AI-assisted article. Not yet reviewed by a qualified medical or scientific professional.

For educational purposes only. This article is not personal medical advice and should not guide individual treatment decisions.

An open wound makes a problem visible. The reason it will not close is usually less visible: pressure, impaired circulation, damaged tissue, repeated inflammation and the limits of existing care. For some people with diabetes, the treatment team can do the fundamentals correctly and still watch healing stall.

This is the clinical problem behind a randomized study published October 2, 2026, in the Journal of Wound Care. Among 170 people with particularly slow-healing diabetic foot ulcers, adding a collagen matrix coated with an antimicrobial agent to standard care increased the proportion of wounds closed at 12 weeks. The result was 42% with the matrix versus 27% with standard care alone. The paper reports an absolute difference of 14.6 percentage points, with a 95% confidence interval of 0.15 to 27.9 points. PREPARE trial.

The result is more informative than a persuasive before-and-after photograph. It is also less certain than a headline reading “42% healed” suggests. To understand its value, start with the unusual four weeks before randomization.

The four-week test before the trial began

PREPARE did not recruit a broad group of all people with diabetic foot ulcers and immediately compare two dressings. Participants first entered a four-week run-in period receiving standardized wound care. Anyone whose target wound area had already shrunk by at least 40% during that period was excluded from randomization.

This matters enormously. A wound that is already healing rapidly under appropriate care is a different research question from a wound that remains resistant despite that care. By filtering out early responders, the researchers concentrated on a harder-to-treat population, one for whom an additional therapy could have a more meaningful role. But the result cannot automatically be extended to every new foot ulcer. Trial design and inclusion criteria.

Across 25 US sites, 96 randomized patients received the collagen matrix plus standard care, and 74 received standard care alone. The target ulcers were associated with diabetes, measured between 1 and 25 square centimeters, and met additional requirements, including adequate limb blood flow and the absence of specified serious complicating infections. The study was not testing treatment of deep bone infection, critically ischemic limbs or every possible diabetic wound.

Two treatments built around the same foundation

Both groups continued a defined clinical protocol: wound cleansing, debridement when indicated, appropriate dressings, and offloading to reduce mechanical pressure. The additional product—purified native type I collagen extracellular matrix coated with polyhexamethylene biguanide, or PHMB—was placed on wounds in the intervention group.

Collagen forms part of the body's connective-tissue architecture. In this application, it is used as a scaffold within a complex wound-care product. PHMB provides an antimicrobial component. That does not mean swallowing collagen peptides, injecting a collagen fragment or applying an unrelated cosmetic product can reproduce the trial result. Identity, formulation, route, wound selection and concurrent clinical care all matter.

The trial's comparison was the whole matrix-based treatment added to the defined standard-care program. It was not a clean test isolating the independent effect of collagen versus PHMB. Methods and intervention.

The result, with its uncertainty left intact

The study's primary endpoint was complete wound closure within 12 weeks, confirmed approximately two weeks later. The additional matrix group reached that endpoint more often: 42% compared with 27% under standard care alone. The authors reported a p-value of 0.047.

A result just below a conventional 0.05 threshold is not necessarily unimportant; nor is it a guarantee that the true effect is large. The 95% confidence interval for the absolute difference ran from 0.15 to 27.9 percentage points. Near its low end, the added clinical benefit would be small; near its upper end, it would be substantial. A replication study could narrow that range.

The mean percentage reduction in wound area was also greater with the matrix, 75.23% versus 58.06%. The paper reports no adverse events or serious adverse events attributable to the matrix in the observed period. That is useful safety information, though it does not establish the rate of uncommon harms or long-term recurrence. Results.

It is useful to keep the endpoint straight. Twelve-week closure is not the same as avoiding amputation, preventing future infection, recovering mobility or remaining healed a year later. All are clinically important, and none can be assumed from a 12-week closure comparison.

Strengths and blind spots of a pragmatic-looking study

The multicenter randomized design and the hard-to-heal population make PREPARE more informative than a series of carefully selected before-and-after photographs. Treating clinicians and patients, however, could not be blinded because the wound product was visibly applied. The investigators addressed one major risk of bias by having an independent panel evaluate deidentified wound photographs without knowing treatment assignments.

The study was funded by Organogenesis, the company manufacturing the product, with operational oversight by contract research organization SerenaGroup. Several authors had relationships with these organizations, and some were speakers for the manufacturer. These disclosures do not make the data false, but they increase the importance of independent replication and transparent methods. Conflict and funding disclosure.

A 12-week endpoint also leaves cost effectiveness unresolved. An adjunct can improve closure rates yet still pose a difficult resource decision if it requires repeated application, specialized care or high product cost. The paper alone is not enough to conclude who should receive it or whether it produces better value than alternative advanced wound-care options.

What this study contributes to regenerative medicine

Regeneration is often depicted as a single extraordinary molecule persuading the body to heal. Chronic-wound treatment provides a different picture. The surrounding environment—pressure, circulation, tissue condition, infection control, dressing selection and repeated care—may matter as much as the new material.

PREPARE is notable because it tested an added intervention on top of a structured treatment foundation and examined an outcome that patients and clinicians recognize: an open wound becoming closed. Its most persuasive feature is also its limitation. The result belongs to a specific, difficult-to-heal group receiving supervised diabetic-foot care.

The next trial should not merely show another attractive closure photograph. It should test whether the observed gain persists, whether ulcers recur less often, whether patients function better and whether the added cost is justified. The result worth pursuing is not a better-looking wound at one visit. It is a foot that remains well enough for its owner to keep moving.

Evidence note and sources

Human multicenter randomized open-label trial, 170 participants, with blinded photo adjudication of the primary closure endpoint; the treatment effect's confidence interval is broad. Sponsor relationships are documented. Findings do not justify home treatment or replacing established wound-care measures. People with diabetic foot wounds need clinical assessment, especially when signs of infection or impaired blood supply arise. Primary sources: PREPARE trial, Journal of Wound Care, October 2, 2026; PubMed record; Trial registry NCT06618612.

For research where laboratory mechanisms are checked against patient outcomes, read Recovery Science Digest.

Related reading

For another example of reading wound-repair evidence cautiously, see the article below. LL-37 research is not evidence of treatment equivalence with a collagen-PHMB matrix.

• LL-37 and the limits of immunity and healing research

AI-assisted editorial article. Human fact-checking and appropriate expert review remain outstanding.