The Graveyard Manifesto

The Shortage After the Shortage

During COVID, the device that was the rage was the oxygen concentrator. The world faced a terrifying shortage. Hospitals ran out. Families ran out. I remember it personally, fighting to get a steady supply of oxygen for my own parents.

And then, what happened?

As the saying goes, I have never seen a shortage not followed by a glut. Within a couple of years, oxygen concentrators were everywhere, donated, purchased, stockpiled, shipped to health systems. The shortage was solved.

Except it wasn't, not really. A large share of those concentrators sit unused today. Nobody budgeted for what came after delivery: who would maintain them, who would fix them when a part failed, who would keep them running once the emergency funding and the emergency attention moved on. The device arrived. The plan for keeping it alive never did.

That same pattern shows up everywhere in medtech, just wearing a different costume. A company builds something that works, gets it cleared, and dies anyway. The product usually isn't what killed it. What killed it is that nobody budgeted for everything that has to happen after clearance: who pays for it, who decides to buy it, and how long the company can survive while the rest of the system catches up.

This is the unfortunate graveyard.

The Autopsy

Look under the hood of enough medtech failures, and the same pattern shows up again and again. The cause is rarely the clinical outcome. The product usually works. What derails it comes after, in what's really post-clinical validation, everything that has to go right commercially once the science is already settled.

This piece is that autopsy. We'll go through real examples, but the goal isn't to be critical of any one company. It's a learning opportunity: if more founders and investors can see this pattern early, more companies can plan for it and thrive.

Category Confusion

In 2020, the FDA cleared EndeavorRx, a video game prescribed to treat ADHD in children. It was real, it worked in trials, and it was the first product of its kind. But nobody had a category ready for it. Doctors had to decide whether to prescribe a video game. Insurers had to decide whether to pay for one. Parents had to decide whether to trust it, and children had to actually keep playing it consistently enough for it to work. None of that is a clinical question. All of it had to get solved before the company could survive.

Akili spent years trying to solve it. Eventually the company shifted toward a non-prescription, consumer version and was acquired by Virtual Therapeutics for around $34 million, a fraction of the roughly $1 billion valuation it carried just two years earlier.

Pear Therapeutics ran into the same wall with Somryst, a prescription app for chronic insomnia cleared by the FDA in March 2020. It needed a doctor willing to prescribe software, a patient willing to believe an app could treat insomnia the way a pill might, and a payer willing to reimburse a category that didn't exist yet. Pear built the product. It couldn't build the category fast enough. The company filed for Chapter 11 in 2023, and Somryst's assets were later picked up by Nox Health.

Different companies, different diseases, same cause of death: a product regulated like a device, used like a consumer app, and reimbursed like neither.

Economics That Don't Survive the Budget

Avinger built the Pantheris family of catheters, devices that let physicians see inside an artery while clearing plaque, instead of relying only on external imaging. The FDA cleared the original Pantheris in 2015, and a newer version, Pantheris LV, in June 2023. In Avinger's own clinical trial, the device hit a 97% technical success rate and a 6.4% six-month rate of needing a repeat procedure, with no reported perforations. This wasn't a product anyone doubted would work.

The problem was how hospitals get paid for the procedure it's used in. Peripheral artery disease treatments are billed as a single bundled payment, one fixed amount covering the whole procedure, regardless of which device the physician uses inside it. A more expensive, image-guided catheter didn't come with a matching bump in what the hospital was paid for the case. Every time a hospital chose Pantheris over a cheaper alternative, it made less money on that case. The device worked. The payment system had no way to reward it for working better.

Avinger received a Nasdaq delisting notice, and in February 2025 the company executed an assignment for the benefit of creditors, an out-of-court process that wound the company down.

The Clock

Proteus Digital Health built an ingestible sensor that could confirm when a patient actually took their medication, technology real enough that Otsuka partnered with it to create Abilify MyCite, the first FDA-approved digital medicine, in 2017. At its peak, Proteus was valued at $1.5 billion.

The technology worked. There was real friction around it too, patients and clinicians were understandably uneasy about a pill that reports back once it's swallowed, especially attached to a mental health drug. That wasn't what ended Proteus.

By late 2019, the company was burning through roughly $2 million a month with no clear date when revenue would catch up. A planned $100 million funding round didn't close. The company furloughed nearly all of its employees. In January 2020, a potential deal with Otsuka fell apart.

Proteus filed for Chapter 11 bankruptcy in June 2020. Otsuka came back and acquired the company's core assets through a bankruptcy sale for $15 million.

When the Clock Doesn't Win

Boston Scientific's EXALT Model D, the first single-use duodenoscope, got FDA clearance in December 2019. The need was never in question, reprocessing failures on reusable scopes were a known infection risk, and EXALT solved it directly. What took time was payment. Medicare's outpatient and inpatient payment systems don't automatically make room for a new device that costs more than what it replaces. Boston Scientific pursued a transitional pass-through payment for outpatient cases, then a New Technology Add-on Payment for inpatient cases, effective October 2021, roughly two years after clearance. They treated that gap as something to go build.

Sequana Medical is fighting the same fight right now. The FDA approved its alfapump System in December 2024 for patients with recurrent fluid buildup from liver cirrhosis. Sequana secured a CMS New Technology Add-on Payment worth up to $21,450 per case, raised €13.6 million in financing through the first half of 2025, and ran a deliberate soft launch, five centers live by February 2026, with a full launch targeted for the second quarter of 2026.

Same disease as Proteus: a gap between clearance and the market being ready to pay. Different outcome, because someone treated the gap itself as the thing to manage.

How Common Is This?

The median time between FDA clearance and a permanent Medicare coverage decision runs about 5.7 years. Across medtech broadly, the typical gap between clearance and meaningful revenue runs three to seven years. The system is built to move at that pace, and few planning processes are built to match it.

The Actual Shape of the Graveyard

The pattern across these cases is consistent. Each company solved the clinical problem and cleared the regulatory bar. What varied was whether category, margin, or time also got solved before the company ran out of runway.

What This Means for the Next Company

Every cause of death in this piece was visible well before it happened, to anyone tracking it closely enough. A few questions worth asking early, alongside the clinical and regulatory ones:

1. 1

   What category is this product actually in, and has anyone besides us agreed to pay for it that way?

2. 2

   What does our margin look like once it runs through the real payment mechanism, not the list price?

3. 3

   How many months of runway do we have left, compared to how long this specific gap has historically taken to close for someone else?

These questions live in a board deck and a financial model, not a clinical trial, which may be exactly why they're easy to miss.

Funding the Plumbing

Go back to the oxygen concentrators. The shortage got solved quickly, because it was visible and urgent. What came after, maintenance, spare parts, a technician on call, was quieter and easier to underfund, even though it determined whether the device kept anyone alive.

Medtech commercialization runs on the same logic. Clearance is visible and urgent. Everything after it, the category, the payment mechanism, the runway to get there, is quieter and easier to underfund, even though it determines whether the company survives.

Fund that part deliberately, the same way the trial gets funded, and the odds shift meaningfully in your favor.