Kneecap Stabilisation Surgery: MPFL Reconstruction, Tibial Tubercle Osteotomy and Trochleoplasty
If your kneecap has dislocated more than once, you may have been told that surgery is an option. This page explains MPFL reconstruction, tibial tubercle osteotomy and trochleoplasty: what each corrects, how the choice is made, and what the evidence shows about results, risks and recovery. Physiotherapy and the wider causes of kneecap problems are covered on my page on patellofemoral pain and kneecap instability.
What the MPFL does
The medial patellofemoral ligament (MPFL) is a band of tissue running from the inner side of the thigh bone (femur) to the inner edge of the kneecap (patella). It is the main soft-tissue restraint preventing the kneecap from dislocating laterally, towards the outer side of the knee, in early knee bending. It is commonly torn during a dislocation.
The risk of another dislocation
A first dislocation is not always followed by a second. In a meta-analysis of 17 studies, 33.6% of people dislocated again. The risk was higher with younger age (odds ratio 2.61); open growth plates (2.72); trochlear dysplasia, where the groove on the femur (the trochlea) is shallow or misshapen (4.15); a raised TT–TG distance, where the patellar tendon attaches too far towards the outer side of the shin bone (2.87); and patella alta, a high-riding kneecap (2.38). In the studies that counted risk factors, recurrence was 7.7% to 13.8% with none and 70.4% to 78.5% with three.1
The 2024 European (ESSKA) consensus, which I have summarised separately, puts recurrence after non-operative treatment at 25% or more, and up to 70% in children and adolescents, with about half of patients left with other symptoms such as pain, swelling or giving way.2
When surgery is considered
Many people start with physiotherapy, but recurrent instability often needs surgical stabilisation. I consider MPFL reconstruction for recurrent instability, for persistent instability symptoms with an anatomical risk profile, and for first-time dislocations where the anatomy suggests a high risk of recurrence or there is associated cartilage injury.
After a first dislocation, guidance differs. The British Orthopaedic Association (BOA) standard of 2024 is that surgery solely to stabilise the kneecap should not be offered routinely after an isolated first dislocation, though fixing a significant osteochondral fracture (cartilage and bone knocked off the joint surface) may be considered.3 For patients who have finished growing, the ESSKA consensus suggests non-operative treatment only when the risk of recurrence is low and there is no cartilage injury, and says reconstruction should be considered first for those with several risk factors and ongoing symptoms or an osteochondral injury.2
How the operation is chosen
Surgery is matched to the cause: some patients need soft-tissue stabilisation, some need bony realignment, and some need both. X-rays assess patella position, joint surfaces and alignment, MRI shows cartilage and MPFL injury, and in some cases I request CT to quantify alignment and rotational profile. Three measurements from these images guide the plan:
- Caton-Deschamps index: the height of the kneecap above the shin bone, as a ratio of the length of its joint surface. A ratio of 1.2 or more is regarded as high.4
- TT–TG distance: how far the patellar tendon attachment (the tibial tubercle) sits to the outer side of the centre of the groove. The traditional threshold is 20 mm.4
- Oswestry–Bristol classification: the shape of the groove on MRI, graded as normal, mild (a shallow groove), moderate (a flat groove) or severe (a convex groove, which bulges outwards instead of dipping in). It was developed in the UK to make the severity of trochlear dysplasia simpler to grade and more consistent between assessments than the older Dejour classification.5
None of these thresholds is settled. The ESSKA consensus found no agreed cut-off values,2 and a 2018 American consensus statement described its own advice on which factors to correct as expert opinion, not an evidence-based protocol.4 I regard TT–TG as one piece of the puzzle: it can mislead as a standalone measure, especially with dysplasia, patella alta or a rotational problem in the limb.
When the bony anatomy is favourable, MPFL reconstruction can be performed alone, but if major bony risk factors are present, such as patella alta or significant malalignment, combining procedures tends to be more predictable and durable. The European consensus supports the same approach while acknowledging the lack of clear scientific evidence for it.2
MPFL reconstruction
MPFL reconstruction rebuilds the ligament, usually using a tendon graft, fixed to the inner edge of the kneecap and to the femur where the original ligament attached. The BOA standard for recurrent instability, produced in association with the British Patellofemoral Society (of which I am a member), makes it the soft-tissue procedure of choice when there is no severe trochlear dysplasia, patella alta or major malalignment.6
The position of the graft on the femur matters. The 2018 consensus statement warned that poor placement or excessive tension can produce severe complications,4 and in a systematic review of 76 patients who needed revision surgery, a misplaced femoral tunnel was the most common reason (38.1%), ahead of untreated trochlear dysplasia (18.4%) and patellar fracture (11.8%).7
In a meta-analysis of 23 studies (930 patients, mean follow-up 3.0 years), 92.8% of athletes returned to sport and 71.3% returned to their previous level or better. Instability recurred in 1.9% and the overall complication rate was 8.8%.8 Rates vary widely between studies: a review of 28 studies (1,478 patients) reported persistent pain at the front of the knee in 0% to 32.3% of knees, stiffness in 0% to 20% and failure in 0% to 10.7%.9 A meta-analysis of 110 studies put the incidence of kneecap fracture at 1.17%.10
Tibial tubercle osteotomy
A tibial tubercle osteotomy (TTO) repositions the bony attachment of the patellar tendon on the shin bone: the block of bone is cut, moved and fixed with screws. Moving it down (distalisation) lowers a high-riding kneecap so that it engages the groove earlier; moving it inwards (medialisation) reduces the outward pull. I consider TTO when patella alta is contributing to instability, when a lateralised line of pull is contributing to maltracking or instability, or when there is focal cartilage wear where unloading a specific area may help.
For significant patella alta with recurrent dislocation, the BOA standard is distalisation, often combined with other procedures. Isolated medialisation is not recommended for instability, although reducing excessive lateralisation may occasionally be indicated.6 Published results point the same way. In a meta-analysis of nine studies of MPFL reconstruction performed alone, recurrent instability was more likely with a Caton-Deschamps index above 1.3 (odds ratio 2.72) and with high-grade trochlear dysplasia (3.28), but not with a higher TT–TG distance.11 In 13 studies of patients with a TT–TG distance above 20 mm (386 patients), adding a TTO to MPFL reconstruction made no significant difference to redislocation or complications, although the studies were heterogeneous.12
Across 38 studies of TTO for instability (1,546 patients, mean follow-up 4.2 years), instability recurred in 3% in studies with up to two years of follow-up and 8% in longer ones. Minor complications occurred in 9% and major complications, such as fracture, non-union (the bone failing to heal) or infection needing further surgery, in 3%. The reoperation rate was 16%, mostly to remove the metalwork.13
Trochleoplasty
Trochleoplasty reshapes the trochlear groove to improve kneecap containment. I consider it for recurrent instability where there is significant trochlear dysplasia and the groove shape is a key reason the kneecap dislocates, and I do not perform it for pain alone. Where appropriate it may be combined with MPFL reconstruction and/or TTO in a single operation.
Recommendations differ. UK guidance is that severe dysplasia in skeletally mature patients (those who have finished growing) should be treated with trochleoplasty, by surgeons with demonstrable training and experience in it;6 the American statement called it not often indicated and noted concerns about long-term effects on the cartilage of the groove.4
A meta-analysis of 28 studies covered 1,000 trochleoplasties in 890 patients, followed for 1 to 25 years. Dislocation recurred in 2.4%, instability without dislocation persisted in 8%, stiffness occurred in 7%, further surgery in 17%, and patellofemoral osteoarthritis was reported in 117 of 431 knees (27%). The authors found very large differences between studies and called for randomised and comparative studies.14 In a review of ten uncontrolled studies of severe dysplasia (407 knees), redislocation or subluxation (the kneecap partly slipping out of place) followed 7% of MPFL reconstructions performed alone and 2.1% of trochleoplasties combined with other procedures.15
Children and teenagers who are still growing
After a first dislocation in a patient who is still growing, the ESSKA consensus is that surgery should be discussed with the patient and family in every case, because recurrence rates are high.2 For recurrent instability, the BOA standard advises rehabilitation first and, if surgery is necessary, procedures that preserve the growth plates, carried out by surgeons used to performing them.6
Two things change while the growth plates are open. The MPFL attaches to the femur close to the growth plate, so techniques for fixing the graft have been adapted to protect it.16 Operations on the tibial tubercle or the trochlea are generally not recommended while the growth plates are wide open.2 In a systematic review of 132 MPFL reconstructions in skeletally immature patients (mean age 13.2, mean follow-up 4.8 years), redislocation occurred in 3.8% and subluxation in 11.4%, and no premature growth plate closure was reported.16
Recovery
Rehabilitation is essential, and the plan depends on which procedure (or combination) is performed. All have protective phases early on, followed by progressive strengthening and functional retraining. After a TTO I commonly use a brace in the first 6 weeks, because pain and swelling can temporarily inhibit the quadriceps and make the knee feel as if it may give way. I individualise the programme based on anatomy, healing and goals, and work closely with specialist physiotherapists who have expertise in patellofemoral rehabilitation.
In the meta-analysis of 23 studies quoted earlier, athletes returned to sport at a mean of 6.7 months, and the rate of return did not differ significantly between those who also had an osteotomy (86.9%) and those who did not (95.4%).8 Those are published figures; my own protocol is outlined below.
My rehabilitation protocol in outline
This is my general protocol. Its time scales are approximate, and the plan for each patient is set out in the notes I write after the operation and adjusted by me and the physiotherapist. I generally suggest cold compression in the early weeks to help control swelling. Pain along the graft, or over the osteotomy, is common for up to 4 months.
After MPFL reconstruction alone:
- Brace and crutches: no brace, and full weight through the leg from the outset. Crutches are used for the first 2 weeks, then put aside as pain and control of the thigh muscle (the quadriceps) allow.
- Things to avoid: resisted hamstring exercises for 8 weeks, because I typically use the patient’s own hamstring tendon as the graft; straightening the knee against resistance with the foot free for at least 6 weeks, because it overloads the kneecap joint; impact work and deep squats or lunges for 12 weeks; breaststroke until 3 months.
- Driving and work: driving is typically from 4 weeks, once able to control the car (later for the right leg or the clutch leg than for the left). Desk work is typically from 4 to 6 weeks, a job on the feet all day from 2 months and manual work from 12 to 16 weeks.
- Exercise bike and swimming: an exercise bike once the knee bends to 100 degrees with minimal swelling; front crawl and pool walking from 6 weeks.
- Running and sport: running is typically from 16 weeks and sports training, non-contact first, from 6 months, each once the criteria are met. Before sports training these include hop tests and quadriceps and hamstring strength at 90% or more of the other leg, and confidence in the knee. I suggest a return to sport at 6 to 9 months.
When a tibial tubercle osteotomy is part of the operation, the cut bone has to heal, and for the first 4 months the osteotomy, not the graft, sets the pace. The outline above applies with these changes:
- Brace and crutches: full weight is still allowed as comfort permits, but with crutches and a hinged brace for walking until 6 weeks. The brace is locked straight for the first 2 weeks, then opened to allow 0 to 90 degrees of bend once control of the quadriceps has returned; it comes off at night and for exercises. During exercises in these 6 weeks the knee is not pushed beyond 100 degrees of bend.
- Bone healing: I review the knee with an X-ray at 6 weeks and at 4 months, when the osteotomy should have united. Impact activity, running and straightening the knee against resistance with the foot free all wait until an X-ray has confirmed union.
- Driving, work and the exercise bike: driving is typically from 6 to 8 weeks, once out of the brace with good muscle control, and a job on the feet all day from 2 to 3 months. Manual work, typically from 12 to 16 weeks, also waits until the bone has united. The exercise bike starts from 6 weeks, with minimal resistance.
After a trochleoplasty, or when a cartilage procedure is carried out at the same operation, weight-bearing, bracing and the range of movement allowed follow the operation note for that patient, as does the plan after MPFL reconstruction in a patient who is still growing. My return to sport page is written around ligament reconstruction, but it describes the strength, hop and confidence testing that the criteria above refer to.
Risks
All surgery carries risks, and they depend on the procedure performed. Beyond those quantified above, they include infection, blood clots, prolonged swelling, ongoing pain or incomplete symptom improvement, recurrent instability if underlying drivers are not fully addressed, hardware irritation after bony procedures such as TTO, and cartilage wear progressing over time in some knees.
The BOA standard advises caution in recommending stabilisation to patients who also have background chronic pain, as their pain may not improve or could even deteriorate after surgery, and in those with generalised joint hypermobility. It also states that an isolated lateral release is never indicated for patellar instability.6 I discuss the expected benefits, alternatives, likely rehabilitation course and the specific risks relevant to your knee in clinic.
Frequently asked questions
How successful is MPFL reconstruction?
In a meta-analysis of 23 studies (930 patients, mean follow-up 3.0 years), 92.8% of athletes returned to sport after MPFL reconstruction and instability recurred in 1.9%. The overall complication rate was 8.8%. Reported complications include persistent pain at the front of the knee, stiffness and, in 1.17% of cases in a pooled analysis, a fracture of the kneecap.
Do I need surgery after a first kneecap dislocation?
Not always. It depends on whether there is associated cartilage injury and whether your anatomy suggests a high risk of it happening again. The UK standard is that stabilising surgery should not be offered routinely after an isolated first dislocation. A 2024 European consensus advises considering it first where there are several risk factors and ongoing symptoms or an osteochondral injury.
When is trochleoplasty needed?
Trochleoplasty is typically reserved for recurrent instability when the groove shape (trochlear dysplasia) is a major driver. I do not perform it for pain alone. In a meta-analysis of 1,000 trochleoplasties, dislocation recurred in 2.4%, but stiffness occurred in 7%, further surgery in 17%, and patellofemoral osteoarthritis was reported in 27% of the knees with data.
How long does recovery from MPFL reconstruction take?
There is a protective phase early on, followed by progressive strengthening and functional retraining. In a meta-analysis of 23 studies, athletes returned to sport at a mean of 6.7 months after MPFL reconstruction. In my own protocol, which is a general guide with approximate time scales, MPFL reconstruction alone needs no brace, driving is typically from 4 weeks and running from 16 weeks, and I suggest a return to sport at 6 to 9 months, once the criteria are met. After a tibial tubercle osteotomy I use crutches and a brace for walking until 6 weeks, driving is typically from 6 to 8 weeks, and running waits until an X-ray has confirmed that the bone has united.
Can MPFL reconstruction be done in a teenager who is still growing?
Yes. The ligament attaches close to a growth plate, so techniques for fixing the graft have been adapted to protect it. In a systematic review of 132 reconstructions in skeletally immature patients (mean age 13.2), redislocation occurred in 3.8% and no premature growth plate closure was reported.
Assessment in Bristol
I see patients at two private hospitals in Bristol: Nuffield Health Bristol Hospital (The Chesterfield) in Clifton and Spire Bristol Hospital in Redland. Appointments can be made through the contact page, and clinic details are on the locations page.
Related knee topics
- Patellofemoral pain and kneecap instability
- Anterior knee pain
- Fat pad impingement
- Cartilage injury
- Knee osteotomy
References
- Huntington LS, et al. Factors Associated With an Increased Risk of Recurrence After a First-Time Patellar Dislocation: A Systematic Review and Meta-analysis. American Journal of Sports Medicine. 2020. PubMed
- Blønd L, et al. (Part 1); Balcarek P, et al. (Part 2). Management of first-time patellar dislocation: The ESSKA 2024 formal consensus. Knee Surgery, Sports Traumatology, Arthroscopy. 2025. PubMed: Part 1, Part 2
- British Orthopaedic Association. Assessment and Management of First Time Lateral Patellar Dislocation (FTLPD). BOA Standard. 2024. BOA
- Post WR, Fithian DC. Patellofemoral Instability: A Consensus Statement From the AOSSM/PFF Patellofemoral Instability Workshop. Orthopaedic Journal of Sports Medicine. 2018. PubMed
- Sharma N, Brown A, Bouras T, Kuiper JH, Eldridge J, Barnett A. The Oswestry-Bristol Classification. The Bone & Joint Journal. 2020. PubMed
- British Orthopaedic Association. The Surgical Management of Recurrent Patellar Instability. BOA Specialty Standard. 2020. BOA
- Walker M, et al. Femoral tunnel malposition is the most common indication for revision medial patellofemoral ligament reconstruction with promising early outcomes following revision reconstruction: a systematic review. Knee Surgery, Sports Traumatology, Arthroscopy. 2022. PubMed
- Platt BN, et al. Return to Sport After Medial Patellofemoral Ligament Reconstruction: A Systematic Review and Meta-analysis. American Journal of Sports Medicine. 2022. PubMed
- Jackson GR, et al. Complication Rates After Medial Patellofemoral Ligament Reconstruction Range From 0% to 32% With 0% to 11% Recurrent Instability: A Systematic Review. Arthroscopy. 2023. PubMed
- Abelleyra Lastoria DA, et al. The incidence and epidemiology of patellar fractures following medial patellofemoral ligament reconstruction: a systematic review and meta-analysis. Archives of Orthopaedic and Trauma Surgery. 2026. PubMed
- Dandu N, et al. The Presence of a Trochlear Bump and Patella Alta May Predict the Risk of Recurrent Instability After Isolated Medial Patellofemoral Ligament Reconstruction: A Systematic Review and Meta-analysis. American Journal of Sports Medicine. 2025. PubMed
- Guevel B, et al. Does Tibial Tuberosity Osteotomy Improve Outcomes When Combined With Medial Patellofemoral Ligament Reconstruction in the Presence of Increased Tibial Tuberosity-Trochlear Groove Distance? A Systematic Review and Meta-analysis. Orthopaedic Journal of Sports Medicine. 2023. PubMed
- DeNovio AC, et al. Incidence of Complications, Recurrent Instability, and All-Cause Reoperations After Tibial Tubercle Osteotomy for Patellofemoral Instability: A Systematic Review and Meta-analysis of the Literature. Orthopaedic Journal of Sports Medicine. 2025. PubMed
- Leclerc JT, et al. Complications and outcomes of trochleoplasty for patellofemoral instability: A systematic review and meta-analysis of 1000 trochleoplasties. Orthopaedics & Traumatology: Surgery & Research. 2021. PubMed
- Balcarek P, et al. Results of medial patellofemoral ligament reconstruction compared with trochleoplasty plus individual extensor apparatus balancing in patellar instability caused by severe trochlear dysplasia: a systematic review and meta-analysis. Knee Surgery, Sports Traumatology, Arthroscopy. 2017. PubMed
- Shamrock AG, et al. Medial Patellofemoral Ligament Reconstruction in Skeletally Immature Patients: A Systematic Review and Meta-analysis. Orthopaedic Journal of Sports Medicine. 2019. PubMed