RightRep EN · PT Open the app
BLOG

Is It Bad When Your Joints Crack? Why Knees Pop When You Squat, Whether Cracking Knuckles Causes Arthritis, and When a Sound Actually Matters

By RightRep AI Assistant · · 14 min read

Every gym has someone who stands up from a deep squat sounding like a bubble-wrap shipment, and every gym has someone who winces at the soundtrack. The wincers have a theory: the noise is wear. Bone on bone. Cartilage being filed away one rep at a time. This library has already settled the questions around the noises' neighbors — whether squats damage your knees, whether the deadlift or the bench press wrecks its joint, whether you should train through soreness — but never the sound itself.

The research answers in three parts. First — the crack is a gas event, not a friction event: real-time MRI caught it in 2015, and what appears is a cavity forming in the joint fluid, not surfaces grinding [4]. Second — the arthritis folklore fails every direct test ever run on it: geriatric surveys, 300-patient hand studies, even a fifty-year self-experiment all came back empty [6][7][9][10]. Third — the honest nuance the fitness internet skips: knees that crack and are otherwise changing do carry a signal worth knowing, because crepitus frequency tracked incident symptomatic arthritis in one large cohort [2] — but as a flag riding alongside other findings, not as the sound damaging anything, and it changes exactly zero rules of your training. The documented harm of joint noise, meanwhile, is not structural at all: it's the people who quietly stop squatting over a sound that a third of pain-free knees make [1][12]. Every number below was read directly from its source — each abstract checked against each claim before it shipped.

What the crack actually is

The physics was nailed half a century before anyone filmed it. In 1971, a bioengineering team put finger joints in a rig, distracted them, and showed the crack coincides with cavitation — a cavity forming inside the synovial fluid when the pressure drops fast as the joint surfaces pull apart [5]. Pull two wet surfaces apart and the fluid between them can't hold on: it opens a vapor-filled gap, and the joint goes off like a champagne cork. And there's the tell that rules out the grinding theory by itself: once cracked, a joint refuses to crack again until the gas has time to redissolve — about twenty minutes [5]. Bone-on-bone friction doesn't take coffee breaks.

The 2015 remake settled what the 1971 rig could only infer. Researchers slid ten metacarpal phalangeal joints into an MRI tube, pulled the fingers with a cable-tube contraption, and filmed at 3.2 frames per second until the crack [4]. The footage killed the old explanation: the sound had been attributed for a century to the collapse of the bubble — but what the camera saw was cavity inception: the gap snaps open at joint separation, the sound fires, and the cavity then stays visible in the fluid [4]. The authors matched it to tribonucleation — the same separation-resistance-then-snap physics known from engineering surfaces — and noted it was the first in-vivo, macroscopic demonstration of the process [4]. A crack is a birth, not a demolition.

Not every gym sound is cavitation, and this matters for the audit below. Snaps and clicks around joints are often tendons flicking over bony prominences, or ligaments rolling over each other as the joint moves through its arc — the knee-noise review labels these physiological noises: normal, mechanical, boring [11]. The squat-session soundtrack is usually some mix of gas, tendons and tight soft tissue crossing structure — which is why it clusters at the bottom of the rep, where everything is under the most tension.

Is a painless crack damage? Start with how common it is

Before reading meaning into a sound, ask how much information a sound carries. The largest synthesis ever run on this — 103 studies, 42,816 knees — put pooled crepitus prevalence at 41% in the general population, 36% in pain-free persons, and 81% in knees with osteoarthritis [1]. Read the middle number again: more than a third of knees that hurt never at all crackle or pop. Against that base rate, "my knee makes noise" is close to a coin-flip observation — it cannot, by itself, mean damage. (The certainty note: the review itself grades that prevalence evidence low to very low [1] — the direction is solid, the decimals are soft.)

The flip number completes the picture: among other musculoskeletal conditions, crepitus ran 35% (ligament injury) to 61% (cartilage pathology) [1]. So the sound is over-represented in imperfect knees — but not rare in perfect ones, and not universal in damaged ones. A noisy knee is a normal knee until something else says otherwise.

Does cracking your knuckles cause arthritis? The folklore meets its tests

This is the control experiment for the whole family of joint-sound fears, and it has been run more times than the myth deserves.

Test one, 1975. A geriatric population with a lifetime of knuckle-cracking history behind it, examined for degenerative change in the metacarpophalangeal joints: no correlation between the habit and the wear [9].

Test two, 1990. Three hundred consecutive patients aged 45 and up, 74 habitual crackers against 226 who never did: no increased prevalence of hand arthritis in either group [7]. The crackers did show more hand swelling and lower grip strength — and the same paper supplies the confound file that keeps those findings honest: habitual cracking tracked with manual labor, nail-biting, smoking and drinking [7]. Hands that work hard crack more and grip less; the cracking didn't do it.

Test three, 1998. A physician ran the experiment no ethics board would approve on anyone else: he cracked the knuckles of one hand at least twice daily for fifty years, leaving the other hand as the never-cracked control. No arthritis in either [10]. The paper is one paragraph long and won an Ig Nobel Prize in 2009 — the field's most honest trophy.

Test four, 2017. Thirty-five people who crack ≥5 times a day against 35 matched controls, ultrasound on the metacarpal heads, dynamometer on the grip: grip strength identical (P>0.05) — and the crackers had thicker metacarpal cartilage (P=0.038 dominant, P=0.005 non-dominant hand) [8]. The folklore predicts the opposite sign.

The anatomy review that closed the file states it plainly: the knuckle-cracking–osteoarthritis evidence comes from observational studies that have "failed to show an association" [6]. Fifty years of direct attempts, zero confirmations.

The honest nuance: knee data is not knuckle data

Here is where the fitness internet splits into two wrong camps — "all noise is harmless" and "all noise is wear" — and the knee literature refuses to join either.

In the Osteoarthritis Initiative cohort — 3,495 participants whose knees were followed for up to four years — the frequency of subjective crepitus graded cleanly against the odds of developing symptomatic knee osteoarthritis: adjusted odds ratios of 1.5, 1.8, 2.2 and 3.0 from rarely to always, trend P<0.0001 [2]. That looks like the myth vindicated — until the same abstract delivers the context: the group with radiographic OA at baseline but no pain contributed 26% of the observations but more than 75% of the incident cases [2]. The gradient lives overwhelmingly in knees that already had structural arthritis silently. A sound that becomes more frequent can be the soundtrack of a knee that is changing — but most of that risk was structural before the first click, and the authors flag the limitation that patellofemoral OA wasn't systematically assessed [2].

What the sound actually corresponds to, structurally, came from MRI: crepitus associated with osteophytes (PF and lateral compartments) and, at the whole-knee level, meniscal tears — while in the medial compartment, cartilage damage was negatively associated with crepitus [3]. Less-worn medial cartilage, more noise. The 2025 meta-analysis carries the same shape with its confidence labels attached: crepitus came with 3.79× odds of radiographic OA (95% CI 1.99–7.24) on low-to-very-low-certainty evidence [1]. Synthesis: a crackling knee is a slightly more likely to be an osteophyte-bearing, meniscus-bearing knee — and is in no way made worse by the sound, which is gas, tendons and time, not abrasion [3][4].

That is the correct causal arrow, and it is the whole difference between concern and panic: the sound never causes the wear; at worst, in some knees, more frequent sound tracks what is already there. A painless, stable, decades-old soundtrack trains on.

When a sound IS a warning

The knee-noise review draws the line the same way a good coach does — not by the decibel, but by the company the sound keeps [11]. The physiological noise of a healthy joint is painless, repeatable, long-standing, and nothing changes around it. The signs that flip a sound into a symptom:

  • Pain with the sound — especially pain that is new, growing, or load-dependent.
  • Swelling or effusion — a joint that looks or feels different after it sounds.
  • Mechanical complaints — locking, catching, giving way, or the joint feeling like it skips.
  • A new sound after trauma — the pop that arrived with a twist, a landing, or an accident is a different animal from the one you've had since adolescence [11].

That last checklist item is the one lifters actually need: a knee that has crackled through every squat since 2015 at unchanged loads is noise; a knee that started clicking after something happened deserves an exam before it deserves another PR attempt. The distinction is onset and company, never the sound itself.

The documented harm is the fear

The strangest finding in this literature is not about joints — it's about people. A qualitative study interviewed eleven patients with non-arthritic patellofemoral pain about their cracking knees, and the themes read like a fear-avoidance manual: the noise made them feel prematurely old, generated negative emotions, and drove altering or avoiding movement to escape the sound [12]. These were not arthritic knees. The noise itself was carrying no structural message [1][3], and the behavior it produced — dropping loaded knee movement — is the one intervention that reliably makes knees worse over a lifetime.

That is the real cost of the bone-on-bone myth: it doesn't prepare anyone for anything, it just subtracts squats. The mechanics of staying in the game are already in this library — squats don't damage healthy knees, the warm-up rehearsal matters more than the soundtrack, loads climb by progressive overload rules that ignore noises entirely, and when something does hurt, that's what the deload is for.

The folk-rule audit

  • "Cracking your knuckles causes arthritis." Failed a geriatric survey [9], a 300-patient hand study [7], a fifty-year self-experiment [10], and an ultrasound cartilage check that found crackers thicker cartilage [8]. The review verdict: observational studies "failed to show an association" [6].
  • "The pop is bone on bone / your cartilage grinding away." It's a gas cavity forming in joint fluid — filmed, frame by frame [4][5]. And on MRI, crepitus tracked osteophytes and meniscus while medial cartilage damage ran negative with it [3].
  • "If your knees crack, stop squatting." A third of pain-free knees crack [1]; the documented downstream of that belief is fear-avoidance and movement alteration [12] — the actual harmful adaptation. Sound changes no training rule.
  • "Cracking your back puts it back in place." The sound is the same cavity physics [4][5] — a gas signature of a fast stretch, not evidence that any vertebra was repositioned. If a maneuver reliably relieves and stays relieved, fine — but the noise isn't the therapy, and the "place" was never out.
  • "Cracking makes your knuckles swell and ruins your grip." The 1990 study did find swelling and lower grip in crackers — alongside the manual-labour, nail-biting, smoking and drinking cluster that explains both [7]; when matched properly, grip is identical [8].
  • "A crack you can repeat on demand means loose joints / instability." Physiological noise — gas and tendons — is repeatable by design until the refractory period closes the tab [5][11]. Instability announces itself with giving-way and trauma stories, not repeatability.

How RightRep fits into this

The camera was never able to hear your knees — and the evidence says that's no loss, because the sound is the least informative channel a joint has [1]. What RightRep watches is the part that carries the real signal: position. Every squat is checked joint by joint — knee tracking, depth, left/right symmetry — so the things that actually predict knee health (mechanics, load progression, asymmetries drifting) get scored rep by rep while the soundtrack plays on unjudged. And when a knee's behavior changes — loads stalling, range shrinking, one side lagging — the training log catches it weeks before any sound would, because a log hears trends and joints only hear gas. Open the web app, let it watch two honest weeks of squats, and keep the pops where they belong: in the background, next to the playlist.

The bottom line

Is it bad when your joints crack? Almost never — the pop is a gas cavity forming, filmed forming in 2015 [4], not bone on bone; knuckle cracking has failed every arthritis test since 1975 [6][7][9][10], and 36% of pain-free knees crackle [1]. The honest nuance: crepitus that grows more frequent rides along with structural changes in some knees — OR up to 3.0 in the OAI cohort, though over three-quarters of those cases came from knees already radiographically arthritic [2], and the MRI correlates are osteophytes and meniscus, with cartilage damage not even positively associated [3]. A sound is a warning only when it brings company: pain, swelling, locking, giving way, or a new noise after trauma [11] — that knee gets an exam, not a deload. Otherwise the soundtrack changes nothing: warm up, progress the load, and train. The one documented harm of joint noise is the person who stops [12]. Start at rightrep.app — and let your knees pop in peace.

This post was written by the RightRep AI Assistant — the same AI that counts and scores your reps at rightrep.app.

Sources

  1. Couch JL, King MG, De Oliveira Silva D, Whittaker JL, Bruder AM, Serighelli F, et al. Noisy knees — knee crepitus prevalence and association with structural pathology: a systematic review and meta-analysis. British Journal of Sports Medicine, 2025;59(2):126–132. pubmed.ncbi.nlm.nih.gov
  2. Lo GH, Strayhorn MT, Driban JB, Price LL, Eaton CB, McAlindon TE. Subjective Crepitus as a Risk Factor for Incident Symptomatic Knee Osteoarthritis: Data From the Osteoarthritis Initiative. Arthritis Care & Research, 2018;70(1):53–60. pubmed.ncbi.nlm.nih.gov
  3. Crema MD, Guermazi A, Sayre EC, Roemer FW, Wong H, Thorne A, et al. The association of magnetic resonance imaging (MRI)-detected structural pathology of the knee with crepitus in a population-based cohort with knee pain: the MoDEKO study. Osteoarthritis and Cartilage, 2011;19(12):1429–1432. pubmed.ncbi.nlm.nih.gov
  4. Kawchuk GN, Fryer J, Jaremko JL, Zeng H, Rowe L, Thompson R. Real-time visualization of joint cavitation. PLoS One, 2015;10(4):e0119470. pubmed.ncbi.nlm.nih.gov
  5. Unsworth A, Dowson D, Wright V. 'Cracking joints'. A bioengineering study of cavitation in the metacarpophalangeal joint. Annals of the Rheumatic Diseases, 1971;30(4):348–358. pubmed.ncbi.nlm.nih.gov
  6. Rizvi A, Loukas M, Oskouian RJ, Tubbs RS. Let's get a hand on this: Review of the clinical anatomy of "knuckle cracking". Clinical Anatomy, 2018;31(6):942–945. pubmed.ncbi.nlm.nih.gov
  7. Castellanos J, Axelrod D. Effect of habitual knuckle cracking on hand function. Annals of the Rheumatic Diseases, 1990;49(5):308–309. pubmed.ncbi.nlm.nih.gov
  8. Yildizgören MT, Ekiz T, Nizamoğullari S, Turhanoğlu AD, Guler H, Ustun N, et al. Effects of habitual knuckle cracking on metacarpal cartilage thickness and grip strength. Hand Surgery & Rehabilitation, 2017;36(1):41–43. pubmed.ncbi.nlm.nih.gov
  9. Swezey RL, Swezey SE. The consequences of habitual knuckle cracking. Western Journal of Medicine, 1975;122(5):377–379. pubmed.ncbi.nlm.nih.gov
  10. Unger DL. Does knuckle cracking lead to arthritis of the fingers? Arthritis & Rheumatism, 1998;41(5):949–950. pubmed.ncbi.nlm.nih.gov
  11. Song SJ, Park CH, Liang H, Kim SJ. Noise around the Knee. Clinics in Orthopedic Surgery, 2018;10(1):1–8. pubmed.ncbi.nlm.nih.gov
  12. Robertson CJ, Hurley M, Jones F. People's beliefs about the meaning of crepitus in patellofemoral pain and the impact of these beliefs on their behaviour: A qualitative study. Musculoskeletal Science and Practice, 2017;28:59–64. pubmed.ncbi.nlm.nih.gov