Background
Estimating the risk of revision after arthroplasty could inform patient and surgeon decision-making. However, there is a lack of well-performing prediction models assisting in this task, which may be due to current conventional modeling approaches such as traditional survivorship estimators (such as Kaplan-Meier) or competing risk estimators. Recent advances in machine learning survival analysis might improve decision support tools in this setting. Therefore, this study aimed to assess the performance of machine learning compared with that of conventional modeling to predict revision after arthroplasty.

Question/purpose
Does machine learning perform better than traditional regression models for estimating the risk of revision for patients undergoing hip or knee arthroplasty?

Methods
Eleven datasets from published studies from the Dutch Arthroplasty Register reporting on factors associated with revision or survival after partial or total knee and hip arthroplasty between 2018 and 2022 were included in our study. The 11 datasets were observational registry studies, with a sample size ranging from 3038 to 218,214 procedures. We developed a set of time-to-event models for each dataset, leading to 11 comparisons. A set of predictors (factors associated with revision surgery) was identified based on the variables that were selected in the included studies. We assessed the predictive performance of two state-of-the-art statistical time-to-event models for 1-, 2-, and 3-year follow-up: a Fine and Gray model (which models the cumulative incidence of revision) and a cause-specific Cox model (which models the hazard of revision). These were compared with a machine-learning approach (a random survival forest model, which is a decision tree–based machine-learning algorithm for time-to-event analysis). Performance was assessed according to discriminative ability (time-dependent area under the receiver operating curve), calibration (slope and intercept), and overall prediction error (scaled Brier score). Discrimination, known as the area under the receiver operating characteristic curve, measures the model’s ability to distinguish patients who achieved the outcomes from those who did not and ranges from 0.5 to 1.0, with 1.0 indicating the highest discrimination score and 0.50 the lowest. Calibration plots the predicted versus the observed probabilities; a perfect plot has an intercept of 0 and a slope of 1. The Brier score calculates a composite of discrimination and calibration, with 0 indicating perfect prediction and 1 the poorest. A scaled version of the Brier score, 1 – (model Brier score/null model Brier score), can be interpreted as the amount of overall prediction error.

Results
Using machine learning survivorship analysis, we found no differences between the competing risks estimator and traditional regression models for patients undergoing arthroplasty in terms of discriminative ability (patients who received a revision compared with those who did not). We found no consistent differences between the validated performance (time-dependent area under the receiver operating characteristic curve) of different modeling approaches because these values ranged between -0.04 and 0.03 across the 11 datasets (the time-dependent area under the receiver operating characteristic curve of the models across 11 datasets ranged between 0.52 to 0.68). In addition, the calibration metrics and scaled Brier scores produced comparable estimates, showing no advantage of machine learning over traditional regression models.

Conclusion
Machine learning did not outperform traditional regression models.

Clinical Relevance
Neither machine learning modeling nor traditional regression methods were sufficiently accurate in order to offer prognostic information when predicting revision arthroplasty. The benefit of these modeling approaches may be limited in this context.

@en

Background: The objective consisted of 2 elements, primarily to define 2 bone geometry variations of the ankle that may be of prognostic value on ankle instability and secondly to translate these bone variations from a 3D model to a simple 2D radiographic measurement for clinical use. Methods: The 3D tibial and talar shape differences derived from earlier studies were translated to two 2D radiographic parameters: the medial malleolar height angle (MMHA) and talar convexity angle (TCA) respectively to ensure clinical use. To assess validity, the MMHA and TCA were measured on 3D polygons derived from lower leg computed tomographic (CT) scans and 2D digitally reconstructed radiographs (DRRs) of these polygons. To assess reliability, the MMHA and TCA were measured on standard radiographs by 2 observers calculating the intraclass correlation coefficient (ICC). Results: The 3D angle measurements on the polygons showed substantial to excellent agreement with the 2D measurements on DRR for both the MMHA (ICC 0.84-0.93) and TCA (ICC 0.88-0.96). The interobserver reliability was moderate with an ICC of 0.58 and an ICC of 0.64 for both the MMHA and TCA, respectively. The intraobserver reliability was excellent with an ICC of 0.96 and 0.97 for the MMHA and the TCA, respectively. Conclusion: Two newly defined radiographic parameters (MMHA and TCA) are valid and can be assessed with excellent intraobserver reliability on standard radiographs. The interobserver reliability was moderate and indicates training is required to ensure uniformity in measurement technique. The current method may be used to translate more variations in bone shape prior to implementation in clinical practice. Level of Evidence: Level III, cohort study.

@en

Background:
Statistical models using machine learning (ML) have the potential for more accurate estimates of the probability of binary events than logistic regression. The present study used existing data sets from large musculoskeletal trauma trials to address the following study questions: (1) Do ML models produce better probability estimates than logistic regression models? (2) Are ML models influenced by different variables than logistic regression models?

Methods:
We created ML and logistic regression models that estimated the probability of a specific fracture (posterior malleolar involvement in distal spiral tibial shaft and ankle fractures, scaphoid fracture, and distal radial fracture) or adverse event (subsequent surgery [after distal biceps repair or tibial shaft fracture], surgical site infection, and postoperative delirium) using 9 data sets from published musculoskeletal trauma studies. Each data set was split into training (80%) and test (20%) subsets. Fivefold cross-validation of the training set was used to develop the ML models. The best-performing model was then assessed in the independent testing data. Performance was assessed by (1) discrimination (c-statistic), (2) calibration (slope and intercept), and (3) overall performance (Brier score).

Results:
The mean c-statistic was 0.01 higher for the logistic regression models compared with the best ML models for each data set (range, −0.01 to 0.06). There were fewer variables strongly associated with variation in the ML models, and many were dissimilar from those in the logistic regression models.

Conclusions:
The observation that ML models produce probability estimates comparable with logistic regression models for binary events in musculoskeletal trauma suggests that their benefit may be limited in this context.@en

Bone shapes, particularly those defining the subtalar joint (STJ), have not received much attention yet as a risk factor for developing chronic ankle instability (CAI) after sustaining a lateral ankle sprain (LAS). This study aimed to compare three-dimensional (3D) shape variations in the STJ bones within individuals with CAI and healthy controls. 3D statistical shape models (SSMs) of the STJ bones were built to describe the bone shape variations observed within a population consisting of 26 individuals with unilateral CAI and 26 healthy controls. Using the SSMs and analysis of covariance test, age- and gender-adjusted shape variations in the bones were compared within individuals with CAI and healthy controls. The mean age of the CAI patients (14 males and 12 females) and healthy controls (12 males and 14 females) was 29 (standard deviation [SD] = 11) and 36 years (SD = 11), respectively. Tali and calcanei did not significantly vary between ipsilateral CAI and their contralateral ankle. Two shape modes, one for the talus (p = 0.015, variations in the curvature of the talar lateral process and the inclination angle of the talar neck relative to the body) and one for the calcaneus (p = 0.003, variations in the medial and lateral tuberosities, and the contour of the anterior articular surface), described significant shape differences between the CAI patients and healthy controls. The CAI patients generally had flatter talar joint surfaces and a flattened calcaneal ground-contact surface. These findings suggest that specific bone shapes may increase the risk of developing CAI after sustaining a LAS.

@en

Objective: Surgical microfracture is considered a first-line treatment for talar osteochondral defects. However, current rigid awls and drills limit access to all locations in human joints and increase risk of heat necrosis of bone. Using a flexible water jet instrument to drill holes can improve the reachability of the defect without inducing thermal damage. The aim of this feasibility study is to determine whether water jet drilling is potentially safe compared with conventional microfracture awls by studying side effects and perioperative complications, as well as the quality of cartilage repair tissue. Design: Talar chondral defects with 6-mm diameter were created bilaterally in 6 goats (12 samples). One defect in each goat was treated with microfracture created with conventional awls, the contralateral defect was treated with holes created with 5-second water jet bursts at a pressure of 50 MPa. Postoperative complications were recorded and after 24 weeks analyses were performed using the ICRS (International Cartilage Repair Society) macroscopic score and modified O’Driscoll histological score. Results: Several practical issues using the water jet in the operating theatre were noted. Water jet drilling resulted in fibrocartilage repair tissue similar to the repair tissue from conventional awls. Conclusions: These results suggest that water jet drilling gives adequate fibrocartilage repair tissue. Furthermore, the results highlight essential prerequisites for safe application of surgical water jet drilling: stable water pressure, water jet beam coherence, stable positioning of the nozzle head when jetting, and minimizing excessive fluid extravasation.

@en

In orthopaedic surgery, water jet drilling provides several advantages over classic drilling with rigid drilling bits, such as the always sharp cut, absence of thermal damage and increased manoeuvrability. Previous research showed that the heterogeneity of bone tissue can cause variation in drilling depth whilst water jet drilling. To improve control over the drilling depth, a new method is tested consisting of two water jets that collide directly below the bone surface. The expected working principle is that after collision the jets will disintegrate, with the result of eliminating the destructive power of the coherent jets and leaving the bone tissue underneath the focal point intact. To assess the working principle of colliding water jets (CWJ), the influence of inhomogeneity of the bone tissue on the variation of the drilling depth and the impact of jet time (twj) on the drilling depth were compared to a single water jet (SWJ) with a similar power. 98 holes were drilled in 14 submerged porcine tali with two conditions CWJ (impact angle of 30° and 90°) and SWJ. The water pressure was 70 MPa for all conditions. The water jet diameter was 0.3 mm for CWJ and 0.4 mm for SWJ. twj was set at 1, 3, 5 and 8 s. Drilling depth and hole diameter were measured using microCT scans. A non-parametric Levene's test was performed to assess a significant difference in variance between conditions SWJ and CWJ. A regression analysis was used to determine differences in influence of twj on the drilling depth. Hole diameter differences were assessed using a one way Anova. A significance level of p<0.05 was set. Condition CWJ significantly decreases the drilling depth variance caused by the heterogeneity of the bone when compared to SWJ. The mean depth for CWJ was 0.9 mm (SD 0.3 mm) versus 4.8 mm (SD 2.0) for SWJ. twj affects the drilling depth less for condition CWJ (p<0.01, R2=0.30) than for SWJ (p<0.01, R2=0.46). The impact angle (30° or 90°) of the CWJ does not influence the drilling depth nor the variation in depth. The diameters of the resulting holes in the direction of the jets is significantly larger for CWJ at 90° than for 30° or a single jet. This study shows that CWJ provides accurate depth control when water jet drilling in an inhomogeneous material such as bone. The maximum variance measured by using the 95% confidence interval is 0.6 mm opposed to 5.4 mm for SWJ. This variance is smaller than the accuracy required for bone debridement treatments (2–4 mm deep) or drilling pilot holes. This confirms that the use of CWJ is an inherently safe method that can be used to accurately drill in bones.@en

The goal of this study was to determine the preferred handle design for two degrees of freedom steerable arthroscopic cutter by performing a two-step development approach. The expected usefulness and usability of control components of three entirely different handles were defined by an on-line survey with 101 students and the actual control by a standardised laboratory study with mock-up models by 16 students. The preferred handle design was integrated in a full functional prototype and optimized by 10 experts performing a meniscectomy on human cadaver knees. Students (survey 70% and task 91%) expected the same control behaviour as the experts (60%): steering a wheel to the right should evoke tip steering to the right regardless the orientation of the beak and moving a ring lever towards the handle’s centre point should evoke closure of the tip. Development of surgical instruments can benefit from expected control behavior based on daily life tools, but requires expert involvement for specific surgical tasks and context.

@en

To improve arthroscopic skills, the preferred means of training is cadaveric tissue, because this gives the most realistic scenario. A drawback of cadaveric training is that objective performance tracking and accompanied feedback cannot be provided due to the absence of a suitable system. The main criteria were that the system should be compatible with any cadaveric joint, be used with any type of instrument, easy to set up, and measure two critical parameters that reflect the task efficiency (task time) and safety (forces due to instrument-tissue interaction). This resulted in the development of a force measurement system which consists of a custom-made universal vice, a custom-designed six degreeof-freedom (DOF) force measurement table (FMT) coupled to a computer equipped with customized software to record the time and forces in all directions. The FMT was calibrated and able to measure forces in the range of 0-750 N, with an accuracy of 0.1 N. During two cadaveric training courses, measurements were performed with the FMT. It was observed that the acquired force data could discriminate between novices and experts or reflect a certain phase of a navigation task performed in a cadaveric cow and human knee. A distinct phase highlighted from the force measurements is the insufficient joint stressing of novices during navigation. This results in too small a joint space for inspection and forces the novices to readjust the stressing. As forces cannot be seen, the FMT can contribute to more efficient training by providing explicit cues on the exerted loads during training. This enables a more precise supervision of the trainees.@en