If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Evaluation of predictors of silent coronary artery disease (SCAD) in high-risk asymptomatic diabetic patients and to evaluate their two-year outcome.
Methods and results
Four hundred diabetic patients without prior CAD but at high CAD risk underwent myocardial perfusion scintigraphy (MPS) in this prospective multicentre outcome trial. MPS were abnormal in 22% of patients. Male sex (OR 2.223, 1.152–4.290; p = 0.017), diabetes duration (OR 1.049,1.015–1.085; p = 0·005), peripheral artery disease (OR 2.134, 1·150–3.961; p = 0.016), smoking (OR 2.064, 1.109–3.839; p = 0·022), systolic blood pressure (OR 1.014, 1.00–1.03, p = 0·056), brain natriuretic peptide (OR 1.002, 1.001–1.004, p = 0·005) independently predicted an abnormal MPS: if <2 and >3 predictors were present, 3.2% and 47% patients had an abnormal MPS, respectively (p < 0·001). Two-year major adverse cardiac event rates increased from 2·9% to 14·6%, cardiac death rates from 0·6% to 4·1% in patients with summed stress scores ≤10 and >10%, respectively (each p < 0.045).
Conclusions
Male sex, diabetes duration, peripheral artery disease, smoking, elevated systolic blood pressure and increased brain-natriuretic peptides independently predicted SCAD. In presence of >3 predictors, almost 50% of patients had an abnormal MPS. They may benefit from screening by MPS since the extent of the MPS abnormality discriminated clearly between a favourable compared to a bad 2-year outcome. However, even highest risk patients without objective evidence of CAD had a benign prognosis without need for specific evaluation or therapy.
Diabetes mellitus is an important risk factor for coronary artery disease (CAD) and it has been shown that silent CAD affects 20–35% of patients with diabetes [
Progression to overt or silent CAD in asymptomatic patients with diabetes mellitus at high coronary risk: main findings of the prospective multicenter BARDOT trial with a pilot randomized treatment substudy.
], mainly based on the Detection of Silent myocardial Ischemia in Asymptomatic Diabetic Subjects trial (DIAD), which implied that general CAD screening does not improve outcome as long as cardiovascular risk factors are treated [
]. In contrast, European guidelines challenged this position arguing that there are high risk patients who may benefit from specific treatment if they can be identified by appropriate screening [
ESC guidelines on diabetes, pre-diabetes, and cardiovascular diseases developed in collaboration with the EASD: the task force on diabetes, pre-diabetes, and cardiovascular diseases of the European Society of Cardiology (ESC) and developed in collaboration with the European Association for the Study of Diabetes (EASD).
]. However, high risk clinical characteristics of such patients need to be better defined.
The present study aimed to 1) evaluate clinical characteristics that may help to identify patients with a high probability of silent CAD documented by an abnormal stress MPS in asymptomatic diabetic patients without any manifestation of CAD, but at high coronary risk according to the American Diabetes Association [
] and 2) to assess the prognostic value of this finding in such patients on 2-year outcome.
2. Methods
2.1 Study design
The present study is an a priori planned analysis as central part of the Basel Asymptomatic high Risk Diabetics' Outcome Trial (BARDOT) of which the protocol and amendment have been published previously [
Progression to overt or silent CAD in asymptomatic patients with diabetes mellitus at high coronary risk: main findings of the prospective multicenter BARDOT trial with a pilot randomized treatment substudy.
]. BARDOT was a prospective multicentre outcome trial in three centres in Germany (district hospital of Schopfheim, Germany; district hospital of Lörrach, Germany) and Switzerland (University Hospital of Basel, Switzerland). Study patients underwent clinical visits and rest/stress myocardial perfusion single photon emission computed tomography (MPS) at baseline and after two years of follow-up as described previously [
Progression to overt or silent CAD in asymptomatic patients with diabetes mellitus at high coronary risk: main findings of the prospective multicenter BARDOT trial with a pilot randomized treatment substudy.
]. Independent clinical predictors of abnormal baseline MPS were evaluated. BARDOT was approved by the ethics committees of all participating centres and all patients gave written informed consent.
2.2 Participants
For inclusion in the BARDOT trial, study patients had to have type-2 diabetes and neither history nor symptoms of CAD i.e. they were free from CAD manifestations (no typical or atypical angina, no history of myocardial infarction or revascularization), i.e. they were “asymptomatic”. They had to be at high risk of CAD documented by end-organ damage (peripheral or carotid occlusive disease, retinopathy, microalbuminuria, autonomic cardiac neuropathy as measured by Ewing et al. [
]), or by the composite of age >55 years, diabetes duration >5 years, and two cardiac risk factors (smoking, hypertension, hypercholesterolemia, or positive family history of CAD) in addition to diabetes.
Exclusion criteria of the study were: age above 75 years, severe illness with a life expectancy of <3 years, known prior CAD, or shortness of breath (NYHA IV).
Between June 2002 and December 2010, 400 out of 2018 patients screened were eligible for this study and consented. The enrolment period was longer than expected because it was more difficult than anticipated to recruit asymptomatic diabetic patients willing to participate in this long-term study with repeated testing.
2.3 Procedures
Clinical examinations, standard laboratory testing and follow-up visits after one and two years were performed in the outpatient clinics of each centre. Rest/stress gated MPS studies were performed for all patients at the core laboratories of the University Hospital Basel following a standard protocol as described before [
Progression to overt or silent CAD in asymptomatic patients with diabetes mellitus at high coronary risk: main findings of the prospective multicenter BARDOT trial with a pilot randomized treatment substudy.
ACC/AHA/ASNC guidelines for the clinical use of cardiac radionuclide imaging—executive summary: a report of the American College of Cardiology/American Heart Association task force on practice guidelines (ACC/AHA/ASNC Committee to revise the 1995 guidelines for the clinical use of cardiac radionuclide imaging).
]. In short, a rest-stress (99mTechnetium sestamibi, 400 MBq/800 MBq) protocol with symptom-limited exercise or adenosine stress and electrocardiographic monitoring was used. Images were scored using a 17-segment model with a five-point scale from 0 = normal to 4 = no uptake [
]. Summed scores (stress, rest and difference scores) were calculated summarizing the perfusion scores of the 17 segments and also converted into % myocardium abnormal. Summed stress scores (SSS) represent the overall perfusion abnormality of the scan, whereas summed difference scores (SDS) represent the severity and extent of ischemia and summed rest scores (SRS) perfusion abnormality at rest. The following categories for SSS, SDS, and SRS were then derived to provide information regarding baseline and follow-up perfusion defects: 0% myocardium, 0.1–4.9%, 5–9.9%, and 10% or higher. All patients with abnormal MPS results were treated with aspirin, a statin and a betablocker while therapy with ACE- and ARB- inhibitors was advised [
Progression to overt or silent CAD in asymptomatic patients with diabetes mellitus at high coronary risk: main findings of the prospective multicenter BARDOT trial with a pilot randomized treatment substudy.
A perfusion scan was considered abnormal with an SSS ≥4, consistent with ≥5% of the myocardium affected. Ischaemia was defined as reversible defect with an SDS of ≥2 (≥3% myocardium ischemic) and scar with at least one non-reversible segment [
Risk stratification in patients with remote prior myocardial infarction using rest-stress myocardial perfusion SPECT: prognostic value and impact on referral to early catheterization.
] and revascularization as symptom-driven revascularization (i.e. revascularizations necessary in patients who became symptomatic and remained so despite medical therapy). Major adverse cardiac events (MACE) were cardiac death, MI or symptom-driven revascularizations.
An independent Critical Events Committee adjudicated all clinical events blinded to baseline MPS results and study group assignment.
2.5 Statistical analysis
Baseline characteristics of patients with versus without evidence of silent CAD and follow-up characteristics with normal MPS at baseline were compared by independent t- or Mann-Whitney U tests (as appropriate) for continuous variables and chi-square tests for binary variables.
Independent predictors of an abnormal MPS were analysed by a binary logistic regression analysis (with binomial error) to estimate odds ratios (OR) with 95% confidence intervals.
With respect to the number of independent variables predicting an abnormal MPS the following continuous variables were divided into 2 categories. If the value was above the defined threshold the predictor was considered to be present: diabetes duration (≥10 years), B-type natriuretic peptide (BNP > 50.3 pg/l [cut-off of the used assay]) and systolic blood pressure (≥140 mm Hg).
Statistical analyses were performed using the IBM SPSS statistics (Version 22).
2.6 Role of the funding source
None of the sponsors mentioned above had any influence on the design and conduct of the study, interpretation of the data, nor the decision to submit the manuscript to publication.
3. Results
3.1 Prevalence of silent CAD at baseline
Study patients had an age range from 35 to 75 years, two thirds were men with an average diabetes history of 11 ± 8 years. The mean scintigraphically determined left ventricular ejection fraction was 58 ± 11%. Overall, 50% of all patients were on insulin and 80% on oral glucose lowering agents. The majority of patients was on cardioactive drugs, particularly antihypertensives, 53% were on antiplatelet therapy and 57% on lipid lowering medications. Baseline MPS was normal in 313 and abnormal in 87 patients (22%), i.e. 22% of patients were considered to have silent CAD.
Baseline characteristics of patients with versus without evidence of CAD are summarized in Table 1. Compared to patients with a normal scan, those with abnormal MPS were older, more frequently male, smoking, suffering from peripheral vascular disease and autonomic cardiac neuropathy and had longer histories of diabetes with higher systolic blood pressures and higher creatinine and BNP values.
Table 1Baseline characteristics of patients with normal versus abnormal MPS Variable.
All clinically relevant variables showing significant differences at baseline between patients with normal versus abnormal MPS findings (Table 1) were tested for their independent predictive value. Independent predictors of an abnormal MPS are summarized in Table 2. It may be of particular note, that age, autonomic cardiac neuropathy, and creatinine were also variables in the logistic regression model but not independent predictors of an abnormal MPS.
Table 2Predictors of an abnormal MPS at baseline.
Univariate predictors
Multivariate predictors
OR
95% - CI
p
OR
95% - CI
p
Age
1.043
1.008–1.079
0.016
Male sex
2.865
1.546–5.310
<0.001
2.223
1.152–4.290
0.017
Diabetes duration (years)
1.052
1.020–1.084
0.001
1.049
1.015–1.085
0.005
Autonomic neuropathy
2.002
1.225–3.272
0.006
Peripheral arterial disease
3.867
1.827–8.187
<0.001
2.134
1.150–3.961
0.016
Smoking
2.226
1.303–3.804
0.003
2.064
1.109–3.839
0.022
Systolic BP (mm Hg)
1.013
1.000–1.026
0.052
1.014
1.000–1.029
0.056
Creatinine
1.015
1.006–1.025
0.001
BNP
1.002
1.001–1.004
0.002
1.002
1.001–1.004
0.005
Abbreviations: OR denotes odds ratio. CI confidence interval, BP blood pressure, BNP brain natriuretic peptide.
The rate of abnormal MPS findings increased with increasing numbers of independent predictors present in a patient (Fig. 1): compared to patients with no or only one of these predictors, the rate of an abnormal MPS increased 6.5-fold in patients with two or three predictors (from 3.2% to 20.7%; p < 0.001) and 14·7-fold in those with >3 predictors (to 47%; p < 0.001).
Fig. 1Relationship between the presence of independent predictors of abnormal MPS and the number of abnormal MPS observed (p < 0·001). Complete data for this analysis in 396 of the 400 patients.
The mean follow-up duration was 743 ± 77 days, complete in 388 patients (97%). During follow-up, 4 patients suffered cardiac death, 6 patients suffered cardiac death or had a non-fatal MI, and 17 experienced a MACE. The 2-year outcome was directly related to the overall abnormality of the MPS results at baseline as shown in Fig. 2. In contrast to patients with <10% myocardium abnormal on MPS, the cardiac death rate was 6·8-fold higher in patients with ≥10% abnormal myocardium (0.6% versus 4.1%, p = 0.045), the cardiac death or MI rate 6.8 fold higher (0.9% versus 6.1%, p = 0.007) and the MACE rate 5-fold higher (2.9% versus 14.6%, p < 0.001). Of note, patients with an abnormal MPS at baseline had good CAD treatment at 2-year follow-up. The 2-year follow-up medication was as follows:
Fig. 2Prognostic impact of the extent of the overall MPS abnormality: Kaplan Meier event rates as a function of % myocardium abnormal (SSS) in patients with less and >10% myocardium abnormal: a) for major adverse cardiac events; b) for cardiac death or myocardial infarction; c) for cardiac death.
The present prospective, a priori planned analysis, a central part of the BARDOT trial in asymptomatic high risk patients with diabetes, identified important independent clinical predictors of silent CAD documented by an abnormal baseline MPS: male sex, smoking, peripheral artery disease, long diabetes duration, elevated systolic blood pressure, and increased BNP value. The rate of an abnormal MPS result increased with the number of predictors present in each patient up to 15-fold. Thus, the presence of these independent predictors in a particular patient may guide further risk stratification. In addition, the 2-year outcome of these patients was significantly related to the extent of the MPS abnormality: cardiac death rate was seven-fold higher in patients with ≥10% myocardium affected versus those with <10% myocardium abnormal.
Since the number of abnormal MPS increased with the number of predictors present, these predictors may identify patients at highest risk for an abnormal MPS and thus relevant silent CAD. Based on our findings, half of diabetic patients with at least four of these predictors had an abnormal MPS and might benefit from further risk stratification. In contrast, patients with less than two predictors had a very low rate of abnormal MPS results (3.2%) and therefore may not need to undergo further risk stratification.
The strength of this prospective multicentre study is its design in asymptomatic patients highly selected for their high baseline clinical cardiac risk despite the absence of prior CAD manifestations. All these patients had a stress MPS study at baseline and were followed for clinical events up to two years. All clinical events were adjudicated by an independent Clinical Events Committee blinded to baseline study findings. Predictors identified are readily available by clinical examination and routine laboratory testing and therefore at the disposition of all physicians caring for these patients. Thus, findings have the potential to influence their management in everyday practice.
4.1 Predictors of an abnormal MPS in asymptomatic diabetic patients
The prevalence of silent CAD in diabetic patients varied widely in published reports, from <10% to >50% [
Silent myocardial ischemia detected by single photon emission computed tomography (SPECT) and risk of cardiac events among asymptomatic patients with type 2 diabetes: a meta-analysis of prospective studies.
], indicating the need to identify patient characteristics that provide hints which patients to select for screening in view of possible preventive treatments.
In the Detection of Silent myocardial Ischemia in Asymptomatic Diabetic Subjects trial (DIAD), the strongest predictor of on abnormal MPS was an abnormal cardiac autonomic function (heart rate response to Valsalva). Although autonomic cardiac function was more frequently diagnosed in the present study in patients with abnormal than with a normal MPS, too, it was not an independent predictor of an abnormal MPS here (p = 0.109). In DIAD, neither demographic variables, traditional risk factors, diabetic complications nor biomarkers of inflammation and thrombosis were independent predictors of an abnormal MPS [
Then, BARDOT patients were selected for high coronary risk: compared with DIAD patients, BARDOT patients were on average 2 years older, had 2.4 years longer diabetes duration, higher hemoglobin A1c values, end-organ damage more frequently (altogether in nearly 90% of patients), and higher rates of standard CAD risk factors, and were more often on insulin therapy (50% vs. 10% in DIAD). These data document the higher baseline risk of BARDOT patients. In addition, only 6% of DIAD patients had perfusion defects of at least 5% of the myocardium as required in BARDOT patients [
The predictive role of an increased BNP-value at baseline is in line with results of a primary prevention study in which BNP was used to identify end organ damage of the heart: in asymptomatic primary prevention patients, BNP screening was able to identify silent heart disease [
], neither an increased creatinine level nor the presence of microalbuminuria were independently predictive of an abnormal MPS in the present analysis. This may be explained partly by the frequent use of ACE-/ARB-inhibitors (77%), the highly selected patient group tested and the multiple predictors analysed in the current versus older studies [
Given the common risk factors for peripheral arterial disease (PAD) and other cardiovascular and cerebrovascular diseases, it is not surprising that people with PAD are more likely to have these other disorders concomitantly and vice versa [
]. Among 5084 Medicare beneficiaries, the prevalence of history of myocardial infarction was 2.5× as high in subjects with PAD (based on ABI < 0.9) versus those without; for angina, congestive heart failure, stroke, and transient ischemic attack, the prevalence rates were 1.9, 3.3, 3.1, and 2.3× as high, respectively [
Incremental prognostic value of myocardial perfusion single photon emission computed tomography for the prediction of cardiac death: differential stratification for risk of cardiac death and myocardial infarction.
Impact of diabetes on the risk stratification using stress single-photon emission computed tomography myocardial perfusion imaging in patients with symptoms suggestive of coronary artery disease.
Cardiac outcomes after screening for asymptomatic coronary artery disease in patients with type 2 diabetes: the DIAD study: a randomized controlled trial.
] and holds true for the BARDOT trial for a high risk diabetic patient population with respect to cardiac death and MACE. However, it is of note that even in these asymptomatic diabetic patients selected for their high coronary risk, a completely normal MPS (SSS = 0% of myocardium) had an annual mortality of 0·35%, only, similar to a comparative normal population of the same age. In contrast, patients with an SSS ≥ 10% of myocardium had an almost six-fold higher annual mortality rate.
4.3 Limitations
This study aimed at the identification of predictors of the presence of CAD and its prognostic implications, but not on the further management of these patients. A large prospective randomized trial would be necessary to show whether such a treatment would be beneficial in high risk asymptomatic patients as identified here, since there was no significant difference in the rates of death and major cardiovascular events between patients undergoing prompt revascularization and those undergoing medical therapy in The Bypass Angioplasty Revascularization Investigation 2 Diabetes (BARI 2D) trial [
CAD was assumed to be present if the MPS was clearly abnormal as defined in this study, but no angiographic proof could be obtained in these asymptomatic patients.
Regarding image interpretation, readers were aware that they were looking at study patients but they were blinded to patient history, ECG results and any symptoms before or during the stress test.
5. Conclusions and implications
Male sex, long diabetes duration, presence of peripheral artery disease, smoking, elevated systolic blood pressure and an increased BNP value were independently predictive of an abnormal MPS in asymptomatic high risk diabetic patients. If >3 of these predictors were present, almost half of the patients had an abnormal MPS. These patients may benefit from screening and risk stratification, since the extent of the MPS abnormality discriminated clearly between a favourable compared to a bad 2-year outcome. Thus, the findings that only 12% of these high risk patients had relevant ischemia (>10% myocardium affected) and only 14% of them experienced a MACE over two years have three important implications: 1) the rate of MACE is too low to warrant general screening, even in high risk patients; 2) clinical predictors as identified here may be used to select the highest risk asymptomatic diabetes patients for further study and 3) among them, patients without objective evidence of CAD have a favourable prognosis without need for a specific cardiac treatment whereas those with non-invasively documented evidence of CAD have a much worse outcome and may benefit from further evaluation and appropriate therapy.
6. Perspectives
6.1 Competency in medical knowledge
Despite the fact that one third of patients with diabetes mellitus have silent CAD, the American Diabetes Association does not recommend routine screening since it does not improve outcome based on one large study. European guidelines challenged this position arguing that there are high risk patients who may benefit from specific treatment if they can be identified by appropriate screening. However high risk clinical characteristics of such patients need to be better defined.
6.2 Translational outlook
The present study identified six readily available clinical parameters independently predictive of an abnormal stress myocardial perfusion scintigram in asymptomatic diabetic patients: male sex, long diabetes duration, presence of peripheral artery disease, smoking, elevated systolic blood pressure and an increased BNP value. If >3 of these predictors were present, 47% of patients had an abnormal MPS versus 3.2% (p < 0.001) if <2 were present. Patients with an abnormal MPS may benefit from further risk stratification and treatment, since the extent of the MPS abnormality discriminated clearly between a favourable compared to a bad 2-year outcome (death, myocardial infarction or major adverse cardiac events separately and combined).
Declaration of interest
M. Zellweger and M. Pfisterer had full access to all the data of the study and take responsibility for the integrity of the data and the accuracy of the data analysis. All authors contributed substantially to conception and design, acquisition or interpretation of the data, revised the manuscript critically and approved the final version. We confirm that no author has a conflict of interest regarding this manuscript. None of the sponsors mentioned above had any influence on the design and conduct of the study, interpretation of the data, nor the decision to submit the manuscript to publication.
Funding
Swiss National Foundation for Research (Nr 3200B0-100620); Swiss Heart Foundation; Basel Foundation for Cardiovascular Research; Diabetes Association, Basel; Roche; Pfizer; Takeda; Heider&Co; Switzerland.
Acknowledgements
The following research assistants collected patient data during the study: Michael Ammon, MD; Claudia Bösch, MD; Miriam Brinkert, MD; Ronny Büchel, MD; Niklas Ehl, MD; Bernhard Friedli, MD; Peter Gnehm, MD; Luca Jörg, MD; Naina Rastalsky, MD; Florian Riede, MD, Myriam Ritter, MD, Philip Haaf, MD, and Ian Russi, MD all from Basel, Switzerland; and Daniel Kammerer, MD, from Schopfheim, Germany.
References
Wackers F.J.
Young L.H.
Inzucchi S.E.
Chyun D.A.
Davey J.A.
Barrett E.J.
et al.
Detection of silent myocardial ischemia in asymptomatic diabetic subjects: the DIAD study.
Progression to overt or silent CAD in asymptomatic patients with diabetes mellitus at high coronary risk: main findings of the prospective multicenter BARDOT trial with a pilot randomized treatment substudy.
ESC guidelines on diabetes, pre-diabetes, and cardiovascular diseases developed in collaboration with the EASD: the task force on diabetes, pre-diabetes, and cardiovascular diseases of the European Society of Cardiology (ESC) and developed in collaboration with the European Association for the Study of Diabetes (EASD).
ACC/AHA/ASNC guidelines for the clinical use of cardiac radionuclide imaging—executive summary: a report of the American College of Cardiology/American Heart Association task force on practice guidelines (ACC/AHA/ASNC Committee to revise the 1995 guidelines for the clinical use of cardiac radionuclide imaging).
Risk stratification in patients with remote prior myocardial infarction using rest-stress myocardial perfusion SPECT: prognostic value and impact on referral to early catheterization.
Silent myocardial ischemia detected by single photon emission computed tomography (SPECT) and risk of cardiac events among asymptomatic patients with type 2 diabetes: a meta-analysis of prospective studies.
Incremental prognostic value of myocardial perfusion single photon emission computed tomography for the prediction of cardiac death: differential stratification for risk of cardiac death and myocardial infarction.
Impact of diabetes on the risk stratification using stress single-photon emission computed tomography myocardial perfusion imaging in patients with symptoms suggestive of coronary artery disease.
Cardiac outcomes after screening for asymptomatic coronary artery disease in patients with type 2 diabetes: the DIAD study: a randomized controlled trial.
There is abundant evidence indicating that the presence of silent myocardial ischemia predicts the destiny of the patients in terms of morbidity and mortality [1]. Silent myocardial ischaemia is a frequent feature in patients with diabetes mellitus (DM) and its prevalence is estimated between 20 and 35%. Ischemia in diabetic patients may be induced by atherosclerotic focal coronary artery disease (CAD), but the role of microvascular disturbances as well as the deleterious effect of hyperglycaemia on endothelial function may not be neglected [2].
30986-5&id=gr1.jpg){kind=link}
30986-5&id=gr2.jpg){kind=link}