🔒
Internal Research

This document is for Remora Medical team members only. Enter your passphrase to continue.

Incorrect passphrase. Please try again.

Nellix EVAS: Learning from a Recalled Device

A structured analysis of why the Endologix Nellix EndoVascular Aneurysm Sealing (EVAS) system was recalled, and how Remora's design avoids every failure mode identified in published post-market data and regulatory investigations.

Confidential · Internal Only Remora Medical Inc.
Research Document
2026

3D Flow Comparison ↗ 3D Mechanism Explorer ↗ Investor Deck ↗ ← Back to main site

What Was the Nellix EVAS System?

The Nellix EndoVascular Aneurysm Sealing (EVAS) System, manufactured by Endologix, was designed to treat infrarenal abdominal aortic aneurysms (AAA) by filling the entire aneurysm sac with a biocompatible polymer encased in dual endobags. Two stent grafts ran through the center of these polymer-filled bags from the aorta into each iliac artery. The concept was elegant: instead of simply rerouting blood through a fabric conduit while the sac remained pressurised, the Nellix device would fill and seal the sac entirely.

The device received CE Mark approval in Europe in Q1 2013 and entered commercial use across European vascular centers shortly after. It never received FDA pre-market approval in the United States. By 2019, Endologix had voluntarily recalled all Nellix inventory, and the CE Mark was suspended by its Notified Body, GMED. Production permanently ended in May 2022.

⚠️

Bottom line: A 43.5% device failure rate at 6-year median follow-up. Freedom from device failure at 6 years was only 32.3%. Open surgical conversion was required in up to 31% of patients at some centers. Thousands of patients worldwide now require lifelong enhanced surveillance and many require explantation surgery.

43.5%
Device failure rate at 6-year follow-up
32.3%
Freedom from device failure at 6 years
~31%
Open surgical conversion rate (some centers)
2019
Voluntary recall and CE Mark suspension
Figure 1 — How the Nellix System Worked (Schematic)
INFRARENAL AORTA Polymer-filled endobags Balloon-expandable dual stent grafts Caudal Migration ↓ "Shelf force" from blood flow

The Nellix system used polymer-filled endobags to seal the aneurysm sac. The proximal stent geometry created a perpendicular "shelf" vulnerable to sustained hemodynamic loading, leading to progressive caudal migration and seal failure.


Timeline of Events

The Nellix story is particularly instructive because short-term results looked strong — it was the 2+ year window where failures began to cascade. This is a critical lesson for Remora's own clinical development timeline.

2013

CE Mark Approved — European Commercial Launch

Nellix receives CE Mark approval and enters commercial use across European vascular centers. Early 30-day and 1-year results appear promising. The concept is praised as a potential breakthrough in AAA treatment.

2015 – 2016

U.S. IDE Trial Begins (EVAS FORWARD)

Endologix initiates the EVAS FORWARD pivotal trial in the U.S., Canada, and Europe. 1-year results released in 2016 show high technical success and low short-term adverse events, generating significant clinical enthusiasm.

2017

Mid-Term Failures Begin to Emerge

2- and 3-year European follow-up data reveals concerning rates of caudal stent migration and type Ia endoleaks. The FDA requires an advisory committee panel review and additional long-term data before any U.S. approval is granted. Endologix obtains an EVAS2 IDE to study a Gen2 version.

2019

Voluntary Recall — CE Mark Suspended

Endologix initiates a voluntary global recall of all Nellix inventory. The CE Mark Notified Body (GMED) suspends the CE Mark. UK MHRA issues a medical device alert requiring enhanced lifetime surveillance for all implanted patients. The recall is attributed to unacceptably high rates of mid- and long-term device failure.

2022

Production Permanently Ended

Endologix ceases all Nellix production in May 2022. Thousands of patients worldwide remain implanted and require lifelong enhanced surveillance under ESVS guidelines. Many surgically fit patients are now recommended for early elective explantation — a high-morbidity intervention.


Why It Failed: Three Root Causes

Post-market data, explant analysis, and regulatory investigations converged on three specific mechanical and material failure modes. Importantly, all three failures stemmed from the fundamental design premise of the device — not from manufacturing defects or off-label use.

⏱

Key pattern: Nellix failures typically emerged more than 2 years after implantation — well beyond the initial clinical trial follow-up windows that had shown positive results. This highlights the danger of drawing regulatory and commercial conclusions from short-term follow-up data alone.

Root Cause 1: Polymer Degradation and Volume Loss

The biostable polymer filling the endobags underwent progressive dehydration in vivo. Explant analysis showed measurable weight and volume loss over time. As the polymer shrank, it lost conformational contact with the aneurysm wall — the primary sealing surface. This single material failure removed the entire fixation strategy, since the device had no secondary anchoring mechanism.

Root Cause 2: Caudal Migration from "Shelf Force"

The dual stent geometry created a proximal "shelf" that was oriented perpendicular to systolic aortic blood flow. Pulsatile hemodynamic pressure repeatedly loaded this shelf with downward (caudal) force — approximately 60–80 times per minute, every day. Over time, these cumulative forces caused the stent constructs to bend, foreshorten, and migrate distally. Migration of even a few millimeters was sufficient to break the proximal aortic neck seal and initiate a type Ia endoleak.

Root Cause 3: Endobag Separation

Documented separation of the endobags from the stent frames and from each other removed the structural cohesion the device relied upon. Once bag separation occurred, there was no redundant mechanism to maintain wall contact. Combined with polymer degradation, this led rapidly to sac repressurisation and aneurysm expansion.

Figure 2 — Three Failure Modes of the Nellix EVAS System
Polymer Degradation Full polymer → Shrunk polymer Gap = endoleak 2–6 year dehydration in vivo Caudal Migration → flow ↓ migrates Pulsatile shelf loading every heartbeat Endobag Separation Separation No backup anchoring mechanism

All three failure modes operate independently and also compound each other — making device failure progressive and difficult to halt once initiated.


Nellix vs. Remora: Side-by-Side Comparison

This comparison should be internalized by every team member. Remora's design fundamentally avoids every failure mode that caused Nellix to fail. These distinctions are our clearest arguments to investors, surgeons, and regulators.

Figure 3 — Mechanism Comparison: Nellix Passive Sealing vs. Remora Active Remodeling
Nellix EVAS (Failed) Remora (Our Approach) Sac still pressurised Passive fill. Sac unchanged. Polymer degrades. Sac reduced ≥10% Active plication. Sac physically reduced. Suction anchored.

Left: Nellix fills the sac with polymer but does not change its geometry. When the polymer degrades, the sac re-pressurises. Right: Remora mechanically cinches the sac smaller, changing its geometry permanently and removing the pressurisation driving force — an active solution, not a passive one.

Parameter Nellix EVAS (Recalled) Remora
Fixation mechanism Passive polymer-filled endobag sealing. Relies entirely on bag volume to maintain wall contact. Biomimetic suction adhesion. Distributes fixation force evenly across full vessel circumference. No hooks, no barbs.
Sac treatment Fills sac with polymer. Sac remains pressurised if polymer degrades or seal is lost. Active coaxial plication physically cinches and reduces sac diameter ≥10%. Eliminates pressurisation dead space.
Material dependency Entire sealing strategy depends on long-term polymer structural integrity in vivo. Dehydration leads to volume loss and seal failure. Nitinol scaffold with well-established cyclic fatigue resistance. No polymer sealing dependency.
Migration risk High. Perpendicular stent geometry creates a "shelf" that amplifies caudal hemodynamic loading. Documented migration in majority of long-term failures. Low. Suction adhesion distributes force tangentially against vessel wall. No perpendicular flow-interception surface.
Endoleak mechanism Type Ia endoleak from proximal seal loss due to migration and polymer shrinkage. Progressive and difficult to halt once initiated. Active plication eliminates sac pressurisation. Suction seal design targets zero-dead-space aneurysm exclusion.
Revisability Very difficult to explant. Polymer-filled bags adhere to aneurysm wall. High surgical morbidity for revision. Catheter-based retrieval. No adherent polymer mass. Revision remains a percutaneous procedure.
FDA regulatory path Never received FDA PMA. Failed to satisfy advisory panel requirements for long-term durability data. De Novo classification (no direct predicate). Targeting Breakthrough Device designation for expedited review.
Long-term outlook ESVS guidelines: lifelong enhanced surveillance for all implanted patients. Early explantation recommended for fit patients showing any failure signs. Active remodeling to reduced sac diameter eliminates the key driver of long-term failure and surveillance burden.

What Remora Learns from This

The Nellix failure is not just a cautionary tale — it is a detailed engineering roadmap of what to avoid. Each failure mode maps directly to a specific decision Remora has already made.

01

Do not rely on a single material as the sole fixation strategy

When Nellix's polymer degraded, the entire sealing system failed with it. There was no secondary mechanism. This is a single-point-of-failure architecture at the most critical level.

Remora: Suction adhesion is a pressure-differential mechanical phenomenon, not a material property. Even if localized suction is disrupted, active plication provides a second layer of sac control.
02

Avoid geometry that amplifies hemodynamic loading

The Nellix shelf was an inadvertent mechanical amplifier of systolic flow forces — only revealed through multi-year follow-up under real pulsatile loading conditions.

Remora: FSI simulation is used specifically to validate hemodynamic load distribution before bench testing. The suction disc geometry is designed to redirect rather than intercept axial flow forces.
03

Design for retrievability from the start

The polymer-endobag design made explantation extremely technically demanding, exposing patients to high morbidity when revision became necessary — which it did, for thousands of patients.

Remora: No polymer mass, no adherent material. The device architecture maintains catheter-based retrieval compatibility throughout the design process.
04

Short-term trial success is not sufficient

Nellix's 30-day and 1-year results were strong enough to support commercial launch in Europe and IDE trial approval in the U.S. The catastrophic failure mode only appeared beyond the 2-year mark.

Remora: Planning bench durability testing at 10M+ loading cycles (approx. 10+ years in vivo) before pivotal clinical trials. We build the long-term durability case before we need it.
05

Active treatment outperforms passive sealing over time

Nellix's passive polymer fill strategy left no active mechanism to compensate for tissue remodeling, neck dilation, or polymer degradation. The device could not adapt.

Remora: Active plication changes the aneurysm's geometry permanently. Even partial reduction in sac diameter reduces wall stress — the device works with the body's remodeling, not against it.
06

Proactively address historical precedent with investors and regulators

Endologix was unable to demonstrate long-term durability data when the FDA required it. By the time long-term data existed, it was damning. The commercial and regulatory window closed.

Remora: The Nellix case is incorporated proactively into our De Novo regulatory strategy. We explain how our mechanism avoids each known failure mode before we are asked.

Investor Q&A Preparation

Any sophisticated medtech investor who performs due diligence on AAA devices will find the Nellix story. Being prepared to address it proactively signals technical depth and regulatory maturity. These questions should be practiced before any investor conversation.

Why won't Remora fail like Nellix?

Nellix failed for three specific mechanical reasons: polymer degradation, shelf-force-driven caudal migration, and endobag separation. Remora has no polymer. Our fixation is pressure-differential suction adhesion — a mechanical force that does not degrade. Our geometry avoids the perpendicular shelf. And our active plication creates a second independent layer of sac control. These are fundamentally different engineering strategies. We designed Remora with the Nellix post-market data in front of us.

Isn't EVAR a crowded, failed space? Several "next-gen" devices have struggled.

The reason next-generation EVAR devices have struggled is that they have remained passive conduits — they reroute blood but do not treat the aneurysm. Nellix tried to seal passively with polymer. Conventional EVAR tries to exclude passively with fabric. Remora is the first device that physically reduces aneurysm geometry. That is a categorically different clinical claim, supported by a categorically different mechanism of action.

How will you prove long-term durability given that short-term data was Nellix's weakness?

We are planning bench durability testing at 10M+ loading cycles — equivalent to approximately 10+ years of in vivo cardiac cycling — before our pivotal clinical trials begin. We also plan FSI simulation validation of hemodynamic load distribution under realistic pulsatile conditions before any animal studies. The FDA will see long-term mechanical durability data supporting our De Novo submission, not 30-day clinical data alone.

💡

Key framing: We do not shy away from Nellix. We lead with it. The fact that we understand exactly why Nellix failed — and have designed around every one of those failure modes — is evidence of technical rigor and regulatory preparedness, not a risk to minimize.


Sources