TKA Stiffness and Arthrofibrosis

A stiff knee replacement is one of the most common reasons for patient dissatisfaction after an otherwise technically successful arthroplasty. Stiffness requiring intervention has an incidence of 1-5%, and the working definition is flexion under 90° at 12 weeks after TKA.

Why 90° is the line. Flexion under 90° prevents normal activities of daily living: climbing stairs, rising from chairs, and getting into and out of vehicles. Mild stiffness, flexion of 90-110°, may not significantly affect function. The flexion each task needs:

• Walking - 0-70°
• Ascending stairs - 0-83°
• Descending stairs - 0-90°
• Sitting comfortably - 90-100°
• Rising from a chair - 95° minimum
• Tying shoes - 110-120°

Extension matters as well. Extension deficits over 10° cause limping, anterior knee pain and difficulty with level walking; the grading is under Clinical Assessment.

The expected course. Motion improves gradually with physiotherapy over the first 6 weeks, by which point the knee should have reached 90° of flexion at minimum. It typically plateaus at 110-120° by 12 weeks, and little further improvement is expected at 6-12 months.

Why timing matters. A knee that is not progressing by 6 weeks, or that plateaus under 90°, should be considered for intervention before the 12-week mark. Outcomes worsen significantly when treatment is delayed, so early recognition is critical.

The knee before surgery. Preoperative flexion under 90° is the strongest predictor of stiffness. The other knee-specific factors:

• Flexion contracture over 10° - soft tissue contracture
• Previous knee surgery - scar tissue and adhesions
• Patella baja - extensor mechanism dysfunction
• Chronic effusion - synovial inflammation

Sex. The only sex-specific data on this page found no difference. In 5,414 revision TKAs the odds of needing manipulation were 1.0 for men versus women (p = 0.85). Female sex is widely quoted as a risk factor for stiffness and may yet prove to be one for primary TKA, but no study cited here supports a numerical multiplier, and the study that measured it directly found nothing. Do not quote a figure.

Age. Age under 50 is real and large: after revision TKA the odds of manipulation are 6.5 times higher (p less than 0.0001). Younger patients also carry higher expectations and activity demands.

Other patient factors. These add to the risk:

• BMI over 35 - mechanical disadvantage
• Diabetes - impaired healing, a tendency to fibrosis
• Smoking - delayed healing, fibrosis
• Inflammatory arthritis - rheumatoid and psoriatic arthritis
• Occult infection - persistent inflammation
• CRPS - sympathetic dysfunction impairs rehabilitation
• Poor pain tolerance and psychological overlay

What can be changed. Previous knee surgery, genetic predisposition, inflammatory arthritis and diabetes are not modifiable. Component positioning, gap balancing, joint line restoration, pain control, early mobilisation and psychological factors are, and prevention concentrates on them.

Technical causes. Each technical error has a recognisable mechanism and clinical finding, and a flexion-extension gap mismatch causes overstuffing.

Prevention starts in clinic. Template for component size, plan the osteophyte removal, set realistic goals from the preoperative range, and counsel patients with flexion under 90° about the risk of stiffness.

Prevention in theatre. These measures prevent the mechanical causes of stiffness:

• Restore joint line height to within 5mm of anatomic
• Balance the flexion and extension gaps to within 2mm
• Avoid oversizing, sizing down if borderline
• Set rotation from the transepicondylar axis and Whiteside's line
• Remove all osteophytes, especially posterior femoral
• Test motion on the table: at least 120° of flexion

Rehabilitation after the primary, the other half of prevention, is set out under Postoperative Care.

Kinematics after TKA. The native knee flexes through 0-135°, and a typical TKA through 0-120°. With flexion the femur translates posteriorly by 10-15mm (rollback) and the tibiofemoral contact point moves posteriorly, while patellofemoral tracking is critical for pain-free flexion. Stiffness disrupts these kinematics through adhesions or mechanical impingement.

Where the scar forms. Primary arthrofibrosis targets the periarticular soft tissues with excessive collagen deposition:

• Suprapatellar pouch - the most common site; adhesions here limit flexion
• Posterior capsule - contracture limits flexion, and scar limits rollback
• Quadriceps and patellar tendons - shortening or tethering impairs gliding
• Periarticular soft tissue - inflammation causes global fibrosis

How the scar develops. Arthrofibrosis passes through four phases, and the phase decides what treatment can still achieve.

Inflammatory phase, 0-2 weeks. Surgical trauma triggers an inflammatory cascade with cytokine release (IL-1, TNF-α), fibrin deposition and myofibroblast activation. Clinically there is swelling, pain, and motion limited by effusion and muscle guarding. Prevention at this stage focuses on controlling inflammation and early mobilisation.

Proliferative phase, 2-6 weeks. Fibroblasts proliferate and deposit immature type III collagen, and adhesions form between the synovium and the surrounding tissues. Motion may plateau as the adhesions mature, with a "rubbery" endpoint to passive flexion. This is the window where aggressive physiotherapy is most effective, before the collagen matures.

Maturation phase, 6-12 weeks. Type III collagen is converted to type I, stronger and less elastic, and cross-linking increases the tensile strength of the scar. The endpoint is firm, adhesions are palpable, and motion is progressively lost if untreated. This is the optimal window for manipulation, before the collagen fully matures: maximal gain with minimal fracture risk.

Chronic phase, over 12 weeks. The scar is dense and mature with extensive cross-linking, and heterotopic ossification may develop in the posterior capsule. The endpoint is hard, physiotherapy brings minimal improvement, and the knee may feel like a "bony block" on examination. Manipulation is less effective and carries a higher fracture risk, so arthroscopic lysis or revision is considered.

Joint line height. Elevating the joint line, through excessive distal femoral resection or an undersized femoral component, produces patella baja (Insall-Salvati ratio under 0.8), anterior impingement between the patella and the polyethylene insert, functional tightening of the extensor mechanism, and a mechanical block to flexion at 70-90°. It is a correctable mechanical cause: it needs revision to restore anatomical joint line height and does not respond to manipulation or physiotherapy alone.

Before labelling a stiff TKA "arthrofibrosis", note which part of the arc is lost: flexion, extension, or both. The pattern localises the mechanical cause and tells you which lever to pull at revision, because a knee tight only in flexion has a different problem from one that cannot fully extend.

The lesion. Patellar clunk syndrome is a complication of posterior-stabilised (PS) designs, which have an intercondylar femoral box. A fibrous nodule forms at the junction of the posterior quadriceps tendon and the superior pole of the patella. As the knee actively extends from deep flexion (around 30-45 degrees) the nodule catches on the superior edge of the box and jumps free.

The symptom. A painful, audible or palpable clunk, classically on rising from a chair or climbing stairs. Suspect it when a "catching" or stiff PS TKA has a discrete, reproducible clunk, and distinguish it from the milder, diffuse patellar crepitus. Newer femoral box geometries have reduced its incidence.

Stiffness is classified by cause, by severity and by timing, and each of the three changes what is done.

By cause. Stiffness is primary, an idiopathic fibrotic response, or secondary to a mechanical or inflammatory cause. Primary arthrofibrosis is a diagnosis of exclusion: correctable mechanical and inflammatory causes are always ruled out first.

By severity. Flexion grades the stiffness, and the grade guides management. Type III stiffness often indicates an underlying mechanical problem that needs revision rather than manipulation alone. Flexion over 110° gives near-normal function.

By timing. The time since surgery tracks the maturity of the scar, and it sets the treatment. The 6-12 week window is critical, and outcomes decline rapidly after it.

History. The pattern of onset and pain points towards the cause:

• Onset - gradual or sudden; sudden suggests a component problem
• Progression - improving, static or worsening
• Pain - anterior points to the patellofemoral joint, global to overstuffing
• Functional loss - the specific activities affected
• Compliance - physiotherapy attendance and home exercise
• Previous surgery - prior procedures increase risk
• CRPS symptoms - pain out of proportion, allodynia

Examination. Measure flexion actively and passively and feel the endpoint, hard or soft. Any flexion contracture over 10° is significant, and an extension deficit shows as a limp. Persistent wound drainage suggests infection, a chronic effusion suggests infection or instability, and patellar maltracking suggests a rotation error. Allodynia and temperature change are the neurovascular signs of CRPS.

Measuring motion. Use a goniometer with the patient supine and the knee over the edge of the bed, measuring actively and passively. Record flexion (normal 0-120°) and extension (normal 0°, a positive number meaning a flexion contracture) separately; the arc of motion is flexion minus extension lag. Extension deficit grades gait:

• 0-5° - normal gait pattern
• 5-10° - mild limp, anterior knee pain
• Over 10° - significant gait abnormality

Differential diagnosis. Each cause of restricted motion has a feature that separates it and a test that confirms it.

Radiographs first. AP, lateral and skyline patellar views. The aim is to identify any correctable mechanical cause before planning manipulation or revision:

• Component position and alignment, with rotation on the AP and lateral
• Joint line height - within 5mm of the contralateral side
• Patellar height - an Insall-Salvati ratio under 0.8 is patella baja
• Overstuffing
• Heterotopic ossification in the posterior capsule
• Loosening - radiolucent lines, subsidence

View larger

Bloods if infection is suspected. Infection must be ruled out before any manipulation or revision:

• ESR over 30mm/hr - concerning
• CRP over 10mg/L - concerning at 6 weeks or more after TKA
• White cell count - usually normal even in infection

Aspiration. If the ESR or CRP is raised, or there is clinical suspicion, the synovial fluid is analysed:

• Cell count over 3000 cells/μL - concerning
• Neutrophil (PMN) percentage over 80% - concerning
• Culture - the gold standard, but may be negative
• Alpha-defensin - high sensitivity for infection

CT for planning. CT with metal artefact reduction is most useful when planning revision for component malposition. It assesses femoral and tibial component rotation, identifies anterior bony or soft-tissue impingement, and measures alignment against the mechanical axis.

MRI. Rarely indicated. It assesses the soft tissues when lysis of adhesions is being considered, and the integrity of the extensor mechanism.

Cause before timing. Infection, CRPS and fracture are ruled out first, and imaging looks for a mechanical cause: component malposition, overstuffing, patella baja or a joint-line error. A mechanical cause goes down the revision pathway at any time, because physiotherapy alone will fail and manipulation alone is not relied on. Only when there is none does timing decide the treatment.

Algorithm

View larger

When the first manipulation fails. Failure means a gain under 10°, or motion lost within 2 weeks. The options:

1. Repeat MUA within 2 weeks - success rate 50-60%
2. Arthroscopic lysis at 3-6 months, if there is no mechanical cause
3. Revision TKA if a mechanical cause is identified
4. Accept the limitation if flexion is over 90° and the patient is satisfied

What not to do is continue physiotherapy indefinitely without progress, or manipulate repeatedly after two attempts, with diminishing returns and an increasing fracture risk.

Before theatre. Rule out infection with a recent CRP and ESR, and consider aspiration. Review the imaging to confirm there is no component malposition, and arrange immediate aggressive physiotherapy for afterwards. The ideal timing is 6-12 weeks; avoid manipulation after 6 months.

Consent. Discuss realistic goals for motion and the specific risks: fracture of the supracondylar femur, tibial plateau or patella; loss of some of the gain over time; the need for a second manipulation if the first is unsuccessful; no improvement; the standard risks of a general or spinal anaesthetic; and revision if an underlying mechanical cause is found. The figures are in the complications table.

Anaesthesia. General or spinal, as long as it gives complete muscle relaxation.

Technique. Gentle progressive force is safer than a single aggressive manipulation, and the whole procedure takes 5-10 minutes.

1. Position. Supine on a standard operating table, the leg free-draped to allow full range, the surgeon at the side of the operative knee. An assistant is optional and can provide counter-pressure. No tourniquet, so that the vascular status can be assessed.
2. Examine under anaesthesia. Document passive flexion and extension before manipulating, feel the endpoint (hard means a mechanical block), and palpate for effusion and crepitus. A hard mechanical block means stopping and obtaining imaging, because revision may be needed.
3. Extension first. This step is often overlooked. With a hand under the heel, apply gentle extension force aiming for full extension (0°). With a flexion contracture over 10° a "pop" may be heard or felt as the adhesions release. Avoid hyperextension force, which risks the PCL.
4. Longitudinal traction. With one hand on the proximal tibia and one on the distal femur, traction along the tibial axis distracts the joint and reduces the compression force on the components during flexion.
5. Progressive flexion. One hand stabilises the femur and the other flexes through the proximal tibia with steady force, in 10-15° increments, listening for crepitus or a "pop" as the adhesions break. The goal is at least 120° of flexion; sudden jerking movements risk fracture.
6. Final assessment. Record the maximum passive flexion achieved and cycle through the range several times. Check MCL and LCL stability for laxity after the manipulation, examine for crepitus (component damage), and palpate for effusion or haematoma.

Afterwards. Both knees can be manipulated in the same session if needed. Aggressive physiotherapy starts on day 1, and CPM can be considered, though it is controversial; the protocol is under Postoperative Care.

The complications of manipulation, with their incidence, what raises the risk, and how each is managed:

What predicts a good manipulation. Five factors separate good results from poor ones:

Motion after a successful manipulation. Some loss over time is expected, but most patients keep functional motion over 90° if the initial manipulation succeeded:

• Immediately - average 120° of flexion
• 6 weeks - typically 110-115°, a 5-10° loss
• 3 months - stabilises at 105-110°
• 1 year - 60-70% maintain motion within 10° of their post-manipulation flexion
• 2 years - 30-40% experience further loss, usually still functional over 90°

Function. With flexion over 90° after successful treatment, independent daily activities, stair climbing and driving return, and patient satisfaction is 70-80%. With persistent flexion under 90°, daily activities are significantly limited, assistive devices are needed, driving is unsafe, and 60-70% are dissatisfied.

Satisfaction. Patients value functional improvement over absolute numbers of degrees. What satisfies them is absolute motion over the 90° threshold, a gain of 20° or more (perceived as meaningful), relief of anterior knee pain, the ability to do what they want, and expectations set by realistic counselling before surgery.