RECONSTRUCTION 2023, Jan, 08 | 12:00 am

Extensor Mechanism Reconstruction in Total Knee Arthroplasty

Introduction

  • Complete disruption or total dysfunction of the extensor mechanism of the knee is fortunately quite rare, occurring in less than 1% of all primary and revision knee arthroplasties (Rand et al., 1989). Nonetheless, its restitution remains perhaps the greatest challenge in all of knee surgery.

  • Adequate functioning of the extensor is necessary not only to enable ambulation and stair climbing, but to preserve the very stability of the prosthesis itself. As with all orthopedic challenges, this problem is better avoided than resolved.

     

     

    P ITFALLS

    • An overall assessment of the patient’s likelihood of success should precede the allograft surgery. In many individuals the anterior skin is scarred, thin, or poorly vascularized. A bulk allograft will consume more space than the tissue it is replacing, and the viability of the skin is critical to the technical success of the procedure.

       

    • Many individuals are indeed compromised hosts, and the ability of the patient to withstand a substantial operation—but more importantly 12 weeks of absolute or relative immobilization—should be considered as well.

     

    Preoperative Considerations

  • Even the most concise discussion of extensor reconstruction must necessarily begin with the observation of several principles. Indeed, ignorance or rejection of these principles is often the primary reason for the initial extensor failure.

  • Extensor problems are largely vascular in origin. Although most involve some element of trauma, that insult is often iatrogenic.

    • Most of these patients have been the recipients of multiple surgical procedures. Even a successful primary total knee arthroplasty is usually accompanied by a medial arthrotomy, a fat pad excision, and occasionally a lateral retinacular release, all of which serve to devascularize the peripatellar tissues.

    • Each subsequent surgery usually serves largely to further decrease the blood supply to this critical area. “Less is truly more” in peripatellar surgery.

      Controversies

      • In some instances, fusion or amputation may be preferable to the heroics of a revision and extensor allograft insertion.

       

  • The forces visited upon the knee during stair climbing and arising from chairs often approximates 7–8 times body weight. Simple reconstructive techniques, therefore, are often doomed to failure by the enormous tension placed on the tissues. Even a limb held in full extension will experience forces 1.5 times body weight simply moving about in bed. Although many varied techniques and different materials have been proposed to facilitate direct primary repair of tendon disruptions, their results are uniformly dismal (Rand et al., 1989).

  • In the author’s experience, well over half of all extensor tendon disruptions occur in the presence of a technically unsatisfactory arthroplasty (Fig. 1).

     

     

     

    407

     

    Extensor Mechanism Reconstruction

     

    FIGURE 1

     

    • Instability in flexion is the most common problem, with anterior subluxation or dislocation of the femur on the tibia resulting in damage to the patellar tendon.

    • Stiff arthroplasties—from component oversizing or malpositioning or joint line irregularities—may undergo manipulations and other therapies that rupture the tendon in either the patellar or quadriceps portion.

    • One must be prepared to revise the preceding arthroplasty as well as repair the extensor tendon to avoid recapitulating the inciting event. It is usually the case that an increased level of constraint is necessary for each subsequent arthroplasty for instability.

      • One must be particularly scrupulous about the possibility of infection in the index arthroplasty.

    • Our overall accuracy in determining the presence of infection is still only approximately 90%,

      and thus revisions, allografts, and exchange arthroplasties are often performed unwittingly in the face of low-grade covert sepsis.

    • The insertion of a large bulk allograft into an infected host would likely transform that covert infection to an apparent one.

       

      408

       

       

      P EARLS

      • One must remember to save every molecule of host tissue, since attempts to excise calcific deposits or patellar fragments can only further devascularize the anterior structures of the knee. Minimal dissection is the order of the day.

       

      Extensor Mechanism Reconstruction

       

      Portals/Exposures

      • The surgical approach for an extensor repair merits preoperative consideration. If there is some residual possibility that the existing extensor mechanism can be repaired or reconstituted, then the previous median parapatellar arthrotomy should be employed.

      • If one is already committed to an extensor allograft, it is frequently easier and better to use a direct midline arthrotomy, splitting the quadriceps and patellar tendons and shucking out the residual patellar bone (Fig. 2). This allows a more open exposure to the knee, less dissection and disruption of the periarticular vascularity, and better margins on both sides of the tendon for reapproximation to the allograft.

         

        P EARLS

        • Many extensor allografts are a complement to even a hinged prosthesis.

         

        Procedure: Autograft Repair

      • Insertion of an extensor allograft remains the definitive treatment for anterior deficiency. Before finally capitulating to an anterior allograft, however, one may choose to consider one of two autografting techniques available for supplementing or substituting for extensor dysfunctions.

      • The most popular technique, described by Cadambi and Engh (1992), involves the use of the semitendinosus tendon. This requires an extensive and difficult dissection, the use of small tendinous strands when compared to the bulk of the natural extensor mechanism, and is fraught with patellar fractures, even in the originators’ hands.

      • The author’s preference for autografting is the use of the iliotibial band, which can be harvested from the lateral side of the knee using a tendon stripper

        (Fig. 3A). A very long and thick strip of tissue can be obtained, woven above the patella, and secured to the medial tibial bone (Fig. 3B).

      • If neither of these techniques appears appropriate, one should proceed directly and without hesitation to a full extensor tendon allograft.

    • Redundant anterior scar tissue should not be excised but rather enfolded in the final suture repair. A useful technique is to tubularize the patulous anterior soft tissues with a vertical running mattress suture (Fig. 4). This creates a cable of extensor tissue that mimics the natural extensor and enhances trochlear tracking.

       

       

       

       

       

      Extensor Mechanism Reconstruction

       

       

       

       

       

       

       

      409

       

      FIGURE 2

       

      A

      FIGURE 3

       

      B

       

      FIGURE 4

       

       

    • Redundant adjacent tissue can be “borrowed” from the proximal quadriceps mechanism and turned down to close transverse defects (Fig. 5).

      • Running or individual sutures will then close the margins of the host tissue, forming a cordlike structure anterior to the allograft and sealing the joint.

    • One should never forsake the opportunity to advance the vastus medialis obliquus over the repair (Fig. 6). This technique serves to improve strength and extensor tracking and also to bring more vascularized tissue into the anterior aspect of the knee.

       

       

       

      410

       

      Extensor Mechanism Reconstruction

       

      FIGURE 5

       

       

       

       

      FIGURE 6

       

      411

       

      Extensor Mechanism Reconstruction

       

      • All soft tissue repairs need to be protected from the extreme force on the anterior knee. It is the author’s preference to use nonabsorbable Mercelene tape in a figure-of-8 configuration rather than pins and wires, which will fragment and probably require subsequent excision

        (Fig. 7).

         

        P EARLS

        • There is no evidence to suggest the merit of donor/host matching in terms of gender, size, or even handedness.

         

        Procedure: Allograft Repair

        Step 1: Allograft Materials

        • Once the surgery has been scheduled, availability of allograft material should be determined.

        • The preparation and sterilization of the allograft are relatively uniform at most tissue banks. However, as a general rule those receiving less irradiation—as with anterior cruciate ligament grafts—will probably have a better record of strength and survival in the future.

        • The graft itself (Fig. 8) demands some specificity.

      • The patellar bone should be intact, as attempts to shave the sides (to allow better infiltration of sterilization chemicals) would weaken the construct.

      • The tibial bone at the distal end of the graft should be at least 6 cm in length and should include the donor bone all the way up to the tibial plateau. This allows the fashioning of a “dovetail” (to be described later) for greater security.

      • Most important, the proximal or quadriceps portion of the graft should include at least 15 cm of donor tendon—or more. Attached muscle can be débrided by the surgeon at the time of implantation, but lesser lengths will not allow adequate material for fixation and ingrowth.

         

         

         

         

        FIGURE 7

         

         

         

        412

         

        Extensor Mechanism Reconstruction

         

        FIGURE 8

         

    • Since there is a good deal of handling and sculpting of these tissues, it is probably prudent to have two grafts available at the time of surgery, in case one graft suffers a mishap.

       

      P EARLS

      • One of the great conveniences of this technique is that the patella can be made to match the host joint line, rather than vice versa as in most revisions.

         

      • The importance of tightening the tendon with the limb in full extension cannot be

        underestimated, as this has been the major failing of the allograft technique in other hands.

         

        P ITFALLS

      • The use of screws for additional fixation of the tibial allograft

      is problematic because of the tendency to split the small and brittle tibial graft and because of the difficulty in directing screws around a revision stem or adjacent methylmethacrylate.

       

      Step 2: Revision Arthroplasty

      • If the antecedent arthroplasty is defective in any regard, a revisional procedure should be performed at this time.

      • It is usually the case that increased levels of constraint are applied with each successive prosthesis.

      • It is the author’s unsubstantiated opinion that, as with other extensor issues such as patellectomy, no less than a posterior stabilized and probably a cruciate condylar constrained prosthesis should be

        used to protect the allograft from anterior/posterior forces.

        Step 3: Extensor Tendon Allografting

      • After determining the appropriate patellar height relative to the prosthetic joint line, the fully thawed allograft is stretched out over the anterior surface of the knee.

      • The relative size and position of the tibial bed should be evaluated, selecting a site slightly medial to the prominence of the host tibial tuberosity to ensure proper patellar tracking.

      • The donor tibial bone is then marked and trimmed, using a small oscillating saw blade, to create a

       

      413

       

      Extensor Mechanism Reconstruction

       

      rectangular block with a proximal dovetail (Fig. 9A). This structure can then be traced with a marking pen or electrocautery on the host bone (Fig. 9B), and a bed excavated to match the graft.

      • The dovetail is a crucial portion of this effort, as it allows the graft to lock into the tibia and to be further secured by future longitudinal tension. The author’s preferred technique is also to leave the graft slightly longer than the bed, to impact the proximal end temporarily, and then to trim the distal end obliquely so that its impaction will further secure the graft against the host bone.

      • This construction is very secure, so that one is tempted not to use further fixation. However, the use of multiple wires (Fig. 10)—either through the

         

         

         

        A B

        FIGURE 9

         

         

         

         

        FIGURE 10

         

        414

         

        Extensor Mechanism Reconstruction

         

        sides of the bed, around the stem of a revisional implant, or even around the entire posterior tibia— should be considered. It has been the author’s preference to use two or three wires to secure the tibial graft.

        • Fixation of the proximal end of the allograft is the most important challenge for surgical success.

    • The author’s preference is to put the limb at full extension and to expose the fullest extent of the medial and lateral halves of the bisected host quadriceps tendon.

    • A nonabsorbable stay suture is then implanted running from the vastus medialis portion of the quadriceps tendon distally, through the proximal allograft in a Krackow stitch, and then proximally once more up the lateral side through the residual vastus lateralis tendon.

    • An assistant can then hold this stay suture under tension with the limb in full extension while further sutures are used to close the residual host tissue over the tendon (Fig. 11). This is performed using a vertical mattress-type suture, first through the medial fragment of the quadriceps tendon, then

      1 cm distal and through the allograft quadriceps, then back 1 cm proximally through the residual vastus lateralis tendon.

    • The offset provided by this suture technique allows further tensioning of the extensor with each suture (Fig. 12).

       

       

       

      FIGURE 11

       

       

       

      415

       

      FIGURE 12

       

      Extensor Mechanism Reconstruction

       

      Postoperative Care and Expected Outcomes

      • Postoperatively, the patient is placed in a cast or brace in full extension for 6 weeks, with no attempts at quadriceps strengthening. One must remember that this is a “bailout” procedure, and the desire to regain early motion or restore quadriceps strength should be subjugated to the imperative of restoring extensor continuity.

      • After 6 weeks, a brace will allow progressive 30° increases in range of motion at 2-week intervals. The priority at this point is ingrowth and incorporation of the graft, and thorough education of physical therapists involved in the patient’s care is critical to avoid inappropriate efforts for motion and strength.

      • Success rates for this technique of extensor allograft in total knee arthroplasty have been reported in the 90% range, although almost every series describes a mild to moderate extensor lag, even in the most

        successful cases (Burnett et al., 2004, 2005; Nazarian and Booth, 1997).

        • The credit for the first description of the allograft technique properly belongs to Emerson et al. (1990), who described success in 12 of 15 patients.

        • Nazarian and Booth (1997) presented a series of 36 of 38 successes, although 6 of these required secondary “adjustments” of the original graft.

        • Rosenberg and Berger reported a modification of the technique in which the graft was sutured in 30° of flexion to avoid loss of motion. None

           

          416

           

          Extensor Mechanism Reconstruction

           

          of their seven patients had a successful result (Leopold et al., 1999). Using the full extension technique described above, they reported another series of patients with the tendon sutured in full extension and with a success rate of 13/13 (Burnett et al., 2004, 2005).

    • At the time of this writing, the author’s experience includes 214 allografts, 2 of which were in natural knees and 212 in arthroplasties (Fig. 13). Of the 212 knees with arthroplasties, 131 required a revision coincident with the allograft insertion. There were 12 failures, 10 from failed incorporation of the proximal soft tissues into the quadriceps tendon. Of the two distal failures, one was from precipitous flexion of the knee by a physical therapist who displaced the graft and one was a rupture of the patellar tendon while kneeling at church. Even the most successful allografts had a 5–10° extensor lag, usually stabilizing 6–12 months after surgery. All of the patients with successful grafts were able to arise from a chair and climb and descend stairs. As a group, they are remarkably pain-free, although the denervation

of the graft may be the primary reason for this fortunate result.

  • These results of extensor tendon allografts compare quite favorably with the relatively disappointing results of primary extensor repairs. Only the enormity of the procedure and the extensive recuperation discourage the use of allografts as a primary

 

 

 

 

FIGURE 13

 

417

 

Extensor Mechanism Reconstruction

 

technique for extensor disruptions. The success of this approach, even in the face of multiple prior surgeries, infections, and deficient soft tissues, as well as the ability to regain limb extension and to stabilize the prosthetic arthroplasty, make this technique highly rewarding, particularly when compared with fusion or amputation.