Small wire external fixation of high energy tibial plateau fractures

Open plate osteosynthesis for high energy tibial plateau fractures with dissociation between the metaphysis and diaphysis has been plagued with frequent soft tissue complications. The Harbor-University of California at Los Angeles Medical Center's experience with small wire external fixation supplemented by limited internal fixation is examined. This alternative method of adequate stable fixation offers the advantage of minimal soft tissue compromise. Twenty-four patients with Schatzker Type VI tibial fractures were treated with small wire external fixation. Supplementary limited internal fixation was used with percutaneous screws in 10 patients and with open reduction in one patient. Sixteen patients had isolated fractures, and eight others suffered multiple injuries. Minimum followup was 12 months. All fractures healed. Complications included one septic knee, two infections at screw sites, and one 10 degrees knee flexion contracture. One knee had Grade 3 radiographic arthrosis, five had Grade 2, 10 had Grade 1, and eight showed no arthrosis. The outcomes (Knee Society clinical rating system) of this study compare favorably with outcomes described in reports published previously for this type of fracture, despite inclusion of eight multiply injured patients. This technique preserves the goals of early range of motion and stable fixation for these devastating injuries, while decreasing the observed major wound complications and nonunion rates. However, longer followup may reveal higher arthrosis rates, specifically in those fractures that were not anatomically reduced.     

The dynamic hip screw with support plate--a reliable osteosynthesis for highly unstable "reverse" trochanteric fractures?

ORIF management of unstable trochanteric fractures of type A3 of the A0 classification is difficult because of lateral dislocation of the proximal fractured segments, particularly when only the sliding hip-screw is used for fixation. A connectable butt-press plate was recently developed in order to prevent this type of dislocation. We review the results with this fixation technique in 22 elderly patients with an average age of 76 years who presented with highly unstable trochanteric fracture of the A3 type. Three patients died of diseases unrelated to the trauma or operation before the fractures had healed. The other 19 were followed prospectively until fracture healing had occurred. Complaints, leg shortening and changes in the CCD angle were recorded. Furthermore, the mobility score was determined. Although the patients were able to bear full weight after the operation, no lateral dislocation of fragments was observed. Only 1 patient had a varus dislocation of 5 degrees until the fracture had consolidated. This was due to severe osteopenia and subsequent dislocation of the screw within the femoral head. No pseudarthrosis, osteitis or rotational malalignment was noted. Five of the surviving patients had a lower mobility score after fracture healing as compared to the status before the fracture was sustained. On the basis of this review, we recommend the use of this new connectable buttress plate with sliding hip screws because it provides sufficient fixation of highly unstable fractures of type A3.     

Monteggia fractures in adults

The records concerning ten consecutive years of experience with Monteggia fractures in adult patients at a level-one trauma center were retrospectively reviewed. Forty-eight patients who had been followed for a minimum of two years (average, 6.5 years; range, two to fourteen years) were identified. There were twenty-five women and twenty-three men, and the average age was fifty-two years (range, eighteen to eighty-eight years). According to the classification of Bado, there were seven type-I, thirty-eight type-II, one type-III, and two type-IV injuries. Twenty-six patients (68 percent) who had a Bado type-II fracture had an associated fracture of the radial head; ten of these patients also had a fracture of the coronoid process as a single large fragment. The ulna was fixed with a tension band-wire construct supplemented with screws in three patients (all of whom had a Bado type-II fracture). An ulnar diaphyseal fracture was fixed with an intramedullary Steinmann pin in one patient. The remaining patients had fixation with a plate and screws. The fracture of the radial head was treated with either complete or partial excision of the fragments in twelve patients (with replacement with a silicone prosthesis in two), open reduction and internal fixation in ten patients, and no intervention in four patients. Nine patients, all of whom had a Bado type-II fracture, needed a reoperation within three months after the initial operation; five had revision of a loose ulnar fixation device, three had resection of the radial head, and one had removal of a wire that had migrated from the radial head into the elbow articulation. Other important complications included proximal radioulnar synostosis in three patients, ulnar malunion in three, posterolateral rotatory instability of the ulnohumeral joint in one, and instability of the distal radioulnar joint in one. At the most recent follow-up examination, which was performed after all of the reoperations and reconstructive procedures had been done, the average score according to the system of Broberg and Morrey was 86 points (range, 15 to 100 points). The result was excellent for eighteen patients, good for twenty-two, fair for two, and poor for six. Six of the eight patients who had an unsatisfactory (fair or poor) result had had a Bado type-II fracture with a concomitant fracture of the radial head. These unsatisfactory results were related to a malunited fracture of the coronoid process in two patients, a proximal radioulnar synostosis in one, a malunited fracture of the coronoid process and a proximal radioulnar synostosis in one, a malunion of the ulna in one, and painfully restricted rotation of the forearm after operative fixation of a comminuted fracture of the radial head in one. The other two unsatisfactory results were in a patient who had had a Bado type-I fracture and in one who had had a Bado type-IV fracture. The results of the present series are much better than those reported in most earlier studies, suggesting that stable anatomical fixation of the ulnar fracture (including associated fracture fragments of the coronoid process) with a plate and screws inserted with use of current techniques of fixation leads to a satisfactory result in most adults who have a Monteggia fracture. The posterior (Bado type-II) fracture is the most common type of Monteggia fracture in adults. Problems with the elbow related to fractures of the coronoid process and the radial head, which are common with Bado type-II Monteggia fractures, remain the most challenging elements in the treatment of these injuries.     

Confirmed abuse cases in public residential facilities for persons with mental retardation: a multi-state study

Our experience with 10 patients who suffered midshaft fractures of the femur and injuries to their ipsilateral hips has resulted in a treatment protocol that we believe can avoid unnecessary complications. The surgical protocol is a three-stage procedure based on (1) intramedullary nailing of the femur with interlocking for preserving the anatomic length and rotation of the femur; (2) treatment of the hip joint injury by means of open reduction and internal fixation of the acetabular fracture and/or reduction and fixation of the fractured femoral neck; and (3) repair of the extensor mechanism. Eight patients received this treatment and had fast recoveries. Union of the femoral fracture and full range of motion of the knee joint were observed within 3 months. In contrast, two patients who had received different surgical treatment had incomplete functional recoveries. This study offers a treatment protocol for ipsilateral disruption of the extensor mechanism, hip joint injury, and midshaft fracture of the femur in the multiply injured patient that can achieve full recovery with no complications. This relatively rare combination of injuries is definitely worthy of special attention.     

Open reduction and internal fixation of forearm fractures in children

We retrospectively reviewed 16 children younger than 13 years with 17 fractures of the shafts of the radius or ulna or both who had undergone an open reduction-internal fixation (ORIF). ORIF was performed when a closed reduction was deemed unacceptable in 14 radius fractures and for three unstable open fractures of the radius. The average age was 9.4 +/- 2.3 years (range, 5.0-12.5). Of the 14 fractures with an unacceptable closed reduction, soft-tissue interposition was encountered in seven. Fixation was secured by plates and screws, percutaneous Steinmann pins, or intramedullary Steinmann pins. There were no delayed unions or nonunions, no infections, and no neurovascular injuries. The average follow-up was 12.3 months; all 17 fractures had excellent results (forearm rotation loss of < 10 degrees). Our study indicates that excellent results can be expected with no increased risk of complications if the treating physician elects to proceed with an ORIF in a pediatric forearm fracture with proper indications.     

Anterior cruciate ligament reconstruction. Bone-patella tendon-bone versus semitendinosus anatomic reconstruction

In this article, the long term (2-10 years; mean, 4.8 years) followup results of two reconstructive procedures for the anterior cruciate ligament are compared. The bone-patella tendon-bone (with interference fit fixation) was performed on 69 knees, and the semitendinosus anatomic reconstruction was performed on 68 knees, in a population of 76 men and 52 women (age range, 15-60 years; average, 31 years). The patients in the two groups showed no difference in subjective results or activity level and no significant difference to manual testing. The semitendinosus procedure group had a slightly higher KT manual maximum failure rate than the patella tendon group (17% versus 11%). Arthrometric stability did not show deterioration, but patient satisfaction decreased in those patients who had meniscectomies. Both procedures showed satisfactory results during the long term followup. However, if the secondary restraints are compromised, the stiffer bone-patella tendon-bone construct is preferred for reconstruction of the chronic anterior cruciate ligament deficient knee.     

Hemispheric titanium porous coated acetabular component without screw fixation

One hundred fifteen hips in 108 patients with primary total hip arthroplasty using the anatomic porous replacement hemispheric acetabular component implanted without adjunctive screw fixation had a mean postoperative followup time of 6 years (range, 5-7.4 years). Clinical evaluation was performed using the Harris hip score and patient self assessment using a modified Short Form-36 questionnaire. Radiographs were measured for radiolucent lines, polyethylene wear, osteolysis, migration, and fractures. No acetabular metal shell had been revised for loosening or was radiographically loose with or without migration (more than 3 mm) at final followup. Reoperation was done in nine (8%) hips because of polyethylene insert wear or disassembly. No fracture of the acetabular bone occurred at the time of surgery or was observed on radiograph. Fixation of the metal shell was stable, with progressive radiolucent lines observed at final followup in 2% of the hips. Osteolysis was recorded in one patient with two acetabular components. The fixation of noncemented hemispheric porous coated acetabular components is more related to the technique of acetabular bone preparation and press fit implantation than to whether additional screws or peg fixation are used. Fixation of this acetabular component without screws at an average of 6 years after surgery is reproducible and predictable in primary hip arthroplasty. The design of modular polyethylene inserts has been improved and should reduce the wear rate of reoperations of the polyethylene insert.     

Traditional methods vs rigid internal fixation of mandible fractures

OBJECTIVE: To compare traditional methods (ie, intermaxillary fixation with interosseous wiring or external fixation) with newer techniques (ie, plating, use of lag screws) of open reduction and fixation of mandible fractures. DESIGN: Retrospective analysis of data from medical records. SETTING: Academic urban medical center. PATIENTS: Nonrandomized sample of 356 patients admitted to the hospital for treatment of mandible fractures from 1987 through 1991; 155 patients treated with open reduction and fixation were studied. INTERVENTIONS: Sixty-nine patients were treated with interosseous wire fixation or external fixation, 86 patients with rigid internal fixation. MAIN OUTCOME MEASURES: Presence of infection, nerve impairment, nonunion, malunion, operative time, and follow-up. RESULTS: No significant difference was noted between the two groups for sex, treating service, delay in presentation, antibiotic coverage, mechanism of injury, or type of fracture. The incidence of infection, nerve injury, and unavailability for follow-up were greater in patients treated by the newer techniques. Overall expense and operative time were greater in the group treated with plates and lag screws. CONCLUSIONS: We advocate traditional techniques for patients with mandible fractures requiring open reduction and fixation.     

2-stage reimplantation for infected total knee replacement

Between 1977 and 1983, 64 infected total knee replacements in 60 patients were treated with a 2-stage protocol for reimplantation. The clinical results and survivorship were determined at an average followup of 7.5 years (range, 2-17 years). Surgical protocol included removal of the prosthesis and all cement with thorough debridement of bone and soft tissues. This was followed by 6 weeks of parenteral antibiotics and then reimplantation of a new prosthesis. At followup, 6 knees (9%) had become reinfected, but only 2 with the same organism. Four knees had been revised: 3 for aseptic loosening and 1 for a periprosthetic femur fracture. Two other knees were impending failures due to aseptic loosening. The average Hospital for Special Surgery knee score was 78 points with 18 excellent, 28 good, 13 fair, and 5 poor results. Seventy-eight percent of patients were satisfied with the overall result of their reimplanted prosthesis and 95% think that they made the right decision in undergoing total knee arthroplasty. The 10-year predicted survivorship of 2-stage reimplantation is 77.4%. Two-stage reimplantation with a 6-week course of parenteral antibiotics is an effective means to eradicate deep infection and to provide a functional knee. The long term functional results and overall survivorship are comparable with the results of revision of aseptic failed total knee arthroplasty.     

Operative treatment of fractures of the tibial plafond. A randomized, prospective study

We performed a randomized, prospective study to compare the results of two methods for the operative fixation of fractures of the tibial plafond. Surgeons were assigned to a group on the basis of the operation that they preferred (randomized-surgeon design). In the first group, which consisted of eighteen patients, open reduction and internal fixation of both the tibia and the fibula was performed through two separate incisions. An additional patient, who had an intact fibula, had fixation of the tibia only through an anteromedial incision. The second group consisted of twenty patients who were managed with external fixation with or without limited internal fixation (a fibular plate or tibial interfragmentary screws). Ten (26 per cent) of the thirty-nine fractures were open, and seventeen (44 per cent) were type III according to the classification of Rüedi and Allg?wer. There were fifteen operative complications in seven patients who had been managed with open reduction and internal fixation and four complications in four patients who had been managed with external fixation. All but four of the complications were infection or dehiscence of the wound that had developed within four months after the initial operation. The complications after open reduction and internal fixation tended to be more severe, and amputation was eventually done in three patients in this group. At a minimum of two years postoperatively (average, thirty-nine months; range, twenty-five to fifty-one months), the average clinical score was lower for the patients who had had a type-II or III fracture, regardless of the type of treatment. With the numbers available, no significant difference was found between the average clinical scores for the two groups. All of the patients, in both groups, who had had a type-II or III fracture had some degree of osteoarthrosis on plain radiographs at the time of the latest follow-up. With the numbers available, there was no significant difference between the two groups with regard to the osteoarthrotic changes. We concluded that external fixation is a satisfactory method of treatment for fractures of the tibial plafond and is associated with fewer complications than internal fixation.     

Indications for surgical treatment of tibial condyle fractures

The residual radioanatomic changes influencing the functional, subjective, and clinical outcome of 131 tibial condyle fractures were studied. Clinical function was found to deteriorate rapidly with increasing values of residual medial tilt of the tibial plateau, whereas lateral tilt of the plateau was well tolerated up to 5 degrees. Articular step-off up to 3 mm and condylar widening up to 5 mm had no adverse effects. Seventy percent of knees with moderate or severe instability were functionally unacceptable. It was concluded that a medial unicondylar fracture with any displacement, and all medially tilted bicondylar fractures, should be operated upon. In fracture of the lateral condyle, open reduction and internal fixation is indicated when lateral tilt or valgus malalignment exceeds 5 degrees, articular step-off exceeds 3 mm, or condylar widening exceeds 5 mm. The same limits apply to laterally tilted bicondylar fractures, provided that the medial condyle is undisplaced. Any displacement seen in the axial bicondylar fracture is an indication for surgical treatment. If there is any mediolateral instability in the extended knee joint after rigid internal fixation, repair of a collateral ligament should be considered. An avulsed anterior cruciate ligament should be fixed, if pathologic laxity exists, but the torn ligament can be ignored and reconstructed later if needed.     

Management of late induced abortions in France

OBJECTIVE: To evaluate the clinical results of comminuted patella fracture fixation after an extensile surgical approach by using a tibial tuberosity osteotomy. DESIGN: Prospective, clinical. PATIENTS: Six knees with displaced comminuted patella fractures had stable internal fixation after an osteotomy of the tibial tubercle. All had immediate postoperative continuous knee motion and were followed for an average of thirty-one months (minimum of eighteen months). OUTCOME MEASURES: Time to clinical and radiographic union, Hospital for Special Surgery (HSS) Knee Scores and comparisons with literature cohort studies. RESULTS: Clinical union of the osteotomy occurred at an average of eight weeks (range 6 to 12 weeks) and of the patella fractures at an average of eleven weeks (range 8 to 21 weeks). There was no radiographic evidence of osteotomy displacement, fracture implant loosening, migration, or failure. All patients had clinical residua, which included loss of motion, thigh muscle atrophy, and barometric complaints. HSS Knee Scores averaged 75 points with four good, one fair, and one poor result. These results were comparable to those of previously published reports of ablative surgery for this type of fracture. CONCLUSION: Comminuted patella fractures are severe injuries that usually result in some lingering morbidity. Internal fixation preserves bone stock, which facilitates future reconstructive procedures. The described tibial tuberosity osteotomy, patella eversion technique of fracture exposure improved visualization and reduction of the articular surface. The osteotomy healed in all cases and did not adversely affect the clinical results.     

Closed management and percutaneous fixation of unstable proximal humerus fractures

BACKGROUND: In dealing with displaced proximal humerus fractures, there is still much controversy in treatment modalities. The latest investigations emphasize the concept of minimal exposure and rigid fixation. METHODS: The technique of closed reduction and percutaneous fixation with cannulated screws and k-pins was performed on 19 patients with two- and three-part proximal humerus fractures. The outcomes were evaluated with a mean follow-up of 21 months. RESULTS: All except one case had a solid union of the fracture. Sixteen of 19 patients (84%) acquired good or excellent results according to Neer's classification. No further collapse or avascular necrosis was found. CONCLUSION: The method of closed reduction and percutaneous fixation, although technically demanding, yields satisfactory results in displaced proximal humerus fracture. Cannulated screws provided rigid fixation that was the underlying tenet for early functional retrieval.     

Operative treatment of displaced proximal humeral fractures: two-year results in 99 cases

The operative treatment of comminuted and displaced fractures of the proximal humerus has been evolving in recent years. Classical open reduction and internal fixation techniques have an increased risk of avascular necrosis. Minimal osteosynthesis procedures often result in a suboptimal fracture reduction and require postoperative immobilization of the arm in some cases. This study reviewed ninety-nine out of 142 patients (70%), an average of 30 months (range 12 to 72 months) after indirect reduction and internal fixation of two-, three- or four-part fractures of the proximal humerus. The surgical procedure includes indirect fracture reduction with no manipulation of the different fracture fragments and subsequent buttress-plate fixation, using a limited deltopectoral approach. Mean age of patients was 63 years (range 17 to 85 years). Twenty percent of patients had associated lesions. Five patients presented with fracture-dislocations. Results were, according to the UCLA- and the Constant-rating system good to excellent in 76 and 69% of cases. Twelve patients had a poor functional outcome. The indirect reduction technique reduces the opening of the fracture site to minimum and thereby limits the risk of iatrogenic damage to local vascularity and the rotator cuff. Complete and partial humeral head necrosis developed in 3% and 1% of cases respectively. Non-union occurred in one case. Plate fixation is an adequate procedure for treating unstable and displaced two- to four-part fractures of the proximal humerus. The incidence of avascular necrosis and non-union are low, when fracture reduction is performed indirectly. Plate fixation enables an early functional treatment, with no need for postoperative immobilization.     

Initial heme uptake from albumin by short-term cultured rat hepatocytes is mediated by a transport mechanism differing from that of other organic anions

Twenty-eight proximal phalangeal fractures secondary to low-velocity gunshot wounds in 27 patients treated by stable fixation were retrospectively reviewed. Definitive fixation was performed within 1 week of injury. Fractures were stabilized with either a plate, intramedullary spacer, or a combination of both. When necessary, supplemental fixation was achieved with cerclage wires or interfragmentary screws. Twenty fractures with bone loss or comminution were primarily supplemented with iliac crest bone graft. After surgery, the fingers were splinted in 90 degrees of metacarpophalangeal (MP) flexion. An aggressive supervised therapy program was initiated within 24 hours of surgery. The average length of follow-up care was 9 months (range, 3-29 months). Primary union was achieved in all fractures. The average range of motion was 83 degrees for the MP joint and 66 degrees for the proximal interphalangeal joint. The average total active motion (TAM) for the involved digits was 200 degrees (range, 65 degrees-250 degrees). Fractures without intra-articular extension had a significantly better average TAM (213 degrees) than did those with intra-articular extension (169 degrees; p = .05). Primary bone grafting did not adversely effect the final TAM. There were no infections. Early stable fracture fixation of these injuries achieved union, alignment, and early rehabilitation with no appreciable increase in morbidity.     

Substrate specificity of ascidian sperm trypsin-like proteases, spermosin and acrosin

We retrospectively reviewed the office records of the senior author--which include two national ballet companies--and identified 35 dancers who sustained distal shaft fractures of the fifth metatarsal. The usual fracture pattern is a spiral, oblique fracture starting distal-lateral and running proximal-medial. Treatment consisted of open reduction and internal fixation for 2 patients, closed reduction and percutaneus fixation for 2 patients, short leg weightbearing cast for 7 patients, and an elastic wrap and treatment of symptoms for 24 patients. Patients with marked displacement of the fracture underwent internal fixation early in the study period; but more recent treatment emphasized nonoperative means, even for displaced fractures. The average time to pain free walking was 6.1 weeks (range, 0 to 16); return to barre exercises, 11.6 weeks (range, 4 to 48); and return to performance, 19 weeks (range, 6 to 52). There was one delayed union (7 months) and one refracture (2 months) that subsequently healed. All patients returned to professional performance without limitation and no patient reported pain with performance at followup. Spiral fractures of the distal shaft of the fifth metatarsal are common injuries and can usually be treated nonoperatively for these high performance athletes without long-term functional sequelae.     

Placental restriction alters adrenal medullary development in the midgestation sheep fetus

This is a retrospective study of 10 patients with combined cruciate ligament and posterolateral instability who underwent surgical reconstruction between 1991 and 1994. All knees had at least 20 degrees increased external rotation at 30 degrees of knee flexion and from 1+ to 3+ varus instability. Five knees with posterior cruciate ligament ruptures had at least a 2+ Lachman test result. (One knee had both anterior and posterior cruciate ligament injuries). In all cases the lateral collateral ligament was reconstructed with a bone-patellar tendon-bone allograft secured with interference screws. Fixation tunnels were placed in the fibular head and at the isometric point on the femur. The cruciate ligaments were reconstructed with autograft or allograft material. The average follow-up was 28 months. Excessive external rotation at 30 degrees of flexion was corrected in all but one knee. Six patients had no varus laxity, and four patients had 1+ varus laxity at 30 degrees of flexion. The posterior drawer test result decreased, on average, to 1+, and the Lachman test result decreased to between 0 and 1+. The average Tegner score was 4.6, with five patients returning to their preinjury level of activity and four returning to one level lower. These results indicate that this is a promising new procedure for patients with instability resulting from lateral ligament injuries of the knee.     

A comparison of results in acute and chronic anterior cruciate ligament ruptures of arthroscopically assisted autogenous patellar tendon reconstruction

We conducted a prospective study of 94 consecutive patients who received a patellar tendon autograft for anterior cruciate ligament rupture. Eighty-seven patients (93%) returned for followup a mean of 28 months postoperatively; 57 had chronic and 30 had acute or subacute ruptures. There were no significant differences between the subgroups for age, sex, articular cartilage lesions, or months of followup. Forty-six meniscal tears were repaired; 27 of these extended into the central avascular region. Rehabilitation emphasized immediate knee motion, but strenuous activity was delayed for at least 4 months. Only one patient had a knee motion complication, and stability (<3 mm, KT-2000 arthrometer, 134 N) was restored in 85% of knees with chronic ruptures and 92% of knees with acute ruptures. Earlier reconstruction should be considered in active persons as symptoms and limitations continued postoperatively in knees with chronic ruptures, leading to overall less satisfactory results. In patient rating of the overall knee condition, 69% of knees with chronic ruptures and 100% of knees with acute ruptures scored in the normal or very good range. Repair of meniscal tears that extend into the central avascular region should be considered, as 24 of the 27 (89%) menisci repaired showed clinical evidence of healing and did not require reoperation.     

Clinical and functional outcomes of internal fixation of displaced pilon fractures

The clinical and functional outcomes for patients treated with open reduction and plate fixation of displaced tibial pilon fractures were determined. A retrospective search of the authors' trauma database was conducted for AO and Orthopaedic Trauma Association Code 43 injuries (pilon fractures) in adults 18 years or older who were treated between December 1988 and December 1992. The group of 64 patients who required open reduction and internal fixation to treat their fractures make up the primary cohort for this analysis. Twenty of these cases required no fibular fixation; the remainder were mostly fixed with 1/3 tubular or 3.5-mm compression plates. Tibial fixation was done using most commonly 3.5-mm cloverleaf plates, 1/3 tubular plates, or both. Of the 64 patients treated with open reduction and internal replacement, four (5%) patients had deep infection develop. Two (7%) of 14 patients had open fractures, and two (4%) of 50 patients had closed fractures. Three of these four patients smoked tobacco products; one was also an intravenous drug abuser. Staphylococcus aureus was the organism in two cases; Enterobacter, in the other two. The infection was controlled with a free flap in two cases, with antibiotics and wound debridement in one and with an arthrodesis in one. Thirty of the 64 patients completed the Short Form-36; two of these patients had bilateral fractures. The study group had significant differences in general health perceptions, physical function, physical role function, emotional role function, social and mental function, and pain and energy levels when compared with age matched population data and patients with tibial plateau fractures. The effect of other injuries on these functional status results cannot be determined specifically.     

Distal realignment of the patellar tendon to correct abnormal patellar tracking

Between January 1980 and January 1994, 31 knees required distal realignment of the extensor mechanism to treat lateral patellar subluxation that could not be corrected with lateral patellar release and vastus medialis advancement during total knee arthroplasty. Fifteen had a preoperative valgus angle of more than 12 degrees, and 16 were undergoing revision total knee arthroplasty. Ten knees had a modified Roux-Goldthwait procedure, 18 had medial tibial tubercle transfer, and three had medial transfer of the medial 1/2 of the patellar tendon. The length of followup ranged from 2 to 16 years. No late patellar subluxations or dislocations have occurred in any of these cases. Three cases of medial tibial tubercle transfer had hematomas develop, with two requiring surgical evacuation; one of these developed a late infection. No fractures or displacements of the tubercle fragment have occurred. No significant patellar complications have occurred in those patients who underwent the modified Roux-Goldthwait procedure or the medial transfer of the medial 1/2 of the patellar tendon. One year after surgery, the mean knee flexion was 113 degrees, four knees had a flexion contracture of 5 degrees, and none had a quadriceps lag.