Prognostic implications of cardiac 82-rubidium positron emission tomography in angina patients with no perfusion defects

Background

Myocardial perfusion imaging with 82-Rubidium positron emission tomography (82Rb-PET) is increasingly used in the assessment of stable coronary artery disease (CAD). Among other variables, it provides quantitative measures of myocardial blood flow (MBF) which has shown the prognostic significance of coronary microvascular dysfunction (CMD), also in patients without perfusion defects. However, other 82Rb-PET variables may also be of prognostic significance in these patients.

Purpose

The purpose of this study was to evaluate the prognostic value of 82Rb-PET in patients with symptoms suggestive of CAD but no perfusion defects.

Methods

A study was conducted with 3726 consecutive patients who underwent 82Rb-PET on the suspicion of stable CAD between January 2018 and August 2020. After exclusion of patients with regional perfusion defects, we examined the association of 82Rb-PET derived parameters with a composite endpoint: all-cause mortality, hospitalization for unstable angina pectoris, acute myocardial infarction, heart failure or ischemic stroke in 2175 patients. CMD was defined as myocardial blood flow reserve (MBFR) <2. Analyses were further stratified to assess differences across gender.

Results

Resting and stress MBF were higher in women, while MBFR was lower and CMD more prevalent (30.5% among women versus 25.3% among men, p=0.008). Over a median follow-up of 1.7 years (IQR 1.1–2.5 years) a total of 148 events were observed. In unadjusted analyses, MBF during stress, MBFR, left ventricular ejection fraction (LVEF), LVEF-reserve, heart rate reserve and Ca-score were associated with adverse outcomes in both genders (Figure 1). A joint multivariable Cox model, for both genders, adjusted for patient characteristics, cardiovascular risk factors and 82Rb-PET variables showed reduced MBFR <2 (HR 1.75, 95% CI 1.24–2.48), resting LVEF (HR 1.38 per 10% decrease, 95% CI 1.24–1.54) and LVEF-reserve (HR 1.19 per 5% decrease, 95% CI 1.07–1.31) to be significant predictors of outcomes (Figure 2). Results were consistent in subgroups defined by gender, previous history of ischemic heart disease (IHD), reduced LVEF and atrial fibrillation.

Conclusion

MBFR, LVEF and LVEF-reserve derived from 82Rb-PET are predictors of adverse outcome and provide prognostic information in patients with no perfusion defects. This may aid in identifying patients at risk and provide opportunity of prevention.

Funding Acknowledgement

Type of funding sources: Public hospital(s). Main funding source(s): Department of Cardiology, Bispebjerg Frederiksberg Hospital, Copenhagen

Coronary microvascular disease assessed by 82Rb-PET-CT is an independent prognostic marker of all-cause mortality

Background

Coronary microvascular disease (CMD) is a major contributor to e.g. heart failure and angina pectoris, as well as being associated with an increased risk of adverse events. CMD is diagnosed by reduced myocardial blood flow reserve (MBFR), preferably by Positron emission tomography myocardial perfusion CT (PET-CT).

Purpose

We aim to determine whether reduced MBFR is associated with an increased hazard of all-cause mortality independently of the extent of perfusion defects in patients suspected of obstructive coronary artery disease.

Method

We conducted a multicenter study of all patients referred for 82Rubidium PET-CT imaging between January 2018 and August 2020. Rest and stress examinations were performed using standard imaging protocols. Percentage of perfusion defects were calculated based on summed rest- and difference score. CMD was defined as MBFR ≤2. Patients were followed for all-cause mortality through national registries with no loss to follow-up.

Results

Among the 7156 patients studied, 61.8% were men, median age was 69 [61–76 IQR] years, 14.1% had LVEF ≤40%, 58.4% had a previous diagnosis of ischemic heart disease (IHD), 20.1% had atrial fibrillation and 38.9% had MBFR ≤2.

A total of 571 (7.8) deaths were observed, more frequently in MBFR ≤2 compared to MBFR >2 (4.2% vs 13.2%, p<0.001). MBFR was significantly associated with reversible hypoperfusion (r2=−0.33, p<0.0001). In Kaplan-Meier estimation MBFR ≤2 was significantly associated to all-cause mortality in the overall population as well as in clinically relevant subgroups defined by the extent of reversible and/or irreversible perfusion defects (p<0.05 for all, fig. 1). In multivariate Cox-analysis adjusting for age, sex, Charlson's Co-morbidity index, eGFR, LVEF and LVEF-reserve and stratifying by diabetes, MBFR ≤2 remained a robust predictor of all-cause mortality with a HR 1.73, 95% CI: 1.62–2.19, p<0.0001 (fig. 2). No interaction was found between MBFR and reversible hypoperfusion. In subgroup analysis including only patients with no reversible perfusion defects (n=3095), MBFR ≤2 was still strongly associated with a HR of 2.00, 95% CI: 1.29–3.11, p<0.001 for all-cause mortality.

Conclusion

MBFR ≤2 is a robust predictor of all-cause mortality independently of the extent of reversible- and/or irreversible perfusion defects. Information of MBFR should be incorporated in the clinical risk stratification of patients being investigated for ischemia.

Funding Acknowledgement

Type of funding sources: Public hospital(s). Main funding source(s): Research Committee, Bispebjerg & Frederiksberg University Hospital, scientific scholarship

Myocardial blood flow reserve assessed by 82Rb-PET-CT is associated with small-vessel disease in the kidney and brain

Background

Coronary microvascular dysfunction (CMD) may be linked to small-vessel disease in other vascular beds as a part of multisystem disorder. However, there are limited data in support of this.

Purpose

We aim to determine whether reduced myocardial blood flow reserve (MBFR) is associated with an increased hazard of small-vessel disease in the kidneys and brain.

Method

We conducted a multicenter study of all patients consecutively referred for 82Rubidium-Positron emission tomography (82Rb-PET) myocardial perfusion CT imaging between January 2018 and August 2020. CMD was defined as MBFR ≤2. Patients were followed through national registries using ICD-10 codes with no loss to follow-up for microvascular events (ME) defined as chronic kidney disease, stroke, affective disorders, and dementia. Despite the heterogeneity of outcomes, they all play a crucial role in ME, with vascular dementia, affective disorders, and both ischemic and hemorrhagic strokes being major contributors to cerebral ME.

Results

Among the 7156 patients studied, 61.8% were men, median age was 69 [61–76 IQR] years, 14.1% had LVEF ≤40%, 58.4% had a previous diagnosis of ischemic heart disease (IHD) and 20.1% had atrial fibrillation. 38.9% had MBFR ≤2. MBFR was significantly associated with eGFR at baseline (r2=0.22, p<0.0001). After multivariable adjustment for demographics, cardiovascular risk factors, LVEF and reversible perfusion defects, MBFR remained significantly associated with eGFR, also in patients with no perfusion defects (β=0.039, 95% Cl 0.03–0.05, p<0.001 in all patients and β=0.039, 95% Cl 0.02–0.05, p<0.001, in patients with ≤5% reversible- and ≤5% irreversible hypoperfusion). During follow-up, a total of 677 (9.5%) ME were observed (480 (6.7%) cerebral ME and 197 (2.7%) renal ME). ME was more frequent in patients with MBFR ≤2 compared to MBFR >2 (11.2% vs. 5.5%, p<0.001).

In crude analysis MBFR ≤2 was significantly associated with ME (p<0.0001, Fig. 1) as well as renal- and cerebral ME (both p<0.001). Similar results were found in subgroup analysis of patients with diabetes, normal kidney function (eGFR ≥60) or no perfusion defects, respectively (Fig. 1). After multivariate adjusting for demographics, IHD, cardiovascular risk factors, Charlson's Comorbidity index, atrial fibrillation and stratifying by chronic kidney disease stages, MBFR remained a significant predictor of ME (HR 1.43, 95% CI 1.15–1.78, p<0.001, fig. 2). In subgroup analysis including only patients with no reversible perfusion defects, MBFR ≤2 was associated with a HR of 2.04, 95% CI 1.43–2.91, p<0.0001 for ME.

Conclusion

This is the first larger cohort study relating CMD to microvascular outcome in the kidneys and brain. We conclude that CMD is an independent predictor of cerebral and renal ME. Data support the hypothesis that CMD is part of a systemic vascular disorder.

Funding Acknowledgement

Type of funding sources: Public hospital(s). Main funding source(s): Research Committee, Bispebjerg & Frederiksberg University Hospital, scientific scholarship