Can collagen really repair knee cartilage?

Three claims in one sentence: that collagen reaches cartilage, graded partly, and that it changes cartilage, graded once and unreplicated

Can collagen really repair knee cartilage is a sentence that hides three claims, and the honest page splits them before grading anything, because the word really is a request to know which parts of the story survive scrutiny and the three parts have three different fates, so that a single yes or no would be wrong about two of them whichever it was. The first claim is that collagen reaches cartilage, which has to be true for anything else to be, and it is partly true. After a dose of hydrolysed collagen, the gut digests most of it to free amino acids that join the general pool, but a fraction survives as very short peptides, pairs and triples of amino acids, and some of these, prolyl-hydroxyproline and hydroxyprolyl-glycine above all, appear in the blood within an hour and remain for several hours at concentrations that are measurable and low, which is the finding that separates collagen from a shelf of amino acids. In dishes, these dipeptides act on fibroblasts and chondrocytes, nudging them toward matrix synthesis and away from the enzymes that degrade it, which is the mechanism the whole cartilage story rests on. And in rodents fed radiolabelled collagen peptides, label has been recovered from cartilage, among other tissues, within hours, which shows that at least some of what survives digestion reaches the tissue in an animal. What is not known is the part the shelf skips: the concentration that reaches a human knee's articular cartilage, a tissue with no blood supply that is fed by diffusion from the joint fluid through a dense matrix, and whether that concentration, which must be lower than the already low concentration in blood, is enough to do in a human joint what the dipeptides do in a dish at the concentrations a dish allows. So the first claim is graded partly true: collagen reaches the blood as specific peptides, reaches cartilage in rodents, and reaches a human knee at an unknown and probably small concentration. The second claim is that collagen changes cartilage, and it rests on a single study, which the honest page describes exactly. A twenty-four week randomised pilot in people with mild knee osteoarthritis gave 10g of collagen hydrolysate a day or placebo and measured cartilage glycosaminoglycan, the component that holds water and resists compression, with a contrast MRI technique in several regions of the knee, and reported that the measure rose in some regions in the collagen group while it fell on placebo, with no difference in other regions and no clinical difference; it was small, its results were regional rather than whole-knee, and it has not been replicated by anyone in the years since, which in a field where a positive imaging result for an oral supplement would be of enormous commercial interest is itself informative. Against it stand the larger imaging and X-ray trials of oral supplements in knee osteoarthritis, glucosamine, chondroitin, collagen hydrolysate, vitamin D and fish oil, which measured cartilage thickness or joint space over years and found no meaningful slowing of loss and no regrowth against placebo, on the supplements rebuild cartilage page; and the joint pain trials of collagen at around 10g a day for twelve to twenty-four weeks, which found a point or two less pain on questionnaires and either scanned no cartilage or found none changed. So the second claim is graded once and unreplicated: one small pilot's glycosaminoglycan signal, unconfirmed, against a body of imaging evidence that found nothing.

The third claim, that collagen repairs cartilage, graded no, what the three grades mean for a reader whose knee hurts, and the worked example

The third claim is the one the question means, that collagen repairs knee cartilage in the sense a reader with a worn or injured knee intends, filling a defect that an injury gouged out or restoring tissue that years of use have worn thin, and it is graded no, with a reason that is biology rather than a shortage of trials. Adult articular cartilage cannot repair a defect because its cells, the chondrocytes, sit locked in lacunae within a matrix they cannot migrate through, so that there are no cells at the edge of a defect that can move across it and lay new tissue, and because the tissue has no blood supply to deliver stem cells or the inflammatory response that begins a repair everywhere else; a defect that reaches the bone fills with fibrocartilage from the marrow, a scar-like tissue without the compressive properties of the original, and a defect that does not reach the bone stays. No trial of any oral agent has ever shown a cartilage defect filled or worn cartilage restored, which is a different and stronger statement than that collagen's trials have not, and the standing refutation is the existence of a surgical field for exactly this problem, cell implantation, microfracture and grafts for focal cartilage defects in the right knees, a field of operating theatres, cultured cells and years of rehabilitation that would not exist if a scoop could do the job, which is the argument on the collagen and knee repair page. So the third claim is no, and the honest page does not soften it, because a reader who believes a defect is being repaired by a powder is a reader who may delay the assessment that decides whether a surgical option exists for their knee while the window for it is open. What the three grades mean for a reader whose knee hurts is this. A reader whose knee hurts less after three months on 10g of collagen has what the joint pain trials found, a questionnaire result, a point or two on a ten-point scale, real and modest and not a repaired joint, since the same trials scanned no cartilage or found none changed; they may keep taking it for that result on its true terms, and they should not read the result as repair or let it replace the things that protect the cartilage they have, quadriceps strength, moderate cyclic load and body weight, which are the astronaut's and the runner's levers on the regenerate knee cartilage page. A reader with a locking, catching or giving-way knee after an injury has a possible focal defect or meniscal tear and needs a scan and a surgeon's opinion within the months in which the surgical options work, not a tub. A reader with a worn knee on X-ray is managing the cartilage that remains, and no claim on this page changes that. The worked example is graded against the same three claims and stated on its true terms: two capsules, 1200mg hydrolysed type 1 marine collagen with 12mg vitamin C at 15% of its reference value plus zinc, 120 for £16.99 at 28p a day, lab tested and UK made, below the 10g dose of every joint trial and of the one pilot, so that even the second claim's single study does not reach it, carrying vitamin C's collagen formation sentence for the normal function of cartilage, which is the register's maintenance sentence and not a repair claim, and no cartilage repair claim of its own, since the register permits none. Can collagen really repair knee cartilage: it partly reaches it, it changed a glycosaminoglycan measure once in a pilot nobody has repeated, and it repairs nothing, which is the honest grading of a sentence the shelf sells whole.

What to check on the label