Faculty Bibliography

Developing effective treatment for osteoarthritis (OA), a prevalent and disabling disease, has remained a challenge, primarily because of limited understanding of its pathogenesis and late diagnosis. In the subchondral bone, rapid bone loss after traumatic injuries and bone sclerosis at the advanced stage of OA are well-recognized hallmarks of the disease. Recent studies have further demonstrated the crucial contribution of subchondral bone in the development of OA. However, the microstructural basis of these bone changes has not been examined thoroughly, and the paradox of how abnormal resorption can eventually lead to bone sclerosis remains unanswered. By applying a novel microstructural analysis technique, individual trabecula segmentation (ITS), to micro-computed tomography (μCT) images of human OA knees, we have identified a drastic loss of rod-like trabeculae and thickening of plate-like trabeculae that persisted in all regions of the tibial plateau, underneath both severely damaged and still intact cartilage. The simultaneous reduction in trabecular rods and thickening of trabecular plates provide important insights to the dynamic and paradoxical subchondral bone changes observed in OA. Furthermore, using an established guinea pig model of spontaneous OA, we discovered similar trabecular rod loss and plate thickening that preceded cartilage degradation. Thus, our study suggests that rod-and-plate microstructural changes in the subchondral trabecular bone may play an important role in the development of OA and that advanced microstructural analysis techniques such as ITS are necessary in detecting these early but subtle changes. With emerging high-resolution skeletal imaging modalities such as the high-resolution peripheral quantitative computed tomography (HR-pQCT), trabecular rod loss identified by ITS could potentially be used as a marker in assessing the progression of OA in future longitudinal studies or clinical diagnosis. © 2017 American Society for Bone and Mineral Research.

Osteoarthritis (OA) is a prevalent joint disease, yet its pathogenesis remains unclear. Studies have shown that detecting and blocking acute subchondral bone loss associated with traumatic injuries may prevent OA. By applying Individual Trabecula Segmentation (ITS) to muCT scans, we previously identified a loss of trabecular rods in the subchondral trabecular bone as an early and consistent microstructural marker of OA in animal and in human. To examine the potentials of rod loss as a quantitative and reliable biomarker to monitor OA in vivo, the sensitivity of ITS to detect these changes clinically was tested using the second-generation HR-pQCT, a valuable clinical scanner that permits in vivo quantification of trabecular microstructure in the knee. OA (n = 4) and control (n = 4) tibial plateaus were collected from total knee replacement patients and cadavers with no history of metabolic bone disease or fracture. Samples were scanned by muCT at 26-mm voxel size and by HR-pQCT at 61-mm voxel size. Images were registered and thresholded by matching BV/TV between the two resolutions. Cubic trabecular regions were selected from the subchondral bone beneath intact and severely damaged cartilage in OA samples and from corresponding control regions. ITS was used to analyze and compare trabecular morphology between OA and control at both resolutions. High correlation in morphological parameters was found in muCT and HR-pQCT scans. Furthermore, ITS analyses of muCT images and HR-pQCT images consistently reported significant increases in PR ratio in the OA subchondral bone beneath both intact and damaged cartilage. Interestingly, this increase was due to a simultaneous loss of trabecular rods by number and thickening of trabecular plates beneath intact cartilage, while solely due to plate thickening beneath damaged cartilage. Most importantly, despite the reduced resolution of HR-pQCT, ITS was still sensitive to these subtle but dynamic changes in trabecular microstructure. We previously showed that rod loss may be an early signature of OA. This study demonstrates the first attempt to correlate ITS based on 2nd generation HR-pQCT with gold standard muCT, and confirms the sensitivity of ITS to this rod loss at clinical resolution. The combination of ITS and HR-pQCT may serve as a useful tool in identifying early signs of bone loss quantitatively, monitoring microstructural changes in disease progression and identifying the best time window for preventative treatments. (Figure Presented)

BACKGROUND: The Center for Medicare and Medicaid Services (CMS) has proposed a move to payment based on patient-reported outcomes (PROs), and failure to report on PROs will result in a penalty of 2% in 2016. However, the cost to the physician to collect PROs is not known. METHODS: Using data from the 2013 Medical Group Management Association Compensation and Financial survey and Center for Medicare and Medicaid Services reimbursement, a calculation was performed to determine the cost to the physician to report on PROs for patients undergoing total knee arthroplasty and total hip arthroplasty. Using Medical Group Management Association and Medicare fee for service rates, calculations were performed based on an annual volume of 200 Medicare operative cases (125 total knee arthroplasties, 75 total hip arthroplasties) with 1000 new patients (level 4) and 2000 established patients (level 3) visits. A range of start-up and annual costs necessary to collect PROs including hardware, software, and personnel costs was calculated and then compared with the calculated 2% Medicare penalty for failing to report PROs in 2016. RESULTS: The cost to collect PROs ranged from $47,973 to $56,288 which far outweighed the penalty of $2954 in 2016 for failing to report these measures. CONCLUSION: With the move toward requiring surgeons to report PROs for reimbursement, the current financial model would prove to be cost prohibitive and the incentive to report PROs might be too costly to gain wide acceptance.

BACKGROUND: Perioperative care pathways are tools used in high-volume clinical settings to standardize care, reduce variability, and improve outcomes. However, the mechanism by which the information is transmitted to other caregivers is often inconsistent and error-prone. At our institution, we developed an online, user-editable ("wiki") database to communicate post-operative protocols. The purpose of this study is to evaluate the hypothesis that implementation of the wiki would improve protocol adherence and reduce unintentional deviations inpatient care. METHODS: We conducted a retrospective review of patients who underwent primary lower extremity arthroplasty at our institution during three 6-month time periods including immediately before, 6 months after, and 2 years following introduction of the wiki. Adherence to defined perioperative care pathways (laboratory studies, post-operative imaging, perioperative antibiotics, and inpatient pain medications) was compared between the groups. RESULTS: After wiki implementation, adherence to protocols improved significantly for laboratory orders (P < .0001), imaging (P < .001), pain control regimen (P = .03), and overall protocol adherence (P < .001). Improvements were seen in some areas almost immediately, while others did not show improvements until 2 years after implementation. Costs associated with unnecessary testing were reduced by 82%. CONCLUSION: Development of an online wiki for tracking post-operative protocols improves care pathway adherence and reduces variability in care while lowering costs associated with unnecessary testing, although some benefits may not be immediately realized. Several practical barriers to implementing the wiki are also discussed, along with proposed solutions.

BACKGROUND: Hemiarthroplasty (HA) has been a mainstay treatment for displaced femoral neck fractures for many years. The purpose of this study was to report the conversion rate of HA to total hip arthroplasty (THA) for displaced femoral neck fractures and compare outcomes between implant constructs (bipolar vs unipolar), fixation options (cemented vs cementless stems), and age groups (<75 years vs >/=75 years). METHODS: We retrospectively reviewed the results of a consecutive cohort of 686 patients who underwent HA for the treatment of femoral neck fractures at our institution between 1999 and 2013 with a minimum of 2-year follow-up. RESULTS: The overall component revision rate, including conversion to THA, revision HA, revision with open reduction internal fixation, and Girdlestone procedure, was 5.6% (39/686). Seventeen patients (2.5%) were converted from HA to THA at an average of 1.9 years after index procedure. A significantly lower conversion rate of 1.4% (7/499 patients) was found in the older patient cohort (>/=75 years old) compared to 5.3% (11/187) in the younger cohort. The most common causes for conversion surgery to THA were acetabular wear (5 patients), aseptic loosening (4 patients), and periprosthetic fracture (3 patients). There was a significantly lower rate of periprosthetic fracture (0.4% vs 2.5%, P value .025) in the cemented implant group compared to the cementless group. We observed a higher rate of dislocations in the bipolar vs unipolar group (3.8% vs 1%, P value .02) and no other significant differences between these groups. CONCLUSION: We observed a low reoperation rate for this cohort of patients, relatively higher conversion rates for the younger population, fewer periprosthetic fractures with the use of cemented stems, and no advantage of bipolar over unipolar prostheses.

BACKGROUND: The goal of this study was to evaluate the effectiveness of the American Orthopaedic Association's Own the Bone secondary fracture prevention program in the United States. METHODS: The objective of this quality improvement cohort study was dissemination of Own the Bone and implementation of secondary prevention (osteoporosis pharmacologic and bone mineral density [BMD] test recommendations). The main outcome measures were the number of sites implementing Own the Bone and implementation of secondary prevention, i.e., orders for BMD testing and/or pharmacologic treatment. The 177 sites participating in the program were academic and community hospitals, orthopaedic surgery groups, and a health system; data were obtained from the first 125 sites utilizing its registry, between January 1, 2010, and March 31, 2015. It included all patients, aged 50 years or older, presenting with fragility fractures (n = 23,132) who were enrolled in the Own the Bone web-based registry. The interventions were education, development of program elements, dissemination, implementation, and evaluation of the Own the Bone program at participating sites. RESULTS: A growing number of institutions implemented Own the Bone (14 sites in 2005-2006 to 177 sites in 2015). After consultation, 53% of patients had a BMD test ordered and/or pharmacologic therapy for osteoporosis. CONCLUSIONS: The Own the Bone intervention has succeeded in improving the behaviors of medical professionals in the areas of osteoporosis treatment and counseling, BMD testing, initiation of pharmacotherapy, and coordination of care for patients who have experienced a fragility fracture.

BACKGROUND: Osteoporotic hip fractures are common injuries typically occurring in patients who are older and medically frail. Studies have suggested that creation of a multidisciplinary team including orthopaedic surgeons, internal medicine physicians, social workers, and specialized physical therapists, to comanage these patients can decrease complication rates, improve time to surgery, and reduce hospital length of stay; however, they have yet to achieve widespread implementation, partly owing to concerns regarding resource requirements necessary for a comanagement program. QUESTIONS/PURPOSES: We performed an economic analysis to determine whether implementation of a comanagement model of care for geriatric patients with osteoporotic hip fractures would be a cost-effective intervention at hospitals with moderate volume. We also calculated what annual volume of cases would be needed for a comanagement program to "break even", and finally we evaluated whether universal or risk-stratified comanagement was more cost effective. METHODS: Decision analysis techniques were used to model the effect of implementing a systems-based strategy to improve inpatient perioperative care. Costs were obtained from best-available literature and included salary to support personnel and resources to expedite time to the operating room. The major economic benefit was decreased initial hospital length of stay, which was determined via literature review and meta-analysis, and a health benefit was improvement in perioperative mortality owing to expedited preoperative evaluation based on previously conducted meta-analyses. A break-even analysis was conducted to determine the annual case volume necessary for comanagement to be either (1) cost effective (improve health-related quality of life enough to be worth additional expenses) or (2) result in cost savings (actually result in decreased total expenses). This calculation assumed the scenario in which a hospital could hire only one hospitalist (and therapist and social worker) on a full-time basis. Additionally, we evaluated the scenario where the necessary staff was already employed at the hospital and could be dedicated to a comanagement service on a part-time basis, and explored the effect of triaging only patients considered high risk to a comanagement service versus comanaging all geriatric patients. Finally, probabilistic sensitivity analysis was conducted on all critical variables, with broad ranges used for values around which there was higher uncertainty. RESULTS: For the base case, universal comanagement was more cost effective than traditional care and risk-stratified comanagement (incremental cost effectiveness ratios of USD 41,100 per quality-adjusted life-year and USD 81,900 per quality-adjusted life-year, respectively). Comanagement was more cost effective than traditional management as long as the case volume was more than 54 patients annually (range, 41-68 patients based on sensitivity analysis) and resulted in cost savings when there were more than 318 patients annually (range, 238-397 patients). In a scenario where staff could be partially dedicated to a comanagement service, universal comanagement was more cost effective than risk-stratified comanagement (incremental cost effectiveness of USD 2300 per quality-adjusted life-year), and both comanagement programs had lower costs and better outcomes compared with traditional management. Sensitivity analysis was conducted and showed that the level of uncertainty in key variables was not high enough to change the core conclusions of the model. CONCLUSIONS: Implementation of a systems-based comanagement strategy using a dedicated team to improve perioperative medical care and expedite preoperative evaluation is cost effective in hospitals with moderate volume and can result in cost savings at higher-volume centers. The optimum patient population for a comanagement strategy is still being defined. LEVEL OF EVIDENCE: Level 1, Economic and Decision Analysis.

BACKGROUND: Acetabular component orientation influences joint stability in total hip arthroplasty (THA). The purpose of this study was to evaluate the effect of cup orientation and other variables on hip dislocation risk and to define a posterior approach specific safe zone. METHODS: A cohort of 1289 posterior approach primary THA cases was prospectively followed and component position measured radiographically. RESULTS: Cup malposition, with respect to the Lewinnek safe zone, was an independent risk factor for dislocation (OR1.88). Modifying the anteversion safe zone limits to 10-25 degrees strongly predicted increased dislocation risk (OR2.69). No dislocations occurred within a zone defined by a circle centered at 41.4 degrees abduction and 17.1 degrees anteversion, radius 4.3 degrees . CONCLUSION: Utilizing a posterior approach specific safe zone of 10-25 degrees anteversion and 30-50 degrees abduction may minimize THA dislocations. LEVEL OF EVIDENCE: Level III.

BACKGROUND: This study evaluated patient-reported outcomes in patients undergoing primary total hip arthroplasty with a polyethylene liner to determine the influence of cup orientation and other variables on patient-reported outcomes. METHODS: A total of 477 cases were prospectively monitored through average 4.7 years follow-up. Cup position was measured on pelvis radiographs. Patients completed the Western Ontario and McMaster Universities Osteoarthritis Index and Short Form 12 Health Survey questionnaires. RESULTS: Average cup abduction was 43.1 degrees +/- 7.5 degrees and anteversion was 13.3 degrees +/- 7.5 degrees . Three hundred cups were within the target zone. All outcomes' improvement from baseline and cup position was not an independent risk factor for the Western Ontario and McMaster Universities Osteoarthritis Index or Short Form 12 Health Survey improvement. CONCLUSION: Accurate cup orientation may not be critical to maximizing patient-perceived outcomes if the combined anteversion is within a normal range, the hip joint is properly balanced, and a polyethylene liner is coupled with a metal or ceramic femoral head.