Medial parapatellar arthrotomy · the gold-standard exposure for primary TKR
- The operation is for symptomatic tricompartmental knee osteoarthritis (about 90 percent of cases) that has failed conservative care for at least 3–6 months, with Kellgren-Lawrence Grade 3–4 changes and significant functional impairment. There is no single global age cut-off; the decision rests on refractory symptoms, radiographic grade and function.
- The medial parapatellar arthrotomy, made 1cm medial to the patella and tibial tubercle and carried 8–10cm into the VMO, preserves the lateral geniculate artery (the main patellar blood supply) and allows safe patellar eversion.
- Restoration of a neutral mechanical axis (0° ± 3°) with balanced rectangular extension and flexion gaps is the goal. Femoral rotation is set from three landmarks — posterior condylar axis plus 3° external rotation, the transepicondylar axis, and Whiteside's line — and verified with a rectangular flexion gap.
- Internal rotation of the femoral component is the leading technical cause of patellar maltracking. Soft-tissue balancing uses staged releases, checking balance with spacers after each step.
- Major registries (AOANJRR, NJR, AJRR, NZJR) report roughly 95–97 percent survivorship at 10 years. Infection (0.5–1 percent) and aseptic loosening are the leading failure modes.
When & Why
Primary indication. Symptomatic tricompartmental knee osteoarthritis — pain on weight-bearing and activity, stiffness and progressive functional loss — with Kellgren-Lawrence Grade 3–4 radiographic changes, that has failed conservative management for a minimum of 3–6 months (NSAIDs, physiotherapy, weight loss, intra-articular injection, bracing). Surgery is offered before prolonged severe disability sets in. Other indications (together under about 10 percent of cases): 1. Inflammatory arthritis (5–8 percent) — rheumatoid arthritis with end-stage destruction, psoriatic arthritis, ankylosing spondylitis. Often younger patients; optimise DMARDs and minimise steroids beforehand; the PCL is frequently attenuated, favouring a posterior-stabilised design. 2. Post-traumatic arthritis (2–5 percent) — after tibial plateau or distal femur fractures; bone defects may need augments or stems, and the revision rate is higher. 3. Avascular necrosis with collapse (1–2 percent) — spontaneous osteonecrosis of the knee (SONK), corticosteroid-induced AVN, sickle cell disease. 4. Failed osteotomy (1–2 percent) — progressive arthritis after a high tibial or distal femoral osteotomy; altered anatomy may require custom guides or navigation. Contraindications - Absolute: active infection (local or systemic), inadequate bone stock for fixation, an absent extensor mechanism, severe peripheral vascular disease with ischaemia. (Neuropathic arthropathy is a relative contraindication — a hinged TKR may be considered.)
- Relative: previous septic arthritis (ensure eradicated, a minimum of 2 years post-treatment); psychiatric illness affecting compliance; severe obesity (BMI greater than 40 — higher complications, consider weight loss first); active malignancy; immunosuppression (HIV, chemotherapy — higher infection risk).
Active UTI (treat with culture-directed antibiotics first); cellulitis or skin lesions (delay until resolved); HbA1c greater than 8 percent (defer for optimisation); bacteraemia from any source (identify and treat); severe malnutrition (albumin under 30 g/L, total lymphocyte count under 1500).
Pre-operative assessment. Document pain severity and character (mechanical versus inflammatory), functional limitation (stairs, walking distance, night pain), ROM and any flexion contracture, coronal deformity and its correctability, ligament stability (which guides implant choice), extensor-mechanism integrity, and the hip and ankle (exclude referred pain, assess overall limb alignment). Review the skin and previous scars. Radiographic assessment & templating. Standing AP, lateral, skyline patella, and a long-leg (hip-knee-ankle) alignment film. Measure the mechanical axis (normal 0° ± 3°; anatomic axis typically 5–7° valgus), assess bone defects (augments or stems), and template component sizes. Add CT for severe deformity, bone loss or previous fracture; MRI if osteonecrosis or infection is suspected; joint aspiration if infection is suspected (synovial WCC greater than 3000, polymorphs greater than 90 percent). Medical optimisation directly lowers infection and complication risk: HbA1c under 7 percent (ideally under 6.5 percent) in diabetics; smoking cessation a minimum of 6 weeks pre-op; weight loss if BMI greater than 35 (every 5 units of BMI raises infection risk about 10 percent); dental clearance; urology review for chronic UTI; dermatology for psoriasis or eczema; MRSA screening and decolonisation; anaemia correction (Hb greater than 120 g/L).
NICE (UK) recommends referral for joint replacement before prolonged severe disability, while AAOS (US) emphasises shared decision-making with documented failure of non-operative treatment. The lowest registry revision rates sit in patients aged 65–75 years; patients under 55 have higher revision rates (higher activity and longer life expectancy).
The Operation
The goal is to resurface the arthritic knee through the medial parapatellar exposure, restore a neutral mechanical axis with balanced rectangular gaps, set correct femoral rotation, and stabilise the knee through full range — all while protecting the popliteal artery, the common peroneal nerve and the patellar tendon. The exposure and the bone cuts are laid out step by step below (and in depth on the medial parapatellar approach to the knee page).
Operative sequence
- Supine on a standard table; a lateral thigh support stops the leg falling into varus during preparation; a foot support allows knee flexion to 90 degrees and beyond.
- General or spinal/epidural anaesthesia; an adductor canal or femoral nerve block for post-operative analgesia — avoid a complete motor block so the extensor mechanism can be assessed intra-operatively.
- Tourniquet on the proximal thigh with adequate padding; exsanguinate with an Esmarch bandage and inflate to 280–300mmHg (about 100mmHg above systolic BP). Document the time and keep it under 120 minutes where possible.
- An anterior midline incision from 5cm proximal to the superior pole of the patella to 2cm distal to the tibial tubercle (typically 12–15cm).
- A midline incision preserves the blood supply to both medial and lateral skin flaps. Use the most lateral of any previous scars, excise old scar tissue, and avoid parallel incisions less than 5cm apart (skin necrosis risk).
- Raise full-thickness flaps in the plane just superficial to the extensor mechanism, medially to the MCL and laterally to the lateral retinaculum, carrying all subcutaneous fat with the skin to preserve perfusion.
- Excise the prepatellar bursa and achieve haemostasis with electrocautery. Avoid undermining more than 2–3cm from the incision (skin necrosis) and handle the skin edges gently with skin hooks, not forceps.
- The medial parapatellar arthrotomy is the gold-standard exposure for primary TKR worldwide. Begin 1cm medial to the tibial tubercle, extend proximally 1cm medial to the medial border of the patella, then curve obliquely into the quadriceps tendon at the junction of the medial third and central third, carrying 8–10cm into the vastus medialis obliquus (VMO).
- Staying 1cm medial to the patella preserves the lateral geniculate artery, which enters the patella at its mid-lateral border and is the main blood supply to the bone. Adequate proximal extension into the VMO is what permits safe patellar eversion; inadequate extension forces eversion and risks the tendon.
- Evert the patella laterally, externally rotate the tibia, and flex the knee to 90 degrees; place a Hohmann retractor over the lateral femoral condyle to protect the lateral structures.
- Adequate medial release is essential before eversion. If the patella will not evert easily, extend the arthrotomy proximally into the VMO or release a little of the medial capsule from the tibia — do not force eversion. In a stiff knee, dislocate the patella laterally rather than everting it.
- Excise the medial and lateral menisci, the ACL, hypertrophic synovium and the osteophytes from the anterior femur and tibia to clear the anterior compartment and improve visualisation.
- The ACL is always excised. The PCL is retained for a cruciate-retaining (CR) design and sacrificed for a posterior-stabilised (PS) design; a deficient or attenuated PCL mandates a PS or more constrained implant.
- Resect 9–10mm from the less worn condyle (typically the lateral condyle in a varus knee, the medial in a valgus knee).
- Cut at 5–7° valgus to the anatomic axis, which is 0° (perpendicular) to the mechanical axis. An intramedullary (IM) guide enters at the centre of the intercondylar notch, anterior to the PCL origin; use an extramedullary (EM) guide instead if there is femoral deformity, previous femoral fixation, or a long-stem hip replacement.
- Resect 10mm from the less worn plateau (typically the lateral plateau in a varus knee — the worn medial side may then have only 2–4mm removed).
- Cut perpendicular to the mechanical axis (0° coronal) with 3–7° posterior slope to match the native slope (typically 5°). Align the EM guide to the centre of the ankle and preserve the tibial tubercle by staying greater than 1cm medial to it.
- Size the femoral component to the anteroposterior dimension of the distal femur, then set rotation from three landmarks: (1) the posterior condylar axis plus 3° external rotation; (2) the transepicondylar axis (line from the medial sulcus to the lateral epicondyle = true 0°); (3) Whiteside's anteroposterior line (trochlear groove to intercondylar notch — rotation is perpendicular to it).
- Verify rotation with a rectangular flexion gap, parallel to the tibial cut and with equal medial-lateral dimensions. If the landmarks disagree by more than 5° (dysplasia, hypoplasia, previous fracture), prioritise the transepicondylar axis and a balanced flexion gap.
- Complete the anterior, posterior and chamfer cuts via the sizing guide; ensure the posterior condyles are fully resected (a tight flexion gap otherwise).
- Verify there is no anterior notching of the femoral cortex (an oversized component or excessive flexion of the cutting block); a notch greater than 3mm multiplies the supracondylar fracture risk — consider downsizing, augments, or prophylactic fixation.
- Assess the extension and flexion gaps with laminar spreaders or a tensor. Aim for a rectangular flexion gap parallel to the tibial cut, with equal medial-lateral laxity of 1–2mm in extension and flexion.
- Varus knee (medial tightness): staged release of the deep MCL, then the posteromedial capsule, semimembranosus and the posterior oblique ligament — checking balance with spacers after EACH release.
- Valgus knee (lateral tightness): release the iliotibial band from Gerdy's tubercle, then popliteus, the lateral capsule (pie-crust technique), and the LCL only if needed (rare, under 5 percent). Identify and protect the common peroneal nerve before any extensive lateral release.
- Resect to a composite thickness (bone plus prosthesis) of at least 15mm — thinner markedly increases the fracture risk. Aim for symmetric resection parallel to the anterior surface, with a central peg.
- Practice varies: routine resurfacing predominates in North America and Australasia; selective resurfacing is common in the UK and Europe. If not resurfacing, patellar denervation (circumferential electrocautery) plus rim osteophyte excision reduces anterior knee pain within 12 months.
- Insert the trials with an appropriate polyethylene thickness (typically 9–11mm).
- Confirm: ROM of 0–5° hyperextension to 110–120° flexion; equal 1–2mm medial-lateral laxity in extension and flexion; central patellar tracking — the no-thumb test (the patella tracks centrally without manual pressure); no anterior impingement and no flexion lift-off. Do not proceed to cementation until all are satisfactory.
- Pulse-lavage all bone surfaces with 1–2L saline, dry thoroughly with suction and sponges, and use antibiotic-loaded bone cement.
- Cement in sequence: tibial component first (pressurise cement into the surface, impact until fully seated), then the femoral component (pressurise all surfaces — distal, anterior, posterior, chamfers), then the patella. Remove excess cement — especially posterior to the femoral condyles, which blocks flexion — before it sets (5–7 minutes). Maintain reduction until polymerised.
- Insert the polyethylene bearing, reduce the joint, and confirm ROM to at least 120°, medial-lateral stability, no anteroposterior translation, and central patellar tracking through the full arc.
- Deflate the tourniquet and achieve meticulous haemostasis. Close the arthrotomy with strong absorbable suture (0 or 1 Vicryl, figure-of-8), the subcutaneous layer (2-0 Vicryl), and the skin (clips, staples or subcuticular 3-0 Monocryl). Dress with wool and crepe; no drain (RCTs show no benefit and a higher transfusion rate).
- Where it lies
- 10–15mm posterior to the posterior capsule
- At risk when
- Posterior osteophyte removal, PCL release, hyperextension
- How to protect it
- Keep the knee flexed when working posteriorly; retractors placed anteriorly; avoid posterior-capsule penetration
- Where it lies
- 15–20mm from the lateral joint line at the fibular neck
- At risk when
- Valgus correction and lateral releases
- How to protect it
- Correct valgus gradually; avoid excessive lateral retraction; identify the nerve in severe valgus (over 20°); keep the knee flexed during release
- Where it lies
- At the tibial tubercle
- At risk when
- Arthrotomy and patellar eversion
- How to protect it
- Stay 1cm medial to the tubercle; adequate medial release before eversion; atraumatic handling — never force eversion
- Where it lies
- 5–8mm from the medial joint line; superficial and deep layers
- At risk when
- Medial releases
- How to protect it
- Staged subperiosteal release from the tibia; check stability after each step; preserve the superficial MCL where possible
- Where it lies
- 4–6mm anterior to the distal femur
- At risk when
- Femoral sizing and the anterior cut
- How to protect it
- Avoid oversizing the femoral component; correct flexion of the cutting block; verify there is no anterior notching
Vascular injury is rare (0.03–0.2 percent) but catastrophic. Signs are an absent distal pulse, a cold pale foot, and capillary refill greater than 3 seconds. Release the tourniquet, call vascular surgery urgently, explore through a medial popliteal approach, and repair primarily, with an interposition vein graft, or by thrombectomy. Perform prophylactic four-compartment fasciotomies if ischaemia time exceeds 4–6 hours. Limb salvage is 80–90 percent with early recognition but falls sharply with delay.
The usual cause is forced eversion without adequate medial release. If it occurs, repair immediately with suture anchors or transosseous tunnels plus augmentation (Achilles allograft or synthetic mesh). Maintain a 1cm margin from the tubercle with every cut and release, and extend the arthrotomy proximally into the VMO rather than levering the patella.
Internal rotation of the femoral component causes lateral patellar subluxation and anterior knee pain, a trapezoidal flexion gap (tight lateral, loose medial) with instability, and accelerated polyethylene wear. Set rotation off the posterior condylar axis plus 3° external rotation and confirm it against the transepicondylar axis and Whiteside's line, verified by a rectangular flexion gap.
The largest meta-analysis (Ahmed et al, Bone Joint J 2021; 41 RCTs, 2819 patients) found a tourniquet was associated with MORE serious adverse events (RR 1.73) and higher day-1 pain, with NO difference in overall blood loss — only a marginally shorter operative time. Many surgeons now operate tourniquet-free, or use it only for the cementation phase and deflate before closure.
Aftercare & Complications
Multimodal analgesia. A regional block (adductor canal or femoral nerve block, 12–24 hours) plus oral paracetamol (1g QID), opioid as required, and an intra-articular injection in some units. Avoid NSAIDs if ketorolac is used. VTE prophylaxis (agent and duration vary by national guideline). Options include LMWH (enoxaparin 40mg SC daily), a DOAC (rivaroxaban 10mg daily, apixaban 2.5mg BD, or dabigatran), or aspirin for standard-risk patients — NICE (UK) and AAOS/ACCP (US) accept aspirin after RCTs showing non-inferiority. Start 12–24 hours post-op; give for 10–14 days, extended toward 35 days in higher-risk patients. Add mechanical prophylaxis (compression stockings, intermittent pneumatic compression) and early mobilisation. Rehabilitation | Phase | Timing | Focus | Immobilisation / activity | |-------|--------|-------|----------------------------| | 1 | 0–24h | Pain control, VTE prophylaxis, wound care | Sit out, stand with physio; WBAT with a frame (cemented TKR allows immediate full weight-bearing) | | 2 | 1–7 days | Twice-daily physiotherapy | ROM 0–90° by discharge; progress frame to crutches/stick; stairs practice | | 3 | 1–6 weeks | ROM, swelling, wound healing | ROM 0–90° by 6 weeks; stick by 2–4 weeks; clips out at 2 weeks | | 4 | 6–12 weeks | Strengthening, gait re-education | ROM 0–110°; stationary bike from week 3–4; pool once wound healed | | 5 | 3–6 months | Function and return to activity | Maximum ROM achieved (average 0–120°); low-impact sport | Return to activity. Driving at 6 weeks (must manage an emergency stop, check insurance); sedentary work at 6–8 weeks; active work at 3 months; low-impact sport (swimming, cycling, golf, walking) at 3–6 months. High-impact sport (running, basketball) is not recommended — it is thought to accelerate polyethylene wear. ROM milestones. Pre-operative ROM predicts post-operative ROM (a patient at 90° pre-op is unlikely to reach 120°). Continuous passive motion (CPM) is not beneficial per RCT and is no longer used. If ROM plateaus under 90°, manipulation under anaesthesia at 6–12 weeks improves ROM by 20–40° (success 60–80 percent if early); arthroscopic arthrolysis adds 15–30° if MUA fails. Complications
- Recognition
- Early (under 3 weeks): wound drainage, erythema, fever, pain. Late (over 3 weeks): pain, swelling, stiffness, loosening. Synovial WCC greater than 3000, PMNs over 90 percent, positive culture; ESR over 30 and CRP over 10 suggestive; alpha-defensin rapid test sensitivity 97 percent, specificity 96 percent
- Prevention
- MRSA screening and decolonisation; chlorhexidine shower; HbA1c under 7 percent; smoking cessation; dental clearance; treat UTI. Intra-op: prophylactic antibiotics within 60 minutes (cefazolin 2g, or vancomycin 15mg/kg if MRSA or penicillin allergy), redose every 4 hours, laminar flow, minimise traffic, antibiotic cement, meticulous haemostasis, limited tourniquet time, wound irrigation 1–2L
- Management
- Acute (under 3 weeks): DAIR — open debridement, modular liner exchange, five tissue cultures, pulse lavage 3–6L, 6 weeks of IV antibiotics; success 50–60 percent. Chronic (over 3 weeks): two-stage revision (international gold standard; single-stage increasingly used in selected UK and European cases) — explant, antibiotic spacer, 6 weeks IV antibiotics, reimplantation at 6–12 weeks; success 80–90 percent
- Recognition
- Mechanical pain worse with activity; implant migration on serial radiographs (over 2mm subsidence or over 5° alignment change); lucent lines over 2mm at the bone-cement interface; component tilt or subsidence. May be asymptomatic until catastrophic failure
- Prevention
- Optimal cement technique (pulse lavage, pressurise with finger, maintain reduction until set); correct alignment (mechanical axis 0° ± 3° — varus over 3° multiplies tibial loosening); avoid malalignment (femoral flexion over 5°, tibial slope over 7°, rotation errors over 5°); cement all components in most patients
- Management
- Confirm aseptic (aspiration, ESR and CRP, PET if doubt); assess bone loss by AORI grade (Type 1 intact, 2A contained, 2B uncontained, 3 major segmental loss). Revision with stems (bypass the defect by 4 cortical diameters) and augments; increase constraint if soft tissue is lax. Success 85–90 percent at 10 years
- Recognition
- Patient reports the knee gives way or buckles. Flexion instability: tight in extension, loose in flexion (instability sitting or bending). Extension instability: loose in extension, tight in flexion (gives way when walking). Examination: medial-lateral laxity over 5mm, anteroposterior translation, recurvatum over 10°
- Prevention
- Balanced gaps intra-operatively (1–2mm medial-lateral laxity in extension and flexion); correct femoral rotation (three landmarks); avoid tibial slope over 7°; appropriate polyethylene 9–11mm; PS design if PCL deficient; CCK if severe instability or bone loss
- Management
- Exclude infection first (aspiration is mandatory). Flexion instability — thicker poly if under 2mm increase is needed, else revision to a larger femur. Extension instability — thicker poly, distal femoral augment, constrained poly. Global instability (attenuated collaterals) — constrained condylar knee (CCK) or rotating hinge
- Recognition
- Maltracking: anterior knee pain, crepitus, a positive J-sign, failed no-thumb test. Patellar fracture: acute pain after trauma, inability to straight-leg-raise if disrupted. Tendon rupture: a palpable gap, inability to extend, patella alta; quadriceps rupture is similar with patella baja
- Prevention
- Correct femoral rotation (internal rotation is the leading technical cause); central patellar component placement; avoid over-resection under 15mm composite; gentle eversion with adequate medial release first; atraumatic handling. Identify high-risk patients (RA, diabetes, chronic steroids, previous surgery, renal failure)
- Management
- Maltracking: conservative first (physiotherapy, VMO strengthening); lateral release if persistent; severe maltracking needs revision with corrected rotation or tibial tubercle osteotomy. Patellar fracture: non-operative if the extensor mechanism is intact and the component stable; tension-band ORIF if displaced over 2mm or the mechanism is disrupted. Tendon rupture: acute primary repair (under 3 weeks) with augmentation; chronic cases need reconstruction (EHL transfer, allograft)
- Recognition
- Cannot flex past 90° (difficulty sitting, stairs, squatting). Document active and passive ROM. Radiographs assess component position (anterior notching, patellofemoral overstuffing, excessive posterior slope)
- Prevention
- Pre-operative counselling (pre-op ROM predicts post-op ROM); aggressive early physiotherapy (CPM is not beneficial per RCT); early mobilisation; adequate analgesia; avoid component malposition and overstuffing; correct a tight flexion gap at the trial stage
- Management
- Early (under 12 weeks): intensive physiotherapy; MUA at 6–12 weeks if ROM plateaus under 90° (success 60–80 percent if early). Late (over 12 weeks): MUA is less effective (30–40 percent); arthroscopic arthrolysis (15–30° improvement); revision if component malposition is identified
- Recognition
- Intra-op: a crack during component impaction or resection. Post-op: acute pain after a fall, inability to weight-bear. Classify the supracondylar femur by Lewis and Rorabeck, the tibial plateau by Backstein, the patella by Ortiguera; CT if the pattern or bone loss is unclear
- Prevention
- Avoid anterior notching; gentle IM-guide insertion (hand pressure only); correct sizing; support bone during impaction; treat osteoporosis; minimise soft-tissue stripping; fall prevention
- Management
- Supracondylar femur (Lewis and Rorabeck): Type I non-displaced, stable prosthesis — non-operative (hinged brace, WBAT); Type II displaced, stable prosthesis — ORIF with a lateral locking plate, consider a retrograde nail; Type III loose prosthesis — revision with a long-stemmed femoral component (bypass by 2 cortical diameters distal and 4 proximal) or a distal femoral replacement. Tibial and patellar fractures follow the same stable-implant versus loose-implant algorithm
- Recognition
- DVT: unilateral calf swelling (over 3cm difference), tenderness; Doppler confirms (95 percent sensitive for proximal DVT). PE: chest pain, dyspnoea, tachypnoea, tachycardia, hypoxia; CTPA is the gold standard (V/Q scan if contrast is contraindicated)
- Prevention
- Mechanical: graduated compression stockings, intermittent pneumatic compression, early mobilisation. Chemical (agent and duration vary by national guideline): LMWH, a DOAC (rivaroxaban, apixaban, dabigatran) or aspirin for standard-risk patients. Duration commonly 10–14 days, up to 35 days for higher-risk patients
- Management
- Therapeutic anticoagulation — LMWH bridged to warfarin (INR 2–3 for 3–6 months) or a DOAC. IVC filter if anticoagulation is contraindicated or PE recurs despite it. Massive PE with instability — consider thrombolysis (alteplase) or embolectomy. Survival over 95 percent with prompt treatment
Common peroneal nerve palsy (0.5–2 percent overall; 2–5 percent in valgus correction over 20°): document the deficit immediately, remove tight dressings, exclude a haematoma (ultrasound/MRI, decompress if present), and observe — 50–70 percent recover over 3–6 months. EMG at 3 months if persistent; an AFO supports function, and tendon transfers (tibialis posterior) are reserved for a permanent deficit. Vascular injury (popliteal artery, 0.03–0.2 percent): see the safety alert in The Operation.
Viva & Exam Focus
BALANCEBALANCE — principles of soft-tissue balancing in TKR
ROTATIONROTATION — establishing correct femoral component rotation
Clinical Decision Scenarios
Practise clinical reasoning and management decisions out loud
“A 68-year-old woman with end-stage tricompartmental knee osteoarthritis is listed for primary TKR. Walk me through your pre-operative assessment, surgical technique, and how you ensure optimal component positioning and soft-tissue balance.”
“You are 30 minutes into cementing the components of a primary TKR when you notice the patient's foot is pale, cold, and has no palpable dorsalis pedis or posterior tibial pulse. What do you do?”
“A 55-year-old man with a severe valgus deformity (20 degrees) and an incompetent PCL is scheduled for TKR. How would you manage the valgus deformity and what implant would you choose? Walk me through your soft-tissue balancing technique.”
Indication
- Symptomatic tricompartmental knee OA failing 3–6 months of conservative care, K-L Grade 3–4, significant functional impairment
- Inflammatory, post-traumatic, AVN, failed osteotomy together under about 10 percent
- No single age threshold — decision by refractory symptoms, K-L 3–4 and function; registries report about 95–97 percent survivorship at 10 years
Exposure & danger zones
- Medial parapatellar arthrotomy 1cm medial to patella and tubercle, 8–10cm into the VMO
- Popliteal artery 10–15mm posterior to capsule; common peroneal nerve 15–20mm from the lateral joint line
- MCL 5–8mm from the medial joint line; anterior femoral cortex 4–6mm anterior — avoid notching
Critical technical steps
- Distal femur 5–7° valgus (0° mechanical), 9–10mm from the less worn condyle
- Proximal tibia 0° coronal, 5° posterior slope, 10mm from the lateral plateau in a varus knee
- Femoral rotation: three landmarks must agree (PCA plus 3° ER, transepicondylar axis, Whiteside's line)
- Gap balancing: 1–2mm medial-lateral laxity in extension and flexion; staged releases checking after each
Finish
- Patellar composite at least 15mm; central peg
- Trial to ROM 0–120°, stability, central tracking (no-thumb test) before cementing
- Cement tibia then femur then patella; remove posterior cement before it sets; no drain
Complications
- PJI 0.5–1 percent — DAIR for acute (under 3 weeks); two-stage revision for chronic
- Aseptic loosening is the leading aseptic failure mode; revise with stems and augments
- Instability 1–2 percent; stiffness under 90° in 3–5 percent (MUA then arthrolysis)
- VTE: risk-stratified LMWH, DOAC or aspirin; common peroneal palsy 2–5 percent in severe valgus
Exam tips
- Cement all components in most patients; CR versus PS broadly equivalent (one large cohort favoured CR at 15 years)
- Tourniquet increases serious adverse events and early pain with no overall blood-loss benefit (Ahmed 2021)
- Exclude infection first (aspiration) before attributing pain to loosening or instability
- Valgus deformity: staged lateral releases, watch the common peroneal nerve, CCK if residual laxity over 3mm
Background & Evidence
Global epidemiology & registry evidence. The world's major arthroplasty registries (AOANJRR Australia, NJR UK, AJRR USA, NZJR, SHAR Sweden) together capture several million primary TKRs and are broadly concordant: roughly 95–97 percent survivorship at 10 years, falling to about 90 percent by 20 years. Revision rates are lowest in patients aged 65–75 years and higher in patients under 55. The leading reasons for revision are infection, aseptic loosening and instability. Implant design — cruciate-retaining versus posterior-stabilised. Both CR and PS are widely used and registry/RCT data show broadly similar mid-term function and patient-reported outcomes; the choice rests on PCL integrity, deformity, bone quality and surgeon preference.
The most common design in several registries. Preserves the native PCL for proprioception and femoral rollback, resects less bone and eases revision. Requires an intact, functional PCL. May develop PCL wear over time.
Predictable kinematics via the cam-post mechanism; corrects a flexion contracture more effectively; suits a PCL-deficient knee (RA, previous trauma). More bone resection; rare cam-post wear and patellar clunk (1–2 percent).
Varus-valgus constraint via a tall tibial post and deep femoral box. For severe instability, gross soft-tissue insufficiency, significant bone loss, or revision; a rotating hinge is used for the most severe cases.
Know both designs and defend your choice. CR is preferred with a normal PCL and mild deformity. PS is preferred when the PCL is deficient or attenuated, with a severe flexion contracture (over 15°), in inflammatory arthritis, or in revision. Either is acceptable in primary OA with an intact PCL — most comparative data show equivalence, though one large cohort (Abdel et al, JBJS 2011) reported better 15-year survival for CR.
Alignment philosophy — mechanical versus kinematic.
- Mechanical alignment
- Neutral mechanical axis (0° ± 3° HKA)
- Kinematic alignment
- Restore the native joint line and pre-arthritic anatomy
- Mechanical alignment
- Distal femur and proximal tibia perpendicular to the mechanical axis
- Kinematic alignment
- Match the native joint surfaces; may leave constitutional varus or valgus (up to 5–6°)
- Mechanical alignment
- The alignment philosophy underpinning registry survivorship
- Kinematic alignment
- Short-term RCTs show improved ROM, lower pain, better function — but more outliers and no mature long-term data
- Mechanical alignment
- Non-anatomic; may leave a subset stiff or dissatisfied
- Kinematic alignment
- Unproven long-term; concerns about medial or lateral overload and wear
References
The clinical content above is supported by the major joint-replacement registries (AOANJRR Australia, NJR UK, AJRR USA, NZ Joint Registry, Swedish Knee Arthroplasty Register) and the national-society guidance on VTE prophylaxis in major lower-limb arthroplasty (NICE; AAOS/ACCP), where the agent and duration vary by jurisdiction. Key source papers: - Insall JN, Binazzi R, Soudry M, Mestriner LA. Total knee arthroplasty. Clin Orthop Relat Res. 1985;(192):13-22. PMID: 3967412. Classic description of gap balancing and measured resection.
- Whiteside LA, Arima J. The anteroposterior axis for femoral rotational alignment in valgus total knee arthroplasty. Clin Orthop Relat Res. 1995;(321):168-172. PMID: 7497664. Description of Whiteside's line for femoral rotation.
- Krackow KA, Mihalko WM. The effect of medial release on flexion and extension gaps in cadaveric knees. Am J Knee Surg. 1999;12(4):222-228. PMID: 10626913. Biomechanics of staged medial releases and gap changes.
- Yuan MC, Ding ZC, Ling TX, Zhou Z. Patellar denervation with electrocautery reduces anterior knee pain within 1 year after TKA: a meta-analysis. Orthop Surg. 2020;13(1):14-27. PMID: 33354916.
Component malrotation causes patellofemoral complications after TKA
- CT study of 30 knees with isolated patellofemoral complications versus 20 well-functioning controls
- The complication group had excessive combined (femoral plus tibial) INTERNAL component rotation; controls were in combined external rotation
- Severity correlated with the magnitude of internal rotation: small (1-4°) tracking or tilt, moderate (3-8°) subluxation, large (7-17°) dislocation or component failure
- The epicondylar axis and tibial tubercle are reproducible CT and intra-operative landmarks for assessing rotation
2018 Evidence-Based Definition of Periprosthetic Hip and Knee Infection
- Multi-institutional study (684 PJI, 820 aseptic) generating a validated weighted scoring system
- Major criteria (diagnostic alone): two positive cultures of the same organism, or a sinus tract communicating with the joint
- Minor criteria are scored — serum CRP, D-dimer, ESR plus synovial WBC count, alpha-defensin, leukocyte esterase, PMN percent and synovial CRP; a score of 6 or more is infected, 2-5 is inconclusive (use intra-operative findings)
- Sensitivity 97.7 percent (superior to the 2011 MSIS and ICM definitions), specificity 99.5 percent
Kinematic versus mechanical alignment in primary TKA (RCT)
- 200 patients randomised to kinematic alignment (patient-specific cutting guides, cruciate-retaining) versus conventional mechanical alignment
- WOMAC and combined Knee Society Score improved significantly in both groups at 12 months
- The kinematic group had significantly better overall WOMAC and KSS scores but also more outliers with poor outcomes
- Deviation of postoperative alignment from the kinematic plan correlated with poorer results
Tourniquet versus no tourniquet in TKA — systematic review and meta-analysis of RCTs
- 41 RCTs, 2819 participants comparing TKA with versus without a tourniquet
- Serious adverse events were significantly more common with a tourniquet (risk ratio 1.73, 95% CI 1.10-2.73)
- Day-1 pain was higher with a tourniquet (mean difference 1.25 points) and hospital stay marginally longer; overall blood loss did NOT differ
- The only finding favouring the tourniquet was a slightly shorter operative time (about 3.7 minutes)
Long-term survival of cruciate-retaining versus posterior-stabilising TKR
- Retrospective registry cohort of 8117 primary TKAs (5389 cruciate-retaining, 2728 posterior-stabilising), 1988-1998
- Fifteen-year aseptic survival 90 percent for cruciate-retaining versus 77 percent for posterior-stabilising (p less than 0.001)
- The survival advantage for cruciate-retaining persisted in knees both with and without preoperative deformity
- After adjustment for age, sex, diagnosis and deformity, the risk of revision was halved with cruciate-retaining (hazard ratio 0.5, 95% CI 0.4-0.6)