Induced membrane technique using enriched bone grafts for treatment of posttraumatic segmental long bone defects.

BACKGROUND: Reconstruction of posttraumatic bone defects represents a difficult challenge. The induced membrane technique is an effective two-stage procedure for bone defect reconstruction. To overcome the problems of autologous bone grafting, different graft substitutes have been investigated. The aim of the present study is to evaluate our clinical experience in reconstruction of critical posttraumatic bone defects using an induced membrane technique based on a combination of autologous graft and allograft (cancellous bone) enriched with platelet-rich plasma (PRP) and bone marrow concentrate aspirate (BMCA). MATERIALS AND METHODS: Between 2009 and 2014, we reconstructed 18 posttraumatic bone defects in 16 patients. Their average length was 6.4 cm (range 1.6-13.2 cm). The defect location was the femur in nine cases (50%), the tibia in eight (44%) cases, and the humerus in one (6%) case. In all cases, we used a combination of autologous and cancellous allograft graft enriched with PRP and BMCA. Bone fixation was achieved using intramedullary nailing in 2 cases (11%), plating in 15 cases (66%), and external fixation in 1 case (6%). RESULTS: Both clinical and radiographic union were achieved in 13 (72%) cases (13 patients). Five (28%) cases (four patients) developed nonunion. Nonunion was observed in two of eight (25%) tibial defects and in three (33%) of nine femoral defects (ns). Three of 4 (75%) double defects had delayed union, whereas 2 of 14 (14%) single defects did not heal (p = 0.016). The average length of the 13 defects that united was 6 cm (range 1.6-11.8 cm), while the length of the 5 defects that did not unite was 10.3 cm (range 6-13.2 cm) (p = 0.009). CONCLUSIONS: In this series using an induced membrane technique based on a combination of autograft and allograft enriched with BMCA and PRP, the healing rate was lower than in other series where autologous bone graft alone was employed. Nonunion was more frequent in longer and double defects. Further research aimed at developing effective alternative options to autogenous cancellous bone graft is desirable. LEVEL OF EVIDENCE: III.

Postsurgical osteomyelitis caused by Enterobacter sakazakii in a healthy young man.

Enterobacter sakazakii, a Gram-negative rod-shaped bacterium, is a rare cause of invasive infections (meningitis, sepsis, necrotizing enterocolitis) with high death rates (40-80%), primarily in newborns. In contrast to the high number of cases in newborns, infants and children, there are only a few reported cases of E. sakazakii infections in adults, generally in subjects with pre-existing conditions such as neoplasms, and just one osteomyelitis of the foot. We report a confirmed case of postsurgical osteomyelitis of the femur caused by E. sakazakii in a young otherwise healthy man.

Lateral collateral ligament reconstruction using a semitendinosus graft.

We evaluated 13 reconstructions of the lateral collateral ligament (LCL) of the knee associated with a reconstruction of the ACL (n=6) and PCL (n=7). In all cases the LCL was reconstructed using a semitendinosus tendon graft through a tunnel in the fibular head and fixed in anatomical insertion of LCL at lateral femoral condyle. Patients were evaluated using the IKDC form. Lateral joint opening and posterior tibial translation were studied by radiographic stress views. The follow-up average was of 60 months (38-93). In the ACL group five patient were symptoms free while one reported moderate pain after strenuous activity. The lateral stress radiographs showed lateral joint opening of 0-2 mm in five knees, and 3-5 mm in one. In the PCL group four patients were symptom free while three complained mild to moderate pain or swelling. Lateral joint opening was 0-2 mm in six knees and 3-5 in one. External rotation at 30 degrees of flexion was 10 degrees in one and between 0-5 degrees in the remainder. Eleven of 13 patients returned to their preinjury level and two returned to one level lower. These findings indicate that the LCL can be successfully reconstructed with a free ST graft at the time of ACL or PCL reconstruction.

Mobile and fixed bearing total knee prosthesis functional comparison during stair climbing.

OBJECTIVE: The purpose of this investigation is to determine the functional performance of the mobile bearing total knee replacement prosthesis as compared to the fixed bearing type total knee replacement prosthesis. DESIGN: Kinematics, kinetics, and electromyography data were gained from 10 patients with mobile bearing and 10 patients with a fixed bearing posterior stabilized Insall Burstein II total knee replacement during ascending and descending stairs. A control group of 10 normal subjects, matched by sex and age, was also analysed. BACKGROUND: No significant biomechanical differences in patients with different total knee replacement designs have been reported from level-walking studies. Slightly better performance of posterior retaining with respect to cruciate sacrificing total knee replacement designs have been claimed from stair climbing studies. Only one study has been conducted regarding mobile versus fixed bearing total knee replacement assessed by gait analysis. This study did not show any biomechanical differences between the two groups. METHODS: Motion analysis was used to quantify the knee kinematics, kinetics, and electromyography (right and left longissimus dorsi, gluteus medius, rectus femoris, biceps femoris, semitendinosus, gastrocnemius and tibialis anterior muscles) during stair ascent and descent. RESULTS: The mobile bearing group demonstrated a reduced knee extensor moment during stair climbing and descending, and a reduced knee adductor moment during stair climbing. When ascending stairs, most of the mobile bearing patients show a peak knee flexion and a peak knee flexion moment at the late stance phase during the double support period. This kinematic and kinetic pattern is absent in normal subject. Both mobile bearing and fixed bearing groups showed abnormal electromyography patterns in both descending and ascending. CONCLUSIONS: During stair climbing, the mobile bearing design demonstrates a different kinematic pattern to the fixed bearing total knee replacement. Lower limb compensatory mechanisms seemed to be adopted particularly by the mobile bearing patients during ascending stairs. RELEVANCE: Total knee replacement patient with mobile bearing design can feel excessive femoro-tibial motion during daily living activities such as stair climbing and descending. Proprioceptive control of this tibio-femoral translation is needed as demonstrated by the lower limb compensatory mechanism. This data suggest that antero-posterior constraint structures (ligamentous or mechanical) are important to obtain reproducible knee kinematics.

The Insall-Burstein II prosthesis: a 5- to 9-year follow-up study in osteoarthritic knees.

We prospectively studied 100 Insall-Burstein Posterior Stabilized II total knee arthroplasties (Zimmer Inc, Warsaw, IN) consecutively implanted in 91 patients with osteoarthritis. We reviewed 92 knees in 85 patients at a mean follow-up of 7.5 years (range, 5.3-9.5 years). According to the Knee Society score, 78 (85%) results were excellent, 11 (12%) were good, and 3 (3%) were fair or poor. The unsatisfactory results were attributed to patellofemoral pain with a low patella (2 knees) and to valgus alignment and medial laxity (1 knee). Flexion at follow-up averaged 116 degrees. Nonprogressive radiolucent lines were present around 30% of the tibial components. There was only 1 small osteolytic lesion around the tibial plateau. We observed no aspetic loosening, infection, or patella stress fracture. Survivorship analysis in the worst-case scenario showed a 90.9% success rate at 8 years.

Patella resurfacing in total knee replacement: functional evaluation and complications.

Three different knee replacements, with three trochlear designs, were prospectively evaluated clinically and radiographically for patellar function and presence of patellar complications. They included the Insall-Burstein (IB) I and the IB-II (posterior cruciate ligament substituting) and the Meniscal Bearing Knee (MBK; posterior cruciate ligament recession). The trochlea of the IB-I was short and shallow with an anterior sharp edge of the intercondylar box (later modified to a smoother edge) and the femoral component had a prominent "shoulder." In the IB-II the trochlea was deeper to allow for soft tissue clearance. In the MBK the trochlea was more prolonged, with R and L components and the "shoulder" was less prominent. In all the cases the patella was resurfaced with an all polyethylene dome prosthesis. Knees with tibiofemoral problems were excluded. From the data of the present study the following conclusions can be drawn: (a) The most frequent problem was impingement (clunks) with the early version of the IB-I. Smoothening of the anterior edge significantly reduced the incidence of clunks to 5% in the modified IB I. (b) With the IB-II deepening the trochlea for soft tissue clearance improved the degree, not the incidence of clunks (4.5%), compared to the modified IB I. (c) With the MBK clunks were very rare and patellar function improved. (d) Throughout the three series patellar stress fractures and instability were rare and loosening or wear not evident. (e) Normal function (including stairs ascending and descending) can be expected in over 80% of category A patients. (f) Of the various radiological parameters only patella baja was correlated with symptoms in the IB prostheses. (g) We still prefer the dome design because is more tolerant and with cold flow may better conform to the trochlea increasing contact area.

Clastogenic effects of two tyrosine kinase inhibitors, tyrphostin 23 and tyrphostin 46, on a transformed (CHO-K1) and on a primary embryonic chinese hamster cell line (CHE).

Protein tyrosine kinases (PTKs) play fundamental roles in signal transduction pathways. Many proliferative diseases are characterized by deregulation of PTK activity, therefore PTKs appear as promising targets in the design of anticancer drugs. Tyrphostins are a family of synthetic compounds which efficiently target specific PTKs without competing for ATP and thus are much less cytotoxic with respect to conventional therapeutic agents. We tested two tyrphostin derivatives, Tyrphostin 23 and Tyrphostin 46, on a transformed (CHO-K1) and on a primary embryonic Chinese hamster cell line (CHE) to determine whether these compounds had a genotoxic effect. We found that the tyrphostins increased sister chromatid exchange frequency in both cell lines, but induced chromosomal aberrations only in the transformed CHO-K1 cell line when treatment was in the S phase of the cell cycle, and not in primary CHE cells. Such a result could have important therapeutic implications: it could mean that deregulation of signal transduction pathways in cells which already have a deficit in cell cycle control could cause chromosomal aberrations.

Femoral tunnel position in anterior cruciate ligament reconstruction using three techniques. A cadaver study.

The possibility of achieving correct deep femoral tunnel positioning during anterior cruciate ligament (ACL) reconstruction with the double incision technique (DI), the transtibial technique (TT), and the anteromedial technique (AM) was evaluated in 30 cadaver knees. A reference hole was made just deep to the insertion of the anteromedial bundle of the ACL through an anteromedial arthrotomy. In the DI technique, a Kirshner wire was inserted outside-in using a rear entry C guide. In the TT and AM techniques, the K-wire was inserted inside-out through the tibial tunnel and through the arthrotomy, respectively. The reference hole could be achieved with each technique. Using lateral radiographs, the superficial aspect of the intra-articular exit of the femoral tunnel was found to be located on average at 36%, 36%, and 34% of the width of the condyles from the posterior margin (NS). None of the holes was more anterior than 40%. In conclusion, a deep femoral tunnel positioning could be achieved with each technique. The choice of technique must be based on the surgeon's preference and clinical results.

The Insall-Burstein total knee replacement in osteoarthritis: a 10-year minimum follow-up.

A total of 99 Insall-Burstein posterior stabilized (IBPS) knee replacements were implanted in 86 osteoarthritic patients. We reviewed 60 knees with a 10- to 15-year follow-up. Using the Knee Society score, 35 knees (58%) were excellent, 15 (25%) good, 4 (7%) fair, and 6 (10%) poor. Flexion at follow-up was 106 degrees on average. Moderate patellofemoral crepitation was present in 5 knees (9%), and 11 knee required excision of a synovial nodule proximal to the patella. Radiographic analysis showed 5 osteolytic lesions (8%) around well-fixed tibial and femoral components and minimal (1 mm) narrowing of the medial polyethylene thickness in 7 knees (12%). There were 6 (10%) failures requiring reoperation because of aseptic loosening (4 knees), deep infection (1 knee), and recurrent patellar dislocation (1 knee). Survivorship analysis using revision as the endpoint showed a cumulative success rate of 92% at 10 years. In this study, the IBPS knee has shown good long-term results with low rates of aseptic loosening and no failures attributable to polyethylene wear.

The relationship between the intercondylar roof and the tibial plateau with the knee in extension: relevance for tibial tunnel placement in anterior cruciate ligament reconstruction.

The relationship between the intercondylar roof and the tibial plateau with the knee in full extension was studied in 100 patients with a unilateral anterior cruciate ligament (ACL) lesion. A lateral view of both knees in full extension with superimposition of the femoral condyles was obtained with the fluoroscope. We found that, in the normal knee, the roof line intersects the tibial plateau (roof-plateau intersection ratio) at 31.5% (SD +/- 5%) of its width and at 33.9% (+/- 5.4%) in the ACL deficient knee (P < .001). The difference was attributable to a subtle anterior tibial displacement in the ACL-deficient knee. Multiple regression analysis showed a direct relationship between roof-plateau intersection ratio and the angle between roof line and tibial plateau (P = .0006). A direct relationship of borderline significance (P = .06) was present with the knee recurvatum measured clinically. In conclusion, the roof-plateau intersection ratio has a wide range of variability (22% to 41%). The larger the roof-plateau angle the more posterior the roof-plateau intersection ratio.

Arthroscopic meniscectomy for discoid lateral meniscus in children and adolescents: 10-year follow-up.

Seventeen adolescents underwent arthroscopic lateral meniscectomy for discoid lateral meniscus. The average age at surgery was 13.6 years (range: 5-18 years). The main preoperative symptom was pain in 16 knees and extension loss in 1 knee. At arthroscopy, 10 menisci were complete, 4 were incomplete, and 3 were Wrisberg type. Arthroscopic total meniscectomy was performed in the 3 Wrisberg types, 2 complete types, and 1 incomplete type. The remaining menisci underwent partial meniscectomy. The average follow-up was 10 years (range: 5-15 years). According to the Ikeuchi rating system, 12 knees were rated as excellent (no symptoms and full range of motion), 4 were rated as good (occasional pain), and 1 was rated as fair (patellofemoral pain in an obese patient). Radiographic evaluation showed development of minor osteophytes in the lateral compartment of 8 knees and <50% narrowing of the lateral joint space in 11 knees. No correlation was found between meniscal type, type of meniscectomy (partial or total), and clinical and radiographic results. Arthroscopic lateral meniscectomy for discoid lateral meniscus in adolescents was effective in relieving symptoms during a 10-year follow-up period. Longer follow-up is needed to ascertain the significance of the radiographic changes seen in this study.

Anatomic versus non-anatomic tibial fixation in anterior cruciate ligament reconstruction with bone-patellar tendon-bone graft.

We prospectively compared two series of 30 anterior cruciate ligament (ACL) reconstructions each where the bone-patellar tendon-bone graft was alternately fixed at the level of the tibial plateau (group A; anatomic fixation) or distal to the plateau level (group B; non-anatomic fixation). In group A, a 35-mm-long tibial tuberosity bone block was harvested. The distal 10-15 mm were resected and fixed proximally to the undersurface of the tendon to shorten it. After an average 18 months' follow-up, there were no significant differences between the two groups concerning subjective evaluation, symptoms, range of motion and objective stability. Tibial tuberosity pain was more frequent in group A (53% vs 17%, P = 0.01). Radiographic evaluation showed that tibial tunnel enlargement was less frequent in group A (23% vs 43%, P = 0.02). There was no correlation between tunnel enlargement and objective stability. In conclusion, fixation of the graft at the tibial plateau level did not improve objective stability in this study. Because of the greater technical difficulty and occurrence of tibial tuberosity pain, this technique is not recommended.

Arthroscopic-assisted anterior cruciate ligament reconstruction with the central third patellar tendon. A 5-8-year follow-up.

We reviewed 89 arthroscopically assisted patellar tendon anterior cruciate ligament (ACL) reconstructions for chronic isolated injuries with an average follow-up of 7 years (range 5.4 to 8.6 years). Pain was present in 7 knees (8%). Giving-way symptoms were reported by 7 patients (8%). A KT-2000 side-to-side difference over 5 mm at 30 lbs was recorded in 12 cases (16%). The pivot shift was glide in 17 cases (19%) and clunk in 10 (11%). A 3 degrees-5 degrees extension loss compared with the normal side was present in 20 knees (22%) and 6 degrees-10 degrees in 4 knees (4%). The intra-articular exit of the femoral tunnel was misplaced in the anterior 50% of the condyles along the roof of the notch in 10% of the knees. This positioning significantly (P = 0.003) increased the frequency of graft failure (62.5%) compared with the cases with a more posterior placement (graft failure 12%). An anterior position of the intra-articular exit of the tibial tunnel (in the anterior 15% of the sagittal width of the tibia) significantly (P = 0.01) increased the frequency of extension loss > 5 degrees. Medial meniscectomy was associated with a 35% incidence of narrowing of the medial joint space, which was significantly higher compared with knees with normal menisci (9%; P = 0.04) or with medial meniscal repair (7%; P = 0.05). In conclusion this study showed satisfactory anterior stability (KT-2000 side-to-side difference up to 5 mm and pivot absent or glide) in 83% of the knees. This percentage increases to 88% in the knees with a correct posterior and proximal femoral tunnel placement. Accuracy in tunnel positioning is essential for the success of ACL surgery. Meniscal repair was effective in decreasing joint space narrowing and should be attempted when possible.

Arthroscopically assisted semitendinosus and gracilis tendon graft in reconstruction for acute anterior cruciate ligament injuries in athletes.

We evaluated 69 arthroscopically assisted anterior cruciate ligament reconstructions for acute tears at an average followup of 60 months. We used a distally based single semitendinosus and gracilis tendon graft passed over the top and fixed to the femur. Combined medial collateral ligament lesions were seen in 30 knees, and they were repaired when found in the distal third (18 knees). The patients were instructed to recover motion preoperatively, and an early range of motion program was used postoperatively. At followup, symptoms of giving way were seen in five knees (7%). Graft failure was seen in seven knees (10%); failure was defined as a positive pivot shift (clunk or gross) or a side-to-side difference in anterior tibial displacement greater than 5 mm, as measured with a KT-1000 arthrometer. Permanent extension loss (3 degrees to 5 degrees) was found in two knees (3%). Patellofemoral crepitation was seen in eight knees (12%), but the condition was symptomatic in only one knee. Forty-six patients (67%) were active in pivoting sports before surgery and 37 (54%) remained active in these sports at followup. We concluded that this operation is simple, effective, and has a low complication rate. Further studies are necessary to elucidate if a stronger graft (e.g., a patellar tendon) would decrease the rate of graft failure without increasing complications.

The Insall-Burstein posterior stabilized total knee replacement in the valgus knee.

Sixty-seven Insall-Burstein (Zimmer Inc, Warsaw, Indiana) posteriorly stabilized total knee replacements were implanted in knees with a preoperative alignment > 10 degrees of valgus (average: 19.5 degrees). An anteromedial approach with release of the lateral ligaments of the femur was used. A lateral patellar release was necessary in 49% of cases. Fifty-one knees (76%) were reviewed with an average follow-up of 6 years. Pain at follow-up was absent or mild in all of the knees except one with aseptic loosening of the tibial and patellar components. The average flexion was 105 degrees. The results according to the Knee Society Rating System were excellent in 53%, good in 39%, fair in 6%, and poor in 2%. A patellofemoral "clunk" was present in three knees (6%), and one knee required arthroscopic debridment. The mechanical axis was within 5 degrees of neutral in 88%. A survivorship analysis was performed using failure requiring reoperation as the end point. There were four failures due to aseptic loosening, symptomatic lateral ligamentous instability, deep infection, and recurrent patellar dislocation. The patellar component had not been implanted in the knee with recurrent patellar dislocation due to severe bony erosion. The cumulative success rate was 95% at 10 years. These results indicate that release of the contracted lateral structures and implantation of a posteriorly stabilized prosthesis achieve a stable and well-aligned knee in 66 out of 67 knees. Attention must be paid to patellar tracking and realignment of the extensor mechanism.

Insall-Burstein posterior-stabilized knee prosthesis in rheumatoid arthritis.

The authors reviewed 65 Insall-Burstein (Zimmer, Warsaw, IN) total condylar posterior-stabilized knee prostheses in 50 patients with rheumatoid arthritis with a follow-up period of at least 5 years (range, 5-13 years). Forty-two knees in 31 patients were evaluated using the Knee Society knee and functional rating scores. Radiographic assessment was performed using standing long radiographs (hip to ankle). Radiolucent lines were studied using fluoroscopic-centered views. Excellent or good clinical results were obtained in 95% of the cases, and the average knee score improved from 22.5 to 90 points. No cases of radiologic loosening were observed. Incomplete radiolucent lines around the tibial component were detected in only 17% and were nonprogressive. Two patients developed hematogenous late deep infection, which required removal of the prosthesis in both, followed, at a second stage, by arthrodesis in one and prosthesis reimplantation in the other. Three knees (7%) had a painful impingement of the patella. Two of these were successfully reoperated with arthroscopic debridement of the peripatellar synovial tissues. Survivorship analysis, based on endpoints such as prosthesis removal for any cause or radiologic loosening (complete radiolucent line thicker than 1 mm, tilt, or subsidence of the component), showed a cumulative success rate of 96.2% at 13 years.

Results of surgical treatment of arthrofibrosis after ACL reconstruction.

We prospectively studied 31 knee arthrolyses performed for loss of motion after intra-articular anterior cruciate ligament (ACL) reconstruction. The arthrolysis was performed on average 10.6 months after the reconstruction (range 4-25). Seven knees were localized forms. They were treated with arthroscopic removal of a fibrous nodule and scar tissue anterior to the ACL, which was preserved. Twenty-four knees were global forms and treated arthroscopically (14) or in open procedure (10). Suprapatellar, medial, and lateral gutter adhesions were sectioned, and fibrous tissue was removed from the anterior compartment. A posteromedial and/or posterolateral capsulotomy was necessary in 7 knees. The ACL graft was nonfunctional and/or malpositioned in 19 knees. The results were evaluated with the IKDC form with an average follow-up of 3.5 years (range 1.5-7). Preoperatively the localized forms had an average extension loss of 11 degrees and an average flexion loss of 14 degrees compared to the opposite knee. At follow-up all the knees were satisfactory for symptoms. All except one achieved a satisfactory motion (within 5 degrees of extension loss and 15 degrees of flexion loss) and a satisfactory final result. Global forms had a greater preoperative flexion loss (average 34 degrees) and extension loss (average 17 degrees). At follow-up 58% were satisfactory for symptoms and 71% for arc of motion. However, the final result was satisfactory in only 37%. In conclusion, local forms have a good prognosis. In global forms motion may be improved by surgery, but the final result is downgraded by symptoms. Arthrolyses performed within 8 months from index operation had a better outcome.

Surgical treatment of recurrent dislocation of the patella.

Sixty-seven patients (69 knees) with recurrent patellar dislocation underwent either a lateral release (20 knees), a proximal realignment (14 knees), a distal realignment (16 knees), or a combined realignment (19 knees) procedure. Average respective followup was 8, 8, 6, and 4 years. Patients undergoing lateral release experienced a 40% recurrence of patellar dislocation. After a realignment procedure, redislocation was uncommon (4%), but pain and swelling were reported by 12% of the patients. Significant patellofemoral crepitation was present in 35% of the realignments. The congruence angle was corrected satisfactorily in the proximal realignments, but it remained abnormal in 25% of the distal realignments. The distance from the tuberosity to the sulcus was restored to normal by transposition of the tibial tuberosity, but remained abnormal in 36% of the proximal realignments. However, it did not preclude a good clinical result. Lateral release cannot be recommended for knees with severe dysplasia of the extensor mechanism. Proximal, distal, and combined realignments yielded similar clinical results. Retensioning of the medial structures and lateral release are effective in reducing the patella within the sulcus. Although transposition of the tuberosity is appealing, clinical advantages are less evident.

Arthroscopic drilling in juvenile osteochondritis dissecans of the medial femoral condyle.

Fourteen children affected by juvenile osteochondritis dissecans of the medial femoral condyle (16 knees) were treated by arthroscopic drilling of the lesion. The average age at operation was 12.8 years, and all the knees had open physes. Conservative treatment including restriction of activities had been unsuccessfully attempted in all cases for a period of > 1 year on average. At surgery each lesion was shown to have intact articular cartilage. All the lesions progressed to healing, and the patients were asymptomatic at an average of 56 months after the operation. Radiographic examination showed reconstitution of a normal or minimally flattened profile of the medial condyle. Arthroscopic drilling of the lesion in knees with juvenile osteochondritis dissecans and intact articular cartilage was effective to promote healing. The low morbidity and easy rehabilitation justify the procedure for these young patients when conservative measures fail.

Patellar tendon versus doubled semitendinosus and gracilis tendons for anterior cruciate ligament reconstruction.

The results of intraarticular anterior cruciate ligament reconstruction with either the patellar tendon or the semitendinosus and gracilis tendons (four strands) were prospectively compared in a consecutive series of 60 patients with chronic injuries. A single surgeon performed arthroscopically assisted reconstructions in an alternating sequence. Preoperative and operative data revealed no significant differences between the two groups. After 28 months of followup there were no significant differences in the incidence of symptoms, and recurrent giving way was present in only one knee with semitendinosus and gracilis tendon graft. Return to sport participation was more frequent in the patellar tendon group (80% versus 43%, P < 0.01). A minor extension loss (< or = 3 degrees) was more frequent in the patellar tendon group (47% versus 3%, P < 0.001). Other differences between the two groups were not significant. KT-2000 arthrometer side-to-side difference of anterior displacement > 5 mm at 30 pounds was present in 13% of the knees with patellar tendon grafts and in 20% of those with semitendinosus and gracilis; a patellofemoral crepitation developed in 17% and 3% of the two groups, respectively. Based on these data we routinely use patellar tendon grafts. Semitendinosus and gracilis tendons are preferred in selected cases: older patients, patients with preexisting patellofemoral problems, and those with failed patellar tendon grafts.