Prevalence and Prognostic Importance of Abnormal Positron Emission Tomography Among Asymptomatic Patients With Diabetes Mellitus

BACKGROUND

Ischemia and reduced global myocardial blood flow reserve (MBFR) are associated with high cardiovascular risk among symptomatic patients with diabetes mellitus (DM).

OBJECTIVES

This study aimed to assess the prevalence and prognostic importance of silent ischemia and reduced MBFR among asymptomatic patients with DM.

METHODS

This study included 2,730 consecutive patients with DM, without known coronary artery disease (CAD) or cardiomyopathy, who underwent rubidium-82 rest/stress positron emission tomography (PET) myocardial perfusion imaging (MPI) from 2010 to 2016. These patients were followed up for all-cause mortality (n = 461) for a median follow-up of 3 years. Patients were considered asymptomatic if neither chest pain nor dyspnea was elicited. Rates of ischemia, reduced MBFR, and coronary microvascular dysfunction on PET were assessed in both groups. Cox regression was used to define the independent association of abnormal MPI markers with mortality.

RESULTS

One-quarter of patients with DM (23.7%; n = 647) were asymptomatic; ischemia was present in 30.5% (n = 197), reduced MBFR in 62.3% (n = 361), and coronary microvascular dysfunction in 32.7% (n = 200). In adjusted analyses, reduced MBFR (HR per 0.1 unit decrease in MBFR: 1.08 [95% CI: 1.03-1.12]; P = 0.001) and reduced ejection fraction (HR per 5% decrease: 1.10 [95% CI: 1.01-1.18]; P = 0.02) were independently prognostic of mortality among asymptomatic patients, but ischemia was not. This was comparable to DM patients with symptoms. Insulin use and older age were significant predictors of reduced MBFR among asymptomatic patients with DM.

CONCLUSIONS

In both symptomatic and asymptomatic patients with DM, impairment in MBFR is common and associated with greater mortality risk.

Prognostic Relationship Between Coronary Artery Calcium Score, Perfusion Defects, and Myocardial Blood Flow Reserve in Patients With Suspected Coronary Artery Disease

BACKGROUND

Coronary artery calcium score (CACS) is an anatomic measure of calcified atherosclerosis. Myocardial perfusion defects and reduced myocardial blood flow reserve (MBFR) are physiological measures of ischemia and coronary circulatory health. We aimed to assess the relative prognostic importance of MBFR, perfusion defects, and CACS in patients with suspected coronary artery disease.

METHODS

A total of 5983 consecutive patients without known history of coronary artery disease or cardiomyopathy, who underwent a CACS and ⁸²Rb positron emission tomography myocardial perfusion imaging between 2010 and 2016, were followed for all-cause death (n=785) over median of 3 years. Prognostic value was assessed using multivariable Cox regression models, and incremental risk discrimination for imaging variables was evaluated by comparing model c-indices after adjusting for clinical risk factors (RF).

RESULTS

Mean age was 67.1 years, 60% were female, and 83% were symptomatic. CACS was 0 in 22%, abnormal perfusion in 19%, and MBFR <2 in 53.3%. When added to RF, the model with MBFR had the best fit (c=0.78, P<0.0001). Addition of CACS to model with RF and perfusion (c=0.77) offered modest improvement in discrimination over the model with RF and perfusion (c=0.76, P=0.02). Adding CACS to a model with RF, perfusion, and MBFR did not provide incremental prognostic value (c=0.785 for both, P=0.16). CACS and MBFR both had independent prognostic value in patients with normal and abnormal myocardial perfusion imaging. Even among patients with CACS of 0, MBFR <2 was present in 37.8%, being associated with higher risk of death (hazard ratio per 0.1↓, 1.10 [1.04-1.15]; P<0.001), but perfusion defects were not.

CONCLUSIONS

Use of anatomic testing such as CACS of 0 to avoid myocardial perfusion imaging in symptomatic patients could lead to missing microvascular dysfunction in 4 out of 10 patients, a finding associated with a high mortality risk. Higher CACS was independently associated with the risk of death but did not provide incremental prognostic value over positron emission tomography with MBFR.

Relationship Between Myocardial Perfusion Imaging Abnormalities on Positron Emission Tomography and Anginal Symptoms, Functional Status, and Quality of Life

BACKGROUND

Myocardial perfusion imaging (MPI) identifies abnormalities that occur early in the ischemic cascade leading to angina. Our aim was to study the association between ischemic measures on positron emission tomography MPI and patients' health status; their symptoms, function, and quality of life.

METHODS

Health status was collected using the Seattle Angina Questionnaire (SAQ-7, 0-100, higher=better) and Rose Dyspnea Score (RDS) on 1515 outpatients with known or suspected coronary artery disease presenting for clinically indicated pharmacological ⁸²Rb positron emission tomography MPI from July 2018 to July 2019. Adjusted multivariable ordinal regression models were used to assess the association between MPI findings of ischemia and the SAQ physical limitation, angina frequency, quality of life, summary score, and the RDS.

RESULTS

The mean SAQ and RDS scores of the cohort (mean age 71.7 years, 55% male, 37.6% prior myocardial infarction or revascularization) were 73.8±28.6 (physical limitation), 87.4±21.7 (angina frequency), 79.0±26.1 (quality of life), 81.3±19.0 (summary score), and 2±2 (RDS). No perfusion, flow or function abnormalities were significantly associated with SAQ angina frequency scores. Low left ventricular ejection fraction reserve (≤0%), low global and regional myocardial blood flow reserve (<2) were independently associated with worse SAQ Physical Limitation score, SAQ summary score, and RDS (30% to 57% greater odds; all P≤0.01), but reversible perfusion defects were not.

CONCLUSIONS

Impaired augmentation of left ventricular ejection fraction and myocardial blood flow with stress is associated with significant angina-associated functional limitation, health status, and dyspnea in patients who underwent positron emission tomography MPI, but not the frequency of their angina. Future studies should evaluate whether therapies that improve stress-induced abnormalities in systolic function and myocardial flow may improve patients' health status.

Relationship between coronary artery calcium score and myocardial blood flow reserve in patients with suspected coronary artery disease

Funding Acknowledgements

Type of funding sources: None.

Background

Both the Coronary Artery Calcium Score (CACS), a non-invasive surrogate for atherosclerotic burden, and reduced myocardial blood flow reserve (MBFR) with normal perfusion, a non-invasive surrogate for coronary vasomotor dysfunction, independently predict future cardiovascular events. The relationship between CACS and MBFR, and potential clinical factors affecting it, is not well understood.

Methods

Among 9467 consecutive patients without known history of CAD who had normal perfusion on 82Rb PET-CT and a concomitantly measured CACS between 01/2010 - 06/2020 within our health system, we assessed the relationship between CACS and MBFR. Multiple linear regression was used to predict MBFR using CACS, adjusted for age, sex, BMI, risk factors, symptoms, resting LVEF and vital signs. Interactions of age, sex, diabetes, and symptoms with CACS were assessed to evaluate if they modified the relationship of CACS with MBFR.

Results

Mean age (SD) of the study cohort was 66.4 (12.6) years, 64% were women, 64% had chest pain and 47% had dyspnea. Reduced MBFR (&lt;2) was present in 44% and CAC &gt;0 in 74% of patients.  There was a modest inverse correlation between MBFR and CACS, r= - 0.18, p = &lt; 0.0001 (Figure). In adjusted analyses, CACS (β for CAC per 100 = -0.013 [95% CI: -0.015, -0.010]) was weakly associated with MBFR, and age, sex, diabetes, or symptoms did not modify this relationship (all interaction p-values &gt;0.1). Older age, female sex, presence of hypertension, diabetes, dyspnea, lower LVEF, higher baseline HR and higher CACS independently predicted reduced MBFR, but explained only 20% of the variance in MBFR (R2 =0.20).

Conclusion

There is a weak relationship between CACS and MBFR, which is not modified by age, sex, symptoms, or other CV risk factors. Coronary calcium burden does not completely reflect the overall disease activity within the coronary circulation, and measures of coronary vasomotor function such as MBFR may offer complementary information on CAD risk to that provided by the total burden of calcified atherosclerosis.