Duncan C. McKeever (Fig 1) was born in rural Kansas but was educated in Kansas City, MO where he attended Westport High School and the Kansas City Junior College. He then attended the University of Kansas College of Medicine, from which he graduated in 1929. McKeever joined the Naval Reserve to help finance his education. After graduation, he interned at the Naval Hospital in Brooklyn and spent 3 years at the Great Lakes Naval Training Station in Chicago. He then returned to Kansas City where he studied orthopaedic surgery at the Dickson Diveley Clinic until 1939.
McKeever began his practice in Houston, TX in 1939. He was called to active duty in the Navy in 1941 and served as the chief of orthopaedics at the Long Beach Naval Hospital and at the Area Hospital in Honolulu. When he was discharged with the rank of Captain in 1945, he returned to Houston to resume his practice.
McKeever was a man of enormous energy which he expended in his busy practice and in promoting new and innovative methods of treatment. He was the cofounder of the Spectator Correspondence Club and was a founding member of the Association of Bone and Joint Surgeons for which he later served as president. He made special efforts to host orthopaedic surgeons from Central and South America. His most important characteristics were his capacity for friendship and his personal charm and magnetism. His tragic death at the age of 55 years was a loss to the orthopaedic community.
The classic article describes one attempt at a solution to the problem of patellofemoral arthritis. Although it was not successful, it was a bold attempt to find a solution to a problem that even now, approximately 50 years later, has not been solved.
Leonard F. Peltier MD, PhD
Repeated trial of patellectomy and various patellar plastic operations and subsequent observation of these cases over a period of years has left many surgeons dissatisfied. In fractures and chondromalacia, with no other internal derangement, patellectomy will undoubtedly produce good results after prolonged physiotherapy, but these are simple problems compared with arthroplasty or the débridement and reconstruction procedures necessary in knee joints badly damaged by trauma, degenerative changes, or disease. In such cases, patellectomy frequently imposes an insurmountable obstacle that leads to failure, when restoration of the extensor mechanism to normal, or nearly normal, would have produced a good result.
Cosmetically, patellectomy leaves much to be desired, particularly in women.
Without a patella, the femoral condyles have no protection from direct trauma in falls and when kneeling. A knee without a patella may be better than one in which the patella has a roughened or misshapen articular surface, but it cannot function as well as a knee with a good patella. It seems logical to conclude that, while patellectomy produces some satisfactory results in the simpler problems involving the extensor mechanism, it is not a satisfactory procedure in more complicated knee-joint problems, and, in any case, it must be possible to do better.
A prosthesis for the articular surface would permit the retention of the patella. A knee joint, normal in mechanics, appearance, and function, should be possible in those patients in whom only the patella is at fault. A much better knee joint should be obtainable in knees having both a defective patella and other mechanical defects, and it is in these cases that the use of a prosthesis could contribute greatly to a good result.
With a normal patella as a model, a Vitallium prosthesis was designed so that the osseous portion of the patella could be fitted into it. A screw transfixing the patella and both lips of the prosthesis hold it in place when the quadriceps is relaxed. When the quadriceps is under tension, the screw is probably non-functional.
This prosthesis has been produced in three sizes. The largest is used in the majority of cases. The medium size, made at the instigation of Dr. Garrett Pipkin, is 12.5 per cent smaller. It is used in small individuals and, in case of question, this size is preferable. The smallest size might be used in children, but to date this has not been tried.
A jig was devised later in order to facilitate fitting the patella to the prosthesis. This instrument may save thirty minutes or more in operative time, but the prosthesis can be fitted without it.
Technique of Application
A median parapatellar incision is recommended because it facilitates orientation in placement of the prosthesis. It is not necessary to disturb the insertion of the patellar tendon. The initial incision is carried directly into the joint. The skin is reflected laterally so as to expose the anterior surface of the tendon overlying the patella. The patella can then be turned over, completely exposing its articular surface. Necessary débridement or reconstructive work on the remaining portion of the joint is carried out first.
In all cases, the patella should be trimmed to fit the prosthesis. If, by chance, the patella will fit the large-size prosthesis without alteration, this size should still not be used because the resulting articular surface of the prosthesis would be larger than that of the original patella and would not fit the condyles.
In every case some portion of the patella will have to be trimmed away. The distal and medial portion should be preserved. The jig is to be placed in such a position that this portion is under the plate which determines the circumference. The portion to be removed is the superior and lateral portion. A motor saw, rongeur, or any other cutting instrument the surgeon may prefer can be used.
The jig is applied with the points on the plate side, sticking into the articular surface of the patella and the ring on the anterior surface of the patella. If the points will not engage, a preliminary cut may be made to convert the articular surface to a flat plane. The jig will be almost at right angles to the axis of the leg, but should be placed without particular regard to this position. In placing the jig accurately in alignment, the axis of the two most prominent parts of the edge of the ring, which will be in line with the vertical ridge on the prosthesis, should be in the axis of the intercondylar groove. This is the guide for the application of the jig. The drill guide position is also disregarded, as it purposely does not coincide with the vertical axis of the patella. It was so placed to prevent loosening of the screw by the torque exerted on the patella as the knee flexes and extends.
The jig is removed, turned over, and reapplied in the same axis of the intercondylar groove, as determined above, and closed tightly. A seven sixty-fourths of an inch drill is passed through the drill guide and the patella, and the chuck is removed leaving the drill in place. That portion of the patella protruding through the ring is cut off flush with the ring. The remnant will be thinner than the original patella and should just fill the prosthesis. In young vigorous individuals the patella is very hard and is cut with difficulty. In older and less active individuals, it is relatively soft and easy to cut. The drill and jig are removed.
The prosthesis is placed on the reshaped portion of the patella so that the small facet is toward the edge of the incision. The prosthesis is clamped tightly on the patella and is held with a lion-jaw forceps in such a position that the screw holes in the patella are in line with the screw holes in the prosthesis. This is made easier by passing a probe or pin through the prosthesis and the hole in the patella before the clamp is applied. An accurate fit is not essential, but the patella should be seated into the prosthesis as deeply as possible. The screw is then inserted tightly, so that the tip of the screw engages the threaded screw hole in the prosthesis. Only one of the screw holes is threaded. It may be at the top or the bottom depending on which side is being fitted. If the prosthesis has been applied properly, it will turn over into functional position so that the smaller facet faces the medial condyle and the larger facet faces the lateral condyle.
Especial care should be taken to see that there is no ridge at the proximal edge of the intercondylar groove so that the prosthesis may pass over this area without catching or interference. If such a ridge is present, it should be removed with an osteotome or a motor burr.
Effusion, as a postoperative complication, can be minimized by making an opening from the suprapatellar pouch into the space beneath the quadriceps muscle, as recommended by Chandler. This procedure is recommended. The incision is closed. These directions may be followed for either knee, as the prosthesis and the jig may be used on either side by merely rotating them 180 degrees.
Postoperatively a cotton and elastic pressure bandage is applied to the knee. This is removed at the end of forty-eight hours and motion is started at once. Weightbearing can be started as soon as the patient is able to raise the extended leg against gravity. When only the patella is damaged, 90 degrees of motion should be easily obtainable within three weeks. When there is extensive involvement of the remaining portion of the knee joint, the restoration of motion and the return of function will depend on the degree of damage and the determination of the patient. If a postoperative effusion occurs, it should be removed by tapping. Hyaluronidase may be used to assist the absorption of the effusion. Hydrocortisone may be used intra-articularly if the surgeon wishes.
The author has used this prosthesis successfully in over forty knees. Some of the patients were operated upon over five years ago. There have been four failures due to infection. There have been no mechanical failures in this series. In two patients, convalescence was prolonged to several months because the prosthesis used was too large.
Other surgeons have used this prosthesis in one or several cases. The principal indication in these cases has been simple chondromalacia. The author considers this to be one of the more important indications for the use of the prosthesis, second only to its use in extensive reconstructive procedures.
The most important lesson derived from this experience is that the restoration of a normal, or nearly normal, quadriceps extensor mechanism is a decisive factor in the success of knee-joint reconstruction and outweighs all other factors in importance. Such a restoration is feasible through the use of this prosthesis.