Abstract
Background
In stable chest pain patients, Rubidium-82 (82Rb) is commonly used for myocardial perfusion imaging (MPI) and allows for quantification of myocardial blood flow (MBF) and myocardial flow reserve (MFR). With increasing implementation, an alternative is [15O]H2O PET MPI, which has optimal tracer kinetics. However, the level of agreement on MBF and MFR between 82Rb and [15O]H2O PET is sparsely elucidated.
Purpose
To examine the agreement between 82Rb and [15O]H2O PET in the measurement of MBF and MFR.
Methods
This is a prospective cohort study conducted at a single hospital. By protocol, consecutive patients referred for clinically indicated routine 82Rb PET MPI underwent additional and subsequent [15O]H2O PET MPI during the same outpatient visit. Quantitative measurements of MBF and MFR were obtained using Syngo.PET for 82Rb PET with correction for non-linear extraction (1) and Carimas CE for [15O]H2O PET. Quantitative MBF and MFR measures were compared using Bland-Altman analysis.
Results
Forty-five patients were included from February to March 2023 with a mean time between 82Rb stress procedure to [15O]H2O rest procedure of 25 minutes (SD 9 minutes). In total, 44 paired rest, 40 paired stress, and 39 paired rest and stress procedures were available for statistical analyses. A lower rate-pressure product (RPP) (mean difference -335 mmHg x bpm, 95% CI -625 to -46) was observed during the 82Rb procedure vs. the [15O]H2O procedure, while other hemodynamic parameters were similar. Rest MBF was normalised to a RPP of 8000 mmHg x bpm.
The mean global rest MBF was 1.1 (SD 0.3) ml/g/min as measured by 82Rb PET versus 0.8 (SD 0.2) ml/g/min by [15O]H2O PET, yielding a mean difference of 0.3 ml/g/min (95% CI 0.1-0.3) and 95% limits of agreement (LoA) of -0.2 to 0.8 ml/g/min (Figure 1A+B). The mean global stress MBF was 2.1 (SD 0.7) ml/g/min as measured by 82Rb PET versus 2.5 (SD 0.9) ml/g/min by [15O]H2O PET, yielding a mean difference of -0.3 ml/g/min (95% CI -0.5 to -0.1) and 95% LoA of -1.6 to 0.9 ml/g/min (Figure 1C+D). The mean global MFR was 1.9 (SD 0.6) as measured by 82Rb PET versus 3.1 (SD 0.9) by [15O]H2O PET, yielding a mean difference of -1.1, (95% CI -1.4 to -0.9).
The median ratio of global MFR measured by 82Rb PET versus [15O]H2O PET was 0.63 (95% CI 0.58-0.68), corresponding to a relative difference of 37% lower (95% CI 32-42) global MFR by 82Rb PET vs. [15O]H2O PET. The 95% LoA of the ratio of global MFR measured by 82Rb PET versus [15O]H2O PET was (0.38-1) corresponding to 95% LoA of the relative difference of 52% lower to 3% higher (Figure 2C+D).
Conclusion
Compared with [15O]H2O PET, 82Rb PET overestimates global MBF at rest, underestimates global MBF at stress, and underestimates global MBFR, despite correction for non-linear extraction. The tracers had wide limits of agreement on MBF and MFR. Caution should be taken in direct comparisons of MBF and MFR obtained by these two tracers.
In stable chest pain patients, Rubidium-82 (82Rb) is commonly used for myocardial perfusion imaging (MPI) and allows for quantification of myocardial blood flow (MBF) and myocardial flow reserve (MFR). With increasing implementation, an alternative is [15O]H2O PET MPI, which has optimal tracer kinetics. However, the level of agreement on MBF and MFR between 82Rb and [15O]H2O PET is sparsely elucidated.
Purpose
To examine the agreement between 82Rb and [15O]H2O PET in the measurement of MBF and MFR.
Methods
This is a prospective cohort study conducted at a single hospital. By protocol, consecutive patients referred for clinically indicated routine 82Rb PET MPI underwent additional and subsequent [15O]H2O PET MPI during the same outpatient visit. Quantitative measurements of MBF and MFR were obtained using Syngo.PET for 82Rb PET with correction for non-linear extraction (1) and Carimas CE for [15O]H2O PET. Quantitative MBF and MFR measures were compared using Bland-Altman analysis.
Results
Forty-five patients were included from February to March 2023 with a mean time between 82Rb stress procedure to [15O]H2O rest procedure of 25 minutes (SD 9 minutes). In total, 44 paired rest, 40 paired stress, and 39 paired rest and stress procedures were available for statistical analyses. A lower rate-pressure product (RPP) (mean difference -335 mmHg x bpm, 95% CI -625 to -46) was observed during the 82Rb procedure vs. the [15O]H2O procedure, while other hemodynamic parameters were similar. Rest MBF was normalised to a RPP of 8000 mmHg x bpm.
The mean global rest MBF was 1.1 (SD 0.3) ml/g/min as measured by 82Rb PET versus 0.8 (SD 0.2) ml/g/min by [15O]H2O PET, yielding a mean difference of 0.3 ml/g/min (95% CI 0.1-0.3) and 95% limits of agreement (LoA) of -0.2 to 0.8 ml/g/min (Figure 1A+B). The mean global stress MBF was 2.1 (SD 0.7) ml/g/min as measured by 82Rb PET versus 2.5 (SD 0.9) ml/g/min by [15O]H2O PET, yielding a mean difference of -0.3 ml/g/min (95% CI -0.5 to -0.1) and 95% LoA of -1.6 to 0.9 ml/g/min (Figure 1C+D). The mean global MFR was 1.9 (SD 0.6) as measured by 82Rb PET versus 3.1 (SD 0.9) by [15O]H2O PET, yielding a mean difference of -1.1, (95% CI -1.4 to -0.9).
The median ratio of global MFR measured by 82Rb PET versus [15O]H2O PET was 0.63 (95% CI 0.58-0.68), corresponding to a relative difference of 37% lower (95% CI 32-42) global MFR by 82Rb PET vs. [15O]H2O PET. The 95% LoA of the ratio of global MFR measured by 82Rb PET versus [15O]H2O PET was (0.38-1) corresponding to 95% LoA of the relative difference of 52% lower to 3% higher (Figure 2C+D).
Conclusion
Compared with [15O]H2O PET, 82Rb PET overestimates global MBF at rest, underestimates global MBF at stress, and underestimates global MBFR, despite correction for non-linear extraction. The tracers had wide limits of agreement on MBF and MFR. Caution should be taken in direct comparisons of MBF and MFR obtained by these two tracers.
| Originalsprog | Engelsk |
|---|---|
| Artikelnummer | jeae142.095 |
| Tidsskrift | European Heart Journal Cardiovascular Imaging |
| Vol/bind | 25 |
| Nummer | Supplement_1 |
| ISSN | 1525-2167 |
| DOI | |
| Status | Udgivet - 27 jun. 2024 |