Patellofemoral Replacement

Twenty-five years ago Dr. Martin Blazina (Fig 1) and his colleagues published this classic article concerning patellofemoral replacement. Dr. Duncan McKeever1 had written a paper on patellar resurfacing nearly 15 years previously, but Dr. Blazina’s contribution took McKeever’s concept one step further. Herein he describes the results of combined resurfacing of both sides of the patellofemoral articulation. In this forthright exposition he presented his results and reviewed the problems, complications, and ancillary procedures necessary to make the operation succeed. He concluded modestly that “Patients for isolated patellofemoral replacements should be carefully selected ... the primary indications are distortion or severe degeneration of the femoral groove.”Fig 1.: Dr. Martin E. Blazina is shown.Martin Blazina, born in May 1924, grew up in Argo, Illinois. He attended the University of Illinois before World War II. He majored initially in chemical engineering but the war interrupted his education from 1942 to 1945 and when he returned from the South Pacific he had decided to become a physician. After his graduation from the University of Illinois with an doctor of medicine degree, he entered orthopaedics as a resident at the Los Angeles County Hospital, finishing in 1959. He practiced medicine in the Los Angeles area for the rest of his life, which ended in November 2002 after a 2-year struggle with pancreatic carcinoma. He served on the full-time faculty at the University of California at Los Angeles (UCLA) for 12 years, during which he provided the medical care required by the varsity and intramural teams at UCLA. His experience there honed his acknowledged expertise in the field of sports medicine in which he established a national and international reputation. Since June, 1974, we have investigated 85 patellofemoral joint replacements. Seventy other replacements of the patellofemoral articulation were performed during the same period, but these procedures were done in conjunction with other compartment replacements in the knee and are not discussed here. Between June, 1974 and March, 1975, the Type I Richards patellofemoral replacement was used and involved 20 cases. Since that time, the Type II Richards patellofemoral replacement with the tonguelike extension of the femoral groove component down towards the intercondylar notch region has been used (figure not shown). We have had no experience with the other types of patellofemoral replacements currently available. This article describes our clinical experience with patellofemoral replacement. An analysis of antecedent surgical procedures performed and subsequent revisionary procedures performed is also included. No attempt is made to compare patellofemoral replacement with patellectomy. This consideration is important, but it is our understanding that the subject of patellectomy will be presented in another discussion in this symposium on the patellofemoral articulation. CLINICAL APPRAISAL In a special follow-up study, 55 patients with 57 patellofemoral replacements (out of the series of 85 replacements) were interviewed and examined. Age distribution was from 19 to 81 years old with an average age of 39 years. The shortest follow-up was 8 months and the longest was 42 months, with an average follow-up period of 21.7 months. Forty-three patients felt that they were much improved, both symptomatically and functionally. Nine patients were unimproved and 3 patients were worse. Preoperatively, 54 patients had moderate to severe pain, whereas postoperatively 12 patients stated they still had severe pain and 11 patients stated that they were experiencing moderate pain. Twenty-one patients, who were unable to perform their usual occupation preoperatively, returned to their previous job postoperatively. Fourteen patients, who were totally disabled preoperatively, returned to work in a modified status. Eleven patients remained totally disabled postoperatively. The remaining 9 patients have not reached a stationary status and have not yet returned to work. In regard to recreational activities (although not encouraged in this direction), 7 patients resumed bicycling and running. Two patients were playing tennis, and one each was golfing, skiing, horseback riding, and bowling. Five patients had resumed swimming. Of the 32 patients who were unable to walk more than 2 to 4 blocks preoperatively, 18 patients remained essentially unchanged postoperatively. Six out of 19 patients continued to limp after the patellofemoral replacement surgery. Thirteen out of 40 patients continued to have significant episodes of effusion postoperatively. Of the 29 patients who had frequent attacks of giving way preoperatively, 13 continued to have some giving way postoperatively. Eleven patients were able to climb stairs preoperatively, and this number increased to 19 postoperatively. Preoperatively, one patient had less than 45° motion, 9 patients had 45-90° motion, 11 patients had 90-120° motion, and 36 patients had more than 120° motion. Postoperatively, one patient had less than 45°, 5 patients had 45-90°, 25 patients had 90-120°, and 26 patients had more than 120° of motion in the knee. The average loss of motion was 4° in this series. ANTECEDENT PROCEDURES Sixty-six out of 85 patients in this series had had surgery performed prior to the patellofemoral replacement (Table 1). Forty-three patients had had one operation, 17 patients had had 2 operations, and 5 patients had had 3 operations.Table 1: Procedures Performed prior to Patellofemoral ReplacementIn reviewing the procedures performed, it was noted that 25 patellaplasties and 2 secondary patellaplasties had been performed and apparently were not efficacious. Twenty-four medial meniscectomies, 4 secondary medial meniscectomies and 8 lateral meniscectomies had been done. Their relationship to the development of significant patellofemoral arthritis has not been well-defined and represents a complex situation. Vastus medialis transposition (23 times), secondary vastus medialis transposition (3 times), medial patellar retinaculum advancement (7 times), secondary medial patellar retinaculum advancement (2 times) all represent attempts at reinforcing the medial pull of the patella as a part of proximal extensor realignment. Vastus lateralis release (23 times) and secondary vastus lateralis release (2 times) represent attempts at obviating the lateral drift of the patella as a part of proximal extensor realignment. Transfer of the tibial tubercle (12 times), the Roux-Goldthwait procedure (4 times), and transfer of the medial ½ of the patellar tendon (2 times), constituted efforts at distal extensor alignment. Other procedures (chondroplasties and removal of loose bodies) relate to articular cartilage damage incurring at other areas in the knee. Incidentally, it is interesting to note that 3 patients had had an open reduction and internal fixation of a fracture of the patella. In reviewing these antecedent procedures, one is led to realize that there are certain failures in extensor realignment surgery and certain failures in chondroplasty of the patella. The incidence of failure has not been clearly defined at the present time. REVISIONARY SURGERY Thirty patients (out of the series of 85 replacements) had further operations performed subsequent to the patellofemoral replacement (Table 2). Twenty of these patients had one subsequent operation and 10 patients had 2 subsequent operations.Table 2: Procedures Performed Subsequent to Patellofemoral ReplacementAn analysis of the subsequent procedures performed indicate 2 main areas of revisionary direction. Procedures Related to Extensor Realignment Twenty procedures involving lateral capsular release, vastus lateralis release, secondary vastus lateralis release, and 2 procedures involving secondary vastus medialis transposition were performed as an attempt to obviate toggling of the patellar prosthesis laterally out of the femoral groove. Seventeen other procedures involving transfer of the tibial tubercle, V-plasty lengthening and Z-plasty lengthening of the patellar tendon, were performed as an attempt to obviate “hooking” of the patellar prosthesis coming out of deep flexion or to decompress the patellofemoral articulation in the deep flexion phase. The patellar prosthesis which migrates down into the notch in deep flexion may leave the femoral groove and occasionally catch momentarily as the knee is extended. Also, it has been noted that pressures on the patellofemoral articulation increase tremendously beyond 90° of flexion. It was felt in some instances, that these phenomena were a result of actual or relative shortening of the patellar tendon as a result of previous surgeries. Procedures Related to the Prosthetic Replacements Five patients had the femoral groove replaced, one patient had the patellar prosthesis changed, and 3 patients had both the femoral groove and patella replaced. These alterations were done because of apparent positioning problems developing at the time of initial insertion of the replacements. On 2 occasions, the remaining protruding edge of the lateral facet of the patella was removed in order to obviate an irritative situation over the lateral capsular region as the replacements moved in flexion and extension of the knee. Two patients had wound healing problems with underlying infection. One patient responded favorably to systemic and local prescription; the other situation was not controlled and the components were removed. At the present time, the patient has stiffness and discomfort in his knee and arthrodesis has been proposed. Another patient had the patellofemoral replacement removed along with the remainder of the patella because of persistent pain and swelling despite repeated negative cultures. His status is considered indefinite at this time. Other Considerations Because of certain instances of the development of postoperative fibroarthrosis, release of intra-articular adhesions and partial excision of a fibrosed fat pad were performed. These procedures usually resulted in increased motion of the knee. DISCUSSION In reviewing the clinical response to patellofemoral replacement, one is impressed by the potential for gain in the area of functional capability. On the other hand, although there has been a statistically favorable response to discomfort, it is not sufficiently predictable in the individual patient and may be related to factors to be discussed below. Range of motion in most patients is usually regained promptly, but in some may be delayed and may only be forthcoming eventually after a secondary procedure. The revisionary procedures outlined above indicate areas of concern such as tracking of the patellar prosthesis, which may lead to “toggling” as it comes into or out of full extension, and certain phenomena occurring in the region of deep flexion, such as “hooking” and crepitation. Insofar as smoothening the tracking is concerned, it is possible that further modification of the femoral groove by making the groove less severe in its central portion and rounding out its peripheral flanges may improve this situation. Also, some change in the contour of the patellar prosthesis may be forthcoming. One suggestion is to make the patellar prosthesis more “anatomic” with an elongated lateral facet and a shortened medial facet. Another suggestion is to design the patellar prosthesis in the shape of a “button” allowing it to “roll” gently without a noticeable “toggle” as it slides in the present groove. In regard to the deep flexion phase, a concerned effort should be made to recess the femoral groove replacement at its termination in the intercondylar notch. Also, assuring adequate length of the patellar tendon is very helpful. A basic consideration is related to the fact that in deep flexion the medial and especially the lateral portions of the patellar prosthesis do articulate with the medial and lateral femoral condyles. It has been ascertained that the present patellar prosthesis, which is made up of ultrahigh-density polymer does not represent the ultimately favorable contact surface against articular cartilage.1 Some patients appear to tolerate this type of interaction whereas others appear to have difficulty-the results are unpredictable. These reactions appear similar to past experience with the interaction between the femur and the quadriceps tendon of the patient who has undergone patellectomy and between the femur and a metallic patellar prosthesis. Perhaps, once again, it would be possible to obviate the extent of contact by redesigning the shapes of the patellar and femoral groove prostheses. On the other hand, the development of a patellar prosthesis using some other material or materials which would predictably lead to less discomfort in the flexion phase would be helpful. Parenthetically, if modification of the patellar prostheses in this direction should occur, then one could seriously consider replacing only the patella in certain situations where the groove is still in satisfactory condition. We have not been aware that contact between the femoral groove implant and the quadriceps tendon, while the knee is in deep flexion, has represented a significant problem in this series, although this phenomenon represents another area of consideration. Every attempt should be made to save the patella in those instances of patellar tracking problems occurring in young people, including the use of patellar decompression procedures. In patients with patellofemoral arthritis without distortion or severe degeneration of the femoral groove, patellectomy can be considered. In instances of a distorted or badly worn femoral groove, femoral groove replacement and, consequently, a patellar replacement may have to be considered (figure not shown). SUMMARY From an analysis of the clinical evaluation of 85 cases and subsequent revisionary procedures, patients for isolated patellofemoral joint replacements should be carefully selected. At the present time, the primary indications are distortion or severe degeneration of the femoral groove.