Patellofemoral instability after total knee arthroplasty


Patellofemoral instability after total knee arthroplasty (TKA) is a serious complication that has significant impact on knee function. The underlying causes are usually related to the surgical technique. Most often, patellar instability results from component malposition, limb malalignment, improper patellar preparation, or soft-tissue imbalance. Understanding the root causes helps to prevent these complications as well as to manage them successfully.

Factors determining patellofemoral stability

Patellofemoral (PF) instability after TKA has been reported in up to 20% of TKAs [1]. Most often, it is caused by technical errors during surgery [1, 2]. Given the complexity of TKA biomechanics, many technical parameters are susceptible to error. Therefore, in most cases, PF instability cannot be traced back to a single cause. More likely, there are multiple contributors.

As a general rule, any manipulation of the normal anatomical and kinematic relationships of knee structures that increases tension in the lateral retinaculum or increases the quadriceps angle, or Q-angle (Figure 1), will produce an abnormal, laterally-directed muscle vector and thus cause lateral maltracking of the patella, instability of the PF joint, and more serious patellar complications if left untreated [3, 4].

Figure 1. The Q-angle is the angle between a line from the ASIS to the midpoint of the patella (dotted line) and the line from the tibial tubercle to the midpoint of the patella. By moving the patella laterally (b), the CQ-angle decreases to zero, whereas the RFQ-angle becomes negative. When the patella is moved medially (c) both CQ and RFQ become larger (more positive). Abbreviations: ASIS: anterior superior iliac spine; CQ: clinical Q-angle; PA: patella; RF: rectus femoris; RFQ: RF-Q-angle; TT: tibial tubercle.

Redrawn from source: Freedman BR, Brindle TJ, Sheehan FT. Re-evaluating the functional implications of the Q-angle and its relationship to in-vivo patellofemoral kinematics. Clin Biomech (Bristol, Avon). 2014;29(10):1139–1145.

Guillermo Bonilla, clinical professor for orthopedic surgery at the Universidad de los Andes in Bogota, Colombia, emphasizes: “In order to adequately treat problems arising from patellar instability, it is essential to understand the underlying biomechanical mechanisms.” So, let us consider and analyze the major factors that may lead to patellar instability.

Guillermo Bonilla, MD

Hospital Universitario Fundación Santa Fe de Bogotá
Universidad de los Andes Bogotá, Colombia

First and foremost, excessive valgus alignment is an important risk factor, since it leads to a mismatch of the trochlear groove and the extensor vector [4]. This encumbers proper patellar tracking and may cause the patella to tilt, subluxate, or even dislocate. A similar mechanism takes effect when a normal Q-angle is not restored. Especially patients with severe preoperative valgus or external rotational deformity, preoperative maltracking [5], and loss of bone stock in the distal lateral condyle are at risk [3]. In patients with pronounced preoperative valgus, the main culprit for this predisposition is usually the retraction of the lateral retinaculum.

Beyond the overall leg alignment, individual component positioning is the most important contributor. Internal rotation of the femoral or the tibial component [5-10] as well as medialization of the femoral component and incorrect placement of the patellar component [10, 11] appear the most obvious.

An internally rotated femoral component shifts the trochlear groove medially, thus increasing the distance to the patella, which tracks laterally relative to the femur. Through the tension exerted by the lateral retinaculum, the patella is pulled sideways. This may lead to patellar tilt, subluxation, or even dislocation [4].

On the other hand, an internally rotated tibial component causes the tibia to rotate externally during knee flexion (Figure 2). This drives the tibial tubercle laterally, which increases the Q-angle and thus leads to lateral tracking. Depending on the severity, this may again lead to patellar tilt, subluxation, or dislocation.

Figure 2. Internal rotation of the tibial component forces the tibia into external rotation during knee flexion, increasing the Q-angle and leading to lateral patellar tracking and subluxation. Diagram on the right showing that rotating the implant (sagittal axis dotted line) internally (medially) relative to the tibia (sagittal axis continuous line) displaces the tibial tubercle laterally and consequently the patella as well. Adapted from: Malo M, Vince KG. The unstable patella after total knee arthroplasty: etiology, prevention, and management. J Am Acad Orthop Surg. 2003;11(5):364–371.

Did you know?

When rotational platforms were introduced, there were high hopes that through the inherent self-alignment, a certain degree of malrotation could be corrected for. Unfortunately, this was not the case. Biomechanical studies analyzing the effect of rotational platforms on patellar tracking in the presence of malrotated femoral components did not find a significant compensatory effect [6, 7], and neither did a large randomized clinical study [12].

Placement of the prosthetic patella also plays a decisive role in determining PF stability. A too far laterally placed patellar button will increase the tension in the lateral retinaculum. This in turn can displace the center of the patella medially and thus lead to a lateral pull and subsequent lateral tracking with the known consequences [3].

A similar effect is seen with medialization of the femoral component [4]. The medialized trochlear groove results in increased contact stresses between the lateral flange of the femoral component and the lateral border of the patellar component. This may create a fulcrum, and in this fashion lead to patellar subluxation [3].

Finally, depending on the preoperative condition and the surgical technique used, soft-tissue imbalance may persist or be created during surgery. This could manifest as medial retinacular insufficiency, vastus medialis oblique weakness, quadriceps contracture, or iliotibial band tightness [9]. The medial patellar restraining structures may be damaged in the course of a medial parapatellar approach [13]. Regardless of the manifestation or root cause, any significant soft-tissue imbalance may have a detrimental effect on patellar stability.

Additional risk factors have been discussed in various publications but are less important. These include overstuffing of the PF joint [3, 4, 14], excessive flexion of the femoral component [15, 16], and the distance from the tibial tubercle to the trochlear groove (TT-TG distance) [17].

As laid out above, there are a multitude of factors that affect patellar stability. Even perfect component positioning does not guarantee perfect patellar tracking and stability. Thoroughly checking the correct patellar path and soft-tissue tension intraoperatively is therefore imperative. Various techniques are available. A commonly used method is the “no thumbs technique” [1]. This test assesses patellar tracking by reducing the patella and taking the knee through the full range of motion without closing the medial arthrotomy or applying any medial force with the thumb to keep the patella in place [3]. If during flexion, the patella does not tilt, subluxate or dislocate, the extensor mechanism can be considered as balanced [18]. A different technique follows the “kissing rule”, which purports that if in maximal flexion, the medial surface of the patella does not make contact with the medial condyle of the femoral component, this is an indication for insufficient PF congruence [19]. Alternatively, lateral retinaculum tightness may be evaluated with the knee in extension by manually trying to subluxate the patella medially with the trial components in place. If the patella crosses the medial prosthetic condyle for half of its diameter, the retinaculum is considered not to be too tight [19].

In summary, ensuring a proper coronal and rotational alignment along with correct patellar thickness and positioning will improve PF tracking, thus reducing the risk of patellar instability.

Diagnostic workup in PF instability

The diagnostic workup has two main objectives. The first objective is to confirm the diagnosis of instability. While dislocation is an obvious condition, more subtle instability requires further investigation of symptoms and signs. The second objective is to understand the causes of PF instability. We will now look at both the clinical and the radiological diagnostic processes.

Patients with PF instability typically present with anterior knee pain upon engaging in stressful activities that involve flexing the knee. In severe cases, they may not be able to fully flex.

For a correct diagnosis, it is important to understand the mechanisms that lead to these signs and symptoms. From a biomechanical viewpoint, the position of the patella in the trochlea depends on several factors: the degree of knee flexion, the position of the foot on the ground, and the relative pull of the different components of the quadriceps muscle. Therefore, the path of the patella from full extension to full flexion is not fixed [20]. During knee flexion, the PF contact point on the patellar surface moves from distal to proximal. This results in a proportional decrease in the patellar ligament/quadriceps tendon stress ratio, which may give rise to the pain patients then experience when climbing stairs or rising from a chair. [21]. The exact location of the pain varies. It can be located in the peripatellar region, the lateral, or the medial aspect of the knee. Also, its intensity varies, from minor to disabling. Its onset usually indicates the root cause. If onset is sudden, after a long asymptomatic period, the cause is likely related to failure of the extensor mechanism or a prosthetic component. If it is persistent, with a long history, it is likely to be related to the surgical technique. Warschawski et al postulated that “in asymptomatic patients with component malrotation, the onset of pain and instability may present in a delayed fashion through two mechanisms. First, component malrotation may, over time, lead to attrition of the ligamentous checkrein mechanisms eventually leading to PF dislocation. Second, failure of PF soft tissue supports may be accelerated or caused by an inciting traumatic event leading to PF dislocation” [9].

In addition to pain, typical complaints are knee stiffness, the inability to perform full flexion, the sensation of the knee slipping out of place, giving way, a buckling sensation, or a subluxation sensation [4, 22].

The clinical workup

As mentioned in the previous section, the patient's history is important because timing of the first occurrence of symptoms and their evolution can give important hints about the etiology.

The physical examination should check for varus or valgus leg alignment and inspect the continuity of the extensor mechanism by palpation throughout the full passive and active range of motion. Patients may identify areas of localized tenderness. Palpating the patella through the range of motion can detect subluxation, dislocation, excess patella mobility, or lateral retinaculum tightness. Being alert for patella alta is important because this condition could place the patella higher in the trochlear groove, where it is shallower and thus more prone to subluxation [23, 24].

Did you know?

Compared to the early TKA designs, the contemporary femoral components have more anatomical trochlear geometries, ie, with steeper sulcus angles. Nonetheless, compared with what is considered a normal trochlear groove, they are still somewhat dysplastic [23].

Manual maneuvers, like attempting to subluxate the patella laterally during active flexion, are also advised. Note that there are also patients with asymptomatic lateral subluxation, ie, with no objective findings except some vague medial knee pain. In these patients, the radiological workup is even more important [22]. After all, maltracking and its inherent abnormal articulation are likely to enhance polyethylene wear which in turn may lead to premature prosthetic loosening.

The radiological workup

The radiological workup needs to assess the structural parameters that lead to patellar instability as laid out in the previous sections. This means the prosthetic components need to be evaluated with regard to their position in all three axes. Beyond that, the Q-angle deserves special attention.

The basic radiographic workup consists of plain x-rays in lateral and in axial (Figure 3), eg, Merchant view.

Figure 3. Patellar dislocation in a knee arthroplasty patient in axial view. Image courtesy of Guillermo Bonilla (MD).

The lateral view enables determination of the patellar thickness, superior or inferior position, and fixation quality, along with positioning of the other prosthetic components. The axial view depicts patellar rotation and alignment in the trochlear groove, ie, potential tilting in the sulcus, as well as the degree of subluxation or dislocation. It also shows the symmetry of the patellar cut and the thickness of the prosthetic components. This information can then be compared to the contralateral knee [22]. Additional AP images of both knees may serve to determine whether the joint line is elevated [1]. This can be important if revision arthroplasty is planned, because restoring the joint line with the contralateral knee as a reference correlates with a good functional outcome [25]. Further essential information can be gleaned from computed tomography (CT). Computed tomographic and magnetic resonance imaging (MRI) have been used alongside computer algorithms to calculate the accurate joint line. Computed tomography has been found to be the most accurate method to assess prosthetic component positioning [26]. Motsis et al describe the method to determine rotational alignment using four scans: “The medial and lateral epicondyles, the tibial plateau immediately below the tibial base plate, the tibial tubercle, and through the tibial insert. The femoral component’s rotation is determined by measuring the angle formed by the line drawn through the medial and lateral epicondyles and the line connecting the posterior flanges of the implant. Tibial component rotation is determined by superimposing the geometric center of the proximal tibia onto the image with the tibial tubercle. The tibial tubercle axis (the line drawn to the highest point of the tubercle) is then placed on the image with the tibial insert with a line drawn perpendicular to the posterior surface of the tibial insert. The normal extent of tibial rotation is 18°” [22]. Even though CT is still the gold standard to evaluate component positioning, recently other authors have described techniques that utilize plain x-rays [27, 28]. These techniques are based on determining the angle between the epicondylar axis and the posterior condylar axis on images taken in knee flexion.

Radiographically suspected component malrotation can additionally be confirmed intraoperatively. This is done by determining appropriate landmarks such as the transepicondylar axis and the femoral anterior tangent line [29, 30].

Depending on the clinical symptoms, ultrasound examination may add relevant information about the morphological integrity of soft-tissue structures around the patella, such as the quadricipital and patellar tendons [31].

Management algorithm

The different causes for patellar instability after TKA are diverse, and so are the corresponding treatment modalities. Most importantly, the chosen treatment modality needs to address the underlying failure mechanism. As this may be a combination of multiple factors, the treatment needs to be tailored to the individual patient's needs.

Bearing in mind that failure typically follows on from a structural problem, most patients ultimately require surgery [1, 4, 22].

Nevertheless, conservative treatment should be attempted before suggesting reoperation. This may include quadriceps exercises, bracing, and avoiding activities that aggravate instability. It has been suggested, that with time, scarring of the retinacular tissues may lead to resolution of symptoms [22].

Guillermo Bonilla points out: “Even if conservative treatment does not achieve the goal to prevent surgery, preoperative strengthening of the quadriceps and, if required, management of retractions, are important aspects of prehabilitation. They will have a significant impact on the postoperative recovery and thus should be included in any treatment strategy.”

Once the decision to proceed with surgery has been made, the first step is to determine whether a revision of the prosthetic components is warranted.

Guillermo Bonilla explains: “When PF instability has been determined to result from malalignment of the prosthetic components, prosthetic revision is usually mandatory. Preoperative radiological diagnosis and intraoperative assessment of the component position in relation to bony landmarks will support the surgical decision. However, clinicians should keep in mind that even after prosthetic components have been revised, soft-tissue procedures could still be needed to correct tissue retractions or laxity.

Prosthetic malalignment may be present in any of the axes or when the patellar button is placed too far lateral. That said, in some cases of coronal malalignment in the presence of a well-fixed prosthesis, closed wedge distal femoral osteotomy (CWDFO) in association with medial PF ligament reconstruction may also be a feasible option [10].

With well aligned prostheses, similar surgical procedures as in native knees can be performed [8]. Management may thus include an initial lateral retinacular release, followed by more extensive procedures such as proximal or distal realignment. The latter, however, has been associated with an increased risk of patellar tendon ruptures, nonunion of the tibial tuberosity osteotomy, and wound complications [1].

The lateral retinacular release may be performed via an intra- or extraarticular approach. In doing so, care should be taken to preserve the lateral superior genicular artery, because it is the main blood supply to the patella [19].

Since soft-tissue imbalance is often caused by insufficiency of the medial restraining structures, such as the medial PF ligament (MPFL) or the vastus medialis, these structures may require reconstruction. A range of tendon reconstruction techniques exist using, for instance, the gracilis, semitendinosus, or quadricipital tendons [13, 32, 33].

If an unphysiological Q-angle was identified as the main cause of the instability, distal realignment may be considered. The procedure can be used alone or in combination with MPFL reconstruction [8]. Whiteside and Ohl developed a technique to improve stability and enhance union of the tibial tuberosity fragment. This technique consists of leaving a bony bridge under the tibial tray. A long osteoperiosteal segment including the tibial tubercle and upper tibial crest is used, and the lateral muscular attachments are left intact to this bone fragment [34]. Although this technique was originally described as an alternate approach during knee revision surgery, it could also be suitable for anterior tubercle realignment after PF instability.

Outcome and conclusion

A wide range of procedures has been performed to address PF instability. Since the etiology is often multifactorial, several procedures may be used in combination. When used wisely, ie, when the underlying cause is identified and addressed accordingly, they have generally led to good clinical results.

Up to now, no comprehensive management algorithms for patellar instability have been published. Given the high etiological variation and the lack of large patient series, this is not a big surprise. Consequently, we lack the clinical evidence to propose a well-founded treatment algorithm. However, what we can provide is an overview of the treatment strategies that have been reported in the literature so far, albeit without claiming to provide sufficient evidence for clinical decision-making.

The broad spectrum of different treatment modalities appears to provide satisfactory outcomes. The greatest challenge is to correctly identify the causes for the instability. Once this is achieved, the foundation is laid for successful treatment.

Contributing experts

This series of articles was created with the support of the following specialists (in alphabetical order):

Guillermo Bonilla, MD

Hospital Universitario Fundación Santa Fe de Bogotá
Universidad de los Andes Bogotá, Colombia

Clemens Gwinner, MD

Center for Musculoskeletal Surgery (CMSC),
Berlin, Germany

Yixin Zhou, MD, PhD

Department of Joint Surgery
Beijing Jishuitan Hospital
The Fourth Clinical College of Peking University
Beijing, China

This issue was  written by Elke Rometsch, AO Innovation Translation Center, Clinical Science, Switzerland.

Additional Resources

Additional AO resources on this topic

Access videos, tools, and other assets to learn more about this topic.


  1. Assiotis A, To K, Morgan-Jones R, et al. Patellar complications following total knee arthroplasty: a review of the current literature. Eur J Orthop Surg Traumatol. 2019 Dec;29(8):1605–1615.
  2. Lee GC, Cushner FD, Scuderi GR, et al. Optimizing patellofemoral tracking during total knee arthroplasty. J Knee Surg. 2004 Jul;17(3):144–149; discussion 149–150.
  3. Eisenhuth SA, Saleh KJ, Cui Q, et al. Patellofemoral instability after total knee arthroplasty. Clin Orthop Relat Res. 2006 May;446:149–160.
  4. Malo M, Vince KG. The unstable patella after total knee arthroplasty: etiology, prevention, and management. J Am Acad Orthop Surg. 2003 Sep-Oct;11(5):364–371.
  5. Wheeless CR. Wheeless' Textbook of Orthopaedics Data Trace; 2020 [cited 2021 January].
  6. Colwell CW, Jr, Chen PC, D'Lima D. Extensor malalignment arising from femoral component malrotation in knee arthroplasty: effect of rotating-bearing. Clin Biomech (Bristol, Avon). 2011 Jan;26(1):52–57.
  7. Kessler O, Patil S, Colwell CW, Jr, et al. The effect of femoral component malrotation on patellar biomechanics. J Biomech. 2008 Dec 5;41(16):3332–3339.
  8. Putman S, Boureau F, Girard J, et al. Patellar complications after total knee arthroplasty. Orthop Traumatol Surg Res. 2019 Feb;105(1s):S43–s51.
  9. Warschawski Y, Garceau S, Frenkel Rutenberg T, et al. Revision total knee arthroplasty for patellar dislocation in patients with malrotated TKA components. Arch Orthop Trauma Surg. 2020 Jun;140(6):777–783.
  10. Saito H, Saito K, Shimada Y, et al. Successful treatment of a habitual patellar dislocation after a total knee arthroplasty with a closing-wedge distal femoral varus osteotomy and medial patello-femoral ligament reconstruction. J Exp Orthop. 2020 Sep 1;7(1):63.
  11. Assi C, Kheir N, Samaha C, et al. Optimizing patellar positioning during total knee arthroplasty: an anatomical and clinical study. Int Orthop. 2017 Dec;41(12):2509–2515.
  12. Pagnano MW, Trousdale RT, Stuart MJ, et al. Rotating platform knees did not improve patellar tracking: a prospective, randomized study of 240 primary total knee arthroplasties. Clin Orthop Relat Res. 2004 Nov(428):221–227.
  13. Lamotte A, Neri T, Kawaye A, et al. Medial patellofemoral ligament reconstruction for patellar instability following total knee arthroplasty: A review of 6 cases. Orthop Traumatol Surg Res. 2016 Sep;102(5):607–610.
  14. Bracey DN, Brown ML, Beard HR, et al. Effects of patellofemoral overstuffing on knee flexion and patellar kinematics following total knee arthroplasty: a cadaveric study. Int Orthop. 2015 Sep;39(9):1715–1722.
  15. Keshmiri A, Maderbacher G, Baier C, et al. The influence of component alignment on patellar kinematics in total knee arthroplasty. Acta Orthop. 2015;86(4):444–450.
  16. Nedopil AJ, Howell SM, Hull ML. What clinical characteristics and radiographic parameters are associated with patellofemoral instability after kinematically aligned total knee arthroplasty? Int Orthop. 2017 Feb;41(2):283–291.
  17. Nakamura S, Shima K, Kuriyama S, et al. Tibial Tubercle-Trochlear Groove Distance Influences Patellar Tilt After Total Knee Arthroplasty. J Arthroplasty. 2019 Dec;34(12):3080–3087.
  18. Scott RD. Prosthetic replacement of the patellofemoral joint. Orthop Clin North Am. 1979 Jan;10(1):129–137.
  19. Gasparini G, Familiari F, Ranuccio F. Patellar malalignment treatment in total knee arthroplasty. Joints. 2013 Mar;1(1):10–17.
  20. Donell S. Patellar tracking in primary total knee arthroplasty. EFORT Open Rev. 2018 Apr;3(4):106–113.
  21. Chalidis BE, Tsiridis E, Tragas AA, et al. Management of periprosthetic patellar fractures. A systematic review of literature. Injury. 2007 Jun;38(6):714–724.
  22. Motsis EK, Paschos N, Pakos EE, et al. Review article: Patellar instability after total knee arthroplasty. J Orthop Surg (Hong Kong). 2009 Dec;17(3):351–357.
  23. Saffarini M, Demey G, Nover L, et al. Evolution of trochlear compartment geometry in total knee arthroplasty. Ann Transl Med. 2016 Jan;4(1):7.
  24. Saffarini M, Ntagiopoulos PG, Demey G, et al. Evidence of trochlear dysplasia in patellofemoral arthroplasty designs. Knee Surg Sports Traumatol Arthrosc. 2014 Oct;22(10):2574–2581.
  25. Clavé A, Le Henaff G, Roger T, et al. Joint line level in revision total knee replacement: assessment and functional results with an average of seven years follow-up. Int Orthop. 2016 Aug;40(8):1655–1662.
  26. Jazrawi LM, Birdzell L, Kummer FJ, et al. The accuracy of computed tomography for determining femoral and tibial total knee arthroplasty component rotation. J Arthroplasty. 2000 Sep;15(6):761–766.
  27. Kanekasu K, Kondo M, Kadoya Y. Axial radiography of the distal femur to assess rotational alignment in total knee arthroplasty. Clin Orthop Relat Res. 2005 May(434):193–197.
  28. Takai S, Yoshino N, Isshiki T, et al. Kneeling view: a new roentgenographic technique to assess rotational deformity and alignment of the distal femur. J Arthroplasty. 2003 Jun;18(4):478–483.
  29. Berger RA, Rubash HE, Seel MJ, et al. Determining the rotational alignment of the femoral component in total knee arthroplasty using the epicondylar axis. Clin Orthop Relat Res. 1993 Jan(286):40–47.
  30. Watanabe H, Gejo R, Matsuda Y, et al. Femoral anterior tangent line of the osteoarthritic knee for determining rotational alignment of the femoral component in total knee arthroplasty. J Arthroplasty. 2011 Feb;26(2):268–273.
  31. Melloni P, Valls R, Veintemillas M. Imaging patellar complications after knee arthroplasty. Eur J Radiol. 2008 Mar;65(3):478–482.
  32. Goto T, Hamada D, Iwame T, et al. Medial patellofemoral ligament reconstruction for patellar dislocation due to rupture of the medial structures after total knee arthroplasty: a case report and review of the literature. J Med Invest. 2014;61(3-4):409–412.
  33. van Gennip S, Schimmel JJ, van Hellemondt GG, et al. Medial patellofemoral ligament reconstruction for patellar maltracking following total knee arthroplasty is effective. Knee Surg Sports Traumatol Arthrosc. 2014 Oct;22(10):2569–2573.
  34. Whiteside LA, Ohl MD. Tibial Tubercle Osteotomy for Exposure of the Difficult Total Knee Arthroplasty. Clinical Orthopaedics and Related Research®. 1990;260:6–9.
Cookies help us improve your website experience.
By using our website, you agree to our use of cookies.