David F. Cuevas Cueto, MD · Updated September 27, 2026 · Sources: SCCT 2026 and CAD-RADS 2.0
CAD-RADS 2.0 summarizes the findings of coronary CT angiography in a single per-patient category, followed by plaque burden and modifiers. This section explains how each category is assigned, the order in which the code is written, and how CT-FFR or CT perfusion results are incorporated. The management summarized here applies to stable chest pain.
Key points
The CAD-RADS category is assigned once per patient and is defined by the clinically most relevant stenosis, which is almost always the highest-grade one.
CAD-RADS 0 requires no plaque and no stenosis, so plaque without stenosis is already enough for CAD-RADS 1.
Left main stenosis of 50% or more is 4B even if no vessel reaches 70%, as is three-vessel disease with 70% or more.
If a segment is nonevaluable and an evaluable segment reaches 50%, the stenosis is graded and /N is added. If none reaches 50%, the category is N.
The code is written in the order category, P, N, HRP, I, S, G and E, separated by slashes, and P is not assigned in CAD-RADS 0.
With CT-FFR measured 1 to 2 cm distal to the lesion, 0.75 or less is I+, 0.76 to 0.80 is I±, and above 0.80 is I−.
Three segments with plaque give P2. The moderate mid LAD stenosis defines category 3, and the mild proximal RCA and minimal proximal LCx stenoses remain in the findings.
Type or paste a string as it appears in a report to see what each piece means, with a link to its term.
Examples:
The reader explains CAD-RADS 2.0 nomenclature; it does not interpret a specific study or recommend management.
Table 1. CAD-RADS 2.0 categories and suggested management in the patient with stable chest pain. These are consensus suggestions, to be interpreted together with the impression and the clinical context.
Category
Maximal stenosis
Interpretation
Suggested management (stable chest pain)
CAD-RADS 0
0% (no plaque or stenosis)
No coronary artery disease
No further testing. Consider nonatherosclerotic causes.
CAD-RADS 1
1–24%, or plaque without stenosis
Minimal nonobstructive disease
No further testing. Preventive therapy according to P, intensive with P3–P4.
CAD-RADS 2
25–49%
Mild nonobstructive disease
No further testing. Preventive therapy according to P, intensive with P3–P4.
Consider invasive angiography, functional or viability assessment. Intensive prevention.
CAD-RADS N
Nondiagnostic study
Obstructive disease cannot be excluded
Additional or alternative evaluation may be needed.
Source: CAD-RADS 2.0, Table 4 (PMC full text, unpaginated); stenosis ranges also in SCCT 2026, p. 7. In the CAD-RADS 5 row, Table 4 reads “total coronary occlusion or sub-total occlusion”; the glossary keeps “total occlusion” because the category requires 100%.
Also seen as:Coronary Artery Disease-Reporting and Data System, CAD-RADS classification, CAD-RADS score(CAD-RADS 2.0 uses it for the stenosis category, Section 3.3), CADRADS
CAD-RADS 2.0 assigns a single category to each patient. It is defined by the clinically most relevant stenosis, usually the highest-grade one, and only vessels 1.5 mm or more are graded.
How to write it
Coronary atherosclerosis with moderate plaque burden (P2; SIS 3). Maximal stenosis in the mid LAD: partially calcified plaque with moderate stenosis (50–69%). CAD-RADS 3/P2.
This example assumes: The left main has no stenosis of 50% or more; if it does, the category is 4B even though the maximal stenosis is moderate.
Do not confuse: Category versus plaque burden: the category follows the clinically most relevant stenosis, and P summarizes all plaque; P3 does not raise a mild stenosis to CAD-RADS 3.
Source: CAD-RADS 2.0, Table 4, footnote · “The CAD-RADS classification should be applied on a per-patient basis for the clinically most relevant (usually highest-grade) stenosis.”
Details
In practice. Each segment is graded first and its stenosis is recorded in the findings. The most severe stenosis defines the category, unless left main stenosis or three-vessel disease raises it to 4B. Lesser lesions are described in the findings.
Pitfall. Assigning a category to each vessel, such as “LAD CAD-RADS 3 and RCA CAD-RADS 2”. The patient receives one category, and each lesion is described in the findings.
Other sources. SCOT-HEART classified each patient into three groups: normal coronary arteries, nonobstructive disease, and obstructive disease, defined as stenosis greater than 70%, or greater than 50% in the left main. CAD-RADS 2.0 assigns one of eight categories. (Williams 2019, SCOT-HEART, Methods)
In CCTA Map
The map takes the highest grade among the marked segments and shows a single category with its reason, for example “Moderate stenosis (50–69%)”. The template describes each vessel in its own paragraph. With a patent graft (graft stenosis below 50%), only the native segments proximal to the anastomosis are excluded from the category, labeled “(bypassed by LIMA-LAD)”; the anastomosis segment is read lesion by lesion and the distal runoff is graded. A severe LAD stenosis distal to the anastomosis of a patent LIMA-LAD, with no other lesions, therefore gives CAD-RADS 4A/P1/G. A lesion marked with a single tap on the anastomosis segment has no position and is read as bypassed.
Literal output of the map's template
Plaque burden: moderate (P2; SIS 3).
Left anterior descending (LAD): mid segment: partially calcified plaque resulting in moderate stenosis (50-69%); the remainder is unremarkable. Branches D1, D2: no plaque.
Circumflex (LCx): proximal segment: calcified plaque causing minimal stenosis (1-24%); the rest of the vessel is unremarkable. Branches OM1, OM2, L-PLB: no plaque.
Right coronary (RCA): proximal segment: calcified plaque causing mild stenosis (25-49%); the rest of the vessel is unremarkable. Branches AM, R-PDA, R-PLB: no plaque.
Also seen as:absence of CAD(CAD-RADS 2.0 interpretation of CAD-RADS 0), minimal nonobstructive CAD(CAD-RADS 2.0 interpretation of CAD-RADS 1)
A study is CAD-RADS 0 only if there is no plaque and no stenosis. Any plaque raises it to at least CAD-RADS 1, even if it does not narrow the lumen, such as plaque with positive remodeling and no stenosis.
How to write it
Coronary atherosclerosis with mild plaque burden (P1; SIS 1). Proximal LAD: non-calcified plaque with positive remodeling, without visible luminal narrowing. CAD-RADS 1/P1.
This example assumes: Positive remodeling is the plaque's only high-risk feature; if it has another, such as low attenuation (<30 HU), /HRP is added: CAD-RADS 1/P1/HRP.
Do not confuse: “Without visible luminal narrowing” versus “without significant stenosis”: the first corresponds to 0% on the CAD-RADS 2.0 scale; the second is not a grade on that scale and does not say whether a minimal or mild stenosis is present.
Source: CAD-RADS 2.0, Table 4, footnote b · “CAD-RADS 1 - This category should also include the presence of plaque with positive remodeling and no evidence of stenosis.”
Details
In practice. Before calling a study normal, review each segment on cross-sectional views and multiplanar reconstructions for small non-calcified plaque. SCCT 2026 reserves the impression of a normal study for the absence of plaque and stenosis. A nonatherosclerotic narrowing, such as that of an anomalous artery with an interarterial course, is not CAD-RADS 0 either: it is graded like any other stenosis and E is added; if there is no plaque, P is omitted, as in CAD-RADS 3/E.
Pitfall. Calling a study CAD-RADS 0 when the Agatston score is 0 but non-calcified plaque is present. That plaque is enough for CAD-RADS 1, and plaque burden is assigned.
In CCTA Map
A plaque marked at 0% raises the category to 1. In category 0 the map does not show P, HRP or I. The template writes a 0% plaque with its type, as in “non-calcified plaque without significant stenosis”. A narrowing without plaque, such as that of an anomalous artery with an interarterial course, is marked with the “Non-atherosclerotic” type: it is graded, adds E and carries no P, so a moderate stenosis without plaque gives CAD-RADS 3/E.
Literal output of the map's template
Left anterior descending (LAD): proximal segment: non-calcified plaque without significant stenosis with positive remodeling as an isolated feature; the remainder is unremarkable. Branches D1, D2: no plaque.
Also seen as:mild nonobstructive CAD(CAD-RADS 2.0 interpretation of CAD-RADS 2)
In CAD-RADS 2 the maximal stenosis is mild (25–49%) and disease is nonobstructive. In CAD-RADS 3 it is moderate (50–69%), and CAD-RADS 2.0 already suggests considering functional testing.
How to write it
Coronary atherosclerosis with mild plaque burden (P1; SIS 2). Maximal stenosis in the proximal LCx: non-calcified plaque with moderate stenosis (50–69%), visually estimated at approximately 55%. CAD-RADS 3/P1.
Source: CAD-RADS 2.0, Section 3.1.1 · “In the presence of CAD-RADS 3, which reflects moderate stenosis (50–69%), there is an option to consider the use of CT-FFR, CTP, or stress testing”
Details
In practice. Stenosis is estimated in at least two perpendicular planes, comparing the minimal luminal diameter with a normal reference segment. When the grade is close to 50%, review the lesion on cross-sectional views before choosing between 2 and 3.
Pitfall. Reading CAD-RADS 3 as a direct indication for invasive angiography. In the stable patient, CAD-RADS 2.0 suggests considering functional testing.
Other sources. In invasive angiography, the 2021 ACC/AHA/SCAI revascularization guideline calls a 40 to 69% stenosis intermediate, which generally warrants physiologic assessment. In CAD-RADS 2.0, a 40 to 49% stenosis is CAD-RADS 2. (ACC/AHA/SCAI 2021 revascularization guideline, Section 4.1)
In CCTA Map
With the 2 and 3 buttons under “Severity”, the template writes the term and its range, such as “moderate stenosis (50-69%)”; the value each button stores only represents the range, is not a measurement and is not printed. With the “Stenosis grade” slider, it also writes the figure as an estimate, as in “moderate stenosis (50-69%; approximately 55%)”.
Literal output of the map's template
Circumflex (LCx): proximal segment: non-calcified plaque causing moderate stenosis (50-69%; approximately 55%); the rest of the vessel is unremarkable. Branches OM1, OM2, L-PLB: no plaque.
Right coronary (RCA): mid segment: partially calcified plaque producing mild stenosis (25-49%); the rest of the vessel is unremarkable. Branches AM, R-PDA, R-PLB: no plaque.
Also seen as:CAD-RADS 4(category that CAD-RADS 2.0 subdivides into 4A and 4B)
CAD-RADS 4A is severe stenosis (70–99%) in one or two vessels. CAD-RADS 4B is left main stenosis of 50% or more, or three-vessel disease with 70% or more.
How to write it
Coronary atherosclerosis with mild plaque burden (P1; SIS 1). Left main: partially calcified plaque with moderate stenosis (50–69%). CAD-RADS 4B/P1.
Do not confuse: Moderate stenosis (50–69%) in three vessels is CAD-RADS 3, not 4B: in CAD-RADS 2.0, three-vessel disease requires 70% or more in each vessel, and only the left main counts from 50%.
Source: CAD-RADS 2.0, Table 4, CAD-RADS 4 row · “B - Left main ≥50% or 3-vessel obstructive (≥70%) disease”
Details
In practice. When a severe stenosis is found, review the left main and all three territories before finalizing the category. In the stable patient, 4A prompts consideration of invasive angiography or functional assessment, and with 4B invasive angiography is recommended.
Pitfall. Coding a moderate left main stenosis as CAD-RADS 3 because no vessel reaches 70%. Left main stenosis of 50% or more is already 4B.
Other sources. In the SYNTAX score, any lesion of 50% or more in vessels 1.5 mm or larger is significant. For category 4B due to three-vessel disease, CAD-RADS 2.0 requires 70% or more in each vessel. (Sianos 2005, SYNTAX, p. 220)
In CCTA Map
The reason appears as “Left main stenosis ≥50%” or “Three-vessel disease (≥70%)”. For three-vessel disease, the map counts the LAD, LCx and RCA, including the posterior descending and posterolateral branches, and excludes diagonals, marginal branches and the ramus intermedius; with right dominance the L-PLB counts as a side branch, like a marginal branch. Both 4B criteria count only atherosclerotic stenosis: a “Non-atherosclerotic” left main narrowing of 50–99% gives 3 or 4A with E, not 4B.
Literal output of the map's template
Left main (LM): partially calcified plaque resulting in moderate stenosis (50-69%).
Impression High-risk obstructive CAD due to left main stenosis (≥50%). CAD-RADS 4B/P1.
CAD-RADS 5 indicates at least one total occlusion (100%). The occlusion may be acute or chronic, and the category does not distinguish between them, because chronicity is inferred from other findings.
How to write it
Coronary atherosclerosis with mild plaque burden (P1; SIS 2). Mid RCA: total occlusion by non-calcified plaque, with distal reconstitution via collaterals from the LAD, a finding that suggests chronic occlusion. CAD-RADS 5/P1.
Do not confuse: Distal opacification versus a patent vessel: because CCTA uses intravenous contrast and delayed acquisition, the vessel distal to a chronic occlusion may opacify; in SCCT 2026, total occlusion is defined by complete nonopacification of the segment.
Source: CAD-RADS 2.0, Section 3.1.1 · “It may be acute or chronic, and, in the context of chronic occlusion, factors such as lesion length, calcification particularly at the proximal aspect, tortuosity and degree of collateralization”
Details
In practice. If the occlusion appears chronic, CAD-RADS 2.0 considers the length of the occluded segment, calcification at its proximal end, tortuosity and degree of collateralization useful for management. In the stable patient, invasive angiography, functional assessment or viability assessment is considered.
Pitfall. Calling every total occlusion chronic. It is described as chronic only when supporting findings are present, such as collateral circulation.
Other sources. In interventional cardiology, chronic total occlusion (CTO) is defined as 100% stenosis with TIMI 0 flow for more than three months. CAD-RADS 5 is not equivalent to CTO: it includes any total occlusion, acute or chronic. (EuroCTO 2019, Galassi, p. 199)
In CCTA Map
The “5” button in the vessel editor, or the “Stenosis grade” slider at 100%, gives category 5 with the reason “Total occlusion (100%)”. Collaterals turned on in the “Collateral circulation” panel of the “Anatomy ▾” menu appear in the template findings only if they reconstitute the occluded territory.
Literal output of the map's template
Left anterior descending (LAD): proximal segment: calcified plaque causing mild stenosis (25-49%); the remainder is unremarkable. Branches D1, D2: no plaque.
Right coronary (RCA): mid segment: non-calcified plaque, total occlusion (100%); the rest of the vessel is unremarkable. Branches AM, R-PDA, R-PLB: no plaque.
Distal reconstitution of the occluded RCA via collaterals (from the LAD · apical PDA–distal LAD anastomosis), consistent with chronic total occlusion.
Impression Total occlusion (100%) of at least one vessel. CAD-RADS 5/P1.
Obstructive CAD (≥50%) in RCA.
The occlusion demonstrates collateral recruitment to the distal territory, supporting chronicity (chronic total occlusion) and relevant to planning the revascularization strategy and access route (e.g., CTO-PCI). Confirm myocardial viability with dedicated imaging, as collaterals alone do not reliably predict it.
Also seen as:CAD-RADS N, N modifier, nondiagnostic modifier, nondiagnostic study(title of modifier N in CAD-RADS 2.0, Section 3.3.1)
N flags segments 1.5 mm or more that cannot be interpreted with confidence. If an evaluable segment reaches 50%, it is graded and /N is added. If none does, CAD-RADS N replaces the number.
How to write it
Coronary atherosclerosis with mild plaque burden (P1). Proximal LAD: calcified plaque with mild stenosis (25–49%). The mid RCA is nonevaluable due to motion artifact; therefore, obstructive disease cannot be excluded in this segment. CAD-RADS N/P1.
This example assumes: Overall plaque burden can be assessed reliably; otherwise, the code is CAD-RADS N, without P.
Do not confuse: Nonevaluable segment versus segment without plaque: N means that the segment cannot be interpreted with confidence, not that plaque is absent; visible plaque is described even if its stenosis cannot be graded.
Source: CAD-RADS 2.0, Section 3.3.1 (Modifier N) and Figure 12 · “for a patient with at least one non-interpretable segment and no stenosis (zero), minimal (1–24%), or no more than mild stenosis (25–49%) in interpretable segments, CAD-RADS N should be used”
Details
In practice. Before declaring a segment nonevaluable, review the reconstructed phases. If the artifact persists, name the segment in the report with its cause, such as motion, blooming or metal artifact. Dense calcified plaque does not require N: SCCT 2026 allows the visualized grade to be reported with the caveat that blooming may overestimate it, although some readers consider a segment with extensive calcified plaque non-interpretable. P is added only if overall plaque burden can be assessed reliably; otherwise, the code carries no P.
Pitfall. Coding a study as CAD-RADS N when it has an evaluable moderate stenosis and one nondiagnostic segment. The stenosis is graded and the modifier is added, as in 3/P2/N.
Other sources. In the SCCT CAC-DRS, N does not indicate a nondiagnostic study: combined with a number, as in A2/N3, it indicates how many arteries contain calcium. In CAD-RADS 2.0, N flags segments that cannot be interpreted with confidence. (CAC-DRS, SCCT 2018, Hecht; Obisesan 2021)
In CCTA Map
The “N” button in the editor marks the segment, which is drawn with a double contour, like a motion-blurred artery. Depending on what is marked, the map shows the category, as in “CAD-RADS N/P1”, or the modifier, as in “3/P1/N”. The “nondiagnostic” option of the “Quality” selector in the “Draft report” panel applies the same rule to the whole study even if no segment is marked: if no stenosis reaches 50%, the category is N; if one does, it is graded and /N is added. The reason shows “Overall study quality nondiagnostic” on its own in category N and after the stenosis reason when /N is added. The other “Quality” options do not change the category. Marking N sets the segment’s plaque aside rather than deleting it, and it comes back when the N is removed; meanwhile, the segment leaves the category and the SIS. Portions of the same stroke in other segments are kept and still graded: with a plaque drawn from the proximal to the mid LAD at 85% with two high-risk features, N on the mid LAD gives CAD-RADS 4A/P1/N/HRP. With “Describe visible plaque”, the report names the plaque in the N segment without grading it.
Literal output of the map's template
Study quality: fair.
Left anterior descending (LAD): proximal segment: calcified plaque causing mild stenosis (25-49%); the remainder is unremarkable. Branches D1, D2: no plaque.
Right coronary (RCA): mid segment: nonevaluable due to nondiagnostic image quality; the rest of the vessel is unremarkable. Branches AM, R-PDA, R-PLB: no plaque.
Impression Nondiagnostic study for obstructive CAD. CAD-RADS N/P1.
Nonobstructive CAD in LAD.
Nonevaluable segments: RCA mid.
No numeric stenosis grade is assigned: CCTA cannot guide management in this setting and obstructive CAD cannot be excluded. Consider additional or alternative evaluation (repeat CCTA, functional/stress imaging, or invasive angiography).
CAD-RADS coding order and modifiers P · N · HRP · I · S · G · E#
Also seen as:CAD-RADS modifiers, CAD-RADS 2.0 modifiers, CAD-RADS coding(term used in CAD-RADS 2.0)
The CAD-RADS 2.0 code follows the order stenosis, plaque burden and modifiers, separated by slashes. After P come N, HRP, I, S, G and E, in that order, and only those that are present.
How to write it
Coronary atherosclerosis with severe plaque burden (P3; SIS 5). Maximal stenosis in the proximal LAD: high-risk non-calcified plaque (low attenuation and positive remodeling) with moderate stenosis (50–69%). Patent stent in the proximal RCA without visible restenosis. CAD-RADS 3/P3/HRP/S.
Do not confuse: P versus the modifiers: P is the plaque burden category and, when assigned, follows the stenosis category or N; the modifiers N, HRP, I, S, G and E follow, in that order.
Source: CAD-RADS 2.0, Section 3.3, coding framework; SCCT 2026, “Impression” section, p. 21 · “First: modifier N (non-diagnostic) Second: modifier HRP (high-risk plaque) Third: modifier I (ischemia) Fourth: modifier S (stent) Fifth: modifier G (graft) Sixth: modifier E (exceptions)”
Details
In practice. Write the stenosis category first, then P, which is omitted in CAD-RADS 0 or when there is no plaque. Then add the modifiers present in their fixed order. When N is the category it takes the place of the number, as in N/P2/S.
Pitfall. Writing /S before /HRP, as CAD-RADS 1.0 did with the former V modifier. In CAD-RADS 2.0, HRP is the second modifier, as in 3/P3/HRP/S.
Other sources. CAD-RADS 1.0 had four modifiers, in the order N, S, G and V, and did not include plaque burden. CAD-RADS 2.0 added P, I and E, and replaced V with HRP, which now precedes S. (CAD-RADS 1.0, Table 2, p. 271)
In CCTA Map
The map builds the code in the order P, N, HRP, I, S, G and E, and omits P, HRP and I in category 0. The impression in the template repeats the full code.
Literal output of the map's template
Left anterior descending (LAD): proximal segment: non-calcified plaque causing moderate stenosis (50-69%) with high-risk plaque features (low attenuation and positive remodeling); mid segment: partially calcified plaque resulting in mild stenosis (25-49%); the remainder is unremarkable. Branches D1, D2: no plaque.
Circumflex (LCx): proximal segment: calcified plaque causing minimal stenosis (1-24%); the rest of the vessel is unremarkable. 1st Obtuse marginal (OM1): calcified plaque causing minimal stenosis (1-24%). Branches OM2, L-PLB: no plaque.
Right coronary (RCA): proximal segment: patent stent, no significant restenosis; mid segment: partially calcified plaque producing mild stenosis (25-49%); the rest of the vessel is unremarkable. Branches AM, R-PDA, R-PLB: no plaque.
Figure 1. The code starts with the category, then plaque burden, and ends with the modifiers present, in a fixed order. When N is the category, it takes the place of the number.
Also seen as:CT-FFR(abbreviation recommended by SCCT 2023), CT fractional flow reserve(term recommended by SCCT 2023), CT-derived FFR, FFRCT(also written FFRct; SCCT 2023 recommends CT-FFR)not recommended, FFR-CT(SCCT 2023 recommends CT-FFR)not recommended, ischemia modifier, translesional gradient, delta CT-FFR(drop in CT-FFR across the stenosis; CAD-RADS 2.0 calls it “delta CT-FFR”)
CT-FFR calculates pressure along the entire coronary tree from the CCTA already acquired. It extracts the anatomy with machine learning and simulates rest and hyperemic flow with computational fluid dynamics, with no additional acquisition or medication. Each value is the ratio between the calculated pressure at that point and aortic pressure during hyperemia. A value of 0.72 means that the pressure at that point is 72% of aortic pressure, and it is interpreted as meaning that maximal flow through the artery is 72% of what it would be without the stenosis. Stress CT myocardial perfusion requires a stress acquisition and assesses perfusion of the myocardium; CT-FFR, in contrast, estimates the pressure drop within the artery. The I modifier is assigned from the lesion-specific value, measured 1 to 2 cm distal to the stenosis. A value of 0.75 or less in a vessel large enough for PCI is I+, 0.76 to 0.80 is I±, and above 0.80 is I−.
How to write it
Coronary atherosclerosis with mild plaque burden (P1; SIS 2). Maximal stenosis in the proximal LAD: partially calcified plaque with moderate stenosis (50–69%). Lesion-specific CT-FFR of 0.72, measured 1 to 2 cm distal to the stenosis. CAD-RADS 3/P1/I+.
Do not confuse: Anatomic stenosis versus functional significance: the category grades the lumen and does not change with CT-FFR; a moderate stenosis with a CT-FFR of 0.85 remains CAD-RADS 3, with I−.
Source: CAD-RADS 2.0, Section 3.3.5.1 and Table 6 (Interpretation of CT-FFR) · “If there is presence of abnormal CT-FFR as defined as a lesion-specific value ≤ 0.75 in a vessel large enough for PCI the designation of “I+” (I=Ischemia) should be included.”
Details
In practice. CT-FFR may be used for stenoses of 50 to 90%, particularly in CAD-RADS 3 and 4A, and can be considered in CAD-RADS 2 if the lesion is proximal and 40% or more. With I±, the decision to proceed to invasive angiography depends on lesion location, symptoms and the translesional gradient. That gradient is the drop in CT-FFR from 1–2 cm proximal to 1–2 cm distal to the stenosis, and it is significant if it exceeds 0.12. According to Nørgaard 2019, every CT-FFR reading begins by reviewing the CCTA to match each pressure drop to a specific lesion. Every lesion with an abnormal CT-FFR is reported, even if it did not seem to cause a significant pressure drop on the initial reading. If CT-FFR is abnormal without a concordant anatomical lesion, CAD-RADS 2.0 indicates I− when the reader is confident that it is a CT-FFR false positive, and I± when the result is indeterminate and the interpretation of both findings is uncertain. In both cases, the stenosis category does not change. In multivessel disease, the physiologically significant lesion may not be the one with the most severe stenosis, and this is clarified in the body of the report and in the impression.
Pitfall. Assigning I from the lowest value in the vessel, which reflects total plaque burden. Use the value 1–2 cm distal to the lesion.
Other sources. For invasive FFR, the FAME trial used 0.80 or less as the threshold for stenting. With CT-FFR, CAD-RADS 2.0 reserves I+ for 0.75 or less and considers the 0.76 to 0.80 range borderline (I±). Nørgaard 2019 had already proposed those cutoffs and supported them, among other data, with two findings: CT-FFR gives values on average 0.03 to 0.05 lower than invasive FFR, and its uncertainty relative to invasive FFR is greatest between 0.76 and 0.80. Because CT-FFR is a calculation and not a measurement of flow and pressure, that document recommends describing the result as a low, borderline or high likelihood of hemodynamic significance rather than as ischemia. (Tonino 2009, FAME, abstract; Nørgaard 2019, “FFRCT Threshold” and “Results”)
CT stenosis versus lesion-specific CT-FFR: how I is coded. In each row, the stenosis described is the patient's maximal stenosis.
CT stenosis
CT-FFR 1 to 2 cm distal
Coding
Interpretation
Moderate (50–69%) in the proximal LAD
0.72
CAD-RADS 3 with I+
The abnormal CT-FFR is concordant with the lesion. In a symptomatic patient likely to benefit from revascularization, invasive angiography may be considered.
Moderate (50–69%) in the proximal LAD, with nonobstructive diffuse disease in the mid LAD
0.85 (0.76 in the terminal segment)
CAD-RADS 3 with I−
The lesion does not cause a significant pressure drop. The terminal value is not associated with a specific stenosis and does not define I. It is appropriate to defer invasive angiography and optimize medical therapy.
Minimal (1–24%) in the proximal LAD
0.75 or less
CAD-RADS 1 with I− or I±
CT-FFR is abnormal without a concordant anatomical lesion. It is I− if the reader is confident that it is a CT-FFR false positive, and I± if the result is indeterminate and the interpretation of both findings is uncertain. The category remains 1.
Source: CAD-RADS 2.0, section 3.3.5.1 and Table 6. The second row reproduces the case in Figure 1a of Nørgaard 2019 (CT-FFR of 0.85 at 16 mm from the stenosis and 0.76 in the terminal segment).
In CCTA Map
In “Modifiers P · I ▾”, the “Ischemia (I)” section has one row per territory (LAD, LCx and RCA). When “CT-FFR” is chosen under “Pathway”, the value is entered, with a decimal point or a decimal comma (0.72 or 0,72), and the app classifies it as I+, I± or I−. Until a result is entered, the territory shows “no test”. The code carries I+ if any territory is I+; otherwise I± if any is borderline; and I− only when every territory with a stenosis of 50% or more has a negative test. If any of those territories has no test, the code carries no I. The template records the value with a decimal point, for example “left anterior descending: CT-FFR 0.72”. If a territory is I+ and its maximal stenosis is less than 50%, the app keeps I+ in the code and shows the warning “Per CAD-RADS, consider I- (false positive, if confident) or I± (if uncertain/discrepant).” To change the result, use the “Direct:” buttons.
Literal output of the map's template
Left anterior descending (LAD): proximal segment: partially calcified plaque causing moderate stenosis (50-69%); the remainder is unremarkable. Branches D1, D2: no plaque.
Circumflex (LCx): distal segment: calcified plaque producing mild stenosis (25-49%); the rest of the vessel is unremarkable. Branches OM1, OM2, L-PLB: no plaque.
Inducible ischemia in the left anterior descending territory by CT-FFR (I+). Detail: left anterior descending: CT-FFR 0.72.
Figure 2. The I modifier is assigned from the value measured 1 to 2 cm distal to the stenosis. The lowest value, at the distal end of the vessel, reflects total plaque burden and should not drive referral to invasive angiography.
Also seen as:myocardial CT perfusion(term recommended by SCCT 2023), myocardial CTP(abbreviation recommended by SCCT 2023), stress CT perfusion, stress CTP, CT myocardial perfusion imaging(SCCT 2023 recommends myocardial CT perfusion)not recommended, CT-MPInot recommended, MCTP(SCCT 2023 recommends myocardial CTP)not recommended
On CT perfusion, a defect that is reversible or larger at stress than at rest is I+. No defect, or a fixed defect from infarction, is I−. A borderline result is I±.
How to write it
Coronary atherosclerosis with mild plaque burden (P1; SIS 1). Distal RCA: partially calcified plaque with severe stenosis (70–99%). Stress and rest myocardial CT perfusion: fixed defect in the RCA territory, consistent with prior infarction; no inducible ischemia. CAD-RADS 4A/P1/I−.
Do not confuse: I− versus ischemia not assessed: I− means that CT-FFR or CT perfusion was performed and the result was negative for ischemia; if neither was performed, the code carries no I.
Source: CAD-RADS 2.0, Section 3.3.5.2 and Table 7 (Interpretation of Stress Myocardial CT Perfusion) · “If no ischemia is detected or if there is presence of a prior fixed myocardial infarct, the Modifier “I−” will be added to CAD-RADS.”
Details
In practice. Stress and rest images are compared. According to SCCT 2026, perfusion assessment is improved with multiple cardiac planes (short axis, four-chamber and two-chamber), 5 to 8 mm averaged MPR or MinIP reconstructions and a narrow window, and MIP must be avoided because it masks defects. A true defect persists across several phases of the cardiac cycle, whereas artifacts, such as beam hardening, may vary in presence and location. If the defect is fixed, the prior infarct is documented in the impression.
Pitfall. Assigning I+ for myocardial hypoattenuation on resting CT angiography. The I modifier requires an ischemia test, either stress perfusion or CT-FFR.
In CCTA Map
With the CTP pathway, “Reversible” gives I+, and “Fixed” or “No defect” give I−; a borderline result is marked with I± in the “Direct:” row. With a fixed defect, the map warns that the infarct should be documented, and the impression in the template describes it as consistent with prior infarction and suggests viability assessment. Until a result is entered, the territory shows “no test” and does not contribute I to the code.
Literal output of the map's template
Right coronary (RCA): distal segment: partially calcified plaque causing severe stenosis (70-99%); the rest of the vessel is unremarkable. Branches AM, R-PDA, R-PLB: no plaque.
Impression Severe obstructive CAD. CAD-RADS 4A/P1/I-.
Obstructive CAD (≥50%) in RCA.
Fixed perfusion defect in the right coronary territory on CT perfusion, consistent with prior infarction; no inducible ischemia (I-). Consider viability assessment. Detail: right coronary: fixed perfusion defect (CTP).
CAD-RADS 2.0 pairs each category with suggested management for the patient with stable chest pain. These are consensus suggestions that complement the impression and do not replace clinical judgment.
How to write it
Suggested management per CAD-RADS 2.0 for CAD-RADS 1/P3 in stable chest pain: consider nonatherosclerotic causes of symptoms; aggressive risk factor modification and preventive pharmacotherapy.
Do not confuse: Stable versus acute chest pain: the category is assigned the same way, but the suggested management differs; with CAD-RADS 3, functional assessment is considered in stable chest pain, and in acute chest pain, hospital admission with cardiology consultation is also considered.
Source: CAD-RADS 2.0, Section 1 and Table 4; SCCT 2026, Table 6, p. 21 · “the clinical management suggestions provided by CAD-RADS should not replace clinical judgment, particularly as there are many patient-specific factors that may influence clinical management.”
Details
In practice. For categories 0 to 2 no further cardiac testing is suggested, and in 1 and 2 preventive therapy is adjusted to P. With 3, functional testing is considered; with 4A, invasive angiography or functional testing; and with 4B, invasive angiography is recommended.
Pitfall. Treating CAD-RADS 1 or 2 as settled without checking P. With P3 or P4, CAD-RADS 2.0 suggests intensive risk factor modification and preventive pharmacotherapy.
Other sources. In stable chest pain, the 2021 AHA/ACC guideline considers CT-FFR most useful for 40 to 90% stenoses in proximal or mid segments. CAD-RADS 2.0 suggests functional assessment starting at CAD-RADS 3 (50–69%). (AHA/ACC 2021 chest pain guideline, Section 5.1.3)
In CCTA Map
“Stable” is selected by default under “Setting”, below the “Draft report” heading. With that option, the “Suggested management (CAD-RADS 2.0 consensus)” in the impression follows the stable chest pain recommendations for the category, with added sentences for P, I, HRP and N, and a closing note that leaves the decision to the treating physician. In categories 1 and 2, the base text says that no additional cardiac workup is needed, in 2 “generally”, and adds “Consider nonatherosclerotic causes of symptoms”; P adds a sentence only with P3 or P4. With I+, or with an anomalous artery with an interarterial or intraseptal course or arising from the “PA”, that base text is omitted (with the anomaly, also in category 0), and P1 or P2 gets a sentence of its own. In categories 3 to 5, any P adds a single aggressive prevention sentence, such as “Plaque burden P1 with ≥50% stenosis: aggressive risk factor modification and preventive pharmacotherapy.”, and without P the same sentence is written without the label. In category 3, if the code already carries I from CT-FFR or CTP, the management does not propose functional testing again, and with I+ it proposes considering invasive angiography, especially if frequent symptoms persist after medical therapy. With a stent or a graft, the category 3 functional testing sentence offers CTP or stress testing and states that CT-FFR is not validated in these patients, unless a CT-FFR value has already been entered. With “Acute”, the management is that for acute chest pain. Categories 0 to 2 begin with “No further evaluation of ACS is required”, and 2 adds when to consider admission; with I+, that base text is omitted in 1 and only the admission sentence remains in 2, and with one of those anomalies the entire base text is omitted. In 3 it proposes considering hospital admission with cardiology consultation, and 4A, 4B and 5 begin with admission and do not add the high-risk plaque sentence.
Literal output of the map's template
Plaque burden: severe (P3; SIS 5).
Left main (LM): calcified plaque resulting in minimal stenosis (1-24%).
Left anterior descending (LAD): proximal segment: calcified plaque causing minimal stenosis (1-24%); mid segment: partially calcified plaque resulting in minimal stenosis (1-24%); the remainder is unremarkable. Branches D1, D2: no plaque.
Circumflex (LCx): proximal segment: calcified plaque causing minimal stenosis (1-24%); the rest of the vessel is unremarkable. Branches OM1, OM2, L-PLB: no plaque.
Right coronary (RCA): proximal segment: calcified plaque causing minimal stenosis (1-24%); the rest of the vessel is unremarkable. Branches AM, R-PDA, R-PLB: no plaque.
Short questions to check the key points. The answer appears once you choose an option.
1In a patient with stable chest pain, the left main has a 55% stenosis, the mid LAD 40% and the proximal RCA 30%. There are no other lesions. Which CAD-RADS category is assigned?
Left main stenosis of 50% or more is enough for CAD-RADS 4B, even if no lesion reaches 70%. The LAD and RCA lesions are described in the findings. See the term
2The mid LAD has a moderate stenosis (50–69%) and plaque burden is P2. Motion artifact prevents evaluation of the mid RCA. How is the study coded?
There is a stenosis of 50% or more in an evaluable segment, so it is graded and N becomes a modifier after P. CAD-RADS N is used only when no evaluable segment reaches 50%. See the term
3There is a patent stent in the proximal RCA. In the proximal LAD, a low-attenuation plaque with positive remodeling causes a moderate stenosis, and plaque burden is P3. What is the correct code?
P follows the category, and the modifiers follow in the order N, HRP, I, S, G and E. The legend of CAD-RADS 2.0 Figure 17 reverses S and HRP, but the ordered list takes precedence. See the term
4The mid LAD has a moderate stenosis. CT-FFR measured 1 to 2 cm distal to the lesion is 0.78, and the lowest value, at the apical end of the LAD, is 0.70. Which modifier is added?
The lesion-specific value, measured 1 to 2 cm distal to the stenosis, is used. A value from 0.76 to 0.80 is borderline. The lowest value in the vessel reflects total plaque burden. See the term
5The distal RCA has a severe stenosis (70–99%). CT perfusion shows an inferior defect of the same size at stress and at rest, with wall thinning. How is ischemia coded?
A fixed defect from prior infarction, without inducible ischemia, is coded I−. CAD-RADS 2.0 reserves the I modifier for ischemia and asks that the infarct be documented in the impression. See the term
Rajiah PS, Alkadhi H, Andreini D, Bullock-Palmer RP, Chinnaiyan K, Chow B, et al. Interpretation and reporting of coronary computed tomographic angiography (2026 update): an expert consensus document of the Society of Cardiovascular Computed Tomography (SCCT). J Cardiovasc Comput Tomogr. 2026. doi:10.1016/j.jcct.2026.08.014 Version of record published online in September 2026 under CC BY 4.0, not yet assigned to an issue; cited pages refer to the PDF of this published version.
Cury RC, Leipsic J, Abbara S, Achenbach S, Berman D, Bittencourt M, et al. CAD-RADS 2.0 - 2022 Coronary Artery Disease-Reporting and Data System: an expert consensus document of the SCCT, ACC, ACR and NASCI. J Cardiovasc Comput Tomogr. 2022;16(6):536-57. doi:10.1016/j.jcct.2022.07.002
Koweek L, Achenbach S, Berman DS, Carr JJ, Cury RC, Ghoshhajra B, et al. Standardized medical terminology for cardiac computed tomography 2023 update: an expert consensus document of the SCCT, AAPM, ACR, NASCI and RSNA. J Cardiovasc Comput Tomogr. 2023;17(5):345-54. doi:10.1016/j.jcct.2023.06.002
Narula J, Chandrashekhar Y, Ahmadi A, Abbara S, Berman DS, Blankstein R, et al. SCCT 2021 expert consensus document on coronary computed tomographic angiography: a report of the Society of Cardiovascular Computed Tomography. J Cardiovasc Comput Tomogr. 2021;15(3):192-217. doi:10.1016/j.jcct.2020.11.001
Neves PO, Andrade J, Monção H. Coronary artery calcium score: current status. Radiol Bras. 2017;50(3):182-9.
Tamburino C, Tomasello SD, Capodanno D, Di Salvo ME, Marzà F, Galassi AR. Long-term follow-up after drug eluting stent implantation in left main trifurcations. EuroIntervention. 2009;5(4):432-7. doi:10.4244/eijv5i4a68
Kovacevic M, Burzotta F, Elharty S, Besis G, Aurigemma C, Romagnoli E, et al. Left main trifurcation and its percutaneous treatment: what is known so far? Circ Cardiovasc Interv. 2021;14(3):e009872. doi:10.1161/CIRCINTERVENTIONS.120.009872
Kim C, Park CH, Lee BY, et al. 2024 consensus statement on coronary stenosis and plaque evaluation in CT angiography from the Asian Society of Cardiovascular Imaging-Practical Tutorial (ASCI-PT). Korean J Radiol. 2024;25(4):331-42. doi:10.3348/kjr.2024.0112
Sources for the “Other sources” notes
Sianos G, Morel MA, Kappetein AP, et al. The SYNTAX Score: an angiographic tool grading the complexity of coronary artery disease. EuroIntervention. 2005;1(2):219-27.
Cury RC, Abbara S, Achenbach S, et al. CAD-RADS: Coronary Artery Disease – Reporting and Data System. J Cardiovasc Comput Tomogr. 2016;10(4):269-81. doi:10.1016/j.jcct.2016.04.005
Williams MC, Moss AJ, Dweck M, et al. Coronary artery plaque characteristics associated with adverse outcomes in the SCOT-HEART study. J Am Coll Cardiol. 2019;73(3):291-301. doi:10.1016/j.jacc.2018.10.066
Hecht HS, Blaha MJ, Kazerooni EA, et al. CAC-DRS: Coronary Artery Calcium Data and Reporting System. An expert consensus document of the Society of Cardiovascular Computed Tomography (SCCT). J Cardiovasc Comput Tomogr. 2018;12(3):185-91. doi:10.1016/j.jcct.2018.03.008
Lawton JS, Tamis-Holland JE, Bangalore S, et al. 2021 ACC/AHA/SCAI guideline for coronary artery revascularization: a report of the American College of Cardiology/American Heart Association Joint Committee on Clinical Practice Guidelines. J Am Coll Cardiol. 2022;79(2):e21-e129. doi:10.1016/j.jacc.2021.09.006
Galassi AR, Werner GS, Boukhris M, et al. Percutaneous recanalisation of chronic total occlusions: 2019 consensus document from the EuroCTO Club. EuroIntervention. 2019;15(2):198-208. doi:10.4244/EIJ-D-18-00826
Tonino PA, De Bruyne B, Pijls NH, et al. Fractional flow reserve versus angiography for guiding percutaneous coronary intervention. N Engl J Med. 2009;360(3):213-24. doi:10.1056/NEJMoa0807611
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: A Report of the American College of Cardiology/American Heart Association Joint Committee on Clinical Practice Guidelines. Circulation. 2021;144(22):e368-e454. doi:10.1161/CIR.0000000000001029
This glossary translates and adapts definitions, tables and figures from the SCCT 2026 consensus: Rajiah PS, et al. Interpretation and reporting of coronary computed tomographic angiography (2026 update): an expert consensus document of the Society of Cardiovascular Computed Tomography (SCCT). J Cardiovasc Comput Tomogr. 2026. doi:10.1016/j.jcct.2026.08.014, published under the CC BY 4.0 license. Changes: plain-language rewording and reorganized tables; the diagrams are dynamic schematics of the map and do not reproduce the guideline figures exactly. The SCCT has not reviewed or endorsed this glossary. CC BY 4.0
