The purpose of this review is to discuss the research presented on selected topics related to adult knee reconstruction during the year 2008. The articles referenced in this update were selected from both The Journal of Bone and Joint Surgery (American Volume) and The Journal of Arthroplasty, as well as select articles from Clinical Orthopaedics and Related Research. All prospective randomized studies related to total knee arthroplasty that were published in these journals in 2008 were reviewed. Other studies were included on the basis of their relevance to the topics that were chosen to be discussed.
A variety of approaches to the knee have been successfully utilized for total knee arthroplasty. There has been recent interest in improving these approaches to limit soft-tissue disruption, to reduce pain, and to avoid some of the functional delays in recovery observed after total knee arthroplasty, but without paradoxically causing more local tissue trauma resulting from excessive traction due to limited exposure. Although certain early functional benefits appear to be derived from less-invasive approaches to total knee arthroplasty, it is not clear whether there will be long-term benefit or detriment.
The midvastus approach involves an arthrotomy that is performed by dissecting diagonally away from the superomedial aspect of the patella into the vastus medialis muscle. Dalury et al.11 conducted a prospective randomized study to compare the electromyographic evaluation of the vastus muscle, radiographs, range-of-motion tests, and Knee Society scores for twenty patients undergoing bilateral total knee arthroplasty (with one arthroplasty being performed through a midvastus approach and one being performed through a medial parapatellar approach). At six weeks and at the time of the latest follow-up at twelve weeks, there were no significant differences in any of the measured parameters and there were no electromyographic or nerve-conduction abnormalities in the knees that had the midvastus approach. However, at the six-week follow-up, six of the twenty patients expressed a preference for the total knee implant that had been inserted through the midvastus approach, one had a preference for the implant that had been inserted through the parapatellar approach, and the remainder had no preference. It appears that the midvastus approach is a safe alternative to the parapatellar approach and may offer some early benefits. The authors suggested that sharp dissection of the vastus, with a closure that includes the capsule but excludes the muscle, may explain the avoidance of electromyographic abnormalities after the midvastus approach.
The mini-subvastus approach, in which the arthrotomy is extended medially under the vastus medialis muscle, avoids incision of the quadriceps muscle and tendon. Schroer et al.12 retrospectively compared 150 total knee arthroplasties that had been performed through the mini-subvastus approach with 150 total knee arthroplasties that had been performed through the traditional medial parapatellar approach. Both groups of patients had similar characteristics and an identical perioperative management protocol. The authors found that patients who had been managed with the mini-subvastus approach were discharged earlier (3.4 compared with 4.1 days; p < 0.05), exhibited a more rapid recovery of quadriceps strength, were less likely to require inpatient rehabilitation, and had improved knee flexion over a two-year period. Comparison of the groups also showed no differences in terms of the number or severity of complications. In a subsequent study, Schroer et al.13 compared implant positioning in patients who underwent total knee arthroplasty with either a traditional medial parapatellar or a mini-subvastus approach, with use of computed tomography scans to determine alignment. Interestingly, there was a trend for increased accuracy in tibial alignment in the mini-subvastus group, perhaps resulting from improved tibial exposure. However, the mini-subvastus approach was also associated with a higher variance in average coronal femoral alignment (1.71° compared with 1.04°). The authors postulated that the increased variation in femoral alignment was due to shortcomings of the instrumentation and not to a lack of exposure.
McAllister and Stepanian14 reported data from a retrospective study in which the early results of 100 consecutive minimally invasive total knee arthroplasties were compared with those of 100 consecutive conventional total knee arthroplasties. The minimally invasive group was defined by avoidance of patellar eversion and tibial dislocation, and the extremity was ideally positioned throughout surgery to maximize visualization and minimize retraction. A medial parapatellar arthrotomy was used in both groups, as was a standardized perioperative protocol of pain management, patient education, and hospital discharge. Patients who had a total knee arthroplasty without patellar eversion or tibial dislocation had a shorter inpatient stay (3.6 compared with 6.4 days; p < 0.01), less pain at the time of discharge as measured with a visual analog scale (3.2 compared with 3.8; p < 0.01), improved flexion for six months postoperatively (122° compared with 113°; p < 0.0001), and a lower rate of manipulation (2% compared with 14%; p < 0.001). By one year after surgery, there were no differences between the groups in terms of range of motion or functional outcome. There were no significant differences between the groups in terms of the complication rate or component positioning. It appears that the avoidance of patellar eversion and tibial dislocation improves the early results of total knee arthroplasty and reduces the need for manipulation in most patients. It must be emphasized that this technique requires specific technical considerations and is safe only when used by a properly educated surgeon.
The discussion about the role of computer navigation in total knee arthroplasty focuses on two primary issues. One is whether navigation improves the accuracy of implant positioning and extremity alignment; the other is whether improvements in the accuracy of implant positioning and alignment enhance clinical results and durability. Several studies that were published in 2008 addressed both of these issues.
Lionberger et al.15 performed forty-six total knee arthroplasties with use of either an electromagnetic or an infrared navigation system. Postoperatively, component and limb alignment were assessed with both computed tomography and full-length radiographs. The authors found that the navigation systems resulted in comparable accuracy in terms of implant positioning, attaining <1° accuracy of the coronal and sagittal component positioning in 95% of the knees. The desired mechanical limb alignment was achieved in 93% of the procedures involving the use of electromagnetic navigation and 90% of the procedures involving the use of infrared navigation. The authors found that the femoral epicondylar axis was difficult to determine, leading them to suggest additional studies on establishing proper femoral rotation. Siston et al.16 similarly identified the difficulty of establishing the rotational axis of the distal part of the femur. They conducted a cadaver study in which they tested a variety of techniques (two kinematic and three anatomic) to establish femoral rotation. The anatomic techniques involved the use of the digitized epicondyles, the Whiteside line, and the posterior condylar line. Computed tomography of each cadaver knee was performed to determine the femoral rotational axis. The authors found that the kinematic methods of defining femoral rotation were neither more accurate nor more precise than the anatomic methods. Overall, the combined methods of attaining the femoral axis of rotation were more accurate than each individual method alone. For example, averaging the results of the epicondylar axis and the Whiteside line (mean error, 1.5°) was more accurate than each individual technique alone (mean error, 5.4° and -2.3°, respectively). The authors recommended the use of combined techniques to improve the accuracy of femoral component implantation when using navigation.
Catani et al.17 measured the alignment deviation between osseous resection and final implant positioning with use of computer-navigation techniques. When bone resection was completed, the instrumented probe was used to measure the alignment of the tibial and femoral surfaces in three planes. After the final implants were cemented, the same measurements were repeated to allow the final implant positioning to be compared with that intended by the bone resection. The authors found that cementation and impaction of the components can result in further alteration in the final implant positions. This variation was >1° in the frontal plane of the femur in 20% of the cases, in the frontal plane of the tibia in 11%, and in the sagittal plane of the tibia in 33%. The deviation was found to be >2° in the sagittal plane of the tibia in 9% of the cases. Even with emerging technologies that provide for more accurate tibial and femoral bone resection, it must be kept in mind that there is still some error in alignment during component implantation when performing a total knee arthroplasty.
Dutton et al.18 conducted a prospective randomized study that included 180 consecutive patients undergoing either a minimally invasive computer-assisted total knee arthroplasty or a traditional total knee arthroplasty. The minimally invasive group had an abbreviated parapatellar arthrotomy with patellar subluxation instead of eversion. All patients received the same perioperative pain-management protocol and adhered to the same discharge guidelines. Despite having a longer operative time (by an average of twenty-four minutes), the patients in the minimally invasive group were discharged from the hospital an average of one day earlier than those in the traditional total knee arthroplasty group (3.3 compared with 4.5 days; p < 0.001) and were more likely to walk independently for thirty minutes by one month after surgery (p = 0.04). The mechanical axis was within 3° of neutral in 92% of the patients in the study group, compared with only 68% of those in the traditional total knee arthroplasty group. The authors observed no complications related to minimally invasive computer-assisted total knee arthroplasty. Carter et al.19, in a retrospective study, described their initial experience with computer-assisted navigation and total knee arthroplasty in a community setting. Implant alignment was determined with computed tomography after 100 total knee arthroplasties performed with navigation and 100 conventional total knee arthroplasties. The authors found that more total knee replacements that had been performed with navigation demonstrated implant positioning within 3° of the surgical goal when compared with conventional total knee replacements, with the largest improvements being observed in the sagittal alignment of the implants.
These studies demonstrated greater consistency in the achievement of neutral bone resection during total knee arthroplasties performed with use of computer navigation as compared with those performed with conventional techniques. However, whether the accuracy in bone resection corresponds with the alignment achieved at the time of implantation is in question17. Additionally, it remains to be seen whether these improvements in bone resection will impact functional outcomes and implant durability.
Improvements in the perioperative management of patients, including pain-management protocols and patient education, have significantly improved the early outcomes after total knee arthroplasty. There has been a trend toward the use of multimodal pain protocols, the avoidance of parenteral narcotics, and the encouragement of early walking in an effort to reduce postoperative pain, the length of inpatient stay, and the prevalence of medical complications such as ileus and deep-vein thrombosis.
Dorr et al.20 reported the results of a prospective study involving thirty-five patients who were managed postoperatively with an epidural anesthetic and thirty-five patients who were managed with a femoral nerve block. These methods were augmented by an oral pain-management protocol that avoided parenteral narcotics, involved the use of preemptive oral analgesics, and included an intraoperative injection of the pericapsular tissues. The authors found that the average daily pain score after surgery was 4 of 10. Postoperative nausea occurred in eight patients (11%) on the first postoperative day. There were no cases of ileus or urinary retention. The authors emphasized the safety, efficacy, and decrease in narcotic side effects when avoiding parenteral narcotics after total knee arthroplasty.
Lavernia et al.21 also reported the results of a multimodal pain-management protocol after total knee arthroplasty. In a retrospective study, 778 procedures in which the patient was managed with traditional pain control were compared with 358 procedures in which the patient received a multimodal pain-management protocol. The pain-management protocol included preemptive treatment with celecoxib, controlled-release oxycodone, acetaminophen, and ondansetron. The patients also received a femoral nerve block and spinal anesthesia. Intraoperatively, the pericapsular tissues were injected with a mixture of Duramorph (morphine sulfate) (Wyeth-Ayerst Laboratories, Philadelphia, Pennsylvania) (10 mg/2 mL), Toradol (ketoralac tromethamine) (Roche Pharmaceuticals, Nutley, New Jersey) (30 mg/3 mL), and bupivacaine (0.25%/10 mL). Postoperatively, pain was treated with oral medications, with the avoidance of parenteral narcotics. The authors reported a significantly decreased rate of knee manipulation in patients receiving the multimodal pain-management protocol (2.24% compared with 4.75%; p = 0.0001). It is clear from a number of studies, including those by Dorr et al.20 and Lavernia et al.21, that multimodal pain management and the avoidance of parenteral narcotics results in improved postoperative pain control and better outcomes without an increase in complications.
Wood et al.22, in a prospective randomized study, addressed the use of autologous reinfusion drains after total knee arthroplasty. Eighty patients who underwent a hip or knee replacement were randomized to have removal of a reinfusion drain at six or twenty-four hours after surgery. Autologous blood was only reinfused if >200 mL was collected from the drain at six hours after surgery. The authors monitored the postoperative drop in hemoglobin level, hospital stay, transfusion rates, complications, wound-healing, and rehabilitation in both groups. They found no significant differences between the groups and noted that postoperative hemoglobin levels and transfusion rates appeared to have no association with the timing of reinfusion drain removal. Therefore, early drain removal seems prudent.
There is ongoing controversy about prophylaxis against venous thromboembolism after total knee arthroplasty. In recent years, prophylaxis against venous thromboembolism has become a performance measure that has been used to evaluate hospitals and surgeons, resulting in an increased use of potent anticoagulants after total knee arthroplasty, some of which are associated with a substantial risk of bleeding that must be weighed against the risk of venous thromboembolism and its sequelae.
Novicoff et al.23 conducted a retrospective review of their experience with anticoagulation after hip and knee arthroplasty in a hospital in which >500 arthroplasties are performed per year. In response to the American College of Chest Physicians (ACCP) guidelines, their hospital mandated prophylaxis against venous thromboembolism starting in July 2005. Patients undergoing a total knee arthroplasty were given warfarin (with a target international normalized ratio of 2.0) on the night of surgery, and those at high risk of venous thromboembolism were also given 40 mg of a low-molecular-weight heparin starting twelve hours after surgery. All patients having a total hip replacement received both warfarin and low-molecular-weight heparin. In order to comply with the ACCP guidelines, the target international normalized ratio was changed to 2.5 in January 2006. The authors found that the rate of complications, including hematoma, seroma, and hemorrhage, increased from 1.4% before the mandated changes to 9.6% after the mandated prophylaxis was initiated. In addition, the rate of readmission increased from 2.2% to 4.4% after the mandate. The rate of venous thromboembolism remained unchanged throughout the study period. After elimination of the use of the low-molecular-weight heparin, except in high-risk patients, and setting of the target international normalized ratio at or below 2.0, a marked decrease in the rate of complications was once again attained.
The use of a multimodal anticoagulation protocol in standard or lower-risk patients was studied by Callaghan et al.24. In that study, 312 consecutive total knee arthroplasties were performed for patients who were considered to have a low risk for venous thromboembolism. Patients were excluded if they had a personal or family history of venous thromboembolism, had a hypercoagulability disorder, had severe venous insufficiency, or were receiving hormonal supplementation. A multimodal approach involving the use of aspirin as the primary mode of chemoprophylaxis was found to be successful, with no deaths for any reason, no readmissions, and no reoperations due to bleeding within ninety days of follow-up. The authors reported that 73% of the patients in the study period had qualified as having a low risk for venous thromboembolism and recommended consideration of this strategy for such patients.
These studies illustrate the disparity between agents with regard to the balance between efficacy in preventing venous thromboembolism and safety in minimizing hemorrhagic complications, which is a very important consideration in the decision regarding the selection of thromboprophylaxis. Galat et al.25 retrospectively reviewed the results for forty-two patients (forty-two knees), seen over a twenty-three-year period, who had a reoperation for the evacuation of a hematoma within thirty days after primary total knee arthroplasty. The study was conducted with matched case controls in order to compare potential risk factors and outcomes. In patients who underwent a reoperation for the treatment of a hematoma after total knee arthroplasty, the subsequent two-year probabilities of requiring another major operation (amputation, muscle flap coverage, or implant removal) or having development of a deep infection were 12% and 11%, respectively. In knees that did not have a reoperation because of bleeding, the rates of subsequent major surgery and infection were only 0.6% and 0.8%, respectively. There is clearly a large penalty associated with increasing bleeding rates after total knee arthroplasty. Although the importance of prophylaxis against venous thromboembolism cannot be overstated, there is also a duty to balance benefits with risks.
Chelly and Schilling26, in a retrospective study of 3588 patients managed with arthroplasty, assessed the risk of hematoma associated with the use of peripheral nerve blocks when combined with thromboprophylaxis. The blocks included continuous lumbar plexus, continuous femoral, and single sciatic block. A total of 6935 blocks were performed in patients who were managed with warfarin (50.0%), fondaparinux (12.8%), dalteparin (11.6%), enoxaparin (1.8%), or aspirin (23.8%). Perineural catheters were removed on the second or third postoperative day, and no perineural hematomas were identified. The authors concluded that perineural catheters can be inserted before the start of anticoagulation and can be safely removed while the patient is receiving prophylaxis against venous thromboembolism.
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, all of which have high-quality study design. In addition to articles already cited in this update, three additional level-I articles were identified that were relevant to this specific review. A list of those articles is appended to this review following the standard bibliography. We have provided a brief commentary about each of the articles to help guide your further reading, in an evidence-based fashion, in this subspecialty area.
Wylde V, Learmonth I, Potter A, Bettinson K, Lingard E. Patient-reported outcomes after fixed- versus mobile-bearing total knee replacement: a multi-centre randomised controlled trial using the Kinemax total knee replacement. J Bone Joint Surg Br. 2008;90:1172-9. Erratum in: J Bone Joint Surg Br. 2008;90:1534.
This multicenter randomized controlled study of 250 knees from four centers compared the two-year outcomes of mobile-bearing and fixed-bearing total knee arthroplasties. Patients were evaluated at intervals of eight weeks, twelve weeks, one year, and two years after the index procedure with regard to a variety of measures, including mental health scores, knee outcome scores, quality-of-life scores, and a validated scale of postoperative satisfaction. The patients had a mean age of sixty-eight years, and 55% of the patients were female. There were no significant differences at any time interval between patients with mobile and fixed-bearing knees. This large study suggests that, at least in the short term, there is no observable difference in outcomes between mobile and fixed-bearing knees. We await the longer-term results for these two patient groups.
Amin A, Watson A, Mangwani J, Nawabi D, Ahluwalia R, Loeffler M. A prospective randomised controlled trial of autologous retransfusion in total knee replacement. J Bone Joint Surg Br. 2008;90:451-4.
The authors conducted a prospective randomized controlled trial to investigate the efficacy of autologous blood retransfusion after total knee arthroplasty. Ninety-two patients who received an autologous retransfusion after surgery were compared with eighty-six patients who were managed with a standard vacuum drain. Hemoglobin levels were measured at twenty-four, forty-eight, and seventy-two hours postoperatively, and the need for allogeneic blood transfusion was monitored. The clinical trigger for transfusion was based on hemoglobin levels and clinical symptoms. The authors found no significant difference in mean hemoglobin levels after surgery and found no difference in the need for allogeneic transfusion after total knee arthroplasty. On the basis of this study, it appears that the retransfusion of drainage blood after total knee arthroplasty does not augment the hemoglobin level or prevent allogeneic transfusion. The technique of autologous retransfusion may be more useful in the setting of revision joint arthroplasty, when larger volumes of blood may be retransfused.
Karachalios T, Giotikas D, Roidis N, Poultsides L, Bargiotas K, Malizos KN. Total knee replacement performed with either a mini-midvastus or a standard approach: a prospective randomised clinical and radiographic trial. J Bone Joint Surg Br. 2008;90:584-91.
The difference between standard and minimally invasive approaches to total knee arthroplasty has been the subject of many studies. This prospective randomized study compared the early results of a standard total knee arthroplasty technique with those of a midvastus technique. Over a three-year period, 100 patients were randomized to one of the two procedures and were followed clinically and radiographically for an average of twenty-three months. The patients who were randomized to the midvastus approach demonstrated significantly better mean functional scores, mean Oxford knee scores, and mean total scores for nine months after the index procedure. However, the authors also noted that more patients who were randomized to the midvastus approach had a malpositioned component. These results generally agree with those of most studies that have compared minimally invasive and standard approaches to total knee arthroplasty. Although the early recovery after total knee arthroplasty appears to be improved for about three to nine months in patients who are managed with a midvastus approach, there may be a simultaneous increase in the overall number of malpositioned implants.