Curr Cardiol Rep (2014) 16:482 DOI 10.1007/s11886-014-0482-7
CARDIAC PET, CT, AND MRI (SE PETERSEN, SECTION EDITOR)
New Insights from Major Prospective Cohort Studies with Cardiac Nuclear Imaging Oliver Gaemperli
Published online: 20 March 2014 # Springer Science+Business Media New York 2014
Abstract For more than two decades, radionuclide imaging has prevailed as a cornerstone in the diagnosis and treatment of patients with cardiac disease. From this experience, large cohort studies have emerged that demonstrate the prognostic value of cardiac radionuclide imaging in a variety of patient groups and conditions. Myocardial perfusion imaging has accrued the most robust evidence for accurate and independent risk stratification over traditional clinical variables. In a variety of patient populations, the presence of myocardial ischemia is a strong predictor of cardiac events on followup. In patients with heart failure, smaller observations have similarly established the prognostic value of viability imaging and imaging of cardiac sympathetic activity. The present review provides a summary of recent cohort studies with radionuclide imaging and a critical appraisal of their clinical implications. Its purpose is to put the available evidence into a clinical context, analyze its potential impact on patient management and identify gaps in knowledge and unanswered questions to be addressed in future randomized trials.
Keywords Cardiac radionuclide imaging . Single photon emission computed tomography . Positron emission tomography . Prognosis . Cardiac nuclear imaging . Prospective cohort studies
This article is part of the Topical Collection on Cardiac PET, CT, and MRI O. Gaemperli Cardiac Imaging and Interventional Cardiology, University Hospital Zurich, Zurich, Switzerland O. Gaemperli (*) Nuclear Cardiology, Cardiovascular Center, University Hospital Zurich, C Hof 109, Raemistrasse 100, CH-8091 Zurich, Switzerland e-mail: [email protected]
Abbreviations CACS coronary artery calcium score CAD coronary artery disease CFR coronary flow reserve ECG electrocardiogram 18 FDG F-fluorodeoxyglucose HF heart failure HMR heart-to-mediastinum ratio LV EF left ventricular ejection fraction MBF myocardial blood flow MI myocardial infarction 123 MIBG I-metaiodobenzylguanidine MPI myocardial perfusion imaging PET positron emission tomography SPECT single photon emission computed tomography
Introduction Every diagnostic test applied in clinical medicine serves a similar purpose: It is used by clinicians to identify suspected disease, to assess severity and prognosis of the disease (usually based on the magnitude of a deranged value), and finally to take action and implement therapies that are able to reverse patient’s outcome. Thus, in order to find acceptance in clinical practice, each clinical test has to undergo a series of steps of clinical validation relating to the following: (1) initial validation of diagnostic accuracy, (2) assessment of prognostic value, (3) impact on patient management and improvement of outcome, and finally (4) cost-effectiveness. Since its introduction in the early 1980s, cardiac radionuclide imaging has played an increasingly important role in the diagnosis and management of patients with cardiac disease. The nuclear cardiology community has done an excellent job
482, Page 2 of 11
in accumulating diagnostic and prognostic data to endorse its clinical value. Based on this evidence, radionuclide imaging has been embraced around the globe as a useful tool to diagnose and stratify patients with cardiac disease and is applied in clinical practice to several subgroups such as patients with cardiovascular risk factors, with suspected or angiographically proven coronary artery disease (CAD), after revascularization or in heart failure (HF) patients. National and international data review bodies have adopted cardiac radionuclide imaging in their guidelines and recommendations for the treatment of patients with the aforementioned conditions [1–5]. Recently, a number of publications and follow-up reports of large cohort studies with radionuclide imaging have been published, adding relevant information. The purpose of the present review is to summarize the data from recent cohort studies, to critically appraise their clinical implications and to assess whether they add new aspects to current clinical practice. Another purpose of this manuscript is to put the available evidence into a clinical context, analyze its potential impact on patient management and identify gaps in knowledge and unanswered questions to be addressed in future randomized trials. For this purpose, the present review focuses on evidence available from radionuclide imaging in four major areas of cardiovascular disease: & & & &
asymptomatic patients (at risk for silent ischemia) symptomatic CAD microvascular dysfunction heart failure (HF).
Screening of Asymptomatic Subjects Approximately 50 % of myocardial infarctions (MI) occur in patients who were asymptomatic until the day of their event, of which 48 % have only one or even none of the traditional risk factors [6]. Therefore, effective screening tools are sought to reduce the high toll of unexpected coronary events by deriving more individual risk estimates than traditional risk factors. Diabetic patients belong to a high-risk category in which screening for asymptomatic CAD may be beneficial [7]. The prevalence of abnormal myocardial perfusion imaging (MPI) studies among asymptomatic diabetic patients varies from 21 to 58 %, with 15 to 20 % having high-risk MPI findings [8–10]. In the Detection of Ischemia in Asymptomatic Diabetics (DIAD) study, 1123 asymptomatic subjects with type 2 diabetes were randomly assigned to undergo screening adenosine-stress radionuclide MPI or standard care (including intensive medical therapy for risk factor modification) and were followed-up for 4.8 years [11]. Overall, cardiac event rates (cardiac death or nonfatal MI) for the entire study
Curr Cardiol Rep (2014) 16:482
population were much lower than anticipated (0.6 % per year). In the screened group, MPI was able to accurately risk stratify patients based on perfusion findings with higher event rates in patients with moderate or large perfusion defects compared to those with only mildly abnormal or normal MPI (2.4 % per year vs 0.4 % per year (hazard ratio (HR), 6.3; 95 % confidence interval (CI), 1.9-20.1; P=.001)). However, screening with radionuclide MPI did not improve the prognosis of the patients with comparable event rates in both groups (2.7 % in the screened group vs. 3.0 % in the not-screened group (P=.73). Due to the low overall risk of the study cohort, the results of the DIAD study cannot be generalized to all asymptomatic diabetics, and therefore the issue of screening asymptomatic diabetics remains unresolved [3]. Others have proposed to use coronary artery calcium scores (CACS) as a gatekeeper to identify asymptomatic subjects at risk for whom further screening radionuclide MPI may be justified [12, 13]. Indeed, abnormal MPI is rare below a CACS threshold of 100, but the prevalence of abnormal scans increases progressively with higher CACS. In 411 largely asymptomatic subjects, He and colleagues reported a rate of only 2.6 % abnormal MPI scans below a CACS of 100, while the prevalence of abnormal MPI rose to 11.3 % and 46 % for CACS of 101-399, and >400, respectively [14]. Berman and colleagues analyzed 1195 subjects of which 51 % were asymptomatic [15]. The frequency of ischemic MPI was 0.80) which were entered into a registry. The prospective nuclear substudy of the COURAGE trial, albeit slightly underpowered to detect significant differences between revascularization versus medical treatment, revealed improved outcomes in patients with ≥5 % reduction in ischemic burden compared to those with
CARDIAC PET, CT, AND MRI (SE PETERSEN, SECTION EDITOR)
New Insights from Major Prospective Cohort Studies with Cardiac Nuclear Imaging Oliver Gaemperli
Published online: 20 March 2014 # Springer Science+Business Media New York 2014
Abstract For more than two decades, radionuclide imaging has prevailed as a cornerstone in the diagnosis and treatment of patients with cardiac disease. From this experience, large cohort studies have emerged that demonstrate the prognostic value of cardiac radionuclide imaging in a variety of patient groups and conditions. Myocardial perfusion imaging has accrued the most robust evidence for accurate and independent risk stratification over traditional clinical variables. In a variety of patient populations, the presence of myocardial ischemia is a strong predictor of cardiac events on followup. In patients with heart failure, smaller observations have similarly established the prognostic value of viability imaging and imaging of cardiac sympathetic activity. The present review provides a summary of recent cohort studies with radionuclide imaging and a critical appraisal of their clinical implications. Its purpose is to put the available evidence into a clinical context, analyze its potential impact on patient management and identify gaps in knowledge and unanswered questions to be addressed in future randomized trials.
Keywords Cardiac radionuclide imaging . Single photon emission computed tomography . Positron emission tomography . Prognosis . Cardiac nuclear imaging . Prospective cohort studies
This article is part of the Topical Collection on Cardiac PET, CT, and MRI O. Gaemperli Cardiac Imaging and Interventional Cardiology, University Hospital Zurich, Zurich, Switzerland O. Gaemperli (*) Nuclear Cardiology, Cardiovascular Center, University Hospital Zurich, C Hof 109, Raemistrasse 100, CH-8091 Zurich, Switzerland e-mail: [email protected]
Abbreviations CACS coronary artery calcium score CAD coronary artery disease CFR coronary flow reserve ECG electrocardiogram 18 FDG F-fluorodeoxyglucose HF heart failure HMR heart-to-mediastinum ratio LV EF left ventricular ejection fraction MBF myocardial blood flow MI myocardial infarction 123 MIBG I-metaiodobenzylguanidine MPI myocardial perfusion imaging PET positron emission tomography SPECT single photon emission computed tomography
Introduction Every diagnostic test applied in clinical medicine serves a similar purpose: It is used by clinicians to identify suspected disease, to assess severity and prognosis of the disease (usually based on the magnitude of a deranged value), and finally to take action and implement therapies that are able to reverse patient’s outcome. Thus, in order to find acceptance in clinical practice, each clinical test has to undergo a series of steps of clinical validation relating to the following: (1) initial validation of diagnostic accuracy, (2) assessment of prognostic value, (3) impact on patient management and improvement of outcome, and finally (4) cost-effectiveness. Since its introduction in the early 1980s, cardiac radionuclide imaging has played an increasingly important role in the diagnosis and management of patients with cardiac disease. The nuclear cardiology community has done an excellent job
482, Page 2 of 11
in accumulating diagnostic and prognostic data to endorse its clinical value. Based on this evidence, radionuclide imaging has been embraced around the globe as a useful tool to diagnose and stratify patients with cardiac disease and is applied in clinical practice to several subgroups such as patients with cardiovascular risk factors, with suspected or angiographically proven coronary artery disease (CAD), after revascularization or in heart failure (HF) patients. National and international data review bodies have adopted cardiac radionuclide imaging in their guidelines and recommendations for the treatment of patients with the aforementioned conditions [1–5]. Recently, a number of publications and follow-up reports of large cohort studies with radionuclide imaging have been published, adding relevant information. The purpose of the present review is to summarize the data from recent cohort studies, to critically appraise their clinical implications and to assess whether they add new aspects to current clinical practice. Another purpose of this manuscript is to put the available evidence into a clinical context, analyze its potential impact on patient management and identify gaps in knowledge and unanswered questions to be addressed in future randomized trials. For this purpose, the present review focuses on evidence available from radionuclide imaging in four major areas of cardiovascular disease: & & & &
asymptomatic patients (at risk for silent ischemia) symptomatic CAD microvascular dysfunction heart failure (HF).
Screening of Asymptomatic Subjects Approximately 50 % of myocardial infarctions (MI) occur in patients who were asymptomatic until the day of their event, of which 48 % have only one or even none of the traditional risk factors [6]. Therefore, effective screening tools are sought to reduce the high toll of unexpected coronary events by deriving more individual risk estimates than traditional risk factors. Diabetic patients belong to a high-risk category in which screening for asymptomatic CAD may be beneficial [7]. The prevalence of abnormal myocardial perfusion imaging (MPI) studies among asymptomatic diabetic patients varies from 21 to 58 %, with 15 to 20 % having high-risk MPI findings [8–10]. In the Detection of Ischemia in Asymptomatic Diabetics (DIAD) study, 1123 asymptomatic subjects with type 2 diabetes were randomly assigned to undergo screening adenosine-stress radionuclide MPI or standard care (including intensive medical therapy for risk factor modification) and were followed-up for 4.8 years [11]. Overall, cardiac event rates (cardiac death or nonfatal MI) for the entire study
Curr Cardiol Rep (2014) 16:482
population were much lower than anticipated (0.6 % per year). In the screened group, MPI was able to accurately risk stratify patients based on perfusion findings with higher event rates in patients with moderate or large perfusion defects compared to those with only mildly abnormal or normal MPI (2.4 % per year vs 0.4 % per year (hazard ratio (HR), 6.3; 95 % confidence interval (CI), 1.9-20.1; P=.001)). However, screening with radionuclide MPI did not improve the prognosis of the patients with comparable event rates in both groups (2.7 % in the screened group vs. 3.0 % in the not-screened group (P=.73). Due to the low overall risk of the study cohort, the results of the DIAD study cannot be generalized to all asymptomatic diabetics, and therefore the issue of screening asymptomatic diabetics remains unresolved [3]. Others have proposed to use coronary artery calcium scores (CACS) as a gatekeeper to identify asymptomatic subjects at risk for whom further screening radionuclide MPI may be justified [12, 13]. Indeed, abnormal MPI is rare below a CACS threshold of 100, but the prevalence of abnormal scans increases progressively with higher CACS. In 411 largely asymptomatic subjects, He and colleagues reported a rate of only 2.6 % abnormal MPI scans below a CACS of 100, while the prevalence of abnormal MPI rose to 11.3 % and 46 % for CACS of 101-399, and >400, respectively [14]. Berman and colleagues analyzed 1195 subjects of which 51 % were asymptomatic [15]. The frequency of ischemic MPI was 0.80) which were entered into a registry. The prospective nuclear substudy of the COURAGE trial, albeit slightly underpowered to detect significant differences between revascularization versus medical treatment, revealed improved outcomes in patients with ≥5 % reduction in ischemic burden compared to those with
