Extensor Allograft Technique

8/17/2019 Extensor Allograft Technique

1/20

COPYRIGHT © 2005 BY THE JOURNAL OF BONE AND JOINT SURGERY, INCORPORATED

Extensor MechanismAllograft ReconstructionAfter Total Knee ArthroplastySurgical TechniqueBy R. Stephen J. Burnett, MD, FRCS(C), Richard A. Berger, MD, Craig J. Della Valle, MD, Scott M. Sporer, MD,Joshua J. Jacobs, MD, Wayne G. Paprosky, MD, and Aaron G. Rosenberg, MD

Investigation performed at the Department of Orthopedic Surgery, Rush University Medical Center, Chicago, Illinois

The original scientific article in which the surgical technique was presented was published in JBJS Vol. 86-A, pp. 2694-2699, December 2004

INTRODUCTION

Extensor mechanism disruption is a devastating complication of to-tal knee arthroplasty. Multiple techniques for repair or reconstruc-tion of a deficient extensor mechanism have been described inassociation with total knee arthroplasty; however, few have beenable to reliably restore a functional extensor mechanism1. Despiteencouraging results reported for direct repair in native knees, at-

tempts at primary repair following a total knee arthroplasty rarelyrestore extensor function. The use of local autogenous tissue toaugment a primary repair has been recommended. These patientshave frequently undergone multiple previous knee procedures, andthese local autogenous tissues may be compromised and unsuitablefor use.

Emerson et al.2,3 reported on the use of a complete knee extensormechanism allograft in total knee arthroplasty to reconstruct the defi-cient extensor mechanism. Although the early clinical results werepromising, extensor lag occurred early. Nazarian and Booth4 modifiedthe technique described by Emerson et al., recommending that the al-

lograft be tightly tensioned in full extension, and they reported im-proved early results. The host tissue-allograft junctions recently havebeen studied5, and the findings have provided useful information insupport of this technique.

In the present report, we describe the surgical techniquethat we have modified and currently use6 to reconstruct the defi-cient extensor mechanism with an extensor mechanism allograftthat is tightly tensioned with the knee in full extension. The criticalconcepts, pitfalls, and technical aspects of this technique arepresented.

ABSTRACT

BACKGROUND:

Disruption of the extensor mecha-

nism is an uncommon but cata-

strophic complication of total

knee arthroplasty. We evaluated

two techniques of reconstructing

a disrupted extensor mechanismwith the use of an extensor mech-

anism allograft in revision total

knee arthroplasty.

METHODS:

Twenty consecutive reconstruc-

tions with the use of an extensor

mechanism allograft consisting

of the tibial tubercle, patellar ten-

don, patella, and quadriceps ten-

don were performed. The first

seven reconstructions (Group I)

were done with the allograft mini-

mally tensioned. The thirteen sub-

sequent procedures (Group II)

were performed with the allograft

tightly tensioned in full extension.

All surviving allografts were evalu-

ated clinically and radiographically

after a minimum duration of

follow-up of twenty-four months.

continued


2/20

THE J OURNAL OF B ON E & JOINT S URGERY · SURGICAL TECHNIQUES SEPTEMBER 2005 · VOLUME 87-A · SUPPLEMENT 1 , PAR T 2 · JBJS.OR G

INITIAL EVALUATION

A deficient extensor mechanismin association with a total kneearthroplasty is one of the mostchallenging problems that theorthopaedic surgeon who per-forms joint replacement surgery

may encounter. The patient isinitially evaluated with a history,directed physical examination ofthe knee and extremity, radio-graphs, and adjunctive investiga-tions. The history should focuson obtaining information about

prior extensor mechanism proce-dures or surgery and the priorand current function of the knee.Symptoms of instability, giving-way, and an inability to extendthe knee should be sought. Thenature of previous surgeries and

ABSTRACT | continued

RESULTS:

All of the reconstructions in GroupI were clinical failures, with an av-

erage postoperative extensor lag

of 59° (range, 40° to 80°) and an

average postoperative Hospital for

Special Surgery knee score of 52

points. All thirteen reconstruc-

tions in Group II were clinical

successes, with an average post-

operative extensor lag of 4.3°

(range, 0° to 15°) (p < 0.0001)

and an average Hospital for Spe-

cial Surgery score of 88 points.

Postoperative flexion did not differsignificantly between Group I (av-

erage, 108°) and Group II (aver-

age, 104°) (p = 0.549).

CONCLUSIONS:

The results of reconstruction with

an extensor mechanism allograft

after total knee arthroplasty de-

pend on the initial tensioning of

the allograft. Loosely tensioned

allografts result in a persistent ex-

tension lag and clinical failure. Al-lografts that are tightly tensioned

in full extension can restore active

knee extension and result in clini-

cal success. On the basis of the

number of knees that we studied,

there was no significant loss of

flexion. Use of an extensor mech-

anism graft for the treatment of a

failure of the extensor mechanism

will be successful only if the graft

is initially tensioned tightly in full

extension.

FIG. 2

A complete fresh-frozen, nonirradiated knee extensor mechanism allograft that includes

the tibia, patellar tendon, patella, and quadriceps tendon is used.

FIG. 1

Assessment of prior incisions over the knee and a careful examination are essential pre-

operatively when considering revision surgery.


3/20


FIG. 4

The host extensor mechanism is sharply dissected longitudinally in the midline, through

the patellar tendon and quadriceps tendon.

the duration of extensor dys-function should be determined.Prior operative reports should bereviewed and scrutinized for theextensor mechanism and how it

was managed in previous surger-ies. A history of infection—re-motely or in association withprior surgery of the knee—war-rants further investigation.Medical comorbidities or im-munosuppressive therapy thatmay impact on wound-healingshould be sought. On physicalexamination, evaluation of thegait pattern and the use of walk-

ing aids are assessed. Prior inci-sions over the knee (Fig. 1) andactive and passive range of mo-tion are recorded. The presenceof an extensor lag should be care-fully measured, and the passiveamount of full extension that isable to be demonstrated shouldbe noted. The presence of a flex-ion contracture and the inability

to passively extend the knee arenoted. The tracking of the ex-tensor mechanism during range-of-motion testing should beexamined closely, as malrotationof the components of the totalknee arthroplasty may be a factorin the extensor mechanism fail-ure. Radiographs are evaluatedfor component alignment, fixa-tion, sizing, remaining host-bonestock, and the design of compo-nents. The extensor mechanismand patellar position are evalu-ated for patella infera, patellaalta, and the presence or absenceof a patella. In addition, the pres-ence of heterotopic ossificationinvolving the extensor mecha-

FIG. 3

Use of the previous incision is preferred. We use a sterile tourniquet as it is easily re-

moved for the allograft-host proximal graft repair.


4/20


nism is noted. The presence ofsuture anchors or staples aroundthe insertion of the patellar ten-don into the tibial tubercle is of-ten an ominous radiographicsign. Patellar tracking is evalu-ated on the axial radiograph. Ifthere is any concern about mal-rotation of the components, werecommend an axial computedtomography scan of the femoraland tibial components to evalu-ate for component internalmalrotation7. The erythrocytesedimentation rate and serumC-reactive protein level are mea-sured to evaluate for infection. Ifthese are elevated, a knee aspira-tion is performed for cell countand synovial fluid culture.

Decision to Reconstruct theExtensor Mechanism with Use ofan Extensor Mechanism Allograft Once the diagnosis and etiology

of a deficient extensor mecha-nism is made, we discuss the sur-gical options with each patient.The indications and contraindi-cations are carefully reviewed.Ongoing infection or repeatedunsuccessful staged reimplanta-tion procedures with persistentinfection are contraindicationsto this procedure. The inabilityto comply with postoperative

immobilization and a directedphysical therapy program arealso contraindications. In theseinstances, bracing and nonop-erative treatment or knee ar-throdesis are discussed with thepatient. If the patient is a candi-date for surgery, the procedureand postoperative rehabilitationare discussed preoperatively. If

FIG. 6

A saw is used to split the patella from anterior to posterior in a longitudinal fashion, in

line with the extensor mechanism arthrotomy.

FIG. 5

The remaining patella or

remnant is split in the

midline with a saw, in line

with the proximal and dis-

tal split.


5/20


the patient has an intact nativepatella and a deficient patellartendon, alternative allograft ex-tensor mechanism reconstruc-tions with the use of an Achillestendon-allograft calcaneal boneblock 8 may also be considered.We always plan to be preparedto revise and address malrotatedtotal knee arthroplasty com-ponents at the time of revisionsurgery.

Allograft Extensor MechanismPreoperatively, we order an al-lograft extensor mechanism ofthe entire knee that includes thetibia or a large portion of theproximal part of the tibia, thepatellar tendon, the patella, andat least 5 cm of quadriceps ten-don (Fig. 2). The allografts arefresh-frozen, nonirradiated spec-imens (Allosource, Centennial,

Colorado). We prefer the fresh-frozen over the freeze-dried al-lografts, given the results previ-ously described by Emerson etal.2,3 and concerns that freeze-drying may weaken the allografttissue, leading to complicationsand failure. The potential to gen-erate a greater risk of a host im-mune response than occurs withfresh-frozen specimens has also

been a concern. Before the pa-tient comes into the operatingroom and before the inductionof anesthesia, we visually inspectthe allograft to ensure that thereis an adequate specimen. Specifi-cally, there must be a proximaltibial allograft that will allowa bone-block harvest of at least5 cm attached to the patellar ten-

FIG. 8

Two sleeves of soft tissue are

reflected off the proximal part

of the tibia in the region of the

tibial tubercle, again maintain-

ing two flaps for later closure.

FIG. 7

The soft tissues around the patella are preserved in continuity with the retinaculum on

the medial and lateral sides of the two fragments.


6/20


don and at least 5 cm of allograftquadriceps tendon proximally.

SURGICAL TECHNIQUE

Patient Positioning We place the patient supine onthe operating table, with a sterilepneumatic tourniquet aroundthe thigh and a padded bump be-neath the trochanter. The leg isprepared and draped free, andthe foot is held in a leg holderduring the procedure to allow

variable amounts of flexion andextension.

Exposure of the Knee The pneumatic tourniquet is in-flated after exsanguination withan Esmarch bandage and flexionof the knee. Previous incisionsare marked (Fig. 3). We prefer amidline skin incision; however,

use of a previous incision is rec-ommended when present. Ifmultiple incisions are present,we use the most lateral incisionclosest to the midline, in orderto preserve blood supply to theskin. Often these are knees thathave had multiple operationsand may have undergone a previ-ous gastrocnemius flap or othersoft-tissue coverage procedure.In this instance, we are carefulnot to disrupt the blood supplyto this coverage and we have aplastic reconstructive surgeonavailable to assist during the ex-posure. The dissection is carrieddown in the midline with con-servative elevation of skin andsubcutaneous flaps. The retinac-ulum and extensor mechanismare then exposed. A midline inci-

FIG. 10

Removal of a malrotated femoral component. If malrotated components are left unad-

dressed, extensor mechanism maltracking will continue, with increased stress on the al-

lograft and early failure.

FIG. 9

Medial and lateral sleeves have been created, allowing direct exposure to the implants

and the anterior aspect of the tibia and tubercle.


7/20


sion is performed through theremaining extensor mechanism(the quadriceps tendon and pa-tellar tendon or scar tissue), cre-ating medial and lateral flaps ofretinaculum and exposing the joint (Fig. 4). Culture specimensare obtained and sent to the mi-crobiology laboratory, and syn-ovial fluid is assessed for cellcount. If there is a native patellaor a remnant, it is osteotomizedin a longitudinal fashion in themidline (Fig. 5), in line with themidline soft-tissue retinacularincision (Figs. 6 and 7). The pa-tellar bone is then shelled outand carefully removed, preserv-ing the soft tissues in continuitywith the medial and lateral reti-nacular flaps. This bone is keptfor autogenous bone graft as nec-essary. The medial and lateralgutters and suprapatellar pouchare recreated. The midline inci-

sion is carried proximally intothe host quadriceps, again main-taining a medial and lateralsleeve of tissue for later closure.The midline incision is carriedover the host tibial tubercle withelevation of medial and lateralsoft-tissue flaps (Figs. 8 and 9).

Total Knee ArthroplastyComponent Revision

and ReimplantationRevision total knee arthroplastythen proceeds as necessary. Ro-tation of the femoral and tibialcomponents is assessed, and ourthreshold for revision of malro-tated components is very low,as they can contribute to exten-sor mechanism maltracking (Fig.10). Balancing of flexion and ex-

FIG. 11

The allograft tibial block is marked for a rectangular cut of 6 to 8 x 2 x 2 cm.

FIG. 12

The rectangular block is then marked for a later bevel cut proximally to create the dovetail.


8/20


tension gaps is then performed,with careful attention to obtain-ing full passive extension of theknee. Trial components are re-moved, and definitive compo-nents are implanted in a routinefashion. The final polyethyleneliner is inserted prior to inser-tion of the extensor mechanismallograft. We have used this pro-cedure with primary cruciate-retaining, posterior stabilized,revision constrained condylardesigns, and constrained hingeknee designs. If stemmed com-ponents are being inserted, itmay be preferable to prepare thehost tibial bone trough and placethe fixation wires through thetibia at this stage, followed byinsertion of the stemmed tibialcomponent.

The revision total knee ar-throplasty implants are now inplace, and the host tissues are

next prepared to accept the ex-tensor mechanism allograft.

Allograft Preparationon the Back Table Simultaneous with the revisionor placement of the total kneearthroplasty components, theallograft specimen may be pre-pared on the back table. The hosttibial trough is not made until

we have harvested the allografttibial block, in order to ensurea press-fit of our allograft tibialblock. We first mark with amarking pen over the allografttibial tubercle and proximal partof the tibia our planned harvestof the allograft tibial bone block,in a rectangular fashion. Thelength of the block should be ap-

FIG. 13

Photograph of the rectangular tibial cut marked for the finishing bevel cut.

FIG. 14-B

Oblique view.

FIG. 14-A

Figs. 14-A through 14-D Finished cut of the tibial allograft segment with a proximal dove-

tail cut. Fig. 14-A Lateral view.


9/20


proximately 6 to 8 cm from thetibial articular surface of the al-lograft to the distal cut. The widthof the block is 2 cm, and the depthis 2 cm (Fig. 11). We cut on theconservative side and make thecuts slightly larger if necessary, asthese may be trimmed or down-

sized as needed. With use of asmall thin microsagittal saw, theallograft block is harvested fromthe allograft tibia (Fig. 12), withcareful attention so as not to dam-age the allograft patellar tendon(Fig. 13). The proximal bevel or“dovetail” on the allograft boneblock is not created during thispart of the harvest, as it is simpler

to perform once the graft has beenremoved from the allograft tibia.

Once the allograft boneblock has been carefully re-moved from the allograft tibia,we next prepare the bevel, ordovetail, on the proximal aspectof the removed bone block (Fig.

12). This serves two purposes.The first is to lock into the hostnative tibial trough and avoidgraft escape. The second is to al-low a press-fit of the graft intothe native tibia. Using a markingpen (Fig. 13), we draw an angleof 30° to 40° (from the perpen-dicular of the graft) as a beveland cut it carefully with the thin

saw blade. The length of the bevelis approximately 20 to 25 mm(Figs. 14-A through 14-D).

Two number-2 nonabsorb-able sutures are then placed in arunning, locked fashion, as de-scribed by Krackow et al.9, alongthe medial and lateral aspects of

the allograft quadriceps tendon,exiting out proximally. These su-tures are kept long, and they areplaced so that the assistant canapply tension and pull the al-lograft tightly proximally onceit has been secured into the pre-pared tibial bed.

The graft and the proximaltwo sutures are then placed care-

FIG. 14-C FIG. 14-D

Fig. 14-C Posterior aspect. Fig. 14-D Anterior aspect.


10/20


fully in a basin on the back table,and attention is turned to thepreparation of the proximal partof the host tibia.

Preparation of the HostProximal Tibial Trough

Using a marking pen, we markout the host proximal tibialtrough (Fig. 15). We typically at-tempt to place the allograft tibialtubercle in a position that is closeto, or slightly medial to, the posi-tion of the native tibial tubercle.In addition, we attempt to leaveat least 15 mm of host bone in-tact below the tibial component

anteriorly to resist proximal mi-gration or escape of the graft, al-though this 15 mm of bone is notalways possible in the revisionsetting with associated bone loss.The rectangular tibial trough isthen marked out for a length of

5 cm and a width of just less than2 cm and a depth of 2 cm. Proxi-mally, the host bone is beveled(Fig. 16) to accept a press-fit ofthe beveled, or dovetailed, allo-graft bone block (Fig. 17). Thisbevel in the host bone should becreated with dimensions slightlysmaller than the allograft boneblock, in order to allow a press-

fit (Fig. 18). Two or three 18-gauge stainless steel wires arethen placed through drill-holesin the tibia from medial to lat-eral (Fig. 19). These wires mustpass deep to the tibial trough. If astemmed tibial component is be-

ing used, it is easier to drill andplace these wires prior to insert-ing the stemmed component.The allograft extensor mecha-nism is then inserted into thehost tibial trough and is gentlypress-fit with a bone tamp orpunch, in an “up and in” fashion,in order to lock the dovetail inplace. The wires are then twisted,

FIG. 15

The host proximal tibial trough is marked. Careful attention to the location of this trough in theregion of the tibial tubercle or slightly medial to it will allow for improved patellar tracking.

CRITICAL CONCEPTS

INDICATIONS:

• Disruption of the extensor

mechanism (extensor lag)

that is not amenable to or

has failed a primary repair

• Patellar tendon rupture, avul-

sion, or prior excision

• Quadriceps tendon rupture,

avulsion, or prior excision

• Patellar fragmentation or

nonreconstructible patellar

fracture

• Severe heterotopic ossification

of the extensor mechanism

• Previous patellectomy with

a total knee arthroplasty

and symptomatic extensor

lag

• Severe patella infera and ar-

throfibrosis of the extensor

mechanism

• Conversion of previous knee

arthrodesis to a total knee

replacement with a fibrosed or

deficient extensor mechanism

continued


11/20


FIG. 16

The proximal host tibial trough is also beveled under to allow for a locking fit of the al-

lograft tibial block. A small curet is useful to complete this bevel.

tightened, cut, and bent overagainst bone to avoid irritationto the soft tissues (Fig. 20). Al-ternatively, a small-fragment

cortical screw and washer maybe added to the fixation at thesurgeon’s preference. This cre-ates a drill-hole in the allograft,and we prefer to avoid this stress-riser, despite the added securityof the screw fixation.

Once we have secured theallograft bone into the host tibia,attention is turned to the proxi-

mal quadriceps medial and lat-eral sleeves and retinaculum.

Preparation and Tensioning

of the Host Distal QuadricepsSimilar to the retention suturesplaced in the allograft quadri-ceps, the host distal quadricepsmedial and lateral soft-tissuesleeves are prepared. We again usea number-2 nonabsorbable suture(FiberWire; Arthrex, Naples, Flor-ida) and place a short runningKrakow suture into both the me-

dial and lateral retinaculum in thedistal quadriceps muscle-tendon junction. This allows a secondassistant to “pull down” the hostquadriceps mechanism (Fig. 21),effectively tensioning the distalhost extensor mechanism (Fig.22). The two previously placed

allograft quadriceps sutures arepulled tightly with the knee in fullextension (Figs. 23-A and 23-B).With use of a suture passer, thesesutures are then pulled from distalto proximal, out and up throughthe more proximal host quadri-ceps. This pulls the allograftquadriceps up and under the hostquadriceps, and simultaneouslypulls or tensions the host quadri-

ceps distally (Fig. 24). With thistension maintained, the allograftis then sutured in place beneaththe host quadriceps with number-5 nonabsorbable suture, in a“vest-over-pants” fashion (seeFig. 24). Throughout this suturerepair, the two assistants main-tain tension on their respectiveretention sutures, in order to

CRITICAL CONCEPTS | continued

CONTRAINDICATIONS:

• Ongoing infection or concur-

rent infection of a total knee

replacement at or near the

operative site

• Reconstructible extensor

mechanism with primary

repair or local autogenous

reinforcement tissue

• An unreliable, noncompliant

patient who is unable to co-

operate with postoperative

rehabilitation

continued


12/20


FIG. 17

Completed anterior tibial trough, which is ready to accept the allograft extensor mechanism.

FIG. 18

The allograft tibial block is press-fit into the host tibial trough.


13/20


FIG. 19

Fixation of the tibial allograft with stainless steel wires, which are drilled through the tibia, beneath the allograft.

FIG. 20

The wire fixation is secured after insertion of the allograft bone into the host tibia.


14/20


maintain tension with the kneein the extended position. Oncethe proximal aspect of the al-lograft is secured, the repair iscontinued along the medial andlateral sides. However, the repairis performed with the host reti-

naculum brought over the topof the allograft, in order to coverthe allograft tissues as much aspossible with the medial and lat-eral sleeves of the host retinacu-lum. We find that we are usuallyable to completely cover the al-lograft with these host sleevesthat have been preserved, in ad-dition to suturing the allograft

underneath these tissues (Fig.25). Distally, the host tissues areclosed over the wires and al-lograft bone block.

CLOSURE

We prefer to not flex the knee to

“test” our repair once it is com-pleted. This should be avoided inorder to not stress the repair andattenuate the allograft host junc-tion. The subcutaneous tissuesare closed in routine fashion. Theskin is closed with staples. If theskin over the distal incision istenuous, nonabsorbable suturemay be used.

FIG. 21

Two running, locked Krakow sutures are placed into the medial and lateral host quadri-

ceps retinaculum.


PITFALLS:

• A fresh-frozen, nonirradiated

allograft specimen consist-

ing of a quadriceps tendon,

patella, patellar tendon,

and tibial bone is required.

It is preferable to have at

least 5 cm of quadriceps

tendon allograft for suture re-

pair into the host quadriceps

mechanism.

• We recommend use of a mid-

line approach through the ex-

tensor mechanism anteriorly.

Large medial and lateral flapsthat provide excellent tissue

for closure over the extensor

mechanism allograft are devel-

oped. If there is native patella

remaining, this is osteoto-

mized transversely in line with

the midline arthrotomy. The pa-

tellar remnant is then shelled

out and removed.

• Component revision is often

necessary. It is important

that the knee be able to bepassively brought to full ex-

tension with the trial implants

in place, in order to ensure

full extension is attainable

postoperatively.

• It is important that the proxi-

mal aspect of the allograft tib-

ial bone and the bone trough

on the native tibia be dove-

tailed in order to lock, or

press-fit, the allograft into the

native tibia and resist proxi-mal migration.

• When the allograft is su-

tured proximally into the

native quadriceps, tension

must be maintained on the al-

lograft with the knee in full

extension.

continued


15/20


POSTOPERATIVE CARE AND

REHABILITATION

In the operating room at thecompletion of the procedure, theknee is placed in full extension.We prefer to use a knee immobi-lizer that is customized to the sizeand diameter of the extremity.

This allows for complete immo-bilization of the knee in full ex-tension and permits access to thewound postoperatively. A poor-fitting brace allows for flexionand movement, which should beavoided in the immediate post-operative period. Alternatively,a cylindrical fiberglass cast maybe placed on the limb in the op-


• It is not desirable to have

an overly long allograft quadri-

ceps tendon. A segment that

is too long will end up being

sewn proximally into the rec-

tus femoris muscle instead

of into the host quadriceps

tendon.

• The host retinaculum medial

and lateral flaps should be

sewn over the allograft as

much as possible in order to

cover the allograft.

• The knee should not be flexed

intraoperatively to assess the

flexion of the construct. The

patient is managed with immo-

bilization of the knee in full ex-

tension with touch-down

weight-bearing for eight weeks,

and then a directed physical

therapy program is begun.

• The allograft patella is not re-

surfaced in order to avoid cre-

ating a stress-riser in it.

continued

FIG. 23-A

Two sutures placed in the allograft quadriceps allow the allograft to be tensioned proximally.

FIG. 22

These two sutures allow the host extensor mechanism to be tensioned by pulling distally.


16/20


FIG. 23-B

The host extensor mechanism and allograft are pulled by two separate assistants into an extension position.

FIG. 24

The sutures in the allograft quadriceps are pulled under the host quadriceps and out proximally through the host extensor mechanism.


17/20


erating room. The disadvantageof the cast is that it must be re-moved if there are concernsabout the wound and in orderto change the dressing postoper-atively. In a patient with border-line compliance, this is the safestform of immobilization, but it isoften poorly tolerated.

Postoperative physical ther-apy follows a protocol that we

developed for this procedure.Patients are maintained with theknee in full extension for eightweeks after surgery. During thisperiod, we allow touch-downweight-bearing only. We have thepatient avoid full weight-bearingin order to reduce the quadri-ceps force on the tibial tubercleand the allograft-host soft-tissue

repair. We do not allow any flex-ion during this eight-week pe-riod. We encourage isometricstatic quadriceps contractions.After eight weeks, 30° of active

flexion is permitted, under thesupervision of a physical thera-pist, with the patient wearing ahinged knee brace with a lock-out against further flexion. Simi-

FIG. 25

The host medial and lateral sleeves of retinaculum, if preserved during the surgery, serve to cover the allograft completely, reducing the ex-

posure of the allograft to the subcutaneous tissues.


AUTHOR UPDATE:

This technique has not been modified since the publication of our original

study. We emphasize that success with this technique requires that several

critical aspects be carefully followed. The midline incision and retention ofhost medial and lateral retinacular tissue is important. Removing the patel-

lar remnant in this way ensures that medial and lateral flaps remain for

closure, and it improves exposure. Tensioning the allograft tightly in full ex-

tension is necessary to help to reduce the risk of allograft attenuation and

extensor lag. Closure of the medial and lateral flaps over the allograft as

much as possible reduces the contact of the allograft with subcutaneous

tissues and, we believe, reduces the risk of infection. We emphasize that

we do not flex the repair once it is completed, as has been recommended

by other authors.


18/20


Figs. 26-A, 26-B, and 26-C Preoperative radiographs showing a patient with a deficient extensor mechanism (patellar tendon atten-

uation) and component malrotation with lateral dislocation of the host extensor mechanism.

FIG. 26-C

FIG. 26-A FIG. 26-B


19/20


Figs. 26-D, 26-E, and 26-F Radiographs made after component revision and reconstruction with an extensor mechanism allograft.

FIG. 26-D

FIG. 26-E

FIG. 26-F


20/20


larly, at eight weeks, patients areadvanced to weight-bearing astolerated. During weight-bearing,we lock the brace in full exten-sion. At twelve weeks, we allowfurther active flexion up to amaximum of 90°, and gentlequadriceps strengthening exer-cises are initiated. Passive flexionis not permitted in order to min-imize the chance of graft failureand early attenuation. Patientsare evaluated at six months andthen on a yearly basis (Figs. 26-Athrough 26-F).

NOTE: The authors thank Regina M. Barden, RN, and Marga-ret Arp for their contribution to the preparation and clinicalsupport for this study.

R. Stephen J. Burnett, MD, FRCS(C)

Department of Orthopaedic Surgery, Barnes

Jewish Hosp ital , Washington Univer sity, 660

South Euclid Avenue, Campus Box 8233,

St. Louis, MO 63110. E-mail address:

[email protected]

Richard A. Berger, MD

Craig J. Della Valle, MD

Scott M. Sporer, MD

Joshua J. Jacobs, MDWayne G. Paprosky, MD

Aaron G. Rosenberg, MD

Rush University Medical Center, Midwest Ortho-

paedics, 1725 West Harrison Street, Suite 1063,

Chicago, IL 60612

In support of their research or preparation of thismanuscript, one or more of the authors received

grants or outside funding from Zimmer. In addi-

tion, one or more of the authors received pay-

ments or other benefits or a commitment or

agreement to provide such benefits from a com-

mercial entity (Zimmer). Also, a commercial entity

(Zimmer) paid or directed, or agreed to pay or

direct, benefits to a research fund, foundation,

educational institution, or other charitable or non-

profit organization with which the authors are affil-

iated or associated.

The line drawings in this article are the work of

Jennifer Fairman ([email protected]).

doi:10.2106/JBJS.E.00442

REFERENCES1. Leopold SS, Greidanus N, Paprosky

WG, Berger RA, Rosenberg AG. High rate

of failure of allograft reconstruction of

the extensor mechanism after total knee

arthroplasty. J Bone Joint Surg Am. 1999;

81:1574-9.

2. Emerson RH Jr, Head WC, Malinin TI. Re-

construction of patellar tendon rupture aftertotal knee arthroplasty with an extensor

mechanism allograft. Clin Orthop Relat Res.

1990;260:154-61.

3. Emerson RH Jr, Head WC, Malinin TI. Ex-

tensor mechanism reconstruction with an

allograft after total knee arthroplasty. Clin

Orthop Relat Res. 1994;303:79-85.

4. Nazarian DG, Booth RE Jr. Extensor mecha-

nism allografts in total knee arthroplasty. Clin


5. Burnett RS, Fornasier VL, Haydon CM, We-

hrli BM, Whitewood CN, Bourne RB. Retrieval

of a well-functioning extensor mechanism al-

lograft from a total knee arthroplasty. Clinical

and histological findings. J Bone Joint Surg

Br. 2004;86:986-90.

6. Burnett RS, Berger RA, Paprosky WG, Della

Valle CJ, Jacobs JJ, Rosenberg AG. Extensor

mechanism allograft reconstruction after to-

tal knee arthroplasty. A comparison of twotechniques. J Bone Joint Surg Am. 2004;

86:2694-9.

7. Berger RA, Crossett LS, Jacobs JJ, Rubash

HE. Malrotation causing patellofemoral com-

plications after total knee arthroplasty. Clin


8. Crossett LS, Sinha RK, Sechriest VF,

Rubash HE. Reconstruction of a ruptured pa-

tellar tendon with achilles tendon allograft

following total knee ar throplasty. J Bone Joint

Surg Am. 2002;84:1354-61.

9. Krackow KA, Thomas SC, Jones LC. Liga-

ment-tendon fixation analysis of a new stitch

and comparison with standard techniques.Orthopedics. 1988;11:909-17.

Documents

Extensor Allograft Technique