Tendon Injury and Repair: How Tendons Heal, and Why It Takes So Long

A tendon injury is damage to the dense collagen cord that carries force from muscle to bone, whether a sudden rupture or a gradual, load-related change called tendinopathy. Tendons repair slowly and imperfectly because of how they are built: few cells, a sparse blood supply and, in the adult Achilles tendon, a collagen core that is barely renewed. That biology shapes tendon research and limits what peptide studies can show.

This guide covers tendon structure, tendon versus ligament, injury types, the three phases of tendon repair, why repair takes months, lab models, and what peptide research has and has not found.

How a tendon is built: from collagen molecule to fascicle

A tendon is a dense, fibrous connective tissue that transmits muscle force to bone. About 70% of its mass is water; of the dry mass, collagen makes up 65 to 80%, mostly type I (see the types of collagen), and elastin about 1 to 2%.

The collagen hierarchy

Each type I collagen molecule is a triple helix of three chains. Molecules assemble into fibrils, which in rat tail tendon measure about 50 to 500 nm across and run in a planar zig-zag called crimp. Fibrils bundle into fibres, the tendon’s basic unit, each wrapped in a fine sheath called the endotenon. Fibres group into larger bundles, the fascicles, 80 to 320 µm across in rat tail tendon. The epitenon wraps the whole tendon, and many tendons also have a loose outer gliding layer, the paratenon.

Cells and matrix

Cells are sparse. Tenoblasts mature into elongated, less active tenocytes, which together make up 90 to 95% of tendon cells and produce the matrix. In 2007, Bi and colleagues identified tendon stem/progenitor cells (TSPCs) in human and mouse tendons, with clonogenicity, multipotency and self-renewal, in a niche organized largely by biglycan and fibromodulin. Small proteoglycans such as decorin, the most abundant in tendon, bind collagen fibrils and help form them; in compressed fibrocartilage near bony insertions, the large proteoglycan aggrecan is abundant.

A sparse blood supply

Healthy tendon is poorly vascularized. Vessels enter at the muscle and bone junctions and through the paratenon, while sheathed tendons, such as those in the hand, depend partly on diffusion. Watershed zones include part of the Achilles tendon and a critical zone about 1 cm from the supraspinatus insertion. Vascular endothelial growth factor (VEGF), which stimulates new vessel growth, is low in intact adult tendon; see angiogenesis and VEGF explained.

Tendon vs ligament: how the two tissues differ

A tendon joins muscle to bone; a ligament spans a joint and is anchored to bone at each end. The ligament figures below come from a review by C. B. Frank of the University of Calgary’s McCaig Centre for Joint Injury and Arthritis Research, drawing on rabbit medial collateral ligament (MCL) studies.

Names can mislead: the patellar tendon joins the kneecap to the tibia, bone to bone, and the US National Library of Medicine’s MeSH vocabulary files it under patellar ligament.

Types of tendon injury: rupture, laceration and tendinopathy

Healthy tendons are injured acutely, by an overload that ruptures them or a laceration that cuts them; pathological tendons fail through overuse or age-related degeneration. How a tendon then repairs depends partly on whether it runs inside a synovial sheath.

Tendinopathy and the continuum model

Persistent, load-related tendon pain was historically called tendinitis. In 2019, an international consensus led by Alex Scott of the University of British Columbia recommended “tendinopathy” for persistent tendon pain and loss of function related to mechanical loading, and did not recommend “tendinosis” without an agreed definition. The continuum model proposed by Cook and Purdam in 2009 describes three stages:

• Reactive tendinopathy: a non-inflammatory, proliferative response in which cells round up and make water-binding proteoglycans while the collagen stays largely intact.
• Tendon dysrepair: more cells and matrix production, disorganized collagen, and sometimes vessel and nerve ingrowth.
• Degenerative tendinopathy: cell death, disordered matrix and little capacity for reversal.

In the model, mechanical load drives movement along the continuum; this article describes only the tissue pathology, not injury management.

Common sites: the rotator cuff and the Achilles tendon

The tendons of supraspinatus, infraspinatus, teres minor and subscapularis form the rotator cuff, which holds the head of the humerus in the shallow glenoid socket. The 2019 consensus group did not agree on “rotator cuff tendinopathy” as a label, because some members felt it could not be diagnosed clinically with so many pain-sensitive structures nearby; impingement is described as the acromion pressing on these tendons where the subacromial space has narrowed. The Achilles tendon joins the calf muscles to the heel bone and is described as the strongest and largest tendon in the human body, with a paratenon rather than a true synovial sheath, a hypovascular area about 2 to 6 cm above its insertion, and a middle segment that is the most common site of injury.

How tendon repair works: three overlapping phases

Tendon repair follows the sequence of inflammation, proliferation and remodelling described in our guide to the wound healing process. The timings below come from a 2006 review by Sharma and Maffulli; the phases overlap and vary with species, tendon and injury.

1. Inflammatory phase, the first days. Inflammatory cells, particularly neutrophils, enter the site, and within 24 hours monocytes and macrophages predominate and clear necrotic material. Their growth factors and cytokines draw in new vessels, fibroblasts and tenocytes, and type III collagen synthesis begins. See inflammatory mediators.
2. Proliferative phase, a few weeks. From a few days after injury, type III collagen synthesis peaks while water and glycosaminoglycan content stay high. Sharma and Maffulli call this stage “remodelling” and the next “modelling”, so names vary between papers.
3. Remodelling phase, about six weeks to a year. A consolidation stage running to about ten weeks turns the tissue fibrous, aligns tenocytes and fibres in the direction of stress and shifts synthesis toward type I collagen. A maturation stage then converts it to scar-like tendon tissue over about a year.

Why repaired tendon stays weaker

The review states that the biochemical and mechanical properties of healed tendon never match those of intact tendon; in a sheep study it cites, transected Achilles tendons left to repair spontaneously reached 56.7% of normal rupture force at 12 months. Rabbit MCL scar shows the same pattern, with more type III and V collagen, smaller fibrils, immature cross-links and failure loads about 50% of normal.

Tendon healing time: why the biology is so slow

Three features of the tissue explain most of the slowness:

• Low metabolism. The 2006 review cites oxygen consumption by tendons and ligaments 7.5 times lower than in skeletal muscle, which lets tendons carry load for long periods without ischemia but slows repair.
• Sparse vessels. Watershed zones in tendons such as the Achilles and patellar tendons have been linked to tendinopathy and rupture.
• Almost no renewal. Using carbon-14 from the 1955–1963 nuclear bomb tests, Heinemeier and colleagues analyzed 28 Achilles tendon core samples from people born between 1945 and 1983. Core carbon-14 matched the atmosphere of roughly the first 17 years of life: the core forms during growth and is essentially not renewed afterwards, while muscle samples showed continuous turnover.

How tendon injury and repair are studied in the lab

Animal models let researchers follow tendon repair over time. A 2014 review by Hast, Zuskov and Soslowsky describes the main approaches:

• Acute models: surgical transection, for example of the rat Achilles tendon or the rotator cuff; the rat is the most common rotator cuff model because its shoulder anatomy most resembles the human shoulder.
• Overuse models: downhill treadmill running in rats, which adds eccentric load.
• Chemical models: collagenase delivered into the tendon to break down matrix, the most widely used Achilles tendinopathy model.
• Outcomes: failure load, stiffness and modulus; histology of collagen organization, cellularity and vascularity; the type I to type III collagen ratio; and imaging.

Growth factors studied in tendon repair

A 2003 review by Molloy, Wang and Murrell describes five growth factors upregulated after tendon injury. IGF-1, expressed early, promotes fibroblast proliferation and migration and then collagen production (see what is IGF-1). TGF-β, active during inflammation, regulates migration and proliferation. VEGF peaks after the inflammatory phase and strongly stimulates angiogenesis. PDGF, released shortly after damage, helps induce other factors including IGF-1. bFGF promotes angiogenesis and cell proliferation. Genetic evidence implicates growth and differentiation factor 5 (GDF5): Achilles tendons in GDF5-deficient mice repaired about one to two weeks more slowly than in littermates and were weaker at five weeks, though not at twelve.

Where peptide research fits

Research peptides enter tendon science at the cell and animal level only; our guide on how to read a peptide study explains why such results do not transfer automatically to people.

Rat tendon and tendon-cell studies

In a 2003 study, rats with a cut Achilles tendon were given BPC-157, a 15-residue peptide, or saline. Over 14 days the BPC-157 groups showed higher load to failure, higher Achilles functional index values and more collagen formation on histology; in culture it reversed the inhibition of tendon-cell growth by 4-hydroxynonenal. A 2011 study of rat Achilles tendon explants and fibroblasts reported faster outgrowth, better survival under hydrogen-peroxide stress, more migration and increased FAK and paxillin phosphorylation, with no direct effect on proliferation. A 2025 systematic review found that 35 of 36 BPC-157 studies in orthopedic sports medicine were preclinical and rated the evidence at levels IV and V. See BPC-157 vs TB-500 and BPC-157 safety research.

Ligament and fibroblast studies

Thymosin beta-4, a 43-residue peptide that sequesters actin monomers, has been tested in a rat ligament model: placed in fibrin sealant in a cut medial collateral ligament, it was followed at four weeks by larger collagen fibrils and better mechanical properties than in controls. GHK-Cu stimulated collagen synthesis in cultured fibroblasts in a 1988 study, independently of any change in cell number; the GHK sequence occurs within the alpha-2 chain of type I collagen. For its alanine analogue AHK-Cu, the one primary study we could confirm used human hair follicles and dermal papilla cells, and we found no tendon or fibroblast-collagen study. See GHK-Cu research and GHK-Cu vs AHK-Cu.

What’s established and what isn’t

Sourcing tendon-research peptides in Canada

Tendon-cell assays read out migration, survival and matrix production, so they are sensitive to what is in the vial. When buying research peptides in Canada:

• Ask for a batch-matched certificate showing identity by mass spectrometry and purity by HPLC; see our peptide COA guide and HPLC testing guide.
• Confirm the sequence. Our catalogue lists thymosin beta-4 (TB-500) and a separate fragment, compared in TB-500 vs TB-500 fragment; GHK-Cu and AHK-Cu differ by one residue.
• Plan controls, such as a vehicle and a scrambled-sequence peptide.
• Store lyophilized material cold, dry and dark; see the peptide storage guide.

BPC-15715-residue peptide used in rat tendon-transection studiesView product TB-500 (Thymosin Beta-4)Actin-binding peptide studied in a rat ligament modelView product GHK-CuCopper tripeptide studied in fibroblast collagen assaysView product AHK-CuAla-His-Lys copper tripeptide; ex-vivo and cell data onlyView product

Frequently asked questions