AI assisted video analysis of the Trendelenburg Test: A Feasibility Study

AI assisted video analysis of the Trendelenburg Test: A Feasibility Study | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Article AI assisted video analysis of the Trendelenburg Test: A Feasibility Study Kieran O'Sullivan, Tom Doyle, Eoghain Quinn, Connor Kilkenny, and 2 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7680719/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 16 Feb, 2026 Read the published version in Scientific Reports → Version 1 posted 12 You are reading this latest preprint version Abstract Background: The Trendelenburg test is widely used to assess hip abductor function but is traditionally graded as a binary sign, limiting sensitivity and reliability. Compensatory trunk lean may mask pelvic drop and further reduces diagnostic accuracy. Advances in artificial intelligence (AI) and markerless motion capture provide an opportunity to quantify pelvic obliquity, trunk lean, and knee alignment objectively using simple video-based systems. Methods: A cross-sectional feasibility study was conducted in an elective hip clinic. Twelve adult patients (seven post-total hip arthroplasty, five with osteoarthritis, performed bilateral single-leg stance while recorded by a single posterior camera (4K, 60 fps). Videos were analysed using an AI-based application (OnForm) to extract pelvic obliquity, trunk lean, and knee alignment change. Feasibility metrics included analysis success rate, retakes required, and workflow timing. Outcomes were reported as medians with interquartile ranges (IQR) and ranges. Results: All patients completed the protocol with analysable recordings (100% technical success). The mean total workflow time was 217 seconds (IQR 206.8–230.3; range 193–244). Median worst pelvic obliquity was 0.0° (IQR –1.0° to +1.5°; range –5° to +6°). Median trunk lean was 4.5° (IQR 2.8°–9.0°; range 2°–10°). Median knee alignment change was 3.0° (IQR 2.0°–4.0°; range 1°–8°). Trunk lean ≥5° was seen in 6 patients (5/7 post-THA), while significant pelvic drop (≤ –4°) occurred in only one patient. Knee deviations ≥3° were observed in 8 patients (67%). Conclusions: AI-assisted single-camera analysis of the Trendelenburg test is feasible, rapid, and clinically informative. The method consistently quantified pelvic, trunk, and knee angles, revealing that post-THA patients frequently compensate with trunk lean rather than pelvic drop. These findings support the integration of objective video-based Trendelenburg assessment into outpatient orthopaedic practice. Health sciences/Anatomy Health sciences/Diseases Health sciences/Health care Health sciences/Medical research Trendelenburg test Hip abductor function Total hip arthroplasty Pelvic obliquity Trunk lean Artificial intelligence Markerless motion analysis Figures Figure 1 Figure 2 Introduction A positive Trendelenburg sign classically manifests as a contralateral pelvic drop when standing on the affected limb, indicating insufficiency of the stance-side gluteal muscles. This clinical sign is observed in a range of hip pathologies, including osteoarthritis and after total hip arthroplasty (THA). However, the traditional binary (positive/negative) interpretation of the Trendelenburg test has notable limitations in reliability and sensitivity. Small drops (<2°) or subtle compensatory movements are often difficult to discern by eye, especially in patients with higher body mass or atypical posture, undermining the test’s utility as an outcome measure in practice (1). Compensatory trunk lean is another critical but under-recognized component of the Trendelenburg response, which can mask a positive pelvic finding. Excessive pelvic drop and trunk lean both alter joint loading: they increase hip joint reaction forces and can effect knee biomechanics over time. (2). The limitations of subjective observation have prompted exploration of objective measurement technologies. Marker-based three-dimensional motion capture is a gold-standard for kinematic analysis but is impractical in routine clinics due to cost, time, and technical requirements. Advances in computer vision and artificial intelligence have enabled markerless motion analysis using standard video cameras. Despite this progress, no published study to date has evaluated a simple clinic-based video system for Trendelenburg test quantification in an orthopaedic population. In particular, it remains unknown whether a smartphone or single-camera setup can feasibly capture pelvic drop, trunk lean, and frontal-plane knee alignment changes during the test as part of routine practice. It is also unclear how these objective measures might differ between post-THA patients and those with native hip pathology, or whether such a tool could reveal secondary deviations that are not captured in the traditional exam. The purpose of this feasibility study was to determine whether an AI-assisted single-camera video analysis can objectively assess Trendelenburg test performance in a standard outpatient hip clinic. We sought to measure key kinematic outcomes – pelvic obliquity, trunk lean, and change in knee alignment – during single-leg stance, and to compare these metrics between post-THA patients and patients with native (non-operative) hip pathology. Methods Study Design and Participants: This was a single-centre, cross-sectional feasibility study conducted in an elective hip clinic at Galway University Hospital. Institutional ethics approval was obtained (Ref: C.A.3416). Patients were recruited prospectively during May – July 2025 from consecutive clinic attendees. Eligible participants were adults (≥18 years) attending the elective orthopaedic hip clinic with either a history of total hip arthroplasty or revision surgery and ongoing hip-related symptoms, or with native hip pathology such as osteoarthritis. Patients were required to provide written consent and be able to perform an unassisted single-leg stance for at least five seconds. Exclusion criteria included inability to safely complete the Trendelenburg test due to neurological disease, severe balance impairment, or acute lower-limb injury; cognitive or communication barriers precluding consent or test comprehension; and any physical condition that prevented reliable video analysis of pelvic and limb landmarks. Patients with active infection or other contraindications to weight-bearing were also excluded. All participants provided written informed consent prior to inclusion, using a Galway University Hospitals-approved consent form. Consent included permission for participation, data collection, and use of anonymised images or data for publication. The individual depicted in Figures 1 and 2 is a volunteer model and not a study participant.” Twelve adults were enrolled (7 post-THA or revision cases, 5 with native hip pathology). The post-operative group had undergone THA or revision surgery and continued to report hip-related symptoms, whereas the native group consisted of patients with hip pain but no arthroplasty. For clinical realism, patients wore their own shoes during testing and only removed heavy outer garments. Very loose clothing that could obscure pelvic landmarks was discouraged, but attire was not otherwise standardized, reflecting evidence that clothing has minimal impact on pose measurements when using consistent setup. Video Capture Protocol: Each patient performed the Trendelenburg test bilaterally while being recorded on a single digital video camera (4K resolution, 60 fps). The camera was positioned 2 m behind the patient and 1 meter high (approximately pelvic height) to obtain a direct posterior-plane view. A horizontal reference (a wall-mounted bar at pelvic height) was in frame to serve as a level baseline for angle measurements. Patients stood facing the camera with arms relaxed at their sides. They were instructed to fix their gaze forward and maintain single-leg stance for 5–10 seconds per trial. Two trials were recorded for each leg. If balance was lost on a first attempt, an additional trial was taken after a brief rest. No external support (e.g. wall or chair) was used during the test. This standardized setup (fixed camera distance/height and a visible horizon) was chosen to maximize measurement consistency based on prior pose-estimation guidelines. Angle Measurement: Recorded videos were analysed offline using a commercial smartphone application (OnForm, OnForm, LLC) that implements markerless pose estimation for video analysis. The software automatically identified anatomical landmarks and allowed manual angle annotation using onscreen tools. We focused on three kinematic measures (illustrated in Figure 1 and Figure 2 ): Pelvic Obliquity: the angle between the line connecting the patient’s left and right hip joints and the horizontal plane. A value of 0° indicates a level pelvis. Negative values denote a drop of the contralateral side, whereas positive values denote a contralateral pelvis rise. Trunk Lean: the angle between the trunk which is found by pose estimation on OnForm and the true horizontal. Larger positive values indicate the torso leaning further toward the stance leg. Coronal Plane Knee Angle Change: the difference in the coronal-plane knee alignment between double-leg standing and single-leg stance. This was derived by measuring the hip–knee–ankle angle of the stance limb during bipedal stance (baseline) and again during the single-leg stance, then taking the change (Δ). Knee angles greater than 180° in single-leg stance were described as “varus” by the amount of increase, whereas angles less than 180° were described as “valgus” by the deficit in degrees. All study procedures were performed in accordance with institutional and international standards, including the Declaration of Helsinki (2013), the HSE National Consent Policy for Health and Social Care Research (2022), and the Data Protection Act 2018 (Health Research) Regulations 2018 and 2021. Data Analysis: Outcome measures for each trial included pelvic obliquity (most negative / positive value, representing the greatest contralateral drop or hike) and the maximum trunk lean observed during the 5–10 s hold, as well as the maximum knee angle change. For each patient, values were aggregated by side (left stance and right stance) and the maximum values were noted. Descriptive statistics were used given the small sample. Continuous outcomes are reported as medians with interquartile ranges (IQR) and total ranges. We also counted the number of patients meeting pre-specified threshold criteria for an abnormal Trendelenburg test: specifically, a pelvic drop of ≥4° (i.e. pelvic obliquity ≤ –4°) and a trunk lean of ≥5°. These cut-offs were based on the criteria proposed by Stevens et al. . All analyses were stratified by group (post-THA vs native hips) for comparison. Results Feasibility: All 12 patients successfully completed the video-based Trendelenburg assessment on both sides. Two patients (17%) required a single retake due to a brief balance loss, but ultimately every patient yielded an acceptable recording for each limb (100% technical success). The workflow was efficient: video capture took an average of 30.2 seconds (mean) (IQR) of 23.8 to 36.0; range 19 to 42 seconds per patient. Video Analysis took an average of 186.8 seconds (mean) IQR of 179.3 to 194.3; range 168 to 207 seconds), for a total of ~3.5 minutes per patient on average (mean 217 seconds, IQR 206.8 TO 230.3; range 193 to 244 seconds) (Table 1). Video quality was consistently sufficient, the horizontal reference was visible in all clips, key landmarks (pelvis, trunk, knees) remained unobstructed, and no frames were discarded due to tracking loss or occlusion. All patients wore their normal footwear and attire, which did not impede the pose detection process, underscoring the method’s practicality in a routine clinic setting. Trendelenburg Angle Outcomes: Table 2 summarizes the kinematic results for each patient. Overall, pelvic tilt deviations were small in this cohort. The median worst contralateral pelvic tilt (i.e. the most negative pelvic obliquity observed, indicating pelvic drop) was 0.0° (IQR of –1.0° to +1.5° ; range –5°to +6°) . In practical terms, half of the patients did not exceed roughly 1° of pelvic drop or lift. By contrast, trunk lean was more pronounced: the median maximum trunk lean was 4.5° (IQR 2.8°–9.0°; range 2°–10°). The median coronal-plane knee angle change (difference between bilateral stance and single-leg stance) was 3.0° (IQR 2.0°–4.0°; range 1°–8°). Group Differences: When comparing the seven post-THA patients to the five native-hip patients, distinct patterns emerged. Post-THA patients showed a greater tendency toward trunk compensation: their median maximal trunk lean was 9.0° (IQR 4.5°–9.5°) versus 3.0° (IQR 2.0°–3.0°) in the native-hip group. In fact, 5 of 7 post-THA patients demonstrated a trunk lean ≥5°, compared to only 1 of 5 in the native-hip group. By contrast, pelvic tilt remained near-neutral in both groups (median worst pelvic obliquity 0.0° in post-THA vs +1.0° in native hips), indicating that large contralateral pelvic drops were uncommon across the board. Knee alignment changes were slightly greater in the arthroplasty group (median 3.0° vs 2.0° in natives) Threshold Criteria: Applying published biomechanical cut-offs for an abnormal Trendelenburg test, only 1 patient (8%) demonstrated a contralateral pelvic drop of ≥4° (signifying significant hip abductor insufficiency). In contrast, 6 patients (50%) exhibited an ipsilateral trunk lean ≥5°; notably, five of these six were in the post-THA cohort. A coronal-plane knee deviation of ≥3° was observed in 8 patients (67%) , including 6 of 7 post-THA cases and 2 of 5 native-hip cases. Importantly, the direction of maximal knee movement was varus (medial collapse) in all 12 patients – in other words, every patient’s greatest knee excursion involved the knee moving medially. None of the patients had their peak knee change occur in the lateral (valgus) direction during the recorded trials. These objective results (Tables 1 and 2) provide a clear quantitative profile of the Trendelenburg test performance in both patient groups, without any need for subjective grading. Table 1. Feasibility results for Trendelenburg video analysis in clinic. Patient Age (Years) Video capture time (seconds) Analysis time (seconds) Total workflow time (seconds) Retake required Videos recorded Successful analysis 1 71 25 168 193 No 1 Yes 2 87 20 177 197 Yes 2 Yes 3 49 19 174 193 No 1 Yes 4 48 30 180 210 No 1 Yes 5 52 35 183 218 No 1 Yes 6 78 23 192 215 Yes 2 Yes 7 72 24 186 210 No 1 Yes 8 54 35 181 216 No 1 Yes 9 60 36 201 237 No 1 Yes 10 49 36 192 228 No 1 Yes 11 53 42 201 243 No 1 Yes 12 50 37 207 244 No 1 Yes Table 2. Patient Trendelenburg Test Outcomes . Patient ID Age (yrs) Status Presenting Complaint Pelvic Obliquity Pelvic Obliquity Trunk Lean Trunk Lean (R) Left Knee Bipedal Left Knee Single Stance Left Knee Change Right Knee Bipedal Right Knee Single Stance Right Knee Change 1 71 Post-Op THA Right hip pain -1 5 3 1 2 4 2 Varus 1 -2 3 Valgus 2 87 Post-Op THA Left hip pain 9 6 9 0 -2 -5 3 Valgus 0 3 3 Varus 3 49 Post-Op THA Right hip revision 3 0 0 9 0 1 1 Varus -8 -16 8 Valgus 4 48 Post-Op THA Post-op (femoral nerve palsy) 0 -5 1 2 -3 0 3 Varus -4 -6 2 Valgus 5 52 Post-Op THA Right hip pain 5 -1 10 3 3 4 1 Varus 1 3 2 Varus 6 78 Post-Op THA MoM hip review 1 3 10 10 -4 -7 3 Valgus -1 -4 3 Valgus 7 72 Post-Op THA Bilateral knee pain 2 0 6 4 3 -2 5 Valgus 4 -3 7 Valgus 8 54 Native Right hip pain 2 0 1 2 -3 -4 1 Valgus -3 -4 1 Valgus 9 60 Native Left hip pain 1 3 3 3 -2 -4 2 Valgus -3 -4 1 Valgus 10 49 Native Right hip pain 4 4 1 3 -2 -3 1 Valgus -2 -4 2 Valgus 11 53 Native Right hip pain 0 -1 2 2 -5 -5 0 -1 -5 4 Valgus 12 50 Native Left hip pain 3 6 5 6 3 -1 4 Valgus -1 -3 2 Valgus * Varus Positive Angle, Valgus Negative Angle. Pelvic Obliquity Positive – Hip Hike, Negative – Hip Drop Measurements in Degrees (x°) Discussion This feasibility study demonstrates that a single-camera AI-based video system can successfully quantify key angles during the Trendelenburg test in a routine clinical setting. All participants were able to complete the protocol, and the automated pose analysis yielded objective measures of pelvic tilt, trunk lean, and knee alignment within minutes. The overall magnitude of observed pelvic drop was quite small, while trunk lean was frequently elevated, especially among post-THA patients, this is possibly consistent with known compensatory patterns in hip abductor weakness. These results support the notion that video-based measurements can capture clinically relevant variations in Trendelenburg test performance that may be subtle or missed by traditional visual observation alone (3). A key finding is that post-THA patients exhibited substantially greater trunk lean than patients with native hip pathology, despite having similarly minimal pelvic drop. This suggests that many post-arthroplasty patients rely on torso compensation (leaning the upper body toward the stance side) to stabilize the pelvis, rather than allowing a pelvic sag on the unsupported side. Prior research indicates that hip abductor strength often remains deficient after THA and that certain surgical/anatomical factors (such as a more superolateral hip centre of rotation or reduced femoral offset) can increase the mechanical demand on the abductors (4, 5). Both of these aspects can contribute to a positive Trendelenburg sign following hip replacement. Leaning the trunk toward the stance leg is a well-recognized compensatory strategy: it reduces the lever arm of body weight, thereby partially offloading the weak hip abductors. Our data align with this mechanism, the THA group in our sample showed frequent and large trunk deviations, whereas large contralateral pelvic drops were rarely seen. In essence, many post-THA patients achieved coronal-plane pelvic stability by substituting trunk lean for pelvic tilt. From a clinical perspective, quantifying these angular deviations offers immediate value. Instead of simply noting a binary “Trendelenburg sign positive/negative,” clinicians can document the degree of pelvic drop and lateral trunk lean in objective terms. These metrics enable tracking changes over time and can inform targeted interventions. For example, a pelvic drop beyond ~ 4° may be interpreted as a significant abductor deficit, which could prompt focused gluteal strengthening and balance training. Similarly, a trunk lean above ~ 5° suggests reliance on a compensation strategy; such patients may benefit from gait retraining to reduce excessive torso sway. In our study, half of all patients, and a large majority of the post-THA subgroup, exceeded the 5° trunk lean threshold, underscoring how common this compensation was in the post-operative population. The inclusion of a knee alignment metric is a novel addition to Trendelenburg test assessment that warrants attention. Although we did not directly measure knee joint moments, a shift of the knee is biomechanically linked to increased loading of the contralateral compartment of the knee. Prior studies have shown that greater lateral trunk lean and contralateral pelvic drop both increase the external knee adduction moment during gait and single-leg stance (2).Therefore, identifying a stance-phase drift of ≥3° on the Trendelenburg test can alert clinicians to potential knee overload in patients with hip abductor dysfunction. This information may guide early preventive interventions such as physiotherapy and advising on limb alignment and weight distribution during daily single-leg tasks, to mitigate undue stress on the knee joint. It is important to recognize the limitations of a 2D single-camera approach for Trendelenburg evaluation. Measurement error in markerless video analysis must be considered when interpreting small angle differences. Experimental work using goniometry suggests that changes smaller than about 4° fall within the typical measurement error for pelvic tilt assessments (6). Likewise, contemporary 2D pose-estimation algorithms have an angle accuracy on the order of ~ 2–3° under ideal conditions. In this context, a difference of only a couple of degrees (for example, a change in pelvic drop from 0° to 2°) should be interpreted with caution, as it might reflect normal variability or minor asymmetry rather than a meaningful change in abductor function. The literature recommends using a conservative threshold (around 4°) for defining true deterioration or improvement in Trendelenburg angles, and repeating the test if values are borderline or unexpected. There are additional technical and design considerations. Because 2D analysis accuracy depends on camera perspective, careful standardization of the setup is required. We kept the camera perpendicular to the coronal plane and used a horizontal reference bar to calibrate pelvic level, but even slight camera tilt or out-of-plane movement by the patient could introduce error (7). We did not strictly control lighting or clothing colour; however, prior evaluations suggest these factors have minimal impact on gross coronal-plane kinematics in clinic settings. All patients performed the test in their own shoes, which might subtly affect balance or knee mechanics, but this choice reflects real-world practice (future studies might adopt a barefoot protocol for consistency). Another limitation is the relatively small sample size . With only 12 patients (7 post-THA and 5 native hips), our study was not powered for formal statistical comparisons between groups. We treated this research as a proof-of-concept feasibility study, so no inferential statistics were performed and we did not attempt to correlate the angle measurements with clinical outcomes like pain or functional scores. The brief duration of each single-leg stance trial (5–10 seconds) is another limitation – shorter than the 30-second holds used in some reference protocols – which may have reduced our ability to observe fatigue-related pelvic drop or trunk sway. These limitations should be kept in mind when generalizing the findings. Despite the above limitations, our findings are in agreement with, and extend, the existing literature on the Trendelenburg test. They reinforce that an objective, video-based approach is both feasible and informative in everyday practice. The pattern we observed, minimal pelvic drop but frequent trunk lean in post-THA patients echoes the logic of prior biomechanical analyses and matches the quantitative criteria proposed by Stevens et al. for an abnormal Trendelenburg response. Using Stevens’ numeric framework as a reference, we found that many patients in our cohort (especially those post-THA) would be classified as having an atypical Trendelenburg result primarily due to excessive trunk lean rather than pelvic drop. This suggests that focusing only on contralateral pelvic drop (the traditional clinical sign) would underestimate the prevalence of gait compensation in post-THA individuals. Furthermore, the inclusion of the simple knee alignment measure provides a more holistic view of single-leg stance, capturing downstream effects on the lower limb kinetic chain that might have implications for long-term joint health. In summary, the corrected results indicate that AI-assisted video analysis can quantitatively distinguish how different patient groups manage the Trendelenburg test: post-THA patients tend to engage a trunk lean compensation more than native-hip patients, even though overt pelvic dropping is minimal in both groups. This quantitative insight, alongside established thresholds from the literature, strengthens the clinical interpretation of the Trendelenburg test and lays the groundwork for improved patient assessment and monitoring. Future Directions Building on this feasibility study, several next steps are recommended to enhance and validate the approach. First , a larger-scale study should be conducted to gather more extensive data and enable direct comparison to established 3D-derived criteria. Second , future research should incorporate additional clinical correlates to explore the underlying causes of the observed compensations. Measuring hip abductor strength (e.g. with hand-held dynamometry) and evaluating post-THA radiographic parameters (such as the reconstructed hip centre position or femoral offset on X-ray) could help discern whether large trunk leans are more attributable to residual muscle weakness or to altered hip biomechanics (or both). Finally , to facilitate clinical adoption of this technique, an effort should be made to develop a simple reporting template or automated output for the Trendelenburg test. Such a template could succinctly record the three key angles – pelvic drop, trunk lean, and knee varus/valgus change, alongside notes on patient symptoms (pain, balance, etc.), providing clinicians with a summary that is far more informative than the traditional binary outcome. These future steps will help validate the utility of AI-assisted Trendelenburg assessment and potentially expand its role in both research and routine orthopaedic practice. Conclusion In summary, this study found that an AI-assisted, single-camera video workflow is a feasible and informative method to quantify the Trendelenburg test in an outpatient setting. The system produced objective measurements of pelvic obliquity, trunk lean, and knee alignment within minutes, with 100% technical success in our sample. An objective video-based Trendelenburg assessment has the potential to improve clinical documentation and patient monitoring. Future work will focus expanding it to larger cohorts to establish its utility as a routine clinical tool. Declarations Ethical Approval Ethical approval was granted by the Clinical Research Ethics Committee (CREC) of Galway University Hospitals, reference number C.A. 3416. Funding Declaration No funding was received for the conduct of this study. Conflicts of Interest The authors declare that they have no conflicts of interest relevant to the content of this manuscript. References McCarney L, Andrews A, Henry P, Fazalbhoy A, Selva Raj I, Lythgo N, et al. Determining Trendelenburg test validity and reliability using 3-dimensional motion analysis and muscle dynamometry. Chiropr Man Therap. 2020;28(1):53. Takacs J, Hunt MA. The effect of contralateral pelvic drop and trunk lean on frontal plane knee biomechanics during single limb standing. J Biomech. 2012;45(16):2791-6. Stevens WR, Jo CH, Tulchin-Francis K. Clinically derived biomechanical criteria for the Trendelenburg test. Clin Biomech (Bristol). 2020;78:105066. Ismailidis P, Kvarda P, Vach W, Cadosch D, Appenzeller-Herzog C, Mündermann A. Abductor Muscle Strength Deficit in Patients After Total Hip Arthroplasty: A Systematic Review and Meta-Analysis. J Arthroplasty. 2021;36(8):3015-27. Kiyama T, Naito M, Shitama H, Maeyama A. Effect of superior placement of the hip center on abductor muscle strength in total hip arthroplasty. J Arthroplasty. 2009;24(2):240-5. Youdas JW, Mraz ST, Norstad BJ, Schinke JJ, Hollman JH. Determining meaningful changes in pelvic-on-femoral position during the Trendelenburg test. J Sport Rehabil. 2007;16(4):326-35. Baldinger M, Reimer LM, Senner V. Influence of the Camera Viewing Angle on OpenPose Validity in Motion Analysis. Sensors (Basel). 2025;25(3). Additional Declarations No competing interests reported. Cite Share Download PDF Status: Published Journal Publication published 16 Feb, 2026 Read the published version in Scientific Reports → Version 1 posted Editorial decision: Revision requested 24 Nov, 2025 Reviews received at journal 13 Nov, 2025 Reviews received at journal 06 Nov, 2025 Reviewers agreed at journal 04 Nov, 2025 Reviewers agreed at journal 01 Nov, 2025 Reviews received at journal 22 Oct, 2025 Reviewers agreed at journal 10 Oct, 2025 Reviewers invited by journal 08 Oct, 2025 Editor assigned by journal 08 Oct, 2025 Editor invited by journal 07 Oct, 2025 Submission checks completed at journal 06 Oct, 2025 First submitted to journal 06 Oct, 2025 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-7680719","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":532379511,"identity":"61bf1dd2-09fb-4f10-a932-3a2204e23324","order_by":0,"name":"Kieran O'Sullivan","email":"data:image/png;base64,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","orcid":"","institution":"University of Galway","correspondingAuthor":true,"prefix":"","firstName":"Kieran","middleName":"","lastName":"O'Sullivan","suffix":""},{"id":532379512,"identity":"d8315687-f757-45c3-81d7-7163611cf72b","order_by":1,"name":"Tom Doyle","email":"","orcid":"","institution":"Royal College of Surgeons in Ireland","correspondingAuthor":false,"prefix":"","firstName":"Tom","middleName":"","lastName":"Doyle","suffix":""},{"id":532379513,"identity":"928284d6-55a0-48ab-9765-5a6ce7c3c6ea","order_by":2,"name":"Eoghain Quinn","email":"","orcid":"","institution":"Royal College of Surgeons in Ireland","correspondingAuthor":false,"prefix":"","firstName":"Eoghain","middleName":"","lastName":"Quinn","suffix":""},{"id":532379514,"identity":"99f91f1e-47e4-4176-ba79-452e253226da","order_by":3,"name":"Connor Kilkenny","email":"","orcid":"","institution":"Royal College of Surgeons in Ireland","correspondingAuthor":false,"prefix":"","firstName":"Connor","middleName":"","lastName":"Kilkenny","suffix":""},{"id":532379515,"identity":"7783b819-b2a5-42ac-b0d3-480e0d69d4c6","order_by":4,"name":"Gordan Daly","email":"","orcid":"","institution":"Royal College of Surgeons in Ireland","correspondingAuthor":false,"prefix":"","firstName":"Gordan","middleName":"","lastName":"Daly","suffix":""},{"id":532379516,"identity":"4dab219c-ea77-4201-a58b-0e1f73b8dc72","order_by":5,"name":"Aiden Devitt","email":"","orcid":"","institution":"University of Galway","correspondingAuthor":false,"prefix":"","firstName":"Aiden","middleName":"","lastName":"Devitt","suffix":""}],"badges":[],"createdAt":"2025-09-22 15:39:12","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-7680719/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-7680719/v1","draftVersion":[],"editorialEvents":[{"content":"https://doi.org/10.1038/s41598-026-38980-z","type":"published","date":"2026-02-16T15:57:54+00:00"}],"editorialNote":"","failedWorkflow":false,"files":[{"id":94049788,"identity":"022b7c2d-3039-428a-85be-e8bbc943a49a","added_by":"auto","created_at":"2025-10-21 23:41:45","extension":"docx","order_by":0,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":275883,"visible":true,"origin":"","legend":"","description":"","filename":"AIassistedvideoanalysisoftheTrendelenburgTestForPublicationV4.docx","url":"https://assets-eu.researchsquare.com/files/rs-7680719/v1/d65485e9021d2892eaac3f7c.docx"},{"id":94049787,"identity":"dbe73da4-794f-44c8-acf1-7bb9fc10def2","added_by":"auto","created_at":"2025-10-21 23:41:44","extension":"json","order_by":1,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":8334,"visible":true,"origin":"","legend":"","description":"","filename":"e9321c3159cf47efbe41b3f8b633e3b6.json","url":"https://assets-eu.researchsquare.com/files/rs-7680719/v1/e7b601efc5454629d6710eab.json"},{"id":94049791,"identity":"7006c76c-3476-4a1f-a699-d9f691272d5c","added_by":"auto","created_at":"2025-10-21 23:41:45","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":196528,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cem\u003eBipedal Stance\u003c/em\u003e\u003c/p\u003e","description":"","filename":"1.png","url":"https://assets-eu.researchsquare.com/files/rs-7680719/v1/85a168d4957c3ec40532ac4c.png"},{"id":94049789,"identity":"64834c07-f845-483c-a6b5-4a1c415d977e","added_by":"auto","created_at":"2025-10-21 23:41:45","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":277753,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cem\u003eSingle Leg Stance including a True Horizontal, Pelvic Obliquity and Trunk Lean Measurement\u003c/em\u003e\u003c/p\u003e","description":"","filename":"2.png","url":"https://assets-eu.researchsquare.com/files/rs-7680719/v1/95fde6f44c8a6d49882a3105.png"},{"id":103251958,"identity":"1ca86fe4-5ad6-4cba-aa4d-2d75ad97b3ba","added_by":"auto","created_at":"2026-02-23 16:12:21","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1558472,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7680719/v1/a5b5452a-17cb-4c1b-9caa-2ff7b8872e70.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"AI assisted video analysis of the Trendelenburg Test: A Feasibility Study","fulltext":[{"header":"Introduction","content":"\u003cp\u003eA positive Trendelenburg sign classically manifests as a contralateral pelvic drop when standing on the affected limb, indicating insufficiency of the stance-side gluteal muscles. This clinical sign is observed in a range of hip pathologies, including osteoarthritis and after total hip arthroplasty (THA). However, the traditional binary (positive/negative) interpretation of the Trendelenburg test has notable limitations in reliability and sensitivity. Small drops (\u0026lt;2°) or subtle compensatory movements are often difficult to discern by eye, especially in patients with higher body mass or atypical posture, undermining the test’s utility as an outcome measure in practice (1).\u003c/p\u003e\n\u003cp\u003eCompensatory trunk lean is another critical but under-recognized component of the Trendelenburg response, which can mask a positive pelvic finding. Excessive pelvic drop and trunk lean both alter joint loading: they increase hip joint reaction forces and can effect knee biomechanics over time. (2).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eThe limitations of subjective observation have prompted exploration of objective measurement technologies. Marker-based three-dimensional motion capture is a gold-standard for kinematic analysis but is impractical in routine clinics due to cost, time, and technical requirements. Advances in computer vision and artificial intelligence have enabled\u0026nbsp;markerless\u0026nbsp;motion analysis using standard video cameras. Despite this progress, no published study to date has evaluated a simple clinic-based video system for Trendelenburg test quantification in an orthopaedic population. In particular, it remains unknown whether a smartphone or single-camera setup can feasibly capture pelvic drop, trunk lean, and frontal-plane knee alignment changes during the test as part of routine practice. It is also unclear how these objective measures might differ between post-THA patients and those with native hip pathology, or whether such a tool could reveal secondary deviations that are not captured in the traditional exam.\u003c/p\u003e\n\u003cp\u003eThe purpose of this feasibility study was to determine whether an AI-assisted single-camera video analysis can objectively assess Trendelenburg test performance in a standard outpatient hip clinic. We sought to measure key kinematic outcomes – pelvic obliquity, trunk lean, and change in knee alignment – during single-leg stance, and to compare these metrics between post-THA patients and patients with native (non-operative) hip pathology.\u003c/p\u003e"},{"header":"Methods","content":"\u003cp\u003e\u003cstrong\u003eStudy Design and Participants:\u003c/strong\u003e This was a single-centre, cross-sectional feasibility study conducted in an elective hip clinic at Galway University Hospital. Institutional ethics approval was obtained (Ref: C.A.3416). Patients were recruited prospectively during May – July 2025 from consecutive clinic attendees. Eligible participants were adults (≥18 years) attending the elective orthopaedic hip clinic with either a history of total hip arthroplasty or revision surgery and ongoing hip-related symptoms, or with native hip pathology such as osteoarthritis. Patients were required to provide written consent and be able to perform an unassisted single-leg stance for at least five seconds. Exclusion criteria included inability to safely complete the Trendelenburg test due to neurological disease, severe balance impairment, or acute lower-limb injury; cognitive or communication barriers precluding consent or test comprehension; and any physical condition that prevented reliable video analysis of pelvic and limb landmarks. Patients with active infection or other contraindications to weight-bearing were also excluded.\u0026nbsp;All participants provided written informed consent prior to inclusion, using a Galway University Hospitals-approved consent form. Consent included permission for participation, data collection, and use of anonymised images or data for publication. The individual depicted in Figures 1 and 2 is a volunteer model and not a study participant.”\u003c/p\u003e\n\u003cp\u003eTwelve adults were enrolled (7 post-THA or revision cases, 5 with native hip pathology). The post-operative group had undergone THA or revision surgery and continued to report hip-related symptoms, whereas the native group consisted of patients with hip pain but no arthroplasty. For clinical realism, patients wore their own shoes during testing and only removed heavy outer garments. Very loose clothing that could obscure pelvic landmarks was discouraged, but attire was not otherwise standardized, reflecting evidence that clothing has minimal impact on pose measurements when using consistent setup.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eVideo Capture Protocol:\u003c/strong\u003e Each patient performed the Trendelenburg test bilaterally while being recorded on a single digital video camera (4K resolution, 60 fps). The camera was positioned 2 m behind the patient and 1 meter high (approximately pelvic height) to obtain a direct posterior-plane view. A horizontal reference (a wall-mounted bar at pelvic height) was in frame to serve as a level baseline for angle measurements. Patients stood facing the camera with arms relaxed at their sides. They were instructed to fix their gaze forward and maintain single-leg stance for 5–10 seconds per trial. Two trials were recorded for each leg. If balance was lost on a first attempt, an additional trial was taken after a brief rest. No external support (e.g. wall or chair) was used during the test. This standardized setup (fixed camera distance/height and a visible horizon) was chosen to maximize measurement consistency based on prior pose-estimation guidelines.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAngle Measurement:\u003c/strong\u003e Recorded videos were analysed offline using a commercial smartphone application (OnForm, OnForm, LLC) that implements markerless pose estimation for video analysis. The software automatically identified anatomical landmarks and allowed manual angle annotation using onscreen tools. We focused on three kinematic measures (illustrated in \u003cstrong\u003eFigure\u0026nbsp;1\u003c/strong\u003e and \u003cstrong\u003eFigure\u0026nbsp;2\u003c/strong\u003e):\u003c/p\u003e\n\u003cul type=\"disc\"\u003e\n \u003cli\u003e\u003cstrong\u003ePelvic Obliquity:\u003c/strong\u003e the angle between the line connecting the patient’s left and right hip joints and the horizontal plane. A value of 0° indicates a level pelvis. Negative values denote a drop of the contralateral side, whereas positive values denote a contralateral pelvis rise.\u003c/li\u003e\n \u003cli\u003e\u003cstrong\u003eTrunk Lean:\u003c/strong\u003e the angle between the trunk which is found by pose estimation on OnForm and the true horizontal. Larger positive values indicate the torso leaning further toward the stance leg.\u003c/li\u003e\n \u003cli\u003e\u003cstrong\u003eCoronal Plane Knee Angle Change:\u003c/strong\u003e the difference in the coronal-plane knee alignment between double-leg standing and single-leg stance. This was derived by measuring the hip–knee–ankle angle of the stance limb during bipedal stance (baseline) and again during the single-leg stance, then taking the change (Δ). Knee angles greater than 180° in single-leg stance were described as “varus” by the amount of increase, whereas angles less than 180° were described as “valgus” by the deficit in degrees.\u003c/li\u003e\n\u003c/ul\u003e\n\u003cp\u003eAll study procedures were performed in accordance with institutional and international standards, including the Declaration of Helsinki (2013), the HSE National Consent Policy for Health and Social Care Research (2022), and the Data Protection Act 2018 (Health Research) Regulations 2018 and 2021.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eData Analysis:\u003c/strong\u003e Outcome measures for each trial included pelvic obliquity (most negative / positive value, representing the greatest contralateral drop or hike) and the \u003cem\u003emaximum\u003c/em\u003e trunk lean observed during the 5–10 s hold, as well as the maximum knee angle change. For each patient, values were aggregated by side (left stance and right stance) and the maximum values were noted. Descriptive statistics were used given the small sample. Continuous outcomes are reported as medians with interquartile ranges (IQR) and total ranges. We also counted the number of patients meeting pre-specified threshold criteria for an abnormal Trendelenburg test: specifically, a pelvic drop of ≥4° (i.e. pelvic obliquity ≤ –4°) and a trunk lean of ≥5°. These cut-offs were based on the criteria proposed by Stevens \u003cem\u003eet al.\u003c/em\u003e. All analyses were stratified by group (post-THA vs native hips) for comparison.\u003c/p\u003e"},{"header":"Results","content":"\u003cp\u003e\u003cstrong\u003eFeasibility:\u003c/strong\u003e All 12 patients successfully completed the video-based Trendelenburg assessment on both sides. Two patients (17%) required a single retake due to a brief balance loss, but ultimately every patient yielded an acceptable recording for each limb (100% technical success). The workflow was efficient: video capture took an average of 30.2 seconds (mean) (IQR) of 23.8 to 36.0; range 19 to 42 seconds per patient. Video Analysis took an average of 186.8 seconds (mean) IQR of 179.3 to 194.3; range 168 to 207 seconds), for a total of ~3.5 minutes per patient on average (mean 217 seconds, IQR 206.8 TO 230.3; range 193 to 244 seconds) (Table 1). Video quality was consistently sufficient, \u0026nbsp;the horizontal reference was visible in all clips, key landmarks (pelvis, trunk, knees) remained unobstructed, and no frames were discarded due to tracking loss or occlusion. All patients wore their normal footwear and attire, which did not impede the pose detection process, underscoring the method\u0026rsquo;s practicality in a routine clinic setting.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eTrendelenburg Angle Outcomes:\u003c/strong\u003e \u003cem\u003eTable\u0026nbsp;2\u003c/em\u003e summarizes the kinematic results for each patient. Overall, pelvic tilt deviations were small in this cohort. The median \u003cstrong\u003eworst contralateral pelvic tilt\u003c/strong\u003e (i.e. the most negative pelvic obliquity observed, indicating pelvic drop) was \u003cstrong\u003e0.0\u0026deg;\u003c/strong\u003e\u0026nbsp; (IQR of \u003cstrong\u003e\u0026ndash;1.0\u0026deg; to +1.5\u0026deg;\u003c/strong\u003e; range \u003cstrong\u003e\u0026ndash;5\u0026deg;to +6\u0026deg;)\u003c/strong\u003e. In practical terms, half of the patients did not exceed roughly 1\u0026deg; of pelvic drop or lift. By contrast, \u003cstrong\u003etrunk lean\u003c/strong\u003e was more pronounced: the median maximum trunk lean was \u003cstrong\u003e4.5\u0026deg;\u003c/strong\u003e (IQR 2.8\u0026deg;\u0026ndash;9.0\u0026deg;; range 2\u0026deg;\u0026ndash;10\u0026deg;). The median \u003cstrong\u003ecoronal-plane knee angle change\u003c/strong\u003e (difference between bilateral stance and single-leg stance) was \u003cstrong\u003e3.0\u0026deg;\u003c/strong\u003e (IQR 2.0\u0026deg;\u0026ndash;4.0\u0026deg;; range 1\u0026deg;\u0026ndash;8\u0026deg;).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eGroup Differences:\u003c/strong\u003e When comparing the seven \u003cstrong\u003epost-THA\u003c/strong\u003e patients to the five \u003cstrong\u003enative-hip\u003c/strong\u003e patients, distinct patterns emerged. Post-THA patients showed a greater tendency toward trunk compensation: their median maximal trunk lean was \u003cstrong\u003e9.0\u0026deg;\u003c/strong\u003e (IQR 4.5\u0026deg;\u0026ndash;9.5\u0026deg;) versus \u003cstrong\u003e3.0\u0026deg;\u003c/strong\u003e (IQR 2.0\u0026deg;\u0026ndash;3.0\u0026deg;) in the native-hip group. In fact, \u003cstrong\u003e5 of 7 post-THA\u003c/strong\u003e patients demonstrated a trunk lean \u0026ge;5\u0026deg;, compared to only \u003cstrong\u003e1 of 5\u003c/strong\u003e in the native-hip group. By contrast, pelvic tilt remained near-neutral in both groups (median worst pelvic obliquity 0.0\u0026deg; in post-THA vs +1.0\u0026deg; in native hips), indicating that large contralateral pelvic drops were uncommon across the board. Knee alignment changes were slightly greater in the arthroplasty group (median 3.0\u0026deg; vs 2.0\u0026deg; in natives)\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eThreshold Criteria:\u003c/strong\u003e Applying published biomechanical cut-offs for an abnormal Trendelenburg test, only \u003cstrong\u003e1 patient (8%)\u003c/strong\u003e demonstrated a contralateral pelvic drop of \u0026ge;4\u0026deg; (signifying significant hip abductor insufficiency). In contrast, \u003cstrong\u003e6 patients (50%)\u003c/strong\u003e exhibited an ipsilateral trunk lean \u0026ge;5\u0026deg;; notably, five of these six were in the post-THA cohort. A coronal-plane knee deviation of \u003cstrong\u003e\u0026ge;3\u0026deg;\u003c/strong\u003e was observed in \u003cstrong\u003e8 patients (67%)\u003c/strong\u003e, including 6 of 7 post-THA cases and 2 of 5 native-hip cases. Importantly, the \u003cstrong\u003edirection\u003c/strong\u003e of maximal knee movement was varus (medial collapse) in \u003cstrong\u003eall 12 patients\u003c/strong\u003e \u0026ndash; in other words, every patient\u0026rsquo;s greatest knee excursion involved the knee moving medially. None of the patients had their peak knee change occur in the lateral (valgus) direction during the recorded trials. These objective results (Tables 1 and 2) provide a clear quantitative profile of the Trendelenburg test performance in both patient groups, without any need for subjective grading.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eTable 1. Feasibility results for Trendelenburg video analysis in clinic.\u003c/strong\u003e\u003c/p\u003e\n\u003ctable border=\"1\" cellspacing=\"0\" cellpadding=\"0\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003ePatient\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eAge (Years)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eVideo capture time (seconds)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eAnalysis time (seconds)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eTotal workflow time (seconds)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eRetake required\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eVideos recorded\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eSuccessful analysis\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e71\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\"\u003e\n \u003cp\u003e25\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e168\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\"\u003e\n \u003cp\u003e193\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eNo\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eYes\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e87\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\"\u003e\n \u003cp\u003e20\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e177\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\"\u003e\n \u003cp\u003e197\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eYes\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eYes\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e3\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e49\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\"\u003e\n \u003cp\u003e19\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e174\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\"\u003e\n \u003cp\u003e193\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eNo\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eYes\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e4\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e48\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\"\u003e\n \u003cp\u003e30\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e180\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\"\u003e\n \u003cp\u003e210\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eNo\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eYes\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e5\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e52\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\"\u003e\n \u003cp\u003e35\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e183\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\"\u003e\n \u003cp\u003e218\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eNo\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eYes\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e6\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e78\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\"\u003e\n \u003cp\u003e23\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e192\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\"\u003e\n \u003cp\u003e215\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eYes\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eYes\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e7\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e72\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\"\u003e\n \u003cp\u003e24\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e186\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\"\u003e\n \u003cp\u003e210\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eNo\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eYes\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e8\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e54\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\"\u003e\n \u003cp\u003e35\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e181\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\"\u003e\n \u003cp\u003e216\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eNo\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eYes\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e9\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e60\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\"\u003e\n \u003cp\u003e36\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e201\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\"\u003e\n \u003cp\u003e237\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eNo\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eYes\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e10\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e49\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\"\u003e\n \u003cp\u003e36\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e192\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\"\u003e\n \u003cp\u003e228\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eNo\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eYes\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e11\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e53\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\"\u003e\n \u003cp\u003e42\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e201\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\"\u003e\n \u003cp\u003e243\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eNo\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eYes\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e12\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e50\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\"\u003e\n \u003cp\u003e37\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e207\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\"\u003e\n \u003cp\u003e244\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eNo\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eYes\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp\u003e\u003cstrong\u003eTable 2. Patient Trendelenburg Test Outcomes\u003c/strong\u003e.\u003c/p\u003e\n\u003ctable border=\"1\" cellspacing=\"0\" cellpadding=\"0\" width=\"604\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 36px;\"\u003e\n \u003cp\u003e\u003cstrong\u003ePatient ID\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 29px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eAge (yrs)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 47px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eStatus\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 64px;\"\u003e\n \u003cp\u003e\u003cstrong\u003ePresenting Complaint\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 45px;\"\u003e\n \u003cp\u003e\u003cstrong\u003ePelvic Obliquity\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 44px;\"\u003e\n \u003cp\u003e\u003cstrong\u003ePelvic Obliquity\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 34px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eTrunk Lean\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 42px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eTrunk Lean (R)\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 38px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eLeft Knee Bipedal\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 38px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eLeft Knee Single Stance\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 57px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eLeft Knee Change\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 38px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eRight Knee Bipedal\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 38px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eRight Knee Single Stance\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 57px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eRight Knee Change\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 36px;\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 29px;\"\u003e\n \u003cp\u003e71\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 47px;\"\u003e\n \u003cp\u003ePost-Op THA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 64px;\"\u003e\n \u003cp\u003eRight hip pain\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 45px;\"\u003e\n \u003cp\u003e-1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 44px;\"\u003e\n \u003cp\u003e5\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 34px;\"\u003e\n \u003cp\u003e3\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 42px;\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 38px;\"\u003e\n \u003cp\u003e2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 38px;\"\u003e\n \u003cp\u003e4\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 57px;\"\u003e\n \u003cp\u003e2 Varus\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 38px;\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 38px;\"\u003e\n \u003cp\u003e-2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 57px;\"\u003e\n \u003cp\u003e3 Valgus\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 36px;\"\u003e\n \u003cp\u003e2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 29px;\"\u003e\n \u003cp\u003e87\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 47px;\"\u003e\n \u003cp\u003ePost-Op THA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 64px;\"\u003e\n \u003cp\u003eLeft hip pain\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 45px;\"\u003e\n \u003cp\u003e9\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 44px;\"\u003e\n \u003cp\u003e6\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 34px;\"\u003e\n \u003cp\u003e9\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 42px;\"\u003e\n \u003cp\u003e0\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 38px;\"\u003e\n \u003cp\u003e-2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 38px;\"\u003e\n \u003cp\u003e-5\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 57px;\"\u003e\n \u003cp\u003e3 Valgus\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 38px;\"\u003e\n \u003cp\u003e0\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 38px;\"\u003e\n \u003cp\u003e3\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 57px;\"\u003e\n \u003cp\u003e3 Varus\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 36px;\"\u003e\n \u003cp\u003e3\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 29px;\"\u003e\n \u003cp\u003e49\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 47px;\"\u003e\n \u003cp\u003ePost-Op THA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 64px;\"\u003e\n \u003cp\u003eRight hip revision\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 45px;\"\u003e\n \u003cp\u003e3\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 44px;\"\u003e\n \u003cp\u003e0\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 34px;\"\u003e\n \u003cp\u003e0\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 42px;\"\u003e\n \u003cp\u003e9\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 38px;\"\u003e\n \u003cp\u003e0\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 38px;\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 57px;\"\u003e\n \u003cp\u003e1 Varus\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 38px;\"\u003e\n \u003cp\u003e-8\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 38px;\"\u003e\n \u003cp\u003e-16\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 57px;\"\u003e\n \u003cp\u003e8 Valgus\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 36px;\"\u003e\n \u003cp\u003e4\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 29px;\"\u003e\n \u003cp\u003e48\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 47px;\"\u003e\n \u003cp\u003ePost-Op THA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 64px;\"\u003e\n \u003cp\u003ePost-op (femoral nerve palsy)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 45px;\"\u003e\n \u003cp\u003e0\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 44px;\"\u003e\n \u003cp\u003e-5\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 34px;\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 42px;\"\u003e\n \u003cp\u003e2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 38px;\"\u003e\n \u003cp\u003e-3\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 38px;\"\u003e\n \u003cp\u003e0\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 57px;\"\u003e\n \u003cp\u003e3 Varus\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 38px;\"\u003e\n \u003cp\u003e-4\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 38px;\"\u003e\n \u003cp\u003e-6\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 57px;\"\u003e\n \u003cp\u003e2 Valgus\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 36px;\"\u003e\n \u003cp\u003e5\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 29px;\"\u003e\n \u003cp\u003e52\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 47px;\"\u003e\n \u003cp\u003ePost-Op THA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 64px;\"\u003e\n \u003cp\u003eRight hip pain\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 45px;\"\u003e\n \u003cp\u003e5\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 44px;\"\u003e\n \u003cp\u003e-1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 34px;\"\u003e\n \u003cp\u003e10\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 42px;\"\u003e\n \u003cp\u003e3\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 38px;\"\u003e\n \u003cp\u003e3\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 38px;\"\u003e\n \u003cp\u003e4\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 57px;\"\u003e\n \u003cp\u003e1 Varus\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 38px;\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 38px;\"\u003e\n \u003cp\u003e3\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 57px;\"\u003e\n \u003cp\u003e2 Varus\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 36px;\"\u003e\n \u003cp\u003e6\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 29px;\"\u003e\n \u003cp\u003e78\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 47px;\"\u003e\n \u003cp\u003ePost-Op THA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 64px;\"\u003e\n \u003cp\u003eMoM hip review\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 45px;\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 44px;\"\u003e\n \u003cp\u003e3\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 34px;\"\u003e\n \u003cp\u003e10\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 42px;\"\u003e\n \u003cp\u003e10\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 38px;\"\u003e\n \u003cp\u003e-4\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 38px;\"\u003e\n \u003cp\u003e-7\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 57px;\"\u003e\n \u003cp\u003e3 Valgus\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 38px;\"\u003e\n \u003cp\u003e-1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 38px;\"\u003e\n \u003cp\u003e-4\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 57px;\"\u003e\n \u003cp\u003e3 Valgus\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 36px;\"\u003e\n \u003cp\u003e7\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 29px;\"\u003e\n \u003cp\u003e72\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 47px;\"\u003e\n \u003cp\u003ePost-Op THA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 64px;\"\u003e\n \u003cp\u003eBilateral knee pain\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 45px;\"\u003e\n \u003cp\u003e2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 44px;\"\u003e\n \u003cp\u003e0\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 34px;\"\u003e\n \u003cp\u003e6\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 42px;\"\u003e\n \u003cp\u003e4\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 38px;\"\u003e\n \u003cp\u003e3\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 38px;\"\u003e\n \u003cp\u003e-2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 57px;\"\u003e\n \u003cp\u003e5 Valgus\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 38px;\"\u003e\n \u003cp\u003e4\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 38px;\"\u003e\n \u003cp\u003e-3\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 57px;\"\u003e\n \u003cp\u003e7 Valgus\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 36px;\"\u003e\n \u003cp\u003e8\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 29px;\"\u003e\n \u003cp\u003e54\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 47px;\"\u003e\n \u003cp\u003eNative\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 64px;\"\u003e\n \u003cp\u003eRight hip pain\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 45px;\"\u003e\n \u003cp\u003e2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 44px;\"\u003e\n \u003cp\u003e0\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 34px;\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 42px;\"\u003e\n \u003cp\u003e2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 38px;\"\u003e\n \u003cp\u003e-3\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 38px;\"\u003e\n \u003cp\u003e-4\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 57px;\"\u003e\n \u003cp\u003e1 Valgus\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 38px;\"\u003e\n \u003cp\u003e-3\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 38px;\"\u003e\n \u003cp\u003e-4\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 57px;\"\u003e\n \u003cp\u003e1 Valgus\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 36px;\"\u003e\n \u003cp\u003e9\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 29px;\"\u003e\n \u003cp\u003e60\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 47px;\"\u003e\n \u003cp\u003eNative\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 64px;\"\u003e\n \u003cp\u003eLeft hip pain\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 45px;\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 44px;\"\u003e\n \u003cp\u003e3\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 34px;\"\u003e\n \u003cp\u003e3\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 42px;\"\u003e\n \u003cp\u003e3\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 38px;\"\u003e\n \u003cp\u003e-2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 38px;\"\u003e\n \u003cp\u003e-4\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 57px;\"\u003e\n \u003cp\u003e2 Valgus\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 38px;\"\u003e\n \u003cp\u003e-3\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 38px;\"\u003e\n \u003cp\u003e-4\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 57px;\"\u003e\n \u003cp\u003e1 Valgus\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 36px;\"\u003e\n \u003cp\u003e10\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 29px;\"\u003e\n \u003cp\u003e49\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 47px;\"\u003e\n \u003cp\u003eNative\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 64px;\"\u003e\n \u003cp\u003eRight hip pain\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 45px;\"\u003e\n \u003cp\u003e4\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 44px;\"\u003e\n \u003cp\u003e4\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 34px;\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 42px;\"\u003e\n \u003cp\u003e3\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 38px;\"\u003e\n \u003cp\u003e-2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 38px;\"\u003e\n \u003cp\u003e-3\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 57px;\"\u003e\n \u003cp\u003e1 Valgus\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 38px;\"\u003e\n \u003cp\u003e-2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 38px;\"\u003e\n \u003cp\u003e-4\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 57px;\"\u003e\n \u003cp\u003e2 Valgus\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 36px;\"\u003e\n \u003cp\u003e11\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 29px;\"\u003e\n \u003cp\u003e53\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 47px;\"\u003e\n \u003cp\u003eNative\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 64px;\"\u003e\n \u003cp\u003eRight hip pain\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 45px;\"\u003e\n \u003cp\u003e0\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 44px;\"\u003e\n \u003cp\u003e-1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 34px;\"\u003e\n \u003cp\u003e2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 42px;\"\u003e\n \u003cp\u003e2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 38px;\"\u003e\n \u003cp\u003e-5\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 38px;\"\u003e\n \u003cp\u003e-5\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 57px;\"\u003e\n \u003cp\u003e0\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 38px;\"\u003e\n \u003cp\u003e-1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 38px;\"\u003e\n \u003cp\u003e-5\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 57px;\"\u003e\n \u003cp\u003e4 Valgus\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 36px;\"\u003e\n \u003cp\u003e12\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 29px;\"\u003e\n \u003cp\u003e50\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 47px;\"\u003e\n \u003cp\u003eNative\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 64px;\"\u003e\n \u003cp\u003eLeft hip pain\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 45px;\"\u003e\n \u003cp\u003e3\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 44px;\"\u003e\n \u003cp\u003e6\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 34px;\"\u003e\n \u003cp\u003e5\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 42px;\"\u003e\n \u003cp\u003e6\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 38px;\"\u003e\n \u003cp\u003e3\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 38px;\"\u003e\n \u003cp\u003e-1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 57px;\"\u003e\n \u003cp\u003e4 Valgus\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 38px;\"\u003e\n \u003cp\u003e-1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 38px;\"\u003e\n \u003cp\u003e-3\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 57px;\"\u003e\n \u003cp\u003e2 Valgus\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cem\u003e*\u0026nbsp;\u003c/em\u003e\u003c/strong\u003e\u003cem\u003eVarus Positive Angle, Valgus Negative Angle.\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003e\u003cem\u003e\u0026nbsp; \u0026nbsp;Pelvic Obliquity Positive \u0026ndash; Hip Hike, Negative \u0026ndash; Hip Drop\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003e\u003cem\u003e\u0026nbsp;Measurements in Degrees\u0026nbsp;\u003c/em\u003e(x\u0026deg;)\u003c/p\u003e"},{"header":"Discussion","content":"\u003cp\u003eThis feasibility study demonstrates that a single-camera AI-based video system can successfully quantify key angles during the Trendelenburg test in a routine clinical setting. All participants were able to complete the protocol, and the automated pose analysis yielded objective measures of pelvic tilt, trunk lean, and knee alignment within minutes. The overall magnitude of observed pelvic drop was quite small, while trunk lean was frequently elevated, especially among post-THA patients, this is possibly consistent with known compensatory patterns in hip abductor weakness. These results support the notion that video-based measurements can capture clinically relevant variations in Trendelenburg test performance that may be subtle or missed by traditional visual observation alone (3).\u003c/p\u003e\n\u003cp\u003eA key finding is that \u003cstrong\u003epost-THA patients exhibited substantially greater trunk lean\u003c/strong\u003e than patients with native hip pathology, despite having similarly minimal pelvic drop. This suggests that many post-arthroplasty patients rely on torso compensation (leaning the upper body toward the stance side) to stabilize the pelvis, rather than allowing a pelvic sag on the unsupported side. Prior research indicates that hip abductor strength often remains deficient after THA and that certain surgical/anatomical factors (such as a more superolateral hip centre of rotation or reduced femoral offset) can increase the mechanical demand on the abductors (4, 5). Both of these aspects can contribute to a positive Trendelenburg sign following hip replacement. Leaning the trunk toward the stance leg is a well-recognized compensatory strategy: it reduces the lever arm of body weight, thereby partially offloading the weak hip abductors. Our data align with this mechanism, the THA group in our sample showed frequent and large trunk deviations, whereas large contralateral pelvic drops were rarely seen. In essence, many post-THA patients achieved coronal-plane pelvic stability by substituting trunk lean for pelvic tilt.\u003c/p\u003e\n\u003cp\u003eFrom a clinical perspective, quantifying these angular deviations offers immediate value. Instead of simply noting a binary “Trendelenburg sign positive/negative,” clinicians can document the \u003cstrong\u003edegree\u003c/strong\u003e of pelvic drop and lateral trunk lean in objective terms. These metrics enable tracking changes over time and can inform targeted interventions. For example, a \u003cstrong\u003epelvic drop beyond ~ 4°\u003c/strong\u003e may be interpreted as a significant abductor deficit, which could prompt focused gluteal strengthening and balance training. Similarly, a \u003cstrong\u003etrunk lean above ~ 5°\u003c/strong\u003e suggests reliance on a compensation strategy; such patients may benefit from gait retraining to reduce excessive torso sway. In our study, half of all patients, and a large majority of the post-THA subgroup, exceeded the 5° trunk lean threshold, underscoring how common this compensation was in the post-operative population.\u003c/p\u003e\n\u003cp\u003eThe inclusion of a knee alignment metric is a novel addition to Trendelenburg test assessment that warrants attention. Although we did not directly measure knee joint moments, a shift of the knee is biomechanically linked to increased loading of the contralateral compartment of the knee. Prior studies have shown that greater lateral trunk lean and contralateral pelvic drop both increase the external knee adduction moment during gait and single-leg stance (2).Therefore, identifying a stance-phase drift of ≥3° on the Trendelenburg test can alert clinicians to potential knee overload in patients with hip abductor dysfunction. This information may guide early preventive interventions such as physiotherapy and advising on limb alignment and weight distribution during daily single-leg tasks, to mitigate undue stress on the knee joint.\u003c/p\u003e\n\u003cp\u003eIt is important to recognize the \u003cstrong\u003elimitations\u003c/strong\u003e of a 2D single-camera approach for Trendelenburg evaluation. Measurement error in markerless video analysis must be considered when interpreting small angle differences. Experimental work using goniometry suggests that changes smaller than about \u003cstrong\u003e4°\u003c/strong\u003e fall within the typical measurement error for pelvic tilt assessments\u0026nbsp;(6). Likewise, contemporary 2D pose-estimation algorithms have an angle accuracy on the order of ~ 2–3° under ideal conditions. In this context, a difference of only a couple of degrees (for example, a change in pelvic drop from 0° to 2°) should be interpreted with caution, as it might reflect normal variability or minor asymmetry rather than a meaningful change in abductor function. The literature recommends using a conservative threshold (around 4°) for defining true deterioration or improvement in Trendelenburg angles, and repeating the test if values are borderline or unexpected.\u003c/p\u003e\n\u003cp\u003eThere are additional technical and design considerations. Because 2D analysis accuracy depends on camera perspective, careful standardization of the setup is required. We kept the camera perpendicular to the coronal plane and used a horizontal reference bar to calibrate pelvic level, but even slight camera tilt or out-of-plane movement by the patient could introduce error (7). We did not strictly control lighting or clothing colour; however, prior evaluations suggest these factors have minimal impact on gross coronal-plane kinematics in clinic settings. All patients performed the test in their own shoes, which might subtly affect balance or knee mechanics, but this choice reflects real-world practice (future studies might adopt a barefoot protocol for consistency). Another limitation is the relatively \u003cstrong\u003esmall sample size\u003c/strong\u003e\u003cstrong\u003e.\u003c/strong\u003e With only 12 patients (7 post-THA and 5 native hips), our study was not powered for formal statistical comparisons between groups. We treated this research as a proof-of-concept feasibility study, so no inferential statistics were performed and we did not attempt to correlate the angle measurements with clinical outcomes like pain or functional scores. The brief duration of each single-leg stance trial (5–10 seconds) is another limitation – shorter than the 30-second holds used in some reference protocols – which may have reduced our ability to observe fatigue-related pelvic drop or trunk sway. These limitations should be kept in mind when generalizing the findings.\u003c/p\u003e\n\u003cp\u003eDespite the above limitations, our findings are in agreement with, and extend, the existing literature on the Trendelenburg test. They reinforce that an objective, video-based approach is both feasible and informative in everyday practice. The pattern we observed,\u0026nbsp;\u003cstrong\u003eminimal pelvic drop but frequent trunk lean in post-THA patients\u003c/strong\u003e echoes the logic of prior biomechanical analyses and matches the quantitative criteria proposed by Stevens \u003cem\u003eet al.\u003c/em\u003e for an abnormal Trendelenburg response. Using Stevens’ numeric framework as a reference, we found that many patients in our cohort (especially those post-THA) would be classified as having an atypical Trendelenburg result primarily due to excessive trunk lean rather than pelvic drop. This suggests that focusing only on contralateral pelvic drop (the traditional clinical sign) would \u003cstrong\u003eunderestimate the prevalence of gait compensation\u003c/strong\u003ein post-THA individuals. Furthermore, the inclusion of the simple knee alignment measure provides a more holistic view of single-leg stance, capturing downstream effects on the lower limb kinetic chain that might have implications for long-term joint health.\u003c/p\u003e\n\u003cp\u003eIn summary, the corrected results indicate that AI-assisted video analysis can quantitatively distinguish how different patient groups manage the Trendelenburg test: post-THA patients tend to engage a trunk lean compensation more than native-hip patients, even though overt pelvic dropping is minimal in both groups. This quantitative insight, alongside established thresholds from the literature, strengthens the clinical interpretation of the Trendelenburg test and lays the groundwork for improved patient assessment and monitoring.\u003c/p\u003e\n\u003ch2\u003e\u003cstrong\u003eFuture Directions\u003c/strong\u003e\u003c/h2\u003e\n\u003cp\u003eBuilding on this feasibility study, several next steps are recommended to enhance and validate the approach. \u003cstrong\u003eFirst\u003c/strong\u003e, a larger-scale study should be conducted to gather more extensive data and enable direct comparison to established 3D-derived criteria. \u003cstrong\u003e\u0026nbsp;Second\u003c/strong\u003e\u003cstrong\u003e,\u003c/strong\u003e future research should incorporate additional clinical correlates to explore the underlying causes of the observed compensations. Measuring \u003cstrong\u003ehip abductor strength\u003c/strong\u003e(e.g. with hand-held dynamometry) and evaluating post-THA radiographic parameters (such as the reconstructed hip centre position or femoral offset on X-ray) could help discern whether large trunk leans are more attributable to residual muscle weakness or to altered hip biomechanics (or both). \u003cstrong\u003eFinally\u003c/strong\u003e, to facilitate clinical adoption of this technique, an effort should be made to develop a simple \u003cstrong\u003ereporting template or automated output\u003c/strong\u003e for the Trendelenburg test. Such a template could succinctly record the three key angles – pelvic drop, trunk lean, and knee varus/valgus change, alongside notes on patient symptoms (pain, balance, etc.), providing clinicians with a summary that is far more informative than the traditional binary outcome. These future steps will help validate the utility of AI-assisted Trendelenburg assessment and potentially expand its role in both research and routine orthopaedic practice.\u003c/p\u003e"},{"header":"Conclusion","content":"\u003cp\u003eIn summary, this study found that an AI-assisted, single-camera video workflow is a feasible and informative method to quantify the Trendelenburg test in an outpatient setting. The system produced objective measurements of pelvic obliquity, trunk lean, and knee alignment within minutes, with 100% technical success in our sample. An objective video-based Trendelenburg assessment has the potential to improve clinical documentation and patient monitoring. Future work will focus expanding it to larger cohorts to establish its utility as a routine clinical tool.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eEthical Approval\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eEthical approval was granted by the Clinical Research Ethics Committee (CREC) of Galway University Hospitals, reference number C.A. 3416.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFunding Declaration\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eNo funding was received for the conduct of this study.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConflicts of Interest\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe authors declare that they have no conflicts of interest relevant to the content of this manuscript.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eMcCarney L, Andrews A, Henry P, Fazalbhoy A, Selva Raj I, Lythgo N, et al. Determining Trendelenburg test validity and reliability using 3-dimensional motion analysis and muscle dynamometry. Chiropr Man Therap. 2020;28(1):53.\u003c/li\u003e\n\u003cli\u003eTakacs J, Hunt MA. The effect of contralateral pelvic drop and trunk lean on frontal plane knee biomechanics during single limb standing. J Biomech. 2012;45(16):2791-6.\u003c/li\u003e\n\u003cli\u003eStevens WR, Jo CH, Tulchin-Francis K. Clinically derived biomechanical criteria for the Trendelenburg test. Clin Biomech (Bristol). 2020;78:105066.\u003c/li\u003e\n\u003cli\u003eIsmailidis P, Kvarda P, Vach W, Cadosch D, Appenzeller-Herzog C, M\u0026uuml;ndermann A. Abductor Muscle Strength Deficit in Patients After Total Hip Arthroplasty: A Systematic Review and Meta-Analysis. J Arthroplasty. 2021;36(8):3015-27.\u003c/li\u003e\n\u003cli\u003eKiyama T, Naito M, Shitama H, Maeyama A. Effect of superior placement of the hip center on abductor muscle strength in total hip arthroplasty. J Arthroplasty. 2009;24(2):240-5.\u003c/li\u003e\n\u003cli\u003eYoudas JW, Mraz ST, Norstad BJ, Schinke JJ, Hollman JH. Determining meaningful changes in pelvic-on-femoral position during the Trendelenburg test. J Sport Rehabil. 2007;16(4):326-35.\u003c/li\u003e\n\u003cli\u003eBaldinger M, Reimer LM, Senner V. Influence of the Camera Viewing Angle on OpenPose Validity in Motion Analysis. Sensors (Basel). 2025;25(3).\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":true,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"scientific-reports","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"scirep","sideBox":"Learn more about [Scientific Reports](http://www.nature.com/srep/)","snPcode":"","submissionUrl":"","title":"Scientific Reports","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Scientific Reports","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"Trendelenburg test, Hip abductor function, Total hip arthroplasty, Pelvic obliquity, Trunk lean, Artificial intelligence, Markerless motion analysis","lastPublishedDoi":"10.21203/rs.3.rs-7680719/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-7680719/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003e\u003cstrong\u003eBackground:\u003c/strong\u003e\u003cbr\u003e\nThe Trendelenburg test is widely used to assess hip abductor function but is traditionally graded as a binary sign, limiting sensitivity and reliability. Compensatory trunk lean may mask pelvic drop and further reduces diagnostic accuracy. Advances in artificial intelligence (AI) and markerless motion capture provide an opportunity to quantify pelvic obliquity, trunk lean, and knee alignment objectively using simple video-based systems.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eMethods:\u003c/strong\u003e\u003cbr\u003e\nA cross-sectional feasibility study was conducted in an elective hip clinic. Twelve adult patients (seven post-total hip arthroplasty, five with osteoarthritis, performed bilateral single-leg stance while recorded by a single posterior camera (4K, 60 fps). Videos were analysed using an AI-based application (OnForm) to extract pelvic obliquity, trunk lean, and knee alignment change. Feasibility metrics included analysis success rate, retakes required, and workflow timing. Outcomes were reported as medians with interquartile ranges (IQR) and ranges.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eResults:\u003c/strong\u003e\u003cbr\u003e\nAll patients completed the protocol with analysable recordings (100% technical success). The mean total workflow time was 217 seconds (IQR 206.8–230.3; range 193–244). Median worst pelvic obliquity was 0.0° (IQR –1.0° to +1.5°; range –5° to +6°). Median trunk lean was 4.5° (IQR 2.8°–9.0°; range 2°–10°). Median knee alignment change was 3.0° (IQR 2.0°–4.0°; range 1°–8°). Trunk lean ≥5° was seen in 6 patients (5/7 post-THA), while significant pelvic drop (≤ –4°) occurred in only one patient. Knee deviations ≥3° were observed in 8 patients (67%).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConclusions:\u003c/strong\u003e\u003cbr\u003e\nAI-assisted single-camera analysis of the Trendelenburg test is feasible, rapid, and clinically informative. The method consistently quantified pelvic, trunk, and knee angles, revealing that post-THA patients frequently compensate with trunk lean rather than pelvic drop. These findings support the integration of objective video-based Trendelenburg assessment into outpatient orthopaedic practice.\u003c/p\u003e","manuscriptTitle":"AI assisted video analysis of the Trendelenburg Test: A Feasibility Study","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-10-21 23:41:40","doi":"10.21203/rs.3.rs-7680719/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2025-11-24T05:32:37+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-11-14T03:21:18+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-11-07T01:35:43+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"87684063309720064019247249975003488218","date":"2025-11-05T02:11:48+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"271590088262681800117575682725817303244","date":"2025-11-01T14:01:38+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-10-22T21:07:59+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"51192818672922782074463369366877366496","date":"2025-10-10T22:16:26+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2025-10-08T20:53:05+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2025-10-08T20:49:59+00:00","index":"","fulltext":""},{"type":"editorInvited","content":"","date":"2025-10-07T14:02:08+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2025-10-06T21:29:19+00:00","index":"","fulltext":""},{"type":"submitted","content":"Scientific Reports","date":"2025-10-06T15:49:38+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"scientific-reports","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"scirep","sideBox":"Learn more about [Scientific Reports](http://www.nature.com/srep/)","snPcode":"","submissionUrl":"","title":"Scientific Reports","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Scientific Reports","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"e62d893c-f731-465d-a91c-c1a07d4825d0","owner":[],"postedDate":"October 21st, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"published-in-journal","subjectAreas":[{"id":56590233,"name":"Health sciences/Anatomy"},{"id":56590234,"name":"Health sciences/Diseases"},{"id":56590235,"name":"Health sciences/Health care"},{"id":56590236,"name":"Health sciences/Medical research"}],"tags":[],"updatedAt":"2026-02-23T16:08:40+00:00","versionOfRecord":{"articleIdentity":"rs-7680719","link":"https://doi.org/10.1038/s41598-026-38980-z","journal":{"identity":"scientific-reports","isVorOnly":false,"title":"Scientific Reports"},"publishedOn":"2026-02-16 15:57:54","publishedOnDateReadable":"February 16th, 2026"},"versionCreatedAt":"2025-10-21 23:41:40","video":"","vorDoi":"10.1038/s41598-026-38980-z","vorDoiUrl":"https://doi.org/10.1038/s41598-026-38980-z","workflowStages":[]},"version":"v1","identity":"rs-7680719","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-7680719","identity":"rs-7680719","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}