 |
 |
 |
|
| home | CME | help | feedback | subscriptions | archive |
|
The Journal of Bone and Joint Surgery (American). 2006;88:1677-1686. doi:10.2106/JBJS.F.00450 © 2006 The Journal of Bone and Joint Surgery, Inc. What's New in Adult Reconstructive Knee Surgery1 New Mexico Orthopaedics, 201 Cedar SE, Suite 6600, Albuquerque, NM 87106. E-mail address for M.J. Archibeck: archibeckmj{at}nmortho.net
Specialty Update has been developed in collaboration with the Council of
Musculoskeletal Specialty Societies (COMSS) of the American Academy of
Orthopaedic Surgeons.
The purpose of this update is to discuss, in summary fashion, topics presented at selected orthopaedic meetings and published in related orthopaedic journals during 2005. 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 Washington, DC, on February 23 through 27, 2005), on Combined Specialty Day at the meeting of The Knee Society (held in Washington, DC, on February 26, 2005), at the interim meeting of The Knee Society (held in New York, NY, on September 8 through 10, 2005), and at the annual meeting of the American Association of Hip and Knee Surgeons (held in Dallas, Texas, on November 4 through 6, 2005).
While total knee replacement is very effective for alleviating pain and improving function in patients of all ages, other approaches have been successful as well. Pearsall et al. reported on the use of fresh osteochondral allografts and autografts for the treatment of full-thickness articular cartilage defects and demonstrated improvement in 80% of sixty-four patients at an average of thirty-six months. Mithoefer et al.1, in a prospective study of forty-eight isolated full-thickness articular cartilage defects that were treated with the microfracture technique, reported good to excellent results in 67% of the patients after a minimum duration of follow-up of two years. Higher scores were found to be correlated with lower bodymass index. The role of osteotomy in the treatment of the arthritic knee was addressed in two studies. Argenson et al. reviewed the results of 299 closing-wedge proximal tibial osteotomies after twelve to twenty-eight years of follow-up and found that only forty-three knees (14%) were revised secondary to the progression of arthritis. In a symposium, Fowler reviewed the benefits of opening-wedge osteotomy, which include the ability to correct biplanar deformities, the need for only one cut, and the fact that the procedure does not violate the anterior compartment of the knee. The procedure does, however, require the use of a graft. Wang and Hsu, in a study of thirty patients with valgus gonarthrosis who were managed with a distal femoral varus osteotomy, reported that 83% of the patients had a satisfactory result and that the ten-year survival rate was 87%. Sisto and Mitchell2 evaluated the results of thirty-seven procedures that were performed with use of the UniSpacer implant (Zimmer, Warsaw, Indiana) for the treatment of medial osteoarthritis. After a mean duration of follow-up of twenty-six months, there were no excellent, ten good, fifteen fair, and twelve poor results. All twelve knees with a poor result were revised to total knee arthroplasty. Conditt et al., in a report on 132 patients, including twenty-seven patients who were managed with a UniSpacer, nineteen patients who were managed with a unicompartmental knee arthroplasty, twenty-one patients who were managed with a total knee arthroplasty, and a matched control group of sixty-five subjects, found no significant differences among the groups with regard to most functional activities (kneeling, squatting, and exercises) and noted similar levels of satisfaction after short-term follow-up.
With the interest in more rapid recovery and shortened hospital stays, perioperative pain management has received a substantial amount of attention. MacDonald et al. reported the results of a randomized, blinded clinical trial of sixty-four patients undergoing total knee arthroplasty who received a periarticular injection (ropivacaine, ketorolac, epimorph, and epinephrine) or no injection. They found that the injection group had lower use of patient-controlled analgesia (p < 0.01), greater satisfaction (p = 0.016), and lower visual analog pain scores (p = 0.04) at four hours postoperatively, with no increase in the rate of wound complications. Vendittoli et al., in a study of forty-two patients undergoing total knee arthroplasty who were randomized to local anesthetic injection or no injection, found lower morphine consumption and less nausea in the injection group. Hartrick et al. studied the effect of a single perioperative epidural injection of DepoMorphine (20 or 30 mg) or a sham injection and found that the groups that received DepoMorphine had reduced time-weighted pain intensity recall scores. However, pruritus and pyrexia were increased in the group that received 30 mg of DepoMorphine. In the study by Dorr et al., patients were randomly assigned to receive femoral nerve block or epidural nerve block for thirty-six hours following total knee arthroplasty. During the first thirty-six hours, patients with the femoral nerve block experienced greater pain relief but had less function (as measured according to walking distance and manual muscle-testing) than did the patients with an epidural block. In the study by Snow Yeo et al., sixty patients who were undergoing unilateral total knee arthroplasty were prospectively randomized to continuous femoral nerve block (with 0.15% or 0.2% ropivacaine) or to a control group. The authors found increased morphine use (p < 0.05) and greater nausea (p < 0.05) in the control group. Peters et al. evaluated a multimodal perioperative anesthetic regimen that included preoperative and postoperative oral narcotics, cyclooxygenase-2 (COX-2) inhibitors, a spinal anesthetic, femoral nerve blocks, and local anesthetic wound infiltration. The authors found that the mean length of stay decreased from 3.1 to 2.5 days and that visual analog pain scores were decreased during the first two postoperative days as compared with the findings associated with other anesthetic techniques. In the study by Marshall et al., ninety-eight patients were randomized into a control group, a femoral nerve block group, and a pericapsular injection (opioid/anesthetic combination) group. The investigators found that the use of supplemental pain management modalities (that is, femoral nerve block or pericapsular injection) reduced the need for postoperative morphine by approximately 30%. In the study by Holmstrom and Hardin, sixty patients (sixty-one total knee arthroplasties) were randomized into a control group, a cold compressive dressing (Cryo/Cuff) group, and an epidural group. The investigators found the cold compressive dressing and epidural groups performed similarly, with significantly less narcotic consumption as compared with the control group (p = 0.028). As many surgeons have begun to use COX-2 inhibitors perioperatively, some concern has arisen regarding the potential negative effects that such therapy could have on osseous growth into cementless devices. Hofmann et al. studied nine patients who underwent staged bilateral total knee arthroplasty with tetracycline labeling being performed twice (before and after the first total knee arthroplasty). Two tantalum plugs were placed in the contralateral medial femoral condyle during the first total knee arthroplasty and were removed at the time of the second total knee arthroplasty. Patients received a COX-2 inhibitor perioperatively following the first total knee arthroplasty only. The authors found better pain control in association with the use of a COX-2 inhibitor and reported no inhibition of bone turnover or bone ingrowth in the tantalum plugs.
Unicompartmental knee arthroplasty, which has been associated with less invasive techniques and improved survival rates, has become increasingly popular. Patil et al., in a kinematic study of six cadaveric specimens, found that a tricompartmental knee replacement significantly changed the kinematics of the knee whereas a unicompartmental knee arthroplasty preserved more normal knee kinematics (p = 0.001). Engh et al., in a review of 411 consecutive medial unicompartmental knee arthroplasties that had been performed between 1984 and 1998 with use of a variety of fixation techniques, polyethylene sterilization techniques, and designs, found an 80% survival rate at nine years. Factors that were associated with revision included younger age, thinner initial polyethylene, longer polyethylene shelf age, and certain designs. O'Rourke et al. reported the clinical results of 136 unicompartmental knee arthroplasties in 103 patients after a minimum duration of follow-up of twenty-one years. Nineteen knees (14%) were revised during the study period because of progression of disease (nine knees), loosening (eight), or pain (two); the mean time to revision was 10.2 years. The authors also reported a significantly higher rate of revision in patients who had been less than sixty-five years of age at the time of surgery (p = 0.005). Berger et al.3 reported the minimum ten-year results of sixty-two consecutive unicompartmental knee arthroplasties that had been performed with a cemented modular Miller/Galante implant (Zimmer). The survival rate was 95.7% at thirteen years with revision or radiographic loosening as the end point. Minimally invasive techniques and computer-assisted surgery have become more commonplace during arthroplasty surgery. Perlick et al. reported on the use of a non-imaging-based navigation system for minimally invasive unicompartmental knee arthroplasty. When compared with conventional techniques, computer-assisted navigation was associated with a significantly improved mechanical axis as well as improved alignment of the tibial and femoral components in the coronal plane (with 95% as compared with 70% of the components being aligned within 4° of ideal) with an added operative time of nineteen minutes. Cossey and Spriggins reviewed thirty medial unicompartmental knee arthroplasties that had been performed with use of computer-assisted navigation and also found improved limb alignment as compared with that achieved with conventional techniques. Some investigators have raised concerns regarding the use of minimally invasive techniques. Lombardi et al., in a review of the early results of seventy-nine medial minimally invasive unicompartmental knee arthroplasties, reported sixteen failures (six cases of tibial loosening, three plateau fractures, four cases of persistent medial pain, one case of progressive arthritis, and two cases of infection) after an average duration of follow-up of thirty-eight months. They also found that a body-mass index of >32 was predictive of failure. Hamilton et al. reported on their first three years of experience with use of minimally invasive techniques in a study involving 221 unicompartmental knee arthroplasties. The results of these procedures were compared with the results of 514 unicompartmental knee arthroplasties that had been performed with use of a standard open procedure involving the same component design. The investigators found that the minimally invasive procedure was associated with higher rates of revision (11.3% compared with 8.6%) and aseptic loosening (3.7% compared with 1.0%).
While the clinical results of total knee arthroplasty continue to be excellent, debate continues with regard to the superiority of various fixation techniques and differing designs. Ma et al., in a study of sixty-four cemented Total Condylar knee implants (Howmedica, Rutherford, New Jersey), reported a twenty-year survival rate of 91.9% with revision or mechanical failure as the end point. Similarly, Vessely et al., in a study of 331 cemented condylar total knee implants that were followed for a mean of 15.7 years, reported fifteen-year survival rates of 93.7%, 95.9%, 97.0%, and 98.8% with component removal, revision for any reason, revision for mechanical failure, and revision for aseptic loosening as the end point, respectively. Dixon et al.4, in a study of 109 modular, fixed-bearing, posterior cruciate ligament-retaining total knee implants (Press-Fit Condylar; Johnson and Johnson, Raynham, Massachusetts) that were followed for a minimum of fifteen years, reported a survival rate of 92.6% at fifteen years with revision or any reoperation as the end point. Plaster, in a study of 174 cementless total knee implants (Natural-Knee; Zimmer) that were followed for a minimum of ten years, reported ten revisions (six for polyethylene exchange, one because of infection, and three because of patellar wear). None of the revisions were performed because of loosening. Tarkin et al.5, in a study of seventy cementless mobile-bearing total knee implants (Low Contact Stress; DePuy, Warsaw, Indiana) that were followed for a mean of sixteen years, reported a seventeen-year survival rate of 97% with aseptic loosening as the end point, with one revision having been performed because of loosening. Mobile-bearing total knee implants received much attention again this year. Callaghan et al.6, in an update of a previous report on 119 cemented, rotating-platform total knee implants (Low Contact Stress; DePuy), reported no cases of aseptic loosening; no revisions for loosening, osteolysis, or wear; an average flexion of 105°; and no bearing dislocations after a minimum duration of follow-up of fifteen years. Kim and Kim compared the results of anteroposterior glide and rotating-platform Low Contact Stress mobile-bearing designs (DePuy) in a study of 190 patients who received one design on each side. After a minimum duration of follow-up of five years, the authors reported favorable and comparable results, with no instances of aseptic loosening, revision, or measurable wear in either group. Haas et al. reported the results of an in vivo kinematic study of thirty-eight patients who had been managed with a posterior cruciate ligament-retaining implant (twenty patients), a posterior stabilized implant (nine), or a posterior cruciate ligament-sacrificing mobile-bearing total knee implant (nine). The investigators found bearing rotation and translation in all patients, with paradoxical anterior translation in deep flexion in the group that had received the cruciate ligament-retaining implant. In a study of eighty mobile-bearing knee replacements, Matsuda et al. found that knees with varus-valgus balance (defined as a <2-mm difference as measured with use of a Telos arthrometer) had improved modified Hospital for Special Surgery and Knee Society pain scores. In contrast, Chung and Shim, in a study of 125 total knee arthroplasties performed with the Low Contact Stress implant (DePuy), found that laxity had no detrimental effect on the early results of the procedure and that mediolateral plane laxity improved the early results. Chiavetta et al. reviewed 540 rotating-platform total knee arthroplasties that had been performed with use of a balance gap technique and found no bearing dislocations. Efforts to obtain greater flexion following total knee arthroplasty were the topic of several studies. Yang et al. evaluated eighty patients who had been prospectively randomized to a standard posterior stabilized total knee arthroplasty or a specialized flexion knee design and found that the flexion knee design was associated with significantly improved flexion at six months (116° compared with 122°), one year (120° compared with 128°), and two years (121° compared with 131°) (p < 0.05). In contrast, Kim et al. performed a bilateral comparison study in which fifty patients received a standard fixed-bearing posterior stabilized total knee implant on one side and a high-flexion posterior stabilized design on the contralateral side7. After an average duration of follow-up of 2.1 years, there was no significant difference with regard to the mean amount of flexion (135.8° compared with 138.6°). Huang et al.8 retrospectively reviewed twenty-five high flexion total knee arthroplasties after a mean duration of follow-up of twenty-eight months and compared the results with those of a matched group of standard posterior stabilized total knee arthroplasties. The authors reported a significant difference between the groups with regard to the amount of final flexion (138° compared with 126°, respectively) (p < 0.05). Several comparison studies of mobile and fixed-bearing designs were reported. In the study by Aglietti et al.9, patients were prospectively randomized to receive a fixed-bearing posterior stabilized total knee prosthesis or a mobile-bearing prosthesis. The mean range of motion was greater in the fixed-bearing group (112° compared with 108°). Bhan et al.10, in a study of thirty-two patients who received a fixed-bearing design on one side and a mobile-bearing design on the other, reported no significant differences with regard to Knee Society scores, the range of flexion, subject preference, or the rate of patellofemoral complications after a mean duration of followup of 4.5 years. There were two reoperations in the mobile-bearing group for the treatment of a bearing dislocation (one knee) and infection (one knee). Similarly, Kim and Kim performed a study of patients with bilateral involvement who received a fixed-bearing design on one side and a mobile-bearing design on the contralateral side. After a mean duration of follow-up of 10.3 years, there were no significant differences with regard to Hospital for Special Surgery scores, revision rates, loosening, or osteolysis. MacDonald et al. performed a randomized, prospective comparison of a cruciate-retaining mobile-bearing implant and two fixed-bearing cruciate-retaining total knee implants. After a mean duration of follow-up of 3.4 years, the authors reported no differences among the groups with regard to multiple outcome measures.
There has been an increasing interest in the use of minimally invasive techniques for total knee arthroplasty. Kolisek et al. reported the results of a multicenter prospective, randomized study in which a minimally invasive midvastus approach was compared with a standard parapatellar approach. After three months of follow-up, the authors reported nearly equal Knee Society scores for the two groups (83 and 86 points), higher SF-12 scores for the minimally invasive group, and similar radiographic results for the two groups. Laskin evaluated the results of 150 consecutive procedures that had been performed with use of the "mini-midvastus" approach and found an average increase of six minutes in operative time, with the patients more rapidly achieving 90° of flexion and stair-climbing postoperatively. Mont et al. evaluated twenty-six total knee arthroplasties that had been performed by means of a minimally invasive lateral approach and reported superior short-term results (a reduced time to straight-leg raising and reduced analgesic use) as compared with those of a standard-incision total knee arthroplasty. Aglietti et al., in a prospective, randomized, observer-blinded study comparing the early results of the quadriceps-sparing and mini-subvastus approaches for total knee arthroplasty, found some early advantages in association with the mini-subvastus approach (including an easier exposure as well as better active straight-leg raising at ten and thirty days), with no differences being observed at three months. Schorer et al. reported the results of a computed tomographic study of postoperative alignment in which the surgical accuracy of a mini-subvastus approach (twenty-five knees) was compared with that of a standard parapatellar approach (twenty-five knees). The authors found that the minimally invasive approach was not associated with any loss of accuracy in terms of tibial alignment but was significantly less accurate in terms of femoral alignment (p = 0.045). Bonutti et al., in a study of twenty consecutive "suspended leg" quadriceps-sparing total knee arthroplasties, reported that there were no complications and that the results of the procedure were similar to those in a control group with respect to operative time, loss of blood, length of stay, and early pain. Pagnano et al. quantified the anatomy of the extensor mechanism on the basis of an intraoperative inspection of the knee in 200 patients, an examination of the knee in forty-five cadavers, and magnetic resonance imaging of the knee in ten patients and found that the vastus medialis obliquus muscle inserted at a mean angle of 48° at the mid-pole of the patella. They argued that any capsular incision that extends proximal to the middle portion of the patella is not a quadriceps-sparing incision. Berger, in a study of fifty selected total knee arthroplasties that were performed with use of a comprehensive protocol of regional anesthesia, a minimally invasive technique, and oral analgesia, reported that forty-eight patients were discharged on the day of surgery. Similarly, Teeny et al. reported on a comprehensive program that led to a 1.3-day reduction in the mean hospital stay, with no significant difference in rehabilitation outcomes, suggesting that a shortened stay does not compromise recovery or results. Stulberg et al. compared five quadriceps-sparing total knee arthroplasties that were performed with use of modified instrumentation with five conventional total knee arthroplasties and found, with use of an image-free navigation system, greater variability in femoral alignment with use of the thinner minimally invasive femoral intramedullary rod, a tendency for the femoral cutting block to move during resection in the group treated with minimally invasive instrumentation, and a tendency to place the tibial component in excessive varus (average, 3°; range, 0° to 7°) with the minimally invasive alignment jig. The addition of computer assistance in surgery has the potential to allow for better alignment of total knee components. The accuracy of computer-assisted orthopaedic surgery is dependent on the surgeons' ability to accurately and reproducibly identify landmarks. Siston et al.11, in an interesting study in which eleven orthopaedic surgeons used five different alignment techniques to establish femoral rotation in ten cadaveric specimens, found no difference between the mean errors of all five techniques (p > 0.11). They concluded that a navigation system that relies on directly digitizing the femoral epicondyles to establish alignment does not provide a more reliable means of establishing femoral rotational alignment than traditional techniques do. Yau et al. performed a similar cadaveric study to identify the intraobserver errors associated with obtaining visually selected anatomic landmarks that are used for registration in a non-image-based computer-assisted total knee arthroplasty system. They found maximum combined errors to be only 1.32° in the mechanical axis (varus-valgus), 4.17° in the coronal plane (flexion-extension), and 8.2° in the transepicondylar axis (rotational alignment). Fehring et al. reported on several clinical situations in which computer-assisted orthopaedic surgery was very helpful, including the cases of sixteen patients with osseous deformity, retained hardware, or a history of osteomyelitis that prevented the use of standard intramedullary guides. Several investigators reported on the value of computer-assisted orthopaedic surgery in total knee arthroplasty alignment. Kim et al. prospectively compared total knee arthroplasties that were performed with imageless navigation with those that were performed with standard techniques and found similar values for the mean mechanical axis but a larger variation in the manual group (with the alignment being within 2° of neutral in 58% of the procedures performed with the manual technique, compared with 78% of those performed with navigation). Anderson et al. similarly found the mechanical axis to be within 3° of neutral in 95% of procedures performed with navigation and 84% of those performed with the conventional technique. Jenny et al. also found improvement in the accuracy of alignment when 235 total knee arthroplasties that had been performed with use of computer navigation were compared with 235 conventional total knee arthroplasties in a multicenter study. Chin et al., in a randomized, controlled trial comparing extramedullary instrumentation, intramedullary instrumentation, and computer navigation, found that computer navigation had greater consistency and accuracy in terms of implant placement. Finally, Decking et al., in a prospective, randomized study, found a significant improvement in mechanical alignment in the group treated with computer-assisted orthopaedic surgery (p < 0.05) but found no significant difference in terms of the femoral or tibial anteroposterior axis or the posterior slope.
Soft-tissue balancing was the topic of several reports. Bellemans and Victor used fluoroscopy-based spatial navigation to document the normal knee laxity in twelve normal human cadaveric specimens and found the mean medial joint line opening to be 2.6 mm in extension, 5.1 mm at 30° of flexion, and 7.1 mm at 90° of flexion. The mean lateral opening was 3.1 mm in extension, 5.9 mm at 30° of flexion, and 8.1 mm at 90° of flexion. The investigators also evaluated measurements following total knee arthroplasty and found that there was stress relaxation in all cases during the first thirty minutes following ligament balancing and that changing the polyethylene thickness had a marked influence on all laxity measurements. Sugama et al.12, in a study of fifty total knee arthroplasties, demonstrated that preparation of the flexion gap resulted in significant widening of the extension gap (approximately 3 mm) (p = 0.0014), especially medially. They suggested a stepwise and conservative medial release to avoid over-release. Ranawat et al.13 performed an excellent review of surgical techniques that are used to address both the osseous and softtissue deformities associated with severe valgus deformities. Efforts to address blood loss following total knee arthroplasty were assessed in several studies. Keating et al., in a randomized study of ninety patients undergoing total knee arthroplasty, evaluated the use of a bipolar sealing device (BPS5.0-VT bipolar sealer; TissueLink Medical, Dover, New Hampshire) and found a significantly smaller decrease in the postoperative hemoglobin level in the treatment group than in the control group (3.3 compared with 3.8 g/dL) (p = 0.01). The rate of blood transfusion was 0% in the treatment group, compared with 4.4% in the control group. McCoy et al., in a retrospective study, found no significant difference between thirty-seven knees in which an autologous platelet gel was applied to potential sources of bleeding (osseous surfaces, synovial tissue, and the wound) and fifty-one controls with regard to postoperative hemoglobin levels, length of stay, or narcotic usage. Shen et al., in a prospective, randomized study of eighty-nine knees, compared four-hour clamping drainage with nonclamping drainage following total knee arthroplasty and found significantly less drainage (514 compared with 843 mL) (p < 0.05) as well as a smaller decrease in hematocrit in the clamping group, with no increase in morbidity. Bailie et al., in a randomized study of 100 knee arthroplasties that were performed with or without a reinfusion drain, found no significant difference between the groups with regard to blood transfusion or complication rates. Husted and Toftgaard Jensen14 prospectively randomized 100 patients to tourniquet inflation with the knee extended or flexed and found no difference between the groups with respect to patellar tracking. They did find marked improvement in patellar tracking with tourniquet deflation. Inspection of patellar tracking with the tourniquet deflated was associated with a 31% reduction in the number of lateral releases. Benjamin and Chilvers found that the use of the "rule of no thumb" test as an indication for lateral retinacular release resulted in a significant (18%) increase in the number of releases. Two excellent surgical technique articles were published in JBJS supplements. Nelson et al.15 described techniques that are used during revision of the stiff total knee replacement. Burnett et al.16, in a well-illustrated review of extensor mechanism allografting after total knee arthroplasty, emphasized the importance of implanting the allograft with significant tension in extension.
Several authors reviewed the results of total knee arthroplasty in specific patient populations. Palmer et al., in a study of eight patients (mean age, 16.8 years) with juvenile rheumatoid arthritis who underwent fifteen total knee arthroplasties, found excellent pain relief and restoration of function after a mean duration of follow-up of 15.5 years but reported mechanical failure in three knees. Crowder et al. reported on forty-seven total knee replacements that were performed with cement in thirty-two patients with rheumatoid arthritis who were fifty-five years old or younger. The patients were followed for a minimum of fifteen years or until death. The authors identified excellent clinical results and reported six revisions, all of which were performed after seventeen years. Five of the six revisions were performed because of polyethylene wear. Silva and Luck, in a review of ninety total knee arthroplasties in sixty-eight patients with hemophilia, found an infection rate of 16%. Twelve components were removed, with nine being removed because of infection. The ten-year survival rates with component removal for any reason, infection, and mechanical failure as the end point were 83%, 77%, and 96%, respectively. Bai et al., in a study of twenty-five total knee arthroplasties in twenty-one patients with hemophilia, found good restoration of function and improvement in range of motion with no osteolysis or loosening after a mean duration of follow-up of 6.2 years. Rajgopal et al., in a review of eighty-four total knee arthroplasties that had been performed for the treatment of spontaneously ankylosed knees (mean preoperative arc of motion, 14°), reported a mean postoperative arc of motion of 75°, good clinical results, and a complication rate of 9% after an average duration of follow-up of nine years. Haidukewych et al., in a study of seventeen patients who were followed for a mean of five years after total knee arthroplasty for the salvage of failed internal fixation or the treatment of nonunion of the distal part of the femur, reported high complication rates (including a 29% rate of intraoperative complications and a 29% rate of postoperative complications) and a five-year survival rate of 91% with revision for aseptic failure as the end point. Lee et al. reported on twenty-one total knee arthroplasties in patients with Paget disease involving the knee. At a mean of nine years, there was good improvement in the Knee Society pain and function scores, from 41 and 36 points to 87 and 67 points, respectively. One knee was revised because of femoral loosening at ten years. Mullaji et al. reviewed the results of 173 total knee arthroplasties in patients with >20° of preoperative varus deformity. The procedures involved the use of a selective posteromedial release, a reduction osteotomy of the posteromedial flare, and, occasionally, an extra-articular osteotomy. There were three cases of tibial loosening, but the results were otherwise successful. Meding evaluated the results of total knee arthroplasty for the treatment of patellofemoral arthritis and found that the patients achieved the same level of function and pain relief as a group of patients undergoing total knee arthroplasty for the treatment of tibiofemoral arthritis. Lachiewicz and Soileau, in a study of fifty-four primary constrained total knee arthroplasties that were performed for the treatment of knee instability, reported an 84% rate of good or excellent results after a mean duration of follow-up of nine years, with a ten-year survival rate of 96%. Certain medical conditions can alter the success rate following total knee arthroplasty. Booth et al. reviewed the results of fifty-six posterior stabilized total knee arthroplasties that had been performed in patients with Parkinson disease. They purposely produced an extension gap that was 2 mm larger than the flexion gap, performed a thorough posterior release, and performed postoperative botulinum toxin injections in the hamstrings. In addition, the patients used extension slings at night postoperatively. The authors reported good functional results and pain relief, with a mean range of motion of 4° to 110°. Seven knees required manipulation. Cohen et al., in a study on the safety of total joint replacement in cirrhotic patients, concluded that elective replacement can be performed in patients with Child class-A or B cirrhosis but reported a mortality rate of 60% (three of five) in association with the use of emergent total hip arthroplasty for the treatment of fracture. Kreder et al. investigated the safety of elective joint replacement in octogenarians and found that such patients were 3.4 times more likely to die, 2.7 times more likely to sustain a myocardial infarction, and 3.5 times more likely to have development of pneumonia when compared with patients between sixty-five and seventy-nine years of age; however, the overall event rate remained low. The authors concluded that the procedures should be offered to these individuals, provided that the complication rates are acceptable to the patient and family.
Many reports dealt with the prevalence and treatment of stiffness following total knee replacement. Vail et al. found that 27% of patients who had a history of severe stiffness following total knee arthroplasty on one side had development of severe postoperative stiffness following treatment on the contralateral side, suggesting that patient-related factors can contribute to this complication. In a similar vein, Parvizi et al. performed a series of molecular studies on periarticular tissue samples retrieved from patients undergoing revision knee surgery for the treatment of stiffness. They found high levels of reactive oxygen and nitrogen species, which can lead to the breakdown and disorganization of collagen, placing a patient at risk for aggressive scar formation and stiffness. Lombardi et al. reported the results of fifty-one primary total knee arthroplasties in patients with a flexion contracture of at least 20°. The procedures were performed with use of a modular, modern total knee arthroplasty system with increasing constraint for greater degrees of contracture. Full extension was achieved in 71% of the knees. With the increasing trend toward earlier hospital discharge following arthroplasty, it is important to ensure that this approach is not placing patients at risk for serious complications at home. Parvizi et al.17 performed a prospective study of intrahospital complications associated with total joint arthroplasty in 611 patients. The authors identified a total complication rate of 32.4% (198 of 611), with forty-seven of these complications considered major. They found that a vast majority of the major complications occurred within a typical three-day hospital stay and cautioned against early hospital discharge. One potentially devastating complication associated with total knee arthroplasty is patellar tendon rupture. Itala et al., in an experimental canine study of patellar tendon healing, found that healing to a porous tantalum surface was achieved under stable mechanical interface conditions, suggesting that fixation and healing to prosthetic devices may be feasible. Dobbs et al.18 found the prevalence of quadriceps tendon tearing after total knee arthroplasty to be 0.1% (twenty-four of 23,800). They reported that partial tears can be treated successfully nonoperatively and found that the results of operative treatment were poor in seven of the eleven patients with a complete tear. Berry et al. reported on the use of magnetic resonance imaging with metal artifact reduction for the evaluation of thirteen total knee implants and found it to be useful for identifying the extent and location of osteolysis and patellar tendon rupture. They suggested that this tool may become useful for the evaluation of failed total knee arthroplasties. Ritter et al. reviewed 1089 total knee arthroplasties and found a 29.8% rate of notching but found no increased rate of femoral fracture and no significant differences in measured outcomes in association with notching.
Infection is one of the most dreaded complications of total knee replacement. Namba et al. reviewed a community joint registry that included data on 5170 primary and 338 revision total knee arthroplasties and performed a multivariate analysis to determine the efficacy of prophylactic measures and the role of risk factors. They found that laminar flow, body suits, drains, surgical time (length), surgeon volume, and hospital volume had no additional effect beyond the use of preoperative antibiotics. Obesity and higher American Society of Anesthesiology (ASA) scores were associated with a higher risk of infection (p < 0.05). Ritter et al., in a review of primary total knee and hip arthroplasties that were performed at their location between 1995 and 2004, found that individuals with a body-mass index of >35 had a 2.1 times greater risk of infection compared with those with a lower body-mass index and that knees had a 1.5 times greater risk of infection compared with hips. In addition, patients with osteonecrosis and rheumatoid arthritis had a 2.2 times greater risk of infection compared with those with osteoarthritis. The diagnosis of periprosthetic infection was the topic of several studies. Deirmengian et al., in a presentation on their exciting research on synovial fluid gene expression patterns, reported that synovial white blood cells in a septic environment express significantly different genes when compared with white blood cells in aseptic conditions (p < 0.001), making the development of an inexpensive synovial fluid test with high sensitivity and specificity a likelihood in the near future. Di Cesare et al.19, in a prospective, case-controlled study of fifty-eight patients undergoing revision because of infection, found that an elevated serum interleukin-6 level (>10 pg/mL) had a sensitivity, specificity, positive predictive value, negative predictive value, and accuracy of 100%, 95%, 89%, 100%, and 97%, respectively. Shui Ko et al. reported the results of intraoperative frozen-section analysis in a study of forty revision cases. With more than five polymorphonuclear leukocytes per high power field being used as the criterion for infection, the sensitivity, specificity, and positive and negative predictive values were found to be 67%, 97%, 86%, and 91%, respectively. With regard to the treatment of infection, Seldes et al. investigated the use of liquid gentamicin combined with bone cement as a temporary spacer and found that the addition of gentamicin caused a marked reduction in compression strength and tensile strength, with the majority of elution of the drug occurring during the initial twenty-four hours. They concluded that the use of this mixture resulted in substantial cost savings and demonstrated adequate elutional characteristics for use as a temporary spacer. Villanueva et al., in a study of twelve infected knees undergoing a two-stage revision, compared the use of an articulating spacer with the use of a static spacer block. The authors found that the knees that were treated with the articulating spacer had better eventual range of motion, required less extensile exposures, and had higher final Knee Society scores.
Wear and osteolysis remain a major concern following total knee arthroplasty. The Chitranjan Ranawat Award of the Knee Society was given to D'Lima et al. for their study of an instrumented tibial prosthesis that was implanted in an eighty-year-old man. The implant demonstrated that peak tibial forces during walking with use of a walker were 1.2 times body weight on the third postoperative day, 1.7 times body weight on the sixth postoperative day, and 2.2 times body weight at six weeks. The authors anticipate that these data will be used to develop better biomechanical knee models and in vitro wear tests to evaluate and improve implant design, bearing surfaces, rehabilitation protocol, and orthotics. Engh performed multiple regression analysis to determine the relative contributions of variables on wear following almost 2000 total knee arthroplasties and 440 unicompartmental knee arthroplasties. The three variables that were consistently correlated with wear were patient age, mechanical axis alignment, and the shelf age of the component. Stiehl et al. performed a kinematic analysis prior to the revision of a failed total knee implant in five patients and found a direct correlation between wear maps of the retrieved components and abnormal kinematics. Currier et al. performed an analysis of polyethylene implants in one patient and found severe fatigue damage despite a minimal shelf time (four months) before implantation, demonstrating that polyethylene that is sterilized with gamma radiation in air can oxidize in vivo. There were several reports on post damage following posterior stabilized total knee arthroplasty. Van Citters et al., in a study of forty-two retrieved polyethylene posterior stabilized tibial inserts with a central post, of five different designs, found visible evidence of contact on the posterior face in 64% of the implants, impingement across the anterior face in 43%, deformation of the anterior corners in 88%, and impingement on the top of the post in 74%. Rubash et al. analyzed nine patients with posterior stabilized total knee implants with use of a dual-orthogonal fluoroscopic imaging system and found anterior post contact at full extension in six patients. Two knee designs were reported to have high failure rates related to tibial post failure. Sugimoto et al. reviewed thirty-seven Interax total knee implants (Stryker, Kalamazoo, Michigan) and found that premature failure of the polyethylene bearing surface occurred in nine patients at twenty to thirty-eight months postoperatively. The failures were attributed to the inadequate implant design of small components and to gamma irradiation of the polyethylene in air. Bal et al. reported on 564 consecutive total knee arthroplasties that had been performed with use of the Encore Foundation 100 Series PS Total Knee System (Encore Medical, Austin, Texas). The polyethylene implants had been sterilized with gamma radiation in an oxygen environment. The authors reported seventy cases of post breakage at a mean of forty months. Backside wear was the topic of several reports. Collier et al., in a study of 365 posterior cruciate ligament-retaining total knee implants, found that the prevalence of osteolysis at five to ten years was 34% when polyethylene that had been sterilized with gamma radiation in air was used on a grit-blasted titanium base and 9% when polyethylene that had been sterilized with gamma radiation in air or polyethylene that had been sterilized with gas plasma was used on a polished cobalt-chromium base. Logistic regression analysis showed that osteolysis was associated with male gender, the use of a grit-blasted titanium tibial base, three polyethylene-related factors (the variety from which it had been machined, the sterilization method, and the shelf age), and femoral component hyperextension. Atwood et al. studied forty-seven retrieved polyethylene implants from knees that had been treated with a Low Contact Stress Rotation Platform total knee implant (DePuy) and found backside abrasive wear of the polyethylene and scratching of the cobalt-chromium tray, pointing to the presence of debris at this interface with resultant third-body wear. Mayor et al. studied eighty-five retrieved PFC bearings (Johnson and Johnson) after nine to 192 months and found that backside wear was consistently greater on the posterior aspect of the implant and near zero on the anterior aspect, with volumetric wear averaging 120 mm3/year. Billi et al. reported on the relative influences of metal surface finish, alloy, and micromotion amplitude on wear in a study of twenty-four implants that were subjected to simulator testing and found the greatest reduction in backside wear to be associated with a highly polished tibial base. Conditt et al.20 studied a series of retrieved total knee components of one design and concluded that the mean volumetric backside wear was 138 ± 95 mm3/year. This amount may be sufficient to induce osteolysis.
Kurtz et al. used the National Inpatient Sample (from 1990 to 2002) and United States Census data to quantify historical trends and to make future projections regarding primary and revision joint arthroplasty. They anticipated that the number of total hip replacements would double (when compared with 2005 numbers) by the year 2026 and that the number of total knee arthroplasties would double by 2016. The number of revision total knee arthroplasties was projected to increase from 37,000 in 2005 to 195,000 in 2030 (a 522% increase). The authors concluded that this demand necessitates a combination of increased economic resources, operative efficiency, technical capacity, and implant longevity. Revision in specific clinical scenarios was reviewed by several authors. Whitesides, in a review of forty-nine knees with a cementless total knee implant that underwent isolated polyethylene exchange for wear, reported only three failures. Keeney et al., in a study of thirty-four knees that were revised with bone-grafting because of large osteolytic lesions, found improvement in terms of pain, flexion, and clinical knee scores, with only one failure, after a mean duration of followup of thirty-three months. Reis et al. reported on patellar augmentation with use of a porous tantalum implant in a study of nineteen knee revisions. The authors reported that failure occurred in six of the eight cases in which residual patellar bone stock was absent. In the eleven cases in which the augment had at least 50% osseous support, all implants remained stable at twelve months. Hanssen et al. reported on the use of porous tantalum cones to address severe tibial bone loss in a study of fifteen knees that were revised with a stemmed component and found osseous integration in all cases, with no reoperations for the treatment of tibial complications after a minimum duration of follow-up of two years. Lotke and Puri, in a review of sixty-four revision total knee arthroplasties in knees with a large bone defect requiring impaction bone-grafting, reported no mechanical failures. Steens et al. analyzed the use of morselized allograft for the treatment of contained defects in a study of thirty-seven knees that were revised with a stemmed implant. The authors reported two failures, with 57% of patients having pain and more than two-thirds of the patients being dissatisfied with the result. They concluded the technique should be viewed critically. Hockman et al., in a study of sixty-five consecutive revision total knee arthroplasties, identified nine failures after a minimum duration of follow-up of five years. The rate of failure following revisions that had been performed with use of bulk allograft was lower than that following revisions that had been performed without bulk allograft (19.2% compared with 42.9%). The authors concluded that modular augments did not effectively address the bone loss and instability encountered in many instances. Stem fixation was the topic of two reviews. Gallagher et al., in a study of 115 consecutive revision total knee arthroplasties that had been performed without cement, reported that six implants were re-revised because of loosening and twenty-four had radiographic evidence of loosening at a mean of five years. They recommended the use of cemented stems for revision total knee arthroplasty with increased constraint. Peters et al.21, in a study of fifty consecutive revision total knee arthroplasties that had been performed with use of metaphyseal cement and a press-fit, diaphyseal engaging cementless stem that was 80 to 160 mm long, reported a 9% rate of reoperation for the treatment of infection but noted no instances of clinical or radiographic loosening after a mean duration of follow-up of thirty-six months. Knee arthrodesis techniques were the topic of three reviews. Crockarell and Mihalko, in a report on fifteen knee arthrodeses that had been performed with use of an intramedullary nail, reported a 100% rate of union and noted six complications (four cases of symptomatic hardware, one case of trochanteric bursitis, and one infection). Kuo et al., in a report on three infected knees that were treated with arthrodesis with use of dual locking plates, reported union in all cases. McQueen et al., in a report on eleven knees that were treated with arthrodesis with use of an intramedullary compression nail, reported union in all cases.
The editorial staff of The Journal reviewed a large number of recently published research studies related to the musculoskeletal system that received a Level of Evidence grade of I. Over 100 medical journals were reviewed to identify these articles, which all have high-quality study design. In addition to articles cited already in this Update, six additional level-I articles were identified that were relevant to adult reconstructive knee surgery. A list of those titles is appended to this review after the standard bibliography. We have provided a brief commentary about each of the articles to help to guide your further reading, in an evidence-based fashion, in this subspecialty area.
Witt C, Brinkhaus B, Jena S, Linde K, Streng A, Wagenpfeil S, Hummelsberger J, Walther HU, Melchart D, Willich SN. Acupuncture in patients with osteoarthritis of the knee: a randomised trial. Lancet. 2005;366:136-43.
Patients with knee osteoarthritis (Kellgren grade Petersen MM, Gehrchen PM, Ostgaard SE, Nielsen PK, Lund B. Effect of hydroxyapatite-coated tibial components on changes in bone mineral density of the proximal tibia after uncemented total knee arthroplasty: a prospective randomized study using dual-energy x-ray absorptiometry. J Arthroplasty. 2005;20:516-20. Sixteen patients were randomized to receive a tibial component either with a hydroxyapatite coating (eight patients) or without a hydroxyapatite coating (eight patients) during routine total knee arthroplasty. The authors then prospectively measured bone density in four areas of interest in the proximal part of the tibia. At two years postoperatively, the only significant difference in bone mineral density was in the lateral tibial condyle, where it had increased by 6.15% in patients with tibial components without hydroxyapatite coating. The authors concluded that hydroxyapatite coating had no significant effect on the bone-remodeling pattern of the proximal part of the tibia. The most apparent weakness of this report was the limited number of patients and the limited (two-year) follow-up. Richards JD, Sanchez-Ballester J, Jones RK, Darke N, Livingstone BN. A comparison of knee braces during walking for the treatment of osteoarthritis of the medial compartment of the knee. J Bone Joint Surg Br. 2005;87:937-9. The use of a simple hinged brace was compared with the use of a valgus corrective brace in this crossover study of twelve patients with Larsen grade-2 to grade-4 osteoarthritis. Knee kinematics, ground-reactive forces, pain, and function were assessed during walking. Significant improvements in terms of pain, function, and loading were seen in association with the valgus brace (p < 0.05). Treatment with a simple brace was only associated with improvements in loading forces. The authors concluded that the valgus brace showed greater benefit. Kim YH, Sohn KS, Kim JS. Range of motion of standard and high-flexion posterior stabilized tota |
Top
Introduction
Treatment of Osteoarthritis...
2) were randomized to
acupuncture (150 patients), minimal acupuncture (superficial needling at
nonacupuncture sites) (seventy-six patients), or a waiting list control
(seventy-four patients). Patients in the acupuncture and minimal acupuncture
groups underwent twelve treatments over eight weeks. All patients completed
WOMAC questionnaires after eight, twenty-six, and fifty-two weeks. The mean
baseline-adjusted WOMAC score at eight weeks was 26.9 in the acupuncture
group, 35.8 in the minimal acupuncture group, and 49.6 in the waiting list
group (p < 0.0002). After fifty-two weeks, the difference between the
acupuncture group and the minimal acupuncture group was no longer significant
(p = 0.08). The authors concluded that after eight weeks of treatment, pain
and joint function are improved more with acupuncture than with minimal
acupuncture or no acupuncture in patients with osteoarthritis of the knee.
However, this benefit decreases over time.