Cod skin beat collagen scaffolds by 12 hospital days, in a study the fish skin's maker funded

The complication gap is the number to hold, not the bed-days. Sixteen per cent against 8 per cent ^[8] means roughly 12 or 13 patients have to receive intact fish skin instead of a synthetic substitute for one case of sepsis or similar to be avoided ^[18]. That is a large effect for a temporary wound covering, and large effects in matched retrospective data are the ones most likely to shrink once somebody randomises.

The arms were nowhere near equal. Ninety-nine patients got GraftGuide and about 372 got approved synthetic substitutes ^[4][16], so the fish skin group is roughly a fifth of the cohort ^[17]. Matching covered age, sex, burn severity and size, and other complications from the injury ^[5], which is sensible as far as it goes; the data still spans 49 centres and a window from 2019 to 2025 ^[3], and a discharge date is partly a property of the institution writing it. Kerecis, which makes the product, funded the work and according to the published account did not analyse the data ^[9]. Fiona Wood at the University of Western Australia puts the milder version of the objection: comparing equivalent patients on different treatments is always difficult, and formal head-to-head trials are what settle it ^[13].

The proposed mechanism is thinner than the outcome. John Lantis at the Icahn School of Medicine at Mount Sinai notes that fish skin resembles human skin under a microscope, so "maybe it works like [our] skin" ^[10], and points to unpublished data in which omega-3 on a burn produces an initial spike of inflammation that stops bleeding and then fades ^[20]. Neither proposition is tested by this dataset.

What is concrete is the processing. Because very few fish viruses are thought to affect people, the skin does not need the harsh detergents that would wash the omega-3 out ^[11], and a cod is large enough to yield a useful amount of skin per animal ^[12]. That is a manufacturing argument rather than a clinical one, and it is the reason a fishery byproduct can sit on the same shelf as a synthesised collagen scaffold ^[2].

The trial being run to firm this up asks a different question. Each participant gets fish skin on part of the burn and a graft on the rest, so every patient is their own control ^[14], which removes the matching problem entirely. But it is testing whether fish skin reduces the need for grafts in second-degree burns ^[14], while the retrospective was in patients whose burns were mostly third-degree or mixed-depth and who went on to be grafted anyway ^[6]. The head-to-head Wood asks for, cod skin against a collagen scaffold in deep burns, is not the one under way.

Scale is why this matters more than a single sponsored analysis should. More than 30,000 people a day develop burns needing medical attention ^[15], which is upwards of 10.9 million a year ^[19]. Only a slice of those reach a burn centre, but against that denominator a 12-day difference in length of stay ^[7] is a bed capacity question, and bed capacity is a bad thing to settle on the manufacturer's own data.