Validation of a Computed Tomography (CT) Based Fractional Flow Reserve (FFR) Software Using the 320 Detector Aquilion ONE CT Scanner.

Conditions

Atherosclerosis, Coronary

Keywords

FFR, Fractional Flow Reserve, Atherosclerosis, Coronary, stenosis

Brief summary

Coronary Computed Tomography Angiography (CCTA) contrast opacification gradients and FFR-CT estimation can aid in the severity estimation of significant atherosclerotic lesions. Currently, FFR-CT algorithms can only be optimized using theoretical models and can only be validated in large multi-center clinical trials. Using patient specific 3D printed coronary phantoms would allow optimization of FFR-CT algorithms with a measured validation technique without the need for large clinical trials. Thus the investigators believe that this study will result in a FFR-CT algorithm/method with a better predictability for arterial lesion severity than those existing on the market today. Flow measurements will be compared with: CT-FFR for both patients and phantoms, angio lab FFR measurements and 30 days follow-up. This pilot clinical study includes \ 50 patients over a year and half at GVI.

Detailed description

Coronary Computed Tomography Angiography (CCTA) contrast opacification gradients and FFR-CT estimation can aid in the severity estimation of significant atherosclerotic lesions. Following this trend, the investigators recently developed a collaboration between Brigham and Women's Hospital (BWH) and Gates Vascular Institute (GVI). The investigators 3D-printed patient specific coronary phantoms at (GVI) and scanned them with a Toshiba Aquilion scanner to test several aspects of the contrast opacification gradients using a method established at BWH. The initial results showed strong correlation between the flow in the phantom and opacification gradients. The investigators believe that this approach could be further developed to test and validate FFR-CT algorithms. Currently, FFR-CT algorithms can only be optimized using theoretical models and can only be validated in large multi-center clinical trials. This phantom approach would allow optimization of FFR-CT algorithms with a measured validation technique without the need for large clinical trials. Thus the investigators believe that this study will result in a FFR-CT algorithm/method with a better predictability for arterial lesion severity than those existing on the market today. The approach is to use the infrastructure at GVI to perform a detailed validation of the FFR-CT method using 3D printed patient specific phantoms. The subject enrollment criteria is: at least one CCTA, at least one lesion with \>50% stenosis or 30-50% and an angio based FFR. Each patient will have a 3D phantom printed, containing the culprit lesion and used in a benchtop flow analysis. Flow measurements will be compared with: CT-FFR for both patients and phantoms, angio lab FFR measurements and 30 days follow-up. This pilot clinical study will include \ 50 patients over a year and half at GVI. The investigators are confident that this approach performed via 3D-phantom testing will prove the validity of FFR-CT based measurements as well as develop a new standard for validating FFR-CT algorithms.

K. Sommer, Lauren Shepard, Dimitrios Mitsouras, Vijay Iyer, Erin Angel et al. (12 authors)

Biomedical Physics & Engineering Express2020-05-0113 citations

10.1088/2057-1976/ab8f6e

Interventions

DIAGNOSTIC_TESTCCTA

Diagnostic Test

Study design

Observational model

COHORT

Time perspective

PROSPECTIVE

Eligibility

Sex/Gender

ALL

Age

18 Years to No maximum

Healthy volunteers

No

Inclusion criteria

* All the patients \>18 yrs of age , who are undergoing CCTA and angio-FFR. Patients who are (1) scheduled for clinically mandated elective invasive coronary angiography (ICA) at Buffalo General Hospital or (2) clinically mandated CTA will be screened.

Exclusion criteria

* Adults unable to consent * Individuals who are not yet adults (infants, children, teenagers) * Pregnant women * Prisoners * atrial fibrillation, * Renal insufficiency (estimated glomerular filtration rate (GFR) \<60 ml/min/1.73 m2), * Active Bronchospasm prohibiting the use of beta blockers * Morbid obesity (body mass index 40 kg/m2) * Contraindications to iodinated contrast. * Emergencies requiring immediate intervention or patients unable to consent. * Patients not showing coronary calcium during Calcium Scoring procedures

Design outcomes

Primary

Measure	Time frame	Description
Comparison of CT Based FFR With Invasive FFR, ROC Analysis	24 hours	Patient CCTA images were imported into Vitrea segmentation software (Vital Images, Minnetonka, MN) for use in the research-based CT based FFR algorithm. The software analyzes four data volumes acquired a 70%, 80%, 90% and 99% of the R-R interval and computes the FFR based on the changes in vessel diameter and computational fluid dynamics. Within the algorithm, the aortic root and three main coronary arteries (LAD, LCX, and RCA) were automatically segmented, and then manually adjusted to obtain accurate centerline and contours. The CT based FFR was calculated and the user adjusted the location of the distal pressure measurement to calculate the CT basedFFR at the same location as Invasive-FFR, two lesion lengths below the distal end of the lesion. Area under the Receiver Operator Characteristic were measured where an Invasive FFR\<=0.8 was considered positive.
Comparison of CT Based FFR With Invasive FFR, Correlation Analysis	24 hours	Patient CCTA images were imported into Vitrea segmentation software (Vital Images, Minnetonka, MN) for use in the research-based CT based FFR algorithm. The software analyzes four data volumes acquired a 70%, 80%, 90% and 99% of the R-R interval and computes the FFR based on the changes in vessel diameter and computational fluid dynamics. Within the algorithm, the aortic root and three main coronary arteries (LAD, LCX, and RCA) were automatically segmented, and then manually adjusted to obtain accurate centerline and contours. The CT based FFR was calculated and the user adjusted the location of the distal pressure measurement to calculate the CT basedFFR at the same location as Invasive-FFR, two lesion lengths below the distal end of the lesion. Pearson Correlation between Invasive FFR and CT based FFR was measured
Comparison of CT Based FFR With Invasive FFR, Sensitivity	24 hours	Patient CCTA images were imported into Vitrea segmentation software (Vital Images, Minnetonka, MN) for use in the research-based CT based FFR algorithm. The software analyzes four data volumes acquired a 70%, 80%, 90% and 99% of the R-R interval and computes the FFR based on the changes in vessel diameter and computational fluid dynamics. Within the algorithm, the aortic root and three main coronary arteries (LAD, LCX, and RCA) were automatically segmented, and then manually adjusted to obtain accurate centerline and contours. The CT based FFR was calculated and the user adjusted the location of the distal pressure measurement to calculate the CT basedFFR at the same location as Invasive-FFR, two lesion lengths below the distal end of the lesion. Sensitivity were measured where an Invasive FFR\<=0.8 was considered positive. Sensitivity reflects the percentage of true positive cases identified by CT-FFR compared to I-FFR
Comparison of CT Based FFR With Invasive FFR, Specificity	24 hours	Patient CCTA images were imported into Vitrea segmentation software (Vital Images, Minnetonka, MN) for use in the research-based CT based FFR algorithm. The software analyzes four data volumes acquired a 70%, 80%, 90% and 99% of the R-R interval and computes the FFR based on the changes in vessel diameter and computational fluid dynamics. Within the algorithm, the aortic root and three main coronary arteries (LAD, LCX, and RCA) were automatically segmented, and then manually adjusted to obtain accurate centerline and contours. The CT based FFR was calculated and the user adjusted the location of the distal pressure measurement to calculate the CT basedFFR at the same location as Invasive-FFR, two lesion lengths below the distal end of the lesion. Specificity was measured, where an Invasive FFR\<=0.8 was considered positive. Specificity reflects the percentage of true negative cases identified by CT-FFR compared to I-FFR

Secondary

Measure	Time frame	Description
Comparison of CT Based FFR With Bench-top FFR Using 3D Printed Patient Specific Phantoms	4 weeks from baseline	CT images were used to measure CT-FFR and to generate patient-specific 3D printed models of the aortic root and three main coronary arteries. Each patient-specific 3D printed model was connected to a programmable pulsatile pump and bench-top FFR (B-FFR) was derived from pressures measured proximal and distal to coronary stenosis using pressure transducers. B-FFR was measured for hyperemic, 500 mL/min by adjusting the model's distal coronary resistance. Linear regression and Pearson correlation was calculated.
Comparison of Bench-top FFR Using 3D Printed Patient Specific Phantoms With Invasive FFR, Specificity	4 weeks from baseline	CT images were used to create patient specific 3d-printed phantom. Each patient-specific 3D printed model was connected to a programmable pulsatile pump and benchtop FFR (B-FFR) was derived from pressures measured proximal and distal to coronary stenosis using pressure transducers. B-FFR was measured for hyperemic, 500 mL/min by adjusting the model's distal coronary resistance. Benchtop-FFR was compared with Invasive-FFR. Specificity was calculated, where an Invasive FFR\<=0.8 was considered positive. Specificity reflects the percentage of true negative cases identified by B-FFR compared to I-FFR
Comparison of Bench-top FFR Using 3D Printed Patient Specific Phantoms With Invasive FFR, ROC Analysis	4 weeks from baseline	CT images were used to create patient specific 3d-printed phantom. Each patient-specific 3D printed model was connected to a programmable pulsatile pump and benchtop FFR (B-FFR) was derived from pressures measured proximal and distal to coronary stenosis using pressure transducers. B-FFR was measured for hyperemic, 500 mL/min by adjusting the model's distal coronary resistance. Benchtop-FFR was compared with Invasive-FFR. Area under the Receiver Operator Characteristic were measured where an Invasive FFR\<=0.8 was considered positive.
Comparison of Bench-top FFR Using 3D Printed Patient Specific Phantoms With Invasive FFR, Pearson Correlation	4 weeks from baseline	CT images were used to create patient specific 3d-printed phantom. Each patient-specific 3D printed model was connected to a programmable pulsatile pump and benchtop FFR (B-FFR) was derived from pressures measured proximal and distal to coronary stenosis using pressure transducers. B-FFR was measured for hyperemic, 500 mL/min by adjusting the model's distal coronary resistance. Benchtop-FFR was compared with Invasive-FFR. Pearson Correlation factor was calculated.
Comparison of Bench-top FFR Using 3D Printed Patient Specific Phantoms With Invasive FFR, Sensitivity	4 weeks from baseline	CT images were used to create patient specific 3d-printed phantom. Each patient-specific 3D printed model was connected to a programmable pulsatile pump and benchtop FFR (B-FFR) was derived from pressures measured proximal and distal to coronary stenosis using pressure transducers. B-FFR was measured for hyperemic, 500 mL/min by adjusting the model's distal coronary resistance. Benchtop-FFR was compared with Invasive-FFR. Sensitivity was measure, where an Invasive FFR\<=0.8 was considered positive.Sensitivity reflects the percentage of true positive cases identified by B-FFR compared to I-FFR

Countries

United States

Participant flow

Participants by arm

Arm	Count
CCTA Patients who are scheduled for clinically mandated elective invasive coronary angiography (ICA) at Buffalo General Hospital. CCTA Coronary: Diagnostic Test	52
Total	52

Withdrawals & dropouts

Period	Reason	FG000
Overall Study	Invasive FFR Measurement cancelled	1
Overall Study	Invasive FFR not performed	21
Overall Study	Poor CT Image Quality	1

Baseline characteristics

Characteristic	CCTA
Age, Categorical <=18 years	0 Participants
Age, Categorical >=65 years	30 Participants
Age, Categorical Between 18 and 65 years	22 Participants
Age, Continuous	64.7 years STANDARD_DEVIATION 10.4
BMI	25.2 kg/m^2 STANDARD_DEVIATION 3.9
Coronary Calcium Score	385 Coronary Calcium Score STANDARD_DEVIATION 388
Creatinine	0.83 mg/ dl STANDARD_DEVIATION 0.45
Diabetes Mellitus No	30 Participants
Diabetes Mellitus Yes	22 Participants
Ethnicity (NIH/OMB) Hispanic or Latino	0 Participants
Ethnicity (NIH/OMB) Not Hispanic or Latino	0 Participants
Ethnicity (NIH/OMB) Unknown or Not Reported	52 Participants
FFR FFR<=0.8	23 Participants
FFR FFR>0.8	29 Participants
Hyperlipidemia No	14 Participants
Hyperlipidemia Yes	38 Participants
Hypertension No	19 Participants
Hypertension Yes	33 Participants
Prior Myocardial Infarction No	48 Participants
Prior Myocardial Infarction Yes	4 Participants
Region of Enrollment Japan	35 Participants
Region of Enrollment United States	17 Participants
Sex: Female, Male Female	20 Participants
Sex: Female, Male Male	32 Participants
Smoking Current	10 Participants
Smoking Former	16 Participants
Smoking Never	26 Participants

Adverse events

Event type	EG000 affected / at risk
deaths Total, all-cause mortality	0 / 52
other Total, other adverse events	0 / 52
serious Total, serious adverse events	0 / 52

Outcome results

Primary

Comparison of CT Based FFR With Invasive FFR, Correlation Analysis

Patient CCTA images were imported into Vitrea segmentation software (Vital Images, Minnetonka, MN) for use in the research-based CT based FFR algorithm. The software analyzes four data volumes acquired a 70%, 80%, 90% and 99% of the R-R interval and computes the FFR based on the changes in vessel diameter and computational fluid dynamics. Within the algorithm, the aortic root and three main coronary arteries (LAD, LCX, and RCA) were automatically segmented, and then manually adjusted to obtain accurate centerline and contours. The CT based FFR was calculated and the user adjusted the location of the distal pressure measurement to calculate the CT basedFFR at the same location as Invasive-FFR, two lesion lengths below the distal end of the lesion. Pearson Correlation between Invasive FFR and CT based FFR was measured

Time frame: 24 hours

Population: From the total number of patients, nine patients had multi-vessel disease and I-FFR was measured in two vessels. In total I-FFR was measured in: 42 LADs, 11 LCXs and 8 RCAs. Invasive -FFR measurement location was from the ostium to two lesion lengths below the distal throat of the lesion .

Arm	Measure	Value (NUMBER)
CCTA	Comparison of CT Based FFR With Invasive FFR, Correlation Analysis	0.75 correlation coefficient

Primary

Comparison of CT Based FFR With Invasive FFR, ROC Analysis

Patient CCTA images were imported into Vitrea segmentation software (Vital Images, Minnetonka, MN) for use in the research-based CT based FFR algorithm. The software analyzes four data volumes acquired a 70%, 80%, 90% and 99% of the R-R interval and computes the FFR based on the changes in vessel diameter and computational fluid dynamics. Within the algorithm, the aortic root and three main coronary arteries (LAD, LCX, and RCA) were automatically segmented, and then manually adjusted to obtain accurate centerline and contours. The CT based FFR was calculated and the user adjusted the location of the distal pressure measurement to calculate the CT basedFFR at the same location as Invasive-FFR, two lesion lengths below the distal end of the lesion. Area under the Receiver Operator Characteristic were measured where an Invasive FFR\<=0.8 was considered positive.

Time frame: 24 hours

Population: From the total number of patients, nine patients had multi-vessel disease and I-FFR was measured in two vessels. In total I-FFR was measured in: 42 LADs, 11 LCXs and 8 RCAs. Invasive -FFR measurement location was from the ostium to two lesion lengths below the distal throat of the lesion .

Arm	Measure	Value (NUMBER)
CCTA	Comparison of CT Based FFR With Invasive FFR, ROC Analysis	0.8 probability of accurate diagnosis

Primary

Comparison of CT Based FFR With Invasive FFR, Sensitivity

Patient CCTA images were imported into Vitrea segmentation software (Vital Images, Minnetonka, MN) for use in the research-based CT based FFR algorithm. The software analyzes four data volumes acquired a 70%, 80%, 90% and 99% of the R-R interval and computes the FFR based on the changes in vessel diameter and computational fluid dynamics. Within the algorithm, the aortic root and three main coronary arteries (LAD, LCX, and RCA) were automatically segmented, and then manually adjusted to obtain accurate centerline and contours. The CT based FFR was calculated and the user adjusted the location of the distal pressure measurement to calculate the CT basedFFR at the same location as Invasive-FFR, two lesion lengths below the distal end of the lesion. Sensitivity were measured where an Invasive FFR\<=0.8 was considered positive. Sensitivity reflects the percentage of true positive cases identified by CT-FFR compared to I-FFR

Time frame: 24 hours

Population: From the total number of patients, nine patients had multi-vessel disease and I-FFR was measured in two vessels. In total I-FFR was measured in: 42 LADs, 11 LCXs and 8 RCAs. Invasive -FFR measurement location was from the ostium to two lesion lengths below the distal throat of the lesion .

Arm	Measure	Value (NUMBER)
CCTA	Comparison of CT Based FFR With Invasive FFR, Sensitivity	82.61 percentage of true positive cases

Primary

Comparison of CT Based FFR With Invasive FFR, Specificity

Patient CCTA images were imported into Vitrea segmentation software (Vital Images, Minnetonka, MN) for use in the research-based CT based FFR algorithm. The software analyzes four data volumes acquired a 70%, 80%, 90% and 99% of the R-R interval and computes the FFR based on the changes in vessel diameter and computational fluid dynamics. Within the algorithm, the aortic root and three main coronary arteries (LAD, LCX, and RCA) were automatically segmented, and then manually adjusted to obtain accurate centerline and contours. The CT based FFR was calculated and the user adjusted the location of the distal pressure measurement to calculate the CT basedFFR at the same location as Invasive-FFR, two lesion lengths below the distal end of the lesion. Specificity was measured, where an Invasive FFR\<=0.8 was considered positive. Specificity reflects the percentage of true negative cases identified by CT-FFR compared to I-FFR

Time frame: 24 hours

Population: From the total number of patients, nine patients had multi-vessel disease and I-FFR was measured in two vessels. In total I-FFR was measured in: 42 LADs, 11 LCXs and 8 RCAs. Invasive -FFR measurement location was from the ostium to two lesion lengths below the distal throat of the lesion .

Arm	Measure	Value (NUMBER)
CCTA	Comparison of CT Based FFR With Invasive FFR, Specificity	76.32 percentage of of true negative cases

Secondary

Comparison of Bench-top FFR Using 3D Printed Patient Specific Phantoms With Invasive FFR, Pearson Correlation

CT images were used to create patient specific 3d-printed phantom. Each patient-specific 3D printed model was connected to a programmable pulsatile pump and benchtop FFR (B-FFR) was derived from pressures measured proximal and distal to coronary stenosis using pressure transducers. B-FFR was measured for hyperemic, 500 mL/min by adjusting the model's distal coronary resistance. Benchtop-FFR was compared with Invasive-FFR. Pearson Correlation factor was calculated.