Identifying Modifiable Cost Drivers of Outpatient Unicompartmental Knee Arthroplasty With Machine Learning.

BACKGROUND: Implementing tools that identify cost-saving opportunities for ambulatory orthopaedic surgeries can improve access to value-based care. We developed and internally validated a machine learning (ML) algorithm to predict cost drivers of total charges after ambulatory unicompartmental knee arthroplasty (UKA). METHODS: We queried the New York State Ambulatory Surgery and Services database to identify patients who underwent ambulatory, defined as <24 hours of care before discharge, elective UKA between 2014 and 2016. A total of 1,311 patients were included. The median costs after ambulatory UKA were $14,710. Patient demographics and intraoperative parameters were entered into 4 candidate ML algorithms. The most predictive model was selected following internal validation of candidate models, with conventional linear regression as a benchmark. Global variable importance and partial dependence curves were constructed to determine the impact of each input parameter on total charges. RESULTS: The gradient-boosted ensemble model outperformed all candidate algorithms and conventional linear regression. The major differential cost drivers of UKA identified (in decreasing order of magnitude) were increased operating room time, length of stay, use of regional and adjunctive periarticular analgesia, utilization of computer-assisted navigation, and routinely sending resected tissue to pathology. CONCLUSION: We developed and internally validated a supervised ML algorithm that identified operating room time, length of stay, use of computer-assisted navigation, regional primary anesthesia, adjunct periarticular analgesia, and routine surgical pathology as essential cost drivers of UKA. Following external validation, this tool may enable surgeons and health insurance providers optimize the delivery of value-based care to patients receiving outpatient UKA. LEVEL OF EVIDENCE: III.

Deep Learning Artificial Intelligence Model for Assessment of Hip Dislocation Risk Following Primary Total Hip Arthroplasty From Postoperative Radiographs.

BACKGROUND: Dislocation is a common complication following total hip arthroplasty (THA), and accounts for a high percentage of subsequent revisions. The purpose of this study is to illustrate the potential of a convolutional neural network model to assess the risk of hip dislocation based on postoperative anteroposterior pelvis radiographs. METHODS: We retrospectively evaluated radiographs for a cohort of 13,970 primary THAs with 374 dislocations over 5 years of follow-up. Overall, 1490 radiographs from dislocated and 91,094 from non-dislocated THAs were included in the analysis. A convolutional neural network object detection model (YOLO-V3) was trained to crop the images by centering on the femoral head. A ResNet18 classifier was trained to predict subsequent hip dislocation from the cropped imaging. The ResNet18 classifier was initialized with ImageNet weights and trained using FastAI (V1.0) running on PyTorch. The training was run for 15 epochs using 10-fold cross validation, data oversampling, and augmentation. RESULTS: The hip dislocation classifier achieved the following mean performance (standard deviation): accuracy = 49.5 (4.1%), sensitivity = 89.0 (2.2%), specificity = 48.8 (4.2%), positive predictive value = 3.3 (0.3%), negative predictive value = 99.5 (0.1%), and area under the receiver operating characteristic curve = 76.7 (3.6%). Saliency maps demonstrated that the model placed the greatest emphasis on the femoral head and acetabular component. CONCLUSION: Existing prediction methods fail to identify patients at high risk of dislocation following THA. Our radiographic classifier model has high sensitivity and negative predictive value, and can be combined with clinical risk factor information for rapid assessment of risk for dislocation following THA. The model further suggests radiographic locations which may be important in understanding the etiology of prosthesis dislocation. Importantly, our model is an illustration of the potential of automated imaging artificial intelligence models in orthopedics. LEVEL OF EVIDENCE: Level III.

Costs of open, arthroscopic and combined surgery for developmental dysplasia of the hip.

A variety of options exist for management of patients with developmental dysplasia of the hip (DDH). Most studies to date have focused on clinical outcomes; however, there are currently no data on comparative cost of these techniques. The purpose of this study was to evaluate in-hospital costs between patients managed with periacetabular osteotomy, hip arthroscopy or a combination for DDH. One hundred and nine patients were included: 35 PAO + HA, 32 PAO and 42 HA. There were no significant differences in the demographic parameters. Operative times were significantly different between groups with a mean of 52 min for PAO, 100 min for HA and 155 min for PAO + HA, (P < 0.001). Total direct medical costs were calculated and adjusted to nationally representative unit costs in 2017 inflation-adjusted dollars. Total in-hospital costs were significantly different between each of the three treatment groups. PAO + HA was the most expensive with a median of $21 852, followed by PAO with a median of $15 124, followed by HA with a median of $11 582 (P < 0.001). There was a significant difference between outpatient median costs of $11 385 compared with $24 320 for inpatients (P < 0.001). Procedures with greater complexity were more expensive. However, a change from outpatient to inpatient status with HA moved that group from the least expensive to similar to PAO and PAO + HA. These data provide an important complement to clinical outcomes reports as surgeons and policymakers aim to provide optimal value.

Revision Total Hip Arthroplasty for the Treatment of Fracture: More Expensive, More Complications, Same Diagnosis-Related Groups: A Local and National Cohort Study.

BACKGROUND: Revision total hip arthroplasty (revision THA) occurs for a wide variety of indications and in the United States it is coded under Diagnosis-Related Groups (DRGs) 466, 467, and 468, which do not account for revision etiology, a potentially substantial driver of cost. This study investigates revision THA costs and 30-day complications by indication, both locally and nationally. METHODS: Hospitalization costs and complication rates for 1,422 aseptic revision THAs performed at a high-volume center between 2009 and 2014 were retrospectively reviewed. Additionally, charges for 28,133 revision THAs in the National Inpatient Sample (NIS) were converted to costs using the Healthcare Cost and Utilization Project cost-to-charge ratios, and 30-day complication rates for 3,224 revision THAs were obtained with use of the American College of Surgeons National Surgical Quality Improvement Program (ACS-NSQIP). Costs and complications were compared between revision THAs performed for fracture, wear/loosening, and dislocation/instability with use of simultaneous and pairwise comparisons and a multivariable model accounting for American Society of Anesthesiologists (ASA) score, age, and sex. RESULTS: Local hospitalization costs for fracture (median, $25,672) were significantly higher than those for wear/loosening ($20,228; p < 0.001) or dislocation/instability ($17,911; p < 0.001), with differences remaining significant even after adjusting for patient comorbidities (p < 0.001). NIS costs for fracture (median, $27,596) were higher than those for other aseptic indications (wear/loosening: $21,176, p < 0.001; dislocation/instability: $16,891, p< 0.001). Local 30-day orthopaedic complication rates for fracture (20.7%) were higher those than for dislocation/instability (9.0%; p = 0.007) and similar to those for wear/loosening (17.6%; p = 0.434). Nationally, combined medical and surgical complication rates for fracture (71.3% of patients with ≥1 complication) were significantly higher than those for wear/loosening (35.2%; p < 0.001) or dislocation/instability (35.1%; p < 0.001). CONCLUSIONS: Hospitalization costs for revision THA for fracture were 33% to 48% higher than for all other aseptic revision THAs, both locally and nationally. This increased cost persisted even after multivariable comorbidity adjustment, the current DRG basis for stratifying revision THA reimbursement. Additionally, 30-day complication rates suggest that increased resource utilization for fracture patients continues even after discharge. Indication-specific coding and reimbursement systems are necessary to maintain sustainable access to revision THA for all patients. LEVEL OF EVIDENCE: Prognostic Level III. See Instructions for Authors for a complete description of levels of evidence.

Cost and Patient Outcomes Associated With Bilateral Total Knee Arthroplasty Performed by 2-Surgeon Teams vs a Single Surgeon.

BACKGROUND: Bilateral total knee arthroplasty (TKA) can be performed under a single-anesthetic (SA) or staged under a two-anesthetic (TA) technique. Recently, our institution began piloting a 2-surgeon team SA method for bilateral TKA. The purpose of this study was to compare the inpatient costs and clinical outcomes in the first 90 days after surgery between the team SA, single-surgeon SA, and single-surgeon TA approaches for bilateral TKA. METHODS: All primary TKAs performed from 2007 to 2017 by the 2 participating surgeons for each of the 3 groups of interest were identified: team SA (N = 42 patients; 84 knees), single-surgeon SA (N = 146 patients; 292 knees), single-surgeon TA (N = 242 patients; 484 knees). No patients were lost to follow-up. RESULTS: Median hospital cost (per TKA) for the episode(s) of care was as follows: team SA $20,962, single-surgeon SA $22,057, single-surgeon TA $31,145 (P < .001 overall; P = .0905 team SA vs single-surgeon SA). Rate of 90-day complications was 2.4% for team SA, 11.0% for single-surgeon SA, and 8.3% for single-surgeon TA (P = .2090). Discharge to skilled nursing facilities or rehab was as follows: team SA 31%, single-surgeon SA 53%, and single-surgeon TA after the second operation 34% (P < .001). CONCLUSION: This pilot project suggests that team SA bilateral TKA is a potentially cost-effective option with fewer complications compared to single-surgeon SA bilateral TKA. The less frequent disposition to skilled nursing facilities in the team SA group in conjunction with more efficient operating room utilization may further enhance the financial benefits.

Single-Anesthetic Versus Staged Bilateral Total Hip Arthroplasty: A Matched Cohort Study.

BACKGROUND: There is debate regarding the role of single-anesthetic versus staged bilateral total hip arthroplasty (THA) for patients with end-stage bilateral osteoarthritis. Studies have shown that single-anesthetic bilateral THA is associated with systemic complications, but there are limited data comparing patient outcomes in a matched setting of bilateral THA. METHODS: We identified 94 patients (188 hips) who underwent single-anesthetic bilateral THA. Fifty-seven percent of the patients were male. Patients had a mean age of 52.2 years and body mass index of 27.1 kg/m. They were matched 1:1 on the basis of sex, age (±1 year), and year of surgery (±3 years) to a cohort of patients undergoing staged bilateral THA. In the staged group, there was <1 year between procedures (range, 5 days to 10 months). Mean follow-up was 4 years for each group. RESULTS: Patients in the single-anesthetic group experienced shorter total operating room time and length of stay. There was no difference (hazard ratio [HR] = 0.73, p = 0.50) in the overall revision-free survival in patients undergoing single-anesthetic or staged bilateral THA. The risks of reoperation (HR = 0.69, p = 0.40), complications (HR = 0.83, p = 0.48), and mortality (HR = 0.47, p = 0.10) were similar. Single-anesthetic bilateral THA reduced the total cost of care (by 27%, p = 0.0001). CONCLUSIONS: In this matched cohort analysis, single-anesthetic bilateral THA was not associated with an increased risk of revision, reoperation, or postoperative complications, while decreasing cost. In our experience, single-anesthetic bilateral THA is a safe procedure that, for certain patients, offers an excellent means to deal with bilateral hip osteoarthritis. LEVEL OF EVIDENCE: Therapeutic Level III. See Instructions for Authors for a complete description of levels of evidence.

Ceramic Femoral Heads for All Patients? An Argument for Cost Containment in Hip Surgery.

Trunnionosis and taper corrosion have recently emerged as problems in total hip arthroplasty (THA). No longer restricted to metal-on-metal bearings, these phenomena now affect an increasing number of metal-on-polyethylene THAs and are exacerbated by modularity. Resulting increases in metal toxicity and patient morbidity, and the added costs of toxicity surveillance and revision surgery, will place a substantial economic burden on many health systems. Although they are more expensive than cobalt-chrome heads, ceramic femoral heads make metal toxicity a nonissue. In this article, we provide a theoretical framework for debating whether use of ceramic femoral heads in all THA patients could represent a more cost-effective option.

Intraoperative angiography provides objective assessment of skin perfusion in complex knee reconstruction.

BACKGROUND: Wound necrosis is a potentially devastating complication of complex knee reconstruction. Laser-assisted indocyanine green angiography (LA-ICGA) is a technology that has been described in the plastic surgery literature to provide an objective assessment of skin perfusion in the operating room. This novel technology uses a plasma protein bound dye (ICG) and a camera unit that is calibrated to view the frequency emitted by the dye. The intention of this technology is to offer real-time visualization of blood flow to skin and soft tissue in a way that might help surgeons make decisions about closure or coverage of a surgical site based on blood flow, potentially avoiding soft tissue reconstruction while preventing skin necrosis or wound breakdown after primary closures, but its efficacy is untested in the setting of complex TKA. QUESTIONS/PURPOSES: The purpose of this study was to evaluate perfusion borders and tension ischemia in a series of complex knee reconstructions to guide optimal wound management. METHODS: Beginning in mid-2011, an LA-ICGA system was used to evaluate soft tissue viability in knee reconstruction procedures that were considered high risk for wound complications. Seven patients undergoing complex primary or revision TKA from 2011 to 2013 were included. These patients were chosen as a convenience sample of knee reconstruction procedures for which we obtained consultation with the plastic surgery service. The perfusion of skin and soft tissue coverage was evaluated intraoperatively for all patients with the LA-ICGA system, and the information was used to guide wound management. Followup was at a mean of 9 months (range, 6-17 months), no patients were lost to followup, and the main study endpoint was uneventful healing of the surgical incision. RESULTS: All seven closures went on to heal without necrosis. One patient, however, was subsequently revised for a deep periprosthetic infection 4 months after their knee reconstruction and underwent flap coverage at the time of that revision. CONCLUSIONS: Implementation of LA-ICGA provides an objective intraoperative assessment of soft tissue perfusion. This technology may help guide the surgeon's decisions about wound closure in real-time to accommodate the perfusion challenges unique to each patient. Specifically, patients with medical risk factors for poor perfusion or wound healing (such as diabetes, peripheral vascular disease, tobacco use, corticosteroid therapy, infection) or anatomical/surgical risk factors (ie, previous surgery about the reconstruction site, trauma wounds, or reconstruction of severe deformity) may benefit from objective intraoperative information regarding perfusion of the wound site. Furthermore, LA-ICGA could be used to prospectively evaluate the physiologic impact of different wound closure techniques. LEVEL OF EVIDENCE: Level IV, therapeutic study. See Guidelines for Authors for a complete description of levels of evidence.