Accuracy of digital templating of uncemented total hip arthroplasty at a certified arthroplasty center: a retrospective comparative study

Can machine learning predict the accuracy of preoperative planning for total hip arthroplasty, basing on patient-related factors? An explorative investigation on Supervised machine learning classification models

Background

The success of Total Hip Arthroplasty (THA) is influenced by preoperative planning, with traditional 2D approaches displaying varied reliability as well. The present study investigates the use of Supervised Machine Learning (SML) models with patient-related features to improve accuracy.

Methods

Preoperative and perioperative data, as well as planning and final implant information, were obtained from 800 consecutive cementless primary THA, which was performed uniformly by a specialized surgical team. Six Supervised Machine Learning models were trained and validated using patient characteristics and implant data: Logistic Regression (LR), Linear Discriminant Analysis (LDA), K-Nearest Neighbors (KNN), Decision Tree (CART), Gaussian Naive Bayes (GN), and Support Vector Classifier (SVC). The models' ability to predict planning reliability and leg length disparity was evaluated.

Results

KNN performed better on the cup model (97.9 %), femur model (96.7 %), and femur size (99.2 %). SVM emerged as the model with the highest accuracy for cup size (60.4 %) and head size (62.1 %). CART had the best accuracy (99 %) when determining leg length discrepancy.

Conclusion

The study demonstrates the utility of Supervised Machine Learning models, specifically KNN, in predicting the accuracy of preoperative planning in THA. The accuracy of these models, which are driven by patient-related characteristics, provides useful information for optimizing patients' selection and improving surgical outcome.

Coxa valga and antetorta configuration leads to underestimation of the femoral component size: a matched case-control study of patients undergoing cementless total hip arthroplasty

BACKGROUND

Total hip arthroplasty (THA) is the gold standard procedure for patients with end-stage osteoarthritis after failed conservative therapy. Digital templating is commonly employed in preoperative preparation for THA and contributes positively to its outcome. However, the impact of coxa valga and antetorta (CVA) configurations on stem size prediction accuracy remains not reported. Previous studies demonstrated that the size of the lesser trochanter (LT) can be used to determine femoral anteversion on pelvis radiographs. This study investigates the accuracy of preoperative digital templating in predicting stem size in patients with CVA undergoing cementless THA.

METHODS

Preoperative radiographs of 620 patients undergoing cementless THA were retrospectively investigated. Radiographs were standardized with patients standing and the leg internally rotated by 15°. A CVA group was established including patients with a CCD angle greater than 140° and a lesser trochanter (LT) size of at least 10 mm for men and 8 mm for women. For the control group, radiographs with a CCD angle ranging from 125-135° and LT size 3-10 mm for men and 3-8 mm for women were selected. Preoperative templating was performed using mediCAD. To reduce confounding factors, case-control matching was carried out for BMI and body height.

RESULTS

After case-control matching, a total of thirty-one matches were analyzed. Stem size was underestimated in 74% (23/31) in the CVA and 13% (4/31) in the control group (p < 0.001). Moreover, patients with CVA were more likely to be underestimated by two sizes compared to controls (p < 0.004). In contrast, the exact stem size was predicted more frequently in the control group (p < 0.001).

CONCLUSION

Stem size in patients with a CVA configuration are at high risk of being underestimated when using digital templating. These findings can be valuable for guiding in intraoperative decisions and lowering the risk of complications associated with an undersized femoral component.

Junior surgeon learning curve for performing digital planning of a first-line uncemented total hip prothesis

PURPOSE

Preoperative digital planning of total hip arthroplasty (THA) anticipates difficulties while increasing implant survival. The objective was to establish the learning curve to produce a reliable and reproducible preoperative planning for THA. We hypothesize that a learning curve exists for planning, and we want to determine the number of procedures required to accomplish it.

METHODS

This prospective study included patients for THA from 02/11/2019 to 01/11/2022. Ten junior (Juniors) and senior surgeons (Seniors) had received dedicated training in the use of the software. Modeling was done blindly by Juniors and Senior before surgery on a standardized front pelvis X-ray (mediCAD 2D Classic Hospital software). Statistical analyses to establish the learning curve were done to compare the Juniors and Seniors.

RESULTS

60% of the Juniors achieved competence after 31.5 ± 12.9 [14-54] planning sessions for the acetabular implant, and 80% after 30.3 ± 8.3 [17-40] planning sessions for the femoral implant. Femoral neck size was achieved by all ten Juniors after 23.1 ± 6.8 [17-38] planning. The offset was correctly restored on the plan by 30% of the Juniors after 33.5 ± 11.6 [18-46] planning.

CONCLUSION

There is a learning curve for 2D planning of uncemented THA. The different planning items seem to have different learning curves. Compared to Seniors, the completion of 75 planning sessions is not sufficient in totality. The competence of the Juniors for the acetabular implant, the length of the neck and the size of the femoral stem are mostly acquired before 75 sessions.

LEVEL OF EVIDENCE

Prospective study-Level II.

Restoration of hip anatomy using computer modelling prior to total hip arthroplasty and its potential consequences in terms of lower limb-length difference

BACKGROUND

Restoring hip anatomy during total hip arthroplasty (THA) is the gold standard. The general goal is for hips to be symmetrical in terms of height. This study aimed to analyse the effect of simulating hip length (HL) equalisation on leg-length difference (LLD) before and after THA with a computerised plan.

METHODS

141 consecutive patients were enrolled following THA. Outcomes of interest were preoperative hip-length difference (HLD) and LLD as determined on CT and final LLD as determined by simulating HL equalisation. We divided patients into 3 groups: no LLD, LLD >5 mm and LLD ⩾10 mm.

RESULTS

The rate of preoperative LLD >5 mm and LLD ⩾10 mm was 37.5% and 14.8%, respectively. HL equalisation did not change LLD distribution in the overall cohort. Among patients with smaller preoperative HLs and LLDs, HL equalisation changed the rate of LLD >5 mm from 100% (n = 24) to 41.7% (n = 10) (p < 0.0001) and of LLD ⩾10 mm from 100% (n = 12) to 16.7% (n = 2) (p < 0.0001). Among patients with no preoperative LLD, HL equalisation changed the rate of LLD ⩽5 mm from 100% (n = 64) to 59.3% (n = 38) (p < 0.0001) and of LLD<10 mm from 100% (n = 76) to 89.5% (n = 68) (p = 0.006).

CONCLUSIONS

Restoring biomechanical hip anatomy by HL equalisation may not be the correct goal for all patients. In patients with no LLD and shorter HLs, equalisation could result in LLD >5 mm in 40% of patients and LLD >10 mm in 10%, demonstrating the necessity to further analyse individuals and propose a personalised stem position.

The Reliability of Preoperative Total Knee Arthroplasty Planning -a Comparison of Two Calibration Markers

BACKGROUND

Preoperative templating is the gold standard in planning elective total knee arthroplasty. This retrospective study analyses preoperative templating by using two different calibration devices.

METHODS

Preoperative radiographic templating with TraumaCad using two different calibration devices was performed in 231 patients.

RESULTS

Overall, 60.6% (n = 140) did not match and 39.4% (n = 91) matched for tibial as well as femoral size. Group I showed significantly more combined matches as well as for just the size of the femoral component.

CONCLUSIONS

In this study, the KingMark calibration should be preferred to standardised ball markers.

THA-Net: A Deep Learning Solution for Next-Generation Templating and Patient-specific Surgical Execution

BACKGROUND

This study introduces THA-Net, a deep learning inpainting algorithm for simulating postoperative total hip arthroplasty (THA) radiographs from a single preoperative pelvis radiograph input, while being able to generate predictions either unconditionally (algorithm chooses implants) or conditionally (surgeon chooses implants).

METHODS

The THA-Net is a deep learning algorithm which receives an input preoperative radiograph and subsequently replaces the target hip joint with THA implants to generate a synthetic yet realistic postoperative radiograph. We trained THA-Net on 356,305 pairs of radiographs from 14,357 patients from a single institution's total joint registry and evaluated the validity (quality of surgical execution) and realism (ability to differentiate real and synthetic radiographs) of its outputs against both human-based and software-based criteria.

RESULTS

The surgical validity of synthetic postoperative radiographs was significantly higher than their real counterparts (mean difference: 0.8 to 1.1 points on 10-point Likert scale, P < .001), but they were not able to be differentiated in terms of realism in blinded expert review. Synthetic images showed excellent validity and realism when analyzed with already validated deep learning models.

CONCLUSION

We developed a THA next-generation templating tool that can generate synthetic radiographs graded higher on ultimate surgical execution than real radiographs from training data. Further refinement of this tool may potentiate patient-specific surgical planning and enable technologies such as robotics, navigation, and augmented reality (an online demo of THA-Net is available at: https://demo.osail.ai/tha_net).

Gender differences in cementless short stem total hip arthroplasty: significantly higher femoral lengthening in female patients

Modern cementless short stems in total hip arthroplasty (THA) enable a precise reconstruction of the native pre-arthritic hip geometry. While gender differences have been reported for older generation straight hip stems, there are hardly reports regarding modern cementless short hip stems. Therefore, we aimed to evaluate the influence of gender differences in hip anatomy in cementless short stem THA. A total of 207 patients (109 females, 98 males) with unilateral THA and absence of contralateral joint space narrowing (Kellgren-Lawrence grade ≤ 2) were included. Acetabular height and offset as well as femoral height and offset were measured on pre- and 3-months-postoperative anteroposterior X-rays of the pelvis and compared to the contralateral hip. Additionally, implant position was evaluated on the postoperative radiograph. In male patients, the loss of acetabular offset was significantly greater than in females (p = 0.012), leading to a compensatory increase in femoral offset (p = 0.041). Femoral height discrepancy was significantly higher in females (p < 0.001), accounting for an increased global hip height discrepancy (p < 0.001). The mean acetabular anteversion was significantly greater in female patients (p < 0.001). Female patients are at higher risk of femoral lengthening in THA with a cementless short stem potentially caused by a further proximally conducted femoral neck osteotomy and show significantly higher cup anteversion angles. Therefore, surgeons should take special care to the level of neck resection and implant positioning in female patients.

Risk factors for templating mismatch of uncemented stems in bipolar hemiarthroplasty for femoral neck fracture

Preoperative templating needs to be precise to optimize hip arthroplasty outcomes. Unexpected implant mismatches can occur despite meticulous planning. We investigated the risk factors for oversized and undersized stem mismatch during uncemented hemiarthroplasty using a double-tapered wedge rectangular stem for femoral neck fracture. Out of 154 consecutive patients who underwent hemiarthroplasty for femoral neck fracture, 104 patients were divided into three groups: (1) oversized (n = 17; 16.3%), (2) matched (n = 80; 76.9%), and (3) undersized stem group (n = 7; 6.7%). A smaller femoral head offset (odds ratio [OR] = 0.89, 95% confidence interval [95% CI] = 0.81-0.98, P = 0.017), smaller isthmus diameter (OR = 0.57, 95% CI = 0.35-0.92, P = 0.021), and smaller canal flare index (OR = 0.20, 95% CI = 0.04-0.98, P = 0.047) were significantly associated with oversized stem insertion, while older age (OR = 1.18, 95% CI = 1.01-1.39, P = 0.037) was associated with undersized stem insertion in logistic regression. In conclusion, when performing hemiarthroplasty for a femoral neck fracture with a double-tapered wedge rectangular stem, surgeons must pay close attention to proximal femoral geometry and patient age during preoperative planning to avoid stem mismatch.

Free Hip Arthroplasty Templating Software - Does it Work?

Background

Preoperative planning is important for successful total hip arthroplasty (THA) and has been historically performed using acetate templates. Digital software templating has been adopted for evaluating implant size, position, and alignment. Commercial software can be expensive, but free programs exist. Detroit Bone Setter (detroitbonesetter.com, Detroit, MI) is a freely available templating program, but hasn't been validated. Our study reports this program's accuracy for templating THA.

Methods

Sixty-five patients undergoing THA between 2017 and 2022 at 2 hospitals were included. All cases were templated by the senior author or orthopaedic trauma fellow prospectively or retrospectively in a blinded fashion. Direct anterior or posterior approaches were used based on attending surgeon's preference. A student's t-test was used to compare means of templated vs actual implant sizes of femoral and acetabular components.

Results

There was no significant difference between implanted (mean [M] = 6.4, standard deviation [SD] = 2.0) and templated femoral component sizes (M = 5.7, SD = 2.1). There was a significant difference between implanted (M = 57.0, SD = 3.9) and templated acetabular component sizes (M = 53.4, SD = 3.0). Bland-Altman testing demonstrated femoral components with positive measurement bias of 0.62, indicating slight overestimation of implant size. Acetabular component size was overestimated with positive measurement bias of 3.6 mm.

Conclusions

Detroit Bone Setter is advantageous as it is freely available and supports most major company implants. It accurately templated femoral component size but consistently overestimated acetabular component size by 3.6 mm. Further studies are needed prior to recommending its routine use for templating THA when other validated methods exist. It could be used with caution when no other methods are available.

Overall Accuracy of Radiological Digital Planning for Total Hip Arthroplasty in a Specialized Orthopaedics Hospital

Preoperative radiological planning is a key factor in the prediction of implant size and positioning that influences surgical time, the risk of complications, and functional outcomes. We have tested the accuracy of the digital templating performed in our hospital for a sample of 215 patients that underwent total hip arthroplasty. We assessed the accuracy of correctly predicting implant size for the femoral and acetabular components, as well as the stem neck length. We found that our method of templating proved accurate (within one size) in 95.8% of cases for the stem and 94.9% for the cup when using the anteroposterior view only, while the lateral view was accurate in 95.8% of cases for the stem and 97.2% for the cup. Exact prediction of the stem size was obtained in 77.7% of cases using the anteroposterior view and 67.0% of cases on the lateral view, and 73.0% and 74.4% of cases for the cup on the AP and LL views, respectively. Stem neck size was predicted exactly in 75.35% of cases and within one size in 93.49% of cases. We concluded that our method of digital templating using dedicated software is highly effective in accurately predicting implant size.

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.

[Application of artificial intelligence preoperative planning system in total hip arthroplasty for adult developmental dysplasia of the hip]

Objective

By comparing with the traditional X-ray template measurement method, to explore the accuracy of artificial intelligence preoperative planning system (AI-HIP) to predict the type of prosthesis and guide the placement of prosthesis before total hip arthroplasty (THA) in adult patients with developmental dysplasia of the hip (DDH).

Methods

Patients with DDH scheduled for initial THA between August 2020 and August 2022 were enrolled as study object, of which 28 cases (28 hips) met the selection criteria were enrolled in the study. Among them, there were 10 males and 18 females, aged from 34 to 77 years, with an average of 59.3 years. There were 12 cases of the left DDH and 16 cases of the right DDH. According to DDH classification, there were 10 cases of Crowe type Ⅰ, 8 cases of type Ⅱ, 5 cases of type Ⅲ, and 5 cases of type Ⅳ. According to Association Research Circulation Osseous (ARCO) staging of osteonecrosis of the femoral head, 13 cases were in stage Ⅲ and 15 cases in stage Ⅳ. The disease duration was 2.5-23.0 years (mean, 8.6 years). The limb length discrepancy (LLD) was 11.0 (8.0, 17.5) mm. Before operation, the prosthesis types of all patients were predicted by AI-HIP system and X-ray template measurement method, respectively. And the preoperative results were compared with the actual prosthesis type during operation in order to estimate the accuracy of the AI-HIP system. Then, the differences in the acetabular abduction angle, acetabular anteversion angle, femoral neck osteotomy position, tip-shoulder distance, and LLD were compared between preoperative planned measurements by AI-HIP system and actual measurement results after operation, in order to investigate the ability of AI-HIP system to evaluate the placement position of prosthesis.

Results

The types of acetabular and femoral prostheses predicted based on AI-HIP system before operation were consistent with the actual prostheses in 23 cases (82.1%) and 24 cases (85.7%), respectively. The types of acetabular and femoral prostheses predicted based on X-ray template measurement before operation were consistent with the actual prostheses in 16 cases (57.1%) and 17 cases (60.7%), respectively. There were significant differences between AI-HIP system and X-ray template measurement (P<0.05). There was no significant difference in acetabular abduction angle, acetabular anteversion angle, femoral neck osteotomy position, and tip-shoulder distance between AI-HIP system and actual measurement after operation (P>0.05). LLD after operation was significantly lower than that before operation (P<0.05). There was no significant difference between the LLD predicted based on AI-HIP system and the actual measurement after operation (P>0.05).

Conclusion

Compared with the traditional X-ray template measurement method, the preoperative planning of AI-HIP system has better accuracy and repeatability in predicting the prosthesis type. It has a certain reference for the prosthesis placement of adult DDH.

Pre-operative templating in THA using a short stem system: precision and accuracy of 2D versus 3D planning method

Background

Total hip arthroplasty (THA) is the most successful orthopaedic surgery of the past century. The current study aimed to compare the accuracy of digital planning using 2D versus 3D templating.

Materials and methods

Ninety-five THAs in 90 patients were included in the current study. Pre- and post-operative X-rays (in two planes) and low-dose rotation computed tomography scans from hip to foot were performed. Paired t-test and regression analyses were conducted to compare 2D and 3D templating accuracy of the definitive implant.

Results

Cup size planned both with 2D (p < 0.0001) and 3D (p = 0.012) templating was significantly different from the definitively used cup size. The difference between the 2D-planned and implanted stem size (p < 0.0001) was statistically significant. In contrast, there were no significant differences in the 3D-planned and implanted stem size (p = 0.181). Three-dimensional templating showed significantly higher accuracy than 2D templating in terms of cup size (1.1 ± 1.4 versus 1.7 ± 1.8; p = 0.007) and stem size (0.3 ± 0.6 versus 0.7 ± 0.7; p < 0.0001).

With increasing body mass index (BMI), 2D templating of the stem became more inaccurate (p = 0.041). Remarkably, 3D templating remained accurate for all components (stem, p = 0.533; cup, p = 0.479) despite increasing BMI.

Conclusion

Despite extended planning time and increased exposure to radiation, 3D-based planning showed higher accuracy than 2D templating, especially in obese patients. On the basis of our results, we believe that 3D-based pre-operative planning in THA is justifiable and beneficial in patients with increased BMI.

Level of Evidence

III.

Minimally Invasive Total Hip Arthroplasty: A Comparison of Restoring Hip Biomechanics With and Without a Traction Table

BACKGROUND/AIM

The aim of the study was to analyze operative time and restoration of hip biomechanics in total hip arthroplasty (THA) via direct anterior approach (DAA) with and without the use of a traction table.

PATIENTS AND METHODS

We retrospectively compared 97 cases where a traction table was used to 92 cases without a table. Ninety-seven patients received THA with a traction table (AMIS® technique) and 92 patients with conventional DAA. Postoperative standard radiographs were used to analyze offset parameters and leg length. Furthermore, time for patient positioning and cut-to-suture time were evaluated.

RESULTS

Cut-to-suture time was statistically significantly shorter in the traction table group (p=0.001), whereas analysis of offset parameters (acetabular, femoral and combined) was comparable between the two groups (p=0.31, p=0.95, p=0.42). Postoperative leg length was statistically significantly different with and without traction table use (p=0.02).

CONCLUSION

Both methods enable restoration of hip biomechanics with high accuracy. Further studies with prospective study designs and larger sample sizes may be needed to confirm these results.