Development of Machine Learning Algorithms to Predict Patient Dissatisfaction After Primary Total Knee Arthroplasty

BackgroundPostoperative dissatisfaction after primary total knee arthroplasty (TKA) that requires additional care or readmission may impose a significant financial burden to healthcare systems. The purpose of the current study is to develop machine learning algorithms to predict dissatisfaction after TKA.MethodsA retrospective review of consecutive TKA patients between 2014 and 2016 from 1 large academic and 2 community hospitals was performed. Preoperative variables considered for prediction included demographics, medical history, flexion contracture, knee flexion, and outcome scores (patient-reported health state, Knee Society Score [KSS], and KSS-Function [KSS-F]). Recursive feature elimination was used to select features that optimized algorithm performance. Five supervised machine learning algorithms were developed by training with 10-fold cross-validation 3 times. These algorithms were subsequently applied to an independent testing set of patients and assessed by discrimination, calibration, Brier score, and decision curve analysis.ResultsOf 430 patients, a total of 40 (9.0%) were dissatisfied with their outcome after primary TKA at a minimum of 2 years postoperatively. The random forest algorithm achieved the best performance in the independent testing set not used for algorithm development (c-statistic: 0.77, calibration intercept: 0.087, calibration slope: 0.74, Brier score: 0.082). The most important factors for predicting dissatisfaction were age, number of medical comorbidities, presence of one or more drug allergies, preoperative patient-reported health state score, and preoperative KSS.ConclusionThe current study developed machine learning algorithms based on partially modifiable risk factors for predicting dissatisfaction after TKA. This model demonstrates good discriminative capacity for identifying those at greatest risk for dissatisfaction and may allow for preoperative health optimization.     

Development of a Novel,Potentially Universal Machine Learning Algorithm for Prediction of Complications After Total Hip Arthroplasty

BackgroundAs the prevalence of hip osteoarthritis increases, the number of total hip arthroplasty (THA) procedures performed is also projected to increase. Accurately risk-stratifying patients who undergo THA would be of great utility, given the significant cost and morbidity associated with developing perioperative complications. We aim to develop a novel machine learning (ML)-based ensemble algorithm for the prediction of major complications after THA, as well as compare its performance against standard benchmark ML methods.MethodsThis is a retrospective cohort study of 89,986 adults who underwent primary THA at any California-licensed hospital between 2015 and 2017. The primary outcome was major complications (eg infection, venous thromboembolism, cardiac complication, pulmonary complication). We developed a model predicting complication risk using AutoPrognosis, an automated ML framework that configures the optimally performing ensemble of ML-based prognostic models. We compared our model with logistic regression and standard benchmark ML models, assessing discrimination and calibration.ResultsThere were 545 patients who had major complications (0.61%). Our novel algorithm was well-calibrated and improved risk prediction compared to logistic regression, as well as outperformed the other four standard benchmark ML algorithms. The variables most important for AutoPrognosis (eg malnutrition, dementia, cancer) differ from those that are most important for logistic regression (eg chronic atherosclerosis, renal failure, chronic obstructive pulmonary disease).ConclusionWe report a novel ensemble ML algorithm for the prediction of major complications after THA. It demonstrates superior risk prediction compared to logistic regression and other standard ML benchmark algorithms. By providing accurate prognostic information, this algorithm may facilitate more informed preoperative shared decision-making.     

Are Functional Outcomes of a Total Hip Arthroplasty Predictive of a Contralateral Total Hip Arthroplasty: A Retrospective Cohort Study

BackgroundCurrent literature suggests that 8%-35% of patients undergoing total hip arthroplasty (THA) undergo a subsequent contralateral THA. This study aims to determine if functional outcomes after primary THA predict outcomes in the subsequent primary THA of the contralateral side.MethodsA retrospective cohort of patients undergoing staged bilateral primary THA was reviewed. The Oxford Hip Score (OHS) was utilized as the functional outcome measurement tool and was assessed preoperatively and at one year postoperatively. The minimal clinically important difference (MCID) was assessed. Based on the first-side THA one-year outcomes, the odds of maintaining an MCID, or not, for the second-side THA were determined.ResultsThe study cohort consisted of 551 patients and 1102 primary THAs. The average postoperative OHSs were similar after the first and second THA. Patients achieving the MCID with the first-side surgery were 2.6 times (95% confidence interval 1.0 to 6.64, P = .04) more likely to achieve the MCID for the second-side surgery than patients failing to reach the MCID for their first-side surgery. After the first THA, 29 (5.3%) patients failed to reach the predefined MCID for the OHS compared with 54 (9.8%) patients undergoing their second THA (odds ratio: 1.96 [95% confidence interval: 1.23 to 3.1], χ² = 8.14, P = .005).ConclusionsFunctional outcomes after the first THA are predictive of functional outcomes of the second THA. Patients are more likely to achieve a clinically significant improvement after their first THA related to higher preoperative OHSs before the second THA.     

Relationships Between Preoperative Mental Health and Improvements in Patient-Reported Outcomes After Total Hip and Knee Arthroplasty

BackgroundPoor preoperative mental health has been associated with worse outcomes after total hip (THA) and total knee arthroplasty (TKA). To fully understand these relationships, we assessed post-THA and post-TKA improvements in patient-reported mental and joint health by preoperative mental health groups.MethodsElective cases (367 THA, 462 TKA) were subgrouped by low (<25th percentile), middle (25th-74th), and high (≥75th) preoperative mental health, using Veterans RAND 12-Item Health Survey Mental Component Summary (MCS) scores. In each subgroup, we assessed the relationship between preoperative MCS and 1-year postoperative change in mental and joint health. Pairwise comparisons and multivariable regression models were applied for THA and TKA separately.ResultsMedian postoperative mental health change was +14.0 points for the low-MCS THA group, +11.1 low-TKA, +2.0 middle-THA and TKA, −4.0 high-THA, and −4.9 high-TKA (between-group differences P < .001). All MCS groups had improved median joint health scores, without significant between-group differences. Preoperative mental health was negatively associated with mental health improvements in all groups (B = −0.94 - −0.68, P < .001-P = .01) but with improvements in joint health only in the low-THA group (B = −0.74, P = .02). Improvements in mental and joint health were positively associated for low and middle (B = 0.61-0.87, P < .001), but not for high-MCS groups, with this relationship differing for the low versus high group.ConclusionPatients who have low preoperative mental health experienced greater postoperative mental health improvement and similar joint health improvement compared to patients who have high preoperative mental health. Findings can guide subgroup-targeted surgical decision-making and preoperative counseling.     

Factors Influencing the Improvement of Gait Ability After Total Hip Arthroplasty

We aimed to investigate the factors influencing gait improvement in the patients who had undergone total hip arthroplasty. We performed gait analysis on 43 female patients with unilateral hip osteoarthritis. All the patients were analyzed before and at 2, 6, and 12 months after the surgery. There were significant reductions in spatiotemporal parameters in the patients with hip osteoarthritis compared with the control group. The mean values of the spatiotemporal parameters of the patients showed considerable improvement by 12 months after surgery; however, they did not reach the same values as those observed in the healthy subjects. The stage of osteoarthritis and the changes in the leg-length discrepancies were the factors that most influenced gait improvement after total hip arthroplasty throughout the follow-up period.     

Patient-Reported Outcomes After Revision of Metal-on-Metal Total Bearings in Total Hip Arthroplasty

Background

Failure of metal-on-metal (MOM) total hip arthroplasty (THA) bearings is often accompanied by an aggressive local reaction associated with destruction of bone, muscle, and other soft tissues around the hip. Little is known about whether patient-reported physical and mental function following revision THA in MOM patients is compromised by this soft tissue damage, and whether revision of MOM THA is comparable with revision of hard-on-soft bearings such as metal-on-polyethylene (MOP).

Preoperative Factors and Patient-Reported Outcomes After Total Hip Arthroplasty: Multivariable Prediction Modeling

BackgroundAlthough total hip arthroplasty (THA) has been established as a cost-effective intervention, some patients experience ongoing pain and dissatisfaction. There is interest in predicting postoperative outcomes based on preoperative data, but the relative contribution of different preoperative factors is unclear. The study’s aim was to develop multivariable prediction models for the assessment of patient-reported outcomes.MethodRegistry data on 1412 patients undergoing THA for osteoarthritis at two hospitals between 2013 and 2018 was used. Potential predictors included age, sex, body mass index, spoken language, education level, previous THA, American Society of Anaesthesiologists (ASA) score, lower back pain, depression/anxiety, other lower limb arthritis, number of other comorbidities, the preoperative expectation of pain and function, EuroQol Visual Analogue Scale (EQ-VAS) and preoperative OHS. Radiographic scores were also used: joint space narrowing (JSN), osteophytes, sclerosis, and an overall grade based on the Kellgren-Lawrence (KL) classification. Outcomes assessed were the patient-rated improvement, satisfaction, and OHS at six months.ResultsJSN or overall KL scores were the most important predictors (P < .001) for all outcomes, with better radiographic scores associated with worse outcomes. Other predictors associated with poorer outcomes were lower back pain and lower expectation (predicting poor improvement), lower education and higher ASA (predicting lower satisfaction) and younger age, female sex, non-English speakers, lower preoperative EQ-VAS, lower education, back pain, and anxiety/depression (predicting lower OHS).ConclusionPreoperative radiological scores are an important predictor of patient-reported outcomes at six months postoperatively. Understanding the relative strengths and significance of different factors in predicting outcomes will help the clinician and patient decision-making for THA.     

Defining Clinically Meaningful Thresholds for Patient-Reported Outcomes in Knee Arthroplasty

BackgroundFor primary knee arthroplasties, clinically meaningful thresholds of patient-reported outcomes that associate with patient satisfaction have not been defined appropriately.MethodsIn this retrospective study of 26,720 primary total knee replacements registered in the Dutch Arthroplasty Register (2016-2018), receiver operating curve analysis was used to define minimal clinically important changes (MCICs) and patient acceptable symptom states (PASSs) with the anchor satisfaction. Patient-reported outcome measures were pain, European Quality of Life 5 Dimensions, Knee disability and Osteoarthritis Outcome Score, and Oxford Knee Score (OKS). Independent analyses were performed for groups, which showed statistically significant interactions with the (change in) score to achieve satisfaction in logistic regression.ResultsIn this cohort, 84.9% completed the anchor questions, of whom 71.3% with a satisfaction score ≥8. Good discriminative abilities (area under the receiving operator curve >0.8) for PASS were achieved by OKS ≥38.5, pain in activity ≤2.5, Knee disability and Osteoarthritis Outcome Score ≤33, and Quality of Life-Index ≥0.813. Discriminative abilities for MCIC were not good. If assessed per baseline tercile, discrimination improved (area under the receiving operator curve >0.8) and yielded different MCICs per preoperative tercile (preoperative OKS ≤19: MCIC ≥19.5; pre-OKS 20-27: MCIC ≥14.5; pre-OKS ≥28: MCIC ≥8.5). For MCIC, the tercile method produced an 11% improved accuracy compared to one threshold for every patient. For the PASS scores, tercile-specific did not improve the accuracy of predicting satisfaction. Demographics were not clinically relevant in determining thresholds.ConclusionEstimating the likelihood of satisfaction with surgery is critical in shared decision-making. Patients with more preoperative symptom severity require larger changes to report satisfaction. Both in the clinic and in science, such differences must be considered when predictions of satisfaction are attempted.     

Metrics of Clinically Important Changes in Total Hip Arthroplasty: A Systematic Review

BackgroundAlthough patient-reported outcome measures (PROMs) have become a regularly used metric, there is little consensus on the methodology used to determine clinically relevant postoperative outcomes. We systematically reviewed the literature for studies that have identified metrics of clinical efficacy after total hip arthroplasty (THA) including minimal clinically important difference (MCID), patient acceptable symptom state (PASS), minimal detectable change (MDC), and substantial clinical benefit (SCB).MethodsA systematic review examining quantitative metrics for assessing clinical improvement with PROMs following THA was conducted according to Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines using the MEDLINE database from 2008 to 2020. Inclusion criteria included full texts, English language, primary THA with minimum 1-year follow-up, use of metrics for assessing clinical outcomes with PROMs, and primary derivations of those metrics. Sixteen studies (24,487 THA patients) met inclusion criteria and 11 different PROMs were reported.ResultsMCIDs were calculated using distribution methods in 7 studies (44%), anchor methods in 2 studies (13%), and both methods in 2 studies (13%). MDC was calculated in 2 studies, PASS was reported in 1 study using anchor-based method, and SCB was calculated in 1 study using anchor-based method.ConclusionThere is a lack of consistency in the literature regarding the use and interpretation of PROMs to assess patient satisfaction. MCID was the most frequently reported measure, while MDC, SCB, and PASS were used relatively infrequently. Method of derivation varied based on the PROM used; distribution method was more frequently used for MCID.     

Social Determinants of Health and Patient-Reported Outcomes Following Total Hip and Knee Arthroplasty in Veterans

BackgroundSocial determinants of health (SDOH) are the conditions in which people are born, grow, live, work, and age. They are associated with disparities in outcomes following total joint arthroplasty (TJA). These disparities occur even in equal-access healthcare systems such as the Veterans Health Administration (VHA). Our goal was to determine whether SDOH affect patient-reported outcome measures (PROMs) following TJA in VHA patients.MethodsPatients scheduled to undergo total hip or knee arthroplasty at VHA Hospitals in Minneapolis, MN, Palo Alto, CA, and San Francisco, CA, prospectively completed PROMs before and 1 year after surgery. PROMs included the Hip disability and Osteoarthritis Outcome Score, the Knee injury and Osteoarthritis Outcome Score, and their Joint Replacement subscores. SDOH included race, ethnicity, marital status, education, and employment status. The level of poverty in each patient’s neighborhood was determined. Medical comorbidities were recorded. Univariate and multivariate analyses were performed to determine whether SDOH were significantly associated with PROM improvement after surgery.ResultsOn multivariate analysis, black race was significantly negatively correlated with knee PROM improvement and Hispanic ethnicity was significantly negatively correlated with hip PROM improvement compared to whites. Higher baseline PROM scores and lower age were significantly associated with lower PROM improvement. Significant associations were also found based on education, gender, comorbidities, and neighborhood poverty.ConclusionMinority VHA patients have lower improvement in PROM scores after TJA than white patients. Further research is required to identify the reasons for these disparities and to design interventions to reduce them.     

Higher Prevalence of Mixed or Solid Pseudotumors in Metal-on-Polyethylene Total Hip Arthroplasty Compared With Metal-on-Metal Total Hip Arthroplasty and Resurfacing Hip Arthroplasty

Background

Pseudotumors are a common finding in metal-on-metal (MoM) total hip arthroplasty (THA) and resurfacing hip arthroplasty (RHA). However, information on pseudotumors in metal-on-polyethylene (MoP) THA is limited.

Midterm Outcomes of Revision Total Hip Arthroplasty With the Use of a Multihole Highly-Porous Titanium Shell

Background

We are reporting on the minimum 5-year outcomes of patients who underwent revision total hip arthroplasty (THA) using a specific highly-porous titanium shell. We assessed (1) aseptic and all-cause survivorship; (2) functional outcomes; (3) complications; and (4) radiographic outcomes.

Influencing Factors for Joint Perception After Total Hip Arthroplasty: Asian Cohort Study

BackgroundPerceiving replaced joints as natural is one of the best scenario after total hip arthroplasty (THA). We investigated the distribution of and influencing factors for patient’s joint perception after THA in Asian cohort, which is not well known.MethodsWe mailed a questionnaire to Asian patients who had undergone THA in our institution between 2012 and 2016, and this study included 318 Asian patients. The questionnaire assessed patient’s joint perception, Oxford Hip Score (OHS), Short Form-12 Health Survey (SF-12) physical, mental, and role component summaries, and satisfaction score. Leg length discrepancy and global femoral offset before and after THA were measured using radiographs. The patients were divided into 2 groups with patient’s joint perception: natural and artificial perception groups. OHS, SF-12, satisfaction, leg length discrepancy, and global femoral offset were compared between natural and artificial perception groups, and which factors significantly influenced joint perception were determined.ResultsOf the 318 patients, 165 patients (51.8%) perceived their replaced joint as a natural joint. OHS, SF-12 physical and role component summaries, and satisfaction score in natural perception group were significantly higher than those in artificial perception group, without significant difference in SF-12 mental component summary. Multivariate analysis showed that less of usual pain, easier to get in or out of a car, and osteoarthritis in the contralateral hip were positively associated with natural joint perception after THA.ConclusionThis study indicated to surgeons that pain relief and improvement in getting in or out of a car after THA could lead to even better patient’s joint perception.     

Predicting Dissatisfaction After Total Hip Arthroplasty: A Study of 850 Patients

We conducted a prospective cohort study investigating the rate and predictors of dissatisfaction among patients after primary total hip arthroplasty (THA). Eight hundred fifty patients were assessed preoperatively and 1 year postoperatively using Patient Reported Outcome Measures. There was a 7% rate of dissatisfaction after THA. After univariate analysis, depression, preoperative Short Form 12 mental component score, and symptomatic arthritis of another major joint predicted dissatisfaction at 1 year, but after multivariate analysis, only symptomatic arthritis in another major joint was significant. The development of a major complication did not predict dissatisfaction. Satisfaction also correlates strongly with postoperative functional scores, relief of pain, restoration of function, and success in meeting patient expectations. Pain relief and expectation management are critical in maximizing patient satisfaction after THA.     

Hospital Consumer Assessment of Healthcare Providers and Systems Scores Do Not Predict Outcomes After Total Hip Arthroplasty

Background

With the establishment of the Hospital Value-Based Purchasing program, the Hospital Consumer Assessment of Healthcare Providers and Systems (HCAHPS) score has been incorporated into the calculation of the total performance score, which determines redistribution of up to 2% of Medicare payments. This study aims to assess whether the HCAHPS score correlates with validated outcome measures after total hip arthroplasty.

Prevention of Dislocation After Total Hip Arthroplasty

Background

Prevention of dislocation after primary total hip arthroplasty (THA) begins with patient preoperative assessment and planning.

Objective Activity Levels and Patient-Reported Outcomes After Total Hip Arthroplasty and Periacetabular Osteotomy: Retrospective Matched Cohort Study at Mean 12-Year Follow-Up

BackgroundThere is increasing interest in improving activity after total hip arthroplasty (THA) and periacetabular osteotomy (PAO). The present study evaluated whether there were differences in the subjective and objective activity levels of THA and PAO patients at mean 12-year follow-up (range 4-20) and what factors influence the objective activity levels.MethodsTHA and PAO patients (30 patients each; mean age: 66 and 63 years, respectively), who had undergone surgery for osteoarthritis due to acetabular dysplasia, were included. Patients were retrospectively matched based on age, gender, body mass index, follow-up duration, and preoperative University of California, Los Angeles activity score (UCLA score). Patients were surveyed for the subjective activity levels using the Oxford Hip Score and UCLA score. Objective activity levels (the number of steps, upright time, and the number of sit-to-stand transitions) were made using an accelerometer.ResultsThere was no significant difference in subjective activity level between THA and PAO patients. The number of steps was significantly higher in PAO than in THA patients. Multivariable analyses revealed that THA and low Oxford Hip Score activities of daily living were negatively associated with the number of steps, that men and high UCLA score were negatively associated with upright time, and that high body mass index was negatively associated with sit-to-stand transitions in THA and PAO patients.ConclusionThis study suggested that although there was no difference in postoperative subjective activity level between THA and PAO patients, there was a difference in objective activity level. Providing guidance to these patients based on objective activity may lead to higher activity levels.     

A Matched Comparison of the Rates of Achieving the Minimal Clinically Important Difference Following Conversion and Primary Total Hip Arthroplasty

BackgroundPatient-reported outcome measures (PROMs) are often lower following conversion total hip arthroplasty (cTHA) compared to matched primary total hip arthroplasty (THA) controls. However, the minimal clinically important differences (MCIDs) for any PROMs are yet to be analyzed for cTHA. This study aimed to (1) determine if patients undergoing cTHA achieve primary THA-specific 1-year PROM MCIDs at comparable rates to matched controls undergoing primary THA and (2) establish 1-year MCID values for specific PROMs following cTHA.MethodsA retrospective case-control study was conducted using 148 cases of cTHA which were matched 1:2 to 296 primary THA patients. Previously defined anchor values for 2 PROM measures in primary THA were used to compare cTHA to primary THA, while novel cTHA-specific MCID values for 2 PROMs were calculated through a distribution method. Predictors of achieving the MCID of PROMs were analyzed through multivariate logistic regressions.ResultsConversion THA was associated with decreased odds of achieving the primary THA-specific 1-year Hip dysfunction and Osteoarthritis Outcome Score for Joint Replacement PROM (Odds Ratio: 0.319, 95% Confidence Interval: 0.182-0.560, P < .001) and Patient Reported Outcomes Measurement Information System Physical Function Short-Form-10a PROM (Odds Ratio: 0.531, 95% Confidence Interval: 0.313-0.900, P = .019) MCIDs in reference to matched primary THA patients. Less than 60% of cTHA patients achieved an MCID. The 1-year MCID of the Hip dysfunction and Osteoarthritis Outcome Score for Joint Replacement and Patient Reported Outcomes Measurement Information System Physical Function Short-Form-10a specific to cTHA were +10.71 and +4.68, respectively.ConclusionWhile cTHA is within the same diagnosis-related group as primary THA, patients undergoing cTHA have decreased odds of achieving 1-year MCIDs of primary THA-specific PROMs.Level of EvidenceLevel III, retrospective comparative study.     

Squatting After Total Hip Arthroplasty: Patient-Reported Outcomes and In Vivo Three-Dimensional Kinematic Study

BackgroundSquatting is an important function for many daily activities, but has not been well documented after total hip arthroplasty (THA). This study investigated the participation rate of squatting and in vivo kinematics during squatting.MethodsA survey questionnaire about squatting was mailed to patients who underwent primary THA and 328 patients returned acceptable responses. Additionally, 32 hips were evaluated for dynamic 3-dimensional kinematics of squatting using density-based image-matching techniques. Multivariate analyses were applied to determine which factors were associated with anterior liner-to-neck distance at maximum hip flexion.ResultsPatients who could easily squat significantly increased this ability postoperatively (23.5% vs 46%, P < .01). In 29.5% of the patients there was still no ability to squat after THA; the main reason was anxiety of dislocation (34.2%). Kinematic analysis revealed that maximum hip flexion averaged 80.7° ± 12.3° with 12.8° ± 10.7° of posterior pelvic tilt and 9.7 ± 3.0 mm of anterior liner-to-neck distance. Neither liner-to-neck, bone-to-bone, nor bone-to-implant contact was observed in any of the hips. Larger hip flexion and smaller cup anteversion were negatively associated with the anterior liner-to-neck distance at maximum hip flexion (P < .05).ConclusionPostoperatively, approximately 70% of patients squatted easily or with support. Anxiety of dislocation made patients avoid squatting after THA. In vivo squatting kinematics suggest no danger of impingement or subsequent dislocation, but excessively large hip flexion and small cup anteversion remain as risks.