Femoroacetabular Impingement (FAI)

Femoroacetabular impingement (FAI) is a mechanical hip disorder: abnormal contact between the femoral head-neck junction and the acetabular rim during hip motion. Repeated impingement damages the labrum and the articular cartilage, and may progress to early hip osteoarthritis.

Two morphologies, usually together. The abnormal contact comes from the femoral side, the acetabular side, or both:

1. Cam - an aspherical femoral head with decreased head-neck offset
2. Pincer - acetabular overcoverage, focal or global
3. Mixed - both components present, and the pattern in 85% of symptomatic FAI

View larger

Who. Symptoms peak at 20-40 years. Cam predominates in males (3:1), pincer is more common in middle-aged females, and prevalence is higher in certain sports (soccer, hockey, dancing). The morphology itself is found in 10-15% of the general population and is often asymptomatic.

Why it matters. FAI is increasingly recognised as a major cause of early hip osteoarthritis in young active patients and may account for 10-15% of early hip OA. By the time of surgery, 86% have a labral tear and 70% have chondral lesions at arthroscopy. The activity limitation is significant in this group and can be career-ending for an athlete if untreated; early recognition and intervention may prevent or delay osteoarthritis, though the long-term data are still emerging.

Risk factors. Symptomatic FAI is more likely with:

• High-impact sport during skeletal development
• Repetitive hip flexion (dancers, martial artists)
• Underlying hip dysplasia or acetabular retroversion
• Previous Perthes disease or slipped capital femoral epiphysis (SCFE)
• Genetic predisposition (familial clustering is observed)

The normal hip. A ball-and-socket joint with exceptional range of motion, built from four parts that matter here:

• Femoral head - spherical, covered by articular cartilage except at the fovea
• Femoral neck - offset from the head, creating a concave waist at the anterosuperior junction
• Acetabulum - a horseshoe-shaped socket covering about 40% of the head
• Labrum - a fibrocartilaginous rim that deepens the socket and seals the joint

Why the offset matters. During flexion and internal rotation the anterosuperior head-neck offset is what lets the femoral neck clear the acetabular rim. Lose it and the neck meets the rim with every deep flexion, producing abnormal contact and shear on the anterosuperior labrum and cartilage. In the normal hip the cartilage sees compression only and the labrum works as a suction seal; in the impinging hip each flexion adds shear microtrauma, the cartilage delaminates progressively, and the end point is full-thickness cartilage loss.

Cam. The head is aspherical (the pistol-grip deformity) and the anterior offset is decreased or absent, so the alpha angle exceeds 55°. Contact lands on the anterosuperior acetabulum, and the damage runs outside-in: the aspherical head strikes the rim first, then shears into the labrum and cartilage like a snowplough, delaminating the cartilage as it goes.

Pincer. The acetabulum covers too much of the head, focally (retroversion) or globally (coxa profunda). Normal acetabular version is 15-20° of anteversion; a retroverted socket produces the crossover sign on the AP pelvis. Contact starts at the anterosuperior rim, but the damage runs inside-out: the rim strikes the femoral neck and levers the head, and the result is a posteroinferior contre-coup labral tear. Chronic rim impingement may also ossify into an os acetabuli.

Position. Cam impinges in flexion and internal rotation, the FADIR position. Pincer is worse in flexion, abduction and external rotation, the FABER position. Mixed hips hurt in more than one position.

View larger

View larger

Femoral Version: The Other Side of the Impingement Equation

Clearance depends on the femur too. Reduced femoral anteversion or frank femoral retroversion brings the neck closer to the anterior acetabular rim in flexion and internal rotation, worsening (or even causing) anterior impingement when the alpha angle is only borderline, and it typically reduces internal rotation on examination. Excessive anteversion does the opposite: more anterior clearance, but a predisposition to posterior extra-articular impingement and anterior microinstability.

Read the two together. The McKibbin combined index is the sum of femoral version and acetabular version. A low combined value indicates global anterior under-clearance and a high value the reverse, which is why neither number is read in isolation: assess femoral version on a CT or MRI rotational profile alongside the acetabular crossover and centre-edge findings. Femoral retroversion is an often-missed driver of persistent anterior impingement and of failure after an isolated cam osteoplasty, and in marked cases a femoral derotation osteotomy, not arthroscopy, is the correct operation.

View larger

Post-SCFE and Post-Perthes Cam: When the Deformity Is Not Idiopathic

Most of what follows concerns the idiopathic cam, a subtle loss of head-neck offset laid down at the anterosuperior junction during adolescent physeal closure. A post-SCFE cam is a different animal, and treating the two as interchangeable is the commonest conceptual error here.

The mechanism is inverted. In an idiopathic cam the head is aspherical. After a slip the head is largely spherical but has moved: the epiphysis translates posteriorly and inferiorly, leaving the anterosuperior metaphysis projecting into the joint. The offending structure is therefore residual metaphyseal prominence, not head asphericity, and it strikes the acetabular rim directly. That is why alpha angles after a slip sit far above the idiopathic range and why internal rotation in flexion is lost so early. It also explains the physical sign, obligate external rotation as the hip is flexed (the Drehmann sign), which reflects the metaphysis having to swing out of the way.

Post-Perthes hips fail differently. Here the problem is rarely a discrete bump. Coxa magna, coxa breva and relative trochanteric overgrowth produce extra-articular impingement between the greater trochanter and the pelvis alongside any intra-articular cam. An arthroscopic osteochondroplasty that ignores the trochanteric side of the equation will not restore clearance, which is why these hips more often need relative femoral neck lengthening and trochanteric transfer.

History. Deep anterior groin pain in 85-90%, activity-related and insidious in onset. Ask what provokes it:

• Prolonged sitting (the theatre sign), getting out of a car, squatting, climbing stairs
• Cutting and pivoting sports, where performance falls off
• Clicking, catching or giving way, which point to a labral tear

Cam and pincer present differently.

Examination. The impingement tests reproduce the patient's pain.

FADIR. Flexion-adduction-internal rotation is the single most sensitive clinical test, at 94-99%. Supine, flex the hip to 90°, then adduct across the body and internally rotate; reproduction of anterior groin pain is positive. The position maximises impingement at the anterosuperior acetabulum. Its specificity is low: it is a good rule-out test.

Range of motion. Internal rotation is the movement FAI restricts most, and external rotation is often preserved, which helps separate it from other pathology.

Gait and posture. Gait is usually normal unless there is severe OA, and antalgic with advanced disease. Patients may sit with the hip abducted and externally rotated to avoid impingement.

The C-sign. The patient cups the hip with thumb and fingers in a C, over the anterior and lateral hip, to show that the pain is deep in the groin and lateral. It is highly specific for intra-articular pathology including FAI and separates it from extra-articular pain such as bursitis or a muscle strain.

Differential diagnosis. Groin pain in a young adult has a long differential.

Two further points. Dysplasia may coexist with FAI or be a separate entity, and a labral tear can be traumatic without any impingement morphology behind it. The extra-articular mimics declare themselves by where the pain is:

• Iliopsoas tendinitis - snapping hip, a different pain pattern
• Greater trochanteric pain syndrome - lateral pain, not groin
• Adductor strain - medial thigh pain

Radiographs first. Three views are mandatory:

• AP pelvis - acetabular coverage, crossover sign, centre-edge angle, Tonnis grade
• Frog-leg lateral - alpha angle and head-neck offset
• False profile - anterior coverage

The measurements that decide management are the alpha angle (greater than 55° on the frog-leg lateral is diagnostic of cam), the centre-edge angle (greater than 40° on the AP suggests pincer overcoverage), the crossover sign of acetabular retroversion, and the Tonnis grade, which must be 0-1 for arthroscopy to be an option.

MRI arthrogram is the gold standard. It measures the alpha angle on axial oblique sequences, the most accurate method; finds labral tears with 90% sensitivity once intra-articular gadolinium is in; assesses the chondral lesions that predict outcome; maps the cam circumferentially; and shows herniation pits, synovial herniation into the femoral neck that marks FAI. The typical picture is:

• Anterosuperior labral tear (the cam pattern)
• Anterosuperior cartilage delamination or defect
• Paralabral cysts where the labrum is torn
• A herniation pit at the anterior femoral neck

View larger

CT is not routine. It is less sensitive than MRI for the labrum and cartilage, and earns its place for:

• Surgical planning in complex acetabular morphology
• 3D reconstruction to map the extent of the cam
• Bony anatomy when MRI is contraindicated
• Precise measurement of version and coverage angles

View larger

View larger

Dynamic fluoroscopy during examination under anaesthesia shows where the hip impinges as it moves, guides the extent of the osteoplasty, confirms clearance after cam resection and picks up additional sources of impingement.

The diagnostic injection. A local anaesthetic injection into the joint under fluoroscopic or ultrasound guidance can be very useful: temporary relief confirms the hip joint as the source, and persistent pain points to extra-articular pathology. It is typically combined with steroid (lidocaine or bupivacaine plus steroid) for therapeutic benefit.

Bloods are generally not needed unless an inflammatory arthropathy is in question:

• ESR and CRP if inflammatory arthritis is suspected
• Rheumatoid factor and anti-CCP if rheumatoid arthritis is suspected
• HLA-B27 if ankylosing spondylitis is suspected

Pre-operative Planning

Consent covers the specific risks of the procedure:

• Failure to improve symptoms - 15-20%, especially with cartilage damage
• Nerve injury - pudendal and lateral femoral cutaneous, 5-8% transient
• Heterotopic ossification - 3-5%, usually asymptomatic
• Infection - under 1%
• Conversion to arthroplasty if severe cartilage damage is found at surgery
• DVT and PE at the standard orthopaedic surgical risk

Equipment is the other half of planning: a hip arthroscopy traction table with a well-padded perineal post; 30° and 70° arthroscopes; curved shavers, burrs and radiofrequency devices; a C-arm positioned for AP and lateral views; suture anchors of 2.3mm or smaller for labral repair; and a high-speed burr and arthroscopic osteotomes for the cam.

Patient Positioning

Supine on the traction table. The well-padded perineal post sits against the medial groin, both feet are secured in traction boots, the contralateral leg is abducted and secured to give the C-arm room, and the operative leg starts in neutral. Padding is where the nerve injuries are prevented: the perineal post for the pudendal nerve, no compression at the ASIS for the lateral femoral cutaneous nerve, the contralateral knee and ankle, and the arms secured across the chest or on boards.

Traction. Apply roughly 25-50 pounds and distract the joint 8-10mm, confirmed on fluoroscopy; adequate distraction is what makes portal placement safe. Monitor it through the case, release periodically, and keep total traction time under 2 hours.

Portal Placement

Anterolateral portal first, under fluoroscopy. It lies at the intersection of a horizontal line from the superior tip of the greater trochanter and a vertical line from the ASIS. A spinal needle confirms the intra-articular position on the image, then skin incision, blunt dissection to the capsule, and the arthroscope sheath and cannula go in. This is the main viewing portal for the central compartment.

Mid-anterior portal second, under direct vision. With the scope looking at the anterolateral capsule from the first portal, a spinal needle comes in outside-in, kept a safe distance from the lateral femoral cutaneous nerve, and the working portal follows by skin incision and blunt dissection. This is the main working portal for labral repair and cam resection.

Additional portals as needed: a posterolateral portal for the posterior labrum and a distal anterolateral accessory portal for peripheral cam resection. Two or three portals are enough for most FAI cases.

Central Compartment Work (Under Traction)

Diagnostic ArthroscopyLabral TreatmentPincer Resection

Look before you cut. The central compartment is examined in a fixed order so nothing is missed:

1. Labrum - the anterosuperior labrum is the usual cam tear site and the posterior labrum is checked for the contre-coup pincer lesion; classify the tear (radial flap, longitudinal, bucket-handle, degenerative) and probe it for an unstable flap
2. Cartilage - map the acetabular damage by the Beck classification; the femoral head cartilage is usually better preserved; look for delamination, the carpet phenomenon of cartilage lifted off bone, and the wave sign that indicates it
3. Ligamentum teres - partial or complete tear; debride if degenerate and symptomatic; not routinely reconstructed
4. Loose bodies - common in FAI from cartilage delamination, in the anterior or posterior recess; remove them

Preserve and repair whenever possible. The labrum is treated according to what the tissue will allow.

Stable partial tear

Treatment

Debridement only

Technique

Shave unstable edges, preserve substance

Outcome

Good outcomes if FAI corrected

Unstable tear, good tissue

Treatment

Labral repair to rim

Technique

Suture anchors, reattach to acetabular rim

Outcome

Best outcomes - restore seal

Degenerative, irreparable

Treatment

Debridement to stable rim

Technique

Remove degenerative tissue only

Outcome

Acceptable if FAI corrected

Calcified, ossified

Treatment

Resection to normal tissue

Technique

Remove calcified portions

Outcome

May limit outcomes

Labral Status	Treatment	Technique	Outcome
Stable partial tear	Debridement only	Shave unstable edges, preserve substance	Good outcomes if FAI corrected
Unstable tear, good tissue	Labral repair to rim	Suture anchors, reattach to acetabular rim	Best outcomes - restore seal
Degenerative, irreparable	Debridement to stable rim	Remove degenerative tissue only	Acceptable if FAI corrected
Calcified, ossified	Resection to normal tissue	Remove calcified portions	May limit outcomes

Repair technique. Anchors of 2.3mm or smaller are placed at the acetabular rim at the chondrolabral junction, with simple or mattress sutures through the labrum to restore the labral seal to the rim; an anterosuperior tear usually takes 2-4 anchors.

Why repair wins. The labrum is a suction seal that maintains negative intra-articular pressure and distributes load, and debridement loses it. In symptomatic FAI, repair gives better patient-reported outcomes, pain and satisfaction than debridement on meta-analysis, with failure and complication rates that did not differ; debridement remains acceptable for irreparable or degenerate tissue.

Find the zone first. Where there is focal or global overcoverage, dynamic examination under fluoroscopy locates the impingement and the rim to be trimmed is marked: typically the anterosuperior rim in focal retroversion, and it may be circumferential in coxa profunda.

Trim. With an arthroscopic burr or osteotomes, resect 3-5mm of rim, keeping the labral attachment where possible or detaching and repairing it, then smooth the rim so no sharp edge remains and confirm clearance on fluoroscopy.

Confirm. Fluoroscopy should show the centre-edge angle reduced to 25-35°, and dynamic testing through the range of motion should show no impingement.

Conservative Resection

Resect conservatively when trimming the acetabular rim. Over-resection creates iatrogenic dysplasia, with instability and accelerated cartilage wear. Aim to normalise coverage (centre-edge angle 25-35°) rather than create undercoverage: it is better to under-resect and revise if needed than to over-resect and cause instability.

View larger

Peripheral Compartment Work (Traction Released)

Release the traction and flex and externally rotate the hip to bring the femoral neck into view. The peripheral compartment lies between the capsule and the neck, and a limited capsulotomy may be needed for access.

Resect the cam. The lesion sits at the anterosuperior head-neck junction. A high-speed burr takes down the aspherical portion of the head, the goal being a restored concave waist at the junction and an alpha angle under 50° on fluoroscopy. The resection typically covers the anterosuperior quadrant, and may extend from the 1 o'clock to the 4 o'clock position on a right hip, with a smooth transition from head to neck and no stress riser; resect conservatively rather than cut a notch into the neck.

Confirm clearance. On the AP view the head-neck offset should be restored; on the lateral the alpha angle should be under 50-55°; and flexing the hip under dynamic fluoroscopy should show no impingement. The arthroscope from the anterolateral portal gives a direct view as well.

View larger

Close the capsule if a capsulotomy was made. Repair with arthroscopic sutures restores capsular integrity, which may reduce postoperative instability; some surgeons leave the capsulotomy open, and the question remains controversial. The reasons for closing it are set out under Complications.

View larger

Closure

Finish as carefully as you started. Release the traction completely, make a final arthroscopic inspection, confirm no retained instruments or debris, and secure haemostasis. Remove the cannulas, close each portal with one or two simple interrupted sutures and apply sterile dressings. A compression dressing, ice and elevation, crutches with partial weight-bearing, and DVT prophylaxis as per protocol complete the case.

Which nerves. The pudendal nerve suffers from perineal post pressure or excessive traction, giving numbness and sexual dysfunction; the lateral femoral cutaneous nerve from portal placement or positioning, giving lateral thigh numbness. Sciatic nerve injury from traction or positioning is rare, and femoral nerve injury very rare.

Heterotopic ossification follows 3-5% of hip arthroscopies and is usually asymptomatic. The risk rises with extensive cam resection, a history of HO, male gender and genetic predisposition, and it is more common in revision cases and after open surgery. Prophylaxis is controversial and not used by every surgeon; the options are indomethacin 75mg daily for 2-3 weeks or single-dose radiation of 7 Gy.

View larger

View larger

Failure to improve is the most common "complication", in 15-20%. The causes:

• Severe cartilage damage at surgery (Outerbridge grade 3-4)
• Incomplete correction of the morphology, leaving residual impingement
• Progression of osteoarthritis despite surgery
• Wrong diagnosis, the pain coming from another source
• Unrealistic expectations
• Poor rehabilitation compliance

Prevention is careful selection (Tonnis 0-1, confirmed morphology, mechanical symptoms), complete correction of both cam and pincer, and realistic counselling.

Hip Microinstability and Capsular Management

What it is. Hip microinstability is symptomatic, extra-physiological motion of the femoral head within the socket without frank dislocation. The hip resists it through three structures, each of which this operation can compromise:

• The capsule, principally the iliofemoral ligament (Y-ligament of Bigelow), the strongest anterior stabiliser limiting external rotation and extension, divided by the interportal or T-capsulotomy used for access
• The labral suction seal, which maintains negative intra-articular pressure and fluid sealing, lost if the labrum is debrided rather than repaired
• The bony architecture, which is why over-resecting the acetabular rim into iatrogenic under-coverage destabilises the joint

Do not trade impingement for instability. Capsular management is therefore part of the operation rather than an afterthought. Many surgeons now repair or plicate the capsule, and in patients with ligamentous laxity, borderline acetabular coverage or revision surgery the threshold for capsular repair or plication and labral preservation, and against aggressive rim resection, is lower; routine closure nevertheless remains debated. The aim is to relieve impingement without destabilising the joint.

Principles. Full recovery takes 6-12 months, and progression is by criteria rather than by the calendar. Weight-bearing is protected for the first 6 weeks, because early full loading stresses the femoral neck; range of motion avoids the impingement positions, because aggressive early motion risks the labral repair; strengthening centres on the gluteals, because hip control is what protects the result. Returning to sport or resuming high-impact activity too early increases the failure risk, because the hip needs full healing first. The therapist needs to know exactly what was done, and persistent symptoms are investigated rather than ignored.

What to expect. Modern hip arthroscopy for FAI gives good to excellent outcomes in about 85% of appropriately selected patients at 2 years, with high satisfaction and significant improvement in pain, function and return to sport. Selection is what moves that number.

Return to sport. Recreational athletes return in 80-90% at 6-12 months and elite or professional athletes reach their pre-injury level in 70-85%, with contact sports possibly lower. The average time is 6 months (range 4-12 months).

Predictors of poor outcome. These patients are counselled about the higher failure risk and the possible need for arthroplasty:

1. Tonnis grade 2-3 osteoarthritis, the strongest negative predictor
2. Severe cartilage damage (Outerbridge grade 3-4 or Beck grade 3-4)
3. Age over 50 years
4. Severe joint space narrowing, under 2mm
5. A worker's compensation claim
6. Isolated pincer morphology, which does worse than cam or mixed
7. Previous hip surgery, revision cases doing worse

Does surgery prevent osteoarthritis? Satisfaction and function are high in the short term (2-5 years) and good outcomes are maintained in most at 5-10 years, but the long-term data are only emerging: at 10 years and beyond, 20-30% may develop OA despite surgery. What decides progression is the severity of cartilage damage at surgery above all, then the completeness of the morphological correction, the age at surgery (younger is better for prevention), and genetic factors and joint loading. Arthroscopy improves symptoms and function and may delay OA; it may not prevent it in every patient, especially with significant cartilage damage at the time of surgery, and studies out to 15-20 years are still lacking.

Revision. Around 10-15% need revision within 5 years, for incomplete initial correction, progression of cartilage damage or adhesions, and revision succeeds less often than primary surgery, at 60-70%.

Conversion to arthroplasty. In the figures usually quoted, 5-10% progress to total hip arthroplasty within 10 years, on average 5-7 years after the failed arthroscopy, with severe cartilage damage at the index operation, older age and incomplete correction the risk factors. Set that against the Menge cohort in the Evidence Base, where 34% converted within 10 years, with older age and a joint space of 2mm or less the independent predictors: the same factors that mark the marginal and poor candidates in the table above.

Global epidemiology:

Major guidelines and consensus, side by side:

Registry evidence:

Unlike arthroplasty, FAI/hip-preservation surgery is tracked mainly through dedicated hip arthroscopy registries rather than the national joint replacement registries. The most informative is the Danish Hip Arthroscopy Registry (DHAR), which reported significant improvements in HAGOS, EQ-5D and pain scores across 2054 procedures (PMID 28630739). National joint replacement registries (NJR, AJRR, AOANJRR, SHAR, NZJR) become relevant only once FAI has progressed to end-stage OA requiring total hip arthroplasty, where they inform implant and bearing selection.

Global practice variation: