ABSTRACT. Sacroiliac (SI) joint dysfunction is responsible for 15–30% of non-radicular lower back pain cases. When conservative treatment fails, minimally invasive sacroiliac joint fusion (SIJF) is often performed using three triangular implants to stabilize the joint. Accurate implant placement is critical but challenging under conventional 2D fluoroscopy due to anatomical variability and limited spatial information. Navigation systems can improve accuracy but are costly and not standard available in peripheral hospitals. Patient-specific surgical guides (PSGs) offer an alternative; however, previous designs required bone exposure, increasing invasiveness and soft tissue damage. This study evaluated the clinical feasibility of a minimally invasive PSG with radiopaque markers that can be aligned with the patient’s anatomy using 2D fluoroscopy to achieve proper positioning. Four patients underwent primary SIJF using the lateral transiliac approach with triangular titanium implants in a standard parallel configuration. Implant trajectories were planned preoperatively and translated to the PSG design for intraoperative execution. Postoperative CT scans were analyzed for positional and angular deviations. Twelve implants were placed successfully using the PSG with a mean 3D positional deviation of 6.9 ± 3.1 mm and angular deviation of 2.5 ± 1.2°. All implants were intraosseous (Grade 0), and no surgical complications occurred. Although the PSG enabled safe implant placement, the observed positional deviations indicate that the workflow lacks robustness for broader clinical adoption, since the accuracy was inferior to the use of virtual surgical planning alone.

ABSTRACT. Hip BMD is commonly measured unilaterally in clinical practice; however, scoliosis-related asymmetry may influence side-to-side BMD differences in patients with ASD. Evidence regarding the magnitude and clinical relevance of such differences in surgically treated ASD patients remains limited. Twenty-nine female patients with ASD undergoing long-segment spinal fusion and 29 age-matched female controls were included. Femoral neck BMD was measured bilaterally using CT-based areal BMD (CT-aBMD) with phantom calibration. Side-to-side correlations and absolute percentage differences were analyzed within and between groups. Non-inferiority was assessed using a predefined margin of 2.5%. The relationship between scoliosis parameters and side-to-side BMD differences was also evaluated. The median absolute side-to-side percentage BMD difference was 2.9% in the ASD group and 4.7% in the control group, with no significant between-group difference. Non-inferiority of side-to-side BMD differences in the ASD group was confirmed, as the upper bound of the one-sided 95% confidence interval did not exceed the predefined margin. No scoliosis-related radiographic parameters were significantly associated with side-to-side BMD differences. In conclusion, side-to-side femoral neck BMD differences in patients with ASD were comparable to those in age-matched controls and were not influenced by scoliosis severity. These findings support the clinical acceptability of unilateral hip BMD

ABSTRACT. Accurate acetabular cup placement is critical for preventing dislocation, polyethylene wear, impingement, and for achieving favorable long-term outcomes after total hip arthroplasty (THA). Although navigation and augmented reality (AR) technologies have improved the reproducibility of cup positioning, the effects of cup geometry and fixation method on navigation accuracy have not been sufficiently investigated. The purpose of this study was to compare navigation accuracy among three types of acetabular cups: streamlined cemented dual-mobility (DM) cups, hemispherical cemented cups, and hemispherical cementless cups, using postoperative radiographs as the reference. We retrospectively reviewed THA cases performed with navigation between 2013 and 2025 and classified them into three groups according to cup type. Radiographic inclination (RI) and radiographic anteversion (RA) were measured on postoperative anteroposterior radiographs. Absolute differences between postoperative radiographic angles and intraoperative navigation angles, as well as between postoperative radiographic angles and preoperative target angles, were calculated for RI and RA and compared among the three groups. RA error between postoperative radiographs and intraoperative angles was greatest in the hemispherical cementless cup group and significantly smaller in both the hemispherical cemented and streamlined cemented DM groups. In contrast, no significant intergroup differences were found in RI error or in the differences between postoperative radiographic angles and preoperative target angles for either RA or RI. These findings suggest that the reproducibility of intraoperative anteversion is more strongly influenced by the fixation method than by cup geometry. Despite their non-hemispherical shape, streamlined cemented DM cups demonstrated stable navigation accuracy when used with navigation systems.

ABSTRACT. Purpose The enhanced workflow and express workflow provided by MAKO robot-assisted artificial hip arthroplasty (THA) show differences in the surgical stage and the scope of data utilization. The purpose of this abstract is to compare and organize the structural differences and clinical application characteristics of the two workflows

Methods Total 49 patients which were programmed by enhanced workflow were enrolled, the rest of 150 patients who were operated on by express workflow were analyzed for comparison. The cup version, stem position, LLD and offset were compared and complications such as dislocation were analyzed for clinical outcomes. Preoperative factors including age, sex, BMI, underlying disease, Koval grade were analyzed. Intraoperatively, ASA, operation time, EBL, transfusion, ICU transfer were analyzed, For the MAKO data, preoperative LLD, postoperative LLD, intraoperative combined offset, anteversion and inclinations were analyzed. Postoperatively, postoperative LLD and anteversion and inclination were analyzed, postoperatively mHHS and complication were analyzed.

Results Operation time was longer (81.59 minutes ± 35.26, p-value: 0.045) than express workflow (74.36 ±15.08). Estimated blood loss was higher (210.20 ㎖ ±192.49, p-value: 0.043) than express workflow (163.19㎖ ±118.17) Postoperatively, anteversion was bigger (34.74˚±48.03, p-value: 0.01) than express workflow (22.44±17.58). Intraoperatively, MAKO anteversion was 25.81˚ (±6.11) in enhanced workflow and 23.06˚ (±4.95, p value: 0.002). Two dislocations had occurred in express workflow. There is no surgical complication in enhanced workflow.

Conclusions The enhanced workflow in MAKO THA showed better surgical outcomes about the cup instability than express workflow. But operation time and estimated blood loss might be a burden to the patients. If the femoral version was lower than normal, enhanced workflow would be more benefit than express workflow in MAKO THA.

ABSTRACT. Reverse total shoulder arthroplasty (rTSA) is increasingly reliant on computer-assisted orthopedic surgery (CAOS) systems to improve glenoid implant positioning and reduce complications. However, the learning curve associated with adopting these technologies remains poorly quantified, limiting their efficient integration into clinical practice. This study aimed to evaluate the number of cases required for surgeons to achieve proficiency with a CAOS system in rTSA and to quantify the time saved between the learning and proficiency phases.

We conducted a retrospective analysis of a multicenter, prospectively collected dataset from the Exactech GPS navigation system, spanning from 2016 to 2025. The study included 8,600 rTSA cases performed by 172 surgeons from 10 countries. Navigation time, automatically recorded by the system, served as the primary outcome. Cumulative Sum Control Chart (CUSUM) analysis, combined with an automated breakpoint detection algorithm, was employed to identify transitions between the learning, plateau, and proficiency phases for each surgeon.

Results demonstrated that surgeons required a median of 10 cases (interquartile range: 5–36) to exit the learning phase and 25 cases (interquartile range: 10–45) to reach proficiency. The transition to proficiency was associated with a significant 21.6% reduction in navigation time, corresponding to a median time savings of 4 minutes and 55 seconds per procedure (from 22 minutes and 43 seconds during the learning phase to 17 minutes and 48 seconds during proficiency, p < 0.0001). These findings align with previous smaller-scale studies but provide robust, large-scale validation of the learning curve for CAOS in rTSA.

The strengths of this study include its unprecedented sample size and the automated, objective analysis of learning curves across a diverse, international cohort of surgeons. However, limitations include the focus on a single navigation system and the exclusion of overall surgical time, including system setup and tracker placement.

This study provides critical evidence-based benchmarks for training programs, enabling more efficient adoption of CAOS technology in rTSA. By quantifying the learning curve, hospitals and surgical teams can better plan resource allocation, optimize training protocols, and ultimately improve patient outcomes through more precise and consistent implant placement.

ABSTRACT. Use of Custom 3D-Printed Titanium Implants for Partial Joint Resurfacing Following Traumatic Injuries: A Preliminary Report Background: Complex intra-articular fractures with severe comminution, bone loss, and cartilage destruction remain a major challenge in orthopaedic trauma surgery. When anatomical reconstruction is not feasible, salvage procedures such as arthrodesis or arthroplasty are often considered. In young and active patients, these options may lead to significant functional limitations. Advances in additive manufacturing now allow the production of patient-specific implants and offer new possibilities for joint preservation. Methods: Four patients with focal post-traumatic joint defects were treated using custom 3D-printed titanium partial articular resurfacing implants. Injuries involved the talar body, medial tibial plateau, lateral femoral condyle, and capitellum following high-energy trauma or penetrating war injuries. High-resolution bilateral CT scans were obtained, and the contralateral healthy anatomy was mirrored and overlaid onto the injured side to guide implant design. Custom polished titanium implants were manufactured with preplanned screw fixation. Intraoperatively, defect beds were debrided to viable bone and implants were fixed with titanium screws. Early range-of-motion rehabilitation was initiated, with protected weight-bearing as indicated. Results: Anatomical restoration of the articular surface was achieved in all cases. At a minimum one-year follow-up, all patients demonstrated near-full, painless range of motion. No radiographic evidence of implant loosening, failure, or progression of post-traumatic osteoarthritis was observed. Conclusion: Custom 3D-printed titanium partial articular resurfacing may represent a promising joint-preserving alternative for selected patients with complex post-traumatic focal joint defects. This technique may restore joint congruity and biomechanics while delaying or avoiding salvage procedures. Larger series with longer follow-up are required to validate long-term outcome

ABSTRACT. Posttraumatic malunions can lead to pain, functional impairment, and osteoarthritis. Corrective osteotomy aims to restore anatomy, but multi-plane corrections are technically demanding. Patient-specific instruments (PSIs), particularly 3D-printed guides, are designed to simplify these procedures by fitting precisely onto bone and standardizing osteotomy planes. This study examined whether general surgical experience or PSI-specific experience influences osteotomy accuracy.

75 long bone corrective osteotomies were performed on four cadaveric specimens by 24 orthopedic and trauma surgeons using PSIs. Depending on correction type, one or two cuts were made per case, totaling 107 saw cuts. Postoperative CT scans compared planned versus actual osteotomy entry points and angles. Accuracy was quantified as deviations in entry point and angle. Surgeons reported overall experience (years) and PSI experience (cases/year).

General experience ranged from 0-25 years (median 9), PSI usage from 0-20/year (median 2). Multiple linear regression showed PSI experience significantly improved accuracy (p = 0.006), while general experience did not (p = 0.46). Accuracy varied by anatomical site, with midshaft femur osteotomies being least accurate (p = 0.003).

Experience with PSIs, rather than general surgical experience, significantly enhances corrective osteotomy accuracy, and highlights the need for training and repeated use.

ABSTRACT. Conventional osteosynthesis plates for tibial plateau fractures often fail to match individual anatomy, potentially compromising reduction accuracy and fixation stability. Patient-specific implants (PSI) offer improved anatomical fit, but their biomechanical and computational performance requires validation. This study compares PSI and conventional plates (CONV) in a standardized bench test and explores finite element modeling (FEM) as a complementary tool.

A Schatzker IV fracture was simulated in synthetic tibial models (Absolute Bones, Sawbone). One construct was fixed with a PSI designed based on CT data, and one with a CONV plate. Both underwent compression testing (3.5 times body weight, 2900 N, 10 N/s) and fatigue testing (10,000 cycles). An intact model served as reference. Additionally, a FEM replicating the test setup was developed to predict displacement and stress distribution.

Experimental stiffness of the construct was ~3.5 kN/mm for PSI and CONV, versus ~4 kN/mm for intact bone. FEM predicted displacements of 1.5 mm (PSI) and 1.6 mm (CONV) at 2900 N, consistent with the experimental tests. Von Mises stresses in the CONV plate exceeded the yield strength near the bone contact point, reflecting bending moments, while PSI stresses remained far below the yield strength.

ABSTRACT. Proximal femoral fractures are prevalent in older adults. They are classified as femoral neck fractures (FNF) or intertrochanteric fractures (ITF), yet the factors that contribute to each fracture type remain unclear. This study aimed to compare muscle volume, muscle quality, proximal femoral morphology, and bone mineral density (BMD) between FNF and ITF using AI-based three-dimensional computed tomography (CT) analysis. One hundred seventy-three female patients with isolated proximal femoral fractures were analyzed. Patients were classified into FNF and ITF groups (cohort A: 86 FNF, 87 ITF), and 1:1 propensity score matching by age and height was performed, resulting in 69 patients per group (cohort B). Using a validated deep-Bayesian U-net model, peri-hip muscles and the proximal femur of the healthy limb were automatically segmented from CT images. Muscle volume, CT attenuation, femoral neck length (FNL), neck–shaft angle (NSA), and BMD were compared between fracture types. In cohort A, ITF showed lower CT attenuation in several muscles than FNF (p < 0.05). After matching (cohort B), this difference remained only in the gluteus medius and vastus medialis (p < 0.05). NSA and BMD were higher in FNF in both groups (p < 0.05). No significant differences were observed in muscle volume and FNL. In conclusion, our study showed that ITF was associated with lower CT attenuation of the gluteus medius and vastus medialis, whereas FNF was characterized by a larger NSA and higher BMD. These fracture-type–specific differences may reflect distinct risk profiles related to muscle quality and proximal femoral morphology.

ABSTRACT. Introduction: Building a statistical shape model (SSM) of the craniomaxillofacial (CMF) complex is challenging due to extensive preprocessing: variability unrelated to CMF morphology should be reduced, while thin and discontinuous bone structures must be preserved. This study proposes an automated segmentation and preprocessing pipeline for the inclusion of data in a cranial SSM and provides a proof of principle.

Methodology: A dataset of n=81 subjects (Dataset A) was used to train a nnU-Net segmentation model. Another dataset of n=85 adults CT scans (Dataset B) was used for validation; 47 predictions were verified, minimally post-processed if necessary, and incorporated into a second training iteration to refine the model on 81 training sets. Dice scores were calculated to assess segmentation accuracy. Segmentations were converted to meshes, remeshed, and mapped using a symmetric template to establish dense correspondence. A weighted PCA-based SSM was constructed and evaluated using compactness, generalization, and specificity.

Results: The refined nnU-Net model achieved a Dice score of 0.926. From Dataset B, 58/85 scans were suitable for SSM inclusion with no or minimal post-processing. The SSM achieved a mean generalization error of 0.99 mm and specificity values of 1.36 to 1.96 mm.

Conclusion: This workflow substantially reduces manual preprocessing and provides a robust basis for CMF SSM construction that is easily extended with future data input.

ABSTRACT. Patient-specific implants have revolutionized reconstruction of large load-bearing bone defects in orthopaedic oncology and complex trauma. However, conventional workflows depend heavily on external vendors for planning, design, and coordination, limiting clinical control, turnaround time, and iterative refinement. To present a fully in-hospital computer-assisted workflow for patient-specific reconstruction of critical bone defects, in which all digital steps from imaging to implant-ready design are performed entirely within the hospital. A multidisciplinary team integrated computer-assisted planning tools within the hospital environment, encompassing multimodal image acquisition, segmentation and fusion, surgical planning, and patient-specific implant design with mechanical evaluation. Anatomical models and surgical guides were fabricated in-house. The in-hospital digital planning and design process was performed under an ISO 13485-compliant quality management framework. This workflow was applied in a series of patients with large oncologic bone defects. The in-hospital digital process enabled rapid iterations between surgeons and engineers, enhancing anatomical fidelity and fixation strategies. All reconstructions achieved accurate execution of the preoperative plan with excellent implant fit and early evidence of osseointegration. The centralization of planning and design within the clinical environment improved responsiveness to surgical needs and reduced reliance on external planning vendors. A fully in-hospital, computer-assisted workflow for patient-specific reconstruction of large load-bearing bone defects is feasible, clinically effective, and scalable. This approach empowers clinical teams with direct control over digital planning and implant design while maintaining quality manufacturing standards, and is broadly applicable to oncology, complex trauma, and other reconstructive scenarios.

ABSTRACT. A major bottleneck in Computer-Assisted Preoperative Planning (CAPP) for fracture reduction is the limited availability of annotated data. While annotated datasets are now available for evaluating bone fracture segmentation algorithms, there is a notable lack of annotated data for the evaluation of automatic fracture reduction methods. Obtaining precise annotations, which are essential for training and evaluating automatic CAPP algorithm, of the reduced bone therefore remains a critical and underexplored challenge. Existing approaches to assess reduction methods rely either on synthetic fracture simulation which often lacks realism, or on manual virtual reductions, which are complex, time-consuming, operator-dependant and error-prone.

To address these limitations, we propose a hybrid physical-digital framework for generating annotated fracture reduction data. Based on fracture CTs, fragments are first 3D printed, physically reduced, fixed and CT scanned to accurately recover transformation matrix applied to each fragment.

To quantitatively assess reduction quality, we introduce a reproducible formulation of clinically relevant 3D fracture metrics, including 3D gap, 3D step-off, and total gap area. The framework was evaluated on 11 clinical acetabular fracture cases reduced by two independent operators. Compared to preoperative measurements, the proposed approach achieved mean improvements of $168.85 ~\mathrm{mm^2}$ in total gap area, $1.82 ~\mathrm{mm}$ in 3D gap, and $0.81 ~\mathrm{mm}$ in 3D step-off.

This hybrid physical–digital framework enables the efficient generation of realistic, clinically relevant annotated fracture reduction data that can be used for the development and evaluation of automatic fracture reduction algorithms.

ABSTRACT. Accurate guidewire placement along the scaphoid axis is essential for surgical fixation of nondisplaced scaphoid waist fractures, however it is technically challenging due to its complex geometry and limited exposure in the standard procedure. Prior work has proposed solutions to improve guidewire placement accuracy, but implementations often require pre-operative planning and specialized equipment, presenting barriers to adoption. This study aims to address this gap by evaluating the feasibility and accuracy of SAFE (Scaphoid Advanced Fixation Equipment), a practical, computer-assisted path planning and drill guidance system compatible with standard fluoroscopy. Four cadaveric wrists underwent guidewire insertion using the SAFE technique. Radio-opaque fiducials were used to estimate the projection geometry in the two fluoroscopic views (posteroanterior (PA) and 45° oblique) used for intraoperative path planning. The drill guide was programmed according to the surgeon-selected path and the guidewire was drilled. Three procedures achieved single-attempt placement. Angular deviations between targeted and achieved paths in PA and oblique views were 1.3±5.3° and -0.21±2.7° respectively; comparable with those previously reported in 2D fluoroscopy-based techniques. Proximal and distal deviations between targeted and achieved paths were less than ~2mm in all cases. Operative times decreased rapidly with successive trials, suggesting this technique is quickly learned. The achieved guidewire positions were all judged to be clinically acceptable, suggesting that the proposed technique is likely feasible for intraoperative use. By achieving accurate guidewire placement without the need for specialized equipment, a technically challenging aspect of the procedure is removed; potentially allowing surgeons to more reliably repair scaphoid fractures.

ABSTRACT. A primary goal of computer-assisted navigation systems used in orthopaedic and related surgeries is to execute accurate bone cuts. Deviations from the plan with manual cuts can be significant and errors can accumulate when multiple cuts are made in sequence. A common method to define the plane of a cut is by digitizing three points on the cut bone surface using a stylus. However, this method is sensitive to irregularities in the cut bone surface and is also time inefficient as it requires three points to be registered for each cut plane. The purpose of this study is to compare the repeatability and time efficiency of both the three-point and plane digitization approaches in the context of the fibular osteotomies performed in mandibular reconstruction surgery.

A custom planar probe tool and corresponding calibration method were developed. They were subsequently used to digitize 10 porcine bone cut planes, 30 times each, recording time and angular deviation from the mean normal.

The planar tool had a smaller average deviation (~2 degrees vs ~3.7 degrees, p = 0.005), lower variance across the bone cuts, and lower digitization duration (3.0s vs 7.0s, p = 0.008). Cuts on the larger tibia-surfaces had on average smaller deviations than on the fibular surfaces. We believe that the differences between the two probes are significant enough to justify recommending use of a planar probe in a mandibular reconstruction application we are targeting.

ABSTRACT. Total hip arthroplasty (THA) requires accurate acetabular cup placement to prevent complications and optimize postoperative outcomes. CT-based navigation has been shown to improve cup placement accuracy. Recently, transverse skin incisions have been increasingly adopted for cosmetic reasons; however, concerns remain regarding limited surgical exposure and potential effects on component positioning accuracy. While these issues have been reported mainly in the direct anterior approach, their impact on the anterolateral supine approach (ALSA) using CT-based navigation remains unclear. The purpose of this study was to investigate whether skin incision orientation affects accuracy and precision in CT-navigated ALSA THA. We reviewed 49 hips in 41 patients who underwent primary THA using CT-based navigation via ALSA by a single experienced surgeon. Patients were divided into a longitudinal incision group (14 hips) and a transverse incision group (35 hips). Cup target angles were determined based on the combined anteversion concept. Postoperative CT scans at one week were used to assess cup inclination and anteversion, and differences from planned values were calculated. There were no significant differences between groups in patient demographics or diagnoses. Operative time and blood loss were significantly lower in the transverse incision group. However, no significant intergroup differences were observed in cup placement accuracy or precision. In conclusion, in CT-navigated ALSA THA, transverse skin incision does not compromise acetabular cup placement accuracy or precision. CT-based navigation suggests mitigation of exposure-related differences, enabling consistent and reliable cup placement regardless of incision orientation.

ABSTRACT. Background: Removing well-fixed bone cement in revision total hip arthroplasty (THA) is complex, in particular below the femoral stem, with a significant risk of cortical perforation and intraoperative fracture. Extended trochanteric osteotomy (ETO) is a commonly performed but invasive procedure with longer and painful recovery, with remaining risks including non-union, intraoperative fractures and soft tissue injury. Several techniques described in literature still require specialized tools and techniques or are difficult to apply with eccentric distal stem orientation. This study presents a simple and patient specific technique for distal cement removal using a 3D printed surgical guide that is based on the prosthetic stem in situ to guide centralized drilling along a planned trajectory. Methods: This technique was applied in 4 consecutive patients undergoing primary revision THA at our institution between February 2021 and January 2025, and were retrospectively evaluated. PSG design evolved to guide the placement of a guidewire to remove cement using a cannulated drill. Intraoperative complications and total operative time were recorded as outcome measures. Results: In 3 cases, no instances of cortical perforations or intraoperative fractures occurred and successful re-implanting of new implants or spacers was achieved. In one case, the conversion to an ETO was needed because of guidewire fracture. The mean operative time was 129.5 ± 38.3 minutes. Conclusions: The use of patient specific 3D printed guides allows for less invasive and effective removal of distal cement in femoral stem revision, potentially decreasing the risk of cortical damage in revision THA and avoiding ETO.

ABSTRACT. Background: Accurate guide-pin insertion is the first and most critical step for reproducing a preoperative glenoid plan in shoulder arthroplasty. We compared fluoroscopy-guided freehand (FL), patient-specific instrumentation (PSI), and CT-based navigation (NAVI) at this step using clinically meaningful tolerance and outlier criteria for angular and entry-point errors to assess plan reproducibility. Methods: Six cadaveric shoulders were used. Three surgeons performed all three techniques in each shoulder (18 insertions per technique). Deviations from the plan (target, 0°/0 mm) were recorded by the navigation system: superior inclination and posterior tilt (degrees) and superior/anterior entry-point displacement (mm). Outcomes were absolute component errors and two-dimensional (2D) composite errors. Optimal accuracy was defined as |inclination| ≤ 5° and |posterior tilt| ≤ 5° and |superior displacement| ≤ 2 mm and |anterior displacement| ≤ 2 mm; outliers were any angular error > 10° or positional error > 4 mm. Results: NAVI had lower 2D angular error (2.33 ± 1.49°) than FL (6.77 ± 4.21°) and PSI (5.68 ± 4.64°) (Friedman p < 0.01). Entry-point 2D error was lowest with NAVI (1.25 ± 0.45 mm), followed by PSI (2.17 ± 1.17 mm) and FL (3.22 ± 1.30 mm) (p < 0.01). Optimal accuracy was achieved in 94.4% (17/18) with NAVI, 44.4% (8/18) with PSI, and 22.2% (4/18) with FL (Cochran’s Q p < 0.01); outlier rates were 0% (0/18), 22.2% (4/18), and 33.3% (6/18), respectively. Conclusion: CT-based navigation most consistently reproduced planned guide-pin placement. PSI improved entry-point accuracy but did not significantly improve angular accuracy compared with FL.

ABSTRACT. This study presents a refined minimally invasive technique for atlantoaxial fixation that reduces surgical morbidity while preserving the biomechanical strength of the traditional Goel–Harms construct. Conventional open C1–2 fixation requires extensive midline dissection and often results in significant venous bleeding, particularly around the C2 nerve root plexus. To address these limitations, the authors developed a below–obliquus capitis inferior (OCI) approach that avoids direct exposure of the vertebral artery and minimizes soft tissue trauma. Twenty four patients underwent this below OCI technique, which uses navigation guided C1 lateral mass and C2 pedicle screws inserted through small bilateral incisions. To compensate for the limited bone grafting surface inherent to minimally invasive exposure, two supplementary strategies were incorporated: C3 transdiscal screws to provide a strong caudal anchor in cases with compromised C2 anatomy, and intra articular atlantoaxial screws to apply direct compression across the C1–2 joint and promote fusion. Compared with the above OCI minimally invasive approach and the open Goel–Harms technique, the below OCI method demonstrated markedly reduced operative time (157 minutes) and blood loss (59.8 g), while maintaining a high fusion rate of 95.8%. No vertebral artery injuries or neurological complications occurred. The C3 transdiscal screw proved particularly valuable in Hangman’s fractures and comminuted C2 fractures, and the intra articular screw enhanced joint fusion through direct compression. Overall, this combined strategy offers a meaningful advancement in minimally invasive C1–2 stabilization, achieving reduced invasiveness without compromising construct stability or fusion success.

ABSTRACT. We have performed total hip arthroplasty (THA) using antero-lateral supine (ALS) approach with CAOS such as CT-based navigation system or Mako robotic arm-assisted system. Developmental dysplasia of the hip is common in our country, consequently the incidence of proximal femoral antetorsion is also frequent. We allow the cementless stem to be inserted in flexed position along the anterior wall of the neck to avoid excessive stem anteversion. The purpose of this study is to evaluate the efficacy of flexed stem insertion in ALS-THA for cases with excessive femoral antetorsion. 50 cases of cementless ALS-THA with estimated stem anteversion angle of 30° or greater preoperatively were included. The evaluations included intraoperative or postoperative complications, the estimated stem anteversion angle (ESA), the actual stem anteversion angle (ASA), and the anterior wall-stem anterior angle (AWSA), calculated for both the flexed stem insertion group (F group) and the neutral stem insertion group (I group). No intraoperative or postoperative complications related to flexed stem insertion were observed. In the F group, ESA was 39.3±8.3°, ASA was 40.1±8.6°, and AWSA was 12.2±9.1°, showing a significant correlation with ASA. (P<0.05) In the I group, ESA was 38.5±6.8° while ASA was 50.0±10.0°, indicating a significantly increased stem anteversion angle in the I group compared to the F group (P<0.05). We concluded that in cases of excessive femoral antetorsion, allowing flexed insertion of the stem may minimize the increase of stem anteversion. ALS-THA with CAOS is considered essential to pursue optimal joint reconstruction such as forgotten joint.

ABSTRACT. Functional alignment total knee arthroplasty (FA-TKA) aims to optimize implant positioning and soft-tissue balance throughout the range of motion, particularly when combined with computer-assisted technologies. However, reliable balancing remains challenging to achieve in knees with severe varus deformity, which are common in Asian populations.

We report our clinical experience comparing two soft-tissue management strategies within a navigated FA workflow: a conventional manual gap-balancing approach using spacer blocks and an instrumented gap-balancing approach using a force-controlled tensioning device. Fifty consecutive primary TKAs with severe varus deformity were retrospectively reviewed, including 25 manual FA-TKAs followed by 25 instrumented FA-TKAs. All procedures were performed by a single surgeon using an image-free navigation system with a tibia-first workflow. The instrumented technique provided continuous medial and lateral gap data under constant distraction force from full extension to deep flexion.

Baseline demographics (age, height, weight, body mass index) did not differ between groups. Femoral component rotational alignment was significantly different between groups (5.2° ± 2.6° , instrumented FA-TKA vs. 1.5° ± 1.0°, manual FA-TKA, p <0.0001). The hip-knee-ankle angle was not significantly different between groups, preoperatively (169.8° ± 4.0° vs. 169.7° ± 3.5°) or postoperatively (176.1° ± 2.1° vs. 176.0° ± 3.1°). The average total KOOS and Forgotten Joint Score-12 scores were not significantly different between groups (79.5 vs. 72.1, and 65.8 vs. 52.3, respectively). The KOOS Sports sub-score was significantly higher in instrumented FA-TKAs than manual FA-TKAs (68.3 vs. 45.8, p =0.029).

With conventional tensioners, accurate gap assessment from mid-flexion to deep flexion has been limited by patellar tendon interference and insufficient precision in defining knee flexion angles, making interpretation of gap values outside extension difficult. In the current study, the findings suggest that instrumented, force-controlled gap balancing within a functional alignment workflow may enhance outcomes in high-demand activities by more accurately addressing patient-specific soft-tissue characteristics. Objective intraoperative gap assessment may be particularly beneficial in TKA for severe varus deformity.