RightRep EN · PT Open the app
BLOG

Can You Train With a Labral Tear? The Hip Tenant Audit: Where Register's Blinded 3-Tesla Census Found Labral Tears in 69% of Hips That Had Never Hurt and Schmitz's Optimized Protocol Pushed the Silent Count Past 80%, Where Tresch's Matched Comparison Put Labral Defects in 44% of Pain-Free Hips Against 61% of FAI Patients — a Difference That Never Reached Significance — and Kopec's Population Survey Found FAI Syndrome in Just 3.0% of Adults While Morphology Refuses to Correlate With Pain, Where the Lancet's UK FASHIoN Trial Saw Physiotherapy Alone Add 14.1 iHOT-33 Points and Surgery 19.6 — a 6.8 Edge Just Over the 6.1 MCID at the Price of an Operation — and the 21-Trial Meta Rules the Pooled Surgical Edge Below Clinical Importance, Where Ganz's Motion-Not-Loading Thesis Meets Bagwell's Stress Map: Stance Angle, Not Bar Weight, Sets the Rim's MPa (9.6 Control vs 28.4 Large Cam at the Same Squat), Where Diamond's Squat Lab Watched FAI Hips Reach the Same Depth by Another Route, Where Agricola's Growth Window Lays Cam Bone Down Between 12 and 16 and Freezes It at Plate Closure With Tak's Dose-Response at 64% Cam for Four-Times-Weekly Football Before 12, and Ramkumar's Twelve-Year Ledger Sits at 48% vs 28% OA Worsening With 70% of Patients Identical: the Tear on Your Scan Is a Tenant, Physiotherapy Is the Consensus First Door, and Surgery Is a Real Door With a Small Key

By RightRep AI Assistant · · 21 min read

There is a fence at the top of this post and it outranks everything below it: a hip that cannot bear weight, groin pain with fever, night pain that wakes you, unexplained weight loss, or sudden giving-way after a fall is a clinician-this-week file, not a programming question — the Zurich consensus explicitly lists tumours, infections and stress fractures among the conditions to exclude before hip-related pain gets its name [7]. The second fence: a limping child or adolescent with groin pain is a pediatric orthopedic urgency, not this audit — slipped capital femoral epiphysis runs on its own clock and is excluded from every file below [7]. The third fence: acetabular dysplasia and the inflammatory arthritides are different dossiers with different rules — a shallow socket is not a cam bump, and rheumatoid arthritis is not a mechanical impingement; neither borrows this post's numbers [7]. And one honesty clause that outranks both internet camps: no exercise ever re-shaped a bone — the trials below measure pain, function and return to sport, never morphology cure; the cam bump that grew between 12 and 16 stays a cam bump [10], and what training changes is what the bump does to your week, not what it looks like on film. With the fences respected, the question underneath is the gym's quietest fear: the resting script ("there's a tear in the socket — stop squatting before it shreds the cartilage") versus the scope-and-shave script ("get the hip cleaned out, you're back in eight weeks"). The evidence grades both of them down.

The short answer: train — the scan is not the verdict, physiotherapy is the consensus first door, and surgery is a real door with a small key

Strip the file to its spine. A labral tear visible on MRI is one of the worst predictors of hip pain in medicine: in Register's blinded 3-tesla study, 45 volunteers with no history of hip pain — their scans read by three musculoskeletal radiologists who could not know who was symptomatic — carried labral tears in 69% of hips [1]; on Schmitz's optimized protocol the silent count reached 85.7% and 80.9% across two independent reads [2]; and when Tresch compared 63 pain-free volunteers with 63 matched FAI patients, labral defects sat in 44% of the volunteers against 61% of the patients — a difference that never reached statistical significance [3]. The rulebook agrees with the census: the Warwick Agreement requires all three of symptoms, positive clinical signs and imaging findings before the word syndrome enters the file, and lists conservative care and rehabilitation as suitable first treatments [6]; the Zurich consensus adds that incidental labral findings in asymptomatic people "should not be confused" with the disease [7]; and in Brazil, 253 surveyed orthopedists vote 86.2% physiotherapy as initial treatment [16]. The surgical edge is real but small: the Lancet's UK FASHIoN trial — 348 patients, 23 hospitals — had personalised hip therapy add 14.1 iHOT-33 points on its own while arthroscopy added 19.6, an adjusted difference of 6.8 (95% CI 1.7–12.0) that just clears the 6.1 MCID [8]; pooled across 21 randomized trials and 1,799 patients, surgery's advantage on the Harris Hip and iHOT scales stays below the thresholds of clinical importance on every measure [9]. The verdict writes itself: the report in your hand is a tenant registry, not a verdict; run the structured hip program first — and the squat rack is not contraindicated, it is re-grooved (stance, depth, route) according to what your hip tells you [12][13].

Who is asking: the scale file

This is the fifth joint of the orthopedic family this library has opened — the shoulder's rotator cuff tear, the knee's meniscus tear, the spine's herniated disc — and the hip's file has a population spine the others lack. Kopec's Metro Vancouver survey is the cleanest prevalence number in the family: 500 randomly-dialled adults aged 20–49, questionnaire plus examination plus radiographs of both hips, scored on the Warwick triad — FAI syndrome in 3.0% of the population (95% CI 1.5–4.5) [4]. Three in a hundred is a real disease; it is also a tenth of the morphology's shadow, because the same study found the physical signs correlating with symptoms but no significant association between either symptoms or examination and the radiographic findings [4] — the bone's shape and the hip's pain live in different apartments. The Brazilian file matches the international one: when Ejnisman's USP–McMaster survey reached 253 Brazilian orthopedists, 88.2% ranked the anterior impingement sign first on examination, 86.2% chose physiotherapy as the initial treatment, and only 38.8% picked arthroscopy as their surgical door — with the authors' own verdict that the "considerable discrepancy" between respondents is the field's honest self-portrait [16]. The gym audience is the center of this bell: deep-hip flexion sports — squats, Olympic lifts, football, futsal, jiu-jitsu — are exactly where a bump that abuts at end-range announces itself [5][12][13].

What the scan is actually telling you: the tenant file

Start with the finding, because both fear camps treat it as a verdict. It is closer to a tenant. Three census studies, three protocols, one direction. Register's is the cleanest design: 45 asymptomatic volunteers, mean age 37.8, unilateral 3.0-tesla MRI with the scans randomly mixed with 19 from symptomatic patients so the three radiologists could not guess the study's purpose, a finding counted only when two of three agreed. Result: labral tears in 69% of hips, and some abnormality in 73% — chondral defects in 24%, paralabral cysts in 13%, with adults over 35 carrying 13.7 times the odds of chondral defect (95% CI 2.4–80) [1]. Schmitz pushed the protocol harder: 42 asymptomatic hips, mean age 34, on an optimized noncontrast 1.5-tesla sequence — acetabular labral tears in 36 of 42 (85.7%) and 34 of 42 (80.9%) on the two radiologists' reads, interobserver κ = .70, paralabral cysts in a quarter [2]. The authors' conclusion deserves its plain reading: labral tears can occur without symptoms [2]. And Tresch's matched design — 63 volunteers against 63 age- and sex-matched symptomatic FAI patients — is the study that closes the argument, because it found more pathology in the patients but could not make the labrum alone reach significance: 44% of pain-free hips versus 61% of patients, P ≤ 0.12, with both groups' defects parked in the same anterosuperior addresses [3]. Read the three together and the imaging question inverts exactly as it did for the disc bulge and the meniscus tear: a labral tear on a hip MRI is a population-level background finding — present in half to four-fifths of hips that never hurt, depending on how hard the sequence looks — and the report is one witness the examination has to cross-examine. This is why the consensus papers wrote the triad into law: the Zurich group's three families of hip-related pain (impingement syndrome, dysplasia, soft-tissue conditions) are diagnoses of symptoms plus signs plus imaging, and the incidental labral finding in a pain-free person is explicitly not to be confused with any of them [7].

The rulebook door: how the diagnosis actually gets made

The 2016 Warwick Agreement is this joint's constitution: 22 panel members and one patient from nine countries and five specialties, meeting a year after the imaging literature had made "impingement" untenable as a radiographic label. The panel's central act was adding the word syndrome — because the morphology alone is not the disease — and ruling that diagnosis requires appropriate symptoms, positive clinical signs and imaging findings, all three [6]. Suitable treatments named by the same panel: conservative care, rehabilitation, and arthroscopic or open surgery — the order of that list is the order the trial file supports [6]. The Zurich 2018 consensus did the differential work: non-musculoskeletal and serious pathology (tumour, infection, stress fracture, slipped capital femoral epiphysis) excluded first; a negative flexion-adduction-internal-rotation test helps rule the hip out though its positive side is weak; plain radiographs — an AP pelvis and a lateral femoral head-neck view — are the initial imaging of choice, with advanced imaging reserved for when more detail genuinely changes the decision [7]. Kopec's population data then showed why the triad matters so much: with 500 community-drawn adults examined and X-rayed, the signs tracked the symptoms, but neither tracked the radiograph — which is precisely the gap the word "syndrome" was coined to police [4]. Translation for the lifter holding a radiology PDF: the scan names a shape, the examination names a pattern, and only the symptoms name a disease [4][6][7].

The mechanism file: motion, not load — and the stress map that proves it

Ganz's 2003 paper founded the entire field on an observation, not a theory: across more than 600 surgical hip dislocations with in-situ inspection, the damage pattern of the nondysplastic hip mapped onto end-range motion abutment — a motion problem more than an axial-loading problem — with the femoral head-neck junction striking the acetabular rim in flexion and internal rotation [5]. Two decades of biomechanics sharpened that thesis into a coaching manual. Bagwell and Powers built finite-element hips from a 28-year-old volunteer's MRI — a control and three cam severities — and drove them all through the same squat kinematics: average acetabular von Mises stress 9.64 MPa in the control, 9.27 with a minimum cam, 11.36 with a moderate cam, and 28.43 MPa with a large cam, with contact migrating out of the cup onto the anterosuperior rim as the bump grows [13]. Their coaching line is the article's most practical sentence: deep squats in cam morphology deserve caution, and a neutral or externally rotated stance may limit bony abutment at high flexion angles [13]. And Diamond's motion-capture lab kept the honesty balanced in the other direction: fifteen symptomatic FAI patients scheduled for arthroscopy against fourteen matched controls — both groups squatted to statistically comparable depth, unconstrained or constrained; the FAI hips just traveled by another route, with greater pelvic rise and a more adducted hip when the task blocked compensation [12]. Stack those three files and the training principle writes itself: the dose dial for an impinging hip is angle and route, not kilograms — the same bar, same depth, walked through a different corridor [5][12][13]. It is the hip's version of what the deadlift-and-back audit found for the spine: technique is the exposure variable.

The trial door: what UK FASHIoN and the pooled file actually showed

The randomized core of this file is the largest in the family. UK FASHIoN enrolled 348 patients with FAI syndrome — symptoms, signs and imaging, no osteoarthritis — across 23 UK National Health Service hospitals, randomized to hip arthroscopy or to personalised hip therapy: an individualized, supervised, progressive physiotherapist-led program [8]. At twelve months, with 92% follow-up, both arms improved by double digits: iHOT-33 from 39.2 to 58.8 with surgery, from 35.6 to 49.7 with therapy [8]. The adjusted between-group difference was 6.8 points (95% CI 1.7–12.0, P = 0.0093), just over the 6.1 minimal clinically important difference — a real edge, honestly small, bought with an operation under anesthesia [8]. The safety ledger is equally plain: muscle soreness was the most common adverse event in both arms (42% surgical vs 47% therapy — the therapy was dosed like training because it was training), against seven serious adverse events overall and one post-operative joint infection [8]. Then the pooled verdict: Ramadanov's multilevel meta-analysis stacked 21 randomized trials and 1,799 patients (674 conservative, 1,125 arthroscopy) and found surgery ahead by 6.5 Harris Hip points within twelve months (P = 0.016) and 9.8 iHOT points within twenty-four (P = 0.035), with HOOS, HOS and the pain scales showing no significant differences — and none of the pooled edges reaching the MCID thresholds [9]. That is the calibrated reading this post carries: surgery works; therapy works; the marginal difference between them is smaller than either arm's own improvement, and the pooled file cannot call it clinically important [8][9]. Unlike the meniscus file — where placebo surgery exposed the operation's benefit as theater compare the meniscus audit — the hip's operation keeps a real, measurable, modest edge, exactly one honest size.

The long game: twelve years of watching both hips

Because the operation's oldest promise was structural — Ganz's group proposed that early correction might decelerate the degenerative process [5] — the twelve-year ledger matters more here than in any sibling file. Ramkumar found the cleanest natural experiment available: 100 patients with bilateral FAI morphology who had one hip scoped and one left alone, mean follow-up 12.0 years, both hips graded on the same radiographs. The result is a study in calibrated honesty: the non-operative hip advanced to a worse Tönnis grade in 48% of cases, the operative hip in 28% — a 42% relative risk reduction in arthritis progression — but in 70% of patients the two hips ended the decade identical, with the operative hip better in 25% and the non-operative better in 5% [14]. The fastest-degenerating subgroup was impingement with borderline dysplasia — the unstable-shaft-plus-bump combination that belongs to the surgeon's judgment, not the gym's [14]. And for the athletes in the room, Schab's 2026 file compared 98 propensity-matched competitive-athlete hips across arthroscopy generations: contemporary techniques (capsular repair, knotless anchors) returned athletes to sport at 97.6% versus 81.0% for early techniques (P = .03), with better patient-reported outcomes throughout [15]. Two doors, one lesson: the modern operation is very good at returning athletes to their sport [15], and over twelve years it may slow the arthritis clock for a meaningful minority — while leaving most patients' structural trajectory untouched [14]. Neither number is an argument for the scalpel before the program; both are arguments for excellent surgery when surgery is genuinely indicated [8][14][15].

The growth-window door: the file that belongs to coaches and parents

The quietest revolution in this literature has nothing to do with the people holding the MRI report — it explains how they got the bump in the first place. Agricola's prospective cohort radiographed 63 preprofessional soccer players at baseline and again 2.4 years later: in the boys who were 12 and 13 at baseline, a flattened head-neck junction climbed from 13.6% to 50.0% (P = .002); across all hips with an open growth plate, a frank prominence rose from 2.1% to 17.7% (P = .002); and — the finding that drew the map — after the proximal femoral growth plate closed, no significant further increase in prevalence or severity occurred [10]. The cam bone is laid down inside a finite window of skeletal growth, as an adaptation to load the maturing femur could not yet distribute. Tak's retrospective dose-response filled in the exposure: among 63 elite adult players and 126 hips, those who started playing football at high frequency (four or more sessions a week) before age 12 carried cam morphology in 64% versus 40% (P = 0.042), and pathological cam (α angle > 78°) in 30% versus 12% (P = 0.038) [11]. Read together, the files do not say "no sport for children" — they say the growing hip in a high-volume kicking sport is a load-management file like any other: diversify early, watch the years around the growth spurt, and do not hand a 12-year-old a six-day football calendar as if bone were tendon [10][11]. For the adult reading this with a bump already banked: the window is closed, the bone is stable [10], and your lever is the one Diamond and Bagwell handed you — the corridor, not the cargo [12][13].

The props: what stance, depth and machines actually do

Both camps sell props; neither prop is the treatment. Stance is the real programming lever: Bagwell's stress map is the arithmetical case for playing with foot angle — a neutral or externally rotated position at high flexion angles may keep the bump out of the rim's address [13], and it costs one warm-up set to test. Depth is the second lever, and Diamond's data refuses to demonize it: FAI hips reached comparable squat depth when allowed to move naturally [12] — which makes graded depth a tool, not a cage: box squats, partial-range work with progressive depth, a wider-stance low-bar route, and an honest note of where the hip starts talking. Substitution is the honest middle while the hip calms down: machines — leg press, hack squat, belt squat, hip thrust — keep the glutes and quadriceps growing along the glutes file's entire menu, which is rich in hip-extension work that lives nowhere near end-range flexion. And the warm-up ramp earns its keep here more than in most files: the symptomatic hip's guarded patterns loosen with progressive exposure, and the trials' therapy arms were progressive, supervised programs — not cautious motioning [8]. What no prop does: re-shape bone, cushion the rim, or substitute for the twelve-week verdict the program exists to deliver [8][9][13].

So what do you actually do? The practical file

Six rules, every one sourced from the files above:

  1. Night pain, fever, inability to bear weight, a post-fall giving-way, or a limping adolescent go to a clinician before they go to a program. The Zurich consensus lists tumours, infections and stress fractures among the exclusions, and a child's groin pain runs on its own clock [7].
  2. Do not treat the MRI report as the diagnosis. Tears live in 69–86% of pain-free hips depending on sequence [1][2], the labrum alone failed to separate patients from volunteers (44% vs 61%, not significant) [3], and the Warwick triad requires symptoms plus signs plus imaging — all three [6].
  3. Run the structured physiotherapy program first — and log it. It is the consensus first door [6][16], it added 14.1 iHOT-33 points on its own in the Lancet trial [8], and the program is what tells the surgeon whether your hip is one that needs the scope [8][9].
  4. Re-groove the lift instead of retiring it. Same depth is reachable by another route [12]; stance angle, not bar weight, sets the rim's stress [13] — test foot angle and width like you test grip width, and grade depth progressively rather than abandoning it.
  5. If the documented program fails, take the surgical consultation seriously. Surgery's edge is real (6.8 iHOT-33 points, CI 1.7–12.0) [8], modern techniques return athletes at 97.6% [15], and over twelve years the scoped hip of a bilateral patient worsened less often (28% vs 48%) [14] — the anti-surgery script is as dishonest as the surgery-first one.
  6. The bump you already have is stable; the bump in the next generation is programmable. Post-closure cam does not keep growing [10] — but a high-frequency single-sport calendar before 12 triples the pathological-cam odds [11], so coach the twelve-year-old's variety like it matters. It does.

How RightRep fits into this

The labral question is ultimately a trend-versus-scan question — which stance, which depth, which week spoke for this hip — and it is unanswerable without a record:

  • The symptom×load ledger. Every session lands with load, reps and per-rep quality scores; beside a dated symptom note it turns "deep squats sometimes ache" into the exact angle-depth-week curve the physiotherapy program and the surgical consultation actually run on [8].
  • Camera-counted honesty around the re-grooved lift. A stance-changed or depth-graded block only screens properly if the substitute work is real — rep-counted, rep-scored form checks keep the modification from quietly becoming detraining, the same honesty the partial-reps file demands of every range decision.
  • The twelve-week block, tracked like a block. The trial protocol is a structured progressive program with a defined endpoint [8] — the same progressive-overload calendar any tracked mesocycle runs on, with the surgeon's decision parked at the end of it instead of at the start, and the tracking file's whole argument behind it.

The bottom line

  • The scan: labral tears in 69% of never-hurt hips on blinded 3-tesla MRI, 85.7%/80.9% on the optimized protocol; volunteers 44% vs patients 61% on the labrum alone — not significant; FAI syndrome itself in just 3.0% of adults [1][2][3][4].
  • The rulebook: Warwick's triad — symptoms + signs + imaging, all three; therapy named a suitable treatment; incidental findings are not the disease [6][7].
  • The mechanism: motion, not axial load — end-range abutment, with squat stress at 9.64 MPa control vs 28.43 MPa large cam, and stance angle as the dose dial [5][13].
  • The squat lab: FAI hips squat to comparable depth by another route — the corridor is programmable [12].
  • The trial door: therapy +14.1 iHOT-33 alone, surgery +19.6, adjusted edge 6.8 (CI 1.7–12.0) just over the 6.1 MCID; pooled across 21 trials the surgical edge stays sub-clinical on every scale [8][9].
  • The long game: twelve years, bilateral morphology — worse Tönnis grade 48% non-operative vs 28% operative (RRR 42%), yet 70% of patients identical; contemporary-technique athletes return at 97.6% vs 81.0% [14][15].
  • The growth window: cam laid down between 12 and 16, frozen at plate closure; high-frequency football before 12 at 64% cam vs 40%, pathological cam 30% vs 12% [10][11].
  • Brazil: 253 orthopedists, 86.2% physiotherapy-first, 88.2% leading with the impingement sign — the field's own discrepancy on display [16].

Move first, re-groove the corridor, log everything — and let the camera keep the reps honest at rightrep.app.

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

Sources

  1. Register B, Pennock AT, Ho CP, Strickland CD, Lawand A, Philippon MJ. Prevalence of abnormal hip findings in asymptomatic participants: a prospective, blinded study. American Journal of Sports Medicine, 2012;40(12):2720–2724. pubmed.ncbi.nlm.nih.gov
  2. Schmitz MR, Campbell SE, Fajardo RS, Kadrmas WR. Identification of acetabular labral pathological changes in asymptomatic volunteers using optimized, noncontrast 1.5-T magnetic resonance imaging. American Journal of Sports Medicine, 2012;40(6):1337–1341. pubmed.ncbi.nlm.nih.gov
  3. Tresch F, Dietrich TJ, Pfirrmann CWA, Sutter R. Hip MRI: Prevalence of articular cartilage defects and labral tears in asymptomatic volunteers. A comparison with a matched population of patients with femoroacetabular impingement. Journal of Magnetic Resonance Imaging, 2017;46(2):440–451. pubmed.ncbi.nlm.nih.gov
  4. Kopec JA, Hong Q, Wong H, et al. Prevalence of femoroacetabular impingement syndrome among young and middle-aged white adults. Journal of Rheumatology, 2020;47(9):1440–1445. pubmed.ncbi.nlm.nih.gov
  5. Ganz R, Parvizi J, Beck M, Leunig M, Nötzli H, Siebenrock KA. Femoroacetabular impingement: a cause for osteoarthritis of the hip. Clinical Orthopaedics and Related Research, 2003;(417):112–120. pubmed.ncbi.nlm.nih.gov
  6. Griffin DR, Dickenson EJ, O'Donnell J, et al. The Warwick Agreement on femoroacetabular impingement syndrome (FAI syndrome): an international consensus statement. British Journal of Sports Medicine, 2016;50(19):1169–1176. pubmed.ncbi.nlm.nih.gov
  7. Reiman MP, Agricola R, Kemp JL, et al. Consensus recommendations on the classification, definition and diagnostic criteria of hip-related pain in young and middle-aged active adults from the International Hip-related Pain Research Network, Zurich 2018. British Journal of Sports Medicine, 2020;54(11):631–641. pubmed.ncbi.nlm.nih.gov
  8. Griffin DR, Dickenson EJ, Wall PDH, et al.; FASHIoN Study Group. Hip arthroscopy versus best conservative care for the treatment of femoroacetabular impingement syndrome (UK FASHIoN): a multicentre randomised controlled trial. The Lancet, 2018;391(10136):2225–2235. pubmed.ncbi.nlm.nih.gov
  9. Ramadanov N, Lettner J, Voss M, et al. Conservative treatment versus hip arthroscopy in patients with femoroacetabular impingement: a multilevel meta-analysis of randomized controlled trials. Bone & Joint Open, 2025;6(4):480–498. pubmed.ncbi.nlm.nih.gov
  10. Agricola R, Heijboer MP, Ginai AZ, et al. A cam deformity is gradually acquired during skeletal maturation in adolescent and young male soccer players: a prospective study with minimum 2-year follow-up. American Journal of Sports Medicine, 2014;42(4):798–806. pubmed.ncbi.nlm.nih.gov
  11. Tak I, Weir A, Langhout R, et al. The relationship between the frequency of football practice during skeletal growth and the presence of a cam deformity in adult elite football players. British Journal of Sports Medicine, 2015;49(9):630–634. pubmed.ncbi.nlm.nih.gov
  12. Diamond LE, Bennell KL, Wrigley TV, Hinman RS, O'Donnell J, Hodges PW. Squatting biomechanics in individuals with symptomatic femoroacetabular impingement. Medicine and Science in Sports and Exercise, 2017;49(8):1520–1529. pubmed.ncbi.nlm.nih.gov
  13. Bagwell JJ, Powers CM. The influence of squat kinematics and cam morphology on acetabular stress. Arthroscopy, 2017;33(10):1797–1803. pubmed.ncbi.nlm.nih.gov
  14. Ramkumar PN, Olsen RJ, Shaikh HJF, Nawabi DH, Kelly BT. Modern hip arthroscopy for FAIS may delay the natural history of osteoarthritis in 25% of patients: a 12-year follow-up analysis. American Journal of Sports Medicine, 2024;52(5):1137–1143. pubmed.ncbi.nlm.nih.gov
  15. Schab AR, Rana K, Wallace I, et al. Outcomes and return-to-sport rates of competitive athletes treated for femoroacetabular impingement syndrome with contemporary hip arthroscopy techniques: a matched comparison to early techniques. American Journal of Sports Medicine, 2026;54(12):2888–2897. pubmed.ncbi.nlm.nih.gov
  16. Ejnisman L, Khan M, Ayeni OR, Bhandari M, Miyahara HS, Vicente JRN. Brazilian orthopedists' opinions and perceptions on femoroacetabular impingement. Acta Ortopedica Brasileira, 2016;24(6):322–326. pubmed.ncbi.nlm.nih.gov