Joints, Cavitation and Tissue Response

Put your hand on the back of your neck and turn your head: dozens of small joints are gliding, opening and closing beneath your fingers, each one wrapped in a capsule, bathed in fluid and wired with sensors. When a joint is stiff, injured or immobilized, the tissues around it change in predictable ways, and when a clinician moves it, sometimes there is a pop. This chapter explains what a joint really is, how examiners feel its motion, what the famous cracking sound is made of, and how injured tissue heals, scars and remodels.

Section 1Synovial joint structure

Three ways bones can meet

A joint, or articulation, is any place where two or more bones meet. Anatomists classify joints in two parallel ways. The structural classification asks what tissue binds the bones together: fibrous joints are held by dense connective tissue (the sutures of the skull, the interosseous membrane of the forearm, the distal tibiofibular syndesmosis); cartilaginous joints are held by cartilage (the hyaline synchondrosis of a growth plate, or the fibrocartilaginous symphyses such as the pubic symphysis and the intervertebral disc); and synovial joints have a fluid filled cavity between the bones, enclosed by a capsule. The functional classification asks how much the joint moves: synarthroses are essentially immovable, amphiarthroses are slightly movable and diarthroses are freely movable. Nearly all diarthroses are synovial.

The spine contains all three structural types. Each mobile segment, which spinal biomechanists call a functional spinal unit or motion segment (two adjacent vertebrae and the soft tissues between them), has a cartilaginous joint in front (the disc between the vertebral bodies) and a pair of synovial joints behind (the left and right zygapophyseal joints). The ligamentum flavum, interspinous ligaments and the fibrous posterior part of the sacroiliac joint add fibrous connections. Because the synovial joints are where the bones glide against each other, and where manual therapies apply most of their force, this chapter focuses on them.

The parts of a synovial joint

Every synovial joint shares the same basic plan, whether it is a knuckle, a knee or a facet joint in the lumbar spine.

• Articular cartilage. The ends of the bones are covered by hyaline cartilage, usually 1 to 4 mm thick in small joints and up to about 6 mm in the knee. It has no blood vessels, no lymphatics and no nerves. It is made of a few cells, the chondrocytes, scattered in a large extracellular matrix of type II collagen fibers and proteoglycans, chiefly aggrecan. Aggrecan carries many negatively charged glycosaminoglycan chains (chondroitin sulfate and keratan sulfate) that attract water, so cartilage is roughly 65 to 80 percent water by weight. The collagen network resists the swelling pressure of the proteoglycans, which gives cartilage its springy stiffness under compression.
• Joint capsule. A sleeve of connective tissue attaches around the margins of the articular surfaces. Its outer fibrous layer is dense, mostly type I collagen, and is often thickened into capsular ligaments. Its inner synovial membrane (synovium) lines every surface inside the joint that is not articular cartilage.
• Synovial membrane. The synovium is a thin, highly vascular layer with one to three cell layers on its surface. Type A synoviocytes are macrophage like cells that clear debris. Type B synoviocytes are fibroblast like cells that secrete hyaluronan (hyaluronic acid) and lubricin into the joint.
• Synovial fluid. The fluid is an ultrafiltrate of blood plasma enriched with hyaluronan and lubricin. A healthy joint contains only a small amount (a normal adult knee holds roughly 1 to 4 mL), clear and pale yellow, with fewer than about 200 white cells per microliter. Hyaluronan makes the fluid viscous and shear thinning: it is thick when the joint moves slowly and thinner when it moves fast. The fluid lubricates the surfaces and carries nutrients to chondrocytes, which have no blood supply of their own.
• Accessory structures. Many synovial joints contain fat pads, fibrocartilage menisci or discs (in the knee, the temporomandibular joint and the sternoclavicular joint), a labrum that deepens a socket (in the hip and shoulder), intracapsular or extracapsular ligaments, and nearby bursae, small synovial sacs that reduce friction where tendons glide over bone.

How a joint stays almost frictionless

Healthy cartilage sliding on cartilage in synovial fluid has a coefficient of friction around 0.001 to 0.01, lower than ice on ice. Three lubrication mechanisms work together. In boundary lubrication, lubricin and surface phospholipids bind to the cartilage surface, so that even when the surfaces press close together a slippery molecular layer separates them. In fluid film lubrication, a thin layer of synovial fluid is trapped and pressurized between moving surfaces. In weeping or hydrostatic lubrication, compression squeezes water out of the cartilage matrix into the gap, so the load is carried mostly by pressurized fluid rather than by the solid matrix. When the load is removed, the cartilage reabsorbs fluid like a sponge.

That sponge action is also how cartilage is fed. Because chondrocytes have no blood vessels, nutrients and waste move by diffusion and by the pumping effect of cyclic loading. A joint that is compressed and released many times a day, through walking and ordinary movement, keeps its cartilage nourished. A joint that is held still or held under constant load does not, a point we return to in Section 5.

Synovial joint shapes

Synovial joints are grouped by the shape of their surfaces, which sets their axes of motion.

The synovial joints of the spine

The zygapophyseal joints, usually called facet joints, form between the inferior articular process of the vertebra above and the superior articular process of the vertebra below. They are true synovial plane joints, with hyaline cartilage, a capsule, synovium and a small volume of fluid (about 1 to 1.5 mL in a lumbar facet joint). Their job is to guide and limit motion: they allow some movements and block others, and in extension they share part of the compressive load that would otherwise go through the disc.

The orientation of the facet surfaces changes from region to region, which largely explains why the neck rotates freely, the thorax rotates but resists flexion and extension, and the lumbar spine flexes and extends but resists rotation.

Other spinal synovial joints include the atlanto-occipital joints, the lateral and median atlantoaxial joints, the costovertebral and costotransverse joints where the ribs meet the thoracic vertebrae, and the anterior part of the sacroiliac joint, which is synovial in front and fibrous behind. The uncovertebral joints (joints of Luschka) between the uncinate processes of C3 to C7 are clefts that develop in the lateral disc during childhood; whether they count as true synovial joints is still debated.

Meniscoids and innervation

Spinal facet joints contain small inward folds of capsule and synovium called intra-articular inclusions or meniscoids. Three types are commonly described: thin connective tissue rims along the capsule edge, adipose tissue pads filling the gaps at the joint poles, and fibroadipose meniscoids, larger folds up to about 5 mm long that project between the articular surfaces. They likely help fill space, spread load and protect exposed cartilage at the margins of the joint as it opens and closes. One hypothesis, discussed in Chapter 4, proposes that a meniscoid can be pinched or trapped at the edge of the joint and contribute to acute locked back or wry neck.

Joint capsules are richly innervated. In the spine, each facet joint is supplied by the medial branches of the dorsal rami, typically from the level of the joint and the level above, which is why one painful facet can refer pain over two segments. Hilton's law captures a general rule: a nerve that supplies a joint also tends to supply the muscles that move it and the skin over it. The capsule and ligaments contain four classic types of receptor (after Wyke):

Section 2Joint play and end feel

Two kinds of joint motion

Ask someone to bend their elbow and you see the forearm swing through an arc. That visible movement of a bone through space is osteokinematic or physiological motion: flexion, extension, abduction, rotation and the other movements a person can make with their own muscles. Underneath it, the joint surfaces are moving relative to each other in small ways that cannot be seen and cannot be produced voluntarily. Those surface motions are arthrokinematics, and the small passive movements a clinician can produce by pushing or pulling one joint surface on the other are called accessory motion or joint play. The physician James Mennell, who popularized the term, argued that normal, pain free physiological movement depends on normal joint play, and that loss of joint play (which he called joint dysfunction) could cause pain and stiffness even when the bones looked normal on film.

Three arthrokinematic motions combine during any movement:

• Roll: new points on one surface meet new points on the other, like a tire rolling on a road.
• Glide (or slide): one point on a surface contacts new points on the other, like a skidding tire.
• Spin: one surface rotates around a fixed point on the other, like a top. The radial head spinning on the capitulum during forearm rotation is the classic example.

Pure rolling would roll a joint off its partner, and pure gliding would jam it. In real joints, roll and glide happen together in proportions set by the shapes of the surfaces.

The convex concave rule

Most joint surfaces are curved, with one convex partner and one concave partner. The convex concave rule (associated with Kaltenborn) predicts the direction of glide:

• When a convex surface moves on a fixed concave surface, it rolls one way and glides the opposite way. Example: as the convex humeral head elevates in the glenoid during abduction, it rolls up and glides down (inferiorly).
• When a concave surface moves on a fixed convex surface, roll and glide go in the same direction. Example: as the concave tibial plateau extends on the femoral condyles in an open chain knee extension, the tibia rolls and glides anteriorly.

Clinicians use the rule to choose the direction of a mobilization. If a shoulder will not abduct well, an inferior glide of the humeral head is the logical accessory motion to restore. The rule is a helpful starting point but a simplification: real joints are not perfect spheres, ligaments and muscles steer the surfaces, and imaging studies sometimes show glide directions that do not match the textbook prediction. The flat spinal facets do not follow it neatly at all; their arthrokinematics are better described as sliding of the inferior articular process up, down or apart relative to the one below.

Close packed and loose packed positions

Every synovial joint has one position in which the surfaces are most congruent, the capsule and ligaments are tightest and accessory motion is at a minimum. This is the close packed position. Examples are full extension of the knee, full extension of the elbow (humeroulnar), and extension with abduction and internal rotation of the hip. In close packing the joint is stable and behaves almost like a single solid unit, which is why it is a common position of injury when a large force strikes, and why it is a poor position for testing joint play.

The loose packed or resting position is where the capsule is most lax and joint play is greatest, typically somewhere in mid range (about 25 degrees of knee flexion, or about 55 degrees of abduction with slight horizontal adduction at the shoulder). Effusion pushes a joint toward this position because it gives the capsule the most volume, so a swollen knee is held slightly bent. Clinicians usually test accessory motion and begin gentle mobilization in the resting position.

Assessing joint play

To test joint play, the examiner stabilizes one bone, places the joint in its resting position and applies a gentle glide, distraction (separation) or compression to the other bone, noting both the quantity of movement and the quality of resistance. Kaltenborn proposed a simple 0 to 6 scale:

In the spine, chiropractors, osteopathic physicians and physical therapists assess segmental motion with static palpation (feeling for tenderness, position and tissue texture), motion palpation (feeling how a segment moves as the patient or examiner moves the spine), and posterior to anterior (PA) spring testing, in which the examiner presses on a spinous or transverse process and judges stiffness and pain. Research on these methods gives a clear, consistent message. Pain provocation (does pressing here reproduce the patient's pain?) has fair to good agreement between examiners. Judgments of motion and position alone have poor to fair inter-examiner reliability, with kappa values often below 0.4, and examiners frequently disagree even about which vertebral level they are touching. These findings do not make palpation useless, but they argue for combining it with history, pain response and regional movement testing rather than treating a single palpation finding as a diagnosis.

End feel

The orthopedic physician James Cyriax taught that the sensation the examiner feels at the end of passive range, the end feel, tells you what tissue is stopping the motion. A normal joint stops in a way that fits its anatomy; a different end feel, or a normal end feel at the wrong point in the range, suggests a problem.

The capsular pattern

Cyriax also observed that when the whole capsule of a joint is inflamed or contracted (as in arthritis or adhesive capsulitis), passive motion is lost in a characteristic proportion for that joint, called the capsular pattern. A loss of motion that does not follow the pattern, a noncapsular pattern, points to something more localized, such as a ligament adhesion, an internal derangement or an extra-articular lesion.

Section 3Cavitation: what the pop is

An old question with a modern answer

People have cracked their knuckles, necks and backs for as long as anyone has recorded, and for most of that history the sound was explained by guesses: bones snapping back into place, ligaments flicking over bone, or tendons snapping. The first careful laboratory work came in the late 1940s, when researchers pulled on finger joints while taking x-rays and found that a clear space appeared inside the joint at the moment of the crack. In 1971, Unsworth and colleagues repeated the experiment with better instruments and gave the explanation that dominated textbooks for decades: a sudden drop in pressure inside the joint causes gas dissolved in the synovial fluid to come out of solution, and the collapse of that bubble makes the sound. They also showed that once a joint has cracked, it cannot crack again for roughly 20 minutes, a refractory period during which the gas slowly redissolves.

The word for this process is cavitation: the formation of a vapor or gas filled cavity in a liquid when local pressure falls low enough. In joints the term is used for the gas event that produces the audible release, and many clinicians simply call the sound a "cavitation" or an "audible release."

The mechanics step by step

Synovial fluid normally sits at slightly below atmospheric pressure, and a healthy joint holds very little of it, spread as a thin film between two closely fitting surfaces sealed inside a capsule. Imagine pulling the two surfaces apart:

1. Initial separation. As force is applied (a pull on a finger, or a thrust on a spinal segment), the surfaces begin to separate. The capsule is a closed container, and the volume inside it increases, but no fluid can rush in from outside to fill the space.
2. Pressure falls. Because the volume increases while the amount of fluid stays the same, pressure inside the joint drops sharply. Measurements in finger joints have recorded strongly negative pressures just before the crack.
3. Gas comes out of solution. Synovial fluid contains dissolved gases, mostly carbon dioxide along with nitrogen and oxygen. When pressure drops below a threshold, a gas cavity suddenly forms, a process called tribonucleation when it occurs as two wetted surfaces are pulled apart.
4. Rapid separation and sound. The joint surfaces jump apart a little further, energy is released abruptly, and an audible crack is heard. On x-ray, a radiolucent (dark) gas shadow appears in the joint space.
5. Refractory period. The gas bubble persists for some time and slowly redissolves. Until it does, the joint cannot be cracked again by the same maneuver. Times of about 20 to 30 minutes are commonly cited.

Formation or collapse? The imaging evidence

Unsworth's model held that the sound came from the collapse of the bubble. That idea had a problem: the bubble can be seen on x-rays taken after the crack, so it clearly had not fully collapsed. In 2015, Kawchuk and colleagues filmed finger joints being pulled inside a magnetic resonance imaging scanner at high frame rates. The cavity appeared at the instant of the crack and remained afterward, supporting the view that the sound is linked to cavity formation (rapid separation and bubble inception) rather than to collapse. In 2018, a mathematical model by Chandran Suja and Barakat suggested that the sound could be produced by partial collapse of the bubble within milliseconds of its formation, leaving a smaller residual bubble behind.

The fairest summary today is this: the audible release is tied to the sudden creation of a gas cavity as the joint surfaces separate, and whether the acoustic energy comes from the formation itself, from rapid partial collapse, or from the surfaces snapping apart is still being studied. What is not in doubt is what the pop is not. It is not a bone being put back into place, and it is not a sign that something has broken.

Cavitation versus other joint sounds

Not every joint noise is cavitation. Clinicians distinguish several kinds:

Does knuckle cracking cause arthritis?

This is one of the most common questions a clinician hears. The best available evidence says no clear link. Cross sectional studies comparing habitual knuckle crackers with non crackers have not found more hand osteoarthritis in the crackers. A famous self experiment by a physician who cracked the knuckles of only one hand for about 60 years found no arthritis in either hand; it is a single person, so it proves little, but it fits the larger studies. The honest answer to a patient is that habitual knuckle cracking does not appear to cause arthritis, although a joint that cracks with pain, swelling or locking deserves examination.

Cavitation during spinal manipulation

A high velocity, low amplitude thrust to a spinal segment often produces one or more audible releases, and both patients and clinicians tend to value the sound as a sign that "something happened." Research tempers that belief in three ways.

• The pop is not always at the target. Studies that placed accelerometers on the skin over several vertebrae found that cavitations during thoracic and lumbar manipulation frequently came from joints one or more levels away from the intended segment, and that a single thrust often produced several cavitations. Spinal manipulation is less segment specific than many practitioners assume.
• Outcomes do not depend on the sound. Clinical trials in people with low back and neck pain have compared patients whose manipulation produced an audible release with those whose manipulation did not. Improvement in pain and function was similar in both groups. The sound appears to be a byproduct rather than the active ingredient.
• The pop does carry meaning for people. Patients often report feeling looser and relieved when they hear it, and this expectation can contribute to the overall response through the same mechanisms that shape any treatment experience (Chapter 4 discusses these nonspecific effects). Clinicians can respect that experience while explaining that the sound is not required.

Cavitation does produce a brief, measurable change. In the finger, joint separation and passive range of motion increase slightly after cracking and the joint feels less stiff for a time, consistent with the gas bubble occupying space and the capsule being slackened.

Section 4Tissue response to injury, healing phases

One process, three overlapping phases

Whether a sprained ligament, a strained muscle or a cut in the skin, vascularized connective tissue heals through the same broad sequence. The phases overlap, and their timing depends on the tissue, the size of the injury, the person's age and health, and how the area is loaded during recovery. The times below are typical teaching values for a moderate soft tissue injury, not fixed deadlines.

The inflammatory phase

Injury tears blood vessels and cells. Within seconds, damaged vessels constrict and platelets aggregate and trigger the clotting cascade, forming a fibrin clot that plugs the leak and serves as a scaffold for incoming cells. Platelets release growth factors such as platelet derived growth factor (PDGF) and transforming growth factor beta (TGF beta) that recruit and activate repair cells. After the brief vasoconstriction, nearby vessels dilate and become leaky under the influence of histamine, bradykinin and prostaglandins.

White cells arrive in a predictable order. Neutrophils dominate the first 24 to 48 hours, killing microbes and clearing debris. Macrophages, derived from blood monocytes, take over by about 48 to 72 hours. They engulf dead tissue and release cytokines and growth factors that switch the area toward repair; without macrophages, healing stalls. The classic signs of inflammation follow directly from these events: rubor (redness) and calor (heat) from increased blood flow, tumor (swelling) from leaky vessels, dolor (pain) from chemical sensitization of nociceptors and mechanical pressure, and functio laesa (loss of function) from pain and swelling together.

Inflammation is the first step of healing, not its enemy. Prostaglandins and other mediators sensitize nearby nociceptors (peripheral sensitization), so ordinary movement hurts and the area is protected in the early days. Problems arise when inflammation is excessive or does not resolve.

The proliferative phase

From about day 3, fibroblasts migrate into the clot along fibrin strands and begin producing new extracellular matrix: glycosaminoglycans, fibronectin and, above all, type III collagen, a thinner, more elastic collagen that can be laid down fast. New capillaries sprout into the area under the influence of vascular endothelial growth factor, and the combination of new vessels, fibroblasts and loose matrix forms granulation tissue, which looks red and granular in an open wound. Some fibroblasts become myofibroblasts, cells with contractile actin filaments that pull the wound edges together.

The new collagen is laid down somewhat randomly, and its cross links are weak. The repair is therefore fragile at first, but tensile strength rises steeply over these weeks. Gentle, progressive loading in this phase helps fibroblasts align new collagen along the direction of stress, which produces a more functional scar.

The remodeling phase

Over months, the scar is reorganized. Type III collagen is gradually replaced by thicker, stronger type I collagen; enzymes called matrix metalloproteinases break down disorganized fibers while fibroblasts lay down new ones; and covalent cross links between collagen molecules mature. Cell numbers and capillaries decline, so the scar becomes paler and less vascular. Fibers increasingly line up with the forces the tissue experiences, which is why mechanical loading during this phase matters so much. In skin, a scar reaches only about 20 percent of normal strength at 3 weeks and roughly 70 to 80 percent by about a year, and it rarely regains full strength. Ligaments follow a similar course: a healed ligament is often longer, weaker and less stiff than the original for a year or more.

Not all tissues heal alike

The main reason tissues differ in healing is blood supply. Repair cells, oxygen and nutrients all arrive by blood, so well vascularized tissues heal faster and more completely.

Notice that bone is the exception that regenerates rather than scarring: once remodeled, a healed fracture can be as strong as the original bone. Notice too that the disc is one of the poorest healers in the body, because its adult nucleus and inner annulus depend on diffusion through the vertebral endplates. A herniated disc often becomes less painful as the herniated fragment is broken down by macrophages, inflammation settles and the nervous system calms, not because the annulus has regrown to its youthful state.

Managing the phases

Advice for acute soft tissue injury has evolved. The old RICE formula (rest, ice, compression, elevation) emphasized rest. It was followed by PRICE (adding protection) and then POLICE (protection, optimal loading, ice, compression, elevation), which replaced rest with the idea of loading the tissue just enough. A more recent framework, PEACE and LOVE, recommends protection, elevation, avoiding anti-inflammatory modalities in the first days, compression and education for the acute stage, followed by load, optimism, vascularization (pain free aerobic activity) and exercise. It reflects concern that suppressing early inflammation could blunt repair signals, and evidence that progressive loading improves tissue quality. The evidence for ice and for routine anti-inflammatory drugs after soft tissue injury is limited and mixed; they can help with pain, but they are not essential to healing.

Section 5Immobilization, adhesion and remodeling

Connective tissue obeys its loads

Living tissues constantly adjust their structure to the forces placed on them. For bone this idea is known as Wolff's law: bone is laid down where it is stressed and resorbed where it is not. The parallel principle for soft tissue is sometimes called Davis's law: ligaments, tendons and fascia that are kept under appropriate tension lengthen and strengthen, while those left slack shorten and weaken. The cellular basis is mechanotransduction, the process by which cells convert mechanical signals into chemical ones. Fibroblasts, chondrocytes, osteocytes and muscle fibers sense stretch, compression and fluid flow through their cell membranes, integrins and cytoskeleton, and respond by changing which genes they express and how much matrix they build or break down.

This is why both too little and too much load are harmful. Moderate, varied, progressive loading keeps tissues healthy and guides healing. Excessive load damages tissue. Removing load entirely, as in casting, bracing, bed rest or guarding a painful area, sets off a cascade of changes that can outlast the original injury.

What immobilization does to each tissue

Much of what we know comes from animal studies in which one limb was immobilized for several weeks and compared with the free limb; human observations broadly agree.

The common theme is that tissues lose properties quickly when unloaded and regain them slowly when loading resumes. A two week immobilization may need two months or more of rehabilitation to reverse. That asymmetry is the strongest argument for keeping immobilization as short and as partial as safety allows.

How adhesions form

The word adhesion describes abnormal fibrous connections between tissue surfaces that should glide independently. In an immobilized joint, three processes work together. First, without movement, collagen fibers that are normally kept apart by water and proteoglycans come into close contact and form new cross links where they touch, binding fibers that should slide past each other. Second, fibrofatty tissue proliferates in the joint recesses and can bridge from synovium to cartilage. Third, if the joint was injured, the collagen of the repair scar is laid down without the orienting influence of motion, so it forms a disordered mesh that tethers neighboring structures. Motion during healing limits all three by keeping fibers sliding and aligning new collagen along the direction of movement.

In chiropractic theory, the concept of a fixation, a spinal segment that has lost some of its normal motion, has long been linked to the idea of intra-articular adhesions, along with other proposed mechanisms such as meniscoid entrapment and segmental muscle spasm. Laboratory work in rats, in which spinal segments were held still with small implanted devices, has shown that adhesions in the zygapophyseal joints become more numerous and larger the longer a segment is immobilized, supporting the idea that spinal synovial joints respond to disuse like other synovial joints. Whether the stiff segments found on examination in people with back pain contain such adhesions, and whether manipulation breaks them, has not been demonstrated in humans. It is reasonable to call adhesion one plausible, partly supported mechanism rather than an established explanation.

Creep, stress relaxation and why sustained stretch works

Ligaments, capsules and tendons are viscoelastic: their response depends on both how much and how fast they are loaded. Two viscoelastic behaviors matter clinically. Creep is continued elongation under a constant load: hold a gentle stretch and the tissue slowly lengthens. Stress relaxation is the gradual fall in tension when tissue is held at a constant length. Both explain why a stretch held for 30 seconds or longer gains more range than a brief bounce, and why sustained or repeated mobilization can restore motion to a stiff capsule.

Permanent lengthening of a contracted capsule requires more than temporary viscoelastic change. It requires remodeling: repeated, tolerable loading over weeks, so that fibroblasts break down old cross linked collagen and lay down new fibers at a longer resting length.

Remodeling: putting load back on purpose

Because tissues remodel according to load, rehabilitation is essentially a program of controlled mechanical signals, sometimes called mechanotherapy. Key principles include:

• Early, protected motion. In the 1970s and 1980s, the surgeon Robert Salter showed in animals that continuous passive motion after joint injury improved cartilage healing and reduced adhesions compared with immobilization. Modern postoperative care, from knee replacement to tendon repair, favors early, controlled movement whenever the repair can tolerate it.
• Progressive loading. Load is increased gradually across weeks, matched to the healing phase, so that the tissue is challenged but not overwhelmed. Tendon rehabilitation programs using slow heavy loading or eccentric exercise are good examples.
• Specificity. Tissues adapt to the specific type of load. Collagen aligns along the direction of tension, bone thickens where it is compressed and bent, and cartilage thrives on cyclic compression. Rehabilitation should load tissues in the directions they will need in daily life.
• Time. Remodeling continues for months. Feeling better does not mean the tissue has reached full strength, which is one reason reinjury is common when people return to full activity too quickly.

Words worth knowing