A Multicenter Review of Patellar Complications Using a Modem Design Total Knee System

Edward T. Habermann, MD, Lester S. Borden, MD, FACS Anthony K. Hedley, MD, FRCS, David S. Hungerford, MD Kenneth A. Krackow, MD

Abstract: This retrospective study reviewed the results of 3,218 total knee arthroplasty cases using one total knee replacement system at five centers from July 1991 to October 1997. Cases were reviewed to determine the occurrence of patellar related complications post-operatively. Events reported included patellar dislocation/subluxation, patellar fracture, component loosening and tendon rupture. The rates were noted to be low in comparison to literature reports of other series. The overall patellar related complication rate was 0.3% (11/3218) for the cases reviewed. Given the historically high rates of patellofemoral complications reported in the literature, this review has shown that good design of the implant helps reduce patellar related problems in total knee arthroplasty patients.

Reports of patellar related complications in total knee replacement cases in the literature range from 1% to 50%(1) and include reports of subluxation/dislocation, patellar bone fracture, component loosening, and in the case of metal backed patellar components, metal/plastic construct related events. Extensor mechanism related complications are a frequent reason for reoperations reported in many aseptic total knee series.(2) Experience gained since the early knee replacement procedures in the 1970's has shown that patient selection, operative technique and implant design all influence the patient outcome.(3) Biomechanical analysis has revealed that the combination of high loads, small contact areas, changing contact points and multiple soft tissue attachmments make the patella particularly vulnerable to problems after total knee arthroplasty.(4) As such, resurfacing the patella remains controversial. Most results have shown that there is relief of pain and improvement in stair climbing ability when the patella is resurfaced compared to not resurfaced. Ultimately, the decision to resurface is left to the surgeon and is influenced his/her past experience and the patient's lifestyle. Inflammation and severe changes in the articular surface with a lack of congruence between the patiet's patella and the natural trochlea frequently lead to a decision in favor of resurfacing. This paper presents the results of a retrospective review of 3,218 total knee replacement cases with patellar resurfacing using one total knee system. Occurrence rates of patellar related complications were determined to assess the results of a modern design prosthesis developed with features to improve patellar tracking, reduce shear forces, resist subluxation, prevent patellar bone degradation and fracture, and enhance articular conformity.

Materials and Methods

Between July 1991 to October 1997, 3,218 total knee cases using the Duracon(r) Total Knee System (Howmedica Osteonics, Inc. Rutherford, NJ, USA) were performed at five centers by five surgeons. This tricompartmental replacement system includes femoral, tibial and patellar components. Two patellar component design versions, a metal backed Asymmetric Patella and an all plastic Asymmetric Patella were used in these cases (Figure 1). The majority of components used were the all plastic design. Case records were reviewed retrospectively for the occurrence of patellar related complications postoperatively.

Figure 1. The Asymmetric Metal Backed Patella and All Plastic Patella Used in This Review


Patellar related complications were identified in 11 of the 3,218 cases for an overall complication rate of 0.3%. Specific events included 2 subluxation/dislocations, 4 patellar fractures, 2 component loosenings, 2 patellar tendon ruptures and 1 quadriceps tendon rupture (Table 1).

Table 1. Patellar Related Complications (N=3,218)
# of Cases
Occurrence Rate
Patellar Fracture
Component Loosening
Patellar Tendon Rupture
Quadriceps Tendon Rupture

The two reports of dislocation/ subluxation were in patients who had an initial lateral release. In one case the event was related to trauma, while the other was due to a spontaneous disruption of the medial retinaculum. Both were repaired and went on to a successful outcome. The O.O6% dislocation/subluxation rate for this series is noted to be much lower than the 14% and 15% reported in literature by Sneppen(5) and Bindelgiass,(6) respectively.

Fractures were reported in four cases and included two displaced (one minimally), one inferior pole and one comminuted type with a loose patellar component. Trauma was noted to be the cause in three of the cases. Treatment included open reduction with internal fixation, patient monitoring and partial patellectomy. In the remaining case, the patient had Parkinson's Disease and reported some mild associated trauma. Treatment included a lower extremity splint. Indeed, the overall fracture rate of 0.12% would be lowered to 0.03% if the cases directly caused by trauma were tracted from the total. These results compare favorably to the 0.33% (9/2719) reported by Grace and Sim(7) for resurfaced patella cases, and when subtracting trauma induced events, compare favorably to their results reported for nonresurfaced patellae, which were quite low at O.05% (3/5530).

There were two cases reporting patellar component loosening. One was in a cemented total knee where the patient had poor bone stock. This patellar component was not replaced on revision. Not replacing the failed patellar component often leads to a good functional result and lack of pain. The other report of loosening was in a 61 year old female at 13 months post surgery with no known etiology. The component was replaced with a cemented patellar component. Given the large number of cases included in this review and the varying degrees of patient activity level represented, overall loosening rate of 0.06% is noted to be very low.

There were two patellar tendon ruptures (0.06%)and one quadriceps tendon rupture (0.03%) reported. One patellar tendon case was noted to be a spontaneous rupture, with significant quadriceps weakness afterward, in a 76 year old female with rheumatoid arthitis. The quadriceps tendon rupture was noted to be post-traumatic secondary to a fall with hyperflexion of knee. All three cases were repaired without difficulty.


Complications related to the TKA patellar component presented in the literature include subluxation/dislocation, patellar tilt, fractures, instability and component loosening. As noted previously, rates range from 1% to 50%.(1) The Duracon(r) Total Knee System patellar conponents incorporate design features that were developed as a result of clinical and biomechanical analyses. The low occurrence rates of patellar complications in large series demonstrates the result of careful paient selection, refined surgical technique and appropriate component design for successful patient outcome. Specific design criteria have been included to address each of the complications frequently reported in the literature and in clinical practice.

The improved anatomic tracking with this system addresses the occurrence of patellar component instability by allowing the trochlea to engage the patella early in flexion, minimizing excessive medial/lateral laxity throughout the range of motion. The occurrence rate of dislocation/ subluxation for this series of 3,218 cases was noted to be only 0.06%. Improved conformity of the articulating surfaces resulting in reduced shear forces and the incorporation of a lateralized deeper trochlear groove are also thought to have contributed to the low rate of instability related complications. The patella is contained as it tracks through the arc of motion, limiting destabilizing laxity. Laboratory testing of these implants using a human cadaveric knee, with a nominal body weight load of 150N, positioned at four different angles and then rotated, measured M/L motion and patellar laxity.(8) The results illustrated in the chart below demonstrate that trochlear geometry contains the conforming pate component, resulting in less M/L motion or 1axity through the range of motion as compared to designs which are non-conforming (Figure 2).

It also has been demonstrated in the laboratory that patellar components used in this clinical series can resist subluxation at up to four times the normal physiological forces experienced at the patellofemoral joint(9)(Figure 3). This may be important in preventing subluxation/dislocation as discussed earlier, as well as component loosening, which was also noted to be low for this series at 0.06%.

Figure 2. Mechanical Testing Results of Patellar Patellar M/L Containment in Varying Degrees of Flexion

The conforming Duracon patellofemoral tracking was compared to that of a non-conforming patellar design at several degrees of flexion, in a cadaveric test. The conforming patella demonstrated less M/L laxity.(8) At 450° and 90° of flexion, the conforming patella shows no destabilizing M/L laxity. This indicates good containment of the patella in the trochlea. This suggests superior anatomic patellar tracking is realized with the conforming Duracon design.


Figure 3. Mechanical Testing Results of Resistance to Lateral Subluxation Force The patellar components used in this study can resist subluxation forces 2 to 4 times the lateral forces experienced in normal activity.

A review of the literature suggests that some causes of fracture of patellar bone after total knee arthroplasty may be weakening and devascularization secondary to excessive bone removal during patellar resurfacing,(10) an increase in patellar joint reaction force secondary to use of a component that is too thick and tightens the patellar tendon,(11,12) and weakening of the bone and increased stress concentration secondary to use of a prosthetic component with a thick central fixation lug.(13) The patellar components used in this series are available in a variety of sizes to allow for appropriate bone resection to optimize the strength of the residual bone. Taking out large segments of centrally located patellar bone can lead to stress risers and ultimately fractures. Thus, the design of the component fixation mechanism and the ability to minimize the amount of bone resected are important features of the prosthetic design. The overall patellar fracture rate for this series was 0.12% (4/3218), while the rate for fractures not secondary to trauma was even lower at 0.03% (1/3218). The component size/thickness options and the bone sparing three fixation peg design combined with careful intraoperative technique in this study address the surgical concerns encountered with the patella in other studies.

Fracture/dissociation of the polyethylene in metal backed patellar components were reported within 2 years in the literature of 5.3%(14) and 10.7%(15)and was associated with polyethylene wear. Additional published reports of component testing have indicated that more conforming patellofemoral designs result in improved wear resistance and greater fatigue performance.(16) The components in this review were designed to improve patellofemoral articular conformity and maintain joint contact area throughout the range of motion with the goal of decreasing stress on the patellar component (Figure 4). No patellar implants in this study fractured or dissociated.

Figure 4. Congruent Patellofemoral Surfaces Have Demonstrated Resistance to Wear.

The literature records a high rate of total knee revisions caused by extensor mechanism complications? It is also encouraging to note the very low incidence of such problems in this series. Careful attention to surgical technique and femoral/patellar designs that recreate normal soft tissue function contribute to these very good results.


Given the historically high rates of patellar component related complications reported in the literature, the overall rate of O.3% for this series using a single total system has shown the important contribution component design can make to achieving a good clinical result. The implants used in this series have been effective in improving anatomic tracking, lowering articulation stresses, reducing sheer forces, increasing resistance to patellar subluxation and improving patient outcomes as evidenced by the low rates of subluxation/dislocation, fracture and component loosening. Extensor mechanism complications, which have plagued other total knee results, were also very low in this series. This can be attributed to sound implant design and good surgical technique. The results observed in this very large study suggest some troubling problems seen in knee arthroplasty may be addressed in part, through implant design.


  1. Boyd AD, Ewald FC, Thomas WH, Poss R, Sledge CB. Long Term Complications after Total Knee Arthroplasty with or without Resurfacing of the Patella. J Bone Surg, 75A(5), 1993.
  2. Brick GN, and Scott RD. The Patellofemoral Components of Total Knee Arthroplasty. Clin Orthop, 321:163-178, 1988.
  3. Huberti HS, and Hayes WC. Patellofemoral Contact Pressures: The Influence of Q-Angle and Tendofemoral Contact, JBJS, Vol. 66-A, June 1984, pp. 715-724
  4. Briard J-L, and Hungerford DS. Patello Femoral Instability in Total Knee Arthroplasty. J Arthroplasty. 1989 Supplement
  5. Sneppen 0, Gudmundsson GH, and Bünger C. Patellofemoral Function in Total Condylar Knee Arthroplasty. Internat Orthop, 9:65-68, 1985
  6. Bindelglass DF, Cohen JL, Dorr LD. Patellar Tilt and Subluxation in Total Knee Arthroplasty. Clin Orthop, 286, 1993
  7. Grace JN, Sim FH. Fracture of the Patella after Total Knee Arthroplasty. Clin Orthop, 230:168-175, 1988
  8. Howmedica R&D Technical Reports/ Data on File
  9. Howmedica R&D Technical Reports/ Data on File
  10. Berman AT, et al. Evaluation of Patellar Complications Total Knee Arthroplasty. Scientific Exhibit, AAOS, 1993
  11. Insall JN, et al. The Posterior Stabilized Condylar Prosthesis. J Bone Joint Surg, 64A(9), 1982
  12. Kitsugi T, et al. Clinical Comparisons of Miller Gala and AMK Total Knee Systems. J Arthroplasty, 9(2), 1994
  13. Clayton ML, et al. Patellar Complications after Condylar Arthroplasty. Clin Orthop, 170, 1982
  14. Lombardi AV Jr, Engh GA, Volz RG, et al. Fracture/Dissociation of the Polyethylene in Metal Backed Patellar Components in Total Knee Arthroplasty. J Bone Joint Surg, 70A:675-679, June 1988
  15. Stalberg DS, Stalberg BN, Hamati Y, and Tsao A. Failure Mechanisms of Metal Backed Patella Components. Clin Orthop, 236:88-105, 1988
  16. Hsu HP, Walker PS. Wear and Deformation of Patellar Components in Total Knee Arthroplasty. Clin Orthop, 246, p. 264, Fig 6B, 1989.