggest to use SE as a first-line imaging test in the presence of good echo windows. In patients free of ferromagnetic material in the body as well as reasonable renal function, stress CMR will be a good second-line imaging option. MPS can provide good functional data if a patient proves claustrophobic. Anatomical imaging remains of limited value, although this might change in the future.