Medtronic's Best Moat Is a Surgeon's Hands. Some of It the DOJ Has Already Drained.

A spine surgeon does not really choose a device at the table. The choice was made years earlier — in a cadaver lab, on the company-funded course, in the thousand small habits of which screw seats in which guide, which tray is laid out in which order, which software the OR already speaks. By the time the patient is open, the surgeon's hands belong to a workflow, and the workflow has a logo on it. That is the most underrated moat in medical devices, and Medtronic built one of the deepest: $33.5 billion in revenue and a 25.7% operating margin sitting on top of it.¹

The official story is that this is a relationship business — surgeons are trained, surgeons trust, surgeons stay. That is true. It is also the part of the story that the Department of Justice has spent two decades pricing, because the line between training a surgeon and buying one turned out to be thinner than the marketing suggested.

Here is the thesis a smart skeptic should be able to argue against, and that we'll argue for anyway: Medtronic's surgeon moat is real, but it is partly a function of regulatory tolerance rather than purely organic stickiness. The same training-and-payments infrastructure that creates lock-in has repeatedly crossed into documented kickbacks. So the moat's depth is set, in part, by how much enforcement looks away — which makes it more fragile than the 'sticky relationships' narrative admits.

Why a trained surgeon almost never switches

Start with the legitimate mechanism, because it's powerful. A device a surgeon implants is what the industry calls a physician-preference item: the surgeon, not the hospital purchasing department, effectively picks the brand, because the surgeon is the one whose hands are on the line. Switching costs here aren't a contract clause — they're cognitive and physical. Relearn a system and your first cases are slower, your complication risk ticks up, your team has to be retrained around you. Nobody wants to be down the learning curve while a patient is open. That friction is genuine, and it's why a margin like 75.1% on spine products could exist at all: that's the average profit margin on its spine products and therapies that Time reported Medtronic's October 2012 SEC filing showed, as cited in widely-reported coverage of the period.⁷ You don't earn three-quarters-margin pricing on a commodity. You earn it on a surgeon who would rather not relearn anything.

Medtronic is now hardening this with technology rather than relationship alone. Its AiBLE spine-surgery ecosystem bundles navigation, robotics, and imaging into one workflow, and in September 2024 it announced a partnership with Siemens Healthineers to co-market and integrate imaging systems — the company's CST division grew high-single-digit on that adoption.⁸ This is the textbook second-generation move: convert a soft, person-to-person moat into hard, system-to-system switching costs that survive any individual surgeon leaving. If it works, the lock-in stops depending on loyalty and starts depending on the OR's plumbing.

The part the marketing leaves out: the moat has a rap sheet

Now the uncomfortable half. The same machinery that trains surgeons also pays them — and the DOJ has repeatedly found that some of those payments weren't education at all. In 2006, Medtronic paid $40 million to settle allegations that its spinal division paid doctors through sham consulting agreements, sham royalty agreements, and lavish trips to induce use of its products between 1998 and 2003.³ The word the DOJ used was 'sham,' not 'consulting.' In 2011, it paid $23.5 million over payments tied to post-market studies and device registries that allegedly functioned as kickbacks to get doctors to implant its pacemakers and defibrillators — including soliciting physicians to convert business away from competitors.⁴ In 2014, it paid $9.9 million over a separate set of pacemaker and defibrillator kickbacks — speaking fees, free marketing plans, and sports tickets — that overlapped in device category but stemmed from a distinct whistleblower complaint.⁹ And in 2020, it paid roughly $9.2 million over a South Dakota neurosurgeon, including funding more than 100 social events at a restaurant the neurosurgeon owned, to push its infusion pumps.⁵ Spine, cardiac rhythm, neuromodulation. Not one anomaly — a pattern across the catalog.

When the surgeon advocacy was partly written in-house

The most revealing case isn't a payment at all — it's the science. Infuse, Medtronic's bone-growth product, is often told as the purest surgeon-relationship win: leading spine surgeons published studies, championed the device, drove adoption. Then a 16-month U.S. Senate Finance Committee investigation, released in October 2012, looked through 5,000 documents across 13 Infuse studies and concluded that Medtronic employees collaborated to edit — and in some cases write — those published papers without public disclosure, and that the company had paid the 13 physician authors and two associated entities $210 million over 15 years.⁶ Read that back. The 'independent clinical endorsement' that fed the moat was, in part, authored by the company and funded by the company. The advocacy wasn't discovered. Some of it was manufactured.

The honest counter: a $33 billion company with a 47-year dividend streak isn't fragile

The fair objection writes itself. Medtronic posted $33.5 billion in revenue, a 25.7% operating margin, and 4.9% organic growth in FY2025;¹ its Cardiovascular Portfolio alone did $11.8 billion in FY2024, with $5.2 billion of free cash flow and a 47th consecutive dividend increase.² You don't compound dividends for 47 years on a moat that's about to collapse. And the settlements, however ugly, were a fraction of the cash this business throws off — well over $80 million across the four cases is real, but it's a rounding error against $5 billion a year in free cash. So why call the moat brittle?

Because the settlements aren't the threat — they're the receipt. The threat is what the receipt tells you about how the moat was partly built. Two forces now press on exactly the soft tissue the DOJ has been documenting. The first is reform of physician-preference-item economics and disclosure: the more transparent the payments and the more the hospital, not the surgeon, controls the purchase, the more the moat reverts to its organic core — which is real but thinner than 75% margins imply. The second is straightforward competition. Medtronic does not own spine; rivals are actively displacing it — analyst surveys of spine surgeons have forecast share gains by Globus Medical and Alphatec at Medtronic's direct expense¹⁰ — which means surgeons can and do switch when given a reason. A moat that's eroding at the edges while leaning on enforcement tolerance at its base is not collapsing — but it is being repriced, and the AiBLE technology bet is best read as Medtronic itself conceding that the relationship moat alone is no longer enough to hold the line.⁸

So here's the sharpened version. A surgeon's trained hands are a wonderful thing to own — when you've earned them with a genuinely better workflow. Medtronic earned plenty of that. But it also rented some, with sham consulting fees, funded dinners, and studies it helped write, and rented stickiness comes due. The deepest moats look identical to the permitted ones right up until the enforcer changes its mind. Medtronic spent two decades and well over $80 million discovering which parts of its moat it actually owned — and which parts it was only being allowed to keep.