Extensor Mechanism Allograft Reconstruction for Extensor Mechanism Failure Following Total Knee Arthroplasty

Transcription

1 279 COPYRIGHT Ó 2015 BY THE JOURNAL OF BONE AND JOINT SURGERY, INCORPORATED A commentary by Robert Booth Jr., MD, is linke to the online version of this article at jbjs.org. Extensor Mechanism Allograft Reconstruction for Extensor Mechanism Failure Following Total Knee Arthroplasty Nicholas M. Brown, MD, Trevor Murray, MD, Scott M. Sporer, MD, Nathan Wetters, MD, Richar A. Berger, MD, an Craig J. Della Valle, MD Investigation performe at the Rush University Meical Center, Chicago, Illinois Backgroun: Extensor mechanism isruption following total knee arthroplasty is a rare but evastating complication. The purpose of this stuy was to report our experience with extensor mechanism allograft reconstruction for chronic extensor mechanism failure. Methos: Fifty consecutive extensor mechanism allograft reconstructions were performe in forty-seven patients with a mean age of 67.6 years who were followe for a mean time of 57.6 months (range, twenty-four to 125 months). The operative technique inclue the use of a fresh-frozen, correctly size full extensor mechanism allograft that was tensione tightly in full extension. Patients were evaluate clinically with use of the Knee Society score, an reconstructions were consiere failures if the patient ha a score of <60 points or a recurrent extensor lag of >30 or if they require revision or removal of the allograft. Results: Nineteen reconstructions (38%) were consiere failures, incluing four revise to a secon extensor mechanism allograft ue to failure of the allograft, five for eep infection, an ten consiere clinical failures seconary to a Knee Society score of <60 points or an extensor lag of >30. The mean Knee Society score improve from 33.9 to 75.9 points (p < ). The estimate Kaplan-Meier survivorship with failure for any reason as the en point was 56.2% (95% confience interval, 39.4% to 70.1%) at ten years. Conclusions: Extensor mechanism isruption following total knee arthroplasty is a ifficult complication to treat, with moest outcomes. Extensor mechanism allograft reconstruction is a reasonable option; however, patients must be informe regaring the substantial risk of complications, an although initial extensor mechanism function may be restore, expectations regaring longer-term outcomes are more guare. Level of Evience: Therapeutic Level IV. See Instructions for Authors for a complete escription of levels of evience. Peer Review: This article was reviewe by the Eitor-in-Chief an one Deputy Eitor, an it unerwent bline review by two or more outsie experts. The Deputy Eitor reviewe each revision of the article, an it unerwent a final review by the Eitor-in-Chief prior to publication. Final corrections an clarifications occurre uring one or more exchanges between the author(s) an copyeitors. Extensor mechanism failure following total knee arthroplasty is a rare 1-3 but serious complication that leas to substantial patient morbiity. Patients often have ifficulty walking, given the loss of active knee extension, an instability that typically manifests as recurrent falls. There are several ifferent options for treatment incluing use of a brace 4 ; Disclosure: None of the authors receive payments or services, either irectly or inirectly (i.e., via his or her institution), from a thir party in support of any aspect of this work. One or more of the authors, or his or her institution, has ha a financial relationship, in the thirty-six months prior to submission of this work, with an entity in the biomeical arena that coul be perceive to influence or have the potential to influence what is written in this work. No author has ha any other relationships, or has engage in any other activities, that coul be perceive to influence or have the potential to influence what is written in this work. The complete Disclosures of Potential Conflicts of Interest submitte by authors are always provie with the online version of the article. J Bone Joint Surg Am. 2015;97:

2 280 primary repair 5 ; an augmentation with native tissue 6,7, allograft 8,9, autograft 1, or synthetic materials 10. The results of operative treatment have been variable with many series reporting poor outcomes 1-3,11-14 ; however, nonoperative treatment is typically not well tolerate. Hence, the optimal form of management is controversial. Prior stuies showing the results of operative treatment in general have escribe small cohorts with relatively short follow-up. We previously reporte the results of twenty consecutive complete extensor mechanism allograft reconstructions 15 that inclue the tibial tubercle, patellar tenon, patella, an quariceps tenon. Seven knees unerwent reconstruction utilizing the technique of Emerson et al. 16,17, in which the allograft was tensione to allow for 60 of passive intraoperative flexion, an all of these reconstructions went on to failure. Thirteen knees unerwent reconstruction using the technique of Nazarian an Booth 18, in which the allograft was tensione tightly in full extension. With a mean follow-up of thirty-seven months, all of these reconstructions were consiere successes. The purpose of the present stuy was to report the results of a larger series of complete extensor mechanism allografts that were performe with the allograft tensione tightly in full extension. We were particularly intereste to etermine whether our originally reporte results were urable over time, what the most common complications were, an if there were any technical or patient-relate factors that preicte success or failure. Materials an Methos Sixty-four consecutive knees reconstructe with a complete extensor mechanism allograft that was tensione tightly in full extension (see Appenix) were performe in sixty-one patients by four surgeons (three of whom [C.J.D.V., R.A.B., an S.M.S.] were authors in this stuy) at the same institution following institutional review boar approval. All knees ha extensor mechanism failure following total knee arthroplasty, an the mean patient age at the time of surgery was 67.6 years (range, forty-one to eighty-seven years). Of the thirteen patients escribe in our original report, eleven were inclue in this analysis as they ha ha aitional follow-up, one was exclue because the extensor mechanism allograft ha been performe as part of a two-stage exchange, an one was lost to follow-up. Aitional patients who were exclue from this analysis inclue five who ha unergone revision extensor mechanism allograft an three who ha unergone the proceure performe as part of a two-stage exchange. One aitional patient was lost to follow-up an three patients ie prior to the two-year follow-up, leaving fifty reconstructions in forty-seven patients (fifteen men an thirty-two women) available for stuy at a minimum two years of follow-up. Of the fifty reconstructions, treatment was for twenty-two patellar tenon tears; twelve quariceps tenon tears; eight patellar fractures that were unite an associate with incompetence of the extensor mechanism; six knee instabilities, extensor mechanism subluxations, an extensor lags without a iscrete extensor mechanism isruption (a prior patellectomy ha been performe in five knees); an two severe patella baja with associate stiffness. For patients with an extensor mechanism isruption, the mean time from iagnosis to final extensor mechanism allograft reconstruction was twenty-six months (range, one to 161 months). Clinical an raiographic evaluation was performe at six weeks, at twelve weeks, at six months, at one year, an yearly thereafter. Knee Society scores were obtaine at each visit. Hospital for Special Surgery knee scores were use in the original stuy 15 ; however, our institution has converte to using the Knee Society score an, therefore, we use this scoring system in the present stuy 19. Failure was efine as a Knee Society score of <60 points, an extensor lag of >30, or revision surgery requiring repeat extensor mechanism allograft reconstruction, amputation, or fusion. The operative technique was unchange from our prior report 15. Briefly, this inclue the use of a same sie, correctly size fresh-frozen extensor mechanism allograft incluing the proximal tibia. Twenty grafts (40%) were nonirraiate, an the balance, after 2005, was sterilize by the supplier (AlloSource, Chicago, Illinois) at 1.0 to 1.3 Mras on the basis of the U.S. Foo an Drug Aministration (FDA) recommenations. A trough was fashione in the proximal tibia to accept the tibial tubercle bone block, which was press-fit into place an then was fixe with two or three 16-gauge wires (forty-seven knees), two 6.5-mm cancellous screws (two knees), or two 16-gauge wires with one screw (one knee). The graft was then tensione tightly in full extension an was covere with as much host tissue as possible; the allograft patella was not resurface. The postoperative regimen inclue immobilization in extension for six weeks followe by progressive range of motion starting at 0 to 30 in a hinge knee brace avancing 10 per week. Seventeen (34%) of the fifty knees ha unergone a prior attempt at primary repair (mean, 1.4 prior attempts [range, one to three prior attempts]) an a concomitant revision of both components was performe in twenty-three knees (46%). Survivorship analysis with failure for any reason was calculate with use of the Kaplan-Meier metho. The Stuent t test was use to analyze continuous variables with significance set at p < Multivariate regression analysis was performe to ientify variables associate with failure. Variables examine inclue age, sex, location of tear, time between tear an reconstruction, time between inex proceure an reconstruction, an concomitant component revision. Source of Funing There was no external funing for this stuy. Results At a mean follow-up time of 57.6 months (range, twentyfour to 125 months), the mean Knee Society score improve from 33.9 to 75.9 points (range, 8 to 100 points) (p < ) (see Appenix). Twenty-one (44.7%) of forty-seven patients were walking with no assistive evice an twenty-five (50%) of fifty knees ha full or near full active extension (an extensor lag of <10 ), with an overall mean extensor lag of 13 (range, 0 to 90 ). Forty-six (92%) of fifty knees ha full or near-full active extension (extensor lag, <10 ) at some point in their early postoperative course (within three months), with a mean extensor lag of 3 (range, 0 to 50 ) at the three-month postoperative time point. Nineteen (38%) of the fifty knees were consiere failures. Four require a repeat extensor mechanism allograft reconstruction (three for recurrent instability seconary to stretching of the graft an one for an extensor mechanism rupture), five faile seconary to a eep infection (two eventually were treate with amputation, two were treate with a knee arthroesis, an one was treate with a revision extensor mechanism allograft reconstruction following a two-stage exchange), an ten were eeme clinical failures seconary to a Knee Society score of <60 points or an extensor lag of >30. Failure occurre at a mean time of twenty-one months after extensor mechanism allograft reconstruction. The Kaplan-Meier estimate survivorship with failure for any reason as the en point was 56.2% (95% confience interval [95% CI], 39.4% to 70.1%) at ten years (Fig. 1). With the numbers available for stuy, no specific variables were ientifie that correlate with failure.

3 281 Fig. 1 Kaplan-Meier survival showing a periprosthetic fracture of the tibia following extensor mechanism allograft reconstruction in a patient in whom concomitant revision was not performe. Aitional complications inclue six revisions of the tibial bone block fixation (all ha been fixe with wires alone an five of the six bone blocks went on to successful union), three periprosthetic fractures of the tibia at the site of the tibial bone block (all were seen in cases in which the tibial component was short an not concomitantly revise or when a short stem was use for the revision) (see Appenix), four periprosthetic femoral fractures seconary to falls, one manipulation uner anesthesia, an one partial quariceps tear requiring operative repair. The periprosthetic tibial fractures were all treate nonoperatively; however, two of the reconstructions were consiere failures, both seconary to a Knee Society score of <60 points. Three of the four periprosthetic femoral fractures were treate with open reuction an internal fixation; one of these was consiere a failure seconary to an extensor lag of >30 an a Knee Society score of 55 points. Discussion Although extensor mechanism isruption is a rare complication following total knee arthroplasty with incience ranging from 0.10% to 2.5% 1-3,20, it results in substantial morbiity an is typically not well tolerate in patients seconary to ifficulty walking an severe knee instability. Hence, operative treatment is typically attempte, but the outcomes have been mixe 8,14,15,17,18,21, an the ieal metho of treatment remains unclear. The current stuy emonstrates that, although our initial experience with a complete extensor mechanism allograft tensione tightly in full extension was goo, the overall complication rate is high an the results have egrae over time with an estimate ten-year survivorship of just over 50% 15. Although some of the ifferences in the reporte results may be ue to a stricter efinition of failure that consiers a Knee Society score of <60 points as a clinical failure, these finings o provie both patients an surgeons with more realistic expectations for the outcomes of the treatment for this complication. When compare with prior stuies of a complete extensor mechanism allograft, our results show a higher rate of failure with longer-term follow-up. Burnett et al. 22 reporte on nineteen patients with extensor mechanism reconstruction at a mean follow-up time of fifty-six months. They emonstrate an increase in the Knee Society score from 27 points preoperatively to 76 points postoperatively, which is consistent with our finings. They cite an 89% patient satisfaction rate; however, three patients were clinical failures applying the same criteria as in our stuy. Aitionally, only nine of the nineteen patients unerwent reconstruction with a complete extensor mechanism composite as was use in our stuy; the other ten patients unerwent an Achilles tenon allograft with a calcaneal bone block. Nazarian an Booth 18 originally escribe the technique an reporte successful clinical results in thirty-four of thirty-six patients at a mean of 3.6 years an showe an average increase in the Knee Society score from 37 to 68 points. However, eight of these patients require a secon revision extensor mechanism allograft for recurrent failure an, if the same criteria from our stuy ha been applie, the failure rate woul have been higher. An alternative for treatment is the use of synthetic mesh (Marlex mesh; C.R. Bar, Murray Hill, New Jersey). Browne

4 282 an Hanssen 10 use this material for reconstruction in a cohort of thirteen patients with a mean follow-up of forty-two months. Three patients ha failure, an one ha evelope an infection, leaving nine (69%) of the original thirteen patients with a wellfunctioning extension mechanism. These results are encouraging; however, this report escribes a small cohort followe for a relatively short perio of time an not unlike our results, clinical outcomes may eteriorate with time. Unfortunately, our own experience with mesh reconstruction is limite to two cases, both of which meet the efinition for failure as efine in our stuy. Achilles tenon allograft reconstruction with a calcaneal bone block for tibial fixation has also been use to reconstruct the extensor mechanism. To our knowlege, the first reporte series of patients revise with this technique was by Crossett et al. 8, who escribe nine patients successfully treate with this metho at a minimum follow-up of two years. However, there were two early failures requiring repeat repair. As previously mentione, ten of the patients in the stuy by Burnett et al. 22 were also reconstructe in this manner. The most comprehensive report on this technique is by Diaz-Leezma et al. 23 who emonstrate a moest 58.6% success rate in twenty-nine knees at a mean follow-up of 3.5 years. Although an avantage of an Achilles tenon allograft in the case of an isolate patellar tenon isruption is preservation of the native patella, our experience is that proximal fixation of the graft is more ifficult as it fans out an thins proximally. Further, it can be ifficult to cover the allograft fully with native tissue, an hence the allograft oftentimes lies just beneath the skin. Hence, although the results of an Achilles tenon allograft are similar, our preference remains a full extensor mechanism allograft. At 10% in our series, eep infection was one of the most common complications an le to many of the poorest outcomes, incluing the nee for amputation an arthroesis. Many of these patients ha unergone multiple prior attempts at surgical treatment, an hence the soft-tissue envelope was likely compromise. Combine with use of a large allograft or other foreign material, this complication has been reporte to have a substantial prevalence in most reporte series 8,10,23 an is probably common to all of the techniques use in contemporary practice other than those that use autograft tissue for augmentation. An unerstaning of patient or technical factors contributing to treatment failures woul be useful for surgeons who are treating this complication. Unfortunately, with the number of patients available for stuy, we are unable to ientify any risk factors for failure or for the occurrence of complications. However, on the basis of the results an given the observe 12% rate of failure of istal fixation of the graft in our stuy, we have altere our surgical technique an now either use screws for ajunctive fixation of the allograft or attempt to pass the fixation wires aroun the stem of the implant, if a concomitant component revision is performe (see Appenix). It is important to recognize that nearly half of the patients in this series require a concomitant revision proceure, as instability, stiffness, an component malposition are common in patients who sustain an extensor mechanism isruption, an hence the surgeon shoul be reay to revise components at the time of extensor allograft reconstruction. Further, given that there were several patients in whom a short stem or a primary component was utilize for reconstruction an who later sustaine a stress-type fracture of the tibia, our threshol to revise the components to a reconstruction with a longer tibial stem is lower. However, in cases in which component size, rotation, an overall stability are ieal, we still will retain the original components 24. Finally, given the frequency with which patients who experience failure report instability, we now typically use a rotating hinge implant if a concomitant revision is performe. In summary, extensor mechanism isruption following total knee arthroplasty is a ifficult complication to treat. Although we still utilize a complete allograft extensor mechanism for treatment of chronic extensor mechanism eficiency, we carefully counsel patients on the seriousness of this complication an that, although initial results may be goo, clinical results may eteriorate with time an the overall complication rate associate with the treatment (particularly eep infection) is high. Given the relatively high rate of failure overall, even in patients with well-functioning allografts, we now recommen the use of an assistive evice when walking, given the impaire proprioception resulting from the replacement of native tissues with both an implant an an allograft extensor mechanism. Appenix Figures showing raiographs of a patient preoperatively with extensor mechanism isruption following a revision total knee arthroplasty an postoperatively following extensor mechanism allograft reconstruction at five years showing healing of the tibial bone block an correction of the previously seen patella alta an raiographs emonstrating a tibial graft fixation wire passe aroun the stem of the revision implant an cerclage wires passe aroun the implant stem for aitional fixation strength an a table showing the summary of results are available with the online version of this article as a ata supplement at jbjs.org. n NOTE: We woul like to thank Aaron Rosenberg, Mario Moric, Vamsi Kancherla, an Laura Quigley for their assistance with this stuy. Nicholas M. Brown, MD Scott M. Sporer, MD Nathan Wetters, MD Richar A. Berger, MD Craig J. Della Valle, MD Rush University Meical Center, 1611 West Harrison Street, Suite 300, Chicago, IL aress for C.J. Della Valle: craigv@yahoo.com Trevor Murray, MD Department of Orthopaeic Surgery, Clevelan Clinic, 9500 Eucli Avenue, A41, Clevelan, OH 44195

5 283 References 1. Caambi A, Engh GA. Use of a semiteninosus tenon autogenous graft for rupture of the patellar ligament after total knee arthroplasty. A report of seven cases. J Bone Joint Surg Am Aug;74(7): Ran JA, Morrey BF, Bryan RS. Patellar tenon rupture after total knee arthroplasty. Clin Orthop Relat Res Jul;(244): Lynch AF, Rorabeck CH, Bourne RB. Extensor mechanism complications following total knee arthroplasty. J Arthroplasty. 1987;2(2): Rosenberg AG. Management of extensor mechanism rupture after TKA. J Bone Joint Surg Br Nov;94(11)(Suppl A): Kim TW, Kamath AF, Israelite CL. Suture anchor repair of quariceps tenon rupture after total knee arthroplasty. J Arthroplasty Aug;26(5): Epub 2011 Mar Jaureguito JW, Dubois CM, Smith SR, Gottlieb LJ, Finn HA. Meial gastrocnemius transposition flap for the treatment of isruption of the extensor mechanism after total knee arthroplasty. J Bone Joint Surg Am Jun;79(6): Busfiel BT, Huffman GR, Nahai F, Hoffman W, Ries MD. Extene meial gastrocnemius rotational flap for treatment of chronic knee extensor mechanism eficiency in patients with an without total knee arthroplasty. Clin Orthop Relat Res Nov;(428): Crossett LS, Sinha RK, Sechriest VF, Rubash HE. Reconstruction of a rupture patellar tenon with Achilles tenon allograft following total knee arthroplasty. J Bone Joint Surg Am Aug;84(8): Malhotra R, Garg B, Logani V, Bhan S. Management of extensor mechanism eficit as a consequence of patellar tenon loss in total knee arthroplasty: a new surgical technique. J Arthroplasty Dec;23(8): Epub 2008 Jan Browne JA, Hanssen AD. Reconstruction of patellar tenon isruption after total knee arthroplasty: results of a new technique utilizing synthetic mesh. J Bone Joint Surg Am Jun 15;93(12): MacCollum MS 3r, Karpman RR. Complications of the PCA anatomic patella. Orthopeics Nov;12(11): Doolittle KH 2n, Turner RH. Patellofemoral problems following total knee arthroplasty. Orthop Rev Jul;17(7): Springer BD, Della Valle CJ. Extensor mechanism allograft reconstruction after total knee arthroplasty. J Arthroplasty Oct;23(7)(Suppl): Leopol SS, Greianus 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 Nov;81(11): Burnett RS, Berger RA, Paprosky WG, Della Valle CJ, Jacobs JJ, Rosenberg AG. Extensor mechanism allograft reconstruction after total knee arthroplasty. A comparison of two techniques. J Bone Joint Surg Am Dec;86(12): Emerson RH Jr, Hea WC, Malinin TI. Reconstruction of patellar tenon rupture after total knee arthroplasty with an extensor mechanism allograft. Clin Orthop Relat Res Nov;(260): Emerson RH Jr, Hea WC, Malinin TI. Extensor mechanism reconstruction with an allograft after total knee arthroplasty. Clin Orthop Relat Res Jun; (303): Nazarian DG, Booth RE Jr. Extensor mechanism allografts in total knee arthroplasty. Clin Orthop Relat Res Oct;(367): Insall JN, Dorr LD, Scott RD, Scott WN. Rationale of the Knee Society clinical rating system. Clin Orthop Relat Res Nov;(248): Dobbs RE, Hanssen AD, Lewallen DG, Pagnano MW. Quariceps tenon rupture after total knee arthroplasty. Prevalence, complications, an outcomes. J Bone Joint Surg Am Jan;87(1): Barrack RL, Stanley T, Allen Butler R. Treating extensor mechanism isruption after total knee arthroplasty. Clin Orthop Relat Res Nov;(416): Burnett RS, Butler RA, Barrack RL. Extensor mechanism allograft reconstruction in TKA at a mean of 56 months. Clin Orthop Relat Res Nov;452(452): Diaz-Leezma C, Orozco FR, Delasotta LA, Lichstein PM, Post ZD, Ong AC. Extensor mechanism reconstruction with achilles tenon allograft in TKA: results of an abbreviate rehabilitation protocol. J Arthroplasty Jun;29(6): Epub 2013 Dec Klein GR, Levine HB, Sporer SM, Hartzban MA. Fracture of the proximal tibia after revision total knee arthroplasty with an extensor mechanism allograft. J Arthroplasty Feb;28(2):375.e9-12. Epub 2012 Jun 14.