Two patch materials, one operation, no clear winner after 4,971 patients
Synthetic and biological neck-artery patches show no meaningful difference in stroke or death risk after 80 months.
By The Weekend · · 5 min read
Surgeons closing a widened neck artery have two choices of patch material — a strip of the patient's own vein or cow tissue, or a sheet of synthetic polymer — and after tracking 4,971 patients for up to 80 months, the evidence says it barely matters which one they pick.
The operation is carotid endarterectomy: scraping fatty plaque out of the carotid artery, the main pipe feeding blood to the brain, then sewing a patch over the opening to widen it and keep it from narrowing again. Left untreated, that plaque is a stroke waiting to happen. The patch is a small technical decision inside a serious procedure, and for decades surgeons have debated which material serves the artery better.
The number the whole debate rests on: 2,021 procedures, one verdict
A Cochrane review pooled 15 studies and 4,971 participants, mostly men in their sixties, operated on across North America, Europe and Australia. Of those, 12 studies covering 2,021 procedures fed the head-to-head comparison between synthetic patches — PTFE or Dacron — and biological ones, meaning either a vein taken from the patient's own body or bovine pericardium, the fibrous sac from a cow's heart.
The result, tracked out to 80 months: probably little or no difference in stroke or death. Probably little or no difference in stroke confined to one side of the body, stroke of any kind, or the artery narrowing back down by more than half. Nine studies covering 1,431 participants specifically checked that re-narrowing outcome, and synthetic patches came out even with biological ones.
That is the figure the story turns on: two materials, thousands of patients, years of follow-up, and the difference between them rounds to zero on the outcomes that matter most.
Why "little or no difference" is a stronger finding than it sounds
A null result is easy to misread as a shrug. Here it isn't. Cochrane reviewers don't report "little or no difference" casually — they grade it against the size and quality of the underlying trials, and this verdict is built on the largest pooled dataset available for the question, not a single underpowered study reaching for a headline.
It also settles something practical. Biological patches, especially a patient's own vein, take longer to harvest and add a second surgical site with its own healing risk. Synthetic patches come off a shelf, cut to size, no second incision. If outcomes are equivalent, that difference in convenience and cost stops being a trade-off against safety and becomes just a logistics choice — which is a real, usable answer for a surgeon standing over an open neck.
Where the equivalence breaks down: the numbers too small to trust yet
The reviewers are explicit about the limits, and the honest version of this story includes them. Three outcomes stayed "very uncertain" rather than settled: stroke on one side of the body during the operation or within 30 days, checked in just 797 participants across five studies; artery rupture during surgery or within a month, in 1,068 participants across six studies; and pseudoaneurysm — a blood-filled bulge where the repair leaks — tracked in 980 participants across five studies.
Those are the acute, in-theatre risks, and they are measured in hundreds of patients, not thousands. A complication rare enough to matter to an individual patient but too rare to show up reliably in a sample that size is exactly the kind of risk this review cannot rule out. The strongest counter-argument to "the patches are equivalent" is that equivalence has only been demonstrated for the outcomes with enough events to measure — and the events with the fewest recorded cases are also the ones most feared by surgeons and patients alike: bleeding on the table, a torn repair, a bulge that shows up months later.
The re-narrowing data, the one the review is most confident about, still only runs to 80 months. Carotid arteries can re-narrow over a decade or more. Nobody has followed a cohort this size long enough to know if the equivalence holds at year twelve the way it does at year six.
What changes at the operating table, and what still doesn't
Nothing here recommends one material over another, and that is itself the practical output. Surgeons will keep choosing based on what is already true outside this review: bovine pericardium and synthetic grafts carry a small, well-documented infection risk profile different from a patient's own vein; using the patient's own vein adds operating time and a second wound but avoids implanting foreign material; PTFE and Dacron are more available and don't depend on the patient having a usable vein to harvest in the first place.
The review doesn't fund any of this decision-making, and the studies inside it were largely run by surgical departments and device makers with an interest in one material or the other doing well — which makes the null result more credible, not less, since a genuine advantage for either side would have been commercially useful to find.
What the number does next is narrow, not close, the debate. The reviewers call for larger studies with longer follow-up, specifically aimed at the operative-window complications — rupture, acute stroke, pseudoaneurysm — where the current sample sizes are a fraction of the 4,971 used for the long-term comparison. Until those exist, the honest clinical position is that patch material is probably not where the stroke risk in this operation lives. It lives somewhere else in the procedure, and this review's real contribution is ruling out one variable so surgeons and researchers can stop scrutinizing the patch and look at what does move the numbers.