Demographic data is more predictive of component size than digital radiographic templating in total knee arthroplasty

Accuracy and clinical role of digital templating for total knee arthroplasty performed on haemophilic knees

INTRODUCTION

In total knee arthroplasty (TKA), choosing the correct implant size is important. There is lack of data on accuracy of templating on haemophilic knees. Our aim was to test the accuracy of 2D digital templating for TKA on haemophilic arthropathy (HA) of knee.

MATERIALS AND METHODS

TKAs performed on HA between January 2011 and January 2022 were screened. Osteoarthritis (OA) group was created as control group by a one-to-one matching regarding type of implant used. Intra- and interobserver correlations were measured in HA, then correlation between templated and implanted sizes was investigated in four assessments (femur AP, femur lateral, tibia AP, tibia lateral), then compared with OA group. Fifty-eight knees in each group included.

RESULTS

Regarding intraobserver correlation in HA, there was excellent correlation for femur AP [.93 (.73-.98)], femur lateral [.98 (.91-.99)], and tibia AP (1.0) templating. Regarding interobserver correlation in HA, excellent correlation was observed for femur lateral [.93 (.74-.98)] and tibia AP templating [.90 (.65-.97)]. Regarding correlation of templated and applied sizes in HA; tibia AP, tibia lateral and femur lateral templating showed good correlation [.81 (.70-.89), .86 (.77-.91), .79 (.67-.87) while femur AP templating showed moderate correlation [.67 (.50-.79)]. Comparing HA and OA, there was no difference in correlation levels regarding femur AP, femur lateral, tibia AP and tibia lateral templating (p = .056, p = .781, p = .761, p = .083, respectively).

CONCLUSION

Although 2D digital templating shows comparable correlation in HA and OA, clinical applicability of templating on HA appears to be limited in its current state.

Reliability of artificial intelligence in predicting total knee arthroplasty component sizes: a systematic review

PURPOSE

This systematic review aimed to investigate the reliability of AI predictive models of intraoperative implant sizing in total knee arthroplasty (TKA).

METHODS

Four databases were searched from inception till July 2023 for original studies that studied the reliability of AI prediction in TKA. The primary outcome was the accuracy ± 1 size. This review was conducted per PRISMA guidelines, and the risk of bias was assessed using the MINORS criteria.

RESULTS

A total of four observational studies comprised of at least 34,547 patients were included in this review. A mean MINORS score of 11 out of 16 was assigned to the review. All included studies were published between 2021 and 2022, with a total of nine different AI algorithms reported. Among these AI models, the accuracy of TKA femoral component sizing prediction ranged from 88.3 to 99.7% within a deviation of one size, while tibial component sizing exhibited an accuracy ranging from 90 to 99.9% ± 1 size.

CONCLUSION

This study demonstrated the potential of AI as a valuable complement for planning TKA, exhibiting a satisfactory level of reliability in predicting TKA implant sizes. This predictive accuracy is comparable to that of the manual and digital templating techniques currently documented in the literature. However, future research is imperative to assess the impact of AI on patient care and cost-effectiveness.

LEVEL OF EVIDENCE III

PROSPERO registration number: CRD42023446868.

Prediction of Total Knee Arthroplasty Sizes with Demographics, including Hand and Foot Sizes

Anticipating implant sizes before total knee arthroplasty (TKA) allows the surgical team to streamline operations and prepare for potential difficulties. This study aims to determine the correlation and derive a regression model for predicting TKA sizes using patient-specific demographics without using radiographs. We reviewed the demographics, including hand and foot sizes, of 1,339 primary TKAs. To allow for comparison across different TKA designs, we converted the femur and tibia sizes into their anteroposterior (AP) and mediolateral (ML) dimensions. Stepwise multivariate regressions were performed to analyze the data. Regarding the femur component, the patient's foot, gender, height, hand circumference, body mass index, and age was the significant demographic factors in the regression analysis (R-square 0.541, p < 0.05). For the tibia component, the significant factors in the regression analysis were the patient's foot size, gender, height, hand circumference, and age (R-square 0.608, p < 0.05). The patient's foot size had the highest correlation coefficient for both femur (0.670) and tibia (0.697) implant sizes (p < 0.05). We accurately predicted the femur component size exactly, within one and two sizes in 49.5, 94.2, and 99.9% of cases, respectively. Regarding the tibia, the prediction was exact, within one and two sizes in 53.0, 96.0, and 100% of cases, respectively. The regression model, utilizing patient-specific characteristics, such as foot size and hand circumference, accurately predicted TKA femur and tibia sizes within one component size. This provides a more efficient alternative for preoperative planning.

Preoperative Templating for Total Hip Arthroplasty: A Method for Calibrating Digital Radiographs Using Patient Demographics and Anthropometric Measurements

Background Preoperative templating aids the surgeon in estimating implant size and placement. Calibration markers are used to set the correct magnification of digital images before templating. Improper marker placement or complete absence can lead to inaccuracy or an inability to calibrate images altogether. Aims This study describes a method for calibrating images using a patient's femoral head size (FHS) predicted using demographics and anthropometric data. Materials and methods A formula predicting the FHS was derived from a cohort of 507 patients who underwent hemiarthroplasty for an intracapsular fractured neck of the femur through multivariate regression analysis. A separate validation cohort (n=50) who had undergone total hip arthroplasty (THA) had postoperative radiographs calibrated using the predicted FHS and the native contralateral hip as a surrogate calibration marker. The THA femoral head implant size was subsequently measured and compared with the actual implant size selected intraoperatively. Measurements were performed by two independent assessors to determine intra- and interobserver reliability. Results Multivariate regression analyses showed four variables significantly correlated with the size of the femoral head: gender (p < 0.001), height (p < 0.001), weight (p < 0.001), and race (Asian) (p = 0.01). Using these, a regression model to predict the FHS was obtained with an R2 value of 0.65 and a standard error of 2.18 mm. The validation cohort showed that THA head implant size could be accurately measured with an average root-mean-squared error (RMSE) of 1.41 mm (SD = 0.97 mm; %RMSE = 4.7%). The implant head size was measured to be within 5%, 10%, and 15% RMSE in 57.5%, 93.0%, and 100.0% of cases, respectively. There was excellent intraobserver (R2 = 0.94 and 0.95) and interobserver (R2 = 0.94) reliability. Conclusions The novel method proposed and validated in this study, using a predicted FHS to calibrate digital images, provides an alternative means of templating THA for fractured neck of the femur patients, in whom external calibration markers are often absent.

Preoperative predictors of implant size in patients undergoing total knee arthroplasty: a retrospective cohort study

BACKGROUND

Traditionally, the size of total knee arthroplasty (TKA) components is predicted by preoperative radiographic templating, which is of limited accuracy. This study aimed to evaluate the role of demographic data and ankle volume in predicting implant size in TKA candidates.

METHODS

In a retrospective study, 415 patients who underwent TKA at a single institution were included. The mean age of the patients was 67.5 ± 7.1 years. The mean BMI of the patients was 31.1 ± 4.7 kg/m2. TKA implants were Zimmer Biomet NexGen LPS-Flex Knee in all cases. The demographic data included age, sex, height, weight, BMI, ethnicity, and ankle volume. Ankle volume was assessed with the figure-of-eight method. Multivariate linear regression analysis was used for predicting factors of implant size.

RESULTS

Multivariate linear regression analysis showed that the Sex (β:1.41, P < 0.001), height (β:0.058, P < 0.001), ankle volume (β:0.11, P < 0.001), and Age (β:0.017, P = 0.004) were significant predictors of tibial component size. Sex (β:0.89, P < 0.001), height (β:0.035, P < 0.001), and ankle volume(β:0.091, P < 0.001) were significant predictors of femoral component size in the multivariate analysis.

CONCLUSION

Demographic data, adjunct with the ankle volume, could provide a promising model for preoperative prediction of the size of tibial and femoral components in TKA candidates.

The Adherence of Digital Templating of Cemented Bicondylar Total Knee Arthroplasty Reveals Gender Differences

INTRODUCTION

Preoperative digital templating is a standard procedure that should help the operating surgeon to perform an accurate intraoperative procedure. To date, a detailed view considering gender differences in templating total knee arthroplasty (TKA), stage of arthrosis, and the surgeons' experience altogether has not been conducted.

METHODS

A series of 521 patients who underwent bicondylar total knee arthroplasty was analyzed retrospectively for the planning adherence of digital templating in relation to sex, surgeon experience, and stage of arthrosis. Pre- and postoperative X-rays were comparably investigated for planned and implanted total knee arthroplasties. Digital templating was carried out through mediCAD version 6.5.06 (Hectec GmbH, 84032 Altdorf, Germany). For statistical analyses, IBM SPSS version 28 (IBM, 10504 Armonk, NY, US) was used.

RESULTS

The general planning adherence was 46.3% for the femur and 41.8% for the tibia. The Mann-Whitney U test revealed a gender difference for templating the femur (z = -5.486; p ≤ 0.001) and tibia (z = -3.139; p = 0.002). The surgeon's experience did not show a significant difference through the Kruskal-Wallis test in the femur (K-W H = 4.123; p = 0.127) and the tibia (K-W H = 2.455; p = 0.293). The stage of arthrosis only revealed a significant difference in the planning of the femur (K-L-score (K-W H = 6.516; p = 0.038) alone.

DISCUSSION/CONCLUSION

Digital templating for total knee arthroplasty brought up gender differences, with oversized implants for women and undersized implants for men. A high stage of femoral arthrosis can lead to the under and oversized planning of the surgeon. Since the surgeon's experience in planning did not show an effect on the adherence to templating, the beneficial effect of digital templating before surgery should be discussed.

Imprecise prediction of implant sizes with preoperative 2D digital templating in total knee arthroplasty

PURPOSE

To analyze the match between preoperatively determined implant size (2D templating) and intraoperatively used implant size in total knee arthroplasty (TKA). Also examined were the factors that might influence templating accuracy (gender, surgeon experience, obesity, etc.).

MATERIALS AND METHODS

The study was retrospective and conducted in a specialized ENDOCERT arthroplasty center. Digital templating was done with the MediCAD software. If the planned and implanted TKA components (both femur and tibia) were the same size, the match was classified "exact." A deviation of ± one size (at the femur or tibia or both) was classified "accurate." A deviation of ± two or more sizes (at the femur or tibia or both) was classified "inaccurate." Obesity, gender, implant type and surgeon experience were investigated for potential influence on templating accuracy. Chi-square tests and Cohen's weighted kappa test were used for statistical analysis.

RESULTS

A total of 482 cases [33.6% male, 66.4% female, age 69 ± 11, body mass index (BMI) 30.3 ± 5.8] were included. When the femur and tibia were taken together, exact size match was observed in 34% (95% CI 29.9-38.3%) of cases, accurate size match in 57.5% (95% CI 53-61.8%) and inaccurate size match in 8.5% (95% CI 6.3-11.2%). Inaccurate size match prolonged operative time (p = 0.028). Regarding the factors potentially influencing templating accuracy, only gender had a significant influence, with templating being more accurate in men (p = 0.004). BMI had no influence on accuracy (p = 0.87). No effect on accuracy was observed for implant type and surgeon experience.

CONCLUSIONS

The accuracy of 2D size templating in TKA is low, even in a specialized ENDOCERT arthroplasty center. The study findings challenge the usefulness of preoperative 2D size templating and highlight the importance of more reliable templating methods.

LEVEL OF EVIDENCE

Level III (retrospective observational study).

Machine learning algorithms predict within one size of the final implant ultimately used in total knee arthroplasty with good-to-excellent accuracy

PURPOSE

To develop a novel machine learning algorithm capable of predicting TKA implant sizes using a large, multicenter database.

METHODS

A consecutive series of primary TKA patients from two independent large academic and three community medical centers between 2012 and 2020 was identified. The primary outcomes were final tibial and femoral implant sizes obtained from an automated inventory system. Five machine learning algorithms were trained using six routinely collected preoperative features (age, sex, height, weight, and body mass index). Algorithms were validated on an independent set of patients and evaluated through accuracy, mean absolute error (MAE), and root mean-squared error (RMSE).

RESULTS

A total of 11,777 patients were included. The support vector machine (SVM) algorithm had the best performance for femoral component size(MAE = 0.73, RMSE = 1.06) with accuracies of 42.2%, 88.3%, and 97.6% for predicting exact size, ± one size, and ± two sizes, respectively. The elastic-net penalized linear regression (ENPLR) algorithm had the best performance for tibial component size (MAE 0.70, RMSE = 1.03) with accuracies of 43.8%, 90.0%, and 97.7% for predicting exact size, ± one size, and ± two sizes, respectively.

CONCLUSION

Machine learning algorithms demonstrated good-to-excellent accuracy for predicting within one size of the final tibial and femoral components used for TKA. Patient height and sex were the most important factors for predicting femoral and tibial component size, respectively. External validation of these algorithms is imperative prior to use in clinical settings.

LEVEL OF EVIDENCE

Case-control, III.

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.

Magnification assessment of radiographs for knee replacement (MARKeR) - A pilot study in a low-resource setting

Background

Selecting the correct size of implants to be used in total knee arthroplasty is critical for a successful outcome. Marker-less templating systems use an institutionally derived magnification factor for all radiographs.

Purpose

To determine the institutional magnification of knee radiographs for patients awaiting total knee arthroplasty.

Material and Methods

Eighty patients awaiting total knee arthroplasty underwent preoperative knee radiographs using a standardized protocol. A marker attached to the patients’ knees at the level of the knee joint was used to calculate the magnification factor on both anteroposterior (AP) and lateral (LAT) views. Two independent observers estimated the magnification to determine the intra and inter-observer reliability.

Results

The mean magnification of the AP (15.3%) radiograph was significantly greater than the LAT (12.1%) radiograph (p< 0.0001). Patients with absent markers on their radiographs were heavier than patients in whom the marker was visible (84.7 kgs vs. 76.6 kgs, p=0.01). No marker was visible on the radiographs in 56.3% (45/80) of patients. There was excellent inter and intra-observer reliability of both the AP and LAT measurements.

Conclusion

After standardizing the protocol for preoperative knee radiographs, our results show significantly greater institutional magnification of the anteroposterior compared with the lateral images. Accurate templating in knee arthroplasty requires both radiographic images. To reduce errors in implant sizing, we recommend surgeons use different institutional magnification factors for the anteroposterior and lateral radiographs.

Artificial Intelligence Based Patient-Specific Preoperative Planning Algorithm for Total Knee Arthroplasty

Previous studies have shown that the manufacturer's default preoperative plans for total knee arthroplasty with patient-specific guides require frequent, time-consuming changes by the surgeon. Currently, no research has been done on predicting preoperative plans for orthopedic surgery using machine learning. Therefore, this study aims to evaluate whether artificial intelligence (AI) driven planning tools can create surgeon and patient-specific preoperative plans that require fewer changes by the surgeon. A dataset of 5409 preoperative plans, including the manufacturer's default and the plans corrected by 39 surgeons, was collected. Features were extracted from the preoperative plans that describe the implant sizes, position, and orientation in a surgeon- and patient-specific manner. Based on these features, non-linear regression models were employed to predict the surgeon's corrected preoperative plan. The average number of corrections a surgeon has to make to the preoperative plan generated using AI was reduced by 39.7% compared to the manufacturer's default plan. The femoral and tibial implant size in the manufacturer's plan was correct in 68.4% and 73.1% of the cases, respectively, while the AI-based plan was correct in 82.2% and 85.0% of the cases, respectively, compared to the surgeon approved plan. Our method successfully demonstrated the use of machine learning to create preoperative plans in a surgeon- and patient-specific manner for total knee arthroplasty.

Validation and performance of a machine-learning derived prediction guide for total knee arthroplasty component sizing

INTRODUCTION

Anticipation of patient-specific component sizes prior to total knee arthroplasty (TKA) is essential to avoid excessive cost associated with additional surgical trays and morbidity associated with imperfect sizing. Current methods of size prediction, including templating, are inconsistent and time-consuming. Machine learning (ML) algorithms may allow for accurate TKA component size prediction with the ability to make predictions in real-time.

METHODS

Consecutive patients receiving primary TKA between 2012 and 2020 from two large tertiary academic and six community hospitals were identified. The primary outcomes were the final femoral and tibial component sizes extracted from automated inventory systems. Five ML algorithms were trained with routinely corrected demographic variables (age, height, weight, body mass index, and sex) using 80% of the study population and internally validated on an independent set of the remaining 20% of patients. Algorithm performance was evaluated through accuracy, mean absolute error (MAE), and root mean-squared error (RMSE).

RESULTS

A total of 17,283 patients that received one of 9 TKA implants from independent manufacturers were included. The SGB model accuracy for predicting ± 4-mm of the true femoral anteroposterior diameter was 83.6% and for ± 1 size of the true femoral component size was 95.0%. The SGB model accuracy for predicting ± 4-mm of the true tibial medial/lateral diameter was 83.0% and for ± 1 size of the true tibial component size was 97.8%. Patient sex was the most influential feature in terms of informing the SGB model predictions for both femoral and tibial component sizing. A TKA implant sizing application was subsequently created.

CONCLUSION

Novel machine learning algorithms demonstrated good to excellent performance for predicting TKA component size. Patient sex appears to contribute an important role in predicting TKA size. A web-based real-time prediction application was created capable of integrating patient specific data to predict TKA size, which will require external validation prior to clinical use.