Background: While large language models (LLMs) have made a tremendous impact on scientific workflow, there are also serious concerns about ethical integrity and nondiverse written language in medical and orthopaedic literature. Recent studies have yielded conflicting results regarding the involvement of artificial intelligence (AI) in writing processes. Purpose: To (1) determine the extent of AI involvement in American Journal of Sports Medicine (AJSM) abstracts across 4 different time periods; (2) to determine whether AI usage differs based on authorship nativity. Study Design: Cross-sectional study. Methods: We retrospectively reviewed abstracts from AJSM and organized these into 4 periods. We chose the year 2000 as a human baseline to establish a threshold for AI involvement. We also selected the years 2015 and 2020 to observe changes in AI probability scores. Considering the release of ChatGPT in November 2022, we included abstracts from 2023 to 2025 to determine the increase in AI usage. We classified articles into native English-speaking, non-native English-speaking, and mixed groups to determine authorship nativity based on institutional affiliations. We used Originality.AI to report AI probability scores. We evaluated differences across years and nativity groups using the Kruskal-Wallis and post hoc tests. Results: Across the 6 time periods, 1885 abstracts were analyzed. AI probability scores from the year 2000 (n = 124; mean, 1.68%) were significantly lower than all subsequent years ( P < .001 for all pairwise comparisons). Scores remained stable between 2015 (n = 348; mean, 9.39%) and 2020 (n = 383; mean, 8.91%; P ≥ .999). However, scores began a rapid and significant increase starting in 2023 (n = 345; mean 15.48%) and continued to rise through 2025 (n = 323; mean, 30.6%; P < .001) compared with all previous years. The percentage of abstracts exceeding the threshold increased consistently throughout the years, from 4% in the 2000 baseline to 62.5% in 2025. When nativity groups were compared within the Generative AI Era, no statistically significant differences in AI probability scores were observed in 2023 ( P = .364), 2024 ( P = .053), or 2025 ( P = .454). Conclusion: Our study demonstrated that the prevalence of AI-like linguistic patterns in AJSM abstracts has increased since the release of ChatGPT. AI adoption across all nativity groups suggests that AI has become a universal standard for efficient article writing, regardless of the language barrier encountered by non-native speakers.
Background: Medial meniscus posterior root tear (MMPRT) is recognized as one of the leading causes of knee osteoarthritis. Given the detrimental effects of MMPRT on knee kinematics and the associated clinical consequences, substantial efforts have been directed toward improving the understanding and management of MMPRT. Purpose: To develop and validate an artificial intelligence (AI)–based prediction model for patient-specific risk assessment of clinical failure at 2 and 5 years after nonsurgical treatment of MMPRT. Study Design: Case-control study; Level of evidence, 3. Methods: The authors retrospectively reviewed a prospectively collected database of 233 patients who underwent nonsurgical treatment for MMPRT between 2006 and 2020. Patient descriptive characteristics, clinical data, and imaging variables were evaluated for their association with clinical failure, defined as conversion to total knee arthroplasty or corrective osteotomy at 2- and 5-year follow-up. Five conventional machine learning models, including Elastic Net logistic regression, multilayer perceptron, support vector machine, random forest, and Extreme Gradient Boosting, as well as a proposed deep learning model, the Grouped Graph Attention (GGAT) network, were developed and internally validated to predict clinical failure. Results: During follow-up, clinical failure occurred in 36 of 233 patients (15.5%) at 2 years and in 54 of 233 patients (23.2%) at 5 years. The deep learning–based GGAT model demonstrated better overall predictive performance compared with conventional machine learning models at both the 2- and 5-year follow-up, with test set accuracy of 0.83 to 0.91, precision of 0.87 to 0.91, sensitivity of 0.92 to 1.00, F1 score of 0.89 to 0.95, Brier score of 0.09 to 0.16, and areas under the receiver operating characteristic curve of 0.69 to 0.77. The most influential predictors of clinical failure included baseline mechanical hip-knee-ankle angle, symptom duration, body mass index, age, bone marrow edema, lateral distal femoral angle, subchondral insufficiency fracture of the knee, cartilage lesion, effusion grade, and medial proximal tibial angle. Conclusion: The deep learning–based GGAT network demonstrated accurate prediction of clinical failure at both 2 and 5 years after nonsurgical treatment of MMPRT. These findings underscore the potential value of deep learning–based risk assessment in supporting clinical decision-making and patient counseling.
Background: Beach volleyball is a globally popular sport that places unique biomechanical and tactical demands on athletes. Although there is some literature on the epidemiology of indoor volleyball injuries, the injury patterns associated with beach volleyball remain comparatively understudied. Purpose/Hypothesis: The purpose of this study was to characterize the epidemiology, mechanisms, and treatment of musculoskeletal injuries sustained by beach volleyball athletes. The authors hypothesized that knee and shoulder injuries would represent the most common injury locations and that the mechanisms of injury would differ from those reported in indoor volleyball. Study Design: Case series; Level of evidence, 4. Methods: A retrospective cross-sectional case series was performed utilizing an institutional electronic medical record and internal search engine from January 1, 2000, to December 31, 2025. Patient characteristics, anatomic injury location, mechanism of injury and treatment characteristics were analyzed. Injury distributions were compared between male and female athletes. Continuous variables are summarized using means and standard deviations. Categorical variables are summarized using counts and percentages. Results: A total of 175 injuries sustained in 160 patients were included in this study. Of these injuries, 102 (58.3%) occurred in male cases and 73 (41.7%) in female cases ( P = .63). The most common injury locations were the knee (74; 42.3%), wrist/hand (29; 16.6%), and shoulder (25; 14.3%). The distribution of injuries by anatomic location did not differ significantly between male and female athletes in any anatomic location. The most common mechanism among all injuries was a noncontact twisting knee event while pursuing the ball (26.5%), followed by overuse injuries (16.2%), ball contact while blocking or spiking (16.2%), and landing from a jump (14.5%). Among injuries with available treatment data, 59 of 168 (35.1%) required surgical intervention, the most common of which were anterior cruciate ligament reconstructions (n = 21) and meniscal surgeries (n = 18). Conclusion: Our study demonstrated that knee injuries were the most common injuries sustained by beach volleyball players. Acute injuries frequently occurred while pursuing the ball, highlighting the unique movement and court coverage demands of beach volleyball. These findings contribute to the limited epidemiological literature on beach volleyball injuries and have the potential to guide future injury prevention strategies.
Background: While large language models (LLMs) have made a tremendous impact on scientific workflow, there are also serious concerns about ethical integrity and nondiverse written language in medical and orthopaedic literature. Recent studies have yielded conflicting results regarding the involvement of artificial intelligence (AI) in writing processes. Purpose: To (1) determine the extent of AI involvement in American Journal of Sports Medicine (AJSM) abstracts across 4 different time periods; (2) to determine whether AI usage differs based on authorship nativity. Study Design: Cross-sectional study. Methods: We retrospectively reviewed abstracts from AJSM and organized these into 4 periods. We chose the year 2000 as a human baseline to establish a threshold for AI involvement. We also selected the years 2015 and 2020 to observe changes in AI probability scores. Considering the release of ChatGPT in November 2022, we included abstracts from 2023 to 2025 to determine the increase in AI usage. We classified articles into native English-speaking, non-native English-speaking, and mixed groups to determine authorship nativity based on institutional affiliations. We used Originality.AI to report AI probability scores. We evaluated differences across years and nativity groups using the Kruskal-Wallis and post hoc tests. Results: Across the 6 time periods, 1885 abstracts were analyzed. AI probability scores from the year 2000 (n = 124; mean, 1.68%) were significantly lower than all subsequent years ( P < .001 for all pairwise comparisons). Scores remained stable between 2015 (n = 348; mean, 9.39%) and 2020 (n = 383; mean, 8.91%; P ≥ .999). However, scores began a rapid and significant increase starting in 2023 (n = 345; mean 15.48%) and continued to rise through 2025 (n = 323; mean, 30.6%; P < .001) compared with all previous years. The percentage of abstracts exceeding the threshold increased consistently throughout the years, from 4% in the 2000 baseline to 62.5% in 2025. When nativity groups were compared within the Generative AI Era, no statistically significant differences in AI probability scores were observed in 2023 ( P = .364), 2024 ( P = .053), or 2025 ( P = .454). Conclusion: Our study demonstrated that the prevalence of AI-like linguistic patterns in AJSM abstracts has increased since the release of ChatGPT. AI adoption across all nativity groups suggests that AI has become a universal standard for efficient article writing, regardless of the language barrier encountered by non-native speakers.
Objective: Osteochondritis dissecans (OCD) lesions frequently present at the capitellum of the elbow, particularly among athletes who participate in overhead sports or impact load through their upper extremity. Bilateral elbow OCD lesions are less common rare and can significantly impact sport participation and elbow function. Previous studies have shown that unique capitellar OCD characteristics are associated with specific sport participation such as gymnasts having more posterior lesions with lesser lesion containment. However, little is known about the presenting characteristics of patients with bilateral elbow OCD lesions. We hypothesized that tumbling athletes including gymnasts, cheerleaders, dancers would have an increased propensity for bilateral OCD lesions. Methods: A retrospective review was performed of a consecutive series of pediatric and adolescent patients with elbow OCD lesions who presented to a single tertiary care pediatric hospital between 2009-2025. Patients were identified using CPT and ICD codes. Data regarding patient demographics, sports participation, lesion characteristics on radiographs and magnetic resonance imaging, and timeline of presentation were collected for patients with bilateral elbow OCD lesions. Standard statistical analyses were performed to compare unilateral and bilateral lesions (p<0.05). Results: A total of 279 consecutive patients with a diagnosis of elbow OCD with a mean age of 13.4 (8.0-19.1) years were included, of which 10.3% (n=29) were determined to have bilateral lesions. Patients with bilateral lesions were younger compared to unilateral lesions (12.9 vs 13.4, p<0.001). Compared to unilateral lesions, bilateral lesions were observed at higher rates among females (79.43% vs 38.2%, p<0.001) and in tumbling athletes (72.4% vs 32.8%, p<0.001). Bilateral lesions were noted at a mean sagittal inclination angle of 38 (±14.1) degrees, indicating that these lesions tended to occur on the posterior aspect of the capitellum. Of the 29 bilateral patients, only 7 (24.1%) had bilateral lesions concomitantly identified at initial presentation, while 22 (75.9%) had their bilateral lesions found sequentially. The majority (26/29, 90.0%) of the secondary lesions were identified after becoming symptomatic, whereas 3 (10.3%) secondary lesions were noted incidentally. Conclusions: In our series, bilateral elbow OCD occurred at an incidence of 10.3%; however, the incidence was over double in tumbling athletes. Bilateral lesions were more common in young females. Due to the high level of prevalence in tumbling athletes, consideration should be given for contralateral elbow screening at initial presentation in these patients, and screening should be promptly performed at the initiation of contralateral symptoms.
Objective: Non-operative management remains the standard of care for first-time patellar dislocation in the absence of an osteochondral fracture or loose body. Extensive research has focused on identifying patients at high risk for recurrence and subsequent surgical intervention; however, limited data exist on the outcomes of individuals who are managed non-operatively and do not report a subsequent dislocation. Limited prior work has demonstrated poorer patient-reported outcomes in this group compared to age-matched controls. The goal of this study is to compare outcomes of this population with patients treated operatively following a first-time patellar dislocation. We hypothesize that these groups will exhibit similar patient-reported outcomes (PROs) within two years of initial treatment. Methods: The study population consisted of a subset of patients enrolled from the prospectively collected multi-center JUPITER (Justifying Patellar Instability by Results) cohort study. Twenty surgeons from eleven academic centers throughout the United States enrolled patients with patellar instability and treated them per their typical clinical practice. The non-operative group included patients treated non-operatively for a first-time patellar dislocation from January 2017 through July 2022 that reported no further patellar dislocations over the next two years. The surgical group included patients treated operatively for a first-time patellar dislocation during the same period. Exclusion criteria included patients who experienced recurrent instability and patients who were missing discrete treatment data. Baseline demographic (age, sex, BMI) and physical exam data (Beighton mobility score, pre-operative J-sign, and pre-operative apprehension testing) were compared between groups and patient-reported outcomes (KOOS, Pedi-FABS, Pedi-IKDC, Banff Patellar Instability Index 2.0 (BPII 2.0), and Kujala) were compared between groups at baseline, one, and two years following presentation. Results: A total of 630 patients experienced a first-time patellar dislocation without recurrence during the study period, including 343 in the non-operative group and 287 in the operative group. There were no significant differences between groups with regards to sex or BMI, however the initial non-operative group was significantly older (16.6 vs. 15.8 years, p< 0.05) and had significantly lower Beighton scores than the initial operative group (3.1 vs. 3.8, p< 0.05). The initial non-operative group also had significantly lower rates of patellar apprehension (70% vs. 82%, p<0.005) as well as lower rates of J-sign seen on presentation than the initial operative group (53% vs 71% vs p<0.005). Baseline patient-reported outcomes (PROs) were significantly higher for the non-operative group than the operative group for KOOS-QOL, KOOS-Symptoms, Pedi-IKDC, Kujala, and BPII 2.0 (p <0.05 ). There were no statistically significant differences between PROs between groups at one or two years following initial presentation. The RTS rate was comparable (84% vs. 78%, p = 0.36) The change in PROs from baseline to two years was significantly greater for the operative group (p<0.05) for Pedi IKDC, Kujala, and BPII 2.0. Conclusions: Patients treated non-operatively following a first-time patellar dislocation who do not experience a recurrent dislocation demonstrate similar patient-reported outcomes scores to those treated operatively following a first-time dislocation.
Background: The quadriceps tendon (QT) graft is increasingly popular in primary anterior cruciate ligament reconstruction (ACLR). However, postoperative quadriceps strength can be significantly weakened after use of a QT graft, compared to using hamstring tendon or bone–patellar tendon–bone grafts. There currently exists no continuous method to predict QT graft size based on preoperative magnetic resonance imaging (MRI) measurements; only dichotomous methods exist at limited cutoff values. Purpose: To (1) assess the accuracy of preoperative MRI QT thickness in predicting graft size dichotomously (<10 mm vs ≥10 mm), (2) assess the accuracy of the formula 2 ×√( WT /π) in predicting graft size continuously (within ±1.0 mm of the actual graft diameter), and (3) determine the interrater reliability of both methods; a minimum threshold of 10-mm QT graft diameter was selected given the young, high-demand athletic population in this cohort. Study Design: Cross-sectional study; Level of evidence, 3. Methods: This is a retrospective review of 130 patients who underwent primary ACLR with full-thickness QT autografts at a single tertiary institution from January 1, 2022, to July 17, 2025. A novel, reproducible method to measure QT thickness on the MRI sagittal view at the most proximal patellar attachment site was developed. Two blinded reviewers independently measured QT thickness. Receiver operating characteristic (ROC) analysis was performed to determine the optimal threshold for predicting a final graft diameter <10 mm, with area under the curve (AUC) as the primary measure of discriminative ability. Sensitivity and specificity at the optimal threshold were additionally reported given their direct clinical relevance to intraoperative decision-making at a fixed cutoff. Results: The mean age was 16.4 years (95% CI, 15.9-16.9 years), with 59 females and 71 males. ROC analysis identified a preoperative QT thickness <8.3 mm as the optimal threshold for predicting a final graft diameter <10 mm, with an AUC of 0.818, a sensitivity of 98.2%, and a specificity of 39.1%. The authors applied the formula 2 ×√( WT /π), demonstrating how graft diameter can be predicted from preoperative MRI, where T is the measured preoperative MRI graft thickness and W is the harvested graft width. Using this formula, the authors found that 90.0% of grafts were within ±1.0 mm of the actual graft diameter. Conclusion: The authors present a novel, numerically continuous way of predicting harvested quadriceps graft size. A preoperative QT thickness <8.3 mm is 98.2% sensitive and 39.1% specific in predicting a final graft diameter <10 mm. Personalizing the width of the harvested QT according to patient's tendon thickness and surgeon's desired graft size may be considered to minimize unnecessary over- or underharvesting of QT in the future.
Background: Literature on long-term return-to-sport (RTS) rates in athletes with borderline hip dysplasia (BHD) is limited. Purpose: To examine patient outcomes and RTS rates in athletes with BHD who underwent primary hip arthroscopy with capsular plication at a minimum 10-year follow-up. Study Design: Cohort study; Level of evidence, 3. Methods: Data were retrospectively reviewed for all patients with BHD (lateral center-edge angle 18°-25°) who underwent primary hip arthroscopy with labral treatment and capsular plication between September 2008 and May 2014. Athletes in the BHD group were propensity matched 1:1 to a control group of patients with normal acetabular coverage based on age, sex, body mass index, follow-up time, Tönnis grade, and preoperative sport competition level. Patient-reported outcome scores (modified Harris Hip Score, Non-Arthritic Hip Score, 12-item International Hip Outcome Tool, Hip Outcome Score Sports Specific Subscale, visual analog scale for pain, and satisfaction), RTS, and continuation-of-sport rates were also analyzed. Results: There were 50 patients (53 hips) in the BHD group and 53 patients (53 hips) in the control group. Both groups demonstrated similar increases in all PROs from preoperative to 10-year follow-up and favorable RTS rates. A total of 46 patients in the BHD group attempted to RTS and 40 (87.0%) were successful while 48 patients in the control group attempted to RTS and 41 (85.4%) were successful at the short to midterm. Of the patients who successfully returned to sport, 28 from the BHD cohort (70.0%) and 30 (73.2%) from the control cohort attempted to continue playing their respective sport 10 years postoperatively. Of these, 23 patients (82.1%) from the BHD cohort and 25 patients (83.3%) from the control cohort successfully continued their sport at a minimum of 10 years postoperatively. Over the course of 10 years of follow-up, 8 patients (16%) out of 50 in the BHD group at some point discontinued their sport, not due to their hip, but because of lifestyle changes or loss of interest. Conclusion: Athletes with BHD who underwent primary hip arthroscopy with capsular plication and labral treatment demonstrated a high rate of RTS at short-term follow-up, high rate of continuation of sport at long-term follow-up, and favorable long-term results. Loss of interest or lifestyle transitions had a greater impact on continuation of sport at long-term follow-up than persistent symptoms related to the hip.
Background: Heterotopic ossification (HO) is a well-known postoperative complication of hip arthroscopy that can lead to pain, limited range of motion, and, in advanced cases, the need for revision surgery. Nonsteroidal anti-inflammatory drugs (NSAIDs), including diclofenac and indomethacin, are widely used for HO prophylaxis. However, there is no consensus on the optimal agent, dosage, and duration of administration. Hypothesis/Purpose: This study aims to compare the efficacy of diclofenac and indomethacin administered for different durations in preventing HO after hip arthroscopy and to evaluate the impact of varying treatment regimens on postoperative clinical outcomes. It was hypothesized that both diclofenac and indomethacin would effectively prevent HO, with no significant difference in HO incidence between them or with prolonged treatment. Study Design: Cohort study; Level of evidence, 3. Methods: A retrospective, multicenter study included 1084 patients who underwent hip arthroscopy for femoroacetabular impingement (FAI) between January 2017 and December 2024. Patients were categorized according to the prophylaxis protocol administered: diclofenac 75 mg for 15 days or indomethacin 75 mg for 15, 30, or 45 days daily. The presence and grade of HO were evaluated according to the Brooker classification. Functional outcomes were assessed pre- and 3, 6, 9, and 12 months postoperatively using the visual analog scale for pain (VAS-Pain) and the Hip Outcome Score–Activities of Daily Living (HOS-ADL). Results: No significant difference in HO development was observed among the study groups ( P = .061), with an overall grade 0 rate of 97.7%. The 45-day indomethacin regimen provided superior postoperative pain control with the mean postoperative VAS-Pain score of 0.5 ± 0.4. In contrast, the 15-day diclofenac and indomethacin regimens achieved the greatest improvement, with mean differences in VAS-Pain scores of 5.4 ± 1.6 and 6.8 ± 0.7, respectively. However, the highest postoperative HOS-ADL score was observed in the 15-day indomethacin group (90.4 ± 2.2), whereas the greatest improvement in HOS-ADL was observed in the 30-day group (47.1 ± 3.9), followed by the diclofenac regimen (47 ± 3.8). Conclusion: The incidence and severity of HO did not differ significantly among diclofenac and indomethacin prophylaxis regimens after hip arthroscopy. Prolonging indomethacin prophylaxis beyond 15 days was not associated with a lower rate of HO. These findings suggest that longer NSAID prophylaxis protocols may not provide additional benefit for HO prevention compared with shorter regimens.
Background: Although recent studies show no clinical outcome differences between cortical button suspensory fixation devices for anterior cruciate ligament reconstruction (ACLR), biomechanical mechanisms of elongation and failure remain unclear for adjustable-loop devices (ALDs) and fixed-loop devices (FLDs). Purpose: To compare biomechanical outcomes of ALDs with button-locking device (BLD), suture-locking device (SLD), and dual-locking device (DLD) mechanisms relative to an industry-standard FLD. Study Design: Controlled laboratory study. Methods: Five femoral cortical suspension devices (n = 11 each; total = 55 constructs) were tested using porcine femurs and bovine tendons simulating doubled hamstring grafts: EndoButton (FLD), Infinity Button (BLD), ProCinch (SLD1), UltraButton (SLD2), and TightRope (DLD). Constructs underwent cyclic loading simulating early (10-250 N, 500 cycles) and late (10-400 N, 500 cycles) rehabilitation, followed by pull-to-failure at 50 mm/min. Outcomes included excursion over simulated early and late rehabilitation and the cumulative total of both loading protocols. Also assessed were maximum load, stiffness, energy to failure, and failure mode, categorized as bone, button, suture loop rupture, tendon, or locking mechanism failure. Results: No significant differences were observed in cyclic loading at 500 cycles of 10 to 250 N or across the cumulative 1000 loading cycles between any ALD and the FLD. During cyclic loading at 10 to 400 N specifically, the BLD (3.34 ± 1.71 mm) and DLD (3.63 ± 1.18 mm) demonstrated greater excursion than the FLD (1.49 ± 0.36 mm; both P < .01). SLD1 (1270.5 ± 204.7 N) and SLD2 (1232.7 ± 185.2 N) withstood higher maximum loads than the BLD (1011.4 ± 95.8 N) and DLD (1018.4 ± 63.1 N; P < .05), although no differences were observed between the FLD and all ALDs combined. Stiffness did not differ across groups. BLD and DLD constructs most often failed by suture rupture, whereas the FLD and SLDs failed by button failure. Suture rupture was associated with greater excursion during cyclic loading at 10 to 400 N, lower maximum load, and reduced energy absorption compared with button failure ( P < .05). Conclusion: All constructs exceeded forces expected during early and total rehabilitation after ACLR, confirming adequate initial biomechanical strength across ALDs and FLDs. Differences in excursion and load to failure were associated with locking mechanism design and resultant failure mode, although clinical implications remain uncertain. Clinical Relevance: Although mechanical behavior varied by device mechanism, all systems provided fixation above physiologic loads, supporting that design-specific differences influence how devices fail but not their ability to provide adequate early postoperative stability in ACLR.
Background: Tobacco use is a known risk factor for poor healing after rotator cuff repair (RCR), but the effects of nontobacco nicotine products remain unclear. Purpose: To compare postoperative complications among nontobacco nicotine users, tobacco users, and nonusers undergoing RCR. Study Design: Cohort study; Level of evidence, 3. Methods: Using the PearlDiver database, the authors identified patients who underwent arthroscopic RCR with ≥2 years of follow-up. Three 1:1 propensity-matched comparisons were conducted: nontobacco nicotine users versus nonusers (n = 4844), tobacco users versus nonusers (n = 19,182), and nontobacco nicotine users vs tobacco users (n = 4882). Outcomes included 90-day complications and 2-year revision surgeries. Results: Nontobacco nicotine users had mildly higher rates of lysis of adhesions than controls (0.7% vs 0.4%; OR, 1.95; P = .033) but no other significant differences. Tobacco users showed increased rates of acute kidney injury (AKI), pneumonia, infections, emergency department (ED) visits, readmissions, and revision surgery compared to controls. Compared to tobacco users, nontobacco nicotine users had moderately strong lower risks of pneumonia (0.6% vs 1.2%; OR, 0.46), ED visits (8.8% vs 13.6%; OR, 0.62), and readmissions (0.8% vs 1.5%; OR, 0.54) (all P < .01). Conclusion: While nontobacco nicotine use was associated with a modestly increased risk of lysis of adhesions after RCR, it did not significantly elevate the risk of major medical complications, infections, or revision surgery when compared to nonnicotine users. In contrast, tobacco users demonstrated consistently worse postoperative outcomes across multiple domains, including higher rates of AKI, pneumonia, surgical site infection, ED visits, readmissions, and revision procedures. Notably, when compared directly to tobacco users, nontobacco nicotine users had significantly lower rates of pneumonia, ED utilization, and readmissions. These findings suggest that although nontobacco nicotine products may not be entirely benign, their perioperative risk profile appears to be less severe than that of traditional tobacco use.
Background: Patellofemoral instability (PFI) is common among adolescents. Dynamic evaluation of extensor mechanism alignment during early knee flexion may improve the understanding of recurrent instability risk. Purpose: To evaluate the use of point of care ultrasound (POCUS) to dynamically measure axial extensor mechanism alignment in patients with and without PFI. Study Design: Cohort study (diagnosis); Level of evidence, 3. Methods: Patients diagnosed with either patellar subluxation or dislocation and aged between 10 and 18 years were eligible. Control participants included those without any knee pain, nor any history of PFI, fracture, or musculoskeletal developmental disorder. Axial extensor mechanism was visualized with POCUS and the distance between the midpoint of the patellar tendon and lateral trochlear ridge (MPT-LTR) and lateral edge of the patellar tendon and LTR (LPT-LTR) was measured under 4 conditions: (1) active knee hyperextension (ie, quadriceps set), (2) full passive extension, (3) 15° of knee flexion, and (4) 30° of knee flexion. Univariate statistical analyses compared MPT-LTR and LPT-LTR distances across cohorts and knee positions. The area under the curve (AUC) evaluated measurement performance. Results: The cohort (PFI; n = 19) and controls (n = 19) were similar with respect to age, height, and weight ( P > .05). In both groups, the MPT-LTR and LPT-LTR were greatest in active hyperextension and progressively medialized with increasing knee flexion. Patients with PFI demonstrated statistically significantly more lateral MPT-LTR and LPT-LTR distance in all testing conditions ( P ≤ .01) when compared with controls. AUC analysis demonstrated acceptable performance across measures, ranging from 0.712 to 0.837. Conclusion: Patients with PFI consistently exhibited a more laterally oriented patella while performing active knee hyperextension and throughout the first 30° of knee flexion when measured using a novel POCUS evaluation.
Objective: A large proportion of individuals affected by medial meniscus posterior root (MMPR) tear progress to knee OA even after surgical repair. Gait analysis could assist in the identification of modifiable biomechanical (kinetic and kinematic) risk factors to optimize knee joint loading through gait modification interventions. The objective of this study was to define gait kinematics and kinetics in patients with MMPR tear and repair, compared to healthy, age-matched controls. We hypothesized that MMPR tear and repair patients would present altered gait characteristics associated with joint degeneration. Methods: In this cross-sectional study, a total of 58 knees were analyzed: MMPR tear (n=20), MMPR repair (n=21), and healthy controls (n=17) (Table 1). All individuals in the repair group had previously undergone transtibial pullout repair performed by a single fellowship-trained surgeon. Gait analysis was performed using a marker-based 3D motion capture system and force plates. Primary outcomes included knee joint kinematics and kinetics. Statistical comparisons were conducted using discrete analysis and Statistical Parametric Mapping to evaluate intergroup differences across the gait cycle. Patient-reported outcomes were also assessed using IKDC, KOOS Jr., VR-12, and VAS pain scores. Group comparisons were performed using the Kruskal-Wallis test. The significance level was set at 5%. Results: Both MMPR tear and repair groups exhibited increased varus thrust range of motion (Tear: 5.0° ± 2.1; Repair: 5.2° ± 2.0; Control: 3.3° ± 0.9), increased tibial internal rotation at loading response (Tear: -6.2° ± 5.3; Repair: -2.1° ± 3.6; Control: 2.2° ± 5.8), and greater knee adduction angle (varus) at midstance (Tear: 0.1° ± 3.9; Repair: -0.1° ± 5.0; Control: -5.9° ± 4.1) (Figure 1 and Table 2). The MMPR repair group exhibited higher functional scores (IKDC: 75.3 ± 19.9 vs. 43.6 ± 14.2; KOOS Jr.: 84.7 ± 15.4 vs. 59.7 ± 8.9), lower pain levels (VAS: 1.5 ± 2.1 vs. 3.8 ± 2.4) (Table 3), and higher tibial internal rotation compared to the tear group (p<0.05). Conclusions: The repair group demonstrated more favorable clinical and biomechanical characteristics than the tear group, however, both displayed established biomechanical risk factors for knee OA progression, such as varus thrust. Motion analysis can be used to detect early biomechanical dysfunction and may help identify MMPR tear and repair candidates for timely gait-modification interventions to optimize long-term joint health.
Objectives: Historically, osteochondral allograft (OCA) transplantation has been indicated for young, active patients with large focal osteochondral lesions. Consequently, 40 years of age has been frequently cited as an arbitrary upper age threshold for OCA transplantation. Large chondral defects in middle-aged patients pose a clinical challenge, as traditional arthroplasty is not ideal due to concerns about implant longevity and incompatibility with patients' postoperative activity goals. However, age-related factors such as prior surgeries, long-standing defects, and high body mass index have been shown to impact graft survival, suggesting that age alone may not be the sole determinant of OCA success. The purpose of this study was to compare graft survivorship and patient-reported outcomes measures (PROMs) following OCA transplantation in patients aged <40 and 40-49. Methods: Patient demographics (age, BMI, gender, surgical history) and surgical details (laterality, diagnosis, anatomic location, number of plug grafts, concomitant procedures, graft size) were extracted from our intuition's IRB-approved prospective OCA registry. Failure was defined as graft removal or conversion to arthroplasty. PROMs and satisfaction were recorded at latest follow-up. Statistical analyses compared demographics, surgical details, failure/reoperation rate, and PROMs between those aged <40 and 40-49. Results: This study included 489 patients (518 knees) with mean ages of 26.4 (391 knees) in patients <40 and 44.0 (127 knees) in patients 40-49. Among demographic and surgical variables, only diagnosis and anatomic location significantly differed by age. No difference in failure rates was observed between patients <40 and those aged 40-49 (p=0.640). Survival probabilities at 5, 10, and 15 years were 90.7%, 86%, and 74.3%, respectively, for patients under 40 and 89.8%, 83.7%, and 80.2% for those aged 40 to 49 (Figure 1.). The log-rank test revealed no significant differences in survival distributions across age groups (p=0.733). After controlling for all variables in logistic regression analysis, age still was not a significant predictor of failure (p = 0.849). Excluding failures, patients <40 and between 40-49 showed significant PROM improvement across all categories, with procedure satisfaction rates of 86.7% and 82.8%, respectively. PROMs and satisfaction did not significantly differ between age groups. Conclusions: Patients aged 40-49 had outcomes comparable to those <40. Surgeons can remain confident in using OCA transplantation as a treatment option for patients in this age group.
Objective: Anterior cruciate ligament reconstruction (ACLR) is one of the most common orthopedic procedures performed in young, active patients, a population particularly susceptible to persistent opioid use and opioid addiction. Despite this, there remains no consensus on optimal postoperative opioid prescribing. A 2025 systematic review by our group found wide variability in ACLR prescribing practices, with significant overprescription of opioids relative to actual consumption, a risk factor for opioid diversion and misuse. Although multimodal pain control strategies, including regional nerve blocks, NSAIDs, and cryotherapy, are widely endorsed, prospective evaluation of their impact on opioid consumption following ACLR remains limited. To address this limitation, our group designed a pragmatic prospective cohort study to evaluate opioid prescriptions and consumption among teens through middle age ACLR patients in the context of a sports medicine center-wide opioid-sparing multimodal pain control protocol with ketorolac as the primary post-discharge analgesic. Methods: The Cleveland Clinic orthopedic department's standardized multimodal pain protocol for ACLR includes preoperative adductor canal nerve block, intraoperative intravenous ketorolac with general anesthesia, intraoperative ropivacaine injection at all incision sites after graft fixation, intra-articular ropivacaine at closure, cryotherapy and knee immobilization initiation in the recovery unit, and postoperative analgesic prescriptions of acetaminophen (650 mg every 8 hours), and ketorolac for the first 72 hours (10 mg every 6 hours) followed by diclofenac extended-release for days 4-10 (75 mg twice daily, if needed). Additionally, four 5 mg oxycodone tablets are prescribed as rescue medicine. Since March 3, 2025, providers have been prompted to complete a pain assessment questionnaire during ACLR patients' first postoperative clinic visit via our institutional Outcome Management and Evaluation (OME) system, an electronic tool for collecting standardized surgical data. Questions include the number of oxycodone tablets prescribed at discharge, refills, if applicable, and patients' report of the number of oxycodone tablets taken, their worst postoperative pain on a 1-10 scale, and their current pain during movement. Among patients who underwent ACLR, with or without meniscal surgery, by July 24, 2025 (N=147), patients were ineligible if they underwent revisions and multi-ligamentous reconstructions (n=17), were >55 years old (n=9), or were enrolled in an experimental study (n=4). Results: Of 117 eligible patients, 1 was loss to follow-up and the questionnaire was not completed for 23, leaving a sample of 93 (79.5%). Patients (50.5% were female, median age 23 years [Q1, Q3: 18-30], 72.2% athletes) were prescribed a median of 4 tablets (Q1, Q3: 4, 5) at discharge. At follow-up, a median of 11 days postop (Q1, Q3: 6, 16) patients reported taking a median of 3 tablets (Q1, Q3: 2, 4); 16.3% took none, resulting in a median of 2 (Q1, Q3: 0, 4) excess oxycodone tablets, not including refills, which 12.9% of patients received. Median worst postoperative pain since surgery was 8 (Q1, Q3: 6, 10). In contrast, at follow-up 6.5% had no pain, 47.8% had no pain at rest/slight on movement, 40.2% had slight pain at rest/moderate on movement, and 5.4% reported moderate pain at rest/severe on movement. All ACLRs were performed by one of seven fellowship-trained sports medicine surgeons. Six of seven surgeons prescribed a median of 4 or 5 oxycodone tablets to patients at discharge, two of whom never prescribed more than 4. See table for additional data by surgeon. No complications related to inadequate pain control or ketorolac use were identified. Conclusions: Early findings from this in-progress cohort study indicate that in the context of a standardized comprehensive ketorolac-centered multimodal pain control protocol in which a limited quantity of oxycodone is prescribed as rescue medicine, most patients undergoing ACL reconstruction consume 3 or fewer tablets, with one in six patients consuming none. This real-world cohort provides preliminary evidence supporting the safety and feasibility of opioid-sparing prescribing practices in sports medicine. Future directions include continuing to enroll patients in this pragmatic prospective cohort and identifying risk factors associated with opioid use. Our group will also explore non-opioid analgesic complimentary and 'rescue' medications, and continue to refine perioperative care pathways, including intraoperative anesthesia and PACU protocols.
Objective: Return to sport (RTS) recommendations after anterior cruciate ligament reconstruction (ACLR) vary based on surgical factors, patient demographics, athletic level, and physician preference. RTS testing is often required before returning to high impact activities, but optimal RTS testing timing remains unclear. Historical guidelines favored RTS testing at 6-months following primary ACLR, though recent recommendations extend this to 9-12 months to reduce reinjury risk. The lack of clarity regarding optimal timing, and patient or surgical factors that may impact optimal timing as well as the utility of RTS testing can be barriers to safe return to sport following ACLR. Therefore, this study aimed to identify surgical, demographic, and timing factors associated with successful RTS without reinjury following ACLR and to correlate standardized RTS testing to each of these factors and the patients reinjury risk. Methods: With IRB approval, patients who underwent primary ACLR between who completed standardized RTS testing with a certified physical therapist, who had documented RTS status and timing and who had a minimum 2 year follow-up were included. Patient demographic factors including age, sex, tobacco use status, marital status, activity level (recreation, occupational, middle school/high school, or college) and a personal history of mental health issues; surgical details including concurrent meniscus repair or meniscectomy, graft type, and pre-operative physical therapy completion; RTS timing defined as the documented timepoint (month) they resumed full activity participation; and RTS testing completed by an outpatient physical therapist were analyzed to identify factors associated with successful RTS without reinjury, defined as ipsilateral graft tear requiring revision or further surgery. Fisher's exact test and Kruskal-Wallis rank sum test was used to analyze categorical and continuous variables with significance set at p<0.05. Results: Of 137 included patients, 113 reported they RTS (82.5%) following ACLR. Patients without mental health issues (p=.002), patients competing at the middle or high school level (p=.017), younger patients (p=.014, mean 18.5 vs. 19.6 years) and patients without a concurrent lateral meniscus repair (p=.016, 27.4% vs. 54.2%) were significantly more likely to RTS. 10 patients (7.3%) had a graft failure at a mean of 18.6 months (range: 7-41 months). No included factors, including returning to sport, RTS testing pass/fail, RTS timing, tobacco use, age, mental health issues, sex, pre-op physical therapy, level of play, graft type, and concurrent procedures were significantly correlated with failure. Of the 57 patients who completed RTS testing at 6-months and eventually fully returned to sport (at any time point), only 2 (3.5%) passed all testing measures at that timepoint. Conclusions: Patients pass standardized RTS testing at a very low rate at 6 months postoperatively after ACLR. No surgical, demographic, timing, or RTS testing factors were significantly associated with successful RTS or reinjury over others following ACLR. These findings suggest that current RTS testing timelines and mechanisms are not definitive for identifying who may be at risk of reinjury. Future research should focus on refining RTS assessments and functional testing to identify individualized criteria associated with safe RTS following ACLR.
Objective: To provide a comprehensive review of return to sport (RTS) following acute traumatic tibial shaft fractures, including RTS rates and timeframes by different fracture types and treatment methods. Methods: A systematic review was performed in accordance with PRISMA guidelines. A search of PubMed, Cochrane, Scopus, and Web of Science databases was performed in October 2024 using the keywords 'tibia', 'tibial shaft,' 'midshaft tibia,' 'tibial diaphysis,' 'fracture,' 'return to sport/play/physical activity/athletic participation.' Nine studies met inclusion criteria: (1) reported on acute traumatic fractures of the tibial shaft and (2) included data on return to sport. Study quality and risk of bias were evaluated by the modified Coleman Methodology Score (mCMS) and the Robins-I tool. Data collected, when available, included sport type, fracture type (open vs closed fracture), treatment method (non-surgical vs surgical), RTS rate, RTS level, RTS timeframe, and rate and time to fracture union. Results: Of the nine studies included there were two randomized controlled trials, one prospective cohort study, four retrospective cohort studies and two case series. Seven studies were rated 'poor' and two rated 'fair' quality by the Modified Coleman Methodology Score. As measured by the Robins-I tool, two studies were found to have serious risk, two moderate risk, and five low risk of bias. Soccer was the most studied sport, being included in five studies, followed by American football. In total, 430 patients (mean age 25.7 years, 97 amateur, 140 professional) were included. Studies had variably reported data. Treatment-specific rates and timeframes for return to sport and fracture union after acute traumatic tibial shaft fractures are summarized in Table 1. The overall RTS rate was 90.4% (142/157) at an average of 38.5 (range, 23.3-55) weeks. Acute traumatic fractures had a 96.5% union rate when the fractures were closed and 90.4% when open. Casting/immobilization had the least successful union rate (70.6%) and return to sport (66.7%). Unreamed intramedullary nail (IMN) demonstrated the quickest time to union (11.3 weeks). Open reduction and internal fixation (ORIF) with a plate-screw construct was associated with the quickest return to sport (23.3 weeks) while external fixation had the slowest (55 weeks). One hundred percent of closed and 89-92% of open fractures treated with reamed or unreamed tibial intramedullary nailing successfully united. Sport-specific rates and timeframes for return to sport and fracture union after tibial shaft fractures are summarized in Table 2. Tibial shaft fractures treated with tibial intramedullary nailing returned to soccer at 35-41 weeks and American football at 42.1 weeks after surgery. Conclusions: Patients sustaining acute traumatic tibial shaft fractures have high RTS and osseous union rates. Reamed or unreamed tibial intramedullary nailing of closed fractures demonstrated superior union rates and timeframes compared with open fractures. Immobilization with casting provides inferior return to sport and union rates compared to operative fixation for acute traumatic tibial shaft fractures. Current data on return to sport after tibial shaft fractures is limited in both quantity and quality with many studies being outdated for contemporary fixation methods. Further research is needed to ascertain these data with current fixation methods.
Objective: Accurate measurement of muscle strength is essential for effective clinical decision-making. Isokinetic dynamometers have been considered the gold standard for measuring muscle strength output; however, their size, cost, and complexity ultimately limit accessibility in many clinical and research settings. Handheld dynamometers (HHD) are generally more cost-effective and portable, providing clinicians a valuable tool for clinical practice; however, their reliability and validity can be influenced by various factors, including tester experience and patient positioning. Recently, the use of crane scales (CS [hanging scale, traction dynamometer]) has gained attention for their potential in measuring muscle strength, both reliably and economically. Despite this growing interest, there is limited research assessing the reliability and validity of CS devices for muscle strength testing. The objective of this study is to evaluate the reliability and validity of a novel method of assessing large muscle groups of the lower extremity (hamstrings, quadriceps) via CS when compared to clinical and research gold standards (HHD, isokinetic device [IKD]). The authors hypothesize that the crane scale will prove to be a valid and reliable alternative to testing lower extremity strength when compared to the gold standards. Methods: Subjects were recruited from a local convenience sampling of healthy volunteers. Two testers of different experience levels (licensed physical therapists, 10+ years vs. 1 year) were utilized for all subjects and blinded to the results of each test. Consented participants were tested on two non-consecutive days at least 72-hours apart for test-retest procedures for each measurement. They performed a standardized warm up and then were tested in a randomized (tester and device) fashion with three maximal-voluntary isometric contractions for each muscle (quadriceps/knee extension and hamstrings/knee flexion) and device (crane scale [CS], Biodex Isokinetic Device [IKD], MicroFet 2 handheld dynamometer [HHD]). See Table 1. All subjects were given adequate rest breaks between contractions and devices. A 1×3 ANOVA (group × device), followed by Bonferroni-corrected pairwise comparisons, was used to assess strength and intra-observer variability (Coefficient of variation, CV) across devices. Intraclass correlations (ICCs) were calculated for inter-observer agreement at the same timepoint and intra-observer consistency across timepoints, with device comparisons via F-tests. Type I error was set at α = 0.05. Results: Twenty-two healthy adult participants (♀=17, ♂=5; 25±5 yrs) were included. Table 2 provides mean/peak torque and variance measures. For knee extension, the crane scale (CS) produced mean and peak torque values comparable to the isokinetic dynamometer (IKD), while the handheld dynamometer (HHD) yielded significantly lower values than both CS and IKD ( p < 0.001). For knee flexion, CS mean torque values were significantly higher than HHD ( p = 0.023), with no significant differences between CS and IKD. Measurement variability (coefficients of variation) was significantly lower for CS and IKD compared with HHD during extension testing ( p < 0.001), while IKD showed the lowest variability in flexion trials ( p = 0.004 vs CS; p = 0.010 vs HHD). Inter-observer reliability was excellent across devices (ICC > 0.7), with CS demonstrating the highest ICCs and narrowest confidence intervals. Intra-observer reliability across sessions was also strong for CS and IKD (ICC ≈ 0.95), whereas HHD consistently exhibited wider confidence intervals, indicating reduced precision. Conclusions: The CS demonstrated strong agreement with the gold standard IKD and superior reliability compared with the HHD for assessing knee torque in large muscle groups. These findings support the CS as a feasible, repeatable, and accurate alternative for strength assessment in clinical and field-based settings. Further validation is warranted in clinical and pathologic populations along with other muscle groups.
Objective: Arthroscopic Bankart repair (ABR) is the most widely performed surgical treatment for anterior shoulder instability, and the Instability Severity Index Score (ISIS), while widely used to guide surgical decision-making, has demonstrated inconsistent reliability. Recent evidence suggests that incorporating advanced imaging findings and refining patient-specific risk factors may enhance predictive accuracy and improve clinical utility. The purpose of this study was to develop and validate an ABR recurrence risk calculator in an independent clinical dataset and assess its performance in stratifying patients by recurrence risk. Methods: The BRACE (Bankart Risk Assessment & Clinical Estimator) score was developed, based on age, sex, glenoid bone loss, Hill-Sachs lesion status, number of preoperative dislocations, ligamentous laxity, and sport participation generating 864 distinct clinical scenarios. Odds ratios were extracted from a systematic review of 110 studies and incorporated into a logistic regression model. The model was then evaluated on an international multicenter cohort including 5 institutions from 3 countries. Discrimination (area under the curve (AUC)), Brier score, calibration (slope, Hosmer-Lemeshow), risk stratification (low; 0-14, moderate; 15%-29%, high; 30+%), decision‑curve analysis and operating characteristics from 10%-30% thresholds were reported. Results: The BRACE score was evaluated on 2535 ABR patients. The interaction between glenoid bone loss >=13.5% and an off-track Hill-Sachs lesion was independently associated with higher recurrence risk. The model demonstrated moderate discriminative ability (AUC = 0.70, 95% CI 0.62-0.78) with appropriate calibration across the range of predicted recurrence probabilities (calibration slope = 0.889, 95% CI 0.62-1.16; Hosmer-Lemeshow p = 0.36). The model achieved a Brier score of 0.067 (95% CI 0.052-0.082), indicating good overall accuracy of the predicted probabilities. At a 15% moderate risk threshold, the specificity was 93.4% (95% CI 91.1%-95.4%), and at a 30% high threshold specificity was 98.0% (95% CI 96.7%-99.1%). Model performance remained consistent across subgroups and predicted risk estimates aligned closely with observed recurrence rates. Conclusions: The ABR recurrence risk calculator demonstrated moderate discrimination, good overall accuracy, and appropriate calibration when applied to a large, multicenter clinical cohort. The tool effectively stratified patients into low-, moderate-, and high-risk groups, with decision-curve analysis confirming clinical utility across relevant thresholds.
Objectives: High tibial osteotomy (HTO) and distal femoral osteotomy (DFO) are powerful tools for correcting coronal malalignment of the lower extremity, and are effective, joint-preserving treatment strategies for addressing early tibiofemoral compartment osteoarthritis in young patients. Historically, these procedures have been performed freehand using intraoperative fluoroscopy, after extensive presurgical planning based on preoperative imaging. More recently, the advent of 3D CT-based patient-specific instrumentation (PSI) has been associated with decreased intraoperative fluoroscopy, improved safety and possibly improved accuracy in a cadaveric model.1 The purpose of this study was to assess correction accuracy of coronal plane modifying osteotomies using PSI in a consecutive series of patients. Methods: All patients who underwent single-level coronal plane correction osteotomy (HTO or DFO) with PSI (Bodycad, Quebec City, Canada) performed by two fellowship trained orthopaedic surgeons at a tertiary orthopaedic center between October 2020 and September 2024 were eligible for study inclusion. Patients who underwent sagittal plane correction osteotomies, dual level osteotomies, or had documented intraoperative hinge fractures were excluded. All included patients underwent preoperative and postoperative long-standing alignment radiographs at a minimum of 3 months postoperatively. Patients without complete imaging were excluded. Preoperative and postoperative 3D CT-based measurements of medial tibial width ratio (MTWR) and joint line congruency angle (JLCA) were obtained from the PSI preoperative plan. Preoperative and postoperative radiographic measurements of MTWR and JLCA were performed by two independent raters. Means +/- standard deviations and medians with ranges were used as descriptive statistics for normally and non-normally distributed variables, respectively. Intraclass correlation coefficients (ICCs) were used to assess both interrater and intermethod reliability for MTWR and JLCA. Wilcoxon signed-rank tests were used to compare the MTWR between 3D CT data and radiographic measurements. Cutoffs of +/-2.5%, +/-5% and +/-10% relative to the planned postoperative 3D CT MTWR were used to assess correction accuracy (Fig 1). Results: 130 patients met study inclusion criteria. Patient characteristic data are found in Table 1. Median age was 40 years (range: 16-59) and 52% of patients were female. Ninety patients (69%) underwent HTO and 40 patients (31%) underwent DFO. The ICC score for interrater reliability on preoperative radiographs was 0.99 (95% CI: 0.99-0.99) for MTWR and 0.92 (95% CI: 0.88-0.95) for JLCA. The ICC score between preoperative 3D CT data and measured values on preoperative radiographs was 0.99 (95% CI: 0.98-0.99) for MTWR and 0.50 (95% CI: 0.36-0.62) for JLCA. There was no significant difference in preoperative MTWR measured on 3D CT compared to preoperative radiographs (difference in means = 0.4% (95% CI: -0.2% -1.0%), p=0.25). Similarly, there was no significant difference in postoperative MTWR planned on 3D CT compared to postoperative radiographic measurement (difference in means = 1.1% (95%CI: -0.5% - 2.7%, p=0.09). The ICC score for the preoperative to postoperative change in MTWR planned on 3D CT and measured on radiographs was 0.97 (95% CI: 0.95-0.98) (Fig 2A). Subgroup analysis demonstrated similar findings for DFO and HTO patients, with ICC scores for preoperative to postoperative changes in MTWR planned on CT and measured on radiographs of 0.86 (95% CI: 0.49-0.97) for DFOs and 0.95 (95% CI: 0.91-0.97) for HTOs (Fig 2B & 2C). When comparing final MTWR measured on postoperative radiographs with the planned 3D CT MTWR, 24.2%, 49.5% and 88.4% of patients had a measured MTWR on postoperative radiographs within 2.5%, 5%, 10% of the planned 3D CT MTWR, respectively (Fig 3A). Subgroup analysis of osteotomy type demonstrated similar results, with 25.0%, 52.9% and 91.2% of HTO patients (Fig 3B) and 22.2%, 40.7% and 81.5% of DFO patients meeting the 2.5%, 5% and 10% accuracy cutoffs, respectively (Fig 3C). Conclusions: These results demonstrate a high degree of interrater reliability for radiographic measurements of MTWR and JLCA. MTWR demonstrated a high degree of intermethod reliability between radiographic measurements and 3D CT data, whereas JLCA demonstrated much lower intermethod reliability. This discrepancy in JLCA is likely related to the non-weight bearing nature of the CT scan versus weight-bearing radiographs. Thus, the MTWR appears to be a highly reliable method for communicating coronal plane correction, as this value is normalized to each patient's tibia and avoids measurement limitations associated with abnormalities in the anatomic axis of the femur and/or tibia. Additionally, the planned degree of correction was highly reliable between the radiographic and CT data. The PSI method for coronal plane osteotomy was highly accurate in achieving the desired correction in coronal plane alignment, with 24.2%, 49.5% and 88.4% of patients' final radiographic coronal alignment falling within 2.5%, 5% and 10% of the preoperative 3D CT plan. Furthermore, this accuracy was maintained for both HTOs and DFOs. HTOs may be slightly more accurate than DFOs, with a higher percentage of patients meeting each of the accuracy cutoffs and a higher ICC score between planned and measured coronal correction, which may be related to the osteotomy cut being closer to the joint line in HTOs compared to DFOs. Overall, this study demonstrates 1) a high degree of concordance between preoperative radiographic measurements of MTWR and 3D CT data and 2) a high degree of accuracy in obtaining the desired coronal correction and final MTWR using PSI for coronal plane osteotomies.
