Extensor Mechanism Disruption after Total Knee Arthroplasty

Extensor Mechanism Disruption after TotalKnee Arthroplasty

Robert J. Schoderbek, Jr., MD*; Thomas E. Brown, MD*;Kevin J. Mulhall, MD, MCh, FRCSI(Tr&Orth)*; Varatharaj Mounasamy, MD, FRCS*;

Richard Iorio, MD†; Kenneth A. Krackow, MD‡; W. Macaulay, MD§; andKhaled J. Saleh, MD, MSc(Epid), FRCSC, FACS*

Extensor mechanism disruption after total knee arthroplastyis a challenging complication for orthopaedic surgeons. Thetreatment options for repair include observation, direct pri-mary repair, direct primary repair with synthetic ligamentor autogenous tissue augmentation, or reconstruction withallograft tissue. A computerized systemic review and litera-ture search was performed to identify the relevant literatureon extensor mechanism disruptions associated with totalknee arthroplasty. A comprehensive review of the literatureand description of relevant treatment options and outcomeswere performed using the information gained with the lit-erature review. A multi-center prospective study on a con-secutive series of patients recruited from the North Ameri-can Knee Arthroplasty Revision (NAKAR) study was per-formed and data collected pre-operatively, intra-operatively,and post-operatively on patients that had a failed total kneearthroplasty using validated health related quality of lifemeasures was analyzed. Six out of 290 patients in the studyhad extensor mechanism disruption and this group of pa-tients had overall worse functional outcomes. The results ofthe study have solidified our knowledge that patients withextensor mechanism disruptions have worse functional out-

comes and will need intensive management and rehabilita-tion.

Level of Evidence: Economic and decision analyses, level III(systematic review of level III studies). See Guide for Authorsfor a complete description of levels of evidence.

Disruption of the extensor mechanism during or after totalknee arthroplasty (TKA) is an infrequent complication thatis a technically challenging but manageable problem fororthopaedic surgeons. The incidence of extensor mecha-nism disruption reported in the literature ranges from0.17–2.5%.10,34,45 Recent evidence from the North Ameri-can Knee Arthroplasty Revision (NAKAR) study supportsthese findings, with six of 290 TKAs being revised be-cause of extensor mechanism disruption. Extensor mecha-nism disruptions can be differentiated into a proximal ex-tensor mechanism disruption being a quadriceps tendonrupture or a distal extensor mechanism disruption being apatella tendon rupture.44 The etiology for extensor mecha-nism disruption is complex and multifactorial, and poorresults emphasize the importance of prevention.13,25,45

There are a variety of surgical options for treating extensormechanism disruption.8,9,10,12,41,45.59 The selected optionis based on the quality of the host tissue, the availability ofhost autograft and allograft tendons to assist with the re-pair, and the patient’s functional demands.51

Surgical management of extensor mechanism disrup-tion after TKA is challenging because of the inconsistentresults associated with the various reported operative tech-niques. In response to the poor results obtained with pri-mary tendon repair, surgical options have expanded toinclude various types of allograft tissue.3,8,11,16,17,32,42

There are numerous treatment options for the repair of adisrupted extensor mechanism ranging from non-operativetreatment with immobilization in a brace or cast to kneearthrodesis.1–3,5,8–10,12–15,22,23,29–34,42,45–47

From the *Department of Orthopaedic Surgery, University of Virginia, Char-lottesville, VA; the †Department of Orthopaedic Surgery, Boston UniversityMedical Center-Lahey Clinic, Burlington, MA; the ‡Department of Ortho-paedic Surgery, Buffalo General Hospital, Buffalo, NY; and the §Center forHip and Knee Replacement, Columbia University, New York City, NY.Each author certifies that he has no commercial associations (e.g., consul-tancies, stock ownership, equity interest, patent/licensing arrangements, etc)that might pose a conflict of interest in connection with the submitted article.One of the authors (KJS) received research funding from Stryker and SmithNephew and is a consultant with Stryker and Smith Nephew.Each author certifies that his institution has approved the human protocol forthis investigation and that all investigations were conducted in conformitywith ethical principles of research, and that informed consent was obtained.Correspondence to: Khaled J. Saleh, MD, MSc(Epid), FRCSC, FACS, De-partment of Orthopaedic Surgery, University of Virginia, PO Box 800159,Charlottesville, VA 22903. Phone: 434-243-0067; Fax: 434-243-0242;E-mail: [email protected]: 10.1097/01.blo.0000218726.06473.26

CLINICAL ORTHOPAEDICS AND RELATED RESEARCHNumber 446, pp. 176–185© 2006 Lippincott Williams & Wilkins

176

Our objectives for this study were two-fold, with thefirst objective being a systematic and bibliographic reviewto identify all relevant literature pertaining to extensormechanism disruptions associated with total knee arthro-plasty focusing on different treatment options and relevantpostoperative outcomes. The second objective was to as-certain through a prospective study, the functional out-comes of extensor mechanism reconstructions using vali-dated health related quality of life measures with ahypothesis that patients who disrupt their extensor mecha-nism have lower functional outcomes after total knee ar-throplasty revision.

MATERIALS AND METHODS

We performed a computerized literature search and bibliographyreview using the PubMed and Medline databases to identify allcitations concerning extensor mechanism disruptions after totalknee arthroplasty with the use of key words and Medical SubjectHeadings (MeSH) terms. The search string included (Arthro-plasty, Replacement, Knee OR Arthroplasty, Replacement ORKnee Replacement OR Arthroplasty, Revision OR Prosthesis)AND (extensor mechanism disruption) AND (patella tendon ORquadriceps tendon). The studies returned by Pub Med and Med-line using the search string were further managed by displayingthe abstracts and searching with the “find” and “highlight” func-tions of Mozilla Firefox (Mozilla Corporation, University ofIllinois, Champaign, IL) web browser by entering the terms “Ex-tensor mechanism disruption” or “ rupture of patella or quadri-ceps tendon”. The bibliography sections in all review articleswere examined and missed citations were retrieved and reviewedfor relevance. We excluded from further review any article thatreported on extensor mechanism disruptions not involving TKA,reports with no relevant postoperative outcomes, animal studies,and non-English language references. We included all appropri-ate articles pertaining to our topic with no limitations on date ofpublication. Due to the limited number of publications present inthe literature pertaining to extensor mechanism disruption aftertotal knee arthroplasty with the inconsistencies in presentationand evaluation of outcomes we were unable to adequately assessand compare study results. Therefore, in the first part of thisstudy we will be presenting a systematic review of the differenttypes of treatment options available supported by the literatureobtained from this review.

The second part of this review will involve the results ob-tained from the NAKAR study. The NAKAR study is a multi-center prospective study on a consecutive series of patients re-cruited from 17 North American sites. It was an observationalstudy of patients with failed TKAs that were deemed candidatesfor revision surgery. Recruitment followed a strictly managedprotocol at each site. Following Institutional Review Board ap-proval of the study at each of the participating units, patientswere considered for enrollment once the attending surgeon iden-tified the need for a total knee arthroplasty revision (TKAR) andwere then specifically approached regarding study inclusion.There were then specific criteria that dictated inclusion in or

exclusion from the study. (Tables 1 and 2) Eligible patients whoagreed to participate were consented by the site investigator andincluded in the study. Data were gathered from patients imme-diately following consent. The study included 290 patients goingon to TKAR, and six were identified as having an extensormechanism disruption causing primary TKA failure. Data col-lection points were at baseline (preoperatively), intraoperatively,6 months postoperatively and 12 months postoperatively. (Table3) The instruments used were specifically the Western Ontarioand McMaster University Osteoarthritis Index (WOMAC), theShort Form 36 (SF-36), the Lower Extremity Activity Scale(LEAS) and the Knee Society Score (KSS).4,33,50,56 TheWOMAC has 3 components that assess stiffness, pain and func-tion.4 The SF-36 has multiple components the main 2 subdivi-sions being the Mental Component Score and the Physical Com-ponent Score.56 The KSS is composed of a functional and anobjective component while the LEAS results in numerical scoreof the patients actual activity level.33,50

Etiology and PreventionThe risk factors associated with extensor mechanism disrup-tion include patient comorbidities, prior knee procedures, ana-tomic constraints, surgical technique, and prosthetic de-sign.1,9,10,12,13,16,17,19,29,42,46,57 Disruption of the extensor mecha-nism can be delineated into patella tendon rupture or quadricepstendon rupture.45 Disruption can occur intraoperatively usuallyrelated to surgical technique or postoperatively usually due tovascular interruption, increased patellofemoral strain, or chronic

TABLE 1. NAKAR Inclusion Criteria

1. At the least, the tibial and/or the femoral component requiredreconstruction

2. Signed informed consent was obtained from the subject3. The patient was greater than 18 years of age4. The patient was cognitively intact, fluent in English, and

capable of completing the self-administered questionnaires andadhering to the study protocol

5. The patient had a primary TKA that had failed (and not arevision)

TABLE 2. NAKAR Exclusion Criteria

1. Re-revision TKA2. Failed unicondylar prostheses3. TKA in need of only a polyethylene exchange4. Metastatic or primary tumor of the knee5. Reflex sympathetic dystrophy of the affected knee6. Subject is medically unsafe to undergo the procedure as

judged by the co-investigator7. Progressive muscular condition, causing deterioration of the

quadriceps muscle8. Neurologic deficit impairing the affected limb9. Knee pain that is associated with back pathology such as

spinal stenosis or vascular occlusion10. Patient declined participation

Number 446May 2006 Extensor Mechanism Disruption after Total Knee Arthroplasty 177

tendon attenuation.42,45 Patella tendon rupture is usually associ-ated with patient comorbidities, tissue devascularization frommultiple previous procedures, trauma or stiff knees.45 Quadri-ceps tendon rupture is usually caused by over resection of thepatella with damage to the quadriceps tendon, vascular injury, orincomplete healing after extended surgical approaches.10,45

Patient comorbidities include obesity, diabetes mellitus, rheu-matoid arthritis (RA), systemic lupus erythematosus, and chronicrenal failure.1,9,12,13,16,17,19,29,42,46,57 Multiple previous surgeriesand patients treated with long-term corticosteroid therapy ormultiple intraarticular steroid preparations are at increased risksecondary to weakened collagenous tissue.11,36,40

Multiple surgical interventions including distal realignmentprocedures about the extensor mechanism weaken and con-tribute to the reduction in blood supply to the surrounding tissueby damaging the vascular anastomosis around theknee.10,34,48,49,52,53,58

Performing a lateral retinacular release to improve patellatracking has also been speculated as contributing to avascularnecrosis (AVN) of the patella but confounding studies have ne-gated this information, making this a controversial conclu-sion.30,39,48,49,52,53 Extensive lateral release toward the quadri-ceps tendon may contribute to initiating the tear and damagingthe arterial blood supply increasing the tendency for the quad-riceps tendon to rupture soon after surgery.13,31

Patients with complex anatomical constraints, multiple pre-vious surgeries, prior infection, previous extensor mechanismrealignment procedures, patella baja, knee ankylosis and patientsundergoing total knee arthroplasty revision (TKAR) who needextensive exposure are at risk due to increased tissue scar-ring.10,16,17,25,32,34,46

Surgical techniques which are inadequate or done in hastemay place excessive strain on the tendon resulting in rupture.Proper component placement is essential to avoid excessivestresses on the extensor mechanism.10,21 Factors that may pre-dispose the knee to extensor mechanism disruption include el-evation in joint line, position and thickness of the patella, oversized TKA components, patellofemoral instability, and malalign-ment of the tibial or femoral components.10,20,21,42,45,47

Prosthetic components characteristics can contribute tochanges in patellofemoral strain and affect the outcome of sur-gical intervention. Constrained (hinged) implants have been im-plicated in causing extensor mechanism disruption.10,20,42,45,47

Chronic tendon attenuation may also be caused by sharp edgesassociated with component design.10

Identifying patients at risk for extensor mechanism disruptionpreoperatively and proper planning is essential. Preoperative his-tory and physical examination is the key to prevention. Evalua-tion of medical comorbidities, possible previous infection, re-view of previous operative technique and implanted prosthesis,and local examination of the knee and extremity (previous scars,skin condition, flexion deformity if any, extensor lag, active andpassive range of motion, patellar position and tracking, vascularstatus of the extremity) are essential. Radiographic evaluation ofcomponent size, position, alignment, fixation, bone quality andpresence of prosthetic remnants of previous surgery is importantin patients planned for revision. Laboratory values to rule outinfection should be routinely performed.

Once identified, precautionary measures should includeavoiding extensive dissection of the tibial attachment of the pa-tella tendon; proper orientation of components; avoiding exces-sive valgus alignment of the knee; and careful surgical tech-nique, especially in knees with limited preoperative motion,which should be undertaken to minimize the risk of disruption ofthe extensor mechanism.47

Surgical TreatmentTreatment for disruption of the extensor mechanism is determined bythe level of the disruption (patellar tendon, patella, or quadriceps ten-don), degree of functional loss (partial versus full tear), acute versuschronic nature of the disruption, availability of viable tissue for eitherdirect primary repair or augmentation, and the health and desires of thepatient.10,16,17,30,41 Treatment options for surgical repair include (1)direct primary repair alone by suturing, stapling,10 or wiring the tendonto the tubercle; (2) primary repair with biologic or synthetic graft aug-mentation (e.g., gracillis tendon, free fascia lata graft, plantaris tendon,semitendinosus tendon, or medial gastrocnemius rotational muscleflap); (3) allograft tissue augmentation with extensor mechanism

TABLE 3. NAKAR Study Results

Outcome Instruments

Disrupted ExtensorMechanism

(Preoperative)

Disrupted ExtensorMechanism(6 Months

Postoperatively)

Intact ExtensorMechanism

(Preoperative)

Intact ExtensorMechanism(6 Months

Postoperatively)

WOMAC function component 41 (SD 15.166) 26.8 (SD 14.990) 34.5 (SD 13.871) 21.92 (SD 14.549)WOMAC stiffness component 4.17 (SD 2.858) 4 (SD 2.449) 4.21 (SD 1.838) 3.18 (SD 1.924)WOMAC pain component 7.67 (SD 7.554) 5.60 (SD 5.505) 10.11 (SD 4.136) 5.66 (SD 4.489)SF-36 mental component score 46.83 (SD 15.039) 54.20 (SD 14.342) 49.13 (SD 12.238) 51.45 (SD 11.638)SF-36 physical component score 26.50 (SD 8.479) 27.8 (SD 4.919) 26.93 (SD 6.983) 33.58 (SD 8.800)LEAS 7.1667 (SD 2.483) 8.2 (SD 3.115) 7.57 (SD 2.635) 8.715 (SD 2.817)KSS function component 5 (SD 12.247) 35 (SD 31.82) 41.35 (SD 21.304) 62.39 (SD 24.924)KSS objective component 45.333 (SD 27.782) 82.5 (SD 2.887) 39.31 (SD 17.626) 76.197 (SD 16.009)

NAKAR = North American Knee Arthroplasty Revision; WOMAC = Western Ontario and McMaster Universities; LEAS = lower extremity activity scale; KSS = KneeSociety score; SD = standard deviation; disrupted extensor mechanism (n = 6); intact extensor mechanism (n = 286)

Clinical Orthopaedicsand Related Research178 Schoderbek, Jr, et al

or Achilles tendon.1–3,7–13,15–19,22–24,29,31,32,34,37,42,44–46,51,55,59

Patellotibial fusion, knee arthrodesis, or above the knee ampu-tation are salvage techniques which may be considered if surgi-cal repair is not possible or if extensor mechanism reconstructionfails.16,17,34,45 It is critical that the reconstructive surgeon fullyunderstand the indications and inherent limitations of each ofthese techniques.

Direct primary repair of a disrupted extensor mechanism withsutures, staples, or wires has typically been the first treatmentoption.1,10,13 Encouraging results have been reported for di-rect repair in native knees that have not had an arthro-plasty,8,15,31,54,55 but attempts at primary repair after TKA haverarely restored extensor function.8,14,32,35,45 Primary repair is thebest option in cases of acute rupture with sufficient tendinoustissue to permit the repair.1,10,13 A study that evaluated 18 kneestreated for patella tendon rupture after TKA reported that directrepair of extensor mechanism disruption has been associatedwith variable results, with only 25% of the patients having asuccessful outcome and the best results were with staple fixa-tion.45 A study performed on 34 patients with quadriceps tendonruptures after TKA (11 with a complete tear and 23 with a partialtear) showed conclusive findings with seven of the eleven com-plete tears having poor outcomes, that operative repair of a com-plete quadriceps tendon rupture with the use of suture fixationalone is unreliable in restoring extensor mechanism function.13

We believe that direct primary repair should be the first lineof treatment, and if the strength of the repair is found to betenuous after evaluation, strong consideration should be given toaugment (auto, allo or synthetic grafts) the primary repair.

The use of synthetic graft for augmentation of the repair of adisrupted extensor mechanism is an additional treatment option.These grafts include Dacron, Gore-tex polypropylene and otherartificial ligaments like Leeds-Keio ligament. Augmentation al-lows the majority of the load during the early post-operativeperiod to be carried by the synthetic ligament, and as the primarytissue heals, the load is gradually transferred to the repairedtissue or to a biologic graft if used in addition to synthetic grafts.Commercial availability and absence of donor site morbidity arethe advantages of using a synthetic graft. Concerns with thisimplant include an increased risk of infection and the risk of poortissue holding strength in patients who have undergone priorTKA revisions.2

One study reported a postoperative ROM of 0º–110º in apatient who sustained a rupture of the quadriceps tendon whichwas repaired with tenoplasty, augmented with Dacron and im-mobilized with an above knee cast with the knee in full extensionfor six weeks.15

A study from Japan reported a postoperative range of 146.4ºusing the Leeds-Keio ligament to repair either disrupted patellar(six) or quadriceps (five) tendons in eleven patients, with onepatient having both repaired. With a follow-up of 3.5 years onlyfour patients had an extensor lag less than 10°.19 Another studyreported a mean ROM from −4.6º of extension to 98º flexionusing the Leeds-Keio ligament to repair a rupture of the extensormechanism in five patients with a minimum follow-up of 38months.2

Autogenous tissue grafts provide healthy collagen thatstrengthens the repair of a disrupted extensor mecha-

nism.9,10,24,29 Different types of autogenous tissue graft availablefor augmentation of extensor mechanism reconstruction includesemitendinosus tendon, gracillis tendon, turndown flap of thequadriceps tendon, free fascia lata graft, and gastrocnemius ro-tational flap9,10,24,29

Hamstring tendon augmentation using the semitendinosusmuscle (Fig 1) has been shown to produce a stronger repair thanfascial strips and free grafts for reconstruction of the patellartendon.10,15,24 Noyes et al evaluated the mechanical properties ofhuman ligamentous grafts in young, cadaveric specimens andfound that the semitendinosus tendon had superior strength com-pared with grafts of the gracilis, distal iliotibial tract, fascia lata,or quadriceps-patella retinaculum.10,43 When tension is appliedto the semitendinosus tendon it is important that the Insall-Salvati ratio10,27 is calculated to ensure that the patella is held ina normal position as tension is applied to the free end of thesemitendinosus tendon and checked in 90º of flexion to avoidundue stress on the graft. Gracilis tendon may also be used tosupplement augmentation of semitendinosus.10

Postoperative care and rehabilitation for patients who havereceived semitendinosus tendon augmentation of the patellar ten-don includes protected extension for 6 weeks followed by pro-tected motion in a brace (0°–60° full extension while ambulat-

Fig 1. This diagram illustrates the reconstruction of the patel-lar ligament with the use of the semitendinosus tendon.

Number 446May 2006 Extensor Mechanism Disruption after Total Knee Arthroplasty 179

ing) for an additional 6 weeks before progressing to normalactivities. Varied outcomes have been reported in patients whohave been augmented with semitendinosus autograft for extensormechanism disruption.9,10,24,29,45

The use of the gastrocnemius rotational flap has been used toaugment repair of the extensor mechanism in patients with ex-posed prosthesis, infection, loss of the extensor mechanism,scarred and contracted soft tissue about the proximal aspect ofthe tibia from multiple operations, failed primary repair of theextensor mechanism, and a history of previous patellec-tomy.9,29,36,37 It provides improved soft tissue coverage and al-lows the return of some function of the extensor mechanism andmay be extended distally to include a portion of the Achillestendon when reconstruction of the entire extensor mechanism isdesired (Figs 2-4).9,29,36,37 It was originally popularized in the1980s for limb salvage procedures associated with reconstruc-tion of functional extensor mechanisms after excision of malig-nant tumors of the proximal aspect of the tibia and insertion ofa prosthesis when lack of soft tissue coverage was a concern.36,37

This rotation flap provides a vascularized muscle bed that pro-motes healing of the reconstructed extensor mechanism and doesnot rely on the bone quality of the patella or proximal tibia, andgives adequate soft tissue envelope for the components, makinglate failure unlikely.29,36,37

After the transfer of the medial gastrocnemius flap, there isminimal loss of function as the remaining lateral head of the

gastrocnemius and soleus function together to provide adequateplantar flexion of the ankle.29,36–38,41,54 The medial and lateralgastrocnemius muscle each receive one primary neurovascularbundle which continues longitudinally for the entire length of themuscle with no additional arterial inflow.4,23,29,51 This makesrotation and elevation of the muscle on its single proximallybased pedicle a reliable and safe procedure. The medial gastroc-nemius rotational flap has been used in two stage TKA re-implantation protocols for patients with wound necrosis and ac-tive chronic infection in which the initial stage involved irriga-

Fig 2A–B. (A) This illustration shows the medial gastrocne-mius flap, which involves dividing the medial gastrocnemius atits distal insertion in the Achilles tendon. (B) This illustrationshows the extended medial gastrocnemius flap, which in-volves obtaining the medial 1⁄3 to 1⁄2 of the Achilles tendon.

Fig 4. This intraoperative photograph shows the medial gas-trocnemius flap used to augment allograft reconstruction. Theuse of the gastrocnemius rotational flap has been used toaugment repair of the extensor mechanism in patients withexposed prosthesis, infection, loss of the extensor mecha-nism, scarred and contracted soft tissue, and failed primaryrepair of the extensor mechanism. It provides improved softtissue coverage and allows the return of some function of theextensor mechanism.

Fig 3. This illustration shows the medial gastrocnemius trans-posed anteriorly to cover the tibial tubercle. The patellar ten-don and anterior joint capsule should then be sutured to themedial border of the gastrocnemius.

Clinical Orthopaedicsand Related Research180 Schoderbek, Jr, et al

tion and debridement of the infected knee and use of thegastrocnemius flap for reconstruction of the extensor mecha-nism.9,29 A split thickness skin graft may be used to cover partof the flap in either technique if approximation of the skin is notpossible.

The technique for the medial gastrocnemius rotational flapwas first described by Jaureguito et al and then by Busfieldet al.9,29 Jaureguito et al stated that it was a reliable option withan improved ROM (70° preoperatively, 100° postoperatively),less extensor lag (24° postoperatively), and improvements infunctional walking status.29 Busfield et al reported an averageROM of 3°–93° and an average extensor lag of 13.5°.9 Similarresults have been reported with complications including repeatsoft tissue flap coverage, need for manipulation due to limitedROM, minimal skin necrosis, and transfemoral amputation.9

The technique involves harvesting the medial gastrocnemiusmuscle including the medial portion of the Achilles tendonthrough the previous TKA incision with extension down the legmedial to the tibia.9,29 The recommended postoperative regimendepends on the surgical strength of the repair. Patients are im-mobilized in full extension for 2–3 weeks and gradual flexion isallowed (0°–30° for 3 weeks, 0°–60° for 3 weeks).9,29 At 12weeks postoperatively, patients are allowed to perform activitiesas tolerated.9,29

Augmentation of a disrupted extensor mechanism in the set-ting of a TKA with an allograft reconstruction technique is usedto restore the structurally strong tissue of the extensor mecha-nism in a setting where the available tissue is mechanically in-sufficient and structurally inadequate for the demands of kneejoint function.3,7,8,12,16,17,33,44,59 Unlimited supply, variety of tis-sue types, absence of donor site, morbidity, and superior strengthwhen compared to some host-tissues are the advantages of usingallograft tissue.

Achilles tendon allografts with a calcaneal bone block andextensor mechanism allografts consisting of quadriceps tendon-patella-patella tendon-tibial-tubercle have been used success-fully to restore disrupted extensor mechanisms.3,16–18 The use ofallografts for restoration of a functional extensor mechanism isbest for patients with poor quality host tissue where primaryrepair and augmentation with autograft is not feasible, patientswith low functional demand or limited life expectancy thatwould be devastated by the lack of quadriceps function, or inpatients with compromised soft tissue from multiple previousoperations or infections.8,16,17,59 Concerns when using allografttissue include immune reaction, disease transmission, and graftstrength.6,11,12,21,28,40 The risk of immune response has beengreatly diminished by deep freezing allografts, and disease trans-mission is extremely rare.6,11,12,21,28,40 Unfortunately, graftstrength and risk of failure are always concerns. Freeze driedallografts are weaker than fresh frozen allografts, and have agreater risk of generating a host immune response.6,11,16,17,26,43

Emerson et al recommended the use of fresh frozen allograftsbecause the two complications occurred in patients with freezedried allografts.16,17

Achilles tendon allograft is indicated for use in repairing theextensor mechanism when the patella and patellar component areintact and can be mobilized to within 3–4 cm of joint line in

extension, or if the quadriceps mechanism is disrupted chroni-cally and retracted too proximally so that an extensor mechanismcomposite graft would not allow sufficient length to restore con-tinuity to the extensor mechanism.3,16,17 The technique involvesprimary repair of the patellar tendon to attempt correct patellarpositioning relative to the joint line (Fig 5). Postoperatively, theknee is immobilized in a hinged knee brace locked in full ex-tension and gradual flexion is allowed over the ensuing 8–12weeks.12,18

The results of Crossett et al on extensor mechanism repairaugmented with Achilles tendon allograft were good and showedimproved walking ability, decreased extensor lag (44° preopera-tively, 3° postoperatively), improved ROM (88° preoperatively,107° postoperatively), and improved knee stability.12

The use of extensor mechanism allograft that includes quad-riceps tendon, patella, patellar tendon, and tibial tubercle forreconstruction of a disrupted extensor mechanism was first usedand popularized by Emerson et al in 1985 in anticipation of animproved functional outcome.16,17 The goal was replacement ofthe deficient tissue with structurally sound tissue, placing thesurgical junctions at the most favorable healing sites, namely thequadriceps tendon and the proximal tibial metaphysis (Fig6).16,17 No autogenous tissue was removed during placement ofthe extensor mechanism allograft, and the patient’s patella rem-nant and scarred patellar tendon were retained to maintain theintegrity of the quadriceps and provide coverage of the graft withhost tissue.16,17 The remaining host tissue worked as load sharingmechanical support to the allograft.16,17 The advantage of theallograft was that it could be anatomically situated over the kneejoint, making it an ideal reconstruction of the extensor mecha-nism.16,17 The postoperative regimen includes placement of ahinged knee brace for 6 weeks until the tibial tubercle is united,

Fig 5. This intraoperative photograph shows an Achilles allo-graft reconstruction of the extensor mechanism. It is used torepair the extensor mechanism when the patella and patellarcomponent are intact and can be mobilized to within 3–4 cm ofjoint line in extension, or if the quadriceps mechanism is dis-rupted.

Number 446May 2006 Extensor Mechanism Disruption after Total Knee Arthroplasty 181

with limited motion up to 60º for the first 6 weeks once the skinincision has healed. Gentle active exercises to 90º are permittedby the end of the second 6 weeks. Protected weight bearing isallowed during bracing. Results when using this technique dem-onstrated improve ambulatory ability, no loss of flexion, andimproved extensor lag.16,17

Patellar resurfacing increased the risk of patellar fracture inthe allograft bone and is no longer advocated.16,17 These resultsare supported by Reuben et al,47 who reported that resurfacingthe patella increases anterior strain in the remaining patella rem-nant and increases the chance of patella fracture.16,17,47

Subsequent publications on the results of this technique re-ported mixed results, with poor functional outcomes for thosecases in which the extensor mechanism allograft was not suturedto the host tissue under maximal tension.32,42 In 2005, Burnett etal7 presented a study that directly compared the results of recon-struction with extensor mechanism allograft and postoperativerehabilitation of two different techniques.7,8,16,17,42 The authorssurmised that tensioning of the extensor mechanism allograft infull extension is a critical determinant of success and yields ahigher rate of clinical success, including restoration of full activeextension and better walking status, without adversely affectingflexion.7,8,16,17,42

RESULTS FROM NAKAR STUDY

The NAKAR study identified 290 patients with failedTKAs that were deemed candidates for TKAR. Six of the290 patients had disruption of their extensor mechanismcausing primary TKA failure. The patient information per-

taining to extensor mechanism disruption did not note theprecise etiology, location or management for the disrup-tion but focused on the pre-operative and post-operativeWOMAC, SF-36, LEAS, and KSS scores (Table 3) andwhether a disruption of the extensor mechanism waspresent. The WOMAC Index showed worse functionalscores, similar stiffness scores, and similar post-operativepain scores for patients with extensor mechanism disrup-tions. These objective findings fit with what we would beexpected from clinical practice. The physical componentof the SF-36 test showed comparable preoperative scoreswith divergent postoperative scores reflecting worse over-all outcomes for patient with extensor mechanism disrup-tions. The LEAS test showed that the patients with exten-sor mechanism disruptions are slightly less active but notby a substantial difference, which is an encouraging find-ing. The functional KSS showed a substantial differencebetween the two groups of patients indicating that al-though not great on general health measures these patientsare worse off when joint specific scores are examined. TheKSS function scores were much worse preoperatively inEMD than in other patients being revised for other causesof failure, but relative to these very low baseline scores,large improvements were actually obtained postopera-tively even though they still substantially lagged behindthe other failures in functional outcome. Interestingly, theobjective score is not as badly affected preoperatively andalso shows notable postoperative improvement. Becauseof the small numbers of patients with extensor mechanismdisruption and lack of information pertaining to etiology,location and management of the extensor mechanism dis-ruption in this study group of six patients with extensormechanism disruption, definitive findings regarding out-comes were difficult to obtain. However, patients withextensor mechanism disruption had worse functional out-comes and will require more intensive management andrehabilitation.

DISCUSSION

Extensor mechanism disruption is a rare but disastrouscomplication after TKA. The etiology is multifactorial andcan include: traumatic rupture, surrounding soft tissue at-tenuation due to associated comorbidities, and infection.The appropriate treatment for disruption of the extensormechanism is determined by the level of the disruption,degree of functional loss, acute versus chronic disruption,availability of viable tissue for direct primary repair oraugmentation, and the health and desires of the pa-tient.10,16,17,34,45 The treatment options for the repair of adisrupted extensor mechanism include observation, directprimary repair, autogenous tissue augmentation, syntheticligament augmentation, and allograft tissue augmenta-

Fig 6. This intraoperative photograph shows the host reti-naculum and soft tissue sleeve, which are closed over theallograft using nonabsorbable sutures (white arrow). A pantsover vest suture technique is used to maximize host-graft tis-sue contact and improve graft incorporation (black arrow =allograft extensor mechanism; gray arrow = distal allograftbone block fixation).

Clinical Orthopaedicsand Related Research182 Schoderbek, Jr, et al

tion.1–3,7–13,15–19,22–24,29,31,32,34,37,42,44–46,51,55,59 It is hardto provide a definitive answer for the best treatment optionfor patients with an extensor mechanism disruption be-cause of the low number of patients in the prospectivestudies, and the lack of randomized studies on treatmentoptions for extensor mechanism reconstruction.

Limitations of this study include the insufficient infor-mation presented from the NAKAR study in relation toprecise etiology, location, and management of the extensormechanism disruptions. This limits the information andoutcomes obtained from the study because it does notprovide us with further guidance into the best treatmentoptions for this population of patients. Limitations of sys-temic and bibliographic review stem from our inability tocorrelate and compare different studies obtained from theliterature review because of the inconsistencies in presen-tation and evaluation of patient outcomes.

The NAKAR study solidified our knowledge that pa-tients with extensor mechanism disruptions have worsefunctional outcomes and will need intensive managementand rehabilitation when compared to patients with failedTKA due to other reasons. What is does not provide isguidance into the best treatment option for patients withextensor mechanism disruption.

Avoiding disruption of the extensor mechanism can beaccomplished by paying close attention to surgical detailwith appropriate knowledge of patients’ medical and sur-gical risk factors. When disruption of the extensor mecha-nism does occur, direct primary repair should be the firstline of treatment for reconstruction.1,8,10,13,14,32,35,45 Di-

rect repair should be evaluated intraoperatively with directvisualization under stress to determine the strength of therepair and whether augmentation is warranted.

Autogenous graft augmentation after primary repair ofa disrupted extensor mechanism has been used for years byproviding healthy collagen tissue to strengthen the repairtissue.9,10,24,29 Its use is appropriate in patients who mayhave weak connective tissue from systemic disease, elder-ly patients with tissues that have been devascularized byprevious operations, infected TKA, exposed prosthesis,and patients who have had a previous patellectomy.8,25

Allograft augmentation of a disrupted extensor mecha-nism in the setting of a TKA has gained popular-ity3,7,8,12,16,17,33,44,59 The goal for allograft augmentationis to restore the tissue of the extensor mechanism in asetting where the tissue is mechanically insufficient andstructurally inadequate for the demands of knee joint func-tion.13,14 Achilles tendon allografts and extensor mecha-nism allografts have been used successfully to restore adisrupted extensor mechanism.3,16–18 Modifications to theoriginally described technique including more vigoroustensioning of the graft and protected flexion postopera-tively have resulted in improved functional outcomes, withno compromise to knee flexion.7,8

This report is the first comprehensive review of theliterature regarding treatment options and outcomes forpatients suffering from extensor mechanism disruptionsafter total knee arthroplasty. The algorithms in Figures 7and 8 may provide a check list for orthopaedic surgeons tofollow for determination of the best operative technique

Fig 7. This algorithm provides acheck list for orthopaedic surgeonsto follow for determination of thebest operative technique and tissuetype to be used for treatment of pa-tients with patellar tendon disruptionassociated with a TKA.

Number 446May 2006 Extensor Mechanism Disruption after Total Knee Arthroplasty 183

and tissue type to be used for treatment of different pa-tients that have sustained an extensor mechanism disrup-tion associated with TKA.

AcknowledgmentsMembers of the North American Knee Arthroplasty RevisionStudy Group: B. Bershadsky, PhD; M. Bostrom, MD; R. Bourne,MD; C. H. Clark, MD; E. Cheng, MD; G. Engh, MD; T. Gioe,MD; H. Ghomrawi, MPH; S. Hass, MD; W. Healy, MD;D. Heck, MD; K. Hepburn, PhD; D. Hungerford, MD; R. Iorio,MD; R. Kane, MD; K. Krackow, MD; R. Kyle, MD; R. Laskin,MD; P. Lotke, MD; W. Maccaulay, MD; S. MacDonald, MD;J. McAuley, MD; R. McCaulden, MD; M. Mont, MD; C. Nel-son, MD; J. Parvizi, MD; C. Rorabeck, MD; T. Sculco, MD;S. Scully, MD; G. Scuderi, MD; M. Swiontkowski, MD;and R. Windsor, MD.

References1. Abril JC, Alvarez L, Vallejo JC. Patellar tendon avulsion after total

knee arthroplasty. A new technique. J Arthroplasty. 1995;10:275–279.

2. Aracil J, Salom M, Aroca JE, Torro V, Lopez-Quiles D. Extensormechanism reconstruction with Leeds-Keio ligament in total kneearthroplasty. J Arthroplasty. 1999;14:204–208.

3. Barrack RL, Stanley T, Allen Butler R. Treating extensor mecha-nism disruption after total knee arthroplasty. Clin Orthop Relat Res.2003;416:98–104.

4. Bellamy N, Buchanan WW, Goldsmith CH, Campbell J, Stitt L.Validation study of WOMAC: a health status instrument for mea-suring clinically-important patient relevant outcomes to antirheu-matic drug therapy in patients with osteoarthritis of the hip or knee.J Rheumatol. 1988;15:1833–1840.

5. Bengtson S, Carlsson A, Relander M, Knutson K, Lidgren L. Treat-

ment of the exposed knee prosthesis. Acta Orthop Scandinavia.1987;58:662–665.

6. Buck BE, Malinin TI, Brown MD. Bone transplantation and humanimmunodeficiency virus. An estimate of risk of acquired immuno-deficiency syndrome (AIDS). Clin Orthop Relat Res. 1989;240:129–136.

7. Burnett SJ, Berger RA, Della Valle CJ, Sporer SM, Jacobs JJ,Paprosky WG, Rosenberg AG. Extensor mechanism allograft re-construction after total knee arthroplasty. J Bone Joint Surg Am.2005;87(Suppl 1:Part 2):175-194.

8. Burnett SJ, Berger RA, Paprosky WG, Della Valle CJ, Jacobs JJ,Rosenberg AG. External Mechanism allograft reconstruction aftertotal knee arthroplasty. A comparison of two techniques. J BoneJoint Surg Am. 2004;86:2694–2699.

9. Busfield BT, Huffman GR, Nahai F, Hoffman W, Ries MD. Ex-tended medial gastrocnemius rotational flap for treatment of chronicknee extensor mechanism deficiency in patient with and withouttotal knee arthroplasty. Clin Orthop Relat Res. 2004;428:190–197.

10. Cadambi A, Engh GA. Use of a semitendinosus tendon autogenousgraft for rupture of the patella ligament after total knee arthroplasty.A report of seven cases. J Bone Joint Surg Am. 1992;74:974–979.

11. Cordrey LJ. A comparative study of fresh autogenous and preservedhomogenous tendon grafts in rabbits. J Bone Joint Surg Br. 1963;45:182–195.

12. Crossett LS, Sinha RK, Sechriest VF, Rubash HE. Reconstructionof a Ruptured Patellar Tendon with Achilles Tendon Allograft fol-lowing Total Knee Arthroplasty. J Bone Joint Surg Am. 2002;84:1354–1361.

13. Dobbs RE, Hanssen AD, Lewallen DG, Pagnano MW. QuadricepTendon Rupture after Total Knee Arthroplasty. J Bone Joint SurgAm. 2005;97:37–45.

14. Doolittle KH 2nd, Turner RH. Patellofemoral problems followingtotal knee arthroplasty. Orthop Rev. 1988;17:696–702.

15. Ecker ML, Lotke PA, Glazer RM. Late reconstruction of the patellartendon. J Bone Joint Surg Am. 1979;61:884–886.

16. Emerson RH Jr, Head WC, Malinin TI. Reconstruction of patellartendon rupture after total knee arthroplasty with and extensormechanism allograft. Clin Orthop Relat Res. 1990;260:154–161.

Fig 8. This algorithm provides acheck list for orthopaedic surgeonsto follow for determination of thebest operative technique and tissuetype to be used for treatment of pa-tients with quadriceps tendon dis-ruption associated with a TKA.

Clinical Orthopaedicsand Related Research184 Schoderbek, Jr, et al

17. Emerson RH Jr, Head WC, Malinin TI. Extensor mechanism recon-struction with an allograft after total knee arthroplasty. Clin OrthopRelat Res. 1994;303:79–85.

18. Falconiero RP, Pallis MP. Chronic rupture of a patellar tendon: Atechnique for reconstruction with Achilles allograft. J Arthroscopyand Related Surg. 1996;12:623–626.

19. Fernandez-Baillo N, Garay EG, Ordonez JM. Rupture of the quad-riceps tendon after total knee arthroplasty. A case report. J Arthro-plasty. 1993;8:331–334.

20. Figgie HE III, Goldberg VM, Heiple KG, Holler HS, Godgon NH.The influence of tibial patellofemoral location on function of theknee in patients with the posterior stabilized condylar knee pros-thesis. J Bone Joint Surg Am. 1986;68:1035–1040.

21. Friedlaender GE, Strong DM, Sell KW. Studies on the antigenicityof bone. I. Freeze-dried and deep-frozen bone allografts in rabbits.J Bone Joint Surg Am. 1976;58:854–858.

22. Fujikawa K, Ohtani T, Matsumoto H, Seedhom BB. Reconstructionof the extensor apparatus of the knee with the Leeds-Keio ligament.J Bone Joint Surg Br. 1994;76:200–203.

23. Gerwin M, Rothaus KO, Windsor RE, Brause BD, Insall JN. Gas-trocnemius muscle flap coverage of exposed or infected knee pros-theses. Clin Orthop Relat Res. 1993;286:64–70.

24. Gustilo RB, Thompson R. Quadriceps and Patellar Tendon Rup-tures Following Total Knee Arthroplasty. In Rand JA, Dorr LD(eds). Total Knee Arthroplasty of the Knee. Proceedings of theKnee Society, 1985-1986. Rockville, Aspen 41-47, 1987.

25. Healy WL, Wasilewski SA, Takei R, Oberlander M. Patellofemoralcomplications following total knee arthroplasty: Correlation withimplant design and patient risk factors. J Arthroplasty. 1995;10:197–201.

26. Hosey G, Kowalchick E, Tesoro D, Balazsy J, Klocek J, PedersonB, Wertheimer SJ. Comparison of the mechanical and histologicproperties of Achilles tendons in New Zealand white rabbits sec-ondarily repaired with Marlex mesh. J Foot Surg. 1991;30:214–233.

27. Insall J, Salvati E. Patella position in the normal knee joint. Radi-ology. 1971;101:101–104.

28. Jackson DW, Windler GE, Simon TM. Intraarticular reaction asso-ciated with the use of freeze-dried, ethylene oxide-sterilized bone-patella tendon-bone allografts in the reconstruction of the anteriorcruciate ligament. Am J Sports Med. 1990;18:1–10.

29. Jaureguito JW, Dubios CM, Smith SR, Gottlieb LJ, Finn HA. Me-dial gastrocnemius transposition flap for the treatment of disruptionof the extensor mechanism after total knee arthroplasty. J BoneJoint Surg Am. 1997;79:866–873.

30. Johnson DP, Eastwood DM. Lateral patellar release in knee arthro-plasty. Effect on wound healing. J Arthroplasty. 1992;7(Suppl):427–431.

31. Larsen E, Lund PM. Ruptures of the extensor mechanism of theknee joint. Clinical results and patellofemoral articulation. Clin Or-thop Relat Res. 1986;213:150–153.

32. Leopold SS, Greidanus N, Paprosky WG, Berger RA, RosenbergAG. High rate of failure of allograft reconstruction of the extensormechanism after total knee arthroplasty. J Bone Joint Surg Am.1999;81:1574–1579.

33. Lingard EA, Katz JN, Wright JR, Wright EA, Sledge CB. Validityand responsiveness of the Knee Society Clinical Rating system incomparison with the SF-36 and WOMAC. J Bone Joint Surg Am.2001;83:1856–1864.

34. Lynch AF, Rorabeck CH, Bourne RB. Extensor mechanism com-plications following total knee arthroplasty. J Arthroplasty. 1987;2:135–140.

35. MacCollum MS 3rd, Karpman RR. Complications of the PCA ana-tomic patella. Orthopedics. 1989;12:1423–1428.

36. Malawer MM, McHale KA. Limb-sparing surgery for high-grademalignant tumors of the proximal tibia. Surgical technique and amethod of extensor mechanism reconstruction. Clin Orthop RelatRes. 1989;239:231–248.

37. Malawer MM, Price WM. Gastrocnemius transposition flap in con-

junction with limb-sparing surgery for primary bone sarcomasaround the knee. Plast Reconstr Surg. 1984;73:741–750.

38. Mathes SJ, Nahai F. Muscle flap transposition with function pres-ervation: technical and clinical considerations. Plast Reconstr Surg.1980;66:242–249.

39. McMahon MS, Scuderi GR, Glashow JL, Scharf SC, Meltzer LP,Scott WN. Scintigraphic determination of patellar viability afterexcision of infrapatellar fat pad and/or lateral retinacular release intotal knee arthroplasty. Clin Orthop Relat Res. 1990;260:10–16.

40. Minami A, Ishii S, Ogino T, Oikawa T, Kobayashi H. Effect of theimmunological antigenicity of the allogenic tendons on tendongrafting. Hand. 1982;14:111–119.

41. Murray MP, Guten GN, Sepic SB, Gardner GM, Baldwin JM. Func-tion of the triceps surae during gait. Compensatory mechanisms forunilateral loss. J Bone Joint Surg Am. 1978;60:473–476.

42. Nazarian DG, Booth RE Jr. Extensor mechanism allografts in totalknee arthroplasty. Clin Orthop Relat Res. 1999;367:123–129.

43. Noyes FR, Butler DL, Grood ES, Zernicke RF, Hefzy MS. Biome-chanical analysis of human ligament grafts used in knee-ligamentrepairs and reconstruction. J Bone Joint Surg Am. 1984;66:344–352.

44. Parker DA, Dunbar MJ, Rorabeck CH. Extensor mechanism failureassociated with total knee arthroplasty: Prevention and manage-ment. J Amer Academy of Ortho Surg. 2003;11:238–247.

45. Rand JA, Morrey BF, Bryan RS. Patellar tendon rupture after totalknee arthroplasty. Clin Orthop Relat Res. 1989;244:233–238.

46. Rasul AT Jr, Fischer DA. Primary repair of quadriceps tendon rup-ture. Results of treatment. Clin Orthop Relat Res. 1993;289:205–207.

47. Reuben JD, McDonald CL, Wodward PL, Hennington LJ. Effect ofpatella thickness on patella strain following total knee arthroplasty.J Arthroplasty. 1991;6:251–258.

48. Ritter MA, Herbst SA, Keating EM, Faris PM, Meding JB. Patel-lofemoral complications following total knee arthroplasty. Effect ofa lateral release and sacrifice of the superior lateral geniculate ar-tery. J Arthroplasty. 1996;11:368–372.

49. Ritter MA, Keating EM, Faris PM. Clinical, roentgenographic, andscintigraphic results after interruption of the superior lateral genicu-lar artery during total knee arthroplasty. Clin Orthop Relat Res.1989;248:145–151.

50. Saleh KJ, Mulhall KJ, Bershadsky B, White LE, Buyea CM, Ghom-rawi H, Krackow KA. Development and validation of a lower-extremity activity scale. Use for patients treated with revision totalknee arthroplasty. J Bone Joint Surg Am. 2005;87:1985–1994.

51. Sanders R, O’Neill T. The gastrocnemius myocutaneous flap usedas a cover for the exposed knee prosthesis. J Bone Joint Surg Br.1981;63:383–386.

52. Scapinelli R. Blood Supply to the Human Patellae. J Bone JointSurg Br. 1967;49:563–570.

53. Scuderi G, Scharf SC, Meltzer LP, Scott WN. The relationship oflateral releases to patella viability in total knee arthroplasty. J Ar-throplasty. 1987;2:209–214.

54. Simon SR, Mann RA, Hagy JL, Larsen LJ. Role of the posterior calfmuscles in normal gait. J Bone Joint Surg Am. 1978;60:465–472.

55. Siwek CW, Rao JP. Ruptures of the extensor mechanism of the kneejoint. J Bone Joint Surg Am. 1981;63:932–937.

56. Ware JE Jr, Sherbourne CD. The MOS 36-item short form healthsurvey (SF-36). I. Conceptual framework and item selection. MedCare. 1992;30:473–483.

57. Webb LX, Toby EB. Bilateral rupture of the patella tendon in anotherwise healthy male patient following minor trauma. J Trauma.1986;26:1045–1048.

58. Wolf AM, Hungerford DS, Krakow KA, Jacobs MA. Osteotomy ofthe tibial tubercle during total knee replacement. A report of twenty-six cases. J Bone Joint Surg Am. 1989;71:848–852.

59. Zanotti RM, Freiberg AA, Matthews LS. Use of patellar allograft toreconstruct a patellar tendon-deficient knee after Total Joint Arthro-plasty. J Arthroplasty. 1995;10:271–274.

Number 446May 2006 Extensor Mechanism Disruption after Total Knee Arthroplasty 185