Journal of the American Academy of Orthopaedic Surgeons 16 Most surgeons agree that the ar- thritic knee with valgus deformity presents a unique set of problems that must be addressed at the time of total knee arthroplasty (TKA). Correction of the deformity and restoration of anatomic alignment should be achieved to maximize the longevity of the replaced compo- nents. Although several authors have suggested TKA as a potential treatment for the knee with severe valgus deformity, 1-9 none has com- prehensively reviewed the extensive kinematic and anatomic variables that need to be understood in at- tempting to balance the TKA in a patient with a valgus deformity. The valgus knee may have any combination of primary or sec- ondary bone and soft-tissue abnor- malities. These include contracted lateral capsular and ligamentous structures, lax medial structures, and acquired or preexisting bony anatomic deficiencies. This constel- lation of pathology makes attaining soft-tissue balance when the knee is returned to physiologic alignment extremely difficult. The recent liter- ature 10,11 has underscored the im- portance of ligament balancing and of assessing this balance throughout a range of motion during the trial reduction of the total knee compo- nents. In the 1970s, Freeman et al 12 and Insall 13 were among the first to emphasize this basic principle with the introduction of a tensor instru- ment and laminar spreader to assess the symmetry of the flexion and extension gaps. The development and more frequent use of both con- strained posterior stabilizing/poste- rior cruciate substituting prostheses and total stabilizing prostheses, as well as the introduction of “mea- sured resection” instrumentation, might suggest that less attention can be paid to this basic principle of bal- ancing the knee. However, if prop- er ligament balancing techniques are used and proper ligament bal- ance is attained, the knee may not require the use of a more con- strained component. A thorough understanding of the pathologic anatomy and a meticulous preoperative evaluation to demonstrate fully the individual patient’s anatomic deficiencies are critical to effective surgical man- agement. Etiology Valgus deformity in adults com- monly is associated with inflamma- tory arthritis as well as with primary osteoarthritis, posttraumatic arthritis, or even overcorrection from a high tibial osteotomy for a preexisting varus deformity. Likely a significant percentage of adult patients with lat- eral compartment osteoarthritis and associated valgus deformity repre- sent unresolved physiologic valgus deformity. Occasionally, persis- tence of genu valgum from child- hood may exist secondary to meta- bolic disorders, such as rickets and Dr. Favorito is Orthopaedic Surgeon, Welling- ton Orthopaedics and Sports Medicine, Cincinnati, Oh. Dr. Mihalko is Associate Pro- fessor, Department of Orthopaedic Surgery, State University of New York at Buffalo, Buffalo, NY. Dr. Krackow is Chief of Ortho- paedics, Kaleida Health System, The Buffalo General Hospital, and Professor of Orthopae- dics, State University of New York at Buffalo, Buffalo. Reprint requests: Dr. Krackow, The Buffalo General Hospital, 100 High Street, Suite B2, Buffalo, NY 14203. Copyright 2002 by the American Academy of Orthopaedic Surgeons. Abstract The valgus knee presents a unique set of problems that must be addressed dur- ing total knee arthroplasty. Both bone and soft-tissue deformities complicate restoration of proper alignment, positioning of components, and attainment of joint stability. The variables that may need to be addressed include lateral femoral condyle or tibial plateau deficiencies secondary to developmental abnor- malities, and/or wear; primary or acquired contracture of the lateral capsular and ligamentous structures; and, occasionally, laxity of the medial collateral ligament. Understanding the specific pathologic anatomy associated with the valgus knee is a prerequisite to selecting the proper surgical method to optimize component position and restore soft-tissue balance. J Am Acad Orthop Surg 2002;10:16-24 Total Knee Arthroplasty in the Valgus Knee Paul J. Favorito, MD, William M. Mihalko, MD, PhD, and Kenneth A. Krackow, MD Paul J. Favorito, MD, et al Vol 10, No 1, January/February 2002 17 renal osteodystrophy. Finally, val- gus alignment may represent a post- traumatic deformity and be the result of a tibial malunion, physeal arrest, or tibial plateau fracture. 14 Despite the well-known associations of valgus deformity of the knee with rheumatoid arthritis, rickets, and renal osteodystrophy, these patients make up a small proportion of those requiring TKA. In four of the five clinical series that utilized TKA in patients with valgus deformity of the knee, primary osteoarthritis was overwhelmingly the most common etiology, with a smaller number of patients having rheumatoid arthritis and posttraumatic arthritis. 5-9 Even less common etiologies included other inflammatory disorders and osteonecrosis. Alignment The normal mechanical axis of the knee is defined as a line that passes from the center of the hip to the cen- ter of the ankle (Fig. 1, A). Normal alignment is defined by the fact that the line passes through the center of the knee. In the knee with valgus deformity, the center of the joint lies medial to the mechanical axis (Fig. 1, B). The anatomic axes of the femur and tibia are represented by lines down the center of their re- spective shafts. The anatomic fem- oral axis is commonly 5 to 6 degrees lateral to or “offset” from the fem- oral mechanical axis, while the mechanical and tibial shaft axes are coincident. The 5- to 6-degree ana- tomic axis of the femur may be decreased by the pressures of coxa valgum or in patients who have undergone total hip arthroplasties. When normal knee alignment exists, the angle between the me- chanical axes of the femur and the tibia is zero. The radiographic de- formity can be defined as the angle drawn between the mechanical axis of the femur (i.e., the middle of the femoral head to the middle of the femoral surface of the knee, not of the entire lower extremity) and the shaft axis of the tibia. Alterna- tively, one can define deformity using the tibiofemoral angle, or the angle created between the anatomic axes of the femur and tibia. It is therefore very important to pro- vide the specifics of the orientation measurement system to avoid con- fusion. In this article, we use the term “tibiofemoral angle” to de- scribe the “anatomic tibiofemoral angle” to emphasize that we are considering the position of the ana- tomic shaft axes. Although various measurements have been reported, 1-9 a valgus knee generally is defined as a tibio- femoral angle >10 degrees. Clearly, a tibiofemoral angle of 7 to 9 de- grees is larger than normal, but patients with smaller deformities have typically not been included in study cohorts evaluating treatment of valgus knees. The definition of the deformity based solely on the position of bony landmarks on radiographs does not provide complete information re- garding the nature of the deformity, especially of the involvement of periarticular soft tissues. A major component of the surgeon’s task in managing deformity is the direct result of the coexisting asymmetry and/or contractures of the soft tis- sues, not just the abnormalities of bony alignment relative to the initial weight-bearing radiograph. For ex- ample, in a patient with acute frac- ture and collapse of the lateral tibial plateau (assuming no preexisting deformity and no secondary con- tracture of soft tissues since frac- ture), placement of a distracting device into the knee joint will “jack” open the lateral compartment so that a space defect is apparent, while the overall alignment of the tibia and femur is normal. This emphasizes the importance of assessing deformity in terms of the “asymmetry” of the soft-tissue sleeve. To assess that asymmetry intraoperatively, a tensor instru- ment as originally introduced by Freeman et al 12 or spacer blocks as described by Insall 13 may be used. Intraoperative Considerations Incision and Exposure A median parapatellar approach is most commonly used during sur- gery even though most of the pathology is on the lateral side of the valgus-deformed line. Patellar eversion is relatively easy because of the combination of the valgus deformity and the relative lateral- ization of the tibial tubercle. The A B Figure 1 A, The mechanical axis of the knee without deformity. B, The mechani- cal axis of the knee with valgus deformity. Total Knee Arthroplasty in the Valgus Knee Journal of the American Academy of Orthopaedic Surgeons 18 knee is then gently flexed and the joint can be exposed. Although most of the required soft-tissue releases are of the lateral structures and rather distant from the anteromedial arthrotomy, the valgus deformity sufficiently facilitates general expo- sure so that access to the posterolat- eral corner of the knee joint is not difficult, even in patients with ex- treme obesity. Keblish 15 has developed and re- ported on the use of a lateral reti- nacular approach for the valgus knee. The potential advantage of this exposure is the direct access to the lateral retinacular and capsular ligamentous tissues for release. In addition, there is no disruption of the medial blood supply to the pa- tella, thus presumably lessening the chances of patellar devascular- ization from sacrifice of the lateral geniculate vessels during aggres- sive lateral retinacular release. This lateral exposure has gained some proponents but has not achieved universal acceptance. There are two main considerations with this approach. The first is that the normal position of the tibial tubercle is lateral to the midline, and this position is accentuated by valgus deformity. A lateral para- patellar exposure therefore does not offer as wide a view of the central and medial aspects of the knee. To overcome this, it is typically neces- sary to perform a modified “wafer” osteotomy of the tibial tubercle, which must carry the inherent risks of postoperative patellar tendon failure and nonunion. The second issue relates to suffi- cient tissue for wound closure. After the lateral side of the knee has had the appropriate releases to achieve balance, there may be only skin and subcutaneous tissue avail- able for closure. Techniques have been developed to address this problem; these include a Z-cut cap- sulotomy or advancing the fat pad over the anterolateral joint line to provide closure elements. Neverthe- less, these two potential causes of morbidity must be weighed against the more direct access to the con- tracted tissues. Bony Architecture Lateral bone deficiency is a common component of the valgus knee de- formity and may involve both the lateral femoral condyle and/or the posterior aspect of the lateral tibial plateau (Fig. 2). Proper femoral component placement for TKA is achieved by referencing the cutting jig system on the posterior femoral condyles or epicondylar axis. How- ever, the epicondylar axis may be difficult to determine intraopera- tively. 16 Therefore, if the lateral femoral condyle is deficient and the posterior femoral condylar axis is used, an inappropriately large amount of bone will be resected from the posterior aspect of the lat- eral condyle. Improper resection may result in internal rotation of the femoral component and obligate medial placement of the patellar groove. Malpositioning of the femoral condyles has the effect of internally rotating the entire extrem- ity. Furthermore, the relationship with the contracted lateral liga- ments creates an abnormal patello- femoral alignment. The end result is an increased Q-angle and abnor- mal patellar tracking. Alternatively, proper femoral component rotation may be best achieved by using the anteropos- terior (AP) axis, as described by Arima et al. 16 The AP axis is de- fined by a line parallel to and bi- secting the intercondylar notch (a line through the deepest part of the patellar groove anteriorly and the center of the intercondylar notch posteriorly) (Fig. 2, B). This line is also approximately perpendicular to the epicondylar axis. In a study by Whiteside and Arima 17 reviewing 107 patients in whom the AP axis was used for TKA in the valgus knee, only one patient required a tibial tubercle transfer for patellar malalignment. Gap Kinematics The relationship of the proximal tibia to the distal femur, as repre- sented by the gap created by bone deficiency, and how it changes while performing ligament balanc- ing, may be referred to as gap kine- matics. Proper balancing of the joint gaps must be achieved so that once the gaps are filled with suit- ably articulating knee components, the static tension of the surrounding soft-tissue sleeve will allow for a stable construct. If asymmetry and inequality exist in the soft-tissue sleeve, properly articulating compo- nents may have no primary support to keep them stable under antago- nistic dynamic forces (Fig. 3). There- fore, it is mandatory to create an essentially stable joint gap through- out the entire range of flexion and extension. The valgus knee is approached by first determining if the deformity is passively correctable during the ini- tial clinical examination under anes- thesia, and then again intraopera- tively once all osteophytes have been removed. If, after the preliminary bone cuts have been performed, the deformity is corrected and the joint Figure 2 Bony and rotational deformity of the distal femur of the valgus knee in full extension (A) and flexion (B). Epicondylar axis AP axis Posterior condylar axis A B Paul J. Favorito, MD, et al Vol 10, No 1, January/February 2002 19 gaps are equal in both flexion and extension, no further lateral stabilizer releases are necessary. However, if there is even a modest residual val- gus deformity or instability, then some lateral release may be required. There is no consensus regarding the sequence in which the struc- tures about the knee should be re- leased. 12,13,18-20 Those structures most commonly addressed for re- lease include the iliotibial band, pos- terolateral capsule, lateral collateral ligament (LCL), popliteal tendon, and the lateral head of the gastroc- nemius muscle. Krackow and Mihalko 21 and Krackow et al 22 developed a kine- matic analysis model of the com- monly released lateral structures in a cadaveric model. After release of the LCL, popliteus, lateral gastroc- nemius, and iliotibial band, <5 degrees of correction could be achieved in full extension if the pos- terior cruciate ligament (PCL) was retained. If the release of the four lateral structures was combined with PCL sacrifice, a 9-degree cor- rection could be achieved. Releasing the LCL first allowed for a more gradual correction, with about 4 degrees obtained after initial LCL release and a gradual increase to about 9 degrees with release of suc- cessive secondary structures. Be- cause the LCL is the primary stabi- lizer of the lateral side of the joint, release of the secondary stabilizers (iliotibial band, popliteal tendon, posterolateral capsule) before the LCL may result in insufficient cor- rection. Subsequent release of the LCL may then result in overcorrec- tion and instability. However, some surgeons think that releasing the LCL first is inap- propriate. They may be correct, de- pending on when in the range of motion the joint remains tight. If the lateral side of the joint is tight in both extension and flexion, then subsequent release of the LCL must be performed to balance the joint gap properly throughout a range of motion. 19 If the joint gap is tight lat- erally only in extension, then release of the iliotibial band or possibly the popliteus may correct the balance in extension. Conversely, if the lateral joint gap is tight only in flexion, that deformity may be corrected by releasing the posterolateral capsule and popliteofibular ligament to equalize the flexion gap. All of these releases are typically done after PCL sacrifice. Another method of progressively releasing the lateral side involves using multiple small incisions with a scalpel blade through the taut pos- terolateral capsule with the knee in full extension. 23 This technique may place the common peroneal nerve at risk. The effectiveness of this method for achieving release has been eval- uated in both the preclinical and clinical settings. With the lateral side of the knee joint and LCL pro- tected, the posterolateral capsule was incised, and effective correction occurred only after the LCL was divided and essentially released. This same technique was then ana- lyzed in the laboratory using cadav- eric kinematic analysis. It was evi- dent that <4 degrees of correction could be obtained if only the pos- terolateral capsule were released without the LCL. 23 In addition, the common peroneal nerve was found to be between 7 to 9 mm from the posterolateral capsule in full exten- sion. These measurements were made, however, in cadaveric knees without deformity and therefore without contracted lateral anatomic structures. The distance from the peroneal nerve to the posterolateral capsule may be even smaller. If release of the lateral structures does not sufficiently stabilize flexion and extension gaps, then the medial side of the joint should be addressed. Several techniques have been de- scribed for successfully and safely “tightening” the incompetent medial collateral ligament (MCL). Krackow et al 5 described MCL advancement off the tibial side, and Krackow 24 described MCL midsubstance divi- sion and imbrication, to equalize the joint gaps (Fig. 4). Healy et al 2 de- scribed recessing the origin of the MCL with a bone block from the femoral epicondyle. Although these procedures are technically demand- ing and may affect ligament strength and isometricity, they may be neces- sary to equalize joint gaps to achieve a stable and durable result. Component Selection Another important consideration in the management of valgus defor- mity is prosthesis selection with regard to the degree of component constraint. Ideally, if proper soft- tissue balance is restored, a mini- mally constrained component then can be implanted. Although most surgeons agree that a more con- strained posteriorly stabilized com- ponent should be used if signifi- cant deformity necessitates PCL sacrifice for soft-tissue balancing, it is not universally accepted. Such a prosthesis provides some degree of posterior stabilization as well as protection against posteromedial, Figure 3 Improper soft-tissue balancing with a lax medial soft-tissue sleeve (A) that allows opening of the joint in distraction (B) or under a valgus-directed force. A B Total Knee Arthroplasty in the Valgus Knee Journal of the American Academy of Orthopaedic Surgeons 20 posterolateral, straight medial, or straight lateral translation, but it will not protect against residual medial laxity, which is one of the major considerations in achieving proper balance. The surgeon should resist the temptation, when possible, to move to a more highly constrained pros- thesis, such as a totally stabilized prosthesis, to compensate for short- comings in achievable soft-tissue balancing. Although highly con- strained components may be neces- sary in difficult revision cases, they are infrequently necessary for pri- mary TKA. The patient with severe valgus knee deformity also may have a stretched or elongated PCL because of the more medial position of the PCL. Therefore, even if the PCL is retained in a severely valgus knee, it may be nonfunctional and require either an ultracongruent or posteriorly stabilized component. Component selection for the val- gus knee with an extremely defi- cient lateral femoral condyle may require the use of component aug- mentation if the femoral component is being cemented. The lateral fem- oral condyle may have had little or no distal femoral bone resected or, similarly, little to no bone resected from the chamfer and posterior cuts, as well. These cuts may require component augmentation. However, if the femoral component is being press-fit, then as long as native bone is resting on one of the chamfer cuts (as is usually the case for the poste- rior bevel or chamfer cut), then the remaining defect can be filled with autograft bone taken from other cuts during the procedure. 8 Surgical Technique After surgical exposure of the joint and débridement of osteophytes, a tension-stress examination is per- formed with the knee in full exten- sion and the tibia distracted from the femur. Small-arc varus and val- gus forces are applied to assess when the medial and lateral aspects of the soft-tissue sleeve are equally taut. With the tibia held steadily at this point, an approximate assess- ment of the overall alignment and existing tibiofemoral angle can be made. In this way, it is possible to make an initial estimate of the de- gree of soft-tissue sleeve asymme- try, which will need to be managed by “balancing.” Next, the general shape of the distal femur is assessed. Both the AP axis of Whiteside and the epicondylar axis are used. In some cases of valgus deformity, the lateral epicondyle is not prominent, and defining a distinct lateral point for the axis may be imprecise. With these axes in view, the amount of posterior femoral condylar deformi- ty is estimated. The knee is then extended and the presence of any flexion contracture or recurvatum can be assessed. The se- verity or degree of these two features may affect the relative proximal- distal positioning of the distal fem- oral cut. With most modern cutting jigs positioned to guide the distal cut, one expects the jig to encounter the more prominent distal medial condyle and thus to stand posi- tioned more distally, away from the lateral condyle. This examination allows a rough estimate of wear or deformity of the lateral femoral condyle distally. The knee is then fully extended, and the cut is refer- enced from the more distal (usually medial) condyle. In the presence of a flexion contracture not completely addressed by capsular release, a somewhat more proximal cut may be made. In the valgus knee with such asymmetric presentation, the decision whether the distal cut should be referenced from the prom- inent condyle, the deficient condyle, or somewhere in between is made based on the situation of the knee at its maximum extension. When the knee comes to full extension, the cut is referenced from the more distal (usually medial) condyle. In the Figure 4 A, The epicondylar origin in this lax MCL is represented by the cross-hatched circle. B, The MCL is removed from the epicondyle and is advanced to remove the laxity. C, A running locking nonabsorbable suture is then used to secure the ligament in its new position, and a surgical staple is placed at the epicondylar origin. (Adapted with permis- sion from Krackow KA: Management of medial collateral ligament loss: Repair and aug- mentation, in Lotke PA, Garino JP (eds): Revision Total Knee Arthroplasty. Philadelphia, Pa: Lippincott-Raven, 1999, pp 227-250.) A B C Paul J. Favorito, MD, et al Vol 10, No 1, January/February 2002 21 presence of a flexion contracture not likely to be completely addressed by capsular and/or osteophyte re- lease, a somewhat more proximal cut may be made. In the rare case of initial recurvatum, an even more distal cut may be necessary. Alter- natively, a smaller component can be selected in an attempt to increase the flexion gap to match the larger than normal extension gap that allows recurvatum to occur. The distal femoral cut is oriented so that it is perpendicular to the mechanical axis of the femur, i.e., the center of femoral head to the center of the distal femur. Usually, the anterior and posterior femoral cuts are made first; however, it is acceptable to perform the tibial cut first and then proceed with most or all of the soft-tissue release. The amount of valgus alignment in the distal femoral cut is determined from preoperative, long-standing radiographs (Fig. 5, A) by measur- ing the angle between a line along the femoral diaphysis through the center of the knee, and one from the center of the knee to the center of the femoral head (Fig. 1, A). After the tibial cut and soft-tissue release, one may assess the rotational position of the femur as it is dis- tracted away from the tibia with the knee in flexion. Although the final determination of the anterior and posterior femoral cuts can be based entirely from the tibia, attempting to make a perfectly rectangular 90- degree flexion space can be danger- ous. Within limits, it may be possi- ble to achieve greater symmetry of the flexion space by some minor ro- tational alterations in accordance with the relative orientation of the cut tibial surface. The amount of tib- ial resection is usually first gauged from the more prominent side (usu- ally medial), ignoring an erosive de- fect if one is present. If the eroded lateral tibial compartment does not allow rim contact, then one can in- crease the size of the cement man- tle or increase the amount of resec- tion until adequate support results. This usually does not require any significant amount of added resec- tion. Generally, we position the fem- oral instrumentation first to select component size and then perform the anterior and posterior femoral cuts. Many jigs utilize “skids,” which contact the posterior femoral condyles, and then use some scheme for off-setting the rotational axis. After assessing the component rota- tion as determined by the instru- mentation, we recheck and confirm that orientation with the Whiteside axis and, if reliably visible, with the epicondylar axis. Any single refer- encing technique or combination of techniques would then be selected and the remaining femoral cuts made. Deciding when to begin and com- plete soft-tissue balancing is influ- enced by surgeon preference and the degree of deformity. If a sub- stantial release is necessary, it may be more appropriate to perform this earlier in the procedure. Repeat evaluation of joint gaps and balance is of paramount importance after each individual structure is divided. The first step is to assess the lateral and posterolateral corner of the knee to determine tension in the iliotibial band. Occasionally, but rarely, the iliotibial band may appear to be the tightest structure; it would therefore require release first. More commonly, the LCL is the tightest structure and is released initially. The LCL is sharply elevated from the lateral epicondyle until it is completely released. During the LCL release, the popliteal tendon should be iden- tified and protected to avoid inad- vertent division. The effect of the LCL release is then assessed by a tension-stress examination in full extension, partial flexion, and 90 de- grees of flexion. It is helpful to place a tagging suture on the stump of the LCL for future identification. The next release is that of the popliteal tendon, either at or near the joint line. Any bridging connections between the popliteus and the LCL or tibia are separated. At this point, there will be definite opening of the lateral aspect of the knee, more pro- nounced in flexion than extension. Other structures that may require release include the posterolateral capsule and femoral origin of the gastrocnemius muscle complex, especially in the setting of flexion contracture. Finally, the iliotibial band can be considered. Release of the biceps femoris tendon and/or exposure of the peroneal nerve gen- erally is not recommended. If both the LCL and the popliteal tendon have been released, they are “repaired” to one another with a locking-loop ligament suture for maximum strength. The purpose of this repair is to provide support in flexion to avoid excessive lateral gapping. Release of both the LCL Figure 5 A, Thirty-six-inch long-standing preoperative radiograph of valgus defor- mity. B, Postoperative thirty-six-inch long- standing radiograph after TKA, showing restoration of proper mechanical align- ment. A B Total Knee Arthroplasty in the Valgus Knee Journal of the American Academy of Orthopaedic Surgeons 22 and popliteus from the lateral fem- oral epicondyle, as described by Insall, 13 has been successful without secondary lateral flexion instability. Even the patient who undergoes a release of all lateral structures to balance the knee does not need ad- ditional care postoperatively beyond that of a routine TKA, except for the addition of a knee brace. If initially there was felt to be stretching of the medial capsular ligamentous complex, it may be impossible to balance the knee by addressing only the lateral side. In this situation, utilization of a tech- nique to tighten the ligamentous structures of the medial side and/ or the use of a more highly con- strained intercondylar prosthesis (nonhinged) may be appropriate. Advancement of the MCL, as previously described 5 (Fig. 4), or division and imbrication of the MCL 24 (Fig. 6), can be done in con- junction with use of a constrained intercondylar prosthesis to protect against gravity distraction of the leg and dissociation of the intercon- necting peg. This is a very simple technique, which together with the constraint of the prosthesis requires no alteration of patient aftercare. Complications Results of past clinical studies, 2,3,5-9,25 clearly indicate that several compli- cations have been reported more fre- quently in this subset of patients. The most commonly reported com- plications in patients with valgus deformities who undergo TKA are tibiofemoral instability (2% to 70%), recurrent valgus deformity (4% to 38%), postoperative motion deficits requiring manipulation (1% to 20%), wound problems (4% to 13%), patel- lar stress fracture or osteonecrosis (1% to 12%), patellar tracking prob- lems (2% to 10%), and peroneal nerve palsy (3% to 4%). 2,3,5-9,25 Idusuyi and Morrey 25 reported 32 postoperative peroneal nerve palsies in more than ten thousand consecu- tive TKAs. Of the 32 palsies, 10 knees had 12 degrees of preopera- tive valgus deformity or more. This problem presumably is caused by lengthening the lateral aspect of the knee during lateral stabilizer release and subsequent traction to the pero- neal nerve. It is generally recom- mended that patients be evaluated carefully for symptoms postopera- tively. If peroneal nerve palsy–type symptoms are discovered, the knee should be flexed to relax the tension that is effectively being placed on the nerve. There are no objective guidelines or data to support the efficacy of any immediate surgical intervention. Clinical Results Krackow et al 5 retrospectively re- viewed 99 arthroplasties for valgus knees in 88 patients and compared them to a control group with mini- mal deformity. They identified three types of valgus knees: type I had a valgus deformity secondary to bone loss in the lateral compartment with medial soft-tissue contracture; type II had obvious attenuation and in- competence of the medial compart- ment; and type III was the result of an overcorrected proximal tibial os- teotomy for varus deformity. All arthroplasties were performed with a minimally constrained PCL-sparing prosthesis. Type I patients were treated with lateral soft-tissue re- lease only, and type II patients were treated with lateral soft-tissue re- lease and MCL advancement. No type III patients were treated. Post- operative rating scores for align- Figure 6 Imbrication of the MCL can also be used to equalize the joint gap. A, Two running locking sutures are placed with the amount of ligament to be imbricated represented by the distance between the two suture ends. B, The MCL is then transected between the two running locking sutures. C, The respective suture ends are tied together to complete the imbrication. (Adapted with permission from Krackow KA: Management of medial collateral ligament loss: Repair and augmentation, in Lotke PA, Garino JP (eds): Revision Total Knee Arthroplasty. Philadelphia, Pa: Lippincott-Raven, 1999, pp 227-250.) A B C Paul J. Favorito, MD, et al Vol 10, No 1, January/February 2002 23 ment and function were equivalent for both type I and II patients. There were more fair (7%) and poor (2%) results in the study group compared with the control group. The authors stated that ligament reconstruction was effective and was important in knee stability even when more con- strained prostheses are used. Stern et al 7 reviewed 134 arthro- plasties in 98 patients with valgus deformity of >10 degrees. Eighty- seven percent of patients were treated with a posteriorly stabilized prosthesis. Ligamentous balancing was done with sequential releases from the lateral side of the femur only and did not include medial ligament reconstruction. The knees were stratified into subgroups based on the severity of preopera- tive deformity. The postoperative alignment goal was 5 to 9 degrees of valgus. At an average follow-up of 4.5 years, 91% of patients had good or excellent results. The authors recognized the difficulty of achieving proper femoral compo- nent rotation because of the defi- cient lateral femoral condyle and concluded that a constrained femoral component is necessary for the severe valgus knee. Laurencin et al 6 retrospectively reviewed 25 arthroplasties in knees with an average preoperative valgus deformity of 25 degrees. In all pa- tients, a lateral retinacular release was performed. The remaining lat- eral structures were released as needed to achieve proper soft-tissue balancing. Twenty-one of 25 pa- tients had an unconstrained PCL- retaining prosthesis. Twenty-four of 25 patients had postoperative ana- tomic valgus alignment of 0 to 10 degrees. Postoperative flexion aver- aged 110 degrees, and average flex- ion contracture measured 2 degrees (range, 0 to 12 degrees). There were significant complications in nine knees, including patellar fractures secondary to osteonecrosis (three knees), patellar instability (one), per- oneal nerve palsy (one), and recur- rence of deformity (one). The au- thors underscored the importance of soft-tissue balancing and preserving the superior lateral geniculate artery when performing a lateral release in conjunction with a medial parapatel- lar approach. Whiteside 8 reviewed 135 knees with valgus deformity treated with a minimally constrained prosthesis. Seventy-one percent of patients with <25 degrees of preoperative valgus had a lateral ligamentous release. In 11 knees with >25 de- grees of deformity, the deficient lat- eral condyle was used as a point of reference, resulting in overresection of the medial distal surface. Six of the 11 patients required medial ligament advancement to achieve stability in extension. Mean postoperative val- gus was 7 degrees compared with 16 degrees preoperatively. There was no deterioration of alignment postoperatively. Knees with >25 degrees of preoperative valgus de- formity had an increased incidence of posterior laxity. Accurate bone resection, correct alignment, and bone graft to fill femoral and tibial defects were thought to be impor- tant factors in achieving good re- sults. Karachalios et al 3 performed a prospective case-control study com- paring severely deformed knees (defined as >20 degrees of varus or valgus malalignment) with minimal- ly deformed knees. All patients were treated with a minimally constrained PCL-retaining prosthesis. The pa- tients demonstrated no significant difference in function postoperatively comparing the group with severe varus to that with valgus deformities. The authors did note a higher inci- dence of postoperative residual val- gus deformity and patellofemoral malalignment in the valgus group. They concluded that failure to obtain full correction of valgus deformity may lead to residual patellar tracking problems. Miyasaka et al 9 reviewed 108 knees in 83 patients with an average follow-up of 14 years. All patients were treated with a standardized sequence of soft-tissue releases from the lateral side, starting with the ilio- tibial band and lateral retinaculum, followed by the LCL and popliteal tendon when necessary. Soft-tissue releases were performed before bone cuts were made. Ninety-eight val- gus knees were reviewed, with 10 patients requiring a highly con- strained component because of excessive instability. Postoperative results were deemed acceptable by the authors despite a 24% rate of knee instability. As a result, a dif- ferent approach was subsequently developed, performing bony cuts first, followed by soft-tissue balanc- ing and pie-crusting of the iliotibial band. Healy et al 2 reviewed eight pa- tients with type II valgus deformi- ties (lateral soft-tissue contractures with lax medial soft-tissue stabiliz- ers) managed with MCL advance- ment. Seven of eight patients had condylar, nonconstrained PCL- retaining components, and one had a PCL-substituting implant. Lateral structure releases included resection of femoral or tibial osteophytes, division of the iliotibial band and popliteal tendon, and release of the arcuate ligament. The MCL with a bone plug and incorporated liga- ment stitch was advanced proximally and laterally and tied over a button or bony bridge on the lateral cortex. Recession of the bone plug was thought to allow bone-to-bone heal- ing while isometrically tightening the MCL. At an average follow-up of 5.8 years, all patients were satis- fied with the procedure, as demon- strated by decreased pain and by improved function. Radiographic tibiofemoral alignment ranged from 3 to 7 degrees of valgus. The authors consider this to be a simple and re- producible technique for eliminating MCL laxity during arthroplasty. Total Knee Arthroplasty in the Valgus Knee Journal of the American Academy of Orthopaedic Surgeons 24 Summary The valgus knee presents a chal- lenge to the joint replacement sur- geon. The principles of TKA must be applied while taking into account preexisting anatomic deformities. Understanding the femoral anatomy and using the AP axis for femoral component placement may help prevent postoperative patellofem- oral maltracking and instability. Recognizing the soft-tissue asym- metry and using the tension-stress examination to evaluate this allows a structured approach to proper bal- ancing. As a result, the surgeon may more confidently achieve soft- tissue balancing, resulting in better load distribution and enhancing component stability and longevity. References 1. Buechel FF: A sequential three-step lat- eral release for correcting fixed valgus knee deformities during total knee ar- throplasty. Clin Orthop 1990;260:170-175. 2. Healy WL, Iorio R, Lemos DW: Me- dial reconstruction during total knee arthroplasty for severe valgus defor- mity. Clin Orthop 1998;356:161-169. 3. Karachalios T, Sarangi PP, Newman JH: Severe varus and valgus deformi- ties treated by total knee arthroplasty. J Bone Joint Surg Br 1994;76:938-942. 4. Keblish PA: The lateral approach to the valgus knee: Surgical technique and analysis of 53 cases with over two- year follow-up evaluation. Clin Orthop 1991;271:52-62. 5. Krackow KA, Jones MM, Teeny SM, Hungerford DS: Primary total knee ar- throplasty in patients with fixed valgus deformity. Clin Orthop 1991;273:9-18. 6. Laurencin CT, Scott RD, Volatile TB, Gebhardt EM: Total knee replacement in severe valgus deformity. Am J Knee Surg 1992;5:135-139. 7. Stern SH, Moeckel BH, Insall JN: Total knee arthroplasty in valgus knees. Clin Orthop 1991;273:5-8. 8. Whiteside LA: Correction of ligament and bone defects in total arthroplasty of the severely valgus knee. Clin Orthop 1993;288:234-245. 9. Miyasaka KC, Ranawat CS, Mullaji A: 10- to 20- year followup of total knee arthroplasty for valgus deformities. Clin Orthop 1997;345:29-37. 10. Hungerford DS: Putting it all togeth- er: The importance of the trial reduc- tion in total knee replacements. Ortho- pedics 1998;21:1032-1033. 11. Chandler HP: Structural bone grafting about the knee. Orthopedics 1998;21: 1044-1045. 12. Freeman MA, Todd RC, Bamert P, Day WH: ICLH arthroplasty of the knee: 1968-1977. J Bone Joint Surg Br 1978;60: 339-344. 13. Insall JN (ed): Surgery of the Knee. New York, NY: Churchill Livingstone, 2000, pp 1558-1562. 14. White GR, Mencio GA: Genu valgum in children: Diagnostic and therapeutic alternatives. J Am Acad Orthop Surg 1995;3:275-283. 15. Keblish PA: Valgus deformity in total knee replacement (TKR): The lateral retinacular approach. Orthop Trans 1985;9:28-29. 16. Arima J, Whiteside LA, McCarthy DS, White SE: Femoral rotational align- ment, based on the anteroposterior axis, in total knee arthroplasty in a val- gus knee: A technical note. J Bone Joint Surg Am 1995;77:1331-1334. 17. Whiteside LA, Arima J: The antero- posterior axis for femoral rotational alignment in valgus total knee arthro- plasty. Clin Orthop 1995;321:168-172. 18. Laskin RS (ed): Total Knee Replacement. London, UK: Springer-Verlag, 1991, pp 46-53. 19. Whiteside LA: Abstract: Selective liga- ment release in total knee replacement of the valgus knee. Proceedings of the American Academy of Orthopaedic Surgeons 66th Annual Meeting, Anaheim, Calif. Rosemont, IL: AAOS, 1999, p 257. 20. Hungerford DS, Lennox DW: Fixed valgus deformity, in Hungerford DS, Krackow KA, Kenna RV (eds): Total Knee Arthroplasty: A Comprehensive Approach. Baltimore: Williams and Wilkins, 1984, pp 167-178. 21. Krackow KA, Mihalko WM: Flexion- extension joint gap changes after later- al structure release for valgus deformi- ty correction in total knee arthroplasty: A cadaveric study. J Arthroplasty 1999; 14:994-1004. 22. Mihalko WM, Miller C, Krackow KA: Total knee arthroplasty ligament bal- ancing and gap kinematics with poste- rior cruciate retention and sacrifice. Am J Orthop 2000;29:610-616. 23. Ranawat C: Live surgical presentation: Techniques in Total Joint Arthroplasty. New York, NY, May 21-25, 1997. 24. Krackow KA: Management of medial collateral ligament loss: Repair and augmentation in Lotke PA, Garino JP (eds): Revision Total Knee Arthro- plasty.Philadelphia, Pa: Lippincott- Raven, 1999, pp 227-250. 25. Idusuyi OB, Morrey BF: Peroneal nerve palsy after total knee arthroplas- ty: Assessment of predisposing and prognostic factors. J Bone Joint Surg Am 1996;78:177-184.