Chapter 100 Unicompartmental Knee Arthroplasty

International Roundtable Discussion

Jean-Noël Argenson, David Murray, Thomas M. Coon, Gerard A. Engh, Richard A. Berger, Hong Zhang, Richard D. Scott

Jean-Noël Argenson: Unicompartmental knee arthroplasty (UKA) has become a popular surgical technique. UKA represents the logical treatment option for an isolated femorotibial lesion of the knee. When the other compartments of the knee are preserved, there is no need to perform a total knee arthroplasty (TKA). When comparing the outcomes of UKA and TKA, several studies have shown that recovery is faster after UKA, and that patient satisfaction is greater, because the knee feels more natural. However, UKA has a specific mode of failure, which is progression of disease in unreplaced compartments. This might be related to surgical technique, but the most critical issue remains patient selection. Improper patient selection combined with limited instrumentation and suboptimal design may explain the less than satisfactory results originally published for UKA. However, in the United States, Europe, and Asia, indications have been more clearly defined, design and instrumentation have shown significant evolution, and the operation can be performed routinely using a minimally invasive technique. Therefore, it is interesting to share experiences within a group of international experts, which includes David Murray, Tom Coon, Jerry Engh, Richard Berger, Hong Zhang, and Richard Scott. Let us start this discussion with patient selection, which is critical for UKA. My own experience has shown that UKA is applicable mainly to osteoarthritis, and in some cases to avascular osteonecrosis. Dick, do you concur with that statement, and regarding osteoarthritis, what type of deformity in varus or in valgus do you think is suitable for UKA?

Richard Scott: I agree that UKA should be limited to patients with a noninflammatory, nonsystemic form of arthritis. I would add post-traumatic arthritis to your list, for example, lateral compartment degeneration that might follow a lateral plateau fracture.

UKA is contraindicated in RA (unless clearly “burned out”) and in any patients with a history of inflammatory synovitis such as gout or pseudogout. The presence of radiographic chondrocalcinosis in the opposite compartment is a relative contraindication.

Severe deformity is a contraindication to UKA because it is difficult to correct. This is represented by more than 10 degrees of mechanical axis deviation in a varus knee, and by 5 to 7 degrees of deviation in a valgus knee. An attenuated medical collateral ligament (MCL) in a valgus knee is also a contraindication.

Jean-Noël Argenson: These indications clinically and radiographically speaking have been nicely outlined by Richard Scott for the femorotibial compartment; however, what about the third compartment represented by the patellofemoral joint (PFJ)? David, is it true to say that in Oxford it is very unlikely to have a UKA indication withdrawn because of the state of the patellofemoral joint?

David Murray: Indications for the mobile-bearing Oxford UKA are different from those for the fixed-bearing UKA. In early studies of the Oxford knee, it was found that the state of the PFJ did not influence the outcome. We therefore advised surgeons to ignore the PFJ and have implanted large numbers of Oxford knees, regardless of the state of the PFJ. As a result, we have been able to study in detail the effect of the PFJ on outcome. Our recent studies have confirmed that for medial UKA, the state of the PFJ can be ignored unless severe lateral PFJ arthritis with bone loss is present. In these rare cases, which involve combined medial tibiofemoral and lateral patellofemoral arthritis, a total knee replacement (TKR) should be implanted. Damage to the trochlear or medial part of the PFJ, however severe, should not be considered a contraindication to mobile-bearing UKA. In addition, anterior knee pain is not a contraindication.

Jean-Noël Argenson: Jerry, it is classical to assume that a functional anterior cruciate ligament (ACL) is mandatory for performing a UKA; however, both our kinematic evaluation of patients with flat fixed tibial bearings and the clinical outcomes of these sedentary patients with a probable secondary distention of the ACL have confirmed that correct clinical function of the knee may be achieved in these low-demand patients using fixed bearings, despite ACL deficiency. Based on your experience, what is your current attitude regarding the state of the ACL?

Jerry Engh: In most of my patients, an absent or attenuated ACL is not a contraindication to a fixed-bearing medial UKA. A patient who does not engage in running or twisting sporting activities and has no symptoms of knee instability prior to surgery will achieve a result similar to that of a patient with a functional ACL following medial unicompartmental arthroplasty.

Historically, patients with ACL deficiency have not done well with mobile-bearing implants. These implants failed by premature tibial component loosening, not by bearing dislocation. The load pattern on the tibial component is different in a knee with ACL deficiency. A mobile-bearing implant is more likely to experience eccentric loading that could cause loosening in a ligamentous deficient knee. In ACL-deficient knees, I prefer to reduce the anteroposterior (AP) slope of the tibial resection to tighten the flexion gap and reduce AP laxity.

Over the past 9 years, I have implanted UKAs in 87 patients (98 knees) with both attenuated (53 knees) and absent (45 knees) ACLs at the time of unicompartmental arthroplasty. Three of these patients died of causes unrelated to the arthroplasty without returning for 2-year follow-up. We revised five patients (six knees) for progression of disease (three knees), aseptic loosening (2 knees), and pain (1 knee). Fifty patients (56 knees) had clinical and radiographic follow-up at a mean of 40 months (range, 24 to 75 months). The mean American Knee Society Clinical Rating System (KS) Clinical Score of these patients was 92 points (range, 75 to 100 points), the mean KS Function Score was 84 points (range, 45 to 100 points), and the mean arc of motion was 127 degrees (range, 101 to 145 degrees). Patients reported less pain in the knee (89%), better function in the knee (88%), and overall satisfaction with the UKA (86%). No patient reported problems with knee instability.

Premature implant wear is another theoretical concern. In the laboratory, wear of polyethylene is increased with multidirectional sliding of metal on polyethylene. The same multidirectional sliding might occur in an ACL-deficient knee. However, I am unaware of any clinical studies that document accelerated wear in ACL-deficient knees.

Jean-Noël Argenson: Thanks Jerry, for reporting this experience, particularly in patients with no complaint of instability before the indication of UKA. Age and weight also represent debating issues for the indication of UKA, because the procedure is often presented as an alternative to osteotomy based on patient age. Our own experience shows that the best indication for osteotomy remains the young, heavy, active male patient with the presence of joint space and bony deformity. Regarding age and based on comparative results at 10 years after UKA and TKA, Richard Scott considered that UKA might assume its role in two groups of patients: middle-aged osteoarthritic patients and osteoarthritic octogenarians who are having their “first and last” arthroplasty. Dick, are there any changes in your current indications?

Richard Scott: I agree with you that osteotomy should always be considered in the young, heavy, active male patient. I still think that UKA is an excellent alternative to osteotomy in the middle-aged patient (especially female) compared with osteotomy. It leads to faster recovery, fewer complications, better longevity, and a better cosmetic result.

If revision should become necessary in the future, this is easy to accomplish following UKA as long as a conservative initial tibial resection has been performed.

I still advocate UKA in octogenarians as long as they are ideal (not borderline) candidates. Recovery is generally faster than after TKA, with better flexion, less medical morbidity, less need for transfusion, and better prosthetic survivorship than are seen with TKA.

Jean-Noël Argenson: I think we now have a more precise perception of the “ideal” patient candidate for a UKA regarding age, etiology, deformity, and state of the joint. Let us move to design considerations, because these might have a direct impact on knee function and on the types of activities performed by the patient following the procedure. The importance of full coverage of the condyle by the femoral component has been recently highlighted to allow full range of motion without impingement. Hong, how important are these activities in high flexion for the patients you have to treat, and more generally, in the Asian world?

Hong Zhang: In Asia, lifestyle and habits depend on religious beliefs, cultural background, and customs. Kneeling and cross-legged and squat positions are part of life for many Asian people when they pray, when they eat, when they get together and chat, and even when they go to the toilet. We understand that many patients expect to go back to their normal lifestyle after knee replacement, so achieving high flexion and full range of motion would be what surgeons strive for. However, although TKA has increased dramatically in Asian countries in recent years, the total number of UKAs performed is still very low. Riddle and colleagues reported that unicompartmental implants accounted for about 8% of all knee arthroplasty procedures in the United States in 2005 (Riddle D, Jiranek M, McGlynn F: Yearly incidence of unicompartmental knee arthroplasty in the United States. J Arthroplasty 23:408–412, 2008). But in Japan in 2007, UKA occurred in less than 1% of those with knee arthroplasty (data from 2007 yearly report of Yano Research Institute Ltd). In China, our UKA numbers are less than 1% as well (data from Beijing Jishuitan Hospital). Most of our patients present with end-stage arthritis when they come to us. It is very hard to find “ideal” patient candidates for UKA.

Jean-Noël Argenson: Yes, this is very true, and I do believe there are regional answers in terms of indications but also for designing philosophy. Rich, for this specific area related to designing considerations, metal backing of the tibial component is controversial because on one side, wear-through or enhanced polyethylene flow has been described with the use of metal backing, but on the other side, the modular tray allows polyethylene exchange without bone invasion through a limited incision. What is your experience?

Richard Berger: I believe that a modular metal-backed tibial component is superior to an all-polyethylene tibial component because of its ability to be easily replaced for wear and its superb long-term results. First, the modular polyethylene insert of a metal-backed component can be easily replaced if significant wear occurs. With an all-polyethylene component, if wear occurs, the entire component must be revised. Second, because a significant amount of polyethylene wear comes from third-body wear from retained cement, the modularity of a metal-backed component facilitates cement removal. Finally, great long-term results have been reported with the metal-backed component. We reported 11-year to 15-year results of UKA with a metal backing. Average follow-up was 13 years (range, 11 to 15 years). No component was radiographically loose, and no osteolysis was seen. Kaplan-Meier survival with loosening or revision for any reason was 98.0% ± 2.0% at 10 years, and 95.7% ± 4.3% at 15 years. Last, even though 64% of these cases had the thinnest polyethylene—only 5.7 mm of actual polyethylene—no revisions for polyethylene wear were performed.

Jean-Noël Argenson: I concur with this experience, including polyethylene liner exchange in UKA, and we have now a few cases at 5 years’ follow-up after the exchange, extending then the overall survival of the UKA to more than 15 years, which is encouraging. However, polyethylene wear is one mode of failure of UKA, besides osteoarthritis progression, and it is multifactorial. Jerry, how important may material properties be in the issue of poly wear?

Jerry Engh: Polyethylene wear was the primary reason for failure of unicondylar arthroplasties when we were using polyethylene that was gamma-irradiated in air. Until the year 2000, 75% of the revisions at our institution were related directly to polyethylene wear. When we examined the revised implants in our lab, all of the gamma-irradiated components had some degree of delamination. During the same time interval, none of the nonirradiated implants were revised. The retrieval lab at Dartmouth reported a similar experience with nonirradiated polyethylene. A group of 16 retrieved unicondylar implants in situ after more than 15 years showed no evidence of fatigue-type wear.

Over the past 9 years at our institution, and since the industry shift to nongamma sterilization or sterilization in an inert environment, we have had no revisions of unicondylar implants for polyethylene wear. Therefore, in response to your question, the material properties appear to no longer be a major issue. However, I do have concern about implants irradiated in an inert environment without quenching of free radicals after the sterilization process. When retrieved components that have been irradiated in an inert environment were analyzed, it appears that oxidation in vivo is a real concern. If we have delayed oxidation only by sterilizing and vacuum packaging the components, polyethylene wear could re-emerge as a factor in the failure of UKAs with a longer time interval in situ.

Jean-Noël Argenson: Thanks for this update regarding properties issues. David, mobile bearings have been presented as the answer to polyethylene wear in UKA, and both retrieval studies and clinical outcomes with the Oxford knee have shown very limited wear. How different are the designing principles of a mobile-bearing meniscal knee in UKA?

David Murray: The Oxford knee was designed to minimize wear by reproducing the function of the normal meniscus. The meniscus, being compliant, achieves fully congruent contact in all positions with the femoral and tibial condyles, so contact stress and thus wear are low. In the artificial situation, because polyethylene cannot change shape, fully congruous contact can be achieved only by using a spherical femoral component. The Oxford knee therefore has a spherical femoral component, a flat tibial component, and an unconstrained bearing with a spherically concave upper surface and a flat lower surface. It has been shown to have very low wear (about 0.01 mm/yr), provided it has been implanted correctly without impingement. It is safe to use polyethylene as thin as 3 mm, so bone stock can be preserved. Unlike fixed-bearing designs, it can be used safely in young, heavy, active men. It also provides full flexion, which is important in Asian countries.

Jean-Noël Argenson: Thank you David, but what can be said to surgeons who are afraid of bearing dislocation and the difficulty of reproducing the surgical technique in every case?

David Murray: With phase 3 instrumentation, which is now generally used, the incidence of bearing dislocation following medial Oxford UKA is very low (less than 1%). So surgeons need not be unduly concerned. In contrast, the dislocation rate for lateral replacement has been high. We have therefore recommended that mobile bearings should not be used in the lateral compartment. However, short-term results of the new design of Oxford lateral UKA, which has a convex tibial component and a biconcave bearing, are encouraging with a low dislocation rate. In the future, this may be a good solution for lateral compartment arthritis.

The phase 3 instrumentation was designed for a minimally invasive approach and is relatively straightforward to use. It accurately balances the flexion and extension gaps. So, as well as preventing dislocation, it reliably restores knee kinematics and function to normal.

Jean-Noël Argenson: As mentioned by David, minimally invasive surgery (MIS) is now the routine approach for UKA without everting the patella, and I must say that the quality of patient recovery and implant positioning has been highly reproducible in our experience. Tom, what are your tricks regarding the soft tissue environment to obtain adequate exposure of the knee?

Tom Coon: Good surgical exposure is critical to good results in UKA. It is desirable to avoid quadriceps muscle trauma to achieve the excellent rehabilitation results possible with the MIS technique. I have found that the best way to facilitate exposure is to use the medial retinacular “T,” as recommended by Tria. This is done midway between the lower edge of the vastus medialis and the medial meniscus. It allows excellent visualization of the medial joint space for removal of osteophytes, placement of implants, or removal of cement, and it provides a good view of the medial tibia for assessment of the tibial slope, as well as for correct placement of the tibial cutting guide. In some cases, it may be useful to perform a medial patellar facetectomy as recommended by Repicci, although I have rarely found this to be necessary.

It is very useful also to utilize the “mobile window,” that is, move the tissue opening from side to side or up and down, and to flex or extend the knee to tense or relax the extensor mechanism as necessary. Tibial exposure can be facilitated by the use of a leg holder, which allows the tibia to be fixed in internal or external rotation, thus allowing easy placement of instruments and implants.

Jean-Noël Argenson: Tom just showed us how to position correctly the tibial cutting guide in the frontal plane, but Hong, what about the tibial slope, which may be increased in some knees, and the direction of the sagittal cut at the upper surface of the tibia?

Hong Zhang: From my personal experience, I try to make a 5 to 7 degree tibial slope. This is the anatomic slope angle, and most manufacturers require the same slope angle as well. When I do my tibial plateau resection, I start with the vertical cut, made with the reciprocating saw. This step helps me confirm the lateral margin of the medial tibial plateau and the origin of the ACL to avoid damaging its fibers. Then I use a 12 mm wide oscillating saw blade to cut about a 2 mm thick plateau by naked eyes, without using the tibial saw guide. I find that without the tibial saw guide, it is more safe and accurate to finish the tibial cut, while not damaging the ACL.

Jean-Noël Argenson: I fully agree with this last point, and I usually bring the knee into extension at this step to decide where to start the vertical cut, according to where the lateral edge of the medial condyle will rest on the tibial plateau. Rich, do you think pin fixation may act as a stress raiser and increase the risk of tibial plateau fracture following UKA?

Richard Berger: Yes, pin fixation does act as a stress raiser, leading to fractures. Perioperatively, we had a high incidence (5%) of medial tibial plateau fractures in our series. This complication has been reported in other series but is rare. The high tibial fracture rate in our series was technique related. These fractures were related to the fixation pins for the tibial alignment guide, which were impacted without predrilling; furthermore, these pins were located in a row at the corner of the tibial resection—the area of greatest stress. Because we recognized the cause of these fractures, the tibial holes were predrilled and relocated; subsequently, tibial fractures have not occurred. Of note, fortunately these tibial fractures in our series did heal without adverse sequelae. Fractures of the medial tibial plateau fracture that were noted on the postoperative radiograph were treated nonoperatively and healed uneventfully. A fracture, when noted intraoperatively, was treated with screw fixation and also healed uneventfully.

Jean-Noël Argenson: Thank you Rich, for reporting this experience and, more important, for describing how such complications can be avoided. Moving now to the femoral component, I always found that mediolateral and rotational placement of the femoral implant was a critical issue in UKA, despite the type of design. Jerry, what is your own experience?

Jerry Engh: Component-to-component positioning is critical to avoid the two major causes of implant failure: aseptic loosening and implant wear. Abnormal medial-to-lateral positioning causes edge loading that can result in tibial bone collapse or aseptic loosening. Malrotation of the components increases contact stresses and accelerates implant wear.

The problem with component positioning relates to instrumentation and surgical technique, not to the design of the implant. Intramedullary instruments set the alignment of the components not to each other, but to the bone to which they are attached. This may or may not be correct for the rotation and translation of components.

The rotational alignment of the femoral component is dictated by the posterior condylar cut from the femur. If this cut is made in varus alignment to the patient’s posterior femoral condyles, the femoral component will be internally rotated. When the knee is in full extension, this internal rotation can result in edge loading of the tibial component. Likewise, if the femoral component is centered on the femoral condyle, it may not articulate with the center of the tibial component. The femoral component should be positioned over the center of the tibial component.

I prefer setting the position of the femoral component relative to the center of the tibial component by using spacer blocks after the tibial resection to orient both the rotation and the translation of the femoral component. I use an extramedullary guide to resect the tibia; I then insert a spacer block to check the flexion/extension gap balance and tibial coverage. Next, I bring the knee to full extension and mark the front of the femur so that when the posterior condylar cut is made, the femur will be correctly oriented and centered on the tibial implant as referenced by this mark. With the knee flexed, I rotate the sizing/cutting guide and align the mark on the front of the femur before making the posterior condylar resection that will set the femoral implant properly.

Jean-Noël Argenson: David, is this true also for mobile bearings?

David Murray: One of the advantages of the mobile-bearing device is that component orientation is not critical. Because the femoral component is spherical, and because the contact area is large (6 cm²), moderate amounts of component malalignment do not matter. We have found that 10 degrees of femoral component malalignment in all directions with 5 degrees of tibial component malalignment in all directions does not compromise the outcome. Therefore, the procedure is very forgiving, provided that the soft tissues are accurately balanced.

Jean-Noël Argenson: Dick, in my experience, lateral UKA is equally successful compared with medial UKA and represents 10% of my uni indications. How different can the surgical principles of lateral UKA be?

Richard Scott: My ratio of medial to lateral UKAs is also 10 to 1. This means that we all have less experience with lateral UKA, and it is technically more difficult than medial arthroplasty. Patellar impingement on the leading edge of the femoral component is more likely to occur. One must be careful to recess the leading edge of the femoral component to avoid this impingement. Little or no posterior slope should be applied to the tibial resection to avoid a postoperative posterior wear pattern. Medial-lateral incongruency is more likely, and one must err toward placing the femoral component laterally on the femoral condyle, and the tibial component medially on the tibial plateau.

Exposure for the arthroplasty can be attained through a short lateral arthroplasty or through a medial parapatellar approach, while sparing the anterior horn of the medial meniscus and subluxing or everting the patella. The medial approach allows easy intraoperative conversion to TKA if appropriate.

Jean-Noël Argenson: Yes, and following these considerations, our published results of lateral UKA compared equally with those reported for medial UKA. Tom, talking now about overall alignment of the limb, how different is the alignment goal following UKA compared with TKA?

Tom Coon: The generally accepted standard of alignment for the TKA is the neutral mechanical axis. This is done to promote equal loading of the tibial component and thus improve longevity of the implant. In the case of the UKA, longevity is promoted by slightly shifting the mechanical axis toward the implant to reduce the chance of progressive degeneration of the contralateral compartment. Toward this end, it is generally better practice to undercorrect the deformity by 1 or 2 degrees; this can be accomplished by filling the existing ligament envelope with an implant thickness sufficient to allow 1 to 2 mm laxity with the existing ligament, and by performing no ligament releases. This translates to undercorrection of 1 to 2 degrees, that is, a final varus of 3 to 5 degrees, or valgus of 7 to 8 degrees.

Jean-Noël Argenson: I also like observing the postoperative mechanical axis of the limb crossing the joint line somewhere between the tibial spines and the middle of my tibial plateau after UKA. Hong, because the possibilities of adjusting soft tissue tension are limited in UKA, what is your rationale for balancing flexion/extension gaps and choosing the appropriate insert thickness?

Hong Zhang: Because most UKA patients have mild varus or valgus deformity, with or without mild flexion contracture, we do not need to do too much soft tissue work during UKA surgery. For example, for 5 degrees of varus deformity, I will peel the medial tibial periosteum about 1 to 2 cm and completely remove the osteophytes and medial meniscus; this is sufficient for soft tissue release. Then I will follow the tibial and femoral cutting instructions to finish the bone cut and put the trail components in to test the tension of the flexion/extension gap. For fixed-bearing UKA, I will keep 1 mm open under the tension or flexion/extension position. I will also make sure that the mechanical alignment of the lower extremity is not overcorrected.

Jean-Noël Argenson: This 1 or 2 mm opening of the compartment is indeed probably a safe way to avoid overcorrection. Dick, what can be the role of computer-assisted surgery in the field of UKA?

Richard Scott: Computer-assisted surgery has promise for the future both in assessing alignment and in preparing the bony surfaces. I must confess that I presently have no personal experience with this technique. I find it to be too time consuming, too expensive, and un-necessary in most cases. I look forward to its eventual use, however, when these issues are resolved.

Jean-Noël Argenson: This is probably correct, and it will be interesting to follow the evolutions in the field of navigation and in using robotic assistance adapted to UKA. The 10-year results of UKA reveal that several centers regularly reported survival rates over 90% and very comparable with those of TKA. Results reported in national and community-based registries showed survivorship at 10 years around 88% for UKA compared with 94% for TKA. The same national registry in Sweden reported that patients are more satisfied after revision of a failed UKA than after revision of a TKA. Rich, it is now commonly supported that revision of a modern UKA is a simple procedure. Don’t you think that this may play a role in differences reported by such registries?

Richard Berger: Yes. I think that we all see total knee patients who have some pain, or who are dissatisfied with their replacement; unless there is a definable and significant source of their pain, few of us recommend a revision. This is because a total knee revision is often a difficult procedure that has significant morbidity. Moreover, even with successful revision, the patient often has residual pain and some dysfunction. Therefore, we tell TKA patients who have pain or are dissatisfied that this is something they must learn to live with.

Conversely, the unicompartmental replacement patient who has some pain or who is dissatisfied, even without a definable or significant source of pain, often gets a conversion to a TKA. This is done because a conversion to a total knee is usually an easy procedure that does not have significant morbidity. Moreover, a successful conversion, as you have pointed out, often results in a perfectly functioning TKA. Therefore, we tell the UKA patient who has pain or is dissatisfied that a conversion to a TKA is warranted. It is a double standard that results in a higher revision rate for unicompartmental replacements than for total knee replacements.

Jean-Noël Argenson: Thanks Rich, for highlighting this point. Tom, what can be the role of the learning curve and providing courses on these results, since UKA is, by definition, a less frequently performed surgery than TKA?

Tom Coon: UKA is indeed less frequently performed than TKA, with UKA accounting for 10T to 25% of the knee replacement practice. For most surgeons who do them infrequently, learning centers are critical to allow improved confidence and better results with this technically demanding procedure. Specific implants have different technical requirements, and surgeons should avail themselves of widely available training courses, so that the nuances of each system to be used can be appreciated. Additionally, as was mentioned previously, newer techniques such as computer navigation and robotics can be learned and may help mitigate problems with earlier systems such as poor implant placement or alignment.

Jean-Noël Argenson: Since we have reached the end of this roundtable, I would like to thank all of you coming from different parts of the world for sharing your experiences with unicompartmental knee arthroplasty. We have been able to better define indications for the procedure, because many early failures are simply related to the wrong indications. Although surgical technique is now currently realized through MIS, this roundtable has explained ways of adhering to the well-established principle of undercorrection, because progression of osteoarthritis in the unreplaced compartment is a frequent mode of failure following UKA. Progress in instrumentation and in design has made the procedure an attractive solution for preserving both ligaments and cartilage for osteoarthritis limited to one femorotibial compartment of the knee.

Only gold members can continue reading. Log In or Register to continue

Share this:

• Click to share on Twitter (Opens in new window)
• Click to share on Facebook (Opens in new window)

Related

Related posts:

Cemented Total Knee Arthroplasty: The Gold Standard Management of Extra-articular Deformity in Total Knee Arthroplasty With Navigation Patellar Instability Posterior Cruciate Ligament Reconstruction: Posterior Inlay Technique Anterior Cruciate Ligament Reconstruction With Central Quadriceps Free Tendon Graft Osteochondritis Dissecans

Stay updated, free articles. Join our Telegram channel

Join

Full access? Get Clinical Tree