Current evidence makes a compelling case for a national registry
Registries for hip and knee replacement procedures are a world-wide reality with growing importance. Many of the preeminent registries—such as those in Sweden, Finland, Norway, Australia, Denmark, and New Zealand—have more than 10 years of experience and are currently collecting data on more than 90 percent of procedures nationally.
Registries are also growing in the complexity of data collected. The minimum recommended data set (http://www.ear.efort.org/E/03/01-03.asp) includes patient, surgeon, and hospital identifiers; core surgical data, date of surgery, diagnostic and treatment codes, laterality, and implant information. This may be augmented to include comorbidity and patient-administered questionnaires. Registries also address important issues such as prophylactic antibiotic and anticoagulant administration by measuring their impact on the incidence of infection and thromboembolic events. Published evidence from registries can now potentially guide important clinical decision-making.
A snapshot of clinical practices
A distinguishing feature of registries is that, with several thousand consecutive patients entered, they have no exclusions—a characteristic of methodologically sophisticated clinical trials. What is sacrificed in terms of potential bias inherent in inappropriate group comparisons, however, is offset by the breadth of applicability of the findings.
Most significantly, registries provide a snapshot of the realities of current clinical practices and real-time assessment of the associated outcomes. The timely feedback of these observations to surgeons may result in significant behavioral change, particularly if a practice pattern can be clearly associated with a desirable or adverse outcome.
National registries, by virtue of their inclusiveness, can detect significant risks to patient health associated with a particular procedure. For example, registry data show that pulmonary embolism (PE) following total hip or knee replacement is rare. In total hip arthroplasty (THA), the 90-day rate of nonfatal PE has been reported to be 0.93 percent in 58,521 Medicare patients who underwent primary THA with or without prophylaxis during 1995 and 1996. Death following PE in THA is very rare, with a 90-day death rate of just 0.22 percent, according to an analysis of 44,785 patients in the Scottish Morbidity Record from 1992 to 2001.
Nonfatal and fatal PE following total knee arthroplasty (TKA) are even less common. A survey of a California discharge database (222,684 patients who had undergone TKA from 1991 to 2001) found a 90-day nonfatal PE rate of 0.41 percent. The rate of fatal PE in 27,000 TKA patients in the Scottish Morbidity Record was 0.15 percent.
Most importantly, registries show that, despite significant changes in venous thromboembolism prophylaxis and surgical techniques over the past 10 to 15 years, the rates of PE and PE-related mortality are remarkably stable. Based on these findings, one might question both the current stereotype of all orthopaedic patients being at high risk for serious thromboembolic complications and the real value of using the newer and more expensive pharmacologic agents for routine thromboprophylaxis in all THA and TKA patients.
Is a U.S. registry needed?
In March 2008, the AAOS presented a symposium on the value of national registries, inviting representatives of several registries to discuss the real and potential benefit of total hip and knee registries. In his introductory comments, William J. Maloney, MD, underscored the following threefold value of registries:
- They provide an early warning system for early implant failure.
- They provide evidence that, if delivered to physicians in a timely and understandable fashion, will positively influence physician behavior to the benefit of patients and society.
- They have the power to ultimately decrease the burden of disease and cost associated with surgical morbidity and mortality, and reduce the volume of premature revision procedures.
The remainder of the symposium contained ample evidence to support these claims, lending weight to arguments for the absolute necessity of developing the American Joint Replacement Registry.
Registries as early warning systems
Early implant failure has not been adequately addressed through the traditional follow-up and publication strategy. Registries have proven more effective in identifying problems with particular implants or new surgical procedures used for implantation.
The Kaiser-Permanente Total Joint Registry, along with several national registries around the world, demonstrates the proficiency of this function by providing data on unicondylar knee replacement, an increasingly popular procedure. Data from multiple registries, however, demonstrate that the aseptic loosening rates of unicondylar knees is significantly higher than those of total knee replacements. At Kaiser, reporting this finding to its physicians led to a decrease in the rate of unicondylar replacements.
After the higher rates of loosening were found to be concentrated among surgeons who performed a lower volume of unicondylar replacements, a subsequent redistribution of the procedures toward higher volume surgeons was noted. The procedure itself was not discredited, but the need to refine the procedure’s indications and surgeon’s qualifications was underscored.
Impact on physicians
During the past several years, minimal incision surgery (MIS) for THA has been aggressively promoted as leading to more rapid recovery times and better short-term outcomes when compared to conventional THA. Much of this promotional information passed from the implant manufacturers directly to the public.
The Kaiser Registry was used to track the increase in volume and the outcomes of this new procedure. Within a few years of its inception in 2001, the registry detected a higher complication rate and compromise in early clinical outcomes associated with two-incision MIS THA. In addition, MIS TKA was found to produce increased pain and a decrease in patient satisfaction during the early postoperative period. When surgeons were informed of these observations, the number of these procedures performed dropped dramatically.
Registries in patient selection
The Australian National Joint Replacement Registry was used to study the impact of the rapid rise in resurfacing hip replacement during the past decade. An analysis of 5-year cumulative data found a significantly higher revision rate for resurfaced hips compared to total hip replacement. Women were found to have more than twice the risk for revision than men.
On further analysis, a femoral head size of less than 50 mm was found to be the primary risk factor. This finding has helped to clarify the population that may be most suitable for undergoing hip resurfacing: Men younger than 65 years old, with osteoarthritis, who require a femoral head size of 50 mm or more.
Reducing the burden of disease
In calling attention to the compelling case for a national total joint registry, David G. Lewallen, MD, chairman of the AAOS American Joint Replacement Registry Project Oversight Board, made the following observations:
During the life of the Swedish Hip and Knee Registries, the revision burden has been reduced from 17 percent to 7 percent.
In 1 year, the Australian National Joint Replacement Registry reported a 0.6 percent decrease in revision knee surgery at a savings value of $8.7 million. Using the volume of revision hip and knee replacements performed in the United States during 2003, a mere 1 percent reduction in the revision rate would have resulted in a 1-year savings of $30 million.
The evidence presented at the AAOS Symposium documented the value of national registries and made an excellent case for moving ahead in cooperation with the international orthopaedic community in developing a national total joint replacement registry in the US. The ability of registries to realize value has been amply demonstrated. A rapidly growing, evidence-based argument favors action now to avoid unnecessary injury to our patients, to promote improved clarity and rationale in developing clinical practice guidelines, and to facilitate the most cost-effective use of our progressively limited healthcare resources.
Norman A. Johanson, MD, is a member of the Evidence-Based Practice Committee. His disclosure information is available at www.aaos.org/disclosure
More information about the Evidence-Based Practice Committee and evidence-based medicine can be found at www.aaos.org/Research/Committee/Evidence/ebpc.asp
Read more...Joshua J. Jacobs, MD, leads a roundtable discussion on "Building a national joint replacement registry"
- Katz JN, Losina E, Barrett J, et al: Association between hospital and surgeon procedure volume and outcomes of total hip replacement in the United States Medicare population. J Bone Joint Surg Am 2001;83(A):1622-1629.
- Howie C, Hughes H, Watts AC: Venous thromboembolism associated with hip and knee replacement over a ten-year period: A population-based study. J Bone Joint Surg Br 2005;87:1675-1680.
- SooHoo NF, Lieberman JR, Ko CY, Zingmond DS: Factors predicting complication rates following total knee replacement. J Bone Joint Surg Am 2006;88:480-485.
- Lie SA, Engesaeter LB, Havelin LI, Furnes O, Vollset SE. Early postoperative mortality after 67,548 total hip replacements: Causes of death and thromboprophylaxis in 68 hospitals in Norway from 1987 to 1999. Acta Orthop Scand 2002;73:392-399.
- American Academy of Orthopaedic Surgeons clinical guideline on prevention of symptomatic pulmonary embolism in patients undergoing total hip or knee arthroplasty. Rosemont, IL: American Academy of Orthopaedic Surgeons (AAOS); 2007.
- Proceedings of the 75th Annual Meeting of the AAOS, San Francisco, March 5-9, 2008, pp 190-205.