© 2009 the American Academy of Orthopaedic Surgeons Locking Plates for Extremity FracturesThe Technology Overview was prepared by an AAOS physician task force using systematic review methodology and summarizes the findings of studies published as of April 1, 2008, on locking plates for extremity fractures. As a summary, this document does not make recommendations for or against the use of locking plates for extremity fractures. It should not be construed as an official position of the American Academy of Orthopaedic Surgeons. Readers are encouraged to consider the information presented in this document and reach their own conclusions about locking plates for extremity fractures. The Technology Overview was adopted by the Board of Directors of the American Academy of Orthopaedic Surgeons on December 6, 2008. The American Academy of Orthopaedic Surgeons has developed and is providing the Technology Overview as an educational tool. Patient care and treatment should always be based on a clinicians independent medical judgment given the individual patients clinical circumstances. Dr. Anglen is Professor and Chairman, Department of Orthopaedic Surgery, Indiana University Medical Center, Indianapolis, IN. Dr. Kyle is Chair, Department of Orthopaedic Surgery, Hennepin County Medical Center, Minneapolis, MN. Dr. Marsh is Professor, Department of Orthopaedics and Rehabilitation, University of Iowa Hospitals and Clinics, Iowa City, IA. Dr. Virkus is Orthopaedic Surgeon, Department of Orthopedic Surgery, Section of Orthopedic Oncology, Rush University Medical Center, Chicago, IL. Dr. Watters is Orthopaedic Surgeon, Bone and Joint Clinic of Houston, Houston, TX. Dr. Keith is Professor, Orthopaedics and Biomedical Engineering, and Chief, Hand Surgery, MetroHealth Medical Center, Case Western Reserve University, Cleveland, OH. Dr. Turkelson is Director, Department of Research and Scientific Affairs, American Academy of Orthopaedic Surgeons, Rosemont, IL. Ms. Wies is Manager, Clinical Practice Guidelines Unit, Department of Research and Scientific Affairs, American Academy of Orthopaedic Surgeons. Mr. Boyer is Research Analyst, Guidelines, Department of Research and Scientific Affairs, American Academy of Orthopaedic Surgeons. Dr. Anglen or a member of his immediate family serves as a board member, owner, officer, or committee member of American Board of Orthopaedic Surgery, American College of Surgeons, and Orthopaedic Trauma Association; is affiliated with the publications Journal of the American Academy of Orthopaedic Surgeons and Journal of Orthopaedic Trauma; has received royalties from Biomet; serves as a paid consultant to or is an employee of Stryker; has received research or institutional support from Stryker and Wyeth; and has received nonincome support (such as equipment or services), commercially derived honoraria, or other non–research-related funding (such as paid travel) from the Journal of the American Academy of Orthopaedic Surgeons. Dr. Kyle or a member of his immediate family serves as a board member, owner, officer, or committee member of Twin Cities Orthopaedic Education Association, Minneapolis Medical Research Foundation, Midwest Orthopaedic Research Foundation, Hennepin Faculty Associates, and Millennium Medical Technologies; is affiliated with the publication Journal of Shoulder and Elbow Surgery; has received royalties from DePuy, Encore Medical, Smith & Nephew, and Zimmer; and has received research or institutional support from DePuy. Dr. Marsh or a member of his immediate family serves as a board member, owner, officer, or committee member of Orthopaedic Trauma Association; has received royalties from Biomet; has received research or institutional support from Smith & Nephew; and has received nonincome support (such as equipment or services), commercially derived honoraria, or other non–research-related funding (such as paid travel) from Oxford University Press and Smith & Nephew. Dr. Virkus or a member of his immediate family serves as a paid consultant to or is an employee of Stryker; has received research or institutional support from Stryker; has stock or stock options held in Stryker; and has received nonincome support (such as equipment or services), commercially derived honoraria, or other non–research-related funding (such as paid travel) from Stryker. Dr. Watters or a member of his immediate family serves as a board member, owner, officer, or committee member of Bone and Joint Decade, USA, North American Spine Society, Intrinsic Therapeutics, Work Loss Data Institute, and American Board of Spine Surgery; is affiliated with the publication The Spine Journal; serves as a paid consultant to or is an employee of Blackstone Medical, Medtronic Sofamor Danek, Stryker, Intrinsic Therapeutics, and McKessen Health Care Solutions; and has stock or stock options held in Intrinsic Therapeutics. Ms. Wies or a member of her immediate family has stock or stock options held in Shering Plough. None of the following authors or a member of their immediate families has received anything of value from or owns stock in a commercial company or institution related directly or indirectly to the subject of this article: Dr. Keith, Dr. Turkelson, and Mr. Boyer. Evidence tables displaying the raw data and information extracted for the Technology Overview are available in a supplemental document available on the AAOS Website, www.aaos.org/technologyoverviews.
Thirty-three peer-reviewed studies met the inclusion criteria for the Overview. Criteria were framed by three key questions regarding indications for the use of locking plates, their effectiveness in comparison with traditional nonlocking plates, and their cost-effectiveness. The studies were divided into seven applications: distal radius, proximal humerus, distal femur, periprosthetic femur, tibial plateau (AO/OTA type C), proximal tibia (AO/OTA type A or C), and distal tibia. Patient enrollment criteria were recorded to determine indications for use of locking plates, but the published studies do not consistently report the same enrollment criteria. Regarding effectiveness, there were no statistically significant differences between locking plates and nonlocking plates for patient-oriented outcomes, adverse events, or complications. The literature search did not identify any peer-reviewed studies that address the cost-effectiveness or cost-utility of locking plates. Internal fixation plates function as splints for a fractured long bone and perform a variety of specific mechanical functions, including compression, buttressing, bridging, and neutralization (ie, protection). Both locking and nonlocking plates can be used to perform any of these functions. Nonlocking plates stabilize bone fragments against deforming forces by the use of friction between the plate and bone, generated by screws that compress the two surfaces together. Locking plates stabilize bone fragments by means of the attachment of the screw to the plate in a rigid, fixed-angle coupling, usually accomplished with threads in the screw head, plate hole, or both. This locking of screw to plate makes the fixation construct more resistant to failure from sequential screw loosening and pullout. Because all the screws in a single bone fragment are locked to the plate at fixed angles, they must fail (ie, pull out) as a unit rather than individually and sequentially.1,2 This feature may be of particular advantage in osteoporotic bone with thinner cortices; in this situation, nonlocking screws cannot generate as much plate-to-bone compression, so the frictional forces resisting motion are less. In addition, the fixed-angle nature of the plate-and-screw fixation resists cantilever bending stresses and reduces the risk of angular deformity in metaphyseal fractures that are comminuted, missing bone, or otherwise mechanically unable to share load. It has been proposed that the reduced plate-to-bone compression afforded by locking plates serves to protect the viability of the bone by maintaining microvascular circulation within the cortex and its investing tissues.3 The clinical importance of this theoretic advantage, however, is unproven. Both traditional (nonlocking) and locking plates can be inserted through less invasive surgical techniques, also known as percutaneous, submuscular, or minimal incision plating. The introduction of locking plate technology has temporally coincided with the development of these less invasive surgical techniques, but it is important that the two concepts be kept separate. Newer surgical techniques that involve smaller incisions, less soft-tissue dissection, less periosteal stripping, and use of intraoperative imaging or navigation are believed by many to improve healing rates and reduce complications, but they are not dependent on the use of locking plate implants. Similarly, the theoretic biomechanical advantages of locking plates (eg, reduced screw pullout in osteopenic bone, reduction of cantilever bending, reduction of angular deformity in deficient metaphyseal bone) are not dependent on minimally invasive surgical techniques. Locking plates can be put in through traditional exposures.
We used systematic processes to locate published studies relevant to this topic. These processes began with the framing of three key questions (see below). We next developed article inclusion criteria (Appendix I, available online at www.aaos.org/technologyoverviews) and then conducted systematic literature searches (Appendix II, available online at www.aaos.org/technologyoverviews). Articles were included only when they met our a priori criteria. A level of evidence was assigned to each article included in this Technology Overview.
Our searches identified 452 citations. Of these, 33 met all inclusion criteria and were used to address the questions below. Four of the studies compared the outcomes of patients receiving locking plates and patients receiving nonlocking plates. One additional study compared the outcomes of patients receiving locking plates and patients receiving intramedullary nailing (for the purposes of this Overview, only the outcomes of the locking plate patients in this study were considered). The remaining 28 studies were case series and reported outcomes only for patients receiving locking plates. The studies were divided into seven applications: distal radius, proximal humerus, distal femur, periprosthetic femur, tibial plateau (AO/OTA type C only), proximal tibia (AO/OTA type A or C), and distal tibia. There were no studies addressing the application of locking plates for extremity fractures in patients with osteoporosis that met all of the inclusion criteria. Several published studies discuss osteoporosis in their study populations; however, none of these studies adequately reports quantitative data, for the outcomes of interest to this Overview, in patients with osteoporosis. Our inclusion criteria specified that surrogate outcome measures, such as fracture union, would be included in this Overview only in the absence of patient-oriented outcome measures. All studies included in this Overview present patient-oriented outcome measures. Therefore, surrogate outcomes are not presented.
For distal radius fractures, two level II randomized controlled trials (RCTs) were included.4,5 An additional level II comparative study,6 which did not compare outcomes between patients receiving locking plates and patients receiving nonlocking plates, was excluded, along with three level III comparative studies7-9 and seven level IV case series studies.10-16 Proximal humerus fractures were addressed with a single level III comparative study17 and nine level IV case series studies.18-26 For distal femur fractures, one level II study comparing the outcomes of patients receiving locking plates and patients receiving intramedullary nailing was included.27 For the purposes of this Overview, only the outcomes of the patients receiving locking plates were considered from this study (as level IV data). An additional seven level IV case series studies were included.28-34 One level IV case series study32 reports outcomes for a subgroup of patients from a previously published study31 that is included in this Overview. For the more recent publication, only the unique, relevant outcomes are reported in this Overview. All relevant outcomes are reported for the previous, original study. Periprosthetic femur fractures are addressed by four level IV case series studies.35-38 Fractures of the tibial plateau (AO/OTA type C only) are addressed by a single level II RCT39 comparing the outcomes of patients receiving locking plates via a single incision and patients receiving nonlocking plates via a double incision. Two additional level IV case series studies were included, as well.40,41 Proximal tibia fractures (AO/OTA type A or C) are addressed by five level IV case series studies.42-46 A single level IV case series study47 was excluded because it was updated by a more recent article43 that is included in this Overview. Distal tibia fractures are addressed by a single level IV case series study.48
To address this question, we recorded the patient enrollment criteria of the included studies. The published studies do not consistently report the same enrollment criteria. In general, published studies enrolled patients with acute, traumatic fractures of varying degrees or severity and typically did not enroll patients with fractures that were not amenable to adequate reduction and/or fixation. In studies with age criteria, no studies enrolled patients under the age of 18 years or not of skeletal maturity. In addition, studies enrolled patients with minimal comorbidities in their medical histories.
To address this question, we recorded the outcomes of studies that compared locking plates with nonlocking plates. Four studies investigated the differences between patients receiving locking plates and patients receiving nonlocking plates. These four studies addressed the application of locking plates in distal radius,4,5 proximal humerus,17 or tibial plateau39 fractures (AO/OTA type C only). There were no statistically significant differences between locking plates and nonlocking plates for patient-oriented outcomes, adverse events, or complications (Figure 1, available online at www.aaos.org/technologyoverviews). We recorded patient-oriented outcomes, adverse events, and complications from case series studies that met our inclusion criteria. Conclusions from case series studies are difficult to interpret because they lack the context that a control group provides. Further, it is difficult to use normative values to interpret the results of such studies because case series often enroll highly selected patients to whom such values may not apply. Making comparisons between studies is difficult because of the wide variety of patient-oriented outcome measures and duration to follow-up used to evaluate patients receiving locking plates in extremity fractures (Table 1).
Patient-oriented outcome measures from case series studies using locking plates for proximal humerus fractures are reported in the included studies. In summary, five studies report Constant-Murley scores at various durations to follow-up (1.5 to 12 months).19,21-24 In three studies reporting the mean Constant-Murley score at 12 months, the scores range from 74.6 to 79.19,21,23 Two studies reporting mean Disabilities of the Arm, Shoulder, and Hand (DASH) or QuickDASH scores indicated low levels of disability 12 months postoperatively.23,25 Two studies reported mean surgical times of 53 and 75 minutes.22,23 Adverse events and complications in patients receiving locking plates for proximal humerus fractures are reported in the included studies, as well. In summary, eight studies reported failure of hardware in 0% to 4% of patients,17-19,21,23-26 eight studies reported osteonecrosis in 0% to 7% of patients,17-19,21-25 four studies reported nerve palsy or paresthesia in 0% to 7% of patients,19,21,23,24 and six studies reported subacromial impingement in 0% to 14% of patients.17,19,21,22,24,25 Six studies reported implant removal in 1% to 8% of patients,17,19,21-23,25 and four studies reported infections in 2% to 5% of patients.18,19,22,24 Eight studies reported revision rates of 6% to 16%.17-21,23-25 Patient-oriented outcome measures from case series studies using locking plates for distal femur fractures are reported in the included studies. Two studies reported mean Lysholm scores. The first study reported a mean score of 48 at 3 months and of 81 at 12 months.27 The second study reported a mean score of 71 at 20 months.28 The mean surgical time ranged from 96 to 183 minutes in three studies.27,29,31 Adverse events and complications in patients receiving locking plates for fractures of the distal femur are reported in the included studies as well. In summary, three studies reported failure of hardware in 0% to 2% of patients,27,29,31 and two studies reported deep infections in 0% and 2% of patients.27,33 Two studies reported pulmonary embolism in 1% and 8% of patients,31,33 and three studies reported implant removal in 3% to 26% of patients.28,29,33 Four studies reported revision rates of 11% to 22%.28,31,33,34 Patient-oriented outcome measures from case series studies using locking plates for periprosthetic femur fractures are reported in the included studies. Two studies reported that 91% and 55% of patients returned to their previous levels of activity.36,38 The mean surgical time ranged from 90 to 104.5 minutes in three studies.35-37 Adverse events and complications in patients receiving locking plates for periprosthetic femur fractures are reported in the included studies, as well. In summary, three studies reported failure of hardware in 0% to 21% of patients,35,36,38 and three studies reported revision rates of 14% to 29%.35-37 Patient-oriented outcome measures from case series studies using locking plates for tibial plateau fractures (AO/OTA type C only) are reported in the included studies. One study reported a mean Medical Outcomes Study 36-Item Short Form physical subscale score of 29 at 6 months and of 40 at 12 months.41 Adverse events and complications in patients receiving locking plates for tibial plateau fractures (AO/OTA type C only) are reported in the included studies, as well. In summary, two studies reported failure of hardware in 0% and 18% of patients,39,41 three studies reported implant removal in 2% to 30% of patients,39-41 and two studies reported irritation from hardware in 6% and 12% of patients.39,41 Two studies reported deep infections in 2% and 7% of patients,39,40 and three studies reported superficial infections in 6% to 10% of patients.39-41 Patient-oriented outcome measures from case series studies using locking plates for proximal tibia fractures (AO/OTA type A or C) are reported in the included studies. No studies used the same patient-oriented outcome measure at similar durations to follow-up. Adverse events and complications in patients receiving locking plates for proximal tibia fractures (AO/OTA type A or C) are reported in the included studies as well. In summary, five studies reported infections in 0% to 6% of patients.42-46 Two studies reported revision to total knee arthroplasty in 3% and 4% of patients,42,44 two studies reported irritation from hardware in 5% and 8% of patients,42,43 and three studies reported implant removal in 5% to 13% of patients.42-44 Two studies reported revision rates in 13% and 15% of patients.43,45
Our literature searches did not identify any peer-reviewed cost-effectiveness or cost-utility studies that directly addressed this question.
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||