Secondary Logo

Journal Logo

Techniques of Medial Retinacular Repair and Reconstruction

Davis, Daniel, K.; Fithian, Donald, C.

Clinical Orthopaedics and Related Research: September 2002 - Volume 402 - Issue - p 38-52

Insufficiency of the passive patellar restraints results in lateral patellar instability by allowing excessive lateral displacement of the patella. Although the surgical approach to patellar instability traditionally has been to realign the dynamic elements (muscle forces) that pull the patella laterally, newer techniques have sought to restore the integrity of key medial passive (ligamentous) stabilizers. An increasing body of evidence indicates that the chief medial ligamentous restraint is the medial patellofemoral ligament. The current authors examine the principles of medial retinacular repair and reconstruction as they relate to patellar stability. Individual techniques and approaches are discussed, including primary repair with or without augmentation, and reconstruction using autogenous tendon, allografts, and synthetic graft materials. These procedures share the common objective of addressing the essential lesion in lateral patellar instability to restore the normal passive restraints against lateral patellar displacement.

From the Southern California Permanente Medical Group.

A portion of this work was supported by community service funds of the Kaiser Foundation Health Plan and a Career Development Award from the Orthopaedic Research and Education Foundation to the senior author (DCF).

Reprint requests to Donald C. Fithian, MD, 250 Travelodge Drive, El Cajon, CA 92020.

DOI: 10.1097/01.blo.0000026961.51742.42

In an article on the semantics of knee instability, Noyes et al 48 defined instability as “a condition of increased mobility between two structures.” For the purposes of the current paper, patellar instability will be defined in that way. There is increasing evidence that patellar mobility is increased in knees with a history of patellar dislocation, 20,34,37,50 that the increased mobility seems to be attributable to abnormal soft tissue laxity in the medial retinacular structures, 20,61 which frequently show signs of injury after patellar dislocation, 21,33,45,54 that specific structures within the medial retinaculum show a fairly consistent pattern of injury in that setting, 21,33,45,54,65 and that in studies done on cadavers, some of these injured tissues seem to be important in resisting lateral patellar motion. 12,18,28

More than 100 procedures have been described for the treatment of patellar instability. 30 The majority of these procedures focus on the dynamic elements that determine the magnitude and direction of forces acting on the patella during active knee extension (the quadriceps angle). 13,32 Most procedures are intended to realign the knee extensor mechanism by manipulating the quadriceps angle to reduce the tendency for the patella to move laterally under quadriceps loading. 6,10,13,22,27,31,35,51,63 Hughston 31 coined the terms proximal and distal reconstruction to differentiate between procedures that manipulate the insertions of individual components of the extensor apparatus into the patella (proximal), from those procedures that relocate the tibial tubercle (distal). All of these realignment procedures have a common objective: to straighten or align the pull of the extensor mechanism to reduce the forces tending to lateralize the patella with respect to the femoral trochlea.

Muscle contraction can have inconsistent effects on joint mobility; it either can cause or prevent abnormal joint motions depending on the magnitude and direction of the resultant muscle force relative to the ligament deficiency. 5,56 Because muscle forces can reduce the apparent limits of joint motion by increasing joint contact force and reducing shear compliance, care must be taken when examining a joint for instability, that the muscles are relaxed. 15,16,57 Alternatively, muscle forces that displace a joint in the direction of its pathologic laxity will result in subluxation. 56 The alignment of the extensor mechanism determines whether quadriceps contraction will tend to reduce the patella in the trochlea or displace it from the trochlea. However, the normal patella cannot be dislocated because the passive restraints prevent it from being displaced from the trochlea. 20,28 There exists no evidence that any amount of malalignment will cause dislocation unless the passive stabilizers are damaged. However, a hypermobile patella is unstable using the definition of Noyes et al 48 even if the muscles are realigned to eliminate lateralizing forces. Using that definition of patellar instability, excessive passive laxity is the essential element in instability of the patellofemoral joint, and the role of extensor alignment and muscle forces is not clear.

To restore normal passive motion limits in an unstable joint, it is necessary to repair or reconstruct the ligamentous structures whose function it is to prevent excessive motion. During the past several decades, the desirability of restoring or reconstructing injured passive motion restraints has become an important surgical principle in the reconstruction of many unstable joints. Concerns about donor site morbidity, loss of joint motion, arthrosis, and pain have stimulated interest in addressing the essential lesion rather than doing muscle or tendon transfers, realignments, arthrodesis, or excision for the treatment of joint instability. 26 Examples of this principle include the Bankart repair in the shoulder, anterior talofibular ligament reconstruction in the ankle, medial collateral ligament reconstruction of the elbow, and intraarticular anterior cruciate ligament reconstruction of the knee. An increasing number of authors have advocated a similar approach for treatment of patellofemoral instability. 19,21,23,41,45,46,54,59 The current authors summarize the biomechanics and pathologic anatomy of lateral patellar instability, and discuss current techniques for medial capsular repair and reconstruction, emphasizing the principles on which each technique is based.

Back to Top | Article Outline

Excessive Mobility (Instability) and Patellar Dislocation

Excessive patellar mobility often is seen in patients with a history of acute lateral patellar dislocation 5,20,34,50,58,61 (Table 1). Bassett 5 described the absence of the normal checkrein when a manually applied lateral force was applied to dislocatable patellas. This is analogous to the absent or soft end point on Lachman testing of anterior cruciate-deficient knees. 16 Subsequently, several authors 20,34,61 reported greater patellar mobility among patients with a history of patellar dislocation than among control subjects. Tables 1 and 2 show displacement data from instrumented measurement and stress radiography studies. Comparison of the results in Tables 1 and 2 suggests that greater displacements are seen when greater force is applied. Table 2 shows data from two studies that studied individual variations in laxity among study subjects, including comparisons between the two knees of a given subject. In a study by Fithian et al, 20 injured and uninjured knees in subjects with unilateral dislocation had greater patellar mobility than was seen among healthy control subjects (p < 0.01) (Table 2). The mean lateral patellar mobility among injured knees of subjects with unilateral dislocation was slightly greater than measured lateral displacements in the uninvolved knees 21 (Table 2). Medial laxity was similar for both knees 20; when medial laxity was used to control for variations in individual subjects’ joint laxity by subtracting medial displacement from lateral displacement (lateral minus medial difference), the difference in lateral laxity between the injured and uninjured knee of subjects with unilateral dislocation was statistically significant (p < 0.01) 21 (Table 2). Among patients with a history of unilateral patellar dislocation, both knees had increased patellar mobility, and the injured knee of subjects with unilateral dislocation had greater lateral patellar mobility than that observed in the contralateral, uninjured knee.





Abnormal passive patellar motion limits in the uninjured knees of these patients may be attributed to patella alta, 3,38 a wide trochlear sulcus angle, 39,40 hypoplastic or attenuated medial retinacular ligaments, 52 dysplastic or injured quadriceps muscle, 5,47,54 or a combination. Considering that no significant differences of bony anatomy have been documented between the two knees of subjects with unilateral dislocation, 3,9 the observed side-to-side differences in patellar mobility must be attributable to greater passive laxity in the medial soft tissues of the injured knees.

Back to Top | Article Outline

Contributions to Patellar Stability

Two components of the knee extensor apparatus primarily affect the limits of mediolateral patellar motion: bony constraint attributable to congruity between the patella and the femoral trochlea, and ligamentous tethers. The combination of articular buttress and ligamentous tension determines the limits of passive patellar displacement.

Acute patellar dislocation has been associated with specific developmental anomalies including patella alta, 38,52 trochlear dysplasia, 17 and rotational and angular bony malalignment. 43 Trochlear dysplasia and patella alta, which reduce the containment of the patella within the femoral trochlea, contribute directly to the risk of recurrent patellar dislocation by increasing the passive motion limits of the patella. Soft tissue dysplasias also are seen more commonly among patients with patellar dislocations than among healthy subjects. 47,52,60 Muscular weakness or imbalance has been associated with patellar instability. It is not known whether it is developmental 47 or the result of dislocations. 5,20,54 Ligamentous hyperlaxity also has been described in patients with patellar instability. 52,60

As in other articulations, the magnitude and direction of joint compressive forces affect patellofemoral kinematics. This particularly is true during active muscle contraction. In a study by Hautamaa et al, 28 the application of as little as 5-lb load to the quadriceps tendon caused a measurable reduction in patellar displacement in response to medially or laterally directed force. Muscle activity can affect patellar motion either by increasing joint reaction force or by generating net medializing or lateralizing force vectors within the patellofemoral joint. Therefore, depending on whether the muscle forces tend to reduce or displace the patella with respect to the trochlea, muscle activity has an inconsistent effect on patellar stability. If quadriceps activation reduces the patella, it prevents medial or lateral displacement and protects against dislocation; if quadriceps activation displaces the patella from the trochlea, it can cause dislocation if the passive medial restraints (ligaments) and lateral trochlear buttress fail to contain the patella.

Even when muscles are aligned so as to center the patella in the trochlea, they must be activated to do so. Although it is possible that passive muscle tension in the vastus medialis obliquus resists lateral patellar displacement, this possibility has not been studied. Muscles are designed to do work; for them to do the work of passive stabilizers is inefficient. Muscle activity requires effort and results in compressive joint forces to compensate for ligamentous laxity. It is possible that the generation of high joint reaction forces may be partially responsible for the arthrosis that can occur after realignment surgery for recurrent patellar dislocation. 2,14 Advancement of the vastus medialis obliquus to increase passive stiffness would have unpredictable effects because the long-term response of vastus medialis obliquus muscle fibers to increased resting length is unknown.

The contribution of specific medial retinacular structures to restraint against lateral patellar displacement has been studied in normal knees from cadavers using sequential cutting methods. 12,18,28 Ligamentous retinacular structures that may be relevant to lateral patellar instability include: (1) the superficial medial patellar retinaculum 49; (2) the medial patellotibial ligament 62; (3) the medial patellomeniscal ligament 12,18,28; and (4) the medial patellofemoral ligament. 8,12,18,28,49 These studies consistently have shown that the medial patellofemoral ligament is the primary ligamentous restraint against lateral patellar displacement.

In studies done on cadavers, the medial patellofemoral ligament contributed 41% to 80% of the restraining force against lateral patellar displacement. 12,18 Desio et al 18 reported that isolated lateral release actually reduces resistance to lateral displacement. Tables 3 and 4 summarize the studies done on cadavers of Hautamaa et al, 28 who used the same technique for measuring force and displacement that was used in earlier studies on live humans. 20 Repair of the medial patellofemoral ligament alone restored lateral mobility to within normal values. 28 Repair of more superficial retinacular tissues, as typically seen with medial reefing was neither necessary nor sufficient to restore stability 28 (Tables 3, 4).





Comparing data from intact knees from cadavers in Tables 3 and 4, 28 with data from knees of healthy volunteers in Tables 1 and 2, 20 lateral displacements and lateral minus medial displacement differences at 5-lb force are similar. Isolated section of the medial patellofemoral ligament resulted in lateral displacements and mediolateral balance similar to that reported in the injured knees of subjects with patellar dislocations.

Back to Top | Article Outline

Soft Tissue Injury in Patellar Dislocation

Surgical and magnetic resonance imaging (MRI) data show typical patterns of medial retinacular injury in the majority of cases of acute primary (first-time) lateral patellar dislocation. 54,66 Clearly, a significant retinacular injury occurs during the primary dislocation event in most cases, and frequently involves the medial patellofemoral ligament and the medial patellomeniscal ligament. Injury to the medial patellofemoral ligament may occur at more than one location along its length during the dislocation. 8,54

Burks et al 8 reported a simulation of patellar dislocation using normal knees from cadavers, which directly compared MRI and gross anatomic findings. The medial patellofemoral ligament was injured in eight of 10 knees. Nomura 45 proposed a classification scheme (Fig 1), which aids in understanding the nature of the medial patellofemoral ligament injury. Seventeen of 18 patients with acute dislocations had an injury to the medial patellofemoral ligament, all at the level of the femur. The one medial patellofemoral ligament that had no visible injury was described as loose. 45 These findings parallel those of Sallay et al, 54 who found tears of the medial patellofemoral ligament off the femur in 15 of 16 patients. In another clinical study in which the medial soft tissues were explored surgically, the medial patellofemoral ligament was disrupted in all 20 patients. 21 Ten were injured near the margin of the patella and 10 were injured posteriorly, at the femoral origin.

Fig 1A–C.

Fig 1A–C.

Back to Top | Article Outline

Biologic Considerations in the Treatment of Medial Patellofemoral Ligament Injury

Similar to the medial collateral ligament, the medial patellofemoral ligament is an extrasynovial ligament. 67 Its biology in recurrent patellar instability has not been studied. It has not been determined whether a rupture of the medial patellofemoral ligament results, after healing, merely in lengthening of the ligament, as in medial collateral ligament injuries, 42 or in a completely incompetent ligament, as in anterior cruciate ligament injuries. The long-term effects of rupture at the epicondyle, in the midsubstance, and at its patellar insertion have not been studied. However, based on its location within the second layer of the medial retinaculum, it is reasonable to expect that its competence will be restored in many cases. The previously reported findings in the dissection of 49 chronic lesions of the medial patellofemoral ligament support this idea. 45 Nomura reported healing in the form of scar in 29 cases, a loose femoral attachment (unhealed avulsion) in nine cases, and inadequate tissue in 11 cases. 45

Back to Top | Article Outline

Repair and Reconstruction of the Medial Patellofemoral Ligament

Given the data showing that posttraumatic lengthening of the medial retinacular ligaments occurs, it is reasonable to hypothesize that the lengthening of these ligaments may be a critical factor in converting the knee from an asymptomatic but predisposed joint into one with frank recurrent instability. Evidence from studies done on cadavers indicates that the medial patellofemoral ligament must be repaired or reconstructed if lateral patellar motion is to be restored to its preinjury limits 12,18,28 (Tables 3, 4). Therefore, one objective of any surgical repair should be to restore the medial patellofemoral ligament to its preinjury length and stiffness.

Back to Top | Article Outline

Primary Repair of the Medial Patellofemoral Ligament

The medial patellofemoral ligament passes from the medial femoral epicondyle and adductor tubercle to the superomedial ⅔ of the patella, fusing anteriorly with the undersurface of the vastus medialis obliquus tendon. 67 Primary medial patellofemoral ligament repair has been approached in various ways. The repair may be done acutely 44 or after a period of initial healing. 19 Repair has been described at the patellar insertion of the medial patellofemoral ligament 19,55 or at its origin on the adductor tubercle. 1,44,54 In the acute setting, the surgical approach must be determined by the location(s) of retinacular injury if repair is to be successful.

If acute operative repair is done, failure to identify any and all locations of disruption may jeopardize the success of the repair. This may explain the relatively low rate of successful repair in the study by Nikku et al. 44 If acute repair is to be done, the entire ligament should be inspected carefully, and all sites of injury should be repaired or reinforced. Preoperative MRI may be of value, 33,54 as may arthroscopy, in determining the nature of the injury (midsubstance tear versus avulsion). Acute midsubstance repairs may be difficult to address surgically because of the small cross section of the medial patellofemoral ligament, whereas an avulsion may be addressed more easily with an anchor.

Garth et al 21 described exploration of the medial retinacular tissues for the purpose of identifying and repairing stretched or torn tissues. To expose the medial patellofemoral ligament, which travels from the medial border of the patella to the adductor tubercle under the distal aspect of the vastus medialis obliquus, the superficial retinaculum first must be reflected posteriorly. This is done through a parapatellar incision extending from the superior pole of the patella to the midaspect of the patellar tendon. The medial patellofemoral ligament is well-observed at this point and may be examined digitally from its patellar insertion to its femoral origin. The deep synovial layer (Layer 3) is dissected carefully off the deep surface of the ligament at this point to aid in inspection and repair. Now isolated in its course, the true nature of medial patellofemoral ligament injury can be inspected fully. It may be seen as an anterior or posterior avulsion, or simply as an attenuation or elongation type of injury. Acutely, hematoma may be appreciated at the site of injury whereas in chronic cases there may be only palpable thinning.

In the report by Garth et al, 24 repair of the medial patellofemoral ligament was dictated by the observed pathologic findings using Number 1 polydioxanone sutures place in a modified Kessler fashion. Tensioning of the soft tissues was judged to be appropriate when lateralization of the patella by passive manipulation was corrected while allowing for full, unrestricted range of motion (ROM) without threatening the repair. The position of the joint and the force applied during passive manipulation were not specified. At this point, Garth et al advocate repair and advancement of the medial patellomeniscal ligament in Layer 3 of the retinaculum.

Fithian and Meier 19 described a technique for medial patellofemoral ligament advancement and repair based on the method reported by Sargent and Teipner. 55 That technique is used in almost all patients with recurrent patellar instability who are treated surgically at the authors’ facilities, whether additional measures such as muscular realignment, trochleoplasty, or lateral release are done, to restore the ligament’s length to its preinjury status.

The approach is through a 10-cm midline incision centered on the patella. The retinaculum is elevated sharply off the medial 5 to 7 mm of the patella. Kocher clamps are applied to the deepest layer of the free edge of the retinaculum. In this way the medial patellofemoral ligament may be observed or palpated through Layer 3 as it runs from the medial epicondyle beneath the distal fibers of the vastus medialis obliquus in a posterior to anterior direction toward the proximal patellar border. A sharp tug on the clamps should result in a firm end point if the medial patellofemoral ligament is competent and anchored firmly. In acute situations or when a medial avulsion has failed to heal to its femoral attachment, this end point may be lost or of poor quality. In that case, a second incision is made over the adductor tubercle and the medial anchor is reestablished before tensioning and repair to the patella.

The repair is completed with the knee flexed to 30° to 40° with the patella manually reduced in the trochlear groove. 19 After the repair the knee is brought passively through the full ROM to assess patellar tracking qualitatively, looking for abrupt or gradual deflections of the patella that might indicate either excessive or insufficient medial tightening. Patellar mobility and Bassett’s checkrein sign 5 are tested to confirm the adequacy of the repair. With a lateral displacing force on the patella, a firm end point or checkrein 5 should be appreciated. Medial and lateral displacement are measured as a 5-lb force is applied to the patella using a Number 5 suture and tension isometer. In healthy knees, a 5-lb force produces 7 to 10 mm of motion in the medial and lateral direction with the knee in 30° to 40° flexion and the muscles relaxed 20,28 (Tables 1–4). The current authors use these data as a rough guide during intraoperative testing of patellar stability. For example, if lateral displacement is more than 10 mm or less than 5 mm, the medial repair is retensioned. However, if medial displacement is less than 5 mm, a lateral release is done on the rationale that the lateral tissues are excessively tight.

Back to Top | Article Outline

Primary Medial Patellofemoral Ligament Repair With Adductor Magnus Augmentation

Avikainen et al 4 described the technique of primary medial patellofemoral ligament exploration and repair with the addition of an adductor magnus tenodesis in acute and chronic cases of patellar instability. Their approach is similar to the approach of Garth et al 21 with an extensile longitudinal incision which is more posteriorly based along the adductor tubercle. Layer 1 is incised and the interval between the vastus medialis obliquus and medial patellofemoral ligament is explored digitally. The medial patellofemoral ligament is cut at its femoral origin for later reinsertion. The adductor magnus then is located at its insertion slightly proximal to the origin of the medial patellofemoral ligament and under the medial intermuscular septum. An 8-cm segment of adductor magnus tendon is harvested with its insertion left intact (Fig 2). The superior medial geniculate artery is identified and protected. The medial patellofemoral ligament with the firm edge of the vastus medialis obliquus is repaired to its epicondylar origin while a medially-displacing force is applied to the patella. The adductor magnus then is turned over the medial patellofemoral ligament and stabilized with absorbable sutures. The retinacular layer then is closed with a running stitch and the tenodesed adductor magnus is checked from 0° to 90° to ensure easy gliding over the femoral epicondyle.

Fig 2.

Fig 2.

Avikainen et al did not do a lateral release. 4 Postoperatively, flexion is limited to 60° with partial weightbearing for 2 weeks. In their series of 14 patients, hemorrhage was observed in every medial patellofemoral ligament at the level of the femoral epicondyle.

Back to Top | Article Outline

Medial Patellofemoral Ligament Reconstruction

In some patients, particularly those with patella alta or trochlear dysplasia, the medial patellofemoral ligament may be structurally deficient. Even if it is otherwise normal, some authors 4,23,41,46 think that a lack of bony constraint can put the ligament at risk for repeated failure if additional measures are not taken to augment or support the native medial tether. Reconstruction of the medial patellofemoral ligament in such cases may be necessary if sufficient collagen is not available to ensure a durable repair. Numerous graft materials have been used including autograft, allograft, and synthetic polyesters. 23,46 Although the use of synthetic fibers for intraarticular ligament reconstruction has fallen out of favor in the United States because of reports of fiber-induced synovial reaction 24 and a high rate of failure, 25,53 their role in the setting of an extraarticular reconstruction remains to be established.

The surgical technique described by Gomes 23 uses three small incisions: one at the lateral aspect of the patella, one at the medial aspect of the patella, and a third incision at the level of medial epicondyle. A 3.2-mm tunnel is drilled from medial to lateral at slightly midway between the proximal and distal poles of the patella with care to avoid penetration of the articular cartilage. A subcutaneous tunnel also is created from the medial patellar incision to the epicondylar incision. A Number 1 steel suture then is passed from lateral to medial through both tunnels and fixed at the femoral epicondyle with a perforated Kirschner wire. The lateral aspect of the suture is fixed to a dynamometer (Smith & Nephew, Andover, MA) and the knee is brought through a full ROM. The femoral point of fixation is altered until dynamometer displacement is less than 5 mm. This indicates a nearly isometric position for the graft. A polyester ligament (Leeds-Keio, Neoligament, Ltd, Leeds, United Kingdom) then is placed through the tunnels. Lateral fixation is made with a knot that prevents migration through the bone tunnel. The knot is countersunk to avoid undue prominence. Gomes then harvests a bone plug at the epicondylar fixation site with a special trephine. This bone washer then is fixed over the ligament with a large-fragment AO screw and washer with the knee in 90° flexion. The knee again is brought through a full ROM and tension in the ligament is adjusted accordingly. Gomes makes a special point in noting that the normal retinaculum is loose in full extension and so should be the graft. A closed lateral release is done in all patients through the lateral incision. The patients are progressed to full weightbearing by the tenth postoperative day with full activities at 4 to 6 months.

Nomura et al 46 reported on their technique of medial patellofemoral ligament reconstruction using the Leeds-Keio polyester ligament (Fig 3). They use a long incision halfway between the patella and the adductor tubercle. A 10-mm wide strip of Layer 1 retinaculum is reflected from anterior to posterior to a level just posterior to the adductor magnus tendon. A rectangular tunnel is created with a 3.5-mm drill bit along the medial patellar border, exiting through the anterior cortex. Its width matches that of the 15-mm artificial ligament. The ligament is fed through the tunnel and temporarily anchored to a stripped region of the femur just distal to the adductor tubercle underneath the elevated strip of retinaculum. They use the integrated double staple system (EDK System BMS KK, Zimmer Division, Tokyo, Japan). The graft initially is tensioned and held with a temporary staple; the knee is flexed to 60° with a tension spacer placed between the ligament and femoral cortex. The patella then is placed manually in the femoral groove and the ligament is tensioned to 0.5-kg force. With the spacer removed, the ligament should create a reasonable checkrein according to Nomura et al, without undue stress on the reconstruction throughout the full ROM. The retinaculum then is repaired back over the ligament. The postoperative protocol of Nomura et al is a bit more accelerated than most, with immediate quadriceps exercises in extension, continuous passive motion on Days 2 to 3, weightbearing as tolerated with a patellar brace on postoperative Day 5, and return to sports at 8 to 12 weeks. They reported good to excellent results using the Crosby and Insall grading system 14 in 26 of 27 patients followed up for a minimum of 4.1 years. Additional subluxation or dislocation occurred in one patient.

Fig 3 A–F.

Fig 3 A–F.

In contrast to the use of a synthetic ligament to reconstruct the medial patellofemoral ligament, Muneta et al 41 prefer to use a double-stranded allograft or autograft placed in the presynovial fatty plane between the Layers 2 and 3 of the medial retinaculum. Their surgical technique begins with a 2.5-cm incision along the medial aspect of the patella. As noted, the interval between the medial patellofemoral ligament (Layer 2) and the synovium (Layer 3) is developed bluntly. A hemostat is passed through this plane down to the femoral epicondyle and a second small incision is made over the tip at the attachment site for the medial patellofemoral ligament. The graft then is passed easily through this tunnel. The patellar tunnel is created with a 4.5-mm drill at the midaspect of the medial patella in the anteroposterior (AP) and proximodistal direction. The drill hole is made parallel to the articular surface to the center of the patella. A 3.2-mm oblique hole then is connected to the tunnel from the anterior surface of the patella. This tunnel configuration theoretically minimizes the stress riser and risk of fracture. The tunnel may need to be enlarged to adjust for a larger graft. Isometry then is checked with a 5-mm cotton tape passing from the anterior patella to the femoral insertion. Muneta et al 41 report that the normal length of the ligament increases with increasing knee flexion. The graft then is passed and fixed on the patellar side with a button and on the femoral side with a spiked staple. The excess length is turned onto itself and sutured to the tendon and surrounding soft tissues. Double staple fixation is difficult because of the limited space on the epicondyle.

Back to Top | Article Outline


More than 100 procedures have been described to address patellar instability with the majority focusing on the dynamic elements (muscles) that affect patellar forces and motion. Attention toward the passive medial restraints is a relatively new development. The proponents of medial patellofemoral ligament repair, reconstruction, or retensioning all share the common goal of restoring the medial checkrein of the patella. Fundamentally, this is similar to other procedures in orthopaedics, which aim to address the essential lesion in cases of excessive mobility.

It now is recognized that excessive lateral patellar mobility is seen after acute patellar dislocation. 20,61 This is attributable to a medial retinacular injury, specifically, avulsion or tearing of the medial patellofemoral ligament. 8,12,18,28 In knees from cadavers, repair of the medial patellofemoral ligament is not only sufficient, but necessary to restore lateral patellar mobility to within a normal range. 28

In identifying and treating medial soft tissue injury in the setting of patellar dislocation, timing may be important. In the largest anatomic study to date, Nomura 45 looked at 18 acute patellar dislocators and 49 chronic dislocators. In seven of the 18 acute patellar dislocators, the medial soft tissue injury represented a femoral avulsion of the medial patellofemoral ligament, whereas 10 of 18 had a rupture or intrasubstance tear of the medial patellofemoral ligament at or adjacent to the adductor tubercle. The chronic dislocators were classified into one of three types of medial patellofemoral ligament injuries: (1) loose femoral attachment (nine of 49); (2) elongated scar posteriorly (29 of 49); and (3) complete absence of the ligament (11 of 49). His findings highlight the various pathologic findings that may be encountered when addressing this lesion. The importance of a thorough intraoperative assessment of the medial structures, especially during acute repairs, cannot be overemphasized.

Sargent and Teipner 55 and Boring and O’Donoghue 7 reported recurrence rates of only 10% after acute repair of the medial soft tissues in combination with lateral release. Sallay et al 54 reported good results with medial patellofemoral ligament repair when the injury occurred posteriorly at the adductor tubercle. Sallay et al did many different procedures in addition to the medial patellofemoral ligament repair, including one adductor magnus augmentation. Although only 58% of 16 patients in the study of Sallay et al returned to their previous level of sports without pain, all were satisfied with their outcomes and no recurrences were observed at minimum 2-year followup. Vainionpaa et al 66 reported 80% good or excellent results in a prospective study of 55 patients having acute medial repair and, in 37 patients with tight lateral retinaculum, lateral release in addition to the medial repair. Vainionpaa et al 66 did not present their criteria for the grading of results, although the frequency of redislocations was reported as 9% at 1 year and 11% at 2 years after surgery. More recently, Ahmad et al 1 reported on eight patients who had acute exploration and repair of the medial patellofemoral ligament with repair of the vastus medialis obliquus where it had torn from the medial patellofemoral ligament and adductor magnus. They reported an average score of Kujala et al 36 of 91.9 at followup. Ahmad et al 1 thought that suturing of the vastus medialis obliquus back to the adductor magnus restoring its normal force vector and medial patellofemoral ligament repair, were critical. Avikainen et al 4 however, reported on 14 patients in whom an 8-cm adductor magnus graft was rotated acutely on its femoral insertion and tenodesed to the medial patellofemoral ligament. Twelve of 14 patients had good results. Only one patient experienced recurrence at 6.9 years after surgery. These results and the results of others suggest that patients who have an acute medial repair do better than those treated nonoperatively, particularly regarding redislocation rates. 4,9,11,21,29,54,66

However, in the only prospective, randomized study in the English language literature comparing acute repair versus nonoperative treatment, Nikku et al 44 found no statistically significant improvement in the risk of recurrent instability among patients who had surgery compared with those who had conservative treatment after the initial event. Nikku et al cited the work of Sallay et al, 54 but few of the procedures described by Nikku et al followed the recommendations of Sallay et al on the technique of repair. The medial retinaculum was repaired, duplicated, or augmented in 63 of 70 patients but the medial patellofemoral ligament was not specifically addressed. Fifty-four patients had a lateral release without clear indications. Multiple surgeons were involved with patients having “individually adjusted procedures”, according to Nikku et al. Although the findings of Nikku et al were not encouraging for acute repair, it is clear that the medial patellofemoral ligament was neither explored nor repaired in their group of patients. In dealing with the medial soft tissues, if the site of the injury is not addressed specifically, a medial tightening procedure may result in additional stretching of the injured medial patellofemoral ligament at the injured location.

Unless indicated for open reduction and internal fixation of an osteochondral fragment, the senior author (DCF) prefers to wait until primary healing of the medial patellofemoral ligament has occurred. Similar to the medial collateral ligament of the knee, the medial patellofemoral ligament is extrasynovial. Unless a full-thickness capsular injury occurs, the medial patellofemoral ligament tear may heal in a fashion similar to the medial collateral ligament, so that the ligament may regain much of its strength when healing is complete. If that is the case, then repair or advancement of the healed ligament either anteriorly or posteriorly might be expected to restore motion limits of the patella to their preinjury state. Garth et al 21 reported on 20 patients with chronic patellar instability. All 20 patients had an exploration and repair, retensioning, or reattachment at the site of medial patellofemoral ligament injury. Ten patients had injuries anteriorly at the patellar border; seven had obvious tears of the medial patellofemoral ligament whereas three had palpable thinning and incompetence of the ligament. Similar findings were seen in the remaining 10 patients: attenuation of the ligament or avulsion of the adductor tubercle. At a minimum 24 months followup, 18 of 20 patients had good or excellent subjective results using the scale of Turba et al, 64 without any recurrent instability. The effect of the medial patellofemoral ligament repair is difficult to assess in the patient sample of Garth et al because of the frequency of procedures done in addition to the medial repair, such as tibial tubercle transfers and lateral releases.

There may exist a subset of individuals with knees at risk for failure after repair or reefing because they lack adequate medial ligamentous substance to withstand the tension under which normal activity would place the retinaculum. This may be from an inherent anatomic predisposition such as medial patellofemoral ligament and/or trochlear dysplasia, or because the ligament failed to heal sufficiently after a previous dislocation. In these patients, a reconstructive procedure may very well be warranted to restore the medial tether of the patella. Gomes, 23 Nomura et al, 46 and Muneta et al 41 prefer to approach patellar instability and the torn medial patellofemoral ligament from this standpoint. Their methods all are similar with the exception that Gomes places the ligament in the subcutaneous tissues whereas the others use the planes between Layers 1 and 3. Muneta et al opt for soft tissue grafts whereas Gomes and Nomura et al use synthetic polyesters. Gomes reported 96% good or excellent results in 23 patients who did not have significant chondral lesions, 83% of whom had an excellent outcome. Patient satisfaction decreased remarkably in the presence of what he described as a “severe” patellar lesion (three of seven good or excellent). Most recently, Nomura et al 46 reported on 27 reconstructions using the Leeds-Keio ligament in 24 patients (three patients had bilateral procedures) at an average 5.9 years followup. Fifty-five percent of patients had excellent results and 41% of patients had good results. One patient had one episode of recurrent instability, which resulted in a poor outcome. The results of Nomura et al are even more remarkable considering that 21 of 27 patients had varying degrees of chondral injury and no other procedures were done with the exception of 10 lateral releases (37%).

Despite the presence of predisposing factors such as dysplasia or generalized ligamentous laxity, medial patellofemoral ligament injury often occurs at the time of patellar dislocation. Medial patellofemoral ligament injury with resultant laxity is a common feature among patients who experience recurrent patellar instability after patellar dislocation. Whatever repair or reconstructive procedure is done to prevent ongoing patellar instability, the injured ligament should be repaired, retensioned, or reconstructed to help ensure a successful outcome. The best method for achieving this goal has not been established. Additional research is needed to determine the optimal approach.

Back to Top | Article Outline


The authors thank the members of the International Patellofemoral Study Group, whose creative energy and generous collaboration inspires us, and Liz W. Paxton, MA for helpful editorial discussions.

Back to Top | Article Outline


1. Ahmad CS, Stein BE, Matuz D, Henry JH: Immediate surgical repair of the medial patellar stabilizers for acute patellar dislocation: A review of eight cases. Am J Sports Med 28:804–810, 2000.
2. Arnbjornsson A, Egund N, Rydling O, Stockerup R, Ryd L: The natural history of recurrent dislocation of the patella: Long-term results of conservative and operative treatment. J Bone Joint Surg 74B:140–142, 1992.
3. Atkin DM, Fithian DC, Marangi KS, Stone ML, Dobson BE, Mendelsohn C: Characteristics of patients with primary acute lateral patellar dislocation and their recovery within the first 6 months of injury. Am J Sports Med 28:472–479, 2000.
4. Avikainen VJ, Nikku RK, Seppanen-Lehmonen TK: Adductor magnus tenodesis for patellar dislocation: Technique and preliminary results. Clin Orthop 297:12–16, 1993.
5. Bassett FH: Acute Dislocation of the Patella, Osteochondral Fractures, and Injuries to the Extensor Mechanism of the Knee. In Burke E (ed). American Academy of Orthopaedic Surgeons Instructional Course Lectures. St Louis, CV Mosby, Inc 40–49, 1976.
6. Blauth W, Mann M: [Medial- and simultaneous anterior-transfer of the tibial tuberosity (authors’ translation)]. Z Orthop Ihre Grenzgeb 115:252–255, 1977.
7. Boring TH, O’Donoghue DH: Acute patellar dislocation: Results of immediate surgical repair. Clin Orthop 136:182–185, 1978.
8. Burks RT, Desio SM, Bachus KN, Tyson L, Springer K: Biomechanical evaluation of lateral patellar dislocations. Am J Knee Surg 11:24–31, 1998.
9. Cash JD, Hughston JC: Treatment of acute patellar dislocation. Am J Sports Med 16:244–249, 1988.
10. Cerullo G, Puddu G, Conteduca F, Ferretti A, Mariani PP: Evaluation of the results of extensor mechanism reconstruction. Am J Sports Med 16:93–96, 1988.
11. Cofield RH, Bryan RS: Acute dislocation of the patella: Results of conservative treatment. J Trauma 17:526–531, 1977.
12. Conlan T, Garth Jr WP, Lemons JE: Evaluation of the medial soft-tissue restraints of the extensor mechanism of the knee. J Bone Joint Surg 75A:682–693, 1993.
13. Cox JS: Evaluation of the Roux-Elmslie-Trillat procedure for knee extensor realignment. Am J Sports Med 10:303–310, 1982.
14. Crosby EB, Insall J: Recurrent dislocation of the patella: Relation of treatment to osteoarthritis. J Bone Joint Surg 58A:9–13, 1976.
15. Daniel DM: Diagnosis of a Ligament Injury. In Daniel DM, Akeson WA, O’Connor JJ (eds). Knee Ligaments: Structure, Function, Injury, and Repair. New York, Raven Press 3–10, 1990.
16. Daniel DM: Assessing the limits of knee motion. Am J Sports Med 19:139–147, 1991.
17. Dejour H, Walch G, Neyret P, Adeleine P: [Dysplasia of the femoral trochlea]. Rev Chir Orthop Reparatrice Appar Mot 76:45–54, 1990.
18. Desio SM, Burks RT, Bachus KN: Soft tissue restraints to lateral patellar translation in the human knee. Am J Sports Med 26:59–65, 1998.
19. Fithian DC, Meier SW: The case for advancement and repair of the medial patellofemoral ligament in patients with recurrent patellar instability. Op Tech Sports Med 7:81–89, 1999.
20. Fithian DC, Mishra DK, Balen PF, Stone ML, Daniel DM: Instrumented measurement of patellar mobility. Am J Sports Med 23:607–615, 1995.
21. Garth Jr WP, DiChristina DG, Holt G: Delayed proximal repair and distal realignment after patellar dislocation. Clin Orthop 377:132–144, 2000.
22. Goldthwait J: Slipping or recurrent dislocation of the patella, with the report of eleven cases. Boston Med Surg J 150:169–174, 1904.
23. Gomes JL: Medial patellofemoral ligament reconstruction for recurrent dislocation of the patella: A preliminary report. Arthroscopy 8:335–340, 1992.
24. Greis PE, Georgescu HI, Fu FH, Evans CH: Particle-induced synthesis of collagenase by synovial fibroblasts: An immunocytochemical study. J Orthop Res 12:286–293, 1994.
25. Grontvedt T, Engebretsen L, Benum P, Fasting O, Molster A, Strand T: A prospective, randomized study of three operations for acute rupture of the anterior cruciate ligament: Five-year follow-up of one hundred and thirty-one patients. J Bone Joint Surg 78A:159–168, 1996.
26. Hamilton WG, Thompson FM, Snow SW: The modified Brostrom procedure for lateral ankle instability [published erratum appears in Foot Ankle 14:180, 1993]. Foot Ankle 14:1–7, 1993.
27. Hauser EDW: Total tendon transplant for slipping patella: A new operation for recurrent dislocation of the patella. Surg Gynecol Obstet 66:199–314, 1938.
28. Hautamaa PV, Fithian DC, Kaufman KR, Daniel DM, Pohlmeyer AM: Medial soft tissue restraints in lateral patellar instability and repair. Clin Orthop 349:174–182, 1998.
29. Hawkins RJ, Bell RH, Anisette G: Acute patellar dislocations: The natural history. Am J Sports Med 14:117–120, 1986.
30. Hughston JC: Subluxation of the patella. J Bone Joint Surg 50A:1003–1026, 1968.
31. Hughston JC: Reconstruction of the extensor mechanism for subluxating patella. J Sports Med 1:6–13, 1972.
32. Hungerford DS, Barry M: Biomechanics of the patellofemoral joint. Clin Orthop 144:9–15, 1979.
33. Kirsch MD, Fitzgerald SW, Friedman H, Rogers LF: Transient lateral patellar dislocation: Diagnosis with MR imaging. Am J Roentgenol 161:109–113, 1993.
34. Kolowich PA, Paulos LE, Rosenberg TD, Farnsworth S: Lateral release of the patella: Indications and contraindications. Am J Sports Med 18:359–365, 1990.
35. Krogius A: Zur operativen behandlung der habituellen luxation der kniescheibe. Zbl Chir 31:254–257, 1904.
36. Kujala UM, Jaakkola LH, Koskinen SK, Taimela S, Hurme M, Nelimarkka O: Scoring of patellofemoral disorders. Arthroscopy 9:159–163, 1993.
37. Kujala UM, Osterman K, Kvist M, Aalto T, Friberg O: Factors predisposing to patellar chondropathy and patellar apicitis in athletes. Int Orthop 10:195–200, 1986.
38. Larsen E, Lauridsen F: Conservative treatment of patellar dislocations: Influence of evident factors on the tendency to redislocation and the therapeutic result. Clin Orthop 171:131–136, 1982.
39. Maldague B, Malghem J: [Significance of the radiograph of the knee profile in the detection of patellar instability: Preliminary report]. Rev Chir Orthop Reparatrice Appar Mot 71:5–13, 1985.
40. Merchant AC, Mercer RL, Jacobsen RH, Cool CR: Roentgenographic analysis of patellofemoral congruence. J Bone Joint Surg 56A:1391–1396, 1974.
41. Muneta T, Sekiya I, Tsuchiya M, Shinomiya K: A technique for reconstruction of the medial patellofemoral ligament. Clin Orthop 359:151–155, 1999.
42. Murphy PG, Frank CB, Hart DA: Characterization of the plasminogen activators and plasminogen activator inhibitors expressed by cells isolated from rabbit ligament and synovial tissues: Evidence for unique cell populations. Exp Cell Res 205:16–24, 1993.
43. Murray TF, Dupont JY, Fulkerson JP: Axial and lateral radiographs in evaluating patellofemoral malalignment. Am J Sports Med 27:580–584, 1999.
44. Nikku R, Nietosvaara Y, Kallio PE, Aalto K, Michelsson JE: Operative versus closed treatment of primary dislocation of the patella: Similar 2-year results in 125 randomized patients. Acta Orthop Scand 68:419–423, 1997.
45. Nomura E: Classification of lesions of the medial patello-femoral ligament in patellar dislocation. Int Orthop 23:260–263, 1999.
46. Nomura E, Horiuchi Y, Kihara M: A mid-term follow-up of medial patellofemoral ligament reconstruction using an artificial ligament for recurrent patellar dislocation. Knee 7:211–215, 2000.
47. Nove-Josserand L, Dejour D: [Quadriceps dysplasia and patellar tilt in objective patellar instability]. Rev Chir Orthop Reparatrice Appar Mot 81:497–504, 1995.
48. Noyes FR, Grood ES, Torzilli PA: Current concepts review: The definitions of terms for motion and position of the knee and injuries of the ligaments. J Bone Joint Surg 71A:465–472, 1989.
49. Reider B, Marshall JL, Koslin B, Ring B, Girgis FG: The anterior aspect of the knee joint. J Bone Joint Surg 63A:351–356, 1981.
50. Reider B, Marshall JL, Warren RF: Clinical characteristics of patellar disorders in young athletes. Am J Sports Med 9:270–274, 1981.
51. Roux C: Luxation habituelle de la rotule. Rev Chir 8:682–689, 1888.
52. Runow A: The dislocating patella: Etiology and prognosis in relation to generalized joint laxity and anatomy of the patellar articulation. Acta Orthop Scand 201(Suppl):1–53, 1983.
53. Safran MR, Harner CD: Technical considerations of revision anterior cruciate ligament surgery. Clin Orthop 325:50–64, 1996.
54. Sallay PI, Poggi J, Speer KP, Garrett WE: Acute dislocation of the patella: A correlative pathoanatomic study. Am J Sports Med 24:52–60, 1996.
55. Sargent JR, Teipner WA: Medial patellar retinacular repair for acute and recurrent dislocation of the patella: A preliminary report. J Bone Joint Surg 53A:386–387, 1971.
56. Shoemaker SC, Adams D, Daniel DM, Woo SL: Quadriceps/anterior cruciate graft interaction: An in vitro study of joint kinematics and anterior cruciate ligament graft tension. Clin Orthop 294:379–390, 1993.
57. Shoemaker SC, Daniel DM: The Limits of Knee Motion: In Vitro Studies. In Daniel DM, Akeson WH, O’Connor JJ (eds). Knee Ligaments: Structure, Function, Injury, and Repair. New York, Raven Press 153–161, 1990.
58. Skalley TC, Terry GC, Teitge RA: The quantitative measurement of normal passive medial and lateral patellar motion limits. Am J Sports Med 21:728–732, 1993.
59. Spritzer CE, Courneya DL, Burk Jr DL, Garrett WE, Strong JA: Medial retinacular complex injury in acute patellar dislocation: MR findings and surgical implications. AJR Am J Roentgenol 168:117–122, 1997.
60. Stanitski CL: Articular hypermobility and chondral injury in patients with acute patellar dislocation. Am J Sports Med 23:146–150, 1995.
61. Teitge RA, Faerber WW, Des Madryl P, Matelic TM: Stress radiographs of the patellofemoral joint. J Bone Joint Surg 78A:193–203, 1996.
62. Terry GC: The anatomy of the extensor mechanism. Clin Sports Med 8:163–177, 1989.
63. Trillat A, DeJour H, Couette A: Diagnostic et traitement des subluxations recidivantes de la rotule. Rev Chir Orthop (Paris) 50:813–824, 1964.
64. Turba J, Walsh W, McLeod W: Long-term results of extensor mechanism reconstruction: A standard for evaluation. Am J Sports Med 7:91–94, 1979.
65. Vainionpaa S, Laasonen E, Patiala H, Rusanen M, Rokkannen P: Acute dislocation of the patella: Clinical, radiographic and operative findings in 64 consecutive cases. Acta Orthop Scand 57:331–333, 1986.
66. Vainionpaa S, Laasonen E, Silvennoinen T, Vasenius J, Rokkanen P: Acute dislocation of the patella: A prospective review of operative treatment. J Bone Joint Surg 72B:366–369, 1990.
67. Warren LA, Marshall JL, Girgis F: The prime static stabilizer of the medical side of the knee. J Bone Joint Surg 56A:665–674, 1974.

Section Description

Kurt P. Spindler, MD; and Edward M. Wojtys, MD—Guest Editors

© 2002 Lippincott Williams & Wilkins, Inc.