Faculty Bibliography

BACKGROUND:Supervised machine learning techniques have been increasingly applied to predict patient outcomes after hip and knee arthroplasty procedures. The purpose of this study was to systematically review the applications of supervised machine learning techniques to predict patient outcomes after primary total hip and knee arthroplasty. METHODS:A comprehensive literature search using the electronic databases MEDLINE, EMBASE, Cochrane Central Register of Controlled Trials, and Cochrane Database of Systematic Reviews was conducted in July of 2021. The inclusion criteria were studies that utilized supervised machine learning techniques to predict patient outcomes after primary total hip or knee arthroplasty. RESULTS:Search criteria yielded n = 30 relevant studies. Topics of study included patient complications (n = 6), readmissions (n = 1), revision (n = 2), patient-reported outcome measures (n = 4), patient satisfaction (n = 4), inpatient status and length of stay (LOS) (n = 9), opioid usage (n = 3), and patient function (n = 1). Studies involved TKA (n = 12), THA (n = 11), or a combination (n = 7). Less than 35% of predictive outcomes had an area under the receiver operating characteristic curve (AUC) in the excellent or outstanding range. Additionally, only 9 of the studies found improvement over logistic regression, and only 9 studies were externally validated. CONCLUSION/CONCLUSIONS:Supervised machine learning algorithms are powerful tools that have been increasingly applied to predict patient outcomes after total hip and knee arthroplasty. However, these algorithms should be evaluated in the context of prognostic accuracy, comparison to traditional statistical techniques for outcome prediction, and application to populations outside the training set. While machine learning algorithms have been received with considerable interest, they should be critically assessed and validated prior to clinical adoption.

PURPOSE/OBJECTIVE:Kinematically designed total knee arthroplasty (TKA) aims to restore normal kinematics by replicating the function of both cruciate ligaments. Traditional posterior-stabilized (PS) TKA designs, on the other hand, simplify knee kinematics and may improve TKA cost-effectiveness. The purpose of this study was to compare outcomes of patients who underwent primary TKA using either a traditional PS or kinematically designed TKA. METHODS:This retrospective study examined all patients who underwent primary TKA using either a kinematically or a traditional PS designed TKA implant, with a minimum follow-up of 2 years. Patient demographics, complications, readmissions, revision rates and causes, range of motion (ROM) and patient reported outcomes (KOOS, JR) were compared between groups. Kaplan-Meier survivorship analysis was performed to estimate freedom from revision, and multivariate regression was performed to control for confounding variables. RESULTS:A total of 396 TKAs [173 (43.7%) with a kinematic design, 223 (56.3%) with a traditional design] with a mean follow-up of 3.48 ± 1.51 years underwent analysis. Revision rates did not differ between groups (9.8% vs. 6.7%, p = 0.418). In Kaplan-Meier analysis at 2-year follow-up, freedom from all-cause revision (96.4% vs. 93.1%, p = 0.139) were similar between groups. The two cohorts had no significant difference in aseptic loosening at 2 years (99.6% vs. 97.1, p = 0.050) and at latest follow up (92.7% vs. 96.4%, p = 0.279). KOOS, JR scores and post-operative ROM were similar between groups. CONCLUSION/CONCLUSIONS:This study demonstrated similar mid-term outcomes following the use of both a kinematically designed and a traditionally designed implant in primary TKA patients. LEVEL OF EVIDENCE/METHODS:Retrospective study-III.

INTRODUCTION/BACKGROUND:Despite the excellent outcomes associated with primary total hip arthroplasty (THA), implant failure and revision continue to burden the healthcare system. The use of computer-assisted navigation (CAN) offers the potential for more accurate placement of hip components during surgery. While intraoperative CAN systems have been shown to improve outcomes in primary THA, their use in the context of revision total hip arthroplasty (rTHA) has not been elucidated. We sought to investigate the validity of using CAN during rTHA. METHODS:A retrospective analysis was performed at an academic medical institution identifying all patients who underwent rTHA using CAN from 2016-2019. Patients were 1:1 matched with patients undergoing rTHA without CAN (control) based on demographic data. Cup anteversion, inclination, change in leg length discrepancy (ΔLLD) and change in femoral offset between pre- and post-operative plain weight-bearing radiographic images were measured and compared between both groups. A safety target zone of 15-25° for anteversion and 30-50° for inclination was used as a reference for precision analysis of cup position. RESULTS:Eighty-four patients were included: 42 CAN cases and 42 control cases. CAN cases displayed a lower ΔLLD (5.74 ± 7.0 mm vs 9.13 ± 7.9 mm, p = 0.04) and greater anteversion (23.4 ± 8.53° vs 19.76 ± 8.36°, p = 0.0468). There was no statistical difference between the proportion of CAN or control cases that fell within the target safe zone (40% vs 20.9%, p =  0.06). Femoral offset was similar in CAN and control cases (7.63 ± 5.84 mm vs 7.14 ± 4.8 mm, p = 0.68). CONCLUSION/CONCLUSIONS:Our findings suggest that the use of CAN may improve accuracy in cup placement compared to conventional methodology, but our numbers are underpowered to show a statistical difference. However, with a ΔLLD of ~ 3.4 mm, CAN may be useful in facilitating the successful restoration of pre-operative leg length following rTHA. Therefore, CAN may be a helpful tool for orthopedic surgeons to assist in cup placement and LLD during complex revision cases.

INTRODUCTION/BACKGROUND:The term "spinopelvic mobility" is most often applied to motion within the spinopelvic segment. It has also been used to describe changes in pelvic tilt between various functional positions, which is influenced by motion at the hip, knee, ankle and spinopelvic segment. In the interest of establishing a consistent language for spinopelvic mobility, we sought to clarify and simplify its definition to create consensus, improve communication, and increase consistency with research into the hip-spine relationship. METHODS:A literature search was performed using the Medline (PubMed) library to identify all existing articles pertaining to spinopelvic mobility. We reported on the varying definitions of spinopelvic mobility including how different radiographic imaging techniques are used to define mobility. RESULTS:The search term "spinopelvic mobility" returned a total of 72 articles. The frequency and context for the varying definitions of mobility were reported. 41 papers used standing and upright relaxed-seated radiographs without the use of extreme positioning, and 17 papers discussed the use of extreme positioning to define spinopelvic mobility. DISCUSSION/CONCLUSIONS:Our review suggests that the definitions of spinopelvic mobility is not consistent in the majority of published literature. We suggest descriptions of spinopelvic mobility independently consider spinal motion, hip motion, and pelvic position, while recognizing and describing their interdependence.

PURPOSE/OBJECTIVE:Aseptic loosening is a common cause of implant failure following total knee arthroplasty (TKA). Cement penetration depth is a known factor that determines an implant's "strength" and plays an important role in preventing aseptic loosening. Tourniquet use is thought to facilitate cement penetration, but its use has mixed reviews. The aim of this study was to compare cement penetration depth between tourniquet and tourniquet-less TKA patients. METHODS:A multicenter retrospective review was conducted. Patients were randomized preoperatively to undergo TKA with or without the use of an intraoperative tourniquet. The variables collected were cement penetration measurements in millimeters (mm) within a 1-month post-operative period, length of stay (LOS), and baseline demographics. Measurements were taken by two independent raters and made in accordance to the zones described by the Knee Society Radiographic Evaluation System and methodology used in previous studies. RESULTS:A total of 357 TKA patients were studied. No demographic differences were found between tourniquet (n = 189) and tourniquet-less (n = 168) cohorts. However, the tourniquet cohort had statistically, but not clinically, greater average cement penetration depth [2.4 ± 0.6 mm (range 1.2-4.1 mm) vs. 2.2 ± 0.5 mm (range 1.0-4.3 mm, p = 0.01)]. Moreover, the tourniquet cohort had a significantly greater proportion of patients with an average penetration depth within the accepted zone of 2 mm or greater (78.9% vs. 67.3%, p = 0.02). CONCLUSION/CONCLUSIONS:Tourniquet use does not affect average penetration depth but increases the likelihood of achieving optimal cement penetration depth. Further study is warranted to determine whether this increased likelihood of optimal cement penetration depth yields lower revision rates.

INTRODUCTION/BACKGROUND:Prior studies have demonstrated an association between time to revision total knee arthroplasty (rTKA) and indication; however, the impact of early versus late revision on post-operative outcomes has not been reported. MATERIALS AND METHODS/METHODS:A retrospective, observational study examined patients who underwent unilateral, aseptic rTKA at an academic orthopedic hospital between 6/2011 and 4/2020 with > 1-year of follow-up. Patients were early revisions if they were revised within 2 years of primary TKA (pTKA) or late revisions if revised after greater than 2 years. Patient demographics, surgical factors, and post-operative outcomes were compared. RESULTS:470 rTKA were included (199 early, 271 late). Early rTKA patients were younger by 2.5 years (p = 0.002). The predominant indications for early rTKA were instability (28.6%) and arthrofibrosis/stiffness (26.6%), and the predominant indications for late rTKA were aseptic loosening (45.8%) and instability (26.2%; p < 0.001). Late rTKA had longer operative times (119.20 ± 51.94 vs. 103.93 ± 44.66 min; p < 0.001). There were no differences in rTKA type, disposition, hospital length of stay, all-cause 90-day emergency department visits and readmissions, reoperations, and number of re-revisions. CONCLUSIONS:Aseptic rTKA performed before 2 years had different indications but demonstrated similar outcomes to those performed later. Early revisions had shorter surgical times, which could be attributed to differences in rTKA indication. LEVEL OF EVIDENCE/METHODS:III, retrospective observational analysis.

AIMS/UNASSIGNED:Revision total knee arthroplasty (rTKA) is a technically challenging and costly procedure. It is well-documented that primary TKA (pTKA) have better survivorship than rTKA; however, we were unable to identify any studies explicitly investigating previous rTKA as a risk factor for failure following rTKA. The purpose of this study is to compare the outcomes following rTKA between patients undergoing index rTKA and those who had been previously revised. METHODS/UNASSIGNED:This retrospective, observational study reviewed patients who underwent unilateral, aseptic rTKA at an academic orthopaedic speciality hospital between June 2011 and April 2020 with > one-year of follow-up. Patients were dichotomized based on whether this was their first revision procedure or not. Patient demographics, surgical factors, postoperative outcomes, and re-revision rates were compared between the groups. RESULTS/UNASSIGNED:= -0.102; p = 0.251). CONCLUSION/UNASSIGNED:Multiply revised TKA had worse outcomes, with higher rates of facility discharge, longer operative times, and greater reoperation and re-revision rates compared to index rTKA.

This study sought to determine common complications and the rates of readmission and revision in total hip arthroplasty patients younger than 65 years. Using the American Joint Replacement Registry, we conducted a retrospective review of all THAs in patients aged 18 to 65 years from 2012 to 2020. We excluded patients aged older than 65 years, revisions, oncologic etiology, conversion from prior surgery, and nonelective cases. Primary outcomes included cumulative revision rate, 90-day readmission rate, and reason for revision. The Kaplan-Meier method and univariate analysis were used. Five thousand one hundred fifty-three patients were included. The average age was 56.7 years (SD 7.8 years), 51% were female, 85% were White, and 89% had a Charlson Comorbidity Index of 0 (1 = 7%, >2 = 4%). The mean follow-up was 39.57 months. Fifty-three patients (1.0%) underwent revision. Seventy-four patients (1.4%) were readmitted within 90 days. Revision was more common in Black patients (P = 0.023). Survivorship was 99% (95% confidence interval, 98.7 to 99.3) and 99% (95% confidence interval, 98.5 to 99.3) at 5 and 8 years, respectively. Infection (21%), instability (15%), periprosthetic fracture (15%), and aseptic loosening (9%) were the most common indications for revision. Total hip arthroplasty performed in young and presumed active patients had a 99% survivorship at 8 years. A long-term follow-up is needed to evaluate survival trends in this growing population.

The use of intraoperative technology (IT), such as computer-assisted navigation (CAN) and robot-assisted surgery (RA), in total knee arthroplasty (TKA) is increasingly popular due to its ability to enhance surgical precision and reduce radiographic outliers. There is disputing evidence as to whether IT leads to better clinical outcomes and reduced postoperative pain. The purpose of this study was to determine if use of CAN or RA in TKA improves pain outcomes. This is a retrospective review of a multicenter randomized control trial of 327 primary TKAs. Demographics, surgical time, IT use (CAN/RA), length of stay (LOS), and opioid consumption (in morphine milligram equivalents) were collected. Analysis was done by comparing IT (n = 110) to a conventional TKA cohort (n = 217). When accounting for demographic differences and the use of a tourniquet, the IT cohort had shorter surgical time (88.77 ± 18.57 vs. 98.12 ± 22.53 minutes; p = 0.005). While postoperative day 1 pain scores were similar (p = 0.316), the IT cohort has less opioid consumption at 2 weeks (p = 0.006) and 1 month (p = 0.005) postoperatively, but not at 3 months (p = 0.058). When comparing different types of IT, CAN, and RA, we found that they had similar surgical times (p = 0.610) and pain scores (p = 0.813). Both cohorts had similar opioid consumption at 2 weeks (p = 0.092), 1 month (p = 0.058), and 3 months (p = 0.064) postoperatively. The use of IT in TKA does not yield a clinically significant reduction in pain outcomes. There was also no difference in pain or perioperative outcomes between CAN and RA technology used in TKA.

AIM/OBJECTIVE:Posterior stabilized (PS) and varus valgus constrained (VVC) knee polyethylene liners have been shown to confer excellent long-term functional results following revision total knee arthroplasty (rTKA). The purpose of this study was to compare outcomes of patients who underwent rTKA using either a PS or VVC liner. METHODS:A retrospective comparative study of 314 rTKA with either PS or VVC liner and a minimum follow-up time of two years was conducted. Patient demographics, complications, readmissions, and re-revision etiology and rates were compared between groups. Kaplan-Meier survivorship analysis was performed to estimate freedom from all-cause revision. RESULTS:Hospital LOS (3.41 ± 2.49 vs. 3.34 ± 1.93 days, p = 0.793) and discharge disposition (p = 0.418) did not significantly differ between groups. At a mean follow-up of 3.55 ± 1.60 years, the proportion of patients undergoing re-revision did not significantly differ (19.1% vs. 18.7%, p = 0.929). In subgroup analysis of re-revision causes, the VVC cohort had superior survival from re-revision due to instability compared to the PS cohort (97.8% vs. 89.4%, p = 0.003). Freedom from re-revision due to aseptic loosening did not significantly differ between groups (85.2% vs. 78.8%, p = 0.436). Improvements in range of motion (ROM) from preoperative to latest follow-up were similar as well. CONCLUSIONS:PS and VVC liners confer similar survivorship, complication rates, and overall knee ROM in rTKA. VVC liners were not associated with increased postoperative aseptic loosening and demonstrated superior freedom from re-revision due to instability. Future studies with longer follow-up are warranted to better determine significant differences in clinical outcomes between the two bearing options. LEVEL III EVIDENCE/METHODS:Retrospective Cohort Study.