Noninvasive evaluation of ischaemic heart disease: myocardial perfusion imaging or stress echocardiography?

Stress echocardiography and myocardial perfusion imaging are commonly used noninvasive imaging modalities for the evaluation of ischaemic heart disease. Both modalities have proved clinically useful in the entire spectrum of coronary artery disease.1–29 Both techniques can detect coronary artery disease and provide prognostic information.1–21 Both techniques can identify low-risk and high-risk subsets among patients with known or suspected coronary artery disease and thus guide patient management decisions.18–21 In patients with acute myocardial infarction, both techniques have been used to identify residual viable tissue and predict improvement of function over time.22–26 In patients with chronic ischaemic left ventricular (LV) dysfunction, viability assessment with either modality can be used to predict improvement of function after revascularisation and thus guide patient treatment.27–29 Hence, the use of noninvasive cardiac imaging can help guide management and potentially reduce healthcare costs.30 The question remains what is the optimal noninvasive cardiac imaging method in which setting? This article evaluates the value of the two modalities in: (1) the detection of coronary artery disease, (2) the prognosis of coronary artery disease in patients with known or suspected coronary artery disease, (3) prediction of functional recovery following acute myocardial infarction and (4) prediction of functional recovery after revascularisation in patients with chronic ischaemic LV dysfunction. To provide the most objective information, only direct comparative studies on stress echocardiography and perfusion imaging in the same patients are included and pooled analysis of the data was performed. The available studies were identified by MEDLINE searches using the following key words: noninvasive imaging, stress echocardiography, dobutamine, dipyridamole, adenosine, myocardial perfusionimaging, technetium-99m sestamibi, technetium-99m tetrofosmin and thallium-201. In addition, a manual search of eight cardiology and nuclear medicine journals (American Heart Journal, American Journal of Cardiology, Circulation, European Heart Journal, Heart, Journal of the American College of Cardiology, Journal of Nuclear Cardiology, Journal of Nuclear Medicine) from January 1975 to 2001 was carried out. Only studies that performed a head-to-head comparison between stress echocardiography and some form of nuclear imaging were selected. From these articles the sensitivity and specificity of the techniques were compared. Studies that did not provide this information were excluded. From the pooled data, weighted sensitivities and specificities were calculated. Comparison of sensitivities and specificities was performed using McNemar testing; a P-value <0.05 was considered significant. Myocardial perfusion imaging vs stress echocardiography in the diagnosis of coronary artery disease MPI=myocardial perfusion imaging; Tc-99m=Technetium-99m; Tl-201=Thallium-201 chloride. Myocardial perfusion imaging vs stress echocardiography in the diagnosis of coronary artery disease MPI=myocardial perfusion imaging; Tc-99m=Technetium-99m; Tl-201=Thallium-201 chloride. The ischaemic cascade represents a sequence of pathophysiologic events caused by coronary artery disease. Nuclear imaging probes an earlier event (hypo-perfusion) in the ischaemic cascade than stress echocardiography does (systolic dysfunction). Sensitivity and specificity of stress echocardiography and nuclear imaging for the detection of coronary artery disease (data based on Refs. 1–17). □=echocardiography; ▪= nuclear. Pharmacological stress can be a useful alternative to exercise stress protocols in patients who are unable to exercise because of neurological, orthopedic, or peripheral vascular disease. Because wall motion abnormalities are a consequence of myocardial ischaemia, dobutamine may be more effective than vasodilator (adenosine or dipyridamole) stress echocardiography.32 In line with this, combined data from seven direct comparative studies demonstrated that dobutamine stress echocardiography had a higher sensitivity for the diagnosis of coronary artery disease than vasodilator stress echocardiography, while specificity was similar. When dobutamine stress echocardiography was compared to dobutamine or vasodilator perfusion scintigraphy, dobutamine echocardiography was equally sensitive but slightly more specific than perfusion scintigraphy.33 Two subgroups of patients were analysed separately: hypertensive patients and female patients. In patients with hypertension, abnormal thallium perfusion results have been demonstrated in the absence of obstructive coronary artery disease;34,35 this may lower specificity. Summarised data from two studies (Table 2 Myocardial perfusion imaging vs stress echocardiography in special patient subsets MPI=myocardial perfusion imaging; Tc-99m=Technetium-99m; Tl-201=Thallium-201 chloride. Myocardial perfusion imaging vs stress echocardiography in special patient subsets MPI=myocardial perfusion imaging; Tc-99m=Technetium-99m; Tl-201=Thallium-201 chloride. The diagnosis of coronary artery disease in women may be more challenging due to the lower prevalence of coronary artery disease. In addition, single-vessel disease is a common finding in women.38,39 The accuracy of perfusion imaging appears to be decreased in women with breast tissue attenuation,39 and the smaller LV chamber size in women.40 Pooled data from three direct comparisons41–43 revealed a comparable sensitivity of the two techniques (71 vs 80%, \(P=ns\) ⁠) (175 patients, Table 2) with a higher specificity of stress echocardiography for the detection of coronary artery disease (72 vs 89%, \(P{<}0.01\) ⁠). Comparative studies on adenosine stress imaging in patients with hypertension and women are not available. Further research on the relative value of adenosine stress echocardiography and nuclear perfusion imaging in these subgroups is needed. Noninvasive cardiac imaging is frequently used for risk stratification of patients with known or suspected coronary artery disease. There are twodirect comparisons available on the prognostic value of myocardial perfusion imaging and stress echocardiography. Geleijnse et al.18 studied 220 patients with chest pain with dobutamine–atropine stress echocardiography and simultaneous technetium-99m sestamibi single photon emission computed tomography (SPECT) imaging. During follow-up of 31±15 months, 24 patients had hard cardiac events (nonfatal myocardial infarction or cardiac death). A normal test was related to a good prognosis, with a low annual cardiac event rate of 0.4% by echocardiography and 0.5% by perfusion imaging. In that study, stress echocardiography and technetium-99m sestamibi SPECT provided comparable prognostic information. Olmos et al.19 studied 248 patients who underwent exercise echocardiography simultaneously with thallium-201 SPECT. During follow-up(obtained in 225 patients with a mean follow-up of 3.7±2.0 years), 64 cardiac events occurred (eight nonfatal infarctions and seven cardiac deaths). A significant difference was observed between patients with normal and abnormal tests for all end points, including death alone, for both modalities. Overall cardiac event rate in patients with normal test results was comparable for both exercise echocardiography and thallium-201 SPECT (1.05 vs 1.13%, ns). Annual cardiac death rate was favorably low for both normal echocardiography and normal SPECT imaging (0.08 vs 0.08%, ns). Since only two direct comparisons are available, more large studies are required to fully elucidate the relative prognostic value of myocardial perfusion imaging and stress echocardiography. The hallmark of viability on dobutamine echocardiography is the improvement of wall motion during the infusion of low-dose dobutamine(5–10μgkg−1min−1). More recent studies have employed a low-high dose protocol (with dosages up to 40μgkg−1min−1, with the addition of atropine). This protocol allows assessment of both viability (response during low-dose dobutamine) and stress-induced ischaemia (response during high-dose dobutamine). For nuclear imaging, different techniques are available. Thallium-201 imaging can be used to evaluate perfusion and cell membrane integrity. Two protocols are used mainly: rest-redistribution imaging and stress-redistribution-reinjection imaging.27–29 While rest-redistribution imaging provides only information on myocardial viability, the reinjection protocol allows assessment of stress-induced ischaemia and viability. Technetium-99m sestamibi can be used to assess perfusion and intact mitochondria. Sestamibi imaging is performed under resting conditions; in the absence of a stress study, this protocol only provides information on viability. It has been demonstrated that the addition of nitrates before tracer administration enhances viabilitydetection.44 Glucose utilisation can be evaluated with F18-fluorodeoxyglucose (FDG). FDG imaging can nowadays be performed with positron emissiontomography and SPECT. The introduction of SPECT imaging has contributed to a more widespread use of FDG.45 Generally, cardiac FDG uptake is compared with regional perfusion. Viability is defined when perfusion/FDG uptake is normal, or when perfusion is reduced with enhanced FDG uptake. Prediction of recovery of function after acute myocardial infarction: myocardial perfusion imaging vs stress echocardiography HDDE=low-high dose dobutamine echocardiography; LDDE=low-dose dobutamine echocardiography; MPI=myocardial perfusion imaging; RI=reinjection; RR=rest-redistribution; Tl-201=thallium-201 chloride. Prediction of recovery of function after acute myocardial infarction: myocardial perfusion imaging vs stress echocardiography HDDE=low-high dose dobutamine echocardiography; LDDE=low-dose dobutamine echocardiography; MPI=myocardial perfusion imaging; RI=reinjection; RR=rest-redistribution; Tl-201=thallium-201 chloride. Sensitivity and specificity of dobutamine stress echocardiography and nuclear imaging for the prediction of functional outcome in acute myocardial infarction (data based on Refs. 22–26). □=DSE; ▪=nuclear. Accuracy of the different viability techniques (data based on pooled analysis of data available in the literature29) DSE=dobutamine stress echocardiography; FDG=F18-fluorodeoxyglucose; MIBI=technetium-99m sestamibi; NPV=negative predictive value; PET=positron emission tomography; PPV=positive predictive value; RR=rest-redistribution; RI=reinjection; Tl-201=thallium-201 chloride. Accuracy of the different viability techniques (data based on pooled analysis of data available in the literature29) DSE=dobutamine stress echocardiography; FDG=F18-fluorodeoxyglucose; MIBI=technetium-99m sestamibi; NPV=negative predictive value; PET=positron emission tomography; PPV=positive predictive value; RR=rest-redistribution; RI=reinjection; Tl-201=thallium-201 chloride. Agreement and disagreement between FDG SPECT and low-dose dobutamine echocardiography in dysfunctional myocardium (reprinted with permission from Cornel et al.49). Head-to-head comparisons between nuclear imaging and dobutamine echocardiography for the prediction of improvement of regional function post-revascularisation (table modified from Ref. 29) FDG=F 18-fluorodeoxyglucose; HDDE=low-high dose dobutamine echocardiography; I=stress-inducible ischaemia; LDDE= low-dose dobutamine echocardiography; PET=positron emission tomography; RI=reinjection; RR=rest-redistribution; Tl-201= thallium-201 chloride; V=viability. Head-to-head comparisons between nuclear imaging and dobutamine echocardiography for the prediction of improvement of regional function post-revascularisation (table modified from Ref. 29) FDG=F 18-fluorodeoxyglucose; HDDE=low-high dose dobutamine echocardiography; I=stress-inducible ischaemia; LDDE= low-dose dobutamine echocardiography; PET=positron emission tomography; RI=reinjection; RR=rest-redistribution; Tl-201= thallium-201 chloride; V=viability. Bar graph illustrating sensitivities and specificities of nuclear imaging and dobutamine stress echocardiography to predict improvement of function post-revascularisation inpatients with chronic ischaemic left ventricular dysfunction (data based on Refs. 50–52,58,60–62). □=DSE; ▪=nuclear. It is conceivable that different levels of ultrastructural cell damage account for this discrepancy: the inotropic response during dobutamine stimulation may be lost while more basal characteristics such as cell membrane integrity and glucose utilisation are still intact. Besides prediction of improvement of function after revascularisation, prediction of long-term survival may be more important. Currently, no direct comparativestudies are available on the prognostic value of nuclear imaging and stress echocardiography in patients with ischaemic cardiomyopathy. Recently, different technical developments have been implemented in both perfusion imaging and stress echocardiography that further optimised the techniques. Myocardial perfusion imaging has been improved by the introduction of new cameras, imaging protocols, attenuation and scatter correction.68,69 For echocardiography, second harmonic imaging has markedly improved endocardial border detection, which could even be further enhanced by intravenous administration of contrast agents that opacify the LV cavity.70–72 Colour kinesis and tissue Doppler echocardiography may allow quantification of function,73,74 which may enhance reproducibility and diagnostic accuracy. Other recent developments have aimed at providing integrated information on function and perfusion. Both gated SPECT imaging and contrast echocardiography allow simultaneous assessment of function and perfusion.75,76 Over the past 5 years ECG-gated SPECT has become state-of-the-art in cardiac myocardial perfusion imaging. Various comparative studies have demonstrated excellentaccuracy of gated SPECT for the assessment of both regional and global LV function.77 Smanio et al.78 have demonstrated that combination of perfusion and function (assessed by gating) resulted in significantly higher diagnostic accuracy for detection/exclusion of coronary artery disease. Moreover, the integrated information of function and perfusion allows superior prognostification in patients with known or suspected coronary artery disease, as demonstrated recently by Sharir et al.79 Contrast echocardiography also allows simultaneous assessment of function and perfusion. With the introduction of contrast agents that can be administered intravenously, the use of contrast echocardiography has now become feasible outside the catheterisation laboratory. In combination with harmonic and intermittent imaging, quantitative assessment of perfusion has become possible. Recently, Kaul et al.80 have shown an excellent concordance between contrast echocardiography and myocardial perfusion imaging. In 30 patients, the agreement for detecting absence/presence of coronary artery disease was 86%. However, two multicentre trials have demonstrated a less favourable agreement between perfusion imaging and contrast echocardiography.81,82 Marwick et al.81 showed that, for the detection of extensive perfusion defects on SPECT perfusion imaging, contrast echocardiography had a sensitivity ranging from 13 to 48% with a specificity ranging from 63 to 100%. Jucquois et al.82 suggested that the discrepancybetween the two techniques could in part be related to attenuation artifacts on SPECT (inferior wall) and suboptimal visualisation of the anterior and lateral walls by echocardiography. However, in experienced settings, adequate quantification of myocardial blood flow and quantification of coronary artery stenoses is possible, as demonstrated by Wei and colleagues.83,84 Besides detection of coronary artery disease, evaluation of patients with acute myocardial infarction is feasible with contrast echocardiography and is particularly useful for assessment of ‘the no-reflow phenomenon’ after thrombolysis or percutaneous interventions.85,86Finally, the feasibility of contrast echocardiography for assessment of viability after acute infarction87 and in the setting of chronic LV dysfunction88 has been demonstrated. Thus, in the near future, the combined assessment of perfusion and function by echocardiography and SPECT will allow a more complete assessment of patients with coronary artery disease. Both myocardial perfusion imaging and stress echocardiography have proved to be extremely useful noninvasive tests for the evaluation of coronary artery disease.1–29 Due to basic differences between the two tests, perfusion imaging is asensitive test with a somewhat lower specificity for the detection of coronary artery disease; the converse is true for stress echocardiography.1–17 This systematic review focuses on direct comparative studies on stress echocardiography and perfusion imaging in order to provide the most objective information. Nevertheless, a potential risk of pooling data from different studies is to mix patients with different clinical characteristics and risk profile. Two available direct head-to-head comparative studies demonstrated a similar prognostic value of perfusion imaging and stress echocardiography. Larger comparisons are needed to draw further conclusions. For the assessment of myocardial viability after acute infarction the modalities seem to be equally sensitive, whereas stress echocardiography is the more specific test.22–26 Hence, for the early assessment of viability stress echocardiography may be preferable. However, specificity is determined by segments that are nonviable that do not improve in function. A lower specificity suggests that a substantial percentage of segments that are viable do not recover in function. Most studies haveevaluated recovery of function at a rather short time-interval after infarction (<3 months) and longer follow-up may be needed. In patients with chronic ischaemic ventricular dysfunction, nuclear imaging has a high sensitivity for the detection of viable myocardium and a low specificity, whereas the converse is true for stress echocardiography.50–67 The lower specificity of nuclear imaging can again be an issue of duration of follow-up. Recent data have demonstrated that a substantial percentage of segments need longer time after revascularisation to (fully) recover in function.89 In addition, large direct comparative studies are required to evaluate the prognostic value of nuclear imaging and stress echocardiography in patients with chronic ischaemic left ventricular dysfunction. In conclusion, the current analysis demonstrated that both techniques are useful in the evaluation of patients with coronary artery disease, although small differences between accuracies exist in different settings. The most important factor for using a test remains the local expertise and availability of these imaging modalities. In addition, patient characteristics (habitus, acoustic window, pregnancy) may influence the choice of the technique, and finally the studies discussed generally reflect the experience in university centers and many of these studies may be influenced by selection and referral bias, which limits application of the results to the general population.