Response to Comments: Non-Invasive Fractional Flow Reserve (FFR) for Stable Ischemic Heart Disease

As a practicing Cardiologist, I use HeartFlow's Fractional Flow Reserve - Computed Tomography (FFRCT) regularly as means of discerning which patients can be managed by medial therapy vs. which are more likely to benefit from invasive revascularization. As a physician who has adopted FFR-CT into my clinical practice, I am pleased to see Palmetto's draft LCD and respectfully submit my comments with regards to the following areas:

1. Expand included coronary stenosis severity range to 40-90%
2. Coronary Stents
3. Remove limitations related to coronary CT angiography (CCTA) image quality

1. Expand included coronary stenosis severity range to 40-90%

I fully support the expansion of stenosis ranges to 40-90%, this will ensure we only send appropriate patients to the cath lab, this is also in agreement with the most current chest pain guidelines by ACC.

• Intermediate risk with acute chest pain and no known coronary artery disease, with coronary artery stenosis of 40-90% in proximal or middle coronary artery on CCTA or
• Intermediate risk with acute chest pain and known coronary artery stenosis of 40-90% in a proximal or middle segment on CCTA or
• Stable nonobstructive coronary artery disease with persistent symptoms requiring further test, and 40-90% stenosis on CCTA and
• Not in conjunction with stress testing (unless FFRCT was not high quality and alternative study needed)

I would like to suggest Palmetto include stable chest pain as well as acute in the indications above, consistent with the revised 2021 ACC/AHA guidelines.

2. Coronary Stents

FFRCT will not be performed on a coronary system that contains a metallic stent; however, this does not mean that we cannot request functional information on any other non-stented coronary system. For example, if there is a stent in the right coronary artery (RCA), we can still image the heart and receive FFRCT results on a suspected stenosis in the left anterior descending artery (LAD). By disqualifying all patients with a stent, you are restricting access to FFRCT results in non-stented vessels. I would recommend that the current non-coverage of bare metal intra-cardiac stents be refined and replaced with the following restrictions on coverage for coronary stents.

1. A stent is present in the left main coronary artery
2. There is a stenosis > 30% in the left main coronary artery and 1 or more stent in the left system
3. There are stents present in 2 or more coronary systems (e.g., the left circumflex artery and the left anterior descending artery)

3. Remove limitations related to coronary CT angiography (CCTA) image quality

There are several coverage restrictions in the proposed LCD for FFRCT that are related to image quality of the CCTA which may affect its acceptability for FFRCT analysis. I would propose that the below restrictions related to CTA image quality be removed, these include:

• Prior placement of prosthetic valves
• Known severe aortic stenosis
• Prior pacemaker or defibrillator lead placement

I personally oversee the CCTA scans of my patients, and with our modern CT scanners I am able to produce high quality images for many patients with prosthetic valves, aortic stenosis, as well as any pacemaker or defibrillator leads in most cases. These proposed restrictions should be removed to ensure adequate access for Medicare beneficiaries. The company also does not perform FFRCT when the image data set is of insufficient quality for accurate interpretation.

Of note, the CAST-FFR study was recently published and validates the use of FFRCT in patients with known severe aortic stenosis.

Conclusion

I appreciate your focus and attention on updating the FFRCT LCD to reflect the guideline updates by the ACC and AHA. I encourage you to make these updates expeditiously so that I can continue to drive better patient care and outcomes leveraging the FFRCT technology. I look forward to ongoing engagement with Palmetto to develop policies that support my ability to focus on delivering high quality care to my patients.

References were provided for review.

In response to the addition of stable chest pain to the indications - stable chest pain has been added to align with ACC/AHA Chest Pain Guidelines.¹

Regarding coronary stents - The request to remove intracoronary metallic stents was accompanied by a suggested limitation which the manufacturer provided as appropriate exclusion in the setting of stents, but without supporting literature. The challenges in reproducible image quality are a known limitation of the technology in the presence of intracoronary stenting. Moreover, the pivotal trials consistently excluded patients with prior percutaneous coronary intervention from participation. Upon review, we find no basis for coverage with the currently available evidence. If availability of evidence to support the efficacy of FFRCT analysis in the presence of intracoronary stents, we can reconsider coverage upon request.

Regarding removal of limitation related to image quality - There are several reasons we will not remove these limitations at this time:

1. This policy is foundational so covers any technology that would provide equivalent service. If a new product that is equivalent to what is currently available emerges it will be held to the requirements within this policy. While current technology returns the images if quality is low, we cannot assume that will be the case for all future technologies.
2. To ensure safety and accuracy of results, literature is needed to support the removal of these limitations. Pivotal trials consistently excluded patients with prior percutaneous coronary interventions.
3. Incomplete reporting of coronary artery anatomy defeats the purpose of this diagnostic test.

If literature or guidelines provide support for the safety and accuracy for CCTA and FFR_CT in this population, that can be evaluated with reconsideration.

1. Gulati M, Levy PD, Mukherjee D, et al. 2021 AHA/ACC/ASE/CHEST/SAEM/SCCT/SCMR Guideline for the Evaluation and Diagnosis of Chest Pain: Executive Summary: A Report of the American College of Cardiology/American Heart Association Joint Committee on Clinical Practice Guidelines. Circulation. 2021. 0000000000001030.
2. Maron DJ, Hochman JS, O'Brien SM, et al. International study of comparative health effectiveness with medical and invasive approaches (ISCHEMIA) trial: Rationale and design. Am Heart J. 2018;201:124-135.
3. Vavere AL, Arbab-Zadeh A, Rochitte CE, et al. Coronary artery stenoses: Accuracy of 64 detector row CT angiography in segments with mild, moderate, or severe calcification – A subanalysis of the CORE-64 trial. Radiology. 2011;261:100-108.
4. Collet C, Miyazaki Y, Ryan N, et al. Fractional flow reserve derived from computed tomographic angiography in patients with multivessel CAD. Journal of the American College of Cardiology. 2018;71(24):2756-2769.
5. De Bruyne B, Pijls NH, Kalesan B, et al. Fractional flow reserve–guided PCI versus medical therapy in stable coronary disease. N Engl J Med. 2012;367(11):991-1001.
6. Michail M, Ihdayhid AR, Comella A, et al. Feasibility and validity of computed tomography - Derived fractional flow reserve in patients with severe aortic stenosis: The CAST-FFR study. Circulation: Cardiovascular Interventions. 2021;14(1):e009586.
7. Collet C, Miyazaki Y, Ryan N, et al. Fractional flow reserve derived from computed tomographic angiography in patients with multivessel CAD. J Am Coll Cardiol. 2018;71(24):2756-2769.

Thank you for the opportunity to comment on the draft LCD and provide additional insights that will ensure appropriate access to FFRCT for Medicare beneficiaries who present with symptoms suggestive of stable ischemic heart disease (SIHD). I applaud the swiftness with which Palmetto has undertaken to revise this LCD in the wake of recent new American College of Cardiology (ACC) and American Heart Association (AHA) Guideline for the Evaluation and Diagnosis of Chest Pain. The expansion of coverage to coronary artery stenoses ranging from 40%-90% to conform to the new Guideline is critically important to ensure access to all Medicare beneficiaries who may benefit from the HeartFlow FFRCT Analysis. In addition, we would like to comment on and propose revisions to the proposed LCD as follows:

Suggested modifications to Indications (details are in following sections):

1. Include coverage for both stable and acute chest pain patients in the indications, to conform to the ACC/AHA Chest Pain Guideline
2. Combine the two indications for “no known” and “known” CAD into one, as once a CCTA is performed and FFRCT ordered, CAD is always “known”
3. Change “FFRCT” into “CCTA” in last indication of the draft LCD

Suggested modifications to Restrictions (details are in following sections):

1. Remove restrictions related to image quality, valves (Restriction 1), known severe aortic stenosis (Restriction 2) and pacemaker or defibrillator leads (Restriction 8)
2. Revise language of the restriction related to intracoronary metallic stents (Restriction 5) to match verbiage in HeartFlow’s Instructions For Use (IFU)
3. Revise Restriction 11 from “<50%” to “<40%” for consistency within the draft LCD

Suggested modification to coverage guidance support paragraph (details are in following sections):

1. Revise “stable” to “stable or acute” in the first sentence.
2. Revise “If higher grade stenoses (i.e., > 90%) are present” to “If higher grade stenoses (i.e., greater than 90%) are present, and no other stenoses 40% - 90% are present.”
3. Revise “less than 30%” to “less than 40%” for consistency within the draft LCD.
4. Remove “if extensive plaque is present a high quality CCTA is unlikely and stress testing is preferred”.

Background

As described in the ‘Summary of Evidence’ section of the proposed LCD, FFRCT allows physicians to choose safely the most appropriate treatment for their patients with coronary artery disease (CAD). Use of FFRCT has been demonstrated to change patient management in 63% of cases and has allowed safe avoidance of invasive coronary angiography (ICA) in > 60% of cases where ICA was initially intended for patients with stable symptoms of SIHD. Long term (1-5 years) outcomes have been studied and reported in over 6,000 patients indicating the safety of patient management strategies that incorporate FFRCT. Finally, clinical pathways that utilize FFRCT significantly reduce healthcare costs.

Rationale for Adding, Removing, or Modifying Coverage Indications and Restrictions in the Proposed LCD

HeartFlow supports the expansion of stenosis ranges from 40-70% to the proposed 40-90% by CCTA. According to the most recent chest pain guidelines, released by AHA and ACC in October 2021, the recommendation is that FFRCT be used in patients with 40-90% stenoses on CCTA. Additionally, because of the clinical utility of FFRCT in patients with moderate (30–50%) and severe (70–90%) stenoses and the potential for markedly different treatment pathways based on FFRCT, the Society of Cardiovascular CT published a Consensus Document in 2021 stipulating that FFRCT is recommended in patients with 30-90% stenosis. We therefore support Palmetto’s update in the LCD to cover use of FFRCT for lesions with a 40–90% lumen reduction in the left anterior descending, circumflex, and right coronary arteries. This will allow physicians to use FFRCT information to make optimal treatment decisions for their patients in accordance with the updated societal chest pain guidelines.

The language for some Indications should be modified

The indications as written in the draft LCD only reference “Acute” chest pain patients. According to numerous studies including many referenced in the draft LCD, as well as the new Chest Pain Guideline from the AHA and ACC, FFRCT is indicated for both “Stable” and “Acute” chest pain patients.

HeartFlow recommends that the first two indications, which are nearly identical except for the reference to “no known” vs “known” CAD, be combined into just one indication. To include “no known” CAD will be confusing to providers as CAD is always “known” when an FFRCT is needed. To be consistent with the updated ACC/AHA Guideline we suggest revising the two indications into one.

Including the “Stable” language as mentioned above, this indication could be written most simply as:

“Intermediate-risk with stable or acute chest pain with coronary artery stenosis from 40-90% on CCTA”

HeartFlow would also like to suggest a revision to the final indication: “Not in conjunction with stress testing (unless FFRCT was not high quality and alternative study needed).”

This should be revised as follows: “Not in conjunction with stress testing (unless CCTA was returned by the FFRCT provider as not sufficient quality for FFRCT, and an alternative study is needed).” The quality of the CCTA as defined by the FFRCT Analysis process determines whether an FFRCT Analysis can be provided. If the images from the CCTA do not meet these image quality standards, no FFRCT Analysis will be provided.

The language for some Restrictions should be modified

The FFRCT Analysis involves creation of a patient-specific anatomical model of the coronary arteries based on CCTA images. Accuracy of the anatomic model is dependent on the ability to: (a) define the lumen boundaries of the entire coronary tree, (b) see all major branches, and (c) segment the myocardium. HeartFlow has a validated CT image quality process that assesses the quality of incoming CCTA images. This process ensures that the above anatomical features can be resolved adequately to enable computation of FFRCT. CCTA datasets that do not meet the quality criteria are rejected, FFRCT results are not produced, and neither the health facility nor the patient is charged. The image quality process has been validated by the FDA and has demonstrated the ability to produce FFRCT accuracy in line with the clinical literature and our FDA label.

Multiple clinical trials have been conducted using this CCTA acceptance process and all have demonstrated that FFRCT has high and consistent accuracy compared to invasive FFR. Investigators in the NXT study demonstrated high per-vessel diagnostic performance [Accuracy 86%, Sensitivity 84%, specificity 86%] for FFRCT compared to invasive FFR. The PACIFIC FFRCT study showed that FFRCT demonstrated higher diagnostic performance than CCTA, SPECT, and PET for vessel-specific ischemia.

This information is shared to demonstrate that regardless of the underlying characteristics of the patient or CCTA image itself, accurate FFRCT results can be provided if the CCTA images pass HeartFlow’s quality criteria. As a reminder, there is no charge for FFRCT for CCTA exams that fail the CCTA acceptance process. The image quality process and HeartFlow’s ability to provide accurate FFRCT results are independent of patient characteristics, which forms the basis for our argument to remove these specific restrictions from the LCD.

With this information in mind HeartFlow proposes that the following Restrictions be removed from the LCD in their entirety.

• Known severe aortic stenosis (Restriction 2)

As stated above, HeartFlow can provide accurate and actionable FFRCT results regardless of a patient’s aortic valve status. The left ventricular hypertrophy which accompanies aortic stenosis is accounted for in the process of generating the HeartFlow Analysis, as myocardial mass determined from the CCTA images is measured and incorporated into the analysis. There is no evidence that the form-function relationships which underlie the HeartFlow Analysis are not preserved in patients with severe aortic stenosis.

Furthermore, CCTA and the HeartFlow Analysis can be highly useful prior to TAVR procedures by enabling avoidance of the need for invasive angiography. There has been a recent publication validating the HeartFlow FFRCT Analysis in this population.

• Prior placement of prosthetic valves (Restriction 1), pacemaker leads, or defibrillator leads (Restriction 8)

The presence of prosthetic valves, pacemaker or defibrillator leads may not interfere with imaging of the coronary arteries by CCTA. As noted above, accurate FFRCT can be produced when the coronary arteries can be sufficiently visualized, and their lumen boundaries clearly defined. The presence of prosthetic valves, or pacemaker or defibrillator leads often does not preclude this. CCTA datasets that do not meet the quality criteria (for example because of imaging artifacts due to leads or prosthetic valves) are rejected, FFRCT results are not produced, and neither the health facility nor the patient is charged. Therefore, this exclusion should be removed.

• Intracoronary metallic stents (Restriction 5)

Regarding the restriction for patients with “intracoronary metallic stents”, Palmetto should modify the final LCD to align with HeartFlow’s IFU. A suggested revision to the Restriction would be as follows, corresponding to the IFU:

“Prior coronary metallic stent placement in the following cases:

1. A metallic stent is present in the left main coronary artery.
2. There is a stenosis > 30% in the left main coronary artery and 1 or more metallic stents in the left system.
3. There are metallic stents present in 2 or more coronary systems (e.g., the left circumflex coronary artery and the left anterior descending coronary artery)”

FFRCT values will be provided for all vessels which do not have a metallic coronary stent, enabling the physician to determine the functional significance of SIHD in vessels which have not been previously stented. The vessel with the metallic stent will appear gray and calculated FFRCT values will not be provided for that vessel.

• Non-obstructing stenoses (Restriction 11)

Lastly, HeartFlow would like to address Restriction 11: “non-obstructing stenosis (<50%) of all major epicardial vessels) on CTA or catherization in the past twelve months, in the absence of a new symptom complex.”

We would suggest that “<50%” be revised to “<40%” for consistency within the LCD.

Modifications for the Coverage Guidance Support paragraph

This section addresses the draft Coverage Guidance Support paragraph to align with the recent ACC/AHA guideline, previous MAC clarifications made by CMDs, and image quality metrics as described above. The support paragraph in question is below, with those points for suggested revisions made bold.

“This service should be performed in patients with stable coronary symptoms. It should not be performed until after the base study (CCTA) has been completed and interpreted. If higher grade stenoses (i.e., greater than 90%) are present, this study is not medically necessary, as the patient should proceed to catheterization. Similarly, low-grade stenoses (less than 30%) do not require additional confirmatory data. This should be performed as an alternative to stress testing. If extensive plaque is present a high quality CCTA is unlikely and stress testing is preferred.”

• We suggest revision from “stable” to “stable or acute” in accordance with 2021 ACC/AHA Guideline.
• Several prior discussions with MAC Medical Directors have clarified for providers that FFRCT is covered for lesions <90% in patients who also have lesions >90%.
  • We suggest revision as follows: “If higher grade stenoses (i.e., greater than 90%) are present, and no other stenoses 40%-90% are identified, this study is not medically necessary.”
• We also suggest revision of “less than 30%” to “less than 40%” for consistency within the LCD and with 2021 ACC/AHA Guideline.
• We suggest removal of the last statement in the support paragraph. High plaque volume does not necessarily impair CCTA quality. Several studies have demonstrated utility in patients with severe multivessel coronary artery disease, and therefore with high plaque burden.

Conclusion

We appreciate your attention to this important matter and your consideration of our comments.

References were provided for review.

We thank you for your comments and have made changes regarding your suggestions. Regarding aortic stenosis and calcified plaques - Previous studies have demonstrated coronary CTA images in the presence of calcified plaques risk overestimation of stenosis severity and reduced agreement between the CTA and invasive angiography.³ The Syntax 3 study⁴ suggests that the overestimation of CTA-SS was corrected by FFR_CT, resulting in higher correlation to invasive angiogram than CTA alone. This was reproduced in the FAME trial.⁵ Therefore, the statement regarding excessive plaque formation is removed from the policy.

The 2021 CAST-FFR study supports FFR_CT in the setting of severe aortic stenosis so this limitation is removed from the policy.⁶ The remainder of the issues addressed in your comments were addressed in response #1, above.

1. Gulati M, Levy PD, Mukherjee D, et al. 2021 AHA/ACC/ASE/CHEST/SAEM/SCCT/SCMR Guideline for the Evaluation and Diagnosis of Chest Pain: Executive Summary: A Report of the American College of Cardiology/American Heart Association Joint Committee on Clinical Practice Guidelines. Circulation. 2021. 0000000000001030.
2. Maron DJ, Hochman JS, O'Brien SM, et al. International study of comparative health effectiveness with medical and invasive approaches (ISCHEMIA) trial: Rationale and design. Am Heart J. 2018;201:124-135.
3. Vavere AL, Arbab-Zadeh A, Rochitte CE, et al. Coronary artery stenoses: Accuracy of 64 detector row CT angiography in segments with mild, moderate, or severe calcification – A subanalysis of the CORE-64 trial. Radiology. 2011;261:100-108.
4. Collet C, Miyazaki Y, Ryan N, et al. Fractional flow reserve derived from computed tomographic angiography in patients with multivessel CAD. Journal of the American College of Cardiology. 2018;71(24):2756-2769.
5. De Bruyne B, Pijls NH, Kalesan B, et al. Fractional flow reserve–guided PCI versus medical therapy in stable coronary disease. N Engl J Med. 2012;367(11):991-1001.
6. Michail M, Ihdayhid AR, Comella A, et al. Feasibility and validity of computed tomography - Derived fractional flow reserve in patients with severe aortic stenosis: The CAST-FFR study. Circulation: Cardiovascular Interventions. 2021;14(1):e009586.
7. Collet C, Miyazaki Y, Ryan N, et al. Fractional flow reserve derived from computed tomographic angiography in patients with multivessel CAD. J Am Coll Cardiol. 2018;71(24):2756-2769.

I appreciate Palmetto’s attention to the draft LCD (DL38278) for FFRCT and request that the comments below are incorporated within, and the finalized LCD is released quickly as the expanded stenosis ranges will support me in extending patient access and improving patient outcomes using FFRCT. With that said, I am in full support of the expansion of the stenosis ranges to include 40-90% by CTA.

There are two other areas of the LCD that I believe could be streamlined. The first are those restrictions related to CTA image quality, specifically prosthetic valves and prior pacemaker and defibrillator leads. These restrictions are unnecessary as HeartFlow has very strict image quality guidelines already in place. If there is artifact in the CTA image that does not allow HeartFlow to clearly see the lumen boundaries, the case is returned, and no analysis is completed. With modern scanners and optimized protocols, we are able to image the majority of patients with no issues.

The second point I would like to make is regarding a statement made within the support paragraph following the restriction section. The statement in question is, “If extensive plaque is present a high quality CCTA is unlikely and stress testing is preferred.” High plaque volume does not necessarily impair CCTA quality. There are many studies that support this including Syntax 2 and 3. I would recommend the removal of this statement as to not limit access to Medicare patients that may have high plaque burden.

Palmetto’s consideration of the comments is appreciated. I look forward to ongoing engagement with Palmetto to develop policies that support clinicians’ ability to focus on delivering high quality care to patients.

References were provided for review.

Thank you for your response, all issues in your response were addressed above.

Regarding the proposed LCD for Non-Invasive Fractional Flow Reserve (FFR) for Stable Ischemic Heart Disease (DL38278), we would like to make the following comments:

Note: We have limited our comments to those scenarios that are addressed by AIM programs and therefore, did not comment on scenarios for use of FFR in the evaluation of ACUTE chest pain.

Covered scenarios

1. The second paragraph under Coverage Guidance suggests that FFRCT is only to be used in patients with "stable" CAD. Then, in contradiction, the first two criteria refer to patients with ACUTE chest pain (which is not usually considered to be stable CAD); and the third scenario is the only one to address those with “stable” CAD.
2. The following language requires clarification:

• “Stable nonobstructive coronary artery disease with persistent symptoms requiring further test, and 40-90% stenosis on CCTA.” Nonobstructive CAD is defined as <50% stenosis (2021 AHA/ACC/ASE/CHEST/SAEM/SCCT/SCMR Guideline for the Evaluation and Diagnosis of Chest Pain). Therefore, the above quoted text would apply ONLY to patients with stenosis in the 40-50% range. We doubt this was the intent of the authors.
• The term “persistent symptoms” also requires further definition. Does this mean that symptoms persist despite optimal medical therapy, or does it mean that the symptoms have been present for some period of time (whether or not optimal medical therapy was instituted)? If the latter, how much time? For example, if a patient with intermittent chest pain for a month undergoes CCTA and is found to have obstructive CAD (i.e., 70% RCA stenosis), should FFR be performed, or should the patient be treated medically per the ISCHEMIA trial protocol?

3. In the indication that addresses FFR for stable CAD, there is no requirement that the stenosis be located in the proximal or middle segment of the coronary artery, as is the case for acute CAD indications. Is this intentional or is it an oversight? We suspect it is an oversight because, in general, distal CAD is not managed with revascularization in which case FFR is not particularly helpful in clinical decision making.
4. Patients who have undergone stress testing are excluded from consideration for FFR. We suggest that the authors consider the following:
   1. If stress testing is inconclusive, CCTA and FFR may have a role in patient management.
   2. If stress testing is performed without imaging, and is positive, and CCTA shows more than one obstructive lesion, FFR may be useful in determination of the flow limiting significance of each lesion thereby guiding PCI decisions.

We thank you for the opportunity to comment on this proposed LCD and we would value the opportunity to discuss it with you in further detail as needed.

The response to most comments have been addressed above. Additionally - Persistent is defined by symptoms that persist despite optimal medical therapy. Optimal medical therapy is defined by ISCHEMIA trial protocol² as unacceptable ischemia-related symptoms despite maximally tolerated medical therapy. This includes at least two anti-anginal drugs from different drug classes added to beta-blocker therapy and titrated to maximally tolerated doses before medical therapy is considered to have failed. This has been added to the policy for clarification.

1. Gulati M, Levy PD, Mukherjee D, et al. 2021 AHA/ACC/ASE/CHEST/SAEM/SCCT/SCMR Guideline for the Evaluation and Diagnosis of Chest Pain: Executive Summary: A Report of the American College of Cardiology/American Heart Association Joint Committee on Clinical Practice Guidelines. Circulation. 2021. 0000000000001030.
2. Maron DJ, Hochman JS, O'Brien SM, et al. International study of comparative health effectiveness with medical and invasive approaches (ISCHEMIA) trial: Rationale and design. Am Heart J. 2018;201:124-135.
3. Vavere AL, Arbab-Zadeh A, Rochitte CE, et al. Coronary artery stenoses: Accuracy of 64 detector row CT angiography in segments with mild, moderate, or severe calcification – A subanalysis of the CORE-64 trial. Radiology. 2011;261:100-108.
4. Collet C, Miyazaki Y, Ryan N, et al. Fractional flow reserve derived from computed tomographic angiography in patients with multivessel CAD. Journal of the American College of Cardiology. 2018;71(24):2756-2769.
5. De Bruyne B, Pijls NH, Kalesan B, et al. Fractional flow reserve–guided PCI versus medical therapy in stable coronary disease. N Engl J Med. 2012;367(11):991-1001.
6. Michail M, Ihdayhid AR, Comella A, et al. Feasibility and validity of computed tomography - Derived fractional flow reserve in patients with severe aortic stenosis: The CAST-FFR study. Circulation: Cardiovascular Interventions. 2021;14(1):e009586.
7. Collet C, Miyazaki Y, Ryan N, et al. Fractional flow reserve derived from computed tomographic angiography in patients with multivessel CAD. J Am Coll Cardiol. 2018;71(24):2756-2769.

Our comments focus on these areas within the LCD and reflect the 2021 AHA/ACC/ASE/CHEST/SAEM/SCCT/SCMR guidelines.

1. Inclusion of intermediate risk patients with stable chest pain, with previously unknown and known CAD, in patients with 40-90% stenosis in the proximal or middle coronary artery on CCTA
2. Inclusion of intermediate risk patients with known CAD, presenting with stable and unstable chest pain
3. Clarification in the coverage guidance to include studies with <90% stenosis co-existing in conjunction with >90% stenosis to allow hemodynamic assessment of the relevant vessel(s)
4. Proposed changes to CCTA disease characteristics that are reasonable and necessary for performance of FFRCT
5. Restrictions more appropriate for coronary CT angiography (CCTA) than Non-Invasive Fractional Flow Reserve (FFRCT)
6. Language prohibiting payment of both FFR CT and stress testing (unless CCTA was not of adequate quality for FFRCT and alternative study is needed)

SCCT provides the following comments on proposed changes to CCTA disease characteristics that are reasonable and necessary for performance of FFRCT:

DL38278 proposes that FFRCT may be considered reasonable and necessary for patients with the following disease characteristics found by CCTA:

Intermediate-risk patients with acute chest pain and no known coronary artery disease, with coronary artery stenosis of 40-90% in proximal or middle coronary artery on CCTA.

OR

Intermediate risk with acute chest pain and known coronary artery stenosis of 40-90% in a proximal or middle segment on CCTA

OR

Stable nonobstructive coronary artery disease with persistent symptoms requiring further test, and 40-90% stenosis on CCTA

AND

Not in conjunction with stress testing (unless FFRCT was not high quality and alternative study needed)

CCTA is effective for diagnosing CAD and cardiovascular risk stratification in stable patients with intermediate-high risk who warrant invasive angiography (Chang 2018, Gulati 2021). In stable patients who are later found to have intermediate-high risk findings on CCTA with coronary artery stenosis of 40- 90% in proximal or middle coronary artery on CCTA, FFRCT has demonstrated ability to safely defer invasive coronary angiography with no major adverse cardiac events while diagnosing vessel specific ischemia.

The SCCT recommends language consistent with the 2021 AHA/ACC/ASE/CHEST/SAEM/SCCT/SCMR guidelines. We recommend inclusion of patients with stable chest pain and stenosis of 40% to 90% in a proximal or middle coronary segment on CCTA in intermediate-high risk patients with no known CAD (Class 2a), as well as patients with known non-obstructive CAD with persistent symptoms despite intensification of medical therapy (Class 2a). In patients with stable chest pain with inconclusive stress testing (stress CMR, Stress PET, Stress SPECT, Stress Echo, or exercise ECG), FFRCT is a Class 2a recommendation for 40-90% stenosis on CCTA.

• Suggestion: recommend adding the following:

“OR Stable patients with no known CAD or with non-occlusive CAD who are at intermediate-high risk and coronary artery stenosis of 40-90% in a proximal or middle segment on CCTA”

DL38278 proposes exclusion of FFRCT coverage in patients with:

1. Prior placement of prosthetic valves
2. Known severe aortic stenosis
3. Prior placement of grafts in coronary bypass surgery
4. Suspicion of acute coronary syndrome (where MI or unstable angina have not been ruled out)
5. Intracoronary metallic stent
6. Status post-heart transplantation
7. Recent MI (30 days or less)
8. Prior pacemaker or defibrillator lead placement
9. Newly diagnosed systolic heart failure, with no prior left heart catherization
10. Left main coronary artery disease with Intermediated Coronary Stenosis (lumen reduction less than or equal to 30%)
11. Non-obstructing stenosis (<50% of all major epicardial vessels) on CTA or catherization in the past twelve months, in the absence of a new symptom complex.
12. If turnaround times may impact prompt clinical care decisions
13. High risk defined by left main stenosis ≥ 50%

The SCCT perceives many of the coverage restrictions within this LCD as more applicable to the performance of CCTA than to FFRCT.

The SCCT proposes removal of items #1, 2, 5, and 8 from the exclusion criteria.

Items #1 and 8: Prior placement of prosthetic valves and pacemaker or defibrillator leads:

Image quality plays a role in the performance of FFRCT. HeartFlow has established criteria for acceptable image quality which can be assessed in the presence of prosthetic valve, pacemaker, or defibrillator leads. The presence of prosthetic valves and pacemakers/defibrillators does not preclude performance of FFRCT. Prosthetic valves and pacemakers/defibrillators only preclude the performance of quality CCTA to the extent that these factors may affect adequate imaging of the lumen of the coronary arteries and prevent the reading physician from interpreting the presence of SIHD. However, these artifacts can be minimized with the use of certain scanning parameters and reconstruction algorithms. HeartFlow can process FFRCT results even for patients who have a prosthetic valve present in the heart if there are no significant artifacts involving coronary arteries. It is the opinion of the SCCT that the restriction on prosthetic valves and pacemakers/defibrillators be removed entirely.

• Suggestion: If CCTA image data quality is unacceptable as determined by HeartFlow, no FFRCT processing takes place, no claim is submitted, and no cost is incurred.

Item #2: Known severe aortic stenosis:

Diagnostic quality images are routinely obtained in patients with severe valvular heart disease, even when vasoactive medications are not administrated prior to image acquisition. Transcatheter aortic valve replacement data demonstrates that coronary anatomy and coronary artery disease severity can be accurately evaluated with CCTA in patients with severe aortic stenosis undergoing cardiac CTA for pre-procedural planning. Lunardi, et al. showed that FFR guidance was associated with favorable outcome in observational study in patients undergoing transcatheter aortic valve implantation. However, the left ventricular hypertrophy induced by aortic stenosis may alter the coronary flow reserve, potentially affecting the results obtained by invasive FFR. Since left ventricular mass is considered in the FFRCT analysis, severe aortic stenosis may not have a bearing on physiologic assessment of coronary stenosis. Recently, CAST-FFR study showed the safety and feasibility of FFRCT use in patients with severe aortic stenosis. The area under the receiver-operating characteristic curve for CT-FFR was 0.83 (0.72-0.93, P<0.0001), which was higher than that of CTA and quantitative coronary angiography (P=0.01 and P<0.001, respectively).

• Suggestion: Aortic stenosis should be removed from the exclusion criteria

Item #5: Intracoronary metallic stents:

HeartFlow can process FFRCT results on patients who have had stents to one vessel. However, they can only interpret the vessels without stent. The IFU for HeartFlow states that there are 3 scenarios where FFRCT cannot be utilized (and therefore would be appropriate to not cover):

1. Stent is present in the left main coronary artery
2. A stenosis > 30% in the left main coronary artery and 1 or more stent in the left system
3. Stents present in 2 or more coronary systems

• Suggestion: intracoronary metallic should be removed from the exclusion criteria

Item #11: Non-obstructing stenosis (<50% of all major epicardial vessels) on CTA or catherization in the past twelve months, in the absence of a new symptom complex:

The SCCT recommends clarification in the language related to item # 11.

• Suggestion: recommend revising item #11 to <40% of all major epicardial vessels to remain consistent with the 2021 ACC/AHA/ASE/CHEST/SAEM/SCCT/SCMR guidelines.

Higher grade stenoses (greater than 90%):

The SCCT recommends coverage for patients with multivessel disease where >90% stenosis coexists with 40-90% stenosis.

DL38278 proposes the following:

If higher grade stenoses (i.e., greater than 90%) are present, this study is not medically necessary, as the patient should proceed to catheterization. Similarly, low-grade stenoses (less than 30%) do not require additional confirmatory data. This should be performed as an alternative to stress testing.

In the case of 3 vessel disease, non-invasive functional Syntax Score (SS) provided by FFRCT reclassified 30% of patients with multivessel CAD from the high and intermediate tertiles to the low risk tertile, where an alternative revascularization strategy (PCI over CABG) could be offered and reducing risk to the patient.

• Suggestion: recommend clarification in the coverage guidance to include studies with <90% stenosis co-existing in conjunction with >90% stenosis to allow hemodynamic assessment of the relevant vessel(s).

Presence of extensive plaque on CCTA:

The SCCT recommends removing the comment suggesting the low likelihood of high quality CCTA in the setting of extensive plaque.

DL38278 proposes the following:

If extensive plaque is present a high quality CCTA is unlikely and stress testing is preferred.

With the development of new technologies in CCTA, presence of extensive plaque or calcification does not necessarily exclude the use of FFRCT. We recommend the use of FFRCT if there is adequate image quality if determined by HeartFlow criteria (described above).

• Suggestion: recommend removing “If extensive plaque is present a high quality CCTA is unlikely and stress testing is preferred.”

Stress testing and FFRCT:

DL38278 proposes the following:

• Not in conjunction with stress testing (unless FFRCT was not high quality and alternative study needed)

The performance of FFRCT is based on the image quality of the CCTA. If the CCTA image quality is not adequate for FFRCT analysis as determined by HeartFlow image quality criteria, then an alternative study such as stress testing would be a reasonable option. SCCT suggests the following language:

• Not in conjunction with stress testing (unless CCTA was not of adequate quality for FFRCT and alternative study needed)

SCCT believes that the above modifications to the current proposed LCD would strike an appropriate balance between providing appropriate access to FFRCT to clinicians while avoiding test layering with an ultimate goal of better precision in selecting patients appropriate for medical therapy versus revascularization. Ultimately, FFRCT has been demonstrated in clinical trials to allow for safe deferral of invasive angiography in patients with coronary stenoses of uncertain significance and save healthcare dollars in the process. Additionally, coverage for FFRCT should not supplant invasive FFR entirely as invasive FFR still remains a valuable and proven gold standard for per-vessel ischemia assessment with robust outcomes data in prospective trials.

Thank you again for the opportunity to comment on the Palmetto GBA proposed LCD on FFRCT.

References were provided for review.

Thank you for your comments. The comment regarding higher grade stenosis was not supported by evidence so no changes in policy are currently planned. All other issues addressed have been outlined in previous responses above.