We will be performing site maintenance on our learning platform at learn.aaos.org on Sunday, February 5th from 12 AM to 5 AM EST. We apologize for the inconvenience.

Paper outlining impact of registries wins 2017 OREF Clinical Research Award


Published 2/1/2017

Value of Arthroplasty Registries Recognized

Each year, the Orthopaedic Research and Education Foundation (OREF) recognizes outstanding clinical research related directly to musculoskeletal disease or injury through the OREF Clinical Research Award. This year, the award will be presented to Henrik Malchau, MD, PhD, primary investigator. Dr. Malchau's coauthors on the paper, "Arthroplasty Implant Registries Over the Past Five Decades: Development, Current and Future Impact," include Daniel J. Berry, MD; Göran Garellick, MD, PhD; and William H. Harris, MD, ScD.

The paper focuses on the impact of four arthroplasty registries—the Mayo Clinic registry, the Harris registry at Massachusetts General Hospital (MGH), and the Swedish knee and hip arthroplasty registries—that have set the standards for infrastructure, analysis and reporting mechanisms, and leadership. The authors note that "as more countries have adopted the concept of national registries, a new area of research is possible by pooling the resources of large registries as is now occurring with the Nordic countries." They point to the globalization of registries in improving and disseminating knowledge gained.

New techniques and designs in joint arthroplasty are not always successful, note the authors. Further, complications including sepsis, thromboembolic events, component breakage, loss of fixation, and wear-induced periprosthetic osteolysis continue to occur. Unfortunately, they note, poor component design problems "often became evident only after several years of an initial period of clinically successful performance."

Today's quality registries, they write, are based on close, detailed follow-up of patients and the development of systems for tracking patients and documenting outcomes. "The purpose of a registry is to collect regional or national data in order to analyze and draw statistically significant conclusions regarding patient-, surgical technique-, and implant-associated risk factors that lead to good or poor outcomes," they write.

Mayo Registry
The first and largest institutionally based joint registry in the United States is the Mayo Clinic total joint registry. Established in 1969, the registry includes detailed information on more than 100,000 primary and revision joint replacements (hip, knee, shoulder, and elbow) performed at the Mayo Clinic in Rochester, Minn.

The Mayo Clinic registry includes detailed operative data on each procedure, joint-specific scores, and detailed information on short- and long-term complications. Uniquely, this registry includes several database cross-references to provide information necessary for studies and a strong follow-up system. "The level of detail, correlation with radiographs, and ability to cross-reference with other clinical databases and individual patient records allows a diversity and depth of study of questions that is different from, and complimentary to, large national registries," wrote the authors.

Data from the Mayo Clinic registry can be used to calculate survivorship of particular implants, including in specific settings. Although implant comparisons are limited those used at the institution, patient, procedure, and complications data enable a detailed and nuanced picture of implant performance. Studies using registry data have compared different categories of cup fixation surface in primary and revision total hip arthroplasty (THA), the performance of cruciate-substituting versus cruciate-sparing total knee arthroplasty (TKA), and the effect of femoral head diameter on risk of hip dislocation.

The detailed patient information also yields insights on the results and complications of joint replacement in particular populations.

Harris Joint Registry
In explaining the origins of the Harris Joint Registry, the authors noted the need for a rigorous system that could quantify and contrast osteotomy and cup arthroplasty outcomes. "Any such system should focus on the outcome by patient, rather than by disease or operation…and it needed to generate a single numerical value for each patient," they wrote. The introduction of the Harris Hip Score in 1969 provided the impetus for a registry to collect, store, and manipulate these data sets.

"To facilitate raw data collection, every office visit was required to generate all the data for the Harris Hip Score and a full set of AP [anteroposterior], frog lateral, and true lateral radiographs," they note. A second critical requirement was to collect and retain all germane radiographs, which proved crucial in assessing issues such as loosening and lysis prior to revision surgery. The advent of digital radiography enabled automatic inclusion of all digital images per THA and TKA patient at MGH, which has expanded the registry concepts to other orthopaedic divisions.

The Harris Joint Registry has made "substantive contributions to a wide array of pressing issues in THA," note the authors, including critical evaluations of the short- and long-term results of various total implant designs, instruments, surgical techniques, and materials.

The Swedish Knee Arthroplasty Registry
Inspired by Professor Göran Bauer, initiated by the Swedish Orthopaedic Society, and funded through the Swedish Medical Research Council, the Swedish Knee Arthroplasty Registry (SKAR) initially had an ambitious program of clinical follow-up, which eventually was abandoned in favor of a focus on a minimal dataset for the primary surgery. "The primary outcome measure became revision with all revisions having a complete review of hospital records to ascertain the reason for failures. It took 35 years before the register again dared ask for additional information regarding prior surgeries, surgical techniques, or the type and length of drug prophylaxis," they said.

SKAR was a pioneer in using actuarial tables to estimate the survival of implants, and its statistical analysis and methods are now quite sophisticated. The registry has been instrumental in identifying issues with postoperative infections and the cost-effectiveness of antibiotic prophylaxis.

Paper outlining impact of registries wins 2017 OREF Clinical Research Award
Fig. 1 This spider diagram displays quality indicators in the Swedish registry.
Courtesy of Henrik Malchau, MD, PhD

"One major aim of the SKAR register is to identify techniques or implants that do not perform well," write the authors. In this, SKAR has effectively supported recommendations on the thickness of the tibial polyethylene in knee implants, the avoidance of unicompartmental implants in knees with inflammatory conditions, and surgeons' training on new implants and techniques.

The Swedish Hip Arthroplasty Registry
The original aims of the Swedish Hip Arthroplasty Registry were to document all reoperations following THA in Sweden, to document complications associated with revision surgery, and to improve long-term results. Complications were found to be more prevalent in revision surgery compared to primary arthroplasty, and one third of the reoperations resulted in further complications.

The registry initiative gained early acceptance and reporting was nearly unanimous. Over the years, reporting was expanded to include primary THAs and implant variables. Although this meant more work for each hospital, the compliance remained high due to several factors including the constant development of online support staff, periodic courses for local staff, statistical support for the individual centers, and feedback to the professional staff.

"As a result of these efforts, the incidence of serious complications and revision rates have continually declined in Sweden over time to the point where only about 10 percent of hip arthroplasty cases are for revision," the researchers write.

Patient-reported outcome measures were introduced in 2002. "Currently the annual report includes data on patient-reported overall satisfaction, pain relief after 1 year, EQ-5D gain [estimation of health-related utility], 90-day mortality rates, completeness of data reporting, 2-year reoperation rates, and 5- and 10-year implant survivorship for each center," they write. The data is typically presented as a balanced scorecard or as a "spider diagram" (Fig. 1). Although no data on individual surgeons are shared publicly, they are available to each center on a confidential basis, enabling review, corrective measures, and improvements.

Data quality and compliance development
"In a perfect world, every registry, at aggregate- and unit level, should report the following five dimensions: 1) coverage, 2) completeness, 3) response rate, 4) missing values, and 5) validation," note the authors. They define each of these dimensions, noting the importance of achieving consensus, appointing a specific contact person, and providing training.

"Efforts should be made to capture a limited number of variables but still get an adequate description of the intervention and outcome. Whether paper or web collection is used the design must be 'intuitive' and user-friendly," they write. The importance of feedback in ensuring compliance cannot be underestimated. "Users must initially see the benefits of the registration burden. Coverage and completeness analyses should be performed and published on a regular schedule. Monitoring of the individual units is a validation process, which will ultimately facilitate 'completeness.'"

The "globalization process"
Based on the success of the Swedish hip and knee arthroplasty registries, other countries began to establish regional and national registries. Globalization created the need for an International Society of Arthroplasty Registers (ISAR). In the United States, they note, the American Joint Replacement Registry is a recent, "very promising concept."

"The purpose of any arthroplasty registry is to improve outcomes of joint replacement surgery," they write. "ISAR [members] have this shared purpose [although] … the focus of the society should be to utilize the strength of linkage and develop a framework that supports the activities of both established and developing registries." The primary goals of ISAR are as follows:

  • to provide a support network for established and developing registries
  • to encourage cooperation and sharing of information, as well as collaborative activities
  • to establish standardized terminology
  • to standardize statistical analysis

The authors also call for an "attributes" database that can be used in implant-based outcomes analysis, identifying features (attributes) common to a variety of prostheses. They also note the potential of an international network in warning surgeons and other registries if an inferior implant or surgical technique is discovered by one of the registries.

Others examples of international registries examined by the authors include the Nordic Arthroplasty Register Association (Danish, Finnish, Norwegian, and Swedish national registries), the International Consortium of Health Outcomes (worldwide registries from various medical fields), the ArthroplastyWatch project (initiated by the Swedish Knee Register and the Swedish National Competence Center and maintained by Lund University), and the International Consortium of Orthopaedic Registries (established by the U.S. Food and Drug Administration and involving 29 joint registries from 14 nations).

Future impact
As the authors note, "Many healthcare systems are struggling today with rising costs, creating a need to find systems to improve the quality of health care while also improving cost-efficiency. An approach offering great potential is the use of national quality registries [to enable] healthcare to identify, implement, and share best practices." They point to a "new clinical trial structure that relies on data from established national registries to support and augment clinical studies, nesting clinical trials within national registries, and the use of pooled registry data to effectively monitor the introduction of new technologies."

In addition, they say, "National registries with high coverage and completeness have a large and unutilized potential as health economic instruments, especially if the registries include patient-reported outcomes and can be linked to reimbursement in health care and other insurance and societal costs." They also note the high level of research activity promoted by registries. Complementary institutional and national databases can foster additional research, including the potential to correlate patient genetic characteristics with the outcomes of joint surgery.