Endovascular Aneurysm Repair — CT-based sizing, IFU screening, oversizing calculation, deployment, and endoleak surveillance for stent-graft exclusion of AAA
Endovascular Aneurysm Repair (EVAR) replaces open surgical aortic replacement with a catheter-delivered stent-graft that excludes the aneurysm sac from arterial blood pressure while preserving flow through a fabric-covered metal skeleton. Every subsequent decision in EVAR — which device, what diameter, what length — is downstream of one thing: precise, reproducible measurement of the patient's aortic anatomy on thin-slice CT angiography (CTA) with centerline reconstruction.
An abdominal aortic aneurysm (AAA) is a permanent, focal dilation of the aorta to ≥1.5× its normal diameter, most commonly infrarenal (below the renal arteries). Left untreated, large aneurysms progressively enlarge and can rupture — a catastrophic event with a mortality rate of 65–85% even with emergency surgery. The traditional treatment, open surgical repair, requires laparotomy, aortic cross-clamping, and sewing in a fabric graft — effective but associated with substantial perioperative morbidity, cardiac stress, and a multi-week recovery.
EVAR, first performed by Juan Parodi in 1991, instead delivers a compressed, self-expanding fabric-covered stent-graft through the femoral arteries under fluoroscopic guidance. Once positioned across the aneurysm, the graft is deployed and expands to form a new, sealed conduit for blood flow — a "sleeve within a sleeve" that diverts pressurized flow away from the weakened aneurysmal wall. Because there is no open incision, aortic cross-clamp, or laparotomy, EVAR carries lower 30-day mortality and faster recovery than open repair, which is why it now accounts for roughly 80% of elective AAA repairs in anatomically suitable patients.
But EVAR trades surgical risk for a different kind of risk entirely: the durability of the repair depends completely on whether the deployed graft achieves a stable, leak-proof seal against the native aortic wall — a mechanical and geometric problem that begins and ends with sizing.
Modern EVAR planning uses centerline-of-flow reconstruction software (e.g., 3mensio, TeraRecon, Aquarius) applied to a thin-slice (≤1.5mm), contrast-enhanced CT angiogram covering from the thoracic aorta through the common femoral arteries. Measurements are taken perpendicular to the vessel centerline — not on axial slices — because an oblique or tortuous aorta will produce falsely elongated diameters on simple axial imaging.
Key measurements:
• Proximal neck diameter — the aortic diameter immediately below the lowest renal artery, measured outer-wall-to-outer-wall on a true cross-section. This is the single most important number in EVAR sizing, since it drives graft diameter selection.
• Neck length — the length of relatively parallel, non-aneurysmal aorta between the lowest renal artery and the start of aneurysmal dilation (where diameter increases >10% from the neck diameter). This is the available "real estate" for the proximal seal.
• Neck angulation — the angle between the suprarenal aortic axis and the infrarenal neck axis (and separately, the angle between the neck and the aneurysm sac axis). Severe angulation makes graft apposition to the wall unreliable.
• Maximal aneurysm sac diameter — the largest transverse diameter of the aneurysmal segment; this determines rupture risk and repair indication, though it does not directly drive device sizing.
• Iliac landing zone diameters and lengths — the common iliac artery diameters and available length distally, where the graft's iliac limbs will achieve distal seal.
All of these numbers together define the "sizing runway" that stages 2 and 3 of this simulator evaluate.
Every commercial stent-graft device is approved with an Instructions For Use (IFU) document that defines the anatomical envelope within which the device has been validated — and, more importantly, the envelope within which durable sealing can reasonably be expected. Treating anatomy outside the IFU ("off-label" use) is sometimes done by experienced operators but carries a measurably higher risk of endoleak, migration, and reintervention.
While exact thresholds vary slightly by manufacturer and device generation, the anatomical criteria that gate standard infrarenal EVAR converge on four pillars:
1. Neck length ≥15 mm — enough parallel, non-diseased aorta to seat the full circumference of the proximal sealing stent with margin. Shorter necks ("hostile necks") force operators toward fenestrated or branched devices, or chimney/snorkel techniques, to preserve renal perfusion while extending the seal zone proximally.
2. Neck diameter 18–32 mm — most infrarenal devices are manufactured in a limited range of diameters (typically 20–36mm for the largest components), and the oversizing math (Stage 3) only works within a validated diameter band. A neck below ~18mm risks graft infolding even at modest oversizing; a neck above ~32mm may exceed the largest available device diameter.
3. Angulation <60° — measured as the angle between the suprarenal and infrarenal aortic segments (or infrarenal-to-sac axis). Steep angulation causes the proximal stent to appose asymmetrically against the outer curve of the bend, leaving a crescent-shaped gap on the inner curve — a direct anatomical setup for a Type Ia endoleak.
4. Absence of significant thrombus or calcification — circumferential mural thrombus >2mm thick, or calcification involving more than 50% of the neck circumference, prevents the stent from achieving full metal-to-wall or fabric-to-wall contact, regardless of how well the diameter is matched.
All four criteria interact: a neck that is borderline on diameter but has severe angulation and thrombus is a substantially worse candidate than a neck that is borderline on any single parameter alone. This is why experienced planners consider the whole "hostile neck" gestalt, not a single pass/fail number.
A neck failing two or more IFU criteria simultaneously is associated with a several-fold increase in Type Ia endoleak and reintervention — the anatomical criteria are not independent checkboxes but compounding risk factors.
Once anatomy is confirmed suitable, sizing becomes a deliberate arithmetic exercise: the nominal graft diameter is chosen larger than the measured native vessel — "oversized" — so that the self-expanding stent-graft exerts continuous, gentle radial force (chronic outward force, COF) against the aortic wall, creating a friction-and-pressure seal without folding the fabric or overstressing the wall.
The core sizing formula used in every EVAR planning session is deceptively simple:
Graft diameter = Neck diameter × (1 + oversizing fraction)
For a 24mm neck with 15% oversizing, the selected graft nominal diameter would be 24 × 1.15 ≈ 27.6mm, rounded to the nearest available device size (commonly 28mm).
Why oversize at all? A self-expanding nitinol stent-graft is manufactured at a fixed nominal diameter. If it were sized to exactly match the native vessel, normal minor variations in vessel geometry, respiratory and cardiac pulsatility, and any imprecision in the original CT measurement could allow the stent to sit slightly loose against the wall — with even a small gap providing a channel for blood to leak around the graft and re-pressurize the aneurysm sac (a Type I endoleak).
Oversizing by 10–20% ensures the stent frame is always pressing outward against the wall with continuous radial force, actively maintaining wall apposition through the cardiac cycle and through any minor geometric change over time (necks can dilate 1–2mm/year after implantation).
But oversizing has an upper limit. Beyond roughly 20–25%, the fabric-covered stent frame — now substantially larger than the vessel it sits in — is forced to accordion or "pleat" as it compresses circumferentially to fit the smaller native lumen. These pleats and infoldings can themselves become channels for endoleak, and in severe cases can encroach on the lumen enough to cause partial or complete limb occlusion. This is why device IFUs specify a validated oversizing window (typically 10–20%) rather than "more is always safer."
Rule of thumb used throughout EVAR planning: oversize the proximal aortic component by 10–20% of the neck diameter, and oversize iliac limbs by roughly 10–15% of the landing-zone diameter — below this range risk migration and Type I endoleak, above it risk infolding, kinking, and limb occlusion.
Diameter sizing determines whether the graft seals; length sizing determines whether it reaches. The main body and limb lengths are chosen so that the fabric-covered portion of the device spans continuously from a point just below the lowest renal artery (proximal seal) to a point within the common iliac artery landing zones (distal seal), fully excluding the aneurysm sac along its entire length.
Modular devices (a main bifurcated body plus one or two iliac limb extensions) require deliberate overlap between components — typically at least one full stent ring, often more — to prevent limb separation ("modular disconnection"), a rare but serious late complication. Overlap length is balanced against total device length: too much overlap can push the distal limb landing zone beyond the usable iliac segment, forcing extension into the external iliac artery and sacrifice of the internal iliac (hypogastric) artery, which carries its own risk of pelvic ischemia and buttock claudication.
Deployment is the moment sizing theory meets mechanical reality. Under continuous fluoroscopic guidance, the delivery system is withdrawn to release the constrained stent-graft, which self-expands against the aortic wall. The operator must confirm, in real time, that the proximal and distal sealing zones achieve full circumferential apposition — because any gap identified now is far easier to fix with an intraoperative maneuver than after the sheaths are withdrawn.
Deployment begins with precise fluoroscopic positioning of radiopaque markers on the delivery system relative to the renal artery origins (confirmed by a marker pigtail catheter injection). Once position is verified — typically with the C-arm angled to align the operator's view perpendicular to the neck, un-foreshortening the proximal landing zone — the outer sheath is withdrawn to release the proximal stent.
Most modern infrarenal devices use a partially constrained proximal stent with suprarenal fixation (bare metal struts extending above the renal arteries, anchoring without covering the renal ostia) to improve resistance to downward migration while the covered fabric portion seals below the renals. As the sheath is withdrawn, the nitinol stent frame self-expands within seconds, driven by the shape-memory property of nitinol returning to its pre-set larger diameter.
After full deployment of the main body and limb components, most operators perform balloon molding — inflating a compliant balloon within the proximal and distal seal zones and at module overlap points — to actively press the stent struts against the wall and iron out any residual folds in the fabric, improving apposition beyond what passive self-expansion alone achieves.
Completion imaging at this stage focuses specifically on the seal zones: does the stent-graft contour follow the native aortic wall contour with no visible gap, kink, or "bird-beak" configuration (where the proximal edge of the graft angles away from the wall, common in angulated necks)? Intravascular ultrasound (IVUS) is an increasingly used adjunct that can directly visualize wall apposition circumferentially, independent of the two-dimensional silhouette that fluoroscopy alone provides.
The procedure is not complete until a formal completion angiogram confirms that the aneurysm sac has been fully excluded from arterial pressure. An endoleak — persistent blood flow within the aneurysm sac outside the lumen of the stent-graft — is the signature failure mode of EVAR, and distinguishing a benign, self-resolving leak from one that mandates immediate correction is the final and arguably most consequential judgment of the case.
An endoleak is defined as persistent blood flow within the aneurysm sac but outside the graft lumen, and is classified by its source:
• Type I — seal-zone failure. Blood leaks around the proximal (Ia) or distal (Ib) attachment site because of inadequate apposition — the direct consequence of undersizing, severe angulation, a short neck, or thrombus/calcification preventing wall contact. Type I endoleaks re-pressurize the sac at near-systemic pressure and require immediate intraoperative correction (additional balloon molding, a proximal cuff, or an endo-anchor).
• Type II — retrograde filling from a branch vessel (typically lumbar arteries or the inferior mesenteric artery) that continues to back-bleed into the excluded sac. This is the most common endoleak type and often the least urgent — many resolve spontaneously as the sac thromboses, though persistent Type II leaks with sac growth eventually warrant embolization.
• Type III — component failure: a mechanical separation between modular graft components, or a tear/defect in the fabric itself, allowing full-pressure flow into the sac. Mechanistically similar in urgency to Type I because the sac is exposed to near-systemic pressure.
• Type IV — transient graft-fabric porosity, seen historically with older graft materials as a faint blush during the procedure that resolves spontaneously once normal coagulation parameters return; rare with modern low-porosity fabrics.
• Type V (endotension) — sac pressurization and continued growth without any radiographically identifiable leak source — a poorly understood entity possibly related to pressure transmission through organized thrombus or fabric.
On this simulator, the seal quality driven by your neck diameter and oversizing selections is used as a proxy for Type I risk: a well-apposed, adequately (but not excessively) oversized graft shows a fully excluded sac on completion angiogram, while an under- or over-sized graft shows a contrast jet leaking into the sac at the seal zone.
The clinical rule that governs the end of every EVAR case: Type I and Type III endoleaks must be identified and corrected before leaving the angio suite, because they leave the aneurysm sac exposed to near-systemic pressure — defeating the entire purpose of the repair. Type II leaks, by contrast, are usually managed with surveillance imaging (CT at 1, 6, and 12 months, then annually) rather than immediate re-intervention.
1. Acquire thin-slice CTA with centerline reconstruction from the thoracic aorta through the common femoral arteries. 2. Measure proximal neck diameter, neck length, angulation, maximal sac diameter, and iliac landing zone diameters/lengths perpendicular to the centerline. 3. Screen the neck against device-specific IFU criteria (length, diameter, angulation, thrombus/calcification burden); flag any hostile-neck features. 4. Calculate nominal graft diameter using 10–20% oversizing of the neck diameter; select limb diameters similarly for the iliac landing zones; select component lengths to span the seal zones with adequate modular overlap. 5. Deploy the device under fluoroscopic guidance with markers referenced to the renal arteries; balloon-mold the seal zones and overlap points. 6. Perform a completion angiogram in at least two projections; assess specifically for Type I and Type III endoleak and confirm sac exclusion. 7. If a Type I or III endoleak is identified, correct it immediately (additional molding, proximal cuff, endo-anchors, or relining) before ending the procedure. 8. Establish a long-term CT surveillance schedule (typically 1, 6, and 12 months, then annually) to monitor sac size, device integrity, and any late-appearing endoleak.