TL;DR
This study investigates the diagnostic power of a new myocardial perfusion imaging (MPI) system (IQ-SPECT) using thallium-201, both with and without CT-based attenuation correction (CT-AC), to predict fractional flow reserve (FFR). The findings suggest that combining IQ-SPECT with CT-AC can improve the prediction of clinically significant coronary artery disease, offering a valuable non-invasive tool.
What Problem Does This Paper Address?
Diagnosing coronary artery disease (CAD) and determining its functional significance is crucial for patient management. Fractional flow reserve (FFR), measured invasively, is the gold standard for assessing the hemodynamic impact of coronary stenoses. However, its invasive nature limits its widespread use. Myocardial perfusion imaging (MPI) using SPECT is a common non-invasive alternative, but its accuracy can be hampered by artifacts, particularly attenuation artifacts caused by variations in tissue density. Computed tomography-based attenuation correction (CT-AC) has been introduced to mitigate these artifacts, improving specificity and accuracy.
While CT-AC has shown promise with conventional SPECT systems, its impact on newer, advanced systems like the IQ-SPECT, which offers faster acquisition and improved image quality, remains unclear. Specifically, the relationship between FFR and the semiquantitative scores derived from IQ-SPECT, with and without CT-AC, and the optimal thresholds for predicting clinically significant FFR were not well-established. This paper aims to bridge this knowledge gap by evaluating the diagnostic value of IQ-SPECT, with and without CT-AC, in predicting FFR in patients with stable CAD.
Methods
The study employed a prospective, single-center design involving 84 patients with stable CAD who underwent adenosine-stress thallium-201 (Tl-201) MPI-IQ-SPECT/CT. The core of the methodology involved comparing the FFR measurements (the reference standard) with visual semiquantitative interpretations of the MPI-IQ-SPECT images. These interpretations were performed twice: once without CT-AC and once with CT-AC.
Key techniques included:
- IQ-SPECT/CT Imaging: A novel MPI system characterized by a multifocal collimator for magnification and cardiocentric, 3D iterative reconstruction, allowing for rapid image acquisition and high image quality.
- CT-based Attenuation Correction (CT-AC): Non-enhanced CT images were acquired to generate attenuation maps (μ maps), which were then used to correct the SPECT data.
- Visual Semiquantitative Analysis: Two independent observers analyzed the SPECT images, dividing the heart into 17 segments and scoring uptake on a scale of 0-4. Regional summed difference scores (rSDS) were calculated, representing the difference between stress and rest scores, and were found to be the most accurate indicator of FFR.
- Fractional Flow Reserve (FFR) Measurement: Invasive FFR was measured using a pressure sensor-tipped guidewire within 4 weeks of the MPI scan.
- Statistical Analysis: Receiver-operating characteristic (ROC) curve analyses were used to determine optimal FFR cut-offs and compare the diagnostic accuracy of images with and without CT-AC. Net reclassification improvement (NRI) and integrated discrimination improvement (IDI) were employed to assess the added value of CT-AC.
Datasets
The study utilized data from 84 consecutive patients prospectively enrolled at a single center. The dataset comprised 136 diseased vessels for which FFR was successfully measured. No new benchmark datasets were introduced; the study relied on patient data collected for the purpose of this research.
Key Insights and Ideas
- rSDS as the Strongest Predictor: The regional summed difference score (rSDS) emerged as the most accurate semiquantitative parameter from MPI-IQ-SPECT for predicting FFR, showing a moderate inverse correlation. This highlights the importance of assessing the difference between stress and rest perfusion defects.
- CT-AC Enhances FFR Prediction for Higher Values: While the diagnostic accuracy for predicting FFR <0.80 did not significantly differ with or without CT-AC, the accuracy for predicting FFR ≥0.82 was significantly better when CT-AC was applied. This suggests CT-AC is particularly beneficial for distinguishing between clinically insignificant and borderline stenoses.
- Combined Interpretation is Powerful: The study proposed a novel approach of combining interpretations from both with and without CT-AC images. This combined approach demonstrated improved sensitivity and specificity for predicting both clinically significant (FFR <0.80) and insignificant (FFR ≥0.81) FFR compared to using either alone.
- Potential for Non-Obese Patients: The findings suggest that CT-AC can be a valuable addition to MPI-IQ-SPECT for FFR prediction even in patients who are not severely obese, challenging the notion that CT-AC is primarily for obese individuals.
Key Results
- Correlation: FFR inversely correlated most accurately with rSDS, with correlation coefficients of -0.584 (without CT-AC) and -0.568 (with CT-AC), both highly statistically significant (P < .001).
- Optimal FFR Cut-off: An optimal FFR cut-off of <0.80 was identified using rSDS for both with and without CT-AC.
- Improved Accuracy with CT-AC: FFR ≥0.82 was diagnosed significantly more accurately with CT-AC than without it.
- Combined Interpretation Performance:
- Regions with rSDS ≥2 predicted FFR <0.80 with 73% sensitivity and 84% specificity (without CT-AC), and 83% sensitivity and 67% specificity (with CT-AC).
- Regions with rSDS <1 predicted FFR ≥0.81 with 79% sensitivity and 84% specificity (without CT-AC), and 75% sensitivity and 67% specificity (with CT-AC).
- Net Reclassification: CT-AC significantly reclassified 20-25% of FFR predictions at cut-offs of 0.82 and 0.84, indicating its added value.
Limitations and Future Work
The authors acknowledge several limitations:
- Single-Center Study: The findings are from a single institution, which may limit generalizability.
- Small Sample Size: The relatively small number of patients and vessels analyzed could affect the robustness of the conclusions.
- Incomplete Segment Assignment: Not all myocardial segments could be definitively assigned to their corresponding coronary artery territories.
- Previous Interventions: A significant proportion of patients had prior percutaneous coronary interventions, which could potentially influence perfusion patterns.
- Tl-201 Dose: While minimizing radiation is important, the study did not attempt to reduce the Tl-201 dose.
Future work suggested includes:
- Multi-center Validation: A larger, multi-center study is needed to validate these findings and generalize them to a broader patient population.
- Further Investigation for FFR ≤0.80: More research is required to determine if FFR ≤0.80 can be reliably predicted using Tl-201 MPI-IQ-SPECT/CT.
- Advanced MPI/CT Modalities: Further investigation into advanced MPI-SPECT/CT modalities, including CZT semiconductor cameras, is warranted to explore potential improvements in diagnostic ability.
Summary updated