Total knee arthroplasty in the presence of severe flexion contracture: a report of 37 cases.

From 1987 to 1994, 37 total knee arthroplasties were performed in 23 patients with severe, fixed flexion contractures averaging 78 degrees (range, 60 degrees -100 degrees). Fourteen of the knees had flexion contractures of greater than 90 degrees and 7 were fused at 90 degrees. There were 19 women and 4 men. The average age at surgery was 42 years (range, 20-57 years). The diagnoses were rheumatoid arthritis in 17 patients, juvenile rheumatoid arthritis in 3, and ankylosing spondylitis in 3. Preoperatively, all patients were Knee Society Category C, with 14 being nonambulatory and 9 minimally ambulatory. Follow-up averaged 4.3 years (range, 2-8 years). Postoperatively, patients were immobilized in extension when not in continuous passive motion or physical therapy. Flexion contractures were corrected to an average of 7 degrees postoperatively (range, 0 degrees -15 degrees). Arc of motion improved from 25 degrees preoperatively to 82 degrees postoperatively. The average Knee Society knee scores improved from 25 points preoperatively to 78 points postoperatively, and the functional scores improved from 0 points preoperatively to 71 points postoperatively. Five knees were manipulated under anesthesia postoperatively. Complications included 3 transient peroneal nerve palsies, 1 transient episode of vascular insufficiency, 6 delayed wound healings, and 1 deep infection. There were no aseptic loosenings. We conclude that although technically difficult, total knee arthroplasty can be performed successfully in this challenging and highly debilitated subset of patients, giving them marked improvement in quality of life.

Revision total knee arthroplasty using the porous-coated anatomic revision prosthesis: six- to twelve-year results.

The porous-coated anatomic (PCA) revision prosthesis was one of the earliest revision total knee systems to feature more constraint for stability, stems for fixation, and thicker femoral components to address bone loss. From 1981 to 1989, 36 revision total knee arthroplasties were performed using the PCA prosthesis. Patients were evaluated by clinical examination, radiographs, and the Hospital for Special Surgery Knee Rating Scale. Six patients died and three were lost to follow-up; 27 knees remained for follow-up. The average age at surgery was 66 years. Follow-up averaged 9.8 years (range, 6-12.4 years). Range of motion averaged 4 degrees to 91 degrees before surgery and 1 degrees to 92 degrees after surgery. Knee scores averaged 48 before surgery and 86 after surgery. Three femoral and 10 tibial components showed lucent lines, with 3 tibial components showing progressive lucency. Six patients required reoperation: 3 for tibial component loosening, 1 for wound infection, 1 for deep infection, and 1 for persistent pain. These results suggest that the use of an implant specifically designed for revision can yield successful long-term results.

Structural allografting in revision total knee arthroplasty.

Between 1985 and 1991, 15 patients underwent structural allografting as part of revision total knee arthroplasty. All patients had large segmental, cavitary, or combination defects of the femur and/or tibia. Seven distal femurs and 12 proximal tibias required allografting. Patients were evaluated with physical examination, radiographs, and The Hospital for Special Surgery knee rating scale. Three patients died, leaving 15 allografts for follow-up study. The average age at surgery was 63 years. The follow-up period averaged 47 months (range, 30-101 months). Average range of motion before surgery was 4 degrees to 93 degrees, and after surgery, 2 degrees to 104 degrees. Average knee score was 47 before and 86 after surgery. Preoperative alignment averaged 5 degrees varus, ranging from 25 degrees valgus to 20 degrees varus, and postoperative alignment averaged 4 degrees valgus, ranging from neutral to 6 degrees valgus. All patients, except one, had improvement of pain and stability. All of the 15 allografts healed to host-bone and 13 showed evidence of incorporation. There were no infections or fractures of the allografts. One complication directly related to the allograft occurred; that patient had a tibial component fracture over a proximal tibial allograft 3 years after surgery. Three other complications occurred. One was tibial loosening in a patient who received a distal femoral allograft, the second was a proximal tibial fracture in a patient who received a distal femoral allograft, and the third was an intraoperative patellar tendon avulsion. These results suggest that structural allografting can provide a satisfactory method of managing large bone defects in the failed total knee arthroplasty.

Noncemented revision total knee arthroplasty.

From 1985 to 1991, 16 patients underwent 17 revision total knee arthroplasties using porous-coated tibial or femoral components, or both, without cement. Eight knees had both components implanted without cement, 4 the femoral component without cement and 5 the tibial component without cement. Fifteen revisions were performed for aseptic loosening, 1 for instability, and 1 for infection. Patients were evaluated by clinical examination, radiographs, and the Hospital for Special Surgery Knee Rating Scale. Two patients died, leaving 15 revision total knee arthroplasties for followup. The average age at surgery was 65 years old. Followup averaged 5.6 years (range, 2.4-10.0 years). Range of motion preoperatively averaged 4 degrees to 96 degrees, and postoperatively 0 degree to 102 degrees. Average knee scores were 52 points preoperatively and 87 points postoperatively. Two tibial components showed lucent lines > or = 1 mm in 1 or more zones, 1 of which showed progressive lucencies. One femoral component showed a 1-mm lucent line in 1 zone, which was not progressive. Two tibial components were revised, 1 for a proximal tibial fracture below the prosthesis, and the other for prosthesis loosening and breakage. There was 1 stitch abcess with intrarticular extension, but no other infections. These results show that successful revision total knee arthroplasty can be accomplished using noncemented porous-coated components, with results comparable with those of cemented revisions.

Results of reconstruction for failed total elbow arthroplasty.

Failure of total elbow arthroplasty leads to difficult and complicated surgical reconstruction. This study evaluates the results of reconstruction after implant removal with respect to pain, motion, and functional ability. Between 1978 and 1985, 11 patients required implant removal. Indications for removal were infection for seven, implant fracture for three, and recurrent dislocation for one. The original diagnosis was rheumatoid arthritis in six elbows and traumatic arthritis in five. The average length of the follow-up period was 5.5 years after implant removal (minimum, two years). Treatment consisted of implant removal and soft-tissue arthroplasty combined with external fixation in ten patients, and attempted arthrodesis with external fixation in one. There were four good, one fair, two poor, and four failed results. Satisfactory results were obtained in seven of the eight elbows in which an anatomic arthroplasty was achieved. This consisted of containment of the ulna by the humeral epicondylar remnants. All eight elbows were pain-free with an average arc of motion of 85 degrees (range, 55 degrees to 120 degrees). They had excellent elbow flexion power; however, triceps strength was often compromised. In the three elbows in which anatomic arthroplasty could not be achieved, one was flail, one was later converted to an arthrodesis with a customized plate, and the third required an immediate arthrodesis. All three were rated as failures. Fractures occurred in five of the 11 elbows. One occurred preoperatively, three occurred intraoperatively, and one occurred postoperatively. All healed satisfactorily during the course of immobilization. The importance of an anatomic arthroplasty when removing a total arthroplasty cannot be overemphasized. Retaining the epicondylar segments is important because satisfactory results were obtained in patients in whom entrapment of the olecranon within the epicondylar ridges was obtained. Such patients can achieve a satisfactory soft-tissue arthroplasty without the use of an interpositional material. When the epicondylar ridges were not retained and there was marked instability, the patients did not achieve satisfactory results.

Salvage of non-union of supracondylar fracture of the humerus by total elbow arthroplasty.

Fourteen patients in whom open reduction and internal fixation could not be achieved satisfactorily had a total elbow arthroplasty for non-union of a supracondylar humeral fracture. All patients had an established non-union, and ten had had from one to four previous attempts at internal fixation. The average age at the time of operation was sixty-five years. The average duration of follow-up was five years, with a minimum of two years. The average preoperative elbow score was 17 points, with both pain and functional disability present. The average postoperative score was 84 points; there were eight good or excellent results and three failures. The latter three patients had an additional operation: one each for dislocation, loosening of the humeral component, and deep infection. Salvage of supracondylar non-union by means of a total elbow arthroplasty is a technically demanding procedure. Strict indications for selection of patients must be followed. A semiconstrained implant is recommended, with preservation of the epicondyles and their muscular attachments in order to achieve balance of the soft tissues.

Total elbow arthroplasty for complete ankylosis of the elbow.

Sixteen patients who received nineteen semiconstrained total elbow replacements for complete ankylosis of the elbow were followed for an average of five and three-quarters years (range, two to twelve years). The average preoperative elbow score was 23 points and the average postoperative score was 84 points. Postoperatively, the average flexion was 115 degrees; extension, 35 degrees; and pronation and supination, 95 degrees. There were fifteen excellent or good results. There was one failure due to a deep infection, but after removal of the prosthesis a satisfactory fascial arthroplasty was achieved in this elbow. Function was improved in all patients, and all patients had relief of the preoperative pain. For the arthroplasty to succeed, the patient must have a good understanding of the procedure and must be willing and able to comply with the postoperative rehabilitation program. The use of a semiconstrained, often custom-fit, implant is necessary. The Bryan-Morrey posteromedial approach to the elbow is recommended for the procedure, since this approach allows early institution of range-of-motion exercises.

Total ankle arthroplasty in rheumatoid arthritis: a long-term follow-up study.

Patients with rheumatoid arthritis who had undergone total ankle arthroplasty and had a minimum of 2 yr follow-up were studied. Of the original 21 patients 17 were available for review. Twenty-three ankle replacements with an average follow-up of 5.6 yr were studied. On follow-up 2 ankles were rated excellent, 13 were rated good, 4 were rated fair, and 4 were rated poor. Thus, 83% were satisfactory on follow-up. Radiographic analysis revealed migration and settling of the talar component in 14 of 15 cases. Bone cement radiolucencies were found in 14 of 15 cases. Bone cement radiolucencies were found in 14 of 15 tibial components with tilting in 12 of these components. The postoperative position of the implant did not correlate with the development of radiolucencies or migration of the implant.