Clinical value of CT-based patient-specific 3D preoperative design combined with conventional instruments in primary total knee arthroplasty: a propensity score-matched analysis

Who Are the Anatomic Outliers Undergoing Total Knee Arthroplasty? A Computed Tomography-Based Analysis of the Hip-Knee-Ankle Axis Across 1,352 Preoperative Computed Tomographies Using a Deep Learning and Computer Vision-Based Pipeline

BACKGROUND

Dissatisfaction after total knee arthroplasty (TKA) ranges from 15 to 30%. While patient selection may be partially responsible, morphological and reconstructive challenges may be determinants. Preoperative computed tomography (CT) scans for TKA planning allow us to evaluate the hip-knee-ankle axis and establish a baseline phenotypic distribution across anatomic parameters. The purpose of this cross-sectional analysis was to establish the distributions of 27 parameters in a pre-TKA cohort and perform threshold analysis to identify anatomic outliers.

METHODS

There were 1,352 pre-TKA CTs that were processed. A 2-step deep learning pipeline of classification and segmentation models identified landmark images and then generated contour representations. We used an open-source computer vision library to compute measurements for 27 anatomic metrics along the hip-knee axis. Normative distribution plots were established, and thresholds for the 15th percentile at both extremes were calculated. Metrics falling outside the central 70th percentile were considered outlier indices. A threshold analysis of outlier indices against the proportion of the cohort was performed.

RESULTS

Significant variation exists in pre-TKA anatomy across 27 normally distributed metrics. Threshold analysis revealed a sigmoid function with a critical point at 9 outlier indices, representing 31.2% of subjects as anatomic outliers. Metrics with the greatest variation related to deformity (tibiofemoral angle, medial proximal tibial angle, lateral distal femoral angle), bony size (tibial width, anteroposterior femoral size, femoral head size, medial femoral condyle size), intraoperative landmarks (posterior tibial slope, transepicondylar and posterior condylar axes), and neglected rotational considerations (acetabular and femoral version, femoral torsion).

CONCLUSIONS

In the largest non-industry database of pre-TKA CTs using a fully automated 3-stage deep learning and computer vision-based pipeline, marked anatomic variation exists. In the pursuit of understanding the dissatisfaction rate after TKA, acknowledging that 31% of patients represent anatomic outliers may help us better achieve anatomically personalized TKA, with or without adjunctive technology.

Automating Linear and Angular Measurements for the Hip and Knee After Computed Tomography: Validation of a Three-Stage Deep Learning and Computer Vision-Based Pipeline for Pathoanatomic Assessment

Background

Variability in the bony morphology of pathologic hips/knees is a challenge in automating preoperative computed tomography (CT) scan measurements. With the increasing prevalence of CT for advanced preoperative planning, processing this data represents a critical bottleneck in presurgical planning, research, and development. The purpose of this study was to demonstrate a reproducible and scalable methodology for analyzing CT-based anatomy to process hip and knee anatomy for perioperative planning and execution.

Methods

One hundred patients with preoperative CT scans undergoing total knee arthroplasty for osteoarthritis were processed. A two-step deep learning pipeline of classification and segmentation models was developed that identifies landmark images and then generates contour representations. We utilized an open-source computer vision library to compute measurements. Classification models were assessed by accuracy, precision, and recall. Segmentation models were evaluated using dice and mean Intersection over Union (IOU) metrics. Contour measurements were compared against manual measurements to validate posterior condylar axis angle, sulcus angle, trochlear groove-tibial tuberosity distance, acetabular anteversion, and femoral version.

Results

Classifiers identified landmark images with accuracy of 0.91 and 0.88 for hip and knee models, respectively. Segmentation models demonstrated mean IOU scores above 0.95 with the highest dice coefficient of 0.957 [0.954-0.961] (UNet3+) and the highest mean IOU of 0.965 [0.961-0.969] (Attention U-Net). There were no statistically significant differences for the measurements taken automatically vs manually (P > 0.05). Average time for the pipeline to preprocess (48.65 +/- 4.41 sec), classify/retrieve landmark images (8.36 +/- 3.40 sec), segment images (<1 sec), and obtain measurements was 2.58 (+/- 1.92) minutes.

Conclusions

A fully automated three-stage deep learning and computer vision-based pipeline of classification and segmentation models accurately localized, segmented, and measured landmark hip and knee images for patients undergoing total knee arthroplasty. Incorporation of clinical parameters, like patient-reported outcome measures and instability risk, will be important considerations alongside anatomic parameters.

Computed tomography-based patient-specific cutting guides used for positioning of the femoral component of implants during unicompartmental knee arthroplasty: a cadaver study

BACKGROUND

To investigate whether patient-specific instrumentation (PSI) improves the femoral component positioning of implants during unicompartmental knee arthroplasty (UKA) using cadaver bone models.

METHODS

Fifty adult cadaveric femoral bone specimens collected from February 2016-2018, were randomized to receive medial UKA with a PSI guide (n = 25) or conventional instrumentation (CI) (n = 25). Standard anteroposterior and lateral view radiographs were obtained postoperatively to assess the coronal and sagittal positioning of the femoral prostheses, respectively. The osteotomy time was recorded to assess the convenience of PSI in guiding osteotomy.

RESULTS

Osteotomy time significantly shortened in the PSI group (3.12 ± 0.65 versus 4.33 ± 0.73 min, p < 0.001). There was a significant difference in the postoperative coronal alignment of the femoral component between the PSI and CI groups (varus/valgus angle: 1.43 ± 0.93° vs. 2.65 ± 1.50°, p = 0.001). The prevalence of outliers in coronal alignment was lower in the PSI than the CI group (2/25, 8% vs. 9/25, 36%). Sagittal posterior slope angle of the femoral component was significantly different between the two groups (8.80 ± 0.65° and 6.29 ± 1.88° in the CI and PSI groups, respectively, p < 0.001). The malalignment rate of the femoral component in the sagittal plane was 60% in the CI group, whereas no positioning deviation was observed in the PSI group.

CONCLUSION

This study used a cadaver model to support the fact that CT-based PSI shows an advantage over CI in optimizing implant positioning for UKAs.

Development and Validation of an Artificial Intelligence Preoperative Planning and Patient-Specific Instrumentation System for Total Knee Arthroplasty

BACKGROUND

Accurate preoperative planning for total knee arthroplasty (TKA) is crucial. Computed tomography (CT)-based preoperative planning offers more comprehensive information and can also be used to design patient-specific instrumentation (PSI), but it requires well-reconstructed and segmented images, and the process is complex and time-consuming. This study aimed to develop an artificial intelligence (AI) preoperative planning and PSI system for TKA and to validate its time savings and accuracy in clinical applications.

METHODS

The 3D-UNet and modified HRNet neural network structures were used to develop the AI preoperative planning and PSI system (AIJOINT). Forty-two patients who were scheduled for TKA underwent both AI and manual CT processing and planning for component sizing, 20 of whom had their PSIs designed and applied intraoperatively. The time consumed and the size and orientation of the postoperative component were recorded.

RESULTS

The Dice similarity coefficient (DSC) and loss function indicated excellent performance of the neural network structure in CT image segmentation. AIJOINT was faster than conventional methods for CT segmentation (3.74 ± 0.82 vs. 128.88 ± 17.31 min, p < 0.05) and PSI design (35.10 ± 3.98 vs. 159.52 ± 17.14 min, p < 0.05) without increasing the time for size planning. The accuracy of AIJOINT in planning the size of both femoral and tibial components was 92.9%, while the accuracy of the conventional method in planning the size of the femoral and tibial components was 42.9% and 47.6%, respectively (p < 0.05). In addition, AI-based PSI improved the accuracy of the hip-knee-ankle angle and reduced postoperative blood loss (p < 0.05).

CONCLUSION

AIJOINT significantly reduces the time needed for CT processing and PSI design without increasing the time for size planning, accurately predicts the component size, and improves the accuracy of lower limb alignment in TKA patients, providing a meaningful supplement to the application of AI in orthopaedics.

Accuracy and reproducibility analysis of different reference axes for femoral prosthesis rotation alignment in TKA based on 3D CT femoral model

BACKGROUND

There are many reference axes to determine the rotational positioning of the femoral prosthesis in total knee arthroplasty (TKA), mainly including the surgical transepicondylar axis (sTEA), anatomical transepicondylar axis (aTEA), Whiteside line, and the posterior condylar line (PCL), etc., but there is still no definite conclusion on which is the most accurate reference axis.

OBJECTIVE

To explore the reproducibility of each reference axis of femoral external osteotomy based on the 3D CT femoral model, compare the deviation of the simulated femoral prosthesis rotation alignment, positioned based on each reference axis, with the gold standard sTEA, and analyze the accuracy of each reference axis.

METHODS

The imaging data of 120 patients with knee osteoarthritis who underwent a 3D CT examination of the knee in our hospital from June 2018 to December 2021 were retrospectively collected. The 3D model of the femur was established by Mimics software. The line relative to PCL externally rotated 3° (PCL + 3°), aTEA, and the vertical line of the Whiteside line were constructed and compared with the gold standard sTEA. Intra-observer, as well as inter-observer reproducibility analysis, was performed by the intra-group correlation coefficient (ICC) and Bland-Altman method.

RESULTS

The angle ∠WS, between the vertical line of Whiteside and sTEA, was 2.54 ± 2.30°, with an outlier of 54.2%; the angle ∠aTEA, between aTEA and sTEA, was 4.21 ± 1.01°, with an outlier of 99.1%; the angle ∠PCL, between PCL + 3° external rotation and sTEA, was 0.50 ± 1.06°, with the highest accuracy and an outlier of 5.8%, and the differences among all three were statistically significant, P < 0.05. The intra-observer ICC values of ∠WS, ∠aTEA, and ∠PCL were 0.975 (0.964-0.982), 0.926 (0.896-0.948), and 0.924(0.892,0.946), respectively, and the reproducibility levels were excellent; the inter-observer ICC values of ∠WS, ∠aTEA, and ∠PCL were 0.968(0.955-0.978), 0.906 (0.868-0.934) and 0.970 (0.957,0.979), respectively, with excellent reproducibility levels; Bland-Altman plots suggested that the scatter points of intra-observer and inter-observer measurement differences more than 95% were within the limits of agreement.

CONCLUSION

The reference axis for locating the distal femoral external rotation osteotomy based on the 3D CT femoral model has good reproducibility. The PCL is easy to operate, has the highest precision, and the lowest outliers among the reference axes is therefore recommended.

Effect of Tibial Component Misalignment on the Lower Limb Joint Kinematics During Squat

The component orientation of the total knee replacement is critical to surgical outcomes. There have been many studies focused on knee movement for different component rotations. However, the effect of component misalignment on a dynamic movement, especially which requires high knee flexion, is not widely studied. The aim of this study is to investigate the effect of tibial component misalignment on a squat motion by predictive simulation. Squat motions with different replacement component alignments were predicted by formulating an optimal control problem. The result indicates that component misalignment on coronal and horizontal planes reduces peak joint flexion angles and the external rotation on the horizontal plane has the most negative impact. Misalignment in external rotation resulted in the greatest reduction of peak joint flexion angles. The simulation was validated by comparison with experimental data, which showed a high level of correlation with the predicted motion.Clinical relevance- The predictive simulation presented in this study can predict the dynamic post-surgery movement of TKR. It has the potential to help surgeons and clinicians at the preoperative planning stage.

Clinical evaluation of the first semi-active total knee arthroplasty assisting robot made in China: a retrospective propensity score-matched cohort study

OBJECTIVE

The precision of overall alignment and knee morphotype after robot-assisted total knee arthroplasty has been fully confirmed. This study aims to conduct a clinical evaluation of the first China-made semi-active total knee arthroplasty assisting robot.

METHODS

After a 1 : 2 propensity score matching, that is, a matched cohort study, patients were matched to the robot group (52 cases) and the conventional group (104 cases). The robot group received osteotomy according to preoperative planning, while the conventional group adopted preoperative planning based on the full-length radiograph and received conventional osteotomy. Perioperative clinical indicators, such as operation time, tourniquet time, hospitalization days, intraoperative bleeding, and hemoglobin level of the two groups were recorded; radiological indicators of postoperative prosthesis position, including hip-knee-ankle angle, frontal femoral component angle, frontal tibial component angle, lateral femoral component angle, and lateral tibial component angle were also recorded; deviations and outliers of the radiological indicators were calculated.

RESULTS

Compared with the conventional group, the operation time and tourniquet time of the robot group were longer, and the postoperative hemoglobin level decreased less, the differences were statistically significant; the lateral tibial component angle of the conventional group was 80.9°±3.6°, which was smaller than 86.7 °±2.3° of the robot group, the difference was statistically significant ( P <0.001); except for lateral femoral component angle, the absolute deviations of the radiological indicators in the robot group were significantly smaller than that in the conventional group ( P ≤0.001); the outliers of the radiological indicators in the robot group were significantly smaller than that in the conventional group with a statistical difference ( P <0.05).

CONCLUSION

Compared with the conventional group, the operation time of the robot group was relatively longer, but the perioperation blood loss was less. The robot group could better control the posterior inclination of the tibial prosthesis, and the absolute deviations and outliers of the prosthesis position were relatively smaller. There was no difference in short-term clinical score between the two groups.

Advanced 3D Visualization and 3D Printing in Radiology

Since the discovery of X-rays in 1895, medical imaging systems have played a crucial role in medicine by permitting the visualization of internal structures and understanding the function of organ systems. Traditional imaging modalities including Computed Tomography (CT), Magnetic Resonance Imaging (MRI) and Ultrasound (US) present fixed two-dimensional (2D) images which are difficult to conceptualize complex anatomy. Advanced volumetric medical imaging allows for three-dimensional (3D) image post-processing and image segmentation to be performed, enabling the creation of 3D volume renderings and enhanced visualization of pertinent anatomic structures in 3D. Furthermore, 3D imaging is used to generate 3D printed models and extended reality (augmented reality and virtual reality) models. A 3D image translates medical imaging information into a visual story rendering complex data and abstract ideas into an easily understood and tangible concept. Clinicians use 3D models to comprehend complex anatomical structures and to plan and guide surgical interventions more precisely. This chapter will review the volumetric radiological techniques that are commonly utilized for advanced 3D visualization. It will also provide examples of 3D printing and extended reality technology applications in radiology and describe the positive impact of advanced radiological image visualization on patient care.

Surgical Transepicondylar Axis Is Not a Reliable Reference when there Was Lateral Femoral Bowing

OBJECTIVE

The surgical transepicondylar axis (sTEA) is believed to be a consistent reference for femoral rotation axis, and the reliability of its orientation seriously affects the accuracy and outcome of total knee arthroplasty (TKA). This study was designed to investigate the relationship between the orientation of sTEA and femoral bowing angle (FBA) and posterior condylar line (PCL) using three-dimensional (3D) computed tomography (CT) reconstruction models to verify its reliability.

METHODS

This study retrospectively collected lower extremity images of 443 southern Chinese osteoarthritic patients (347 women, 96 men; 234 left, 209 right; mean age 66.5 ± 9.3 years) from August 2016 to June 2018. The hip-knee-ankle angle (HKA) was measured on anteroposterior weight-bearing full lower extremity standing radiographs. Measurements on 3D CT models of the femurs included lateral angle between the femoral mechanical axis and sTEA coronal angle, angle between sTEA and distal joint line (distal condylar axis angle, DCA); angle between sTEA and PCL (sTEA axial angle); angle between anatomical axis of proximal femur and anatomical axis of distal femur in the plane they form (actual FBA) and its projection on the coronal (lateral FBA) and sagittal (anterior FBA) planes. The correlations between sTEA coronal angle, sTEA axial angle and actual FBA, lateral FBA, anterior FBA, HKA, DCA were explored using the Spearman correlation test.

RESULTS

The mean value of actual FBA is (14.4 ± 3.6)°, of lateral FBA is (6.0 ± 4.0)°, and of anterior FBA is (12.7 ± 3.0)°. The mean value of sTEA coronal angle is (88.7 ± 3.6)°, of sTEA axial angle is (2.1 ± 2.8)°. The sTEA coronal angle was positively correlated with actual FBA (r = 0.320, P < 0.01), lateral FBA (r = 0.448, P < 0.01), anterior FBA (r = 0.113, P < 0.05), HKA (r = 0.482, P < 0.01) and DCA (r = 0.486, P < 0.01). The sTEA axial angle was positively correlated with DCA (r = 0.168, P < 0.01), but not significantly correlated with all FBAs and HKA (NS).

CONCLUSION

The sTEA cannot be used as a stable reference when there was an obvious femoral bowing deformity. As the lateral femoral bowing increases, the orientation of sTEA becomes more varus, no matter the knee is varus or valgus.

Could surgical transepicondylar axis be identified accurately in preoperative 3D planning for total knee arthroplasty? A reproducibility study based on 3D-CT

BACKGROUND

Surgical transepicondylar axis (sTEA) is frequently used for positioning of femoral component rotation in total knee arthroplasty (TKA). Previous studies showed that intraoperative identification of sTEA was not reliable. While surgeons or engineers need to identify sTEA with three-dimensional (3D) computer-aid techniques pre- or intraoperatively, the reproducibility of sTEA identification on preoperative 3D images has not been explored yet. This study aimed to investigate the reproducibility of identifying sTEA in preoperative planning based on computed tomography (CT).

METHODS

Fifty-nine consecutive patients (60 knees involved) who received TKA in our center from April 2019 to June 2019 were included in this study. Six experienced TKA surgeons identified sTEA three times on 3D model established on the basis of knee CT data. The projection angle of each sTEA and the posterior condyle axis on the transverse plane were measured and analyzed.

RESULTS

The overall intra-observer reproducibility was moderate. The median intra-observer variation was 1.27°, with a maximum being up to 14.07°. The median inter-observer variation was 1.24°, and the maximum was 11.47°. The overall intra-class correlation coefficient (ICC) for inter-observer was 0.528 (95% CI 0.417, 0.643).

CONCLUSION

The identification of sTEA on a 3D model established on the basis of knee CT data may not be reliable. Combined with the previous cadaveric and surgical studies, caution should be exercised in determining femoral component rotation by referencing sTEA both preoperatively and intraoperatively.

LEVEL OF EVIDENCE

III.

Patient-specific instrumentation makes sense in total knee arthroplasty

INTRODUCTION

Patient-specific instrumentation (PSI) for total knee arthroplasty (TKA) surgery was initially developed to increase accuracy. The potential PSI benefits have expanded in the last decade, and other advantages have been published. However, different authors are critical of PSI and argue that the advantages are not such and do not compensate for the extra cost. This article aims to describe the recently published advantages and disadvantages of PSI.

AREAS COVERED

Narrative description of the latest publications related to PSI in accuracy, clinical and functional outcomes, operative time, efficiency, and other benefits.

EXPERT COMMENTARY

We have published high accuracy of the system, with a not clinically relevant loss of accuracy, significantly higher precision with PSI than with conventional instruments, and a high percentage of cases in the optimal range and similar to that obtained with computer-assisted navigation, greater imprecision for tibial slope, a significant blood loss reduction, and time consumption, an acceptable and non-significant increase in the cost per procedure and no difference in complications during hospital admission and at 90 days. We think that PSI will not follow the Scott Parabola and that it will continue to be a valuable type of device in some instances of TKA surgery.

Robotics versus personalized 3D preoperative planning in total knee arthroplasty: a propensity score-matched analysis

Purpose

Lower limb alignment is crucial in total knee arthroplasty (TKA). Previous studies have shown that robotics and personalized three-dimensional (3D) preoperative planning could improve postoperative alignment accuracy compared with conventional TKA, but comparison between the above two techniques has never been reported. The authors hypothesized that robotics may be superior to personalized 3D preoperative planning in terms of postoperative alignment in primary TKA, with similar patient-reported outcome measures (PROMs) but higher cost and longer operative time.

Methods

A consecutive series of patients who received TKA in our center from September 2020 to January 2021 were enrolled retrospectively. After 1:2 matching, 52 and 104 patients were included and divided into study group for robotics and control group for personalized 3D preoperative planning, respectively. Multiple postoperative alignment angles were measured, and clinical features such as operation and tourniquet time, length of hospital stay and hemoglobin (Hb) were recorded. Knee Society Score (KSS) and Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC) were used to evaluate clinical results.

Results

Compared with control group, robotics group had significantly lower frontal femoral component angle (FFC) and frontal tibial component angle (FTC) absolute deviation (P < 0.05). It also had less outliers in hip–knee–ankle angle (HKA), FTC, lateral femoral component angle (LFC) and lateral tibial component angle (LTC) (P < 0.05). Hb loss of robotics group was significantly lower than control group (P < 0.001), while the operation and tourniquet time were longer (P < 0.001). There was no significant difference in KSS and WOMAC scores between two groups.

Conclusion

Compared with control group, patients in robotics group had significantly less malalignment, malposition, Hb loss, but similar PROMs. The operations in robotics group spent longer operation time and cost more compared with control group.

Trial registration: The Chinese Clinical Trial Registry, ChiCTR2000036235. Registered 22 August 2020, http://www.chictr.org.cn/showproj.aspx?proj=59300.

Level of evidence

III.

Slight femoral under-correction versus neutral alignment in total knee arthroplasty with preoperative varus knees: a comparative study

Background

This study aimed to compare the short-term clinical results of slight femoral under-correction with neutral alignment in patients with preoperative varus knees who underwent total knee arthroplasty.

Methods

The medical records and imaging data were retrospectively collected from patients who had undergone total knee arthroplasty in our hospital from January 2016 to June 2019. All patients had varus knees preoperatively. Upon 1:1 propensity score matching, 256 patients (256 knees) were chosen and divided into a neutral alignment group (n=128) and an under-correction group (n=128). The patients in the neutral group were treated with the neutral alignment. In the under-correction group, the femoral mechanical axis had a 2° under-correction. The operative time, tourniquet time and the length of hospital stay in the two groups were recorded. The postoperative hip-knee-ankle angle, frontal femoral component angle and frontal tibial component angle were measured. Patient-reported outcome measures were also compared.

Results

The operative time, tourniquet time and the length of hospital stay in the under-correction group were significantly shorter than the neutral alignment group (P<0.05). At the 2-year follow-up, the under-correction group had a larger varus alignment (P<0.05) and a larger frontal femoral component angle (P<0.05), and the frontal tibial component angles of the two groups were comparable. Compared with the neutral alignment group, the slight femoral under-correction group had significantly better patient-reported outcome measures scores (P<0.05).

Conclusion

For varus knees treated with total knee arthroplasty, alignment with a slight femoral under-correction has advantages over the neutral alignment in terms of the shorter operative time and better short-term clinical results.

Level of evidence

III