Clinical Orthopaedics and Related Research®
Clin Orthop Relat Res (2015) 473:76–81 DOI 10.1007/s11999-014-3716-5
A Publication of The Association of Bone and Joint Surgeons®
SYMPOSIUM: 2014 KNEE SOCIETY PROCEEDINGS
Preoperative Pain and Function Profiles Reflect Consistent TKA Patient Selection Among US Surgeons David C. Ayers MD, Wenjun Li PhD, Leslie Harrold MD, MPH, Jeroan Allison MD, MS, Patricia D. Franklin MD, MBA, MPH
Published online: 24 June 2014 Ó The Association of Bone and Joint Surgeons1 2014
Abstract Background As the number of primary total knee arthroplasties (TKAs) performed in the United States increases, policymakers have questioned whether the indications and timing of TKA have evolved so that surgery is offered earlier. Questions/purposes We analyzed data from a US national TKA cohort to evaluate variation in surgeon selection criteria for elective unilateral TKA based on preoperative patient-reported pain and function scores. Methods Preoperative SF-36 (Physical Component Summary [PCS]/physical function) scores and Knee Injury and Osteoarthritis Outcome Score (KOOS) (pain, activities of daily living/function) of 4900 patients undergoing elective unilateral TKA enrolled in this national database of prospectively followed patients from 22 states were evaluated. The 25th, 50th, and 75th percentile pain and function scores
During the study period, the authors’ institution has received funding from the Agency for Healthcare Research and Quality (Grant P50HS018910). All ICMJE Conflict of Interest Forms for authors and Clinical Orthopaedics and Related Research1 editors and board members are on file with the publication and can be viewed on request. Each author certifies that his or her institution approved the human protocol for this investigation, that all investigations were conducted in conformity with ethical principles of research, and that informed consent for participation in the study was obtained. D. C. Ayers (&), W. Li, L. Harrold, P. D. Franklin Department of Orthopedics and Physical Rehabilitation, University of Massachusetts Medical School, 55 Lake Avenue North, Worcester, MA 01655, USA e-mail: [email protected]; [email protected] J. Allison Department of Quantitative Health Sciences, University of Massachusetts Medical School, 55 Lake Avenue North, Worcester, MA 01655, USA
123
for patients cared for in 24 orthopaedic offices with 20 or more patients in the database were compared to assess whether consistent preoperative criteria are used in selecting patients undergoing TKA across settings. Results The preoperative global function (PCS median, 32.6; national norm, 50; SD, 10) and knee-specific function (KOOS median, 51.5; maximum score, 100; SD, 17) percentile scores represented substantial patient disability, because both values approached 2 SDs below ideal. Consistency in patients across 24 surgeon offices, and more than 100 surgeons, was noted because site-specific medians varied from the national median by less than the minimum clinically important change. Conclusions These data suggest that despite the rapidly growing use of TKA, surgeons in the participating sites use consistent patient criteria in scheduling TKA. Today’s patients report significant pain and disability, supporting the need for TKA.
Introduction TKA is one of the most common surgical procedures in the United States with more than 600,000 TKAs performed in 2011 with aggregate charges (our ‘‘national bill’’) of over USD 33 billion [16]. In 2007, Kurtz et al. [17] projected the number of procedures would rise to over 3 million by 2030.The rise in the number of total joint arthroplasties (TJAs) can be viewed as an indication of the success of the surgery to safely reduce pain and improve function in an aging population. The large increase in TKA use is often attributed to the aging Medicare population combined with the growing prevalence of conditions that predispose to osteoarthritis such as obesity [9, 22]. In a recent analysis of Medicare data, researchers observed that growth is related
Volume 473, Number 1, January 2015
to both aging and increased per-capita demand with the volume of Medicare-funded TKAs increasing 160% from 93,000 in 1991 to 242,000 in 2010. In parallel, the percapita TKA use rate doubled from 30 to 60 TKAs per 10,000 Medicare enrollees [6]. However, many policymakers regard this rise as evidence of overuse in an environment rewarding high-cost procedures [4]. This tremendous growth in TKA use thus prompts concern as to whether too many (or too few) of these procedures are performed both in aggregate and among key patient subgroups. Any effort to answer these concerns raises the issues of TKA indication, timing, and appropriateness. Cram et al. [6] noted ‘‘it is difficult to determine the extent to which the growth in TKA use represents appropriate use of a highly effective procedure or the overuse of a well-reimbursed procedure for which indications depend on clinical judgment. It is likely that both factors are at play.’’ We therefore compared early data from a recently established US national TKA cohort including patients from 123 surgeons in 22 states to evaluate preoperative variability in patient selection across surgeons. In particular, we asked whether surgeons use similar selection criteria based on preoperative patient-reported outcomes (PROs) when selecting patients to undergo TKA.
Data From a US National TKA Cohort
77
prospectively maintained database of patient-reported outcomes in a cohort of patients recruited from more than 100 surgeons who practice in 22 states.
Study Population/Measures Study Sample
We conducted an analysis of a consecutive sample of patients undergoing primary TKA or THA from a large,
The Agency for Health Care Research and Quality funded a USD 12 million research program, Function and Outcomes Research in Comparative Effectiveness Research in Total Joint Replacement (FORCE-TJR), to assemble a national cohort of patients undergoing TJA with comprehensive data elements [11]. FORCE-TJR includes patients from a national consortium of 136 orthopaedic surgeons representing all regions of the United States (Fig. 1) as well as varied hospital and surgeon practice settings. To date 19,200 patients have been enrolled; 6900 have reached 12 months followup. The database has allowed us to publish three articles [2, 11, 12] and submit several others that primarily deal with outcome analysis of this population. The first consecutive 4900 FORCE-TJR patients undergoing primary, unilateral TKA from 22 sites and with signed informed consent and enrolled in 2011 and 2012 were included in this analysis. Patients had a mean age of 66 years and were drawn from 24 orthopaedic practices (more than 100 surgeons). Sixty-two percent were female. By study design FORCE returns scored data quarterly, because PROs are not scored real-time in the office. This means that the patient and surgeon made the surgical decision before the surgeon received the scored presurgical
Fig. 1 This figure shows the locations of the FORCE-TJR participating sites across the United States: (1) Core Clinical Centers: University of Massachusetts Medical School, Worcester, MA, USA (blue star); Connecticut Joint Replacement Institute, Hartford, CT, USA (red star);
The University of Rochester Medical Center, Rochester, NY, USA (green star); Medical University of South Carolina, Charleston, SC, USA (gray star); Baylor College of Medicine, Houston, TX, USA (light blue star); and (2) Community Sites (green dots).
Materials and Methods Study Design
123
78
Clinical Orthopaedics and Related Research1
Ayers et al.
PROs. The decision to schedule elective TKA was based on the surgeons’ routine indications and clinical judgment.
We then observed whether any individual site median differed from (1) the national median or (2) all other site medians on pain and function scores to a degree greater than the minimal clinical important difference.
Surgeons and Patients in the Database FORCE-TJR surgeons vary in annual TJA volume with 13% performing an estimated 0 to 50 TJAs/year, 28% performing 51 to 200 TJAs/year, and 59% performing more than 200 TJAs/year. Five are academic hospitals/centers and eight are rural hospitals. Community-based surgeons represent a large portion (75%) of this national cohort, consistent with national TJA practice. Enrolled surgeons include both fellowship-trained and general orthopaedic surgeons. Patients in this database have private, public, and HMO insurance.
Measures FORCE-TJR collects extensive patient demographic information, complete medical and musculoskeletal comorbidities, surgical care data, implant data, postoperative complications, adverse events, and readmission. FORCE-TJR collects several validated PROs, directly from patients, to estimate functional outcomes and pain relief. This study presents data using a global function measure, the SF-36 Physical Component Summary (PCS) [21] and a knee-specific outcomes measure, the Knee Injury and Osteoarthritis Outcome Score (KOOS) [19], that includes a pain score and activities of daily living (ADL)/function score questionnaires. Of note, the KOOS ADL/function score is equivalent to the widely used WOMAC function index [3] but is nonproprietary [19]. For reference, the PCS scores range from 0 to 80 with the highest score representing greatest function. For the PCS, 50 is the national adult norm. The KOOS ADL score ranges from 0 to 100 with 100 representing perfect function. For the KOOS pain, a score of 100 represents no pain, and pain increases as scores approach 0. FORCE-TJR collects data before surgery, 6 and 12 months postoperatively, and yearly thereafter. All patients in this study completed preoperative PROs. By study design FORCE returns data quarterly, because PROs are not scored real-time in the office. This means that the surgical decision was made before the surgeon received the scored pre-PROs. To assure stable PRO estimates for each site, only surgeons who enrolled more than 20 patients by mid-2012 were included in this analysis.
Results The median baseline (preoperative) PCS score was 32.6 with a range of medians across sites from 29.3 to 34.4 (Fig. 2). Preoperative scores were almost 2 SDs below the general population PCS norm (mean, 50; SD, 10) [20] and were consistent with advanced disability resulting from arthritis. For context, the minimum clinically important change (MCIC) for PCS is 5 points [20]. The lowest median (29.3) and highest median (34.4) were less than 3 points from the national median, or less than the MCIC. This suggests no clinical difference could be detected in the median (typical) patient across all sites. The 25th percentile PCS mean for the study sample was 26.9 (range, 23.9–28), whereas the 75th percentile PCS mean was 38.1 (range, 35.3–41). More than 81% had preoperative scores greater than, or equal to, 1 SD below the national norm of 50. The median KOOS pain score was 44.4 with a range of medians across sites from 36.1 to 50, representing clinically important pain because a score of 90 to 100 represents none or minimal pain (Fig. 3). The MCIC is 7 to 8 points. One SD represents approximately 15 points. Thus, the median pain scores across sites differed less than the MCIC, meaning not clinically different. Like with PCS, we observed a tight clustering of 25th and 75th percentile for the KOOS pain scores by site (Fig. 3). The median KOOS ADL, or function, score was 51.5 with a range of medians across sites from 39.7 to 57.8 (Fig. 4) as compared with the MCIC of 7 to 8. There was a tight clustering of 25th and 75th percentile for the KOOS ADL scores by site.
Statistical Analysis We analyzed the data descriptively. We estimated select descriptive characteristics (mean, 25th, 50th, and 75th percentiles) of function and pain scores by participating site.
123
Fig. 2 This figure depicts the baseline primary TKA SF-36 PCS by site. The red line represents median across sites. The national PCS norm for the general population is 50.
Volume 473, Number 1, January 2015
Fig. 3 This figure depicts the baseline primary TKA KOOS pain score by site. The red line represents median across sites. Being painfree corresponds to a score of 90 to 100.
Fig. 4 This figure depicts the baseline primary TKA KOOS/ADL score by site. The red line represents median across sites. The ideal function corresponds to a score between 90 and 100.
Discussion The rapid growth in use of TKA is often attributed to the aging US population because more than 60% of adults older than 65 years of age report knee or hip osteoarthritis. However, the growing number of patients younger than 65 years raises the question as to whether surgeon selection of appropriate patients is evolving. We hypothesized that if surgeon selection is changing, the typical patient undergoing TKA would report less pain and disability at the time of surgery and we would see variation in typical patients across surgeons. The preoperative data from this 2011 to 2012 national sample of patients undergoing primary TKA suggest that surgeons across the United States applied consistent patient selection criteria as defined by preoperative pain and function scores. Both the median preoperative global function (PCS) and median knee-specific pain and function (KOOS) scores represent a large degree of pain and disability at the time of TKA. This suggests that the rapid increase in number of patients
Data From a US National TKA Cohort
79
undergoing TKA cannot be explained fully by clinicians extending surgical indications to patients with less severe symptoms. This study had a number of limitations. First, the typical patient undergoing TKA is described by self-report of pain and disability, because we do not have the physical examination or radiographic findings that informed the surgeon’s decision for TKA. Future research may compare patient-reported symptoms as well as physical examination findings (eg, contractures or deformities) and severity of arthritis changes on radiographs to fully illuminate the clinical decision process. Second, although FORCE-TJR surgeons perform diverse TKA volumes and practice in varied settings, these observations are based on a sample of US arthroplasty surgeons. It would be valuable to validate these findings in another national sample before drawing conclusions about the entire US surgical community. Finally, we noted some variation in the higher and lower quartiles in function and pain scores. Future research should examine the full clinical rationales for surgery, including the physical examination and supporting radiographic images, before any conclusions can be drawn about the timing and need for TKA. The international literature documents variation in the objective criteria associated with the optimal timing of TKA. A 2006 literature review highlighted large variation among orthopaedic surgeons as to the indications for TKA [7]. Similarly, a Canadian study attributed much of the regional variation in TKA use to differences in surgeon opinions when recommending patients for TKA [23]. Research in Australia and Europe found that variation in age, pain, stiffness, and physical function before TKA and THA are associated with substantial differences in timing of surgery [1]. Ackerman et al. [1] conclude that, ‘‘Possible contributing factors [to premature or delayed TKR timing] include patient preferences, the absence of concrete indications for surgery, and the capacity of health care systems.’’ In contrast to these reports of diverse indications for TKA, the typical (median) patient pain and disability scores in this sample correspond with advanced clinical symptoms. Furthermore, these patients reported similar median scores across sites. By definition, 50% of patients had more pain and poorer function than the median. It is possible that the studies describing varied TKA indications have only focused on patients reporting the upper 25% to 30% of pain and function scores. In this cohort, additional clinical data are needed to make any assessment about the indication for surgery among these patients reporting less severe symptoms. Finally, it is possible that surgeons who agreed to participate in FORCE-TJR are conservative in their patient selection criteria, because they know our analysts will review the data. Although this is possible,
123
80
Clinical Orthopaedics and Related Research1
Ayers et al.
only three of the participating surgeons had used PROs in their clinical practice before their participation in FORCETJR so most surgeons had no established PRO criteria for surgical timing and depended on physical examination and radiographs alone to make a recommendation. Few studies have estimated appropriateness measures for TKA use. In a multicenter prospective observational study involving 15 hospitals in Spain, Cobos et al. [5] estimated that as many as 25% of the TKAs and THAs performed might be considered inappropriate. The authors attributed this finding to variability among the clinical criteria used by surgeons when selecting patients for surgery [5]. Although the current study cannot comment on the appropriateness of the TKA surgical decision, the PRO scores at the median are consistent with advanced pain and functional limitations. The Osteoarthritis Research Society International (OARSI)/Outcome Measures in Rheumatology (OMERACT) group recently concluded that the most discriminating criteria in regard to indication for TKA were the symptoms and duration of pain and physical function impairment [15]. This recommendation parallels the National Institutes of Health consensus statement on TKA issued in 2003 [18]. In the United States, the landmark patient outcomes research team (PORT) TKA study in the 1980s defined PRO norms at the time of primary TKA [14]. Of note, the median pain and function scores in this sample are similar to those reported by the PORT study in the 1980s. The timing of TKA is a crucial decision and surgeons agree that patients should have severe knee pathology to warrant surgery. Pain relief is achieved in the vast majority of patients undergoing TKA, and measurable post-TKA functional gain is reported among at least two-thirds of patients undergoing TKA. However, patients with the poorest function before surgery are more likely to have smaller functional gains [10, 13]. In a sample of 8300 patients undergoing primary TKA from a national implant database, variation in functional outcome after TKA was associated with patient attributes, including poorer preoperative function [13]. Of note, the mean preoperative PCS was 30 (SD, 7.8) in this national sample accrued 10 years ago, which is similar to that in our study. Our data suggest that despite the growth in the number of patients undergoing TKA during the past decade, today’s typical patient has comparable preoperative disability as estimated by the SF-36 PCS function score. Further analysis of the upper 25th percentile of patients reporting less pain and functional limitation could provide insight into the role that patient demand to maintain higher levels of activity and tolerate less pain may play in TKA timing. It is possible that both observations offer different sides to the story. Although the typical patient, and the lower 50%, have symptoms of
123
advanced joint disease, a subgroup of the patients with lesser symptoms may reflect evolving patient demands. Specific guidelines to inform surgeon and patient decisions for TKA would be helpful. To this end, OARSI/ OMERACT recently attempted to define a ‘‘theoretical indication for TKA’’ using the severity and duration of symptoms of pain, physical function as well as structural knee damage assessed by radiographs. However, variations among individual patients, surgeons, and countries made it difficult to pinpoint relevant cut points consistent with patient need for TKA. Radiographic evidence did not correlate with symptoms of pain and physical function. Again, the most discriminating criteria identified with appropriate timing for TKA were symptoms and duration of pain and physical function [13]. In conclusion, patients differ in both their clinical presentation and demand for physical activity, so surgeon evaluation guided by the patient’s history is critical to individualized decisions for TKA. The future availability of real-time, scored, and trended patient-reported outcomes may inform in-office decisions for surgery [2]. Finally, new national cohorts with patient-reported outcomes such as FORCE-TJR may help define normative guidelines to support patient selection and TKA timing [8, 11]. Our data, which represent an early iteration of FORCE-TJR, suggest that surgeons use consistent patient criteria in scheduling the median or typical patient undergoing TKA. In addition, these patients report significant levels of disability at the time of surgery. In the future, the addition of national norms for patient-reported pain and function measures may be an important complement to clinical data to support consistency in patient selection of patients most likely to achieve an optimal TKA outcome. Acknowledgments We gratefully acknowledge the more than 120 surgeons who participate in the FORCE-TJR national consortium, especially the core site Principal Investigators, Courtland Lewis MD, Philip Noble PhD, Regis O’Keefe MD, PhD, and Vincent Pellegrini MD. We also appreciate the editorial assistance of Sylvie Puig PhD.
References 1. Ackerman IN, Dieppe PA, March LM, Roos EM, Nilsdotter A, Brown GC, Sloan KE, Osborne RH. Variation in age and physical status prior to total knee and hip replacement surgery: a comparison of centers in Australia and Europe. Arthritis Rheum. 2009;61:166–173. 2. Ayers DC, Zheng H, Franklin PD. Integrating patient-reported outcomes (PROs) into orthopaedic clinical practice: proof of concept from FORCE-TJR. Clin Orthop Relat Res. 2013;471: 3419–3425. 3. Bellamy N, Buchanan WW, Goldsmith CH, Campbell J, Stitt LW. Validation study of WOMAC: A health status instrument for measuring clinically important patient relevant outcomes to
Volume 473, Number 1, January 2015
4. 5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
antirheumatic drug therapy in patients with osteoarthritis of the hip or knee. J Rheumatol. 1988;12:1833–1840. Chernew M, Goldman D, Axeen S. How much savings can we wring from Medicare? N Engl J Med. 2011;365:e29. Cobos R, Latorre A, Aizpuru F, Guenaga JI, Sarasqueta C, Escobar A, Garcia L, Herrera-Espineira C. Variability of indication criteria in knee and hip replacement: an observational study. BMC Musculoskelet Disord. 2010;11:249.. Cram P, Lu X, Kates SL, Singh JA, Li Y, Wolf BR. Total knee arthroplasty volume, utilization, and outcomes among Medicare beneficiaries, 1991–2010. JAMA. 2012;308:1227–1236. Cross WW 3rd, Saleh KJ, Wilt TJ, Kane RL. Agreement about indications for total knee arthroplasty. Clin Orthop Relat Res. 2006;446:34–39. Dieppe P, Lim K, Lohmander S. Who should have knee joint replacement surgery for osteoarthritis? Int J Rheum Dis. 2011;14:175–180. Felson DT, Chaisson CE. Understanding the relationship between body weight and osteoarthritis. Baillieres Clin Rheumatol. 1997;11:671–681. Fortin PR, Penrod JR, Clarke AE, St-Pierre Y, Joseph L, Belisle P, Liang MH, Ferland,D, Phillips CB, Mahomed N, Tanzer M, Sledge C, Fossel AH, Katz JN. Timing of total joint replacement affects clinical outcomes among patients with osteoarthritis of the hip or knee. Arthritis Rheum. 2002;46:3327–3330. Franklin PD, Allison JJ, Ayers DC. Beyond joint implant registries: a patient-centered research consortium for comparative effectiveness in total joint replacement. JAMA. 2012;308:1217–1218. Franklin PD, Harrold LR, Ayers DC. Incorporating patient reported outcomes in total joint arthroplasty registries: challenges and opportunities. Clin Orthop Relat Res. 2013;471:3482– 3488. Franklin PD, Li W, Ayers DC. The Chitranjan Ranawat Award: Functional outcome after total knee replacement varies with patient attributes. Clin Orthop Relat Res. 2008;466:2597–604. Freund D, Lave J, Clancy C, Hawker G, Hasselblad V, Keller R, Schneiter E, Wright J. Patient outcomes research teams: contribution to outcomes and effectiveness research. Annu Rev Public Health. 1999:337–359.
Data From a US National TKA Cohort
81
15. Gossec L, Paternotte S, Bingham CO 3rd, Clegg DO, Coste P, Conaghan PG, Davis AM, Giacovelli G, Gunther KP, Hawker G, Hochberg MC, Jordan JM, Katz JN, Kloppenburg M, Lanzarotti A, Lim K, Lohmander LS, Mahomed NN, Maillefert JF, Manno RL, March LM, Mazzuca SA, Pavelka K, Punzi L, Roos EM, Rovati LC, Shi H, Singh JA, Suarez-Almazor ME, Tajana-Messi E, Dougados M. OARSI/OMERACT initiative to define states of severity and indication for joint replacement in hip and knee osteoarthritis. An OMERACT 10 special interest group. J Rheumatol. 2011;38:1765–1769. 16. HCUPnet. Healthcare Cost and Utilization Project (HCUP). Nationwide Inpatient Sample (NIS). Rockville, MD, USA: Agency for Healthcare Research and Quality; 2011. Available at: http://hcupnet.ahrq.gov/. Accessed October 11, 2013. 17. Kurtz S, Ong K, Lau E, Mowat F, Halpern M. Projections of primary and revision hip and knee arthroplasty in the united states from 2005 to 2030. J Bone Joint Surg Am. 2007;89:780–785. 18. Rankin EA, Alarcon GS, Chang RW, Cooney LM Jr, Costley LS, Delitto A, Dey RA, Donaldson SK, Hochberg MC, MacLean CH, Yelin EH, Marciel K. NIH consensus statement on total knee replacement, December 8–10, 2003. J Bone Joint Surg Am. 2004;86:1328–1335. 19. Roos EM, Roos HP, Lohmander LS, Ekdahl C, Beynnon BD. Knee Injury and Osteoarthritis Outcome Score (KOOS)—development of a self-administered outcome measure. J Orthop Sports Phys Ther. 1998;28:88–96. 20. Ware JE, Kosinski M, Bjomer JB, Turner-Bowker DM, Gandek B, Maruish ME. User’s Manual for the SF-36v2 Health Survey. 2nd ed. Lincoln, RI, USA: QualityMetric Incorporated; 2007. 21. Ware JE Jr, Sherbourne CD. The MOS 36-item short-form health survey (SF-36). I. Conceptual framework and item selection. Med Care. 1992;30:473–483. 22. Woolf AD, Breedveld F, Kvien TK. Controlling the obesity epidemic is important for maintaining musculoskeletal health. Ann Rheum Dis. 2006;65:1401–1402. 23. Wright JG, Hawker GA, Bombardier C, Croxford R, Dittus RS, Freund DA, Coyte PC. Physician enthusiasm as an explanation for area variation in the utilization of knee replacement surgery. Med Care. 1999;37:946–956.
123
Clin Orthop Relat Res (2015) 473:76–81 DOI 10.1007/s11999-014-3716-5
A Publication of The Association of Bone and Joint Surgeons®
SYMPOSIUM: 2014 KNEE SOCIETY PROCEEDINGS
Preoperative Pain and Function Profiles Reflect Consistent TKA Patient Selection Among US Surgeons David C. Ayers MD, Wenjun Li PhD, Leslie Harrold MD, MPH, Jeroan Allison MD, MS, Patricia D. Franklin MD, MBA, MPH
Published online: 24 June 2014 Ó The Association of Bone and Joint Surgeons1 2014
Abstract Background As the number of primary total knee arthroplasties (TKAs) performed in the United States increases, policymakers have questioned whether the indications and timing of TKA have evolved so that surgery is offered earlier. Questions/purposes We analyzed data from a US national TKA cohort to evaluate variation in surgeon selection criteria for elective unilateral TKA based on preoperative patient-reported pain and function scores. Methods Preoperative SF-36 (Physical Component Summary [PCS]/physical function) scores and Knee Injury and Osteoarthritis Outcome Score (KOOS) (pain, activities of daily living/function) of 4900 patients undergoing elective unilateral TKA enrolled in this national database of prospectively followed patients from 22 states were evaluated. The 25th, 50th, and 75th percentile pain and function scores
During the study period, the authors’ institution has received funding from the Agency for Healthcare Research and Quality (Grant P50HS018910). All ICMJE Conflict of Interest Forms for authors and Clinical Orthopaedics and Related Research1 editors and board members are on file with the publication and can be viewed on request. Each author certifies that his or her institution approved the human protocol for this investigation, that all investigations were conducted in conformity with ethical principles of research, and that informed consent for participation in the study was obtained. D. C. Ayers (&), W. Li, L. Harrold, P. D. Franklin Department of Orthopedics and Physical Rehabilitation, University of Massachusetts Medical School, 55 Lake Avenue North, Worcester, MA 01655, USA e-mail: [email protected]; [email protected] J. Allison Department of Quantitative Health Sciences, University of Massachusetts Medical School, 55 Lake Avenue North, Worcester, MA 01655, USA
123
for patients cared for in 24 orthopaedic offices with 20 or more patients in the database were compared to assess whether consistent preoperative criteria are used in selecting patients undergoing TKA across settings. Results The preoperative global function (PCS median, 32.6; national norm, 50; SD, 10) and knee-specific function (KOOS median, 51.5; maximum score, 100; SD, 17) percentile scores represented substantial patient disability, because both values approached 2 SDs below ideal. Consistency in patients across 24 surgeon offices, and more than 100 surgeons, was noted because site-specific medians varied from the national median by less than the minimum clinically important change. Conclusions These data suggest that despite the rapidly growing use of TKA, surgeons in the participating sites use consistent patient criteria in scheduling TKA. Today’s patients report significant pain and disability, supporting the need for TKA.
Introduction TKA is one of the most common surgical procedures in the United States with more than 600,000 TKAs performed in 2011 with aggregate charges (our ‘‘national bill’’) of over USD 33 billion [16]. In 2007, Kurtz et al. [17] projected the number of procedures would rise to over 3 million by 2030.The rise in the number of total joint arthroplasties (TJAs) can be viewed as an indication of the success of the surgery to safely reduce pain and improve function in an aging population. The large increase in TKA use is often attributed to the aging Medicare population combined with the growing prevalence of conditions that predispose to osteoarthritis such as obesity [9, 22]. In a recent analysis of Medicare data, researchers observed that growth is related
Volume 473, Number 1, January 2015
to both aging and increased per-capita demand with the volume of Medicare-funded TKAs increasing 160% from 93,000 in 1991 to 242,000 in 2010. In parallel, the percapita TKA use rate doubled from 30 to 60 TKAs per 10,000 Medicare enrollees [6]. However, many policymakers regard this rise as evidence of overuse in an environment rewarding high-cost procedures [4]. This tremendous growth in TKA use thus prompts concern as to whether too many (or too few) of these procedures are performed both in aggregate and among key patient subgroups. Any effort to answer these concerns raises the issues of TKA indication, timing, and appropriateness. Cram et al. [6] noted ‘‘it is difficult to determine the extent to which the growth in TKA use represents appropriate use of a highly effective procedure or the overuse of a well-reimbursed procedure for which indications depend on clinical judgment. It is likely that both factors are at play.’’ We therefore compared early data from a recently established US national TKA cohort including patients from 123 surgeons in 22 states to evaluate preoperative variability in patient selection across surgeons. In particular, we asked whether surgeons use similar selection criteria based on preoperative patient-reported outcomes (PROs) when selecting patients to undergo TKA.
Data From a US National TKA Cohort
77
prospectively maintained database of patient-reported outcomes in a cohort of patients recruited from more than 100 surgeons who practice in 22 states.
Study Population/Measures Study Sample
We conducted an analysis of a consecutive sample of patients undergoing primary TKA or THA from a large,
The Agency for Health Care Research and Quality funded a USD 12 million research program, Function and Outcomes Research in Comparative Effectiveness Research in Total Joint Replacement (FORCE-TJR), to assemble a national cohort of patients undergoing TJA with comprehensive data elements [11]. FORCE-TJR includes patients from a national consortium of 136 orthopaedic surgeons representing all regions of the United States (Fig. 1) as well as varied hospital and surgeon practice settings. To date 19,200 patients have been enrolled; 6900 have reached 12 months followup. The database has allowed us to publish three articles [2, 11, 12] and submit several others that primarily deal with outcome analysis of this population. The first consecutive 4900 FORCE-TJR patients undergoing primary, unilateral TKA from 22 sites and with signed informed consent and enrolled in 2011 and 2012 were included in this analysis. Patients had a mean age of 66 years and were drawn from 24 orthopaedic practices (more than 100 surgeons). Sixty-two percent were female. By study design FORCE returns scored data quarterly, because PROs are not scored real-time in the office. This means that the patient and surgeon made the surgical decision before the surgeon received the scored presurgical
Fig. 1 This figure shows the locations of the FORCE-TJR participating sites across the United States: (1) Core Clinical Centers: University of Massachusetts Medical School, Worcester, MA, USA (blue star); Connecticut Joint Replacement Institute, Hartford, CT, USA (red star);
The University of Rochester Medical Center, Rochester, NY, USA (green star); Medical University of South Carolina, Charleston, SC, USA (gray star); Baylor College of Medicine, Houston, TX, USA (light blue star); and (2) Community Sites (green dots).
Materials and Methods Study Design
123
78
Clinical Orthopaedics and Related Research1
Ayers et al.
PROs. The decision to schedule elective TKA was based on the surgeons’ routine indications and clinical judgment.
We then observed whether any individual site median differed from (1) the national median or (2) all other site medians on pain and function scores to a degree greater than the minimal clinical important difference.
Surgeons and Patients in the Database FORCE-TJR surgeons vary in annual TJA volume with 13% performing an estimated 0 to 50 TJAs/year, 28% performing 51 to 200 TJAs/year, and 59% performing more than 200 TJAs/year. Five are academic hospitals/centers and eight are rural hospitals. Community-based surgeons represent a large portion (75%) of this national cohort, consistent with national TJA practice. Enrolled surgeons include both fellowship-trained and general orthopaedic surgeons. Patients in this database have private, public, and HMO insurance.
Measures FORCE-TJR collects extensive patient demographic information, complete medical and musculoskeletal comorbidities, surgical care data, implant data, postoperative complications, adverse events, and readmission. FORCE-TJR collects several validated PROs, directly from patients, to estimate functional outcomes and pain relief. This study presents data using a global function measure, the SF-36 Physical Component Summary (PCS) [21] and a knee-specific outcomes measure, the Knee Injury and Osteoarthritis Outcome Score (KOOS) [19], that includes a pain score and activities of daily living (ADL)/function score questionnaires. Of note, the KOOS ADL/function score is equivalent to the widely used WOMAC function index [3] but is nonproprietary [19]. For reference, the PCS scores range from 0 to 80 with the highest score representing greatest function. For the PCS, 50 is the national adult norm. The KOOS ADL score ranges from 0 to 100 with 100 representing perfect function. For the KOOS pain, a score of 100 represents no pain, and pain increases as scores approach 0. FORCE-TJR collects data before surgery, 6 and 12 months postoperatively, and yearly thereafter. All patients in this study completed preoperative PROs. By study design FORCE returns data quarterly, because PROs are not scored real-time in the office. This means that the surgical decision was made before the surgeon received the scored pre-PROs. To assure stable PRO estimates for each site, only surgeons who enrolled more than 20 patients by mid-2012 were included in this analysis.
Results The median baseline (preoperative) PCS score was 32.6 with a range of medians across sites from 29.3 to 34.4 (Fig. 2). Preoperative scores were almost 2 SDs below the general population PCS norm (mean, 50; SD, 10) [20] and were consistent with advanced disability resulting from arthritis. For context, the minimum clinically important change (MCIC) for PCS is 5 points [20]. The lowest median (29.3) and highest median (34.4) were less than 3 points from the national median, or less than the MCIC. This suggests no clinical difference could be detected in the median (typical) patient across all sites. The 25th percentile PCS mean for the study sample was 26.9 (range, 23.9–28), whereas the 75th percentile PCS mean was 38.1 (range, 35.3–41). More than 81% had preoperative scores greater than, or equal to, 1 SD below the national norm of 50. The median KOOS pain score was 44.4 with a range of medians across sites from 36.1 to 50, representing clinically important pain because a score of 90 to 100 represents none or minimal pain (Fig. 3). The MCIC is 7 to 8 points. One SD represents approximately 15 points. Thus, the median pain scores across sites differed less than the MCIC, meaning not clinically different. Like with PCS, we observed a tight clustering of 25th and 75th percentile for the KOOS pain scores by site (Fig. 3). The median KOOS ADL, or function, score was 51.5 with a range of medians across sites from 39.7 to 57.8 (Fig. 4) as compared with the MCIC of 7 to 8. There was a tight clustering of 25th and 75th percentile for the KOOS ADL scores by site.
Statistical Analysis We analyzed the data descriptively. We estimated select descriptive characteristics (mean, 25th, 50th, and 75th percentiles) of function and pain scores by participating site.
123
Fig. 2 This figure depicts the baseline primary TKA SF-36 PCS by site. The red line represents median across sites. The national PCS norm for the general population is 50.
Volume 473, Number 1, January 2015
Fig. 3 This figure depicts the baseline primary TKA KOOS pain score by site. The red line represents median across sites. Being painfree corresponds to a score of 90 to 100.
Fig. 4 This figure depicts the baseline primary TKA KOOS/ADL score by site. The red line represents median across sites. The ideal function corresponds to a score between 90 and 100.
Discussion The rapid growth in use of TKA is often attributed to the aging US population because more than 60% of adults older than 65 years of age report knee or hip osteoarthritis. However, the growing number of patients younger than 65 years raises the question as to whether surgeon selection of appropriate patients is evolving. We hypothesized that if surgeon selection is changing, the typical patient undergoing TKA would report less pain and disability at the time of surgery and we would see variation in typical patients across surgeons. The preoperative data from this 2011 to 2012 national sample of patients undergoing primary TKA suggest that surgeons across the United States applied consistent patient selection criteria as defined by preoperative pain and function scores. Both the median preoperative global function (PCS) and median knee-specific pain and function (KOOS) scores represent a large degree of pain and disability at the time of TKA. This suggests that the rapid increase in number of patients
Data From a US National TKA Cohort
79
undergoing TKA cannot be explained fully by clinicians extending surgical indications to patients with less severe symptoms. This study had a number of limitations. First, the typical patient undergoing TKA is described by self-report of pain and disability, because we do not have the physical examination or radiographic findings that informed the surgeon’s decision for TKA. Future research may compare patient-reported symptoms as well as physical examination findings (eg, contractures or deformities) and severity of arthritis changes on radiographs to fully illuminate the clinical decision process. Second, although FORCE-TJR surgeons perform diverse TKA volumes and practice in varied settings, these observations are based on a sample of US arthroplasty surgeons. It would be valuable to validate these findings in another national sample before drawing conclusions about the entire US surgical community. Finally, we noted some variation in the higher and lower quartiles in function and pain scores. Future research should examine the full clinical rationales for surgery, including the physical examination and supporting radiographic images, before any conclusions can be drawn about the timing and need for TKA. The international literature documents variation in the objective criteria associated with the optimal timing of TKA. A 2006 literature review highlighted large variation among orthopaedic surgeons as to the indications for TKA [7]. Similarly, a Canadian study attributed much of the regional variation in TKA use to differences in surgeon opinions when recommending patients for TKA [23]. Research in Australia and Europe found that variation in age, pain, stiffness, and physical function before TKA and THA are associated with substantial differences in timing of surgery [1]. Ackerman et al. [1] conclude that, ‘‘Possible contributing factors [to premature or delayed TKR timing] include patient preferences, the absence of concrete indications for surgery, and the capacity of health care systems.’’ In contrast to these reports of diverse indications for TKA, the typical (median) patient pain and disability scores in this sample correspond with advanced clinical symptoms. Furthermore, these patients reported similar median scores across sites. By definition, 50% of patients had more pain and poorer function than the median. It is possible that the studies describing varied TKA indications have only focused on patients reporting the upper 25% to 30% of pain and function scores. In this cohort, additional clinical data are needed to make any assessment about the indication for surgery among these patients reporting less severe symptoms. Finally, it is possible that surgeons who agreed to participate in FORCE-TJR are conservative in their patient selection criteria, because they know our analysts will review the data. Although this is possible,
123
80
Clinical Orthopaedics and Related Research1
Ayers et al.
only three of the participating surgeons had used PROs in their clinical practice before their participation in FORCETJR so most surgeons had no established PRO criteria for surgical timing and depended on physical examination and radiographs alone to make a recommendation. Few studies have estimated appropriateness measures for TKA use. In a multicenter prospective observational study involving 15 hospitals in Spain, Cobos et al. [5] estimated that as many as 25% of the TKAs and THAs performed might be considered inappropriate. The authors attributed this finding to variability among the clinical criteria used by surgeons when selecting patients for surgery [5]. Although the current study cannot comment on the appropriateness of the TKA surgical decision, the PRO scores at the median are consistent with advanced pain and functional limitations. The Osteoarthritis Research Society International (OARSI)/Outcome Measures in Rheumatology (OMERACT) group recently concluded that the most discriminating criteria in regard to indication for TKA were the symptoms and duration of pain and physical function impairment [15]. This recommendation parallels the National Institutes of Health consensus statement on TKA issued in 2003 [18]. In the United States, the landmark patient outcomes research team (PORT) TKA study in the 1980s defined PRO norms at the time of primary TKA [14]. Of note, the median pain and function scores in this sample are similar to those reported by the PORT study in the 1980s. The timing of TKA is a crucial decision and surgeons agree that patients should have severe knee pathology to warrant surgery. Pain relief is achieved in the vast majority of patients undergoing TKA, and measurable post-TKA functional gain is reported among at least two-thirds of patients undergoing TKA. However, patients with the poorest function before surgery are more likely to have smaller functional gains [10, 13]. In a sample of 8300 patients undergoing primary TKA from a national implant database, variation in functional outcome after TKA was associated with patient attributes, including poorer preoperative function [13]. Of note, the mean preoperative PCS was 30 (SD, 7.8) in this national sample accrued 10 years ago, which is similar to that in our study. Our data suggest that despite the growth in the number of patients undergoing TKA during the past decade, today’s typical patient has comparable preoperative disability as estimated by the SF-36 PCS function score. Further analysis of the upper 25th percentile of patients reporting less pain and functional limitation could provide insight into the role that patient demand to maintain higher levels of activity and tolerate less pain may play in TKA timing. It is possible that both observations offer different sides to the story. Although the typical patient, and the lower 50%, have symptoms of
123
advanced joint disease, a subgroup of the patients with lesser symptoms may reflect evolving patient demands. Specific guidelines to inform surgeon and patient decisions for TKA would be helpful. To this end, OARSI/ OMERACT recently attempted to define a ‘‘theoretical indication for TKA’’ using the severity and duration of symptoms of pain, physical function as well as structural knee damage assessed by radiographs. However, variations among individual patients, surgeons, and countries made it difficult to pinpoint relevant cut points consistent with patient need for TKA. Radiographic evidence did not correlate with symptoms of pain and physical function. Again, the most discriminating criteria identified with appropriate timing for TKA were symptoms and duration of pain and physical function [13]. In conclusion, patients differ in both their clinical presentation and demand for physical activity, so surgeon evaluation guided by the patient’s history is critical to individualized decisions for TKA. The future availability of real-time, scored, and trended patient-reported outcomes may inform in-office decisions for surgery [2]. Finally, new national cohorts with patient-reported outcomes such as FORCE-TJR may help define normative guidelines to support patient selection and TKA timing [8, 11]. Our data, which represent an early iteration of FORCE-TJR, suggest that surgeons use consistent patient criteria in scheduling the median or typical patient undergoing TKA. In addition, these patients report significant levels of disability at the time of surgery. In the future, the addition of national norms for patient-reported pain and function measures may be an important complement to clinical data to support consistency in patient selection of patients most likely to achieve an optimal TKA outcome. Acknowledgments We gratefully acknowledge the more than 120 surgeons who participate in the FORCE-TJR national consortium, especially the core site Principal Investigators, Courtland Lewis MD, Philip Noble PhD, Regis O’Keefe MD, PhD, and Vincent Pellegrini MD. We also appreciate the editorial assistance of Sylvie Puig PhD.
References 1. Ackerman IN, Dieppe PA, March LM, Roos EM, Nilsdotter A, Brown GC, Sloan KE, Osborne RH. Variation in age and physical status prior to total knee and hip replacement surgery: a comparison of centers in Australia and Europe. Arthritis Rheum. 2009;61:166–173. 2. Ayers DC, Zheng H, Franklin PD. Integrating patient-reported outcomes (PROs) into orthopaedic clinical practice: proof of concept from FORCE-TJR. Clin Orthop Relat Res. 2013;471: 3419–3425. 3. Bellamy N, Buchanan WW, Goldsmith CH, Campbell J, Stitt LW. Validation study of WOMAC: A health status instrument for measuring clinically important patient relevant outcomes to
Volume 473, Number 1, January 2015
4. 5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
antirheumatic drug therapy in patients with osteoarthritis of the hip or knee. J Rheumatol. 1988;12:1833–1840. Chernew M, Goldman D, Axeen S. How much savings can we wring from Medicare? N Engl J Med. 2011;365:e29. Cobos R, Latorre A, Aizpuru F, Guenaga JI, Sarasqueta C, Escobar A, Garcia L, Herrera-Espineira C. Variability of indication criteria in knee and hip replacement: an observational study. BMC Musculoskelet Disord. 2010;11:249.. Cram P, Lu X, Kates SL, Singh JA, Li Y, Wolf BR. Total knee arthroplasty volume, utilization, and outcomes among Medicare beneficiaries, 1991–2010. JAMA. 2012;308:1227–1236. Cross WW 3rd, Saleh KJ, Wilt TJ, Kane RL. Agreement about indications for total knee arthroplasty. Clin Orthop Relat Res. 2006;446:34–39. Dieppe P, Lim K, Lohmander S. Who should have knee joint replacement surgery for osteoarthritis? Int J Rheum Dis. 2011;14:175–180. Felson DT, Chaisson CE. Understanding the relationship between body weight and osteoarthritis. Baillieres Clin Rheumatol. 1997;11:671–681. Fortin PR, Penrod JR, Clarke AE, St-Pierre Y, Joseph L, Belisle P, Liang MH, Ferland,D, Phillips CB, Mahomed N, Tanzer M, Sledge C, Fossel AH, Katz JN. Timing of total joint replacement affects clinical outcomes among patients with osteoarthritis of the hip or knee. Arthritis Rheum. 2002;46:3327–3330. Franklin PD, Allison JJ, Ayers DC. Beyond joint implant registries: a patient-centered research consortium for comparative effectiveness in total joint replacement. JAMA. 2012;308:1217–1218. Franklin PD, Harrold LR, Ayers DC. Incorporating patient reported outcomes in total joint arthroplasty registries: challenges and opportunities. Clin Orthop Relat Res. 2013;471:3482– 3488. Franklin PD, Li W, Ayers DC. The Chitranjan Ranawat Award: Functional outcome after total knee replacement varies with patient attributes. Clin Orthop Relat Res. 2008;466:2597–604. Freund D, Lave J, Clancy C, Hawker G, Hasselblad V, Keller R, Schneiter E, Wright J. Patient outcomes research teams: contribution to outcomes and effectiveness research. Annu Rev Public Health. 1999:337–359.
Data From a US National TKA Cohort
81
15. Gossec L, Paternotte S, Bingham CO 3rd, Clegg DO, Coste P, Conaghan PG, Davis AM, Giacovelli G, Gunther KP, Hawker G, Hochberg MC, Jordan JM, Katz JN, Kloppenburg M, Lanzarotti A, Lim K, Lohmander LS, Mahomed NN, Maillefert JF, Manno RL, March LM, Mazzuca SA, Pavelka K, Punzi L, Roos EM, Rovati LC, Shi H, Singh JA, Suarez-Almazor ME, Tajana-Messi E, Dougados M. OARSI/OMERACT initiative to define states of severity and indication for joint replacement in hip and knee osteoarthritis. An OMERACT 10 special interest group. J Rheumatol. 2011;38:1765–1769. 16. HCUPnet. Healthcare Cost and Utilization Project (HCUP). Nationwide Inpatient Sample (NIS). Rockville, MD, USA: Agency for Healthcare Research and Quality; 2011. Available at: http://hcupnet.ahrq.gov/. Accessed October 11, 2013. 17. Kurtz S, Ong K, Lau E, Mowat F, Halpern M. Projections of primary and revision hip and knee arthroplasty in the united states from 2005 to 2030. J Bone Joint Surg Am. 2007;89:780–785. 18. Rankin EA, Alarcon GS, Chang RW, Cooney LM Jr, Costley LS, Delitto A, Dey RA, Donaldson SK, Hochberg MC, MacLean CH, Yelin EH, Marciel K. NIH consensus statement on total knee replacement, December 8–10, 2003. J Bone Joint Surg Am. 2004;86:1328–1335. 19. Roos EM, Roos HP, Lohmander LS, Ekdahl C, Beynnon BD. Knee Injury and Osteoarthritis Outcome Score (KOOS)—development of a self-administered outcome measure. J Orthop Sports Phys Ther. 1998;28:88–96. 20. Ware JE, Kosinski M, Bjomer JB, Turner-Bowker DM, Gandek B, Maruish ME. User’s Manual for the SF-36v2 Health Survey. 2nd ed. Lincoln, RI, USA: QualityMetric Incorporated; 2007. 21. Ware JE Jr, Sherbourne CD. The MOS 36-item short-form health survey (SF-36). I. Conceptual framework and item selection. Med Care. 1992;30:473–483. 22. Woolf AD, Breedveld F, Kvien TK. Controlling the obesity epidemic is important for maintaining musculoskeletal health. Ann Rheum Dis. 2006;65:1401–1402. 23. Wright JG, Hawker GA, Bombardier C, Croxford R, Dittus RS, Freund DA, Coyte PC. Physician enthusiasm as an explanation for area variation in the utilization of knee replacement surgery. Med Care. 1999;37:946–956.
123
