Navigated augmented reality through a head-mounted display leads to low deviation between planned, intra- and post-operative parameters during glenoid component placement of reverse shoulder arthroplasty: a proof-of-concept case series.

BACKGROUND: Navigated augmented reality (AR) through a head-mounted display (HMD) has led to accurate glenoid component placement in reverse shoulder arthroplasty (RSA) in an in-vitro setting. The purpose of this study is to evaluate the deviation between planned, intra-, and postoperative inclination, retroversion, entry point and depth of the glenoid component placement during RSA, assisted by navigated AR through a HMD, in a surgical setting. METHODS: A prospective, multicenter study was conducted. All consecutive patients undergoing RSA in two institutions, between August 2021 and January 2023, were considered potentially eligible for inclusion in the study. Inclusion criteria were: age >18 years, surgery assisted by AR through a HMD, and postoperative computed tomography (CT) scans at six weeks. All participants agreed to participate in the study and an informed consent was provided in all cases. Preoperative CT scans were undertaken for all cases and used for three-dimensional (3D) planning. Intra-operatively, glenoid preparation and component placement were assisted by a navigated AR system through a HMD in all patients. Intraoperative parameters were recorded by the system. A postoperative CT scan was undertaken at 6 weeks, and 3D reconstruction was used for obtaining postoperative parameters. The deviation between planned, intra-, and postoperative inclination, retroversion, entry point, and depth of the glenoid component placement was calculated. Outliers were defined as >5° for inclination and retroversion and >5 mm for entry point. RESULTS: 17 patients (9 females, 12 right shoulders) with a mean age of 72.8±9.1 years old (range, 47.0 to 82.0) met inclusion criteria. The mean deviation between intra- and postoperative measurements was 1.5°±1.0° (range, 0.0° to 3.0°) for inclination, 2.8°±1.5° (range, 1.0° to 4.5°) for retroversion, 1.8±1.0 mm (range, 0.7mm to 3.0mm) for entry point, and 1.9±1.9 mm (range, 0.0mm to 4.5mm) for depth. The mean deviation between planned and postoperative values was 2.5°±3.2° (range, 0.0° to 11.0°) for inclination, 3.4°±4.6° (range, 0.0° to 18.0°) for retroversion, 2.0±2.5 mm (range, 0.0° to 9.7°) for entry point, and 1.3±1.6 mm (range, 1.3mm to 4.5mm) for depth. There were no outliers between intra- and postoperative values and there were three outliers between planned and postoperative values. The mean time (minutes:seconds) for the tracker unit placement and the scapula registration was 03:02 (range, 01:48 to 04:26) and 08:16 (range, 02:09 to 17:58), respectively. CONCLUSION: The use of a navigated AR system through a HMD in RSA led to low deviations between planned, intra-operative and postoperative parameters for glenoid component placement.

Deep-Learning-Based Segmentation of the Shoulder from MRI with Inference Accuracy Prediction.

Three-dimensional (3D)-image-based anatomical analysis of rotator cuff tear patients has been proposed as a way to improve repair prognosis analysis to reduce the incidence of postoperative retear. However, for application in clinics, an efficient and robust method for the segmentation of anatomy from MRI is required. We present the use of a deep learning network for automatic segmentation of the humerus, scapula, and rotator cuff muscles with integrated automatic result verification. Trained on N = 111 and tested on N = 60 diagnostic T1-weighted MRI of 76 rotator cuff tear patients acquired from 19 centers, a nnU-Net segmented the anatomy with an average Dice coefficient of 0.91 ± 0.06. For the automatic identification of inaccurate segmentations during the inference procedure, the nnU-Net framework was adapted to allow for the estimation of label-specific network uncertainty directly from its subnetworks. The average Dice coefficient of segmentation results from the subnetworks identified labels requiring segmentation correction with an average sensitivity of 1.0 and a specificity of 0.94. The presented automatic methods facilitate the use of 3D diagnosis in clinical routine by eliminating the need for time-consuming manual segmentation and slice-by-slice segmentation verification.

Peri-Prosthetic Joint Infection of the Knee Causes High Levels of Psychosocial Distress: A Prospective Cohort Study.

Background: Peri-prosthetic joint infection (PJI) is a major complication of knee arthroplasty that can cause long-term disability. In addition to its physical impact, there is a clear psychological burden that has not been measured yet. We hypothesized that the psychosocial burden of PJI can be assessed quantitatively using standardized questionnaires and may be correlated with treatment stage. Methods: Thirty-one patients were enrolled in this longitudinal prospective cohort study from August 2015 to November 2016. Participants had clinically established knee PJI after primary total knee replacement in osteoarthritis according to the Musculoskeletal Infection Society criteria and underwent a standardized two-stage protocol. After explantation of the prosthesis and implantation of a polymethylmethacrylate knee spacer, patients were treated with organism-specific intravenous antibiotics for two weeks, followed by oral antibiotics for four weeks; and then reimplantation was performed in all cases. Psychometrically validated standardized questionnaires were used to measure psychosocial stress via self-assessment at four time points: (1) Before explantation of the prosthesis; (2) after explantation; (3) after the antibiotic treatment before reimplantation; and (4) three months after reimplantation (follow-up). The Patient Health Questionnaire (PHQ)-4, Short Form (SF)-12 (including PSK and KSK), Questions about Life Satisfaction (FLZM) and Fear of Progression (PA-F-KF) (titles and abbreviations in German) scores were interpreted according to cut-off values for depression, fear of progression, anxiety, and quality of life. Results: Eighteen patients (58.1%) showed a PHQ-4 score above the cut-off value for depression at least once, with the highest score before reimplantation (time point 3). On the SF-12, the mean subtest mental scale (PSK) score was 42.6 (± 14.5), and the mean subtest physical scale (KSK) score was 26.9 (± 7.5) over the four time points, which was significantly lower than that of the general German population (PSK 53.1, KSK 44.0; p < 0.05). The SF-12 scores did not change significantly over time. On the FLZ, health was least satisfactory, followed by recreational activities and work. On the PA-F-KF, patients had the greatest fear of being dependent on outside help, drastic medical interventions, and infection progression. The mean PA-F-KF value was 31.24 (± 9.60; values ≥34 are regarded as critical). Conclusion: Peri-prosthetic joint infection is a measurable, relevant psychosocial stressor for patients. Their quality of life and fear of the disease progressing are comparable to those of oncology patients. Routine screening should be conducted to identify affected patients early for appropriate treatment, improving long-term outcomes. Orthopaedic surgeons who treat patients with PJI should initiate by psychologists as well in order to maintain the patient's long-term quality of life.