Predictive Models for Clinical Outcomes in Total Knee Arthroplasty: A Systematic Analysis

Feasibility of prehabilitation for patients awaiting total knee arthroplasty; a pilot study

Objective

To examine the feasibility of conducting a preoperative home-based prehabilitation program for total knee arthroplasty patients at risk for delayed in-hospital recovery, and to explore the pre- and postoperative impact of this program.

Design

A retrospective cohort study with matched controls, enabling subgroup analyses.

Setting

Home-based.

Subjects

Patients awaiting primary unilateral total knee arthroplasty between 2019 and 2020, were compared with matched historic cases from 2016 to 2017. Matching criteria were scoring ≤17 points on the De Morton Mobility Index and >12.5 s on the timed-up-and-go test.

Intervention

Supervised home-based prehabilitation program versus no prehabilitation.

Outcomes

Feasibility, determined by recruitment rate, adherence, and safety of the program. Preoperative impact, assessed for the intervention group by differences in mean values for aerobic capacity, muscle strength and functional mobility between the first and last sessions. Postoperative impact was defined as the time needed to achieve in-hospital independence of physical function and was measured by the differences in mean values between the intervention and control groups.

Results

Recruitment rate was 71 %; adherence and drop-out rates were 88 % and 12 % respectively. No adverse events were reported. Preoperatively, the intervention group showed significant improvements in aerobic capacity on the 2-min walking test (84.29 m-98.06 m; p = 0.007) and 2-min step test (40.35 steps to 52.95 steps; p = 0.014), muscle strength on the 30 s chair stand test (7.3 stands to 10.1 stands; p = 0.002), and functional mobility as seen in the timed-up-and-go-test (19.52 s-15.85 s; p = 0.031). Postoperatively, the intervention group achieved in-hospital independence of physical function earlier (mean rank 16.11) than the control group (mean rank 24.89; p=<0.01).

Conclusions

It is feasible to conduct a prehabilitation program 4-6 weeks preoperatively, for high-risk patients awaiting total knee arthroplasty. Additionally, the program appears to have a positive impact on physical fitness both pre- and postoperatively.

Strategies to Manage Poorer Outcomes After Hip or Knee Arthroplasty: A Narrative Review of Current Understanding, Unanswered Questions, and Future Directions

PURPOSE

Although hip or knee arthroplasty is generally a successful intervention, it is documented that 15%-30% of patients undergoing arthroplasty report suboptimal outcomes. This narrative review aims to provide an overview of the key findings concerning the management of poorer outcomes after hip or knee arthroplasty.

METHOD

A comprehensive search of articles was conducted up to November 2023 across three electronic databases. Only studies written in English were included, with no limitations applied regarding study design and time.

RESULT

Efficiently addressing poorer outcomes after arthroplasty necessitates a thorough exploration of appropriate methods for assessing recovery following hip or knee arthroplasty, ensuring accurate identification of patients at risk or experiencing poorer recovery. When selecting appropriate outcome measure tools, various factors should be taken into consideration, including understanding patients' priorities throughout the recovery process, assessing psychometric properties of outcome measure tools at different time points after arthroplasty, understanding how to combine/reconcile provider-assessed and patient-reported outcome measures, and determining the appropriate methods to interpret outcome measure scores. However, further research in these areas is warranted. In addition, the identification of key modifiable factors affecting outcomes and the development of interventions to manage these factors are needed.

CONCLUSION

There is growing attention paid to delivering interventions for patients at risk or not optimally recovering following hip or knee arthroplasty. To achieve this, it is essential to identify the most appropriate outcome measure tools, factors associated with poorer recovery and management of these factors.

The Role of Risk Tolerance in a Patient's Decision to Undergo Total Knee and Hip Arthroplasty

BACKGROUND

A patient's decision-making process to undergo surgery is crucial for surgeons to understand for patient-counseling purposes. Total knee and hip arthroplasty, like any other major surgery, is associated with serious, sometimes life-threatening, complications. Using the results of discrete choice experiments (DCEs), we aimed to understand the relationship between a patient's risk tolerance and choosing to undergo surgery in real life.

METHODS

This is a retrospective study of prospectively collected DCE results for 142 potential knee or hip arthroplasty clinic patients from October 2021 to March 2022. The DCE presented the patient with 2 scenarios, each of which was made up of different combinations of attributes and levels. A hierarchal Bayesian model was used to obtain a risk score that reflected the risk attributes chosen by each patient. Logistic regressions were then used to evaluate the association between a patient's willingness to incur risk and their decision to undergo a total joint arthroplasty.

RESULTS

Of the 142 patients enrolled in the DCE, 89 (62.3%) underwent a total joint arthroplasty. Risk score (odds ratio [OR] = 2.6, 95% confidence interval [CI] 1.1 to 6.6, P = 0.04), men (OR = 2.5, 95% CI 1.1 to 5.9, P = 0.028), and patients who have hip osteoarthritis (OR = 2.4, 95% CI 1.1 to 5.5, P = 0.036) increased the odds of undergoing arthroplasty, whereas physical function of at least 75% at the initial visit (OR = 0.3, 95% CI 0.1 to 0.7, P = 0.004) decreased these odds.

CONCLUSIONS

We found that a patient's willingness to incur risk, lower baseline physical function, and men were all independently associated with undergoing total knee arthroplasty. We believe that these findings prompt much-needed future studies that focus solely on the relationship between patients' inherent risk behavior and surgical and patient-reported outcomes.

Lateral Subvastus Lateralis versus Medial Parapatellar Approach for Total Knee Arthroplasty: Patient Outcomes and Kinematics Analysis

The conventional approach for total knee arthroplasty (TKA) is a medial parapatellar approach (MPA). We aimed to study patient outcomes and kinematics with a quadriceps sparing lateral subvastus lateralis approach (SLA). Patients with neutral/varus alignment undergoing primary TKA were consented to undergo the SLA. At 1-year postoperative, patients underwent radiostereometric analysis. Patients were administered the Short Form 12 (SF-12), Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC), and Knee Society Score (KSS). Kinematics and outcome data were compared to a group undergoing TKA via conventional MPA. Fourteen patients underwent TKA via SLA with a mean age 71.5 ± 8.0 and mean body mass index (BMI) 31.0 ± 4.5. The MPA group had 13 patients with mean age 63.4 ± 5.5 (p = 0.006) and mean BMI 31.2 ± 4.6 (p = 0.95). The SLA resulted in a significantly more posterior medial contact point at 0 (p = 0.011), 20 (p = 0.020), and 40 (p = 0.039) degrees of flexion. There was no significant difference in medial contact point from 60 to 120 degrees, lateral contact point at any degree of flexion, or axial rotation. There was no difference in improvement in postoperative WOMAC, SF-12, KSS function, and total KSS knee scores between groups. The MPA group had a significantly greater improvement in KSS knee scores at 3 months (p < 0.001), 1 year (p = 0.003), and 2 years (p = 0.017). The SLA resulted in increased medial femoral rollback early in flexion. Although both approaches resulted in improved postoperative outcomes, the MPA group showed significantly greater improvements in KSS knee scores at 3 months, 1 year, and 2 years. Further studies are required to identify any benefits that the SLA may offer. LEVEL OF EVIDENCE:  Therapeutic Level II.

Surgical factors play a critical role in predicting functional outcomes using machine learning in robotic-assisted total knee arthroplasty

PURPOSE

Predictive models help determine predictive factors necessary to improve functional outcomes after total knee arthroplasty (TKA). However, no study has assessed predictive models for functional outcomes after TKA based on the new concepts of personalised surgery and new technologies. This study aimed to develop and evaluate predictive modelling approaches to predict the achievement of minimal clinically important difference (MCID) in patient-reported outcome measures (PROMs) 1 year after TKA.

METHODS

Four hundred thirty robotic-assisted TKAs were analysed in this retrospective study. The mean age was 67.9 ± 7.9 years; the mean body mass index (BMI) was 32.0 ± 6.8 kg/m2. The following PROMs were collected preoperatively and 1-year postoperatively: knee injury and osteoarthritis outcome score for joint replacement, Western Ontario and McMaster Universities osteoarthritis index (WOMAC) Function, WOMAC Pain. Demographic data, preoperative CT scan, implant size, implant position on the robotic system and characteristics of the joint replacement procedure were selected as predictive variables. Four machine learning algorithms were trained to predict the MCID status at 1-year post-TKA for each PROM survey. 'No MCID' was chosen as the target. Models were evaluated by class discrimination (F1-score) and area under the receiver operating characteristic curve (ROC-AUC).

RESULTS

The best-performing model was ridge logistic regression for WOMAC Function (area under the curve [AUC]  = 0.80, F1 = 0.48, sensitivity = 0.79, specificity = 0.62). Variables most strongly contributing to not achieving MCID status were preoperative PROMs, high BMI and femoral resection depth (posterior and distal), supporting functional positioning principles. Conversely, variables contributing to a positive outcome (achieving MCID) were medial/lateral alignment of the tibial component, whether the procedure was an outpatient surgery and whether the patient received managed Medicare insurance.

CONCLUSION

The most predictive variables included preoperative PROMs, BMI and surgical planning. The surgical predictive variables were valgus femoral alignment and femoral rotation, reflecting the benefits of personalised surgery. Including surgical variables in predictive models for functional outcomes after TKA should guide clinical and surgical decision-making for every patient.

LEVEL OF EVIDENCE

Level III.

Efficiency assessment of follow-up methodology of patients with knee replacement to predict post-surgical functionality: a protocol for randomised control PROKnee trial

INTRODUCTION

Even when total knee arthroplasty (TKA) is an extended treatment, most patients experience a suboptimal evolution after TKA. The objectives of this study are the following: (1) to determine the effectiveness of two different prosthesis stabilisation systems on the functionality in activities of daily life, and (2) to determine prognostic biomarkers of knee prosthesis function based on radiological information, quantification of cytokines, intra-articular markers and biomechanical functional evaluation to predict successful evolution.

METHODS AND ANALYSIS

The PROKnee trial was designed as a randomised controlled patient-blinded trial with two parallel groups that are currently ongoing. The initial recruitment will be 99 patients scheduled for their first TKA, without previous prosthesis interventions in lower limbs, who will be randomly divided into two groups that differed in the stabilisation methodology incorporated in the knee prosthesis: the MEDIAL-pivot group and the CENTRAL-pivot group. The maximum walking speed will be reported as the primary outcome, and the secondary results will be patient-reported questionnaires related to physical status, cognitive and mental state, radiological test, laboratory analysis and biomechanical instrumented functional performance, such as the 6-minute walking test, timed up-and-go test, gait, sit-to-stand, step-over, and ability to step up and down stairs. All the results will be measured 1 week before TKA and at 1.5, 3, 6 and 12 months after surgery.

ETHICS AND DISSEMINATION

All procedures were approved by the Ethical Committee for Research with Medicines of the University Clinical Hospital of Valencia on 8 October 2020 (order no. 2020/181). Participants are required to provide informed consent for the study and for the surgical procedure. All the data collected will be treated confidentially since they will be blinded and encrypted. The results from the trial will be published in international peer-reviewed scientific journals, regardless of whether these results are negative or inconclusive.

TRIAL REGISTRATION NUMBER

ClinicalTrials.gov Registry (NCT04850300).

Medium-term patient's satisfaction after primary total knee arthroplasty: enhancing prediction for improved care

BACKGROUND

Patient-reported satisfaction after total knee arthroplasty (TKA) is low compared to other orthopedic procedures. Although several factors have been reported to influence TKA outcomes, it is still challenging to identify patients who will experience dissatisfaction five years after surgery, thereby improving their management. Indeed, both perioperative information and follow-up questionnaires seem to lack statistical predictive power.

HYPOTHESIS

This study aims to demonstrate that machine learning can improve the prediction of patient satisfaction, especially when classical statistics fail to identify complex patterns that lead to dissatisfaction.

PATIENTS AND METHODS

Patients who underwent primary TKA were included in a Registry that collected baseline data and clinical outcomes at different follow-ups. The patients were divided into satisfied and dissatisfied groups based on a satisfaction questionnaire administered five years after surgery. Satisfaction was predicted using linear statistical models compared to machine learning algorithms.

RESULTS

A total of 147 subjects were analyzed. Regarding statistics, significant differences between satisfaction levels started emerging only six months after the intervention, and the classification was close to random guessing. However, machine learning algorithms could improve the prediction by 72% soon after the intervention, and an improvement of 178% was possible when including follow-ups up to one year.

DISCUSSION

This study demonstrates the feasibility of a registry-based approach for monitoring and predicting satisfaction using ML algorithms.

LEVEL OF EVIDENCE

III.

Does preoperative opioid use predict outcomes to 6 months following primary unilateral knee or hip arthroplasty for osteoarthritis? A data-linked retrospective study

BACKGROUND

Few Australian studies have examined the incidence of prescribed opioid use prior to primary total knee or total hip arthroplasty (TKA, THA) and whether it predicts post-surgery outcomes. A recent Australian study demonstrated that the prevalence of pre-arthroplasty opioid use was approximately 16%. In the United States, approximately 24% of people undergoing TKA or THA are chronic opioid users preoperatively.

PURPOSE

This study aimed to determine (i) the proportion of TKA and THA patients who use prescribed opioids regularly (daily) before surgery (i.e., opioid use reported between the time of waitlisting and any time up to 3 months before surgery), (ii) if opioid use before surgery predicts (a) complication/readmission rates to 6-months post-surgery, and (b) patient-reported outcomes to 6-months post-surgery.

METHODS

A retrospective cohort study of patients who underwent TKA or THA between January 2013 and June 2018 from two Australian public hospitals was undertaken utilizing linked individual patient-level data from two prospectively collected independent databases comprising approximately 3,500 and 9,500 people (database contained known opioid usage data within the 5-year time frame). Inclusion criteria included (i) primary diagnosis of osteoarthritis of the index joint, (ii) primary elective THA or TKA, and (iii) age ≥ 18 years. Exclusion criteria included (i) revision arthroplasty, (ii) non-elective arthroplasty, (iii) hip hemiarthroplasty, (iv) uni-compartmental knee arthroplasty, and (v) previous unilateral high tibial osteotomy.

RESULTS

Analysis was completed on 1,187 study participants (64% female, 69% TKA, mean (SD) age 67 [9.9]). 30% were using regular opioids preoperatively. Adjusted regression analyses controlling for multiple co-variates indicated no significant association between preoperative opioid use and complications/readmission rates or patient-reported outcomes to 6 months post-surgery. Model diagnostics produced poor discrimination for area under the curves and non-significant goodness of fit tests. Pre-arthroplasty opioid use was associated with lower health-related quality of life (EuroQol-Visual Analogue Scale) compared to non-opioid users undergoing primary THA (mean difference -5.04 [-9.87, -0.22], P = 0.04, Adjusted R2 = 0.06) CONCLUSION: In this study, 30% of patients were using prescribed opioids daily prior to primary TKA or THA. Pre-arthroplasty opioid use was not associated with postoperative adverse events or patient-reported pain, function, or global perceived improvement up to six months post-surgery.

Do preoperative PROMIS scores independently predict 90-day readmission following primary total knee arthroplasty?

INTRODUCTION

The rising demand for total knee arthroplasty (TKA) is expected to increase the total number of TKA-related readmissions, presenting significant public health and economic burden. With the increasing use of Patient-Reported Outcomes Measurement Information System (PROMIS) scores to inform clinical decision-making, this study aimed to investigate whether preoperative PROMIS scores are predictive of 90-day readmissions following primary TKA.

MATERIALS AND METHODS

We retrospectively reviewed a consecutive series of 10,196 patients with preoperative PROMIS scores who underwent primary TKA. Two comparison groups, readmissions (n = 79; 3.6%) and non-readmissions (n = 2091; 96.4%) were established. Univariate and multivariate logistic regression analyses were then performed with readmission as the outcome variable to determine whether preoperative PROMIS scores could predict 90-day readmission.

RESULTS

The study cohort consisted of 2170 patients overall. Non-white patients (OR = 3.53, 95% CI [1.16, 10.71], p = 0.026) and patients with cardiovascular or cerebrovascular disease (CVD) (OR = 1.66, 95% CI [1.01, 2.71], p = 0.042) were found to have significantly higher odds of 90-day readmission after TKA. Preoperative PROMIS-PF10a (p = 0.25), PROMIS-GPH (p = 0.38), and PROMIS-GMH (p = 0.07) scores were not significantly associated with 90-day readmission.

CONCLUSION

This study demonstrates that preoperative PROMIS scores may not be used to predict 90-day readmission following primary TKA. Non-white patients and patients with CVD are 3.53 and 1.66 times more likely to be readmitted, highlighting existing racial disparities and medical comorbidities contributing to readmission in patients undergoing TKA.

Functional positioning principles for image-based robotic-assisted TKA achieved a higher Forgotten Joint Score at 1 year compared to conventional TKA with restricted kinematic alignment

PURPOSE

Few comparative studies assessed the current concept of individualised alignment for total knee arthroplasty (TKA) and their outcomes at mid-term. This study aimed to evaluate the functional outcomes at 1 year of primary TKA performed with a functional positioning technique based on an image-based robotic-assisted system, compared to conventional TKA performed with a restricted kinematic alignment technique.

METHODS

This retrospective comparative study included 100 primary TKAs performed with functional positioning principles using an image-based robotic-assisted system. A control group included 100 primary TKAs with the same posterior-stabilised implant as the robotic group but performed with manual instrumentation and restricted kinematic alignment technique. In the robotic group, the mean age was 69.2 years old ± 7.9; the mean body mass index was 29.7 kg/m2 ± 4.6. The demographic characteristics were similar between both groups. Kujala score, Forgotten Joint Score (FJS), Knee Society Score (KSS) knee and KSS function were collected 12 months postoperatively. Normally distributed continuous variables were compared using the Student t test. For non-normally distributed continuous variables, the Mann-Whitney test was used.

RESULTS

FJS was significantly higher in the robotic group (76.3 ± 13 vs. 68.6 ± 16.9 in the conventional group; p = 0.026). At a 1-year follow-up, there was no significant difference in the KSS knee and KSS function scores and the Kujala score between both groups. The mean KSS knee score was 90.8 ± 11.4 in the robotic group versus 89.4 ± 9.6 in the conventional group (p = 0.082). The mean KSS function score was 91.4 ± 12.3 versus 91.3 ± 12.6, respectively (p = 0.778).

CONCLUSION

Functional positioning principles using an image-based robotic-assisted system achieved a higher Forgotten Joint Score 1 year after TKA compared to restricted kinematic alignment. Personalised alignment and implant positioning are interesting paths to improve the functional outcomes after TKA.

LEVEL OF EVIDENCE

III.

The effect of body mass index and preoperative weight loss in people with obesity on postoperative outcomes to 6 months following total hip or knee arthroplasty: a retrospective study

BACKGROUND

Few studies have investigated the association between obesity, preoperative weight loss and postoperative outcomes beyond 30- and 90-days post-arthroplasty. This study investigated whether body mass index (BMI) and preoperative weight loss in people with obesity predict postoperative complications and patient-reported outcomes 6 months following total knee or hip arthroplasty.

METHODS

Two independent, prospectively collected datasets of people undergoing primary total knee or hip arthroplasty for osteoarthritis between January 2013 and June 2018 at two public hospitals were merged. First, the sample was grouped into BMI categories, < 35 kg/m2 and ≥ 35 kg/m2. Subgroup analysis was completed separately for hips and knees. Second, a sample of people with BMI ≥ 30 kg/m2 was stratified into participants who did or did not lose ≥ 5% of their baseline weight preoperatively. The presence of postoperative complications, Oxford Hip Score, Oxford Knee Score, EuroQol Visual Analogue Scale and patient-rated improvement 6 months post-surgery were compared using unadjusted and adjusted techniques.

RESULTS

From 3,552 and 9,562 patients identified from the datasets, 1,337 were included in the analysis after merging. After adjustment for covariates, there was no difference in postoperative complication rate to 6 months post-surgery according to BMI category (OR 1.0, 95%CI 0.8-1.4, P = 0.8) or preoperative weight loss (OR 1.1, 95%CI 0.7-1.8, P = 0.7). There was no between-group difference according to BMI or preoperative weight change for any patient-reported outcomes 6 months post-surgery.

CONCLUSION

Preoperative BMI or a 5% reduction in preoperative BMI in people with obesity was not associated with postoperative outcomes to 6 months following total knee or hip arthroplasty.

Improved performance of machine learning models in predicting length of stay, discharge disposition, and inpatient mortality after total knee arthroplasty using patient-specific variables

BACKGROUND

This study aimed to compare the performance of ten predictive models using different machine learning (ML) algorithms and compare the performance of models developed using patient-specific vs. situational variables in predicting select outcomes after primary TKA.

METHODS

Data from 2016 to 2017 from the National Inpatient Sample were used to identify 305,577 discharges undergoing primary TKA, which were included in the training, testing, and validation of 10 ML models. 15 predictive variables consisting of 8 patient-specific and 7 situational variables were utilized to predict length of stay (LOS), discharge disposition, and mortality. Using the best performing algorithms, models trained using either 8 patient-specific and 7 situational variables were then developed and compared.

RESULTS

For models developed using all 15 variables, Linear Support Vector Machine (LSVM) was the most responsive model for predicting LOS. LSVM and XGT Boost Tree were equivalently most responsive for predicting discharge disposition. LSVM and XGT Boost Linear were equivalently most responsive for predicting mortality. Decision List, CHAID, and LSVM were the most reliable models for predicting LOS and discharge disposition, while XGT Boost Tree, Decision List, LSVM, and CHAID were most reliable for mortality. Models developed using the 8 patient-specific variables outperformed those developed using the 7 situational variables, with few exceptions.

CONCLUSION

This study revealed that performance of different models varied, ranging from poor to excellent, and demonstrated that models developed using patient-specific variables were typically better predictive of quality metrics after TKA than those developed employing situational variables.

LEVEL OF EVIDENCE

III.

Prehabilitation for Patients Undergoing Orthopedic Surgery: A Systematic Review and Meta-analysis

Importance

Prehabilitation programs for patients undergoing orthopedic surgery have been gaining popularity in recent years. However, the current literature has produced varying results.

Objective

To evaluate whether prehabilitation is associated with improved preoperative and postoperative outcomes compared with usual care for patients undergoing orthopedic surgery.

Data Sources

Bibliographic databases (MEDLINE, CINAHL [Cumulative Index to Nursing and Allied Health Literature], AMED [Allied and Complementary Medicine], Embase, PEDRO [Physiotherapy Evidence Database], and Cochrane Central Register of Controlled Trials) were searched for published trials, and the Institute for Scientific Information Web of Science, System for Information on Grey Literature in Europe, and European clinical trials registry were searched for unpublished trials from January 1, 2000, to June 30, 2022.

Study Selection

Randomized clinical trials (RCTs) comparing prehabilitation with standard care for any orthopedic surgical procedure were included.

Data Extraction and Synthesis

Two independent reviewers screened trials. Data were pooled using a random-effects model. Recommendations were determined using the Grading of Recommendations Assessment, Development and Evaluation system and the study was conducted according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) reporting guideline.

Main Outcomes and Measures

Pain, function, muscle strength, and health-related quality of life (HRQOL).

Results

Forty-eight unique trials involving 3570 unique participants (2196 women [61.5%]; mean [SD] age, 64.1 [9.1] years) were analyzed. Preoperatively, moderate-certainty evidence favoring prehabilitation was reported for patients undergoing total knee replacement (TKR) for function (standardized mean difference [SMD], -0.70 [95% CI, -1.08 to -0.32]) and muscle strength and flexion (SMD, 1.00 [95% CI, 0.23-1.77]) and for patients undergoing total hip replacement (THR) for HRQOL on the 36-item Short Form Health Survey (weighted mean difference [WMD], 7.35 [95% CI, 3.15-11.54]) and muscle strength and abduction (SMD, 1.03 [95% CI, 0.03-2.02]). High-certainty evidence was reported for patients undergoing lumbar surgery for back pain (WMD, -8.20 [95% CI, -8.85 to -7.55]) and moderate-certainty evidence for HRQOL (SMD, 0.46 [95% CI, 0.13-0.78]). Postoperatively, moderate-certainty evidence favoring prehabilitation was reported for function at 6 weeks in patients undergoing TKR (SMD, -0.51 [95% CI, -0.85 to -0.17]) and at 6 months in those undergoing lumbar surgery (SMD, -2.35 [95% CI, -3.92 to -0.79]). Other differences in outcomes favoring prehabilitation were of low to very low quality of evidence.

Conclusions and Relevance

In this systematic review and meta-analysis of RCTs, moderate-certainty evidence supported prehabilitation over usual care in improving preoperative function and strength in TKR and HRQOL and muscle strength in THR, high-certainty evidence in reducing back pain, and moderate-certainty evidence in improving HRQOL in lumbar surgery. Postoperatively, moderate-certainty evidence supported prehabilitation for function following TKR at 6 weeks and lumbar surgery at 6 months. Prehabilitation showed promising results for other outcomes, although high risk of bias and heterogeneity affected overall quality of evidence. Additional RCTs with a low risk of bias investigating preoperative and postoperative outcomes for all orthopedic surgical procedures are required.

An orthopaedic intelligence application successfully integrates data from a smartphone-based care management platform and a robotic knee system using a commercial database

PURPOSE

To evaluate the feasibility of using a smartphone-based care management platform (sbCMP) and robotic-assisted total knee arthroplasty (raTKA) to collect data throughout the episode-of-care and assess if intra-operative measures of soft tissue laxity in raTKA were associated with post-operative outcomes.

METHODS

A secondary data analysis of 131 patients in a commercial database who underwent raTKA was performed. Pre-operative through six week post-operative step counts and KOOS JR scores were collected and cross-referenced with intra-operative laxity measures. A Kruskal-Wallis test or a Wilcoxon sign-rank was used to assess outcomes.

RESULTS

There were higher step counts at six weeks post-operatively in knees with increased laxity in both the lateral compartment in extension and medial compartment in flexion (p < 0.05). Knees balanced in flexion within < 0.5 mm had higher KOOS JR scores at six weeks post-operative (p = 0.034) compared to knees balanced within 0.5-1.5 mm.

CONCLUSION

A smartphone-based care management platform can be integrated with raTKA to passively collect data throughout the episode-of-care. Associations between intra-operative decisions regarding laxity and post-operative outcomes were identified. However, more robust analysis is needed to evaluate these associations and ensure clinical relevance to guide machine learning algorithms.

Pre- and Post-Operative Rehabilitation Interventions in Patients at Risk of Poor Outcomes Following Knee or Hip Arthroplasty: Protocol for Two Systematic Reviews

Objective

Total knee (TKA) and hip arthroplasty (THA) are successful procedures in treating end-stage osteoarthritis when nonoperative treatments fail. However, a growing body of literature has been reporting suboptimal outcomes following TKA and THA. While pre- and post-operative rehabilitation is imperative to recovery, little is known about their effectiveness for patients at risk of poor outcomes. In the 2 systematic reviews with identical methodology, we aim to evaluate the effectiveness of (a) pre-operative and (b) post-operative rehabilitation interventions for patients at risk of poor outcomes following TKA and THA.

Methods

The 2 systematic reviews will follow the principles and recommendations outlined in the Cochrane Handbook. Only randomized controlled trials (RCTs) and pilot RCTs will be searched in 6 databases: CINAHL, MEDLINE, Embase, Web of Science, Pedro, and OTseeker. Eligible studies including patients at risk of poor outcomes and evaluating rehabilitation interventions following and preceding arthroplasty will be considered for inclusion. Primary outcomes will include performance-based tests and functional patient-reported outcome measures, and secondary outcomes will include health-related quality of life and pain. The quality of eligible RCTs will be evaluated using the Cochrane's risk of bias tool, and the strength of evidence will be assessed using the Grades of Recommendation, Assessment, Development, and Evaluation (GRADE).

Discussion

These reviews will synthesize the evidence regarding the effectiveness of pre-and post-operative rehabilitation interventions for patients at risk of poor outcomes, which in turn may inform practitioners and patients in planning and implementing the most optimal rehabilitation programs to achieve the best outcomes after arthroplasty.

Systematic Review Registration

PROSPERO CRD42022355574.

Pain Catastrophizing Predicts Opioid and Health-Care Utilization After Orthopaedic Surgery: A Secondary Analysis of Trial Participants with Spine and Lower-Extremity Disorders

BACKGROUND

Most individuals undergoing elective surgery expect to discontinue opioid use after surgery, but many do not. Modifiable risk factors including psychosocial factors are associated with poor postsurgical outcomes. We wanted to know whether pain catastrophizing is specifically associated with postsurgical opioid and health-care use.

METHODS

This was a longitudinal cohort study of trial participants undergoing elective spine (lumbar or cervical) or lower-extremity (hip or knee osteoarthritis) surgery between 2015 and 2018. Primary and secondary outcomes were 12-month postsurgical days' supply of opioids and surgery-related health-care utilization, respectively. Self-reported and medical record data included presurgical Pain Catastrophizing Scale (PCS) scores, surgical success expectations, opioid use, and pain interference duration.

RESULTS

Complete outcomes were analyzed for 240 participants with a median age of 42 years (34% were female, and 56% were active-duty military service members). In the multivariable generalized additive model, greater presurgical days' supply of opioids (F = 17.23, p < 0.001), higher pain catastrophizing (F = 1.89, p = 0.004), spine versus lower-extremity surgery (coefficient estimate = 1.66 [95% confidence interval (CI), 0.50 to 2.82]; p = 0.005), and female relative to male sex (coefficient estimate = -1.25 [95% CI, -2.38 to -0.12]; p = 0.03) were associated with greater 12-month postsurgical days' supply of opioids. Presurgical opioid days' supply (chi-square = 111.95; p < 0.001), pain catastrophizing (chi-square = 96.06; p < 0.001), and lower extremity surgery (coefficient estimate = -0.17 [95% CI, -0.24 to -0.11]; p < 0.001), in addition to age (chi-square = 344.60; p < 0.001), expected recovery after surgery (chi-square = 54.44; p < 0.001), active-duty status (coefficient estimate = 0.58 [95% CI, 0.49 to 0.67]; p < 0.001), and pain interference duration (chi-square = 43.47; p < 0.001) were associated with greater health-care utilization.

CONCLUSIONS

Greater presurgical days' supply of opioids and pain catastrophizing accounted for greater postsurgical days' supply of opioids and health-care utilization. Consideration of several modifiable factors provides an opportunity to improve postsurgical outcomes.

LEVEL OF EVIDENCE

Prognostic Level III . See Instructions for Authors for a complete description of levels of evidence.

Can Patient-Reported Outcome Measurement Information System Measures Differentiate Patients Who Will Undergo Hip and Knee Total Joint Arthroplasty: A Retrospective Case-Control Study

BACKGROUND

The Patient-Reported Outcome Measurement Information System (PROMIS) can be used to monitor patients in population-health-based programs. However, it is unknown which measures are most appropriate to differentiate patients who will undergo hip or knee total joint arthroplasty (TJA) in a cohort of patients with osteoarthritis.

METHODS

A retrospective cohort of new patients consulting for treatment from November 17, 2017 to April 20, 2020 (cases: hip: n = 157, knee: n = 112; randomly selected nonsurgical controls: hip: n = 314, knee: n = 224) was extracted from the electronic health record. We recorded demographics, comorbidity, and PROMIS scores for 8 domains (physical function, pain interference, pain intensity, anxiety, depression, sleep disturbance, ability to participate in social roles and activities, and fatigue). We performed descriptive statistics to characterize the cohorts and baseline PROMIS scores and conducted logistic regression models to determine which PROMIS domains differentiated patients undergoing hip and knee TJA.

RESULTS

In univariate comparisons of PROMIS domains, the hip and knee surgical cohorts differed from controls in physical function (P < .01), pain interference (P < .01), and ability to participate in social roles and activities (P < .02). In logistic regression models informed by univariate analyses, PROMIS physical function was the only PROMIS measure to differentiate undergoing surgery in both hip and knee cohorts (P < .01).

CONCLUSION

PROMIS physical function can differentiate TJA cases from nonsurgical controls in both hip and knee patients. These findings have implications for considering which PROMIS measures to administer in patients with hip and knee osteoarthritis.

Patient-Reported Outcomes for Function and Pain in Total Knee Arthroplasty Patients

BACKGROUND

Some patients undergoing total knee arthroplasty successfully manage their condition postoperatively, while others encounter challenges in regaining function and controlling pain during recovery at home.

OBJECTIVE

To use traditional statistics and machine learning to develop prediction models that identify patients likely to have increased care needs related to managing function and pain following total knee arthroplasty.

METHODS

This study included 201 patients. Outcomes were changes between baseline and follow-up in the functional and pain subcomponents of the Oxford Knee Score. Both classification and regression modeling were applied. Twenty-one predictors were included. Tenfold cross-validation was used, and the regression models were evaluated based on root mean square error, mean absolute error, and coefficient of determination. Classification models were evaluated based on the area under the receiver operating curve, sensitivity, and specificity.

RESULTS

In classification modeling, random forest and stochastic gradient boosting provided the best overall metrics for model performance. A support vector machine and a stochastic gradient boosting machine in regression modeling provided the best predictive performance. The models performed better in predicting challenges related to function compared to challenges related to pain.

DISCUSSION

There is valuable predictive information in the data routinely collected for patients undergoing total knee arthroplasty. The developed models may predict patients who are likely to have enhanced care needs regarding function and pain management. Improvements are needed before the models can be implemented in routine clinical practice.

Artificial intelligence in knee arthroplasty: current concept of the available clinical applications

Background

Artificial intelligence (AI) is defined as the study of algorithms that allow machines to reason and perform cognitive functions such as problem-solving, objects, images, word recognition, and decision-making. This study aimed to review the published articles and the comprehensive clinical relevance of AI-based tools used before, during, and after knee arthroplasty.

Methods

The search was conducted through PubMed, EMBASE, and MEDLINE databases from 2000 to 2021 using the 2009 Preferred Reporting Items for Systematic Reviews and Meta-Analyses Protocol (PRISMA).

Results

A total of 731 potential articles were reviewed, and 132 were included based on the inclusion criteria and exclusion criteria. Some steps of the knee arthroplasty procedure were assisted and improved by using AI-based tools. Before surgery, machine learning was used to aid surgeons in optimizing decision-making. During surgery, the robotic-assisted systems improved the accuracy of knee alignment, implant positioning, and ligamentous balance. After surgery, remote patient monitoring platforms helped to capture patients’ functional data.

Conclusion

In knee arthroplasty, the AI-based tools improve the decision-making process, surgical planning, accuracy, and repeatability of surgical procedures.

Predictive capacity of four machine learning models for in-hospital postoperative outcomes following total knee arthroplasty

Background

Machine learning (ML) methods have shown promise in the development of patient-specific predictive models prior to surgical interventions. The purpose of this study was to develop, test, and compare four distinct ML models to predict postoperative parameters following primary total knee arthroplasty (TKA).

Methods

Data from the Nationwide Inpatient Sample was used to identify patients undergoing TKA during 2016-2017. Four distinct ML models predictive of mortality, length of stay (LOS), and discharge disposition were developed and validated using 15 predictive patient and hospital-specific factors. Area under the curve of the receiver operating characteristic curve (AUCROC) and accuracy were used as validity metrics, and the strongest predictive variables under each model were assessed.

Results

A total of 305,577 patients were included. For mortality, the XGBoost, neural network (NN), and LSVM models all had excellent responsiveness during validation, while random forest (RF) had fair responsiveness. For predicting LOS, all four models had poor responsiveness. For the discharge disposition outcome, the LSVM, NN, and XGBoost models had good responsiveness, while the RF model had poor responsiveness. LSVM and XGBoost had the highest responsiveness for predicting discharge disposition with an AUCROC of 0.747.

Discussion

The ML models tested demonstrated a range of poor to excellent responsiveness and accuracy in the prediction of the assessed metrics, with considerable variability noted in the predictive precision between the models. The continued development of ML models should be encouraged, with eventual integration into clinical practice in order to inform patient discussions, management decision making, and health policy.