 |
 |
 |
|
| home | CME | help | feedback | subscriptions | archive |
|
© 2003 The Journal of Bone and Joint Surgery, Inc.
What's New in Adult Reconstructive Knee Surgery
Michael J. Archibeck, MD The purpose of this update is to discuss, in summary fashion, topics presented at selected orthopaedic meetings and reports published in related orthopaedic journals during 2002. The sources for this review are articles published in The Journal of Bone and Joint Surgery (American edition) and The Journal of Arthroplasty. The podium presentations mentioned in this article include those given at the annual meeting of the American Academy of Orthopaedic Surgeons (held in Dallas, Texas, on February 13-17, 2002), on Specialty Day at the meeting of The Knee Society (held in Dallas, Texas, on February 16, 2002), at the Interim Meeting of the Knee Society (held in Philadelphia, Pennsylvania, on September 19-21, 2002), and at the meeting of the American Association of Hip and Knee Surgeons (held in Dallas, Texas, on November 1-3, 2002). Treatment of Osteoarthritis without Arthroplasty While total knee replacement has been very successful for alleviating pain and improving function in patients of all ages, other approaches remain the initial treatment option for many patients with knee arthritis. Leopold et al., in a randomized study that was not supported by industry, compared intra-articular corticosteroids with Synvisc (hylan GF-20) for the treatment of knee osteoarthritis and found no differences in pain or function at six months, with significantly less response in women relative to men in both groups (p < 0.05). Wai et al., in a review of the results of 14,391 arthroscopic knee débridements performed in patients more than fifty years of age, reported that 18% (1146) of 6212 patients required total knee arthroplasty within three years 1 . They concluded that arthroscopic débridement may be overutilized in elderly patients with osteoarthritis. Gross et al. reviewed the use of fresh osteochondral allografts to treat posttraumatic knee defects measuring >3 cm in diameter (with the grafting procedure combined with osteotomy in cases of malalignment) and found a ten-year survival rate of 85% for the femoral condyle grafts. Only two papers discussed the role of osteotomy in the arthritic knee. Hart and Sekel wrote a review article in support of proximal tibial and distal femoral osteotomies for varus and valgus deformities, respectively. Babis et al., in a review of twenty-nine double-level (distal femoral and proximal tibial) osteotomies of the knee that had been performed to correct varus malalignment and to retain joint-line obliquity, reported a 100-month survival rate of 96%. Unicompartmental Knee Replacement Unicompartmental knee replacement has become increasingly popular because of improved rates of prosthetic survival and the development of less invasive techniques. Fisher et al. compared minimally invasive unicompartmental arthroplasty (eighty-eight knees), open unicompartmental arthroplasty (sixty-four knees), and total knee arthroplasty (fifty-four knees) and found that alignment was most accurate in the knees that had been treated with total knee replacement and that variations in anteroposterior tibial placement and overall alignment occurred more commonly in the knees that had been treated with the minimally invasive unicompartmental procedure than in those that had been treated with the open procedure. Hernigou and Deschamps reviewed patellofemoral complications following ninety-nine unicompartmental knee replacements and found, after an average duration of follow-up of fourteen years, that twenty-nine knees had radiographic patellofemoral osteoarthritic changes and twenty-eight had impingement of the patella on the femoral component 2 . Impingement was more common in knees that had been treated with a lateral unicompartmental arthroplasty, and symptoms were more severe in the knees with impingement. Several investigators reported on the clinical results of unicompartmental knee replacement. Price et al., in a report on 420 patients, found the fifteen-year survival rate of the Oxford Medial Unicompartmental Knee Arthroplasty (Biomet, Warsaw, Indiana) to be 94.3% (95% confidence interval, 85.6% to 100%), with excellent clinical results (mean Hospital for Special Surgery score, 86 points) at ten years. Argenson et al., in a report on 160 Miller-Galante (Zimmer, Warsaw, Indiana) unicompartmental knee replacements, found that the mean Hospital for Special Surgery score was 96 points after a mean duration of follow-up of sixty-six months and that the ten-year survival rate was 94% ± 3% with revision for any reason or radiographic loosening as the end point. Skyrme et al. reviewed the results of twenty-six porous-coated anatomic unicompartmental knee arthroplasties after a mean duration of follow-up of 6.9 years and found a 42% revision rate, with a mean time to revision of thirty-eight months. The most common modes of failure included femoral loosening (55%), polyethylene wear (55%), loosening and wear (72%), and fracture of the femoral component (18%). Perkins and Gunckle, in a review of forty unicompartmental knee replacements that had been performed in a community hospital setting, reported a ten-year survival rate of 97% (with revision or loosening as the end point). Those investigators found that an age of more than sixty-five years and postoperative alignment positively affected the clinical outcome and survivorship, respectively. Soave et al. used roentgen stereophotogrammetric analysis to investigate the stability of the tibial component of twenty Duracon UNI prostheses (Howmedica, Limerick, Ireland). After a mean duration of follow-up of thirty months, only one patient showed progressive migration (3.2 mm at four years) and there was no evidence of polyethylene deformation (except in the case of one patient who had loosening). Argenson et al. used videofluoroscopy to investigate the in vivo kinematics in twenty subjects who had had a unicompartmental knee replacement and found that the anterior cruciate ligament plays an important role in these patients. Polyethylene wear has been found to be a common mode of failure following unicompartmental knee replacement. McGovern et al. reported on seventy-five unicompartmental knee replacements and found a significant inverse linear correlation between the shelf life of the polyethylene and the time to revision (p < 0.01) when the polyethylene had been sterilized by gamma irradiation in air 3 . Early severe wear was identified in cases of storage for more than 4.4 years. Engh et al. reviewed sixty-eight retrieved unicompartmental knee replacements (of seven designs) and found that wear scores were correlated with time in situ and the age of polyethylene at the time of revision (p < 0.01). The wear scores were not correlated with shelf life (p = 0.27) or initial polyethylene thickness (p = 0.91). Clinical Results of Primary Total Knee Replacement The correlation between the volume of total knee replacements performed by a surgeon and outcome was the subject of one review. Hervey et al. reported on a national stratified probability sample from the 1997 Health Care Utilization Project—National Inpatient Survey Release 6 and found that patients who were treated by providers with a lower caseload had higher rates of mortality, infection, and deep venous thrombosis. The superiority of cruciate-retaining or posterior stabilized total knee designs continues to be debated. Tanzer et al., in a prospective, randomized, double-blind trial comparing twenty posterior cruciate-retaining and twenty posterior stabilized total knee replacements, found no differences in the clinical, functional, or radiographic scores of the Knee Society at two years. Trousdale et al. reviewed 11,606 primary total knee arthroplasties and found that the ten-year survival of posterior stabilized knee implants was significantly worse than that of posterior cruciate-retaining knee implants (76% compared with 91%, p < 0.0001). The mechanisms of failure among 2994 posterior stabilized total knee implants included extensor mechanism complications (36%), infection (25%), tibial loosening (18%), polyethylene wear (10%), instability (3%), fracture (2%), and a combination of mechanisms (6%). Gill et al., in a study of 887 posterior cruciate-retaining total knee replacements, reported an 88% rate of implant survival at twenty-four years, with a 2.7% rate of revision for aseptic loosening. Fetzer et al. reported on 101 posterior cruciate-retaining total knee replacements after a mean duration of follow-up of 10.5 years. The twelve-year survival rate was 93.3%, with only one knee having radiographic signs of failure, and the mean range of motion was 1° to 110°. Indelli et al., in a report on the five-to-nine-year results of 100 total knee arthroplasties performed with the Insall-Burstein II posterior stabilized prosthesis (Zimmer), found a 97% rate of excellent or good results according to the scoring system of the Knee Society, a mean flexion of 116°, only one osteolytic lesion, and no cases of aseptic loosening or infection. The surgical technique of total knee arthroplasty has been addressed in several studies. Maloney et al., in a review of the minimum five-year results of ninety-seven total knee arthroplasties that had been performed with a tibial surface cementation technique (cementation of the cut surface of the tibia without cementation of the stem), reported that the rate of tibial aseptic loosening was 14.4% (fourteen knees) and concluded that this technique should not be used. Winemaker reported on the use of a tensor-balancer device during eighty-three total knee arthroplasties, with restoration of the flexion-extension symmetry to within 1° in all knees. Green et al. evaluated the effect of varus tibial alignment on proximal tibial strain in a study of fourteen paired fresh-frozen cadaveric tibiae (with one specimen from each pair cut in neutral and the other specimen cut in 5° of varus). The investigators found an area in the posteromedial part of the tibia that demonstrated significantly increased strain in the varus knees (p < 0.05). Forster et al., in a report on the minimum five-year results of 249 total knee arthroplasties performed with use of the Kinemax Plus device (Howmedica, Allendale, New Jersey) and an all-polyethylene tibial component, found an 86% rate of good and excellent results (according to the scoring system of the Knee Society) and a 94.5% rate of survival at seven years. Iorio et al., in a matched-pair analysis in which ninety-eight patients with an all-polyethylene tibial component were compared with ninety-eight patients with a metal-backed tibial component, reported no reoperations or revisions in either group and found no differences between the groups in terms of range of motion, pain scores, Hospital for Special Surgery scores, Knee Society scores, or SF-36 scale scores. Extensor mechanism complications continue to be common and were the topic of concern in four studies. Whiteside and Nakamura, in a study of knees with unresurfaced patellae, compared one design that had a shallow patellar groove and a flat femoral component surface (Ortholoc II; Dow Corning Wright, Arlington, Tennessee [thirty-eight knees]) with two designs that had a more anatomic trochlear groove (Advantim; Wright Medical Technology, Arlington, Tennessee [222 knees] and Profix; Smith and Nephew, Memphis, Tennessee [330 knees]). They found that the knees with a more anatomic trochlear groove had less anterior pain (p < 0.002). Mahoney et al. also reviewed the effect of total knee implant design on extensor mechanism function. They found that patients who had received an implant with a single flexion-extension axis achieved flexion earlier, were able to arise from a chair earlier, and had less anterior knee pain with this activity than did patients who had received an implant with a multi-radius design (p < 0.003). Kawano et al. evaluated sixty-two knees with regard to postoperative patellar tracking and found that patellar component position (medialization), patellar resection angle, and lateral retinacular release positively affected tracking. Valdivia et al. reported the intermediate-term results associated with thirty-nine press-fit and 133 cemented highly congruent patellar components. At a mean of six years, the cemented implants had a higher prevalence of maltracking (p = 0.005) but there was no significant difference between the groups with regard to Knee Society scores or complications (p > 0.05). The results of total knee arthroplasty in specific subgroups of patients were discussed in several reports. Mont et al. reported on the results of total knee arthroplasty in thirty patients who were less than fifty years of age at the time of the procedure. After a mean duration of follow-up of eighty-six months, 97% of the knees had an excellent or good Knee Society objective score and only one knee had been revised (because of unexplained pain). Joshi and Gill, in a review of twenty total knee arthroplasties that had been performed in patients who were more than ninety years old, reported excellent pain relief in all patients, a mean hospital stay of 10.1 days, one early postoperative death, and a 26.3% rate of medical complications. Deshmukh et al., in a study of 180 obese patients, found that body weight did not adversely influence the outcome of total knee arthroplasty, at least in the short term. Mont et al. reviewed thirty total knee arthroplasties that had been performed primarily for patellofemoral arthritis and found a 97% rate of good or excellent results after a mean duration of follow-up of eighty-one months. Mont et al. also reviewed thirty-two total knee arthroplasties that had been performed for osteonecrosis and reported a successful clinical outcome in 97% of the knees, with no progressive radiolucent lines, after a mean duration of follow-up of 108 months. Mobile-Bearing Total Knee Replacements The advantages of mobile-bearing knee designs remain theoretical, as the following studies suggest. Hartford et al. reported on 139 consecutive mobile-bearing knee replacements (LCS; DePuy, Warsaw, Indiana) after a mean duration of follow-up of ten years and found that the rate of prosthetic survival was 90% at nine years and 63% at fifteen years 4 . They found that the dramatic decrease in the survival rate at fifteen years was due primarily to mobile-bearing fracture. Aglietti et al., in a prospective, randomized study, compared a fixed-bearing posterior stabilized total knee implant (LPS; Zimmer) with a mobile-bearing total knee implant (MBK; Zimmer). After three years of follow-up, there were no significant differences between the groups with regard to maximum flexion, extension loss, pain, or knee scores. Rasquinha et al. examined the rate of lateral release in a study of 100 posterior stabilized total knee replacements and 100 mobile-bearing total knee replacements and found a significantly lower rate of lateral release in the mobile-bearing group (0% compared with 10%, p < 0.05), which was thought to be due to the self-correcting tibial rotational alignment that occurs with this implant. The Difficult Primary Total Knee Replacement Extra-articular deformity can be addressed with asymmetrical intra-articular resection (which can make ligamentous balancing difficult) or with correctional osteotomy performed prior to or at the time of primary total knee arthroplasty. Wang and Wang, in a study of fifteen patients who had an extra-articular deformity, reported that intra-articular bone resection is an effective procedure in patients with <20° of femoral deformity or <30° of tibial deformity in the coronal plane 5 . Radke and Radke reported on ten patients with >15° of extra-articular tibial deformity who were treated with an unconstrained total knee replacement and a tibial osteotomy. Union of the osteotomy site and correction of alignment were observed in all cases. Catonne et al., in a review of eleven total knee arthroplasties that had been performed in association with a corrective femoral or tibial osteotomy, also reported good clinical results and deformity correction. Infection and implant longevity in immunocompromised patients remain concerns and were reviewed in several studies. Norian et al. reviewed the results of fifty-three total knee arthroplasties that had been performed in thirty-eight patients with hemophilic arthropathy (twenty-nine of whom were positive for the human immunodeficiency virus [HIV]) and found a five-year implant survival rate of 90%, with seven infections (prevalence, 13%) that were independent of HIV status. Lewis et al. reviewed the results of sixty total joint replacements that had been performed in thirty-two HIV-positive patients and found only one infection after a mean duration of follow-up of eight years. They concluded that these patients did not demonstrate a dramatically increased rate of infection or a poor clinical outcome when compared with historical controls. Meding et al. reviewed the results of 363 total knee arthroplasties that had been performed in patients with diabetes mellitus. After a mean duration of follow-up of fifty-two months, these patients had less pain and higher knee scores when compared with nondiabetic patients who had had a total knee replacement during the same time-period. The rate of deep infection was only slightly higher in the group of diabetic patients (1.3% compared with 0.7%). Giori and Lewallen identified fifteen patients (sixteen knees) with poliomyelitis who underwent primary total knee arthroplasty and reported a high complication rate 6 . Knee Society pain and knee scores were improved in patients with at least antigravity quadriceps strength preoperatively, but pain relief was inferior in patients with less than antigravity strength. McAuley et al. reported on twenty-seven total knee arthroplasties in patients with a poor preoperative range of motion (mean, 30°; range, 0° to 50°). After a mean duration of follow-up of six years, the mean range of motion was 74° (range, 15° to 110°), the complication rate was 41%, and the revision rate was 18.5%. However, 86% of the patients were satisfied and reported improved function. Parvizi et al., in a report on forty total knee arthroplasties that had been performed in twenty-nine patients with neuropathic arthropathy, reported six revisions (prevalence, 15%) after a mean duration of follow-up of 7.9 years. Berry et al. reviewed the results of thirty total knee arthroplasties that had been performed with use of a cemented tibial component with a wedge augment for bone deficiencies. Only two tibial components had been revised for loosening after a mean duration of follow-up of 12.2 years, demonstrating that this is a successful technique in cases of tibial bone loss. The Kinematics of Total Knee Replacement Several studies have focused on the kinematics of mobile-bearing designs. Haas et al. compared the kinematics of five posterior stabilized mobile-bearing implants with those of five posterior cruciate-sacrificing mobile-bearing implants by means of video fluoroscopy during gait and knee flexion to 90°. They found that both groups of knees had similar kinematics during gait and that the group with the posterior stabilized implant had more posterior rollback of the lateral condyle during knee flexion. Stiehl et al. used a frontal plane fluoroscopy system to evaluate coronal plane liftoff and medial-lateral translation in ten knees (five with an LCS rotating platform implant [DePuy] and five with an LCS posterior stabilized rotating platform implant [DePuy]) during normal gait. The group with the posterior stabilized implant was found to have less medial-lateral translation and less condylar lift-off (p < 0.05). Elias et al. studied the in vitro kinematic patterns of six cadaveric knees in the intact state, after implantation of a fixed-bearing total knee prosthesis, and after implantation of a mobile-bearing prosthesis. They found no significant difference between the two types of prostheses. Walker et al. reported on the kinematics of a mobile-bearing knee implant that allowed ±20° of rotation and 4.5 mm of anteroposterior translation. The motion patterns varied among subjects and, in general, did not reproduce normal knee motion. Rees et al. performed a randomized, controlled trial of seven patients who had a posterior cruciate-retaining mobile-bearing knee implant on one side and a fixed-bearing posterior cruciate-retaining implant on the other. The subjects were studied with in vivo fluoroscopic analysis. The investigators found that the knees with the posterior cruciate-retaining mobile-bearing implant demonstrated a linear, more consistent behavior during extension against gravity, flexion against gravity, and a step-up exercise (p < 0.039). Rubash et al. studied ten mobile-bearing total knee implants in cadaveric specimens with use of an in vitro robotic experimental setup and found that mobile bearings did not enhance tibial rotation when compared with a fixed-bearing design. There were fewer kinematic studies on fixed-bearing designs. Draganich et al. studied patients who had received different posterior stabilized knee designs and found that patients with implants that enforced early rollback (such as the Maxim posterior stabilized implant [Biomet] and the TRAC posterior stabilized implant [Biomet]) demonstrated more normal quadriceps efficiency during stair-stepping. Komistek et al. studied the in vivo anteroposterior contact positions in twenty subjects who had either a posterior cruciate-retaining total knee implant or a posterior stabilized total knee implant. No difference was detected when the subjects were in a seated position. When the subjects were in a kneeling position, the contact point shifted more anteriorly in the group with the posterior cruciate-retaining implant and remained posterior in the group with a posterior stabilized implant. In contrast with previous reports on the erratic kinematics of posterior cruciate-retaining total knee implants, Bertin et al. evaluated twenty subjects during a deep knee bend and found that nineteen experienced posterior femoral rollback of the lateral condyle (mean, -3.9 mm) and thirteen experienced posterior femoral rollback of the medial condyle (mean, -3.1 mm) 7 . Blood Management
The practice of autologous predonation has been called into question by several studies. Cushner et al., in a study of 376 unilateral total knee arthroplasties that had been performed in patients who had donated one unit of autologous blood preoperatively, found a marked resultant preoperative anemia, with 2.3% of the patients still requiring allogeneic transfusion. The investigators concluded that preoperative donation programs are not effective and should be abandoned. Friederichs et al. reported on 200 consecutive patients who had been managed with total joint replacement and had undergone perioperative blood salvage and retransfusion. The rate of allogeneic transfusion was 3.8% (five of 132) for patients managed with total knee replacement and 4.4% (three of sixty-eight) for patients managed with total hip replacement. The risk of allogeneic transfusion was 1.2% (two of 173) for patients in whom the preoperative hematocrit was Complications Following Total Knee Replacement Several articles have examined neurovascular complications related to total knee replacement. Reis reported on nine patients who suffered skin necrosis after total knee arthroplasty, eight of whom had predisposing factors. He concluded that necrosis over the patellar tendon or tubercle usually requires muscle flap coverage, whereas necrosis over the patella may be treated effectively with local wound-care and skin-grafting. Bruce et al. studied the effect of the ankle brachial pressure index and the use of a tourniquet in seventy-three patients. They found that it was safe to proceed with surgery in patients with nonpalpable foot pulses or claudication as long as the femoral pulse was present and there were no active ulcerations and no pain at rest. Patil et al. studied the effects of total knee replacement on blood flow to the limb in forty patients (thirty-three of whom had normal vascular status and seven of whom had chronic ischemia) and found that the ankle brachial pressure index was not affected by total knee replacement in either group. Sierra et al. studied sixty-nine above-the-knee amputations that were performed following total knee replacement. The cause of amputation was unrelated to the total knee arthroplasty in thirty-six limbs (53%), uncontrollable infection in twenty-three (33%), periprosthetic fracture in three, severe pain in three, severe bone loss in three, and a vascular complication in one. Seventy-five percent of the patients were unable to walk independently following the above-the-knee amputation. Clarke et al. evaluated the risk of peroneal nerve injury in association with the use of a pie-crust technique of lateral release in a study of sixty patients who were evaluated with magnetic resonance imaging. The mean distance between the nerve and the tibia was found to be 1.29 cm, a distance that was correlated with the anteroposterior dimensions of the leg (correlation coefficient, 0.6). The investigators concluded that the technique is safe if performed carefully. Extensor mechanism complications remain a difficult problem. Crossett et al. reviewed their experience with the use of fresh-frozen Achilles tendon allografts in nine patients with patellar tendon ruptures 9 . Two patients had early failure of the graft, and the remaining patients had a mean extensor lag of 3°. Pollock et al., in a review of 459 total knee arthroplasties performed with three different designs, identified twenty-seven knees that required surgical treatment because of synovial entrapment. They found this complication to be very design-dependent, with posterior stabilized femoral components with a proximally positioned or wide femoral box to be at greatest risk. Periprosthetic femoral fracture was the topic of two articles. Wang and Wang reported on ten supracondylar femoral fractures that occurred proximal to the site of a total knee arthroplasty. All fractures were treated with open reduction and internal fixation with use of a plate and cortical strut allografts, and all united at a mean of 17.6 weeks. Bong et al., in a study of paired human cadaveric femora, compared the LISS plate (Less Invasive Stabilization System; Synthes USA, Paoli, Pennsylvania) with a retrograde nail for fixation of simulated periprosthetic femoral fractures and found that the retrograde nail provided greater stability 10 . The etiology of stiffness following total knee replacement can be multifactorial. Boldt et al., in a matched case-control study of thirty-eight patients, found a significant correlation (p < 0.00001) between internal malrotation of the femoral component on computed tomography scans and arthrofibrosis 11 . Nazarian et al. evaluated the use of postoperative radiation therapy in a study of sixty patients with a history of heterotopic ossification who underwent revision total knee replacement. The patients were randomized to treatment with either 700 rads of radiation or a placebo dose. Twenty-four of the thirty patients who received radiation had no recurrence of heterotopic ossification, compared with only twelve of the thirty patients who did not receive radiation. Maloney, in a study of posterior stabilized total knee arthroplasties, found the prevalence of arthrofibrosis (defined as <90° of flexion at six weeks) to be 11% (twenty-four of 223). Manipulation under anesthesia resulted in an improvement in the mean range of motion from 67° before manipulation to 111° immediately after manipulation and to 114° at one year. Medical complications following total knee replacement were reported in three presentations at the annual meeting of the American Academy of Orthopaedic Surgeons in Dallas in 2002. Colwell et al. presented a North American, multicenter, randomized, double-blind trial in which an oral direct thrombin inhibitor, ximelagatran, was compared with warfarin for the prevention of venous thromboembolism in 680 patients undergoing total knee replacement 12 . The investigators found that ximelagatran, started the morning after surgery, was at least as effective as warfarin and did not require routine monitoring or dose adjustment. Reitman et al. reported on 1300 total knee replacements in patients who were treated with intraoperative heparin, epidural hypotensive anesthesia, external pneumatic compression boots, and aspirin. Screening with duplex venous ultrasound prior to discharge demonstrated that the prevalence of deep venous thrombosis was 3.94%. The prevalence of bleeding complications was 1.04%. In a review of mortality following 1108 total hip and 610 total knee replacements, Nunley et al. reported no intraoperative deaths. The in-hospital mortality rate following total knee arthroplasty was 0.16%, and the one-year mortality rate was 1.53%. All deaths were unrelated to the surgery. Infection at the Site of Total Knee Replacement The routine use of antibiotics in cement for patients undergoing primary total knee arthroplasty was addressed in one study. Chiu et al. performed a prospective, randomized study of 340 primary total knee arthroplasties in which Group 1 (178 knees) received cefuroxime-impregnated cement and Group 2 (162 knees) received plain cement 13 . There were no deep infections in Group 1 and five deep infections (prevalence, 3.1%) in Group 2 (p = 0.024). Nazarian et al. reported on a two-stage approach for the treatment of arthritic knees that had chronic infection (recurrent infection or osteomyelitis). These patients underwent a radical débridement, primary bone cuts, and placement of an antibiotic-impregnated cement spacer block followed by a six-week course of parenteral antibiotics. Total knee implantation was performed no longer than two months after the index procedure, with no recurrent infections. The treatment of infection at the site of a total knee arthroplasty can be a very challenging undertaking. Deirmengian et al. reviewed the outcomes for thirty-one patients in whom an acute gram-positive infection at the site of a total knee arthroplasty was treated with open débridement and component retention. The mean duration of symptoms was nine days. This method of treatment was associated with a success rate of 35% (eleven of thirty-one). Only one of thirteen patients who had an infection with Staphylococcus aureus was successfully treated with this technique, compared with 56% of patients who had an infection with Streptococcal species or Staphylococcus epidermidis (p = 0.007), suggesting that successful treatment by this means may depend largely on the causative bacteria. Nazarian et al. also reported the results for 124 knees in which an infection at the site of a total knee arthroplasty was treated with a two-stage reimplantation with use of a static intervening antibiotic spacer. The infection was successfully eradicated in 112 (90%) of the 124 knees, and the average postoperative arc of motion was from 3° (range, 0° to 15°) to 95° (range, 85° to 125°). Wang et al. reviewed the clinical outcome for twenty-six knees in which an infection at the site of a total knee arthroplasty had been successfully treated. After more than two years of follow-up, approximately 50% of the reimplanted total knee components were associated with mild to moderate pain. Polyethylene Wear and Osteolysis The measurement of wear at the site of a total knee arthroplasty generally has been qualitative. Hoshino et al. described a new technique for the evaluation of linear polyethylene wear in vivo with use of a single standing anteroposterior radiograph and three-dimensional transformation parameters 14 . The sixty-nine patients in that study had an average annual wear rate of 0.23 mm (two times greater than that commonly reported after total hip arthroplasty). Miura et al. found that their "wear index" (defined as a numeric model of deformation multiplied by the sliding velocity) was predictive of polyethylene wear in a knee simulator model. Conditt et al. used laser surface profilometry in combination with computer-aided-design (CAD) software to quantitate the volumetric backside wear associated with AMK knee implants (DePuy) that had been in situ for thirty-six to ninety-six months and found a mean loss of 1064 mg of polyethylene, a magnitude large enough to generate osteolysis. The Mark Coventry Award of the Knee Society was given to Jacobs et al. for their evaluation of wear and deformation on the backside of twenty-five posterior cruciate-retaining total knee arthroplasty inserts from one manufacturer at a mean of sixty-four months after implantation. They found limited backside damage with polishing in twenty-one inserts (84%), abrasive wear in five (20%), and pitting in twenty-one (84%). While many recommendations have been made with regard to the minimum acceptable polyethylene thickness, Edwards et al. found that manufacturers' reported values of minimum thickness (8, 9, and 10 mm) differed from actual measurements (range, 5.6 to 6.2 mm). Lachiewicz and Soileau evaluated 165 consecutive Insall-Burstein II posterior stabilized total knee replacements (Zimmer) after a mean duration of follow-up of seven years and reported a 3.6% prevalence of osteolysis (six knees) and no tibial loosening. O'Rourke et al. reported on 176 cemented modular Insall-Burstein II posterior stabilized total knee implants (Zimmer) after a mean duration of follow up of 6.4 years. While there had been no revisions for loosening or osteolysis, osteolysis was observed radiographically in seventeen of 105 knees (16%) and was believed to be secondary to post-cam impingement and backside wear. Vagner et al. examined ninety-three retrieved posterior stabilized tibial inserts and found evidence of post-cam contact in 86% of the inserts and burnishing of the post in 81%. The investigators concluded that posterior stabilized knee implants do not tolerate substantial misalignment without sustaining localized damage to the tibial post. Weber et al. reviewed more than 1000 AGC total knee implants (Biomet) after five to eleven years of follow-up and found that compression-molded monoblock designs (used in 698 knees) were associated with a lower prevalence of osteolysis, radiolucent lines, and revision when compared with ram-extruded modular designs (used in 353 knees). Keating, in a study of 4583 AGC total knee implants that had a compression-molded nonmodular tibial component, reported a 99% survival rate at fifteen years with no revisions for polyethylene wear or osteolysis. Huang et al. reviewed eighty revisions (performed with thirty-four LCS implants [DePuy] and forty-six fixed-bearing knee implants) and found the prevalence of osteolysis to be significantly higher in the mobile-bearing group (47% compared with 13%; p = 0.003) 15 . These lesions were predominantly on the femoral side and were underestimated on prerevision radiographs. Conditt et al. examined the contribution of third-body abrasion to wear in a study of 166 cemented total knee implants that were retrieved at the time of revision (one to twenty-six months after implantation). The severity of wear was found to be greater in posterior stabilized total knee implants than in non-posterior stabilized total knee implants (p < 0.02), with cement particles embedded in 11% of the posterior stabilized inserts. Puloske et al. evaluated the preimplantation surface roughness of nine posterior stabilized femoral components from different manufacturers. The articular surface of the posterior stabilizing cam device was found to be rougher than the condylar surface (p < 0.05) in six of the nine implants, indicating a potential site for the generation of wear debris. Potential solutions to the problem of wear include modification of the bearing surfaces and biologic methods. Muratoglu et al. examined eight explanted highly crosslinked and seventy-one conventional polyethylene tibial inserts. The total damage score was reduced in the highly crosslinked group (47.3 compared with 58.6). Melt recovery analysis "healed" nearly all surface damage in this group, suggesting that plastic deformation rather than material removal had occurred. Laskin reported that an oxidized zirconium surface demonstrated an overall 85% reduction in polyethylene wear compared with cobalt chromium. With regard to biologic solutions to the wear problem, Ulrich-Vinther et al. used gene therapy in the form of a recombinant adeno-associated virus vector co-expressing osteoprotegerin (a natural decoy protein that inhibits osteoclast activation and bone resorption) in a mouse calvarial model and found inhibition of osteolysis 16 . Revision Total Knee Replacement Two articles focused on exposure techniques in revision total knee arthroplasty. Mendes et al. reviewed the results of fifty-one tibial tubercle osteotomies and found the union rate to be 88% (forty-five of fifty-one). Fehring et al. reported on the patellar inversion method for exposure in a study of 420 revision total knee arthroplasties. This technique involves a lateral retinacular release with no attempt to evert the patella and a progressive subperiosteal release of the medial sleeve combined with external rotation of the tibia. Several reports focused on the use of stems in revision total knee arthroplasty. Fehring et al. reviewed 209 revision total knee arthroplasties that had been performed with use of metaphyseal-only-engaging stems. After a mean duration of follow-up of fifty-five months, ten (9%) of the 106 cement stems and twenty-five (24%) of the 103 cementless stems were characterized as loose or possibly loose radiographically, indicating that cemented metaphyseal-only stems are perhaps more stable. Shannon et al. examined thirty revision total knee arthroplasties that had been performed with cement fixation of the components with press-fit stems (noncemented) and found a 16% rate of mechanical failure at about six years. Parsley et al., in a comparison study of cemented stems, non-canal-filling cementless stems, and canal-filling cementless stems, found that the ability to achieve tibial alignment in the anteroposterior and lateral planes was more predictable with canal-filling cementless stems and least predictable with cemented stems. Nazarian et al. compared 161 stemmed and sixty-five nonstemmed Insall-Burstein Constrained Condylar revision total knee implants (Zimmer) and found no significant difference in loosening rates. The John Insall Award of the Knee Society was given to Sharkey et al. for their review of the etiology of failure of 212 revisions that had been performed between nine days and twenty-eight years after implantation. The causes of failure included polyethylene wear (25%), aseptic loosening (24.1%), instability (21.1%), infection (17.5%), arthrofibrosis (14.6%), malalignment or malpositioning (11.8%), a deficient extensor mechanism (6.6%), patellar osteonecrosis (4.2%), periprosthetic fracture (2.8%), and isolated patellar resurfacing (0.9%). More than one cause was identified in 32% of the patients. Cameron, in a review of ninety-eight revisions in which the original femoral component was retained, reported no instances of subsequent femoral component loosening after two to twelve years of follow-up. They concluded that a well-fixed, well-aligned femoral component can be successfully retained in a low-demand patient in spite of some surface scratching. Babis et al. reviewed fifty-six isolated polyethylene insert exchanges after a mean duration of follow-up of 8.3 years 17 . Fourteen (25%) of the fifty-six knees required a subsequent rerevision at a mean of only three years after the tibial insert exchange, and the survival rate at 5.5 years was only 64%. The investigators concluded that isolated tibial polyethylene exchange should be undertaken with caution. Christensen et al. reported on eleven revisions that had been performed with a posterior stabilized condylar prosthesis for the treatment of stiffness. After a mean duration of follow-up of thirty-eight months, the range of motion had increased from 40° to 83°, pain scores had improved, and all eleven patients were satisfied. Berger et al. reported the five-year results of thirty-eight revision total knee arthroplasties that had been performed with a semiconstrained modular prosthesis (Insall-Burstein-CCK; Zimmer). There were two aseptic revisions (one for instability and one for tibial loosening), for a five-year survival rate of 92.3%. Nelson et al. compared the results of sixteen revisions in which severe patellar bone loss had been treated with a porous tantalum patellar shell with those of nine revisions in which severe patellar bone loss had been treated with patelloplasty. The investigators found diminished anterior knee pain in the group that had received the tantalum shell but reported no difference between the groups with regard to Knee Society scores after short-term follow-up. Lonner et al. reviewed the results of 202 revision total knee arthroplasties in which the original, well positioned and stable patellar component was retained. After a mean duration of follow-up of seven years, the prevalence of anterior knee pain was 10% (twenty-one knees) and only patellar components that had been gamma irradiated in air (eleven knees) had failed. Springer et al. reported on twenty-six knees that had been treated with a rotating-hinge implant for nonneoplastic limb salvage. After a mean duration of follow-up of fifty-nine months, there was significant improvement (p < 0.05) in range of motion as well as in the knee and functional scores. However, the rate of complications was high (35%; nine patients). The most common complication was deep infection (five patients). Sierra et al., in a review of 1893 revision knee arthroplasties, found that 287 knees (15.2%) required reoperation. The investigators concluded that patients undergoing revision total knee arthroplasty have a surprisingly high risk of subsequent problems that will require reoperation.
References
  CiteULike  Connotea  Del.icio.us  Technorati What's this?
This article has been cited by other articles:
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
37% and 22% (six of twenty-seven) for those in whom the preoperative hematocrit was <37% (p < 0.01). Salido et al., in a retrospective study of 296 patients (370 joint replacements), found a significant relationship (p = 0.0001) between the preoperative hemoglobin level and postoperative transfusion
| home | CME | help | feedback | subscriptions | archive | site index |
|
Stanford University Libraries' HighWire Press (TM) assists in the publication of JBJS Online. |