Anonymization and validation of three-dimensional volumetric renderings of computed tomography data using commercially available T1-weighted magnetic resonance imaging-based algorithms

Purpose

Previous studies have demonstrated that three-dimensional (3D) volumetric renderings of magnetic resonance imaging (MRI) brain data can be used to identify patients using facial recognition. We have shown that facial features can be identified on simulation-computed tomography (CT) images for radiation oncology and mapped to face images from a database. We aim to determine whether CT images can be anonymized using anonymization software that was designed for T1-weighted MRI data.

Approach

Our study examines (1) the ability of off-the-shelf anonymization algorithms to anonymize CT data and (2) the ability of facial recognition algorithms to identify whether faces could be detected from a database of facial images. Our study generated 3D renderings from 57 head CT scans from The Cancer Imaging Archive database. Data were anonymized using AFNI (deface, reface, and 3Dskullstrip) and FSL's BET. Anonymized data were compared to the original renderings and passed through facial recognition algorithms (VGG-Face, FaceNet, DLib, and SFace) using a facial database (labeled faces in the wild) to determine what matches could be found.

Results

Our study found that all modules were able to process CT data and that AFNI's 3Dskullstrip and FSL's BET data consistently showed lower reidentification rates compared to the original.

Conclusions

The results from this study highlight the potential usage of anonymization algorithms as a clinical standard for deidentifying brain CT data. Our study demonstrates the importance of continued vigilance for patient privacy in publicly shared datasets and the importance of continued evaluation of anonymization methods for CT data.

Prediction of Progression After Cervix Cancer Radiotherapy Using a Machine-Learning Model on Pre-Treatment MRI

PURPOSE/OBJECTIVE(S)

MRI may be useful to identify women with cervical cancer at high risk of disease progression to test strategies of treatment intensification. The purpose of this study was to determine the value of a machine-learning model built on pre-treatment MRI for prediction of risk of progression after radiation therapy.

MATERIALS/METHODS

MagneticResonance Imaging (MRI) data for women with cervical cancer was collected from The Cancer Genome Atlas Cervical Squamous Cell Carcinoma and Endocervical Adenocarcinoma Collection (TCGA-CESC) on the Cancer Imaging Archive (TCIA), which reported clinical, treatment, and imaging data from a single institution. 27 patients who had received radiation for cervical cancer were selected for input into a custom 3-D Residual Neural Network (ResNet) model with added custom layers specific to DICOM data in tensorflow python package. One T2 MRI per patient was used to predict recurrence free survival after radiation treatment, where patients were predicted to be "high risk" or "low risk" for disease recurrence as the output of the model. All slices of the T2 MRI were used. The model was validated using five-fold cross validation; 80% of the data was used to train each fold and 20% was used for testing. Final model statistical significance was confirmed through shuffle test at the p < 0.01 level. The clinical outcomes of patients and the model's "low-risk" and "high-risk" prediction were compared.

RESULTS

There were 27 patients in the study with mean age of 51 years (range 29-79). 20 patients had squamous cell carcinoma and 7 patients had adenocarcinoma. The stage breakdown consisted of 9 women IB, 2 IIA, 9 IIB, 2 IIIA, 2 IIIB, and 3 stage IV. 10 women were treated with radiation alone and 17 with chemo-radiation. 5 women received surgery in addition to radiation or chemoradiation. 21 patients received brachytherapy. Median follow-up of patients was 29 months (range 3-64). The model predicted 7 patients as "high risk" for recurrence; all 7 developed a recurrence during follow up. None of the 20 patients predicted to be "low risk" developed disease recurrence. Among all patients in the study, the two-year progression free survival (PFS) was 82.0%. Patients identified as "low risk" and "high risk" by model had two-year PFS of 100% and 43%, respectively. Among patients with recurrence, 3 developed local recurrence and 4 developed distant metastases. The ResNet model achieved cross-validated accuracy of 92% for prediction of progression-free survival (p<0.01).

CONCLUSION

A 3-D ResNet machine-learning model using pretreatment MRI image data can accurately predict clinical outcomes for cervical cancer following radiation therapy. Future work to confirm generalizability should focus on validation with a larger clinical dataset.

CFPNet-M: A light-weight encoder-decoder based network for multimodal biomedical image real-time segmentation

-Deep learning techniques are proving instrumental in identifying, classifying, and quantifying patterns in medical images. Segmentation is one of the important applications in medical image analysis. The U-Net has become the predominant deep-learning approach to medical image segmentation tasks. Existing U-Net based models have limitations in several respects, however, including: the requirement for millions of parameters in the U-Net, which consumes considerable computational resources and memory; the lack of global information; and incomplete segmentation in difficult cases. To remove some of those limitations, we built on our previous work and applied two modifications to improve the U-Net model: 1) we designed and added the dilated channel-wise CNN module and 2) we simplified the U-shape network. We then proposed a novel light-weight architecture, the Channel-wise Feature Pyramid Network for Medicine (CFPNet-M). To evaluate our method, we selected five datasets from different imaging modalities: thermography, electron microscopy, endoscopy, dermoscopy, and digital retinal images. We compared its performance with several models having a variety of complexities. We used the Tanimoto similarity instead of the Jaccard index for gray-level image comparisons. The CFPNet-M achieves segmentation results on all five medical datasets that are comparable to existing methods, yet require only 8.8 MB memory, and just 0.65 million parameters, which is about 2% of U-Net. Unlike other deep-learning segmentation methods, this new approach is suitable for real-time application: its inference speed can reach 80 frames per second when implemented on a single RTX 2070Ti GPU with an input image size of 256 × 192 pixels.

CaraNet: context axial reverse attention network for segmentation of small medical objects

Purpose

Segmenting medical images accurately and reliably is important for disease diagnosis and treatment. It is a challenging task because of the wide variety of objects' sizes, shapes, and scanning modalities. Recently, many convolutional neural networks have been designed for segmentation tasks and have achieved great success. Few studies, however, have fully considered the sizes of objects; thus, most demonstrate poor performance for small object segmentation. This can have a significant impact on the early detection of diseases.

Approach

We propose a context axial reverse attention network (CaraNet) to improve the segmentation performance on small objects compared with several recent state-of-the-art models. CaraNet applies axial reserve attention and channel-wise feature pyramid modules to dig the feature information of small medical objects. We evaluate our model by six different measurement metrics.

Results

We test our CaraNet on segmentation datasets for brain tumor (BraTS 2018) and polyp (Kvasir-SEG, CVC-ColonDB, CVC-ClinicDB, CVC-300, and ETIS-LaribPolypDB). Our CaraNet achieves the top-rank mean Dice segmentation accuracy, and results show a distinct advantage of CaraNet in the segmentation of small medical objects.

Conclusions

We proposed CaraNet to segment small medical objects and outperform state-of-the-art methods.

A teacher–student framework with Fourier Transform augmentation for COVID-19 infection segmentation in CT images

Automatic segmentation of infected regions in computed tomography (CT) images is necessary for the initial diagnosis of COVID-19. Deep-learning-based methods have the potential to automate this task but require a large amount of data with pixel-level annotations. Training a deep network with annotated lung cancer CT images, which are easier to obtain, can alleviate this problem to some extent. However, this approach may suffer from a reduction in performance when applied to unseen COVID-19 images during the testing phase, caused by the difference in the image intensity and object region distribution between the training set and test set. In this paper, we proposed a novel unsupervised method for COVID-19 infection segmentation that aims to learn the domain-invariant features from lung cancer and COVID-19 images to improve the generalization ability of the segmentation network for use with COVID-19 CT images. First, to address the intensity difference, we proposed a novel data augmentation module based on Fourier Transform, which transfers the annotated lung cancer data into the style of COVID-19 image. Secondly, to reduce the distribution difference, we designed a teacher–student network to learn rotation-invariant features for segmentation. The experiments demonstrated that even without getting access to the annotations of the COVID-19 CT images during the training phase, the proposed network can achieve a state-of-the-art segmentation performance on COVID-19 infection.

A Distance-based Separability Measure for Internal Cluster Validation

To evaluate clustering results is a significant part of cluster analysis. Since there are no true class labels for clustering in typical unsupervised learning, many internal cluster validity indices (CVIs), which use predicted labels and data, have been created. Without true labels, to design an effective CVI is as difficult as to create a clustering method. And it is crucial to have more CVIs because there are no universal CVIs that can be used to measure all datasets and no specific methods of selecting a proper CVI for clusters without true labels. Therefore, to apply a variety of CVIs to evaluate clustering results is necessary. In this paper, we propose a novel internal CVI – the Distance-based Separability Index (DSI), based on a data separability measure. We compared the DSI with eight internal CVIs including studies from early Dunn (1974) to most recent CVDD (2019) and an external CVI as ground truth, by using clustering results of five clustering algorithms on 12 real and 97 synthetic datasets. Results show DSI is an effective, unique, and competitive CVI to other compared CVIs. We also summarized the general process to evaluate CVIs and created the rank-difference metric for comparison of CVIs’ results.

Common functional connectivity alterations in focal epilepsies identified by machine learning

OBJECTIVE

This study was undertaken to identify shared functional network characteristics among focal epilepsies of different etiologies, to distinguish epilepsy patients from controls, and to lateralize seizure focus using functional connectivity (FC) measures derived from resting state functional magnetic resonance imaging (MRI).

METHODS

Data were taken from 103 adult and 65 pediatric focal epilepsy patients (with or without lesion on MRI) and 109 controls across four epilepsy centers. We used three whole-brain FC measures: parcelwise connectivity matrix, mean FC, and degree of FC. We trained support vector machine models with fivefold cross-validation (1) to distinguish patients from controls and (2) to lateralize the hemisphere of seizure onset in patients. We reported the regions and connections with the highest importance from each model as the common FC differences between the compared groups.

RESULTS

FC measures related to the default mode and limbic networks had higher importance relative to other networks for distinguishing epilepsy patients from controls. In lateralization models, regions related to somatosensory, visual, default mode, and basal ganglia showed higher importance. The epilepsy versus control classification model trained using a 400-parcel connectivity matrix achieved a median testing accuracy of 75.6% (median area under the curve [AUC] = .83) in repeated independent testing. Lateralization accuracy using the 400-parcel connectivity matrix reached a median accuracy of 64.0% (median AUC = .69).

SIGNIFICANCE

Machine learning models revealed common FC alterations in a heterogeneous group of patients with focal epilepsies. The distribution of the most altered regions supports the hypothesis that shared functional alteration exists beyond the seizure onset zone and its epileptic network. We showed that FC measures can distinguish patients from controls, and further lateralize focal epilepsies. Future studies are needed to confirm these findings by using larger numbers of epilepsy patients.

COVID-19 CT Image Synthesis With a Conditional Generative Adversarial Network

Coronavirus disease 2019 (COVID-19) is an ongoing global pandemic that has spread rapidly since December 2019. Real-time reverse transcription polymerase chain reaction (rRT-PCR) and chest computed tomography (CT) imaging both play an important role in COVID-19 diagnosis. Chest CT imaging offers the benefits of quick reporting, a low cost, and high sensitivity for the detection of pulmonary infection. Recently, deep-learning-based computer vision methods have demonstrated great promise for use in medical imaging applications, including X-rays, magnetic resonance imaging, and CT imaging. However, training a deep-learning model requires large volumes of data, and medical staff faces a high risk when collecting COVID-19 CT data due to the high infectivity of the disease. Another issue is the lack of experts available for data labeling. In order to meet the data requirements for COVID-19 CT imaging, we propose a CT image synthesis approach based on a conditional generative adversarial network that can effectively generate high-quality and realistic COVID-19 CT images for use in deep-learning-based medical imaging tasks. Experimental results show that the proposed method outperforms other state-of-the-art image synthesis methods with the generated COVID-19 CT images and indicates promising for various machine learning applications including semantic segmentation and classification.

Machine Learning Detects Pattern of Differences in Functional Magnetic Resonance Imaging (fMRI) Data between Chronic Fatigue Syndrome (CFS) and Gulf War Illness (GWI)

BACKGROUND

Gulf War Illness (GWI) and Chronic Fatigue Syndrome (CFS) are two debilitating disorders that share similar symptoms of chronic pain, fatigue, and exertional exhaustion after exercise. Many physicians continue to believe that both are psychosomatic disorders and to date no underlying etiology has been discovered. As such, uncovering objective biomarkers is important to lend credibility to criteria for diagnosis and to help differentiate the two disorders.

METHODS

We assessed cognitive differences in 80 subjects with GWI and 38 with CFS by comparing corresponding fMRI scans during 2-back working memory tasks before and after exercise to model brain activation during normal activity and after exertional exhaustion, respectively. Voxels were grouped by the count of total activity into the Automated Anatomical Labeling (AAL) atlas and used in an "ensemble" series of machine learning algorithms to assess if a multi-regional pattern of differences in the fMRI scans could be detected.

RESULTS

A K-Nearest Neighbor (70%/81%), Linear Support Vector Machine (SVM) (70%/77%), Decision Tree (82%/82%), Random Forest (77%/78%), AdaBoost (69%/81%), Naïve Bayes (74%/78%), Quadratic Discriminant Analysis (QDA) (73%/75%), Logistic Regression model (82%/82%), and Neural Net (76%/77%) were able to differentiate CFS from GWI before and after exercise with an average of 75% accuracy in predictions across all models before exercise and 79% after exercise. An iterative feature selection and removal process based on Recursive Feature Elimination (RFE) and Random Forest importance selected 30 regions before exercise and 33 regions after exercise that differentiated CFS from GWI across all models, and produced the ultimate best accuracies of 82% before exercise and 82% after exercise by Logistic Regression or Decision Tree by a single model, and 100% before and after exercise when selected by any six or more models. Differential activation on both days included the right anterior insula, left putamen, and bilateral orbital frontal, ventrolateral prefrontal cortex, superior, inferior, and precuneus (medial) parietal, and lateral temporal regions. Day 2 had the cerebellum, left supplementary motor area and bilateral pre- and post-central gyri. Changes between days included the right Rolandic operculum switching to the left on Day 2, and the bilateral midcingulum switching to the left anterior cingulum.

CONCLUSION

We concluded that CFS and GWI are significantly differentiable using a pattern of fMRI activity based on an ensemble machine learning model.

Breast cancer detection using synthetic mammograms from generative adversarial networks in convolutional neural networks

The convolutional neural network (CNN) is a promising technique to detect breast cancer based on mammograms. Training the CNN from scratch, however, requires a large amount of labeled data. Such a requirement usually is infeasible for some kinds of medical image data such as mammographic tumor images. Because improvement of the performance of a CNN classifier requires more training data, the creation of new training images, image augmentation, is one solution to this problem. We applied the generative adversarial network (GAN) to generate synthetic mammographic images from the digital database for screening mammography (DDSM). From the DDSM, we cropped two sets of regions of interest (ROIs) from the images: normal and abnormal (cancer/tumor). Those ROIs were used to train the GAN, and the GAN then generated synthetic images. For comparison with the affine transformation augmentation methods, such as rotation, shifting, scaling, etc., we used six groups of ROIs [three simple groups: affine augmented, GAN synthetic, real (original), and three mixture groups of any two of the three simple groups] for each to train a CNN classifier from scratch. And, we used real ROIs that were not used in training to validate classification outcomes. Our results show that, to classify the normal ROIs and abnormal ROIs from DDSM, adding GAN-generated ROIs in the training data can help the classifier prevent overfitting, and on validation accuracy, the GAN performs about 3.6% better than affine transformations for image augmentation. Therefore, GAN could be an ideal augmentation approach. The images augmented by GAN or affine transformation cannot substitute for real images to train CNN classifiers because the absence of real images in the training set will cause over-fitting.

Breast cancer detection using synthetic mammograms from generative adversarial networks in convolutional neural networks

The convolutional neural network (CNN) is a promising technique to detect breast cancer based on mammograms. Training the CNN from scratch, however, requires a large amount of labeled data. Such a requirement usually is infeasible for some kinds of medical image data such as mammographic tumor images. Because improvement of the performance of a CNN classifier requires more training data, the creation of new training images, image augmentation, is one solution to this problem. We applied the generative adversarial network (GAN) to generate synthetic mammographic images from the digital database for screening mammography (DDSM). From the DDSM, we cropped two sets of regions of interest (ROIs) from the images: normal and abnormal (cancer/tumor). Those ROIs were used to train the GAN, and the GAN then generated synthetic images. For comparison with the affine transformation augmentation methods, such as rotation, shifting, scaling, etc., we used six groups of ROIs [three simple groups: affine augmented, GAN synthetic, real (original), and three mixture groups of any two of the three simple groups] for each to train a CNN classifier from scratch. And, we used real ROIs that were not used in training to validate classification outcomes. Our results show that, to classify the normal ROIs and abnormal ROIs from DDSM, adding GAN-generated ROIs in the training data can help the classifier prevent overfitting, and on validation accuracy, the GAN performs about 3.6% better than affine transformations for image augmentation. Therefore, GAN could be an ideal augmentation approach. The images augmented by GAN or affine transformation cannot substitute for real images to train CNN classifiers because the absence of real images in the training set will cause over-fitting.

Application of unsupervised learning to hyperspectral imaging of cardiac ablation lesions

Atrial fibrillation is the most common cardiac arrhythmia. It is being effectively treated using the radiofrequency ablation (RFA) procedure, which destroys culprit tissue and creates scars that prevent the spread of abnormal electrical activity. Long-term success of RFA could be improved further if ablation lesions can be directly visualized during the surgery. We have shown that autofluorescence-based hyperspectral imaging (aHSI) can help to identify lesions based on spectral unmixing. We show that use of k -means clustering, an unsupervised learning method, is capable of detecting RFA lesions without a priori knowledge of the lesions' spectral characteristics. We also show that the number of spectral bands required for successful lesion identification can be significantly reduced, enabling the use of increased spectral bandwidth. Together, these findings can help with clinical implementation of a percutaneous aHSI catheter, since by reducing the number of spectral bands one can reduce hypercube acquisition and processing times, and by increasing the spectral width of individual bands one can collect more photons. The latter is of critical importance in low-light applications such as intracardiac aHSI. The ultimate goal of our studies is to help improve clinical outcomes for atrial fibrillation patients.

Mid-term results of anatomical total shoulder arthroplasty for primary osteoarthritis using a short-stemmed cementless humeral component

Aims

The aim of this study was to evaluate the clinical and radiological outcome of using an anatomical short-stem shoulder prosthesis to treat primary osteoarthritis of the glenohumeral joint.

Patients and Methods

A total of 66 patients (67 shoulders) with a mean age of 76 years (63 to 92) were available for clinical and radiological follow-up at two different timepoints (T1, mean 2.6 years, sd 0.5; T2, mean 5.3 years, sd 0.7). Postoperative radiographs were analyzed for stem angle, cortical contact, and filling ratio of the stem. Follow-up radiographs were analyzed for timing and location of bone adaptation (cortical bone narrowing, osteopenia, spot welds, and condensation lines). The bone adaptation was classified as low (between zero and three features of bone remodelling around the humeral stem) or high (four or more features).

Results

The mean Constant score improved significantly from 28.5 (sd 11.6) preoperatively to 75.5 (sd 8.5) at T1 (p < 0.001) and remained stable over time (T2: 76.6, sd 10.2). No stem loosening was seen. High bone adaptation was present in 42% of shoulders at T1, with a slight decrease to 37% at T2. Cortical bone narrowing and osteopenia in the region of the calcar decreased from 76% to 66% between T1 and T2. Patients with high bone adaptation had a significantly higher mean filling ratio of the stem at the metaphysis (0.60, sd 0.05 vs 0.55, sd 0.06; p = 0.003) and at the diaphysis (0.65 sd 0.05 vs 0.60 sd 0.05; p = 0.007). Cortical contact of the stem was also associated with high bone adaptation (14/25 shoulders, p = 0.001). The clinical outcome was not influenced by the radiological changes.

Conclusion

Total shoulder arthroplasty using a short-stem humeral component resulted in good clinical outcomes with no evidence of loosening. However, approximately 40% of the shoulders developed substantial bone loss in the proximal humerus at between four and seven years of follow-up. Cite this article: Bone Joint J 2018;100-B:603–9.

Optimization of wavelength selection for multispectral image acquisition: a case study of atrial ablation lesions

In vivo autofluorescence hyperspectral imaging of moving objects can be challenging due to motion artifacts and to the limited amount of acquired photons. To address both limitations, we selectively reduced the number of spectral bands while maintaining accurate target identification. Several downsampling approaches were applied to data obtained from the atrial tissue of adult pigs with sites of radiofrequency ablation lesions. Standard image qualifiers such as the mean square error, the peak signal-to-noise ratio, the structural similarity index map, and an accuracy index of lesion component images were used to quantify the effects of spectral binning, an increased spectral distance between individual bands, as well as random combinations of spectral bands. Results point to several quantitative strategies for deriving combinations of a small number of spectral bands that can successfully detect target tissue. Insights from our studies can be applied to a wide range of applications.

The medium- and long-term outcome of total shoulder arthroplasty for primary glenohumeral osteoarthritis in middle-aged patients

Aims

Promising medium-term results from total shoulder arthroplasty (TSA) have been reported for the treatment of primary osteoarthritis in young and middle-aged patients. The aim of this study was to evaluate the long-term functional and radiological outcome of TSA in the middle-aged patient.

Patients and Methods

The data of all patients from the previous medium-term study were available. At a mean follow-up of 13 years (8 to 17), we reviewed 21 patients (12 men, nine women, 21 shoulders) with a mean age of 55 years (37 to 60). The Constant-Murley score (CS) with its subgroups and subjective satisfaction were measured. Radiological signs of implant loosening were analysed.

Results

Two shoulders (two patients) were revised and in two shoulders of two different patients, revision surgery was recommended. The mean CS increased from 23.3 (10 to 45) pre-operatively to 56.5 (26 to 81; p < 0.0001), but with a decrease in CS from 62.8 (38 to 93) to 56.5 (26 to 81) between medium- and long-term follow-up (p = 0.01). Without revision surgery, 18 patients (95%) rated their result as good or very good. The mean radiolucent line score for the glenoid components increased from 1.8 (0 to 6) to 8.2 (2 to 18) between medium- and long-term follow-up (p < 0.001).

Conclusion

TSA in young and middle-aged patients leads to improvement in clinical function and a relatively high satisfaction rate. However, clinical or radiological glenoid loosening worsens in the long term. Further studies are needed to optimise the treatment options in this patient population. Cite this article: Bone Joint J 2017;99-B:939–43.

A multicentre study of the long-term results of using a flat-back polyethylene glenoid component in shoulder replacement for primary osteoarthritis

We report the long-term clinical and radiological outcomes of the Aequalis total shoulder replacement with a cemented all-polyethylene flat-back keeled glenoid component implanted for primary osteoarthritis between 1991 and 2003 in nine European centres. A total of 226 shoulders in 210 patients were retrospectively reviewed at a mean of 122.7 months (61 to 219) or at revision. Clinical outcome was assessed using the Constant score, patient satisfaction score and range of movement. Kaplan-Meier survivorship analysis was performed with glenoid revision for loosening and radiological glenoid loosening (SD) as endpoints. The Constant score was found to improve from a mean of 26.8 (SD 10.3) pre-operatively to 57.6 (SD 20.0) post-operatively (p < 0.001). Active forward flexion improved from a mean of 85.3° (SD 27.4) pre-operatively to 125° (SD 37.3) postoperatively (p < 0.001). External rotation improved from a mean of 7° (SD 6.5) pre-operatively to 30.3° (SD 21.8°) post-operatively (p < 0.001). Survivorship with revision of the glenoid component as the endpoint was 99.1% at five years, 94.5% at ten years and 79.4% at 15 years. Survivorship with radiological loosening as the endpoint was 99.1% at five years, 80.3% at ten years and 33.6% at 15 years. Younger patient age and the curettage technique for glenoid preparation correlated with loosening. The rate of glenoid revision and radiological loosening increased with duration of follow-up, but not until a follow-up of five years. Therefore, we recommend that future studies reporting radiological outcomes of new glenoid designs should report follow-up of at least five to ten years.

Humeral surface replacement for the sequelae of fractures of the proximal humerus

Fractures of the proximal humerus can lead to malalignment of the humeral head, necrosis and post-traumatic osteoarthritis. In such cases surface replacement might be a promising option. A total of 28 shoulders with glenohumeral arthritis subsequent to a fracture underwent surface replacement arthroplasty of the humeral head in patients with a mean age of 60 years (35 to 83). On the basis of the inclination of the impacted head, post-traumatic arthritis was divided into three types: type 1, an impacted fracture of the head in an anatomical position (seven cases); type 2, a valgus impacted fracture (13 cases); type 3, a varus impacted fracture (eight cases). The outcome was measured by means of the Constant score. According to the Boileau classification of the sequelae of fractures of the proximal humerus, all 28 patients had a final result of intra-capsular category 1. The mean Constant score for the 28 shoulders increased from 23.2 points (2 to 45) pre-operatively to 55.1 points (20 to 89) at a mean of 31 months (24 to 66) post-operatively. Valgus impacted fractures had significantly better results (p < 0.039). Surface replacement arthroplasty can provide good results for patients with post-traumatic osteoarthritis of the shoulder. Their use avoids post-operative complications of the humeral shaft, such as peri-prosthetic fractures. Further surgery can be undertaken more easily as the bone stock is preserved.

[Significance of bone mineral density and modern cementing technique for in vitro cement penetration in total shoulder arthroplasty]

AIM

Loosening of the glenoid component is one of the major causes of failure in total shoulder arthroplasty. Possible risk factors for loosening of cemented components include an eccentric loading, poor bone quality, inadequate cementing technique and insufficient cement penetration. The application of a modern cementing technique has become an established procedure in total hip arthroplasty. The goal of modern cementing techniques in general is to improve the cement-penetration into the cancellous bone. Modern cementing techniques include the cement vacuum-mixing technique, retrograde filling of the cement under pressurisation and the use of a pulsatile lavage system. The main purpose of this study was to analyse cement penetration into the glenoid bone by using modern cement techniques and to investigate the relationship between the bone mineral density (BMD) and the cement penetration. Furthermore we measured the temperature at the glenoid surface before and after jet-lavage of different patients during total shoulder arthroplasty. It is known that the surrounding temperature of the bone has an effect on the polymerisation of the cement. Data from this experiment provide the temperature setting for the in-vitro study.

METHOD

The glenoid surface temperature was measured in 10 patients with a hand-held non-contact temperature measurement device. The bone mineral density was measured by DEXA. Eight paired cadaver scapulae were allocated (n = 16). Each pair comprised two scapulae from one donor (matched-pair design). Two different glenoid components were used, one with pegs and the other with a keel. The glenoids for the in-vitro study were prepared with the bone compaction technique by the same surgeon in all cases. Pulsatile lavage was used to clean the glenoid of blood and bone fragments. Low viscosity bone cement was applied retrogradely into the glenoid by using a syringe. A constant pressure was applied with a modified force sensor impactor. Micro-computed tomography scans were applied to analyse the cement penetration into the cancellous bone.

RESULTS

The mean temperature during the in-vivo arthroplasty of the glenoid was 29.4 °C (27.2-31 °C) before and 26.2 °C (25-27.5 °C) after jet-lavage. The overall peak BMD was 0.59 (range 0.33-0.99) g/cm (2). Mean cement penetration was 107.9 (range 67.6-142.3) mm (2) in the peg group and 128.3 (range 102.6-170.8) mm (2) in the keel group. The thickness of the cement layer varied from 0 to 2.1 mm in the pegged group and from 0 to 2.4 mm in the keeled group. A strong negative correlation between BMD and mean cement penetration was found for the peg group (r (2) = -0.834; p < 0.01) and for the keel group (r (2) = -0.727; p < 0.041). Micro-CT shows an inhomogenous dispersion of the cement into the cancellous bone.

CONCLUSIONS

Data from the in-vivo temperature measurement indicate that the temperature at the glenohumeral surface under operation differs from the body core temperature and should be considered in further in-vitro studies with human specimens. Bone mineral density is negatively correlated to cement penetration in the glenoid. The application of a modern cementing technique in the glenoid provides sufficient cementing penetration although there is an inhomogenous dispersion of the cement. The findings of this study should be considered in further discussions about cementing technique and cement penetration into the cancellous bone of the glenoid.

[Cementless humeral surface replacement arthroplasty in patients less than 55 years of age]

BACKGROUND

Cementless humeral surface replacement arthroplasty is a viable treatment option for degenerative diseases of the shoulder joint. The aim of this prospective study was to analyse the results of this treatment option in patients younger than 55 years of age with different pathologies of the shoulder.

PATIENTS AND METHODS

Twenty-three patients (26 implants) treated with cementless humeral surface replacement arthroplasty were included in this study. Mean follow-up was 2.5 years (1-6 years). Ten patients had posttraumatic osteoarthritis, seven had primary osteoarthritis, and six had osteonecrosis. Patients were evaluated using the Constant score, shoulder motion, and subjective satisfaction.

RESULTS

The mean Constant score increased significantly from 33 points preoperatively (8-69 points) to 61 points postoperatively (25-83 points; p<0.0001), adjusted to age and gender from 38% (8-86%) to 70% (28-114%; p<0.0001). Significant improvement for the whole cohort was found regarding patients' pain, activity, mobility, shoulder flexion and abduction, and internal and external rotation (p<0.001). In one case, reoperation was necessary due to a superficial wound infection, and in another case, implant revision to a total shoulder replacement was performed because of glenoid erosion.

CONCLUSION

Cementless humeral surface replacement arthroplasty is a viable bone-preserving treatment option for young and active patients. Later conversion to total shoulder replacement is possible. Good clinical results, a low complication rate, and high patient satisfaction were found in the short and mid term. Long-term investigations are necessary to confirm these observations.

Mid-term survivorship analysis of a shoulder replacement with a keeled glenoid and a modern cementing technique

We have investigated the mid-term outcome of total shoulder replacement using a keeled cemented glenoid component and a modern cementing technique with regard to the causes of failure and loosening of the components. Between 1997 and 2003 we performed 96 total shoulder replacements on 88 patients, 24 men and 64 women with a mean age of 69.7 years (31 to 82). The minimum follow-up was five years and at the time of review 87 shoulders (77 patients) were examined at a mean follow-up of 89.1 months (60 to 127). Cumulative survival curves were generated with re-operations (accomplished and planned), survivorship of the proshesis, loosening of the glenoid (defined as tilt > 5 degrees or subsidence > 5 mm), the presence of radiolucent lines and a Constant score of < 30 as the endpoints. There were two re-operations not involving revision of the implants and the survival rate of the prosthesis was 100.0% for the follow-up period, with an absolute Constant score of > 30 as the endpoint the survival rate was 98%. Radiological glenoid loosening was 9% after five years, and 33% after nine years. There was an incidence of 8% of radiolucent lines in more than three of six zones in the immediate post-operative period, of 37.0% after the first year which increased to 87.0% after nine years. There was no correlation between the score of Boileau and the total Constant score at the latest follow-up, but there was correlation between glenoid loosening and pain (p = 0.001). We found that total shoulder replacement had an excellent mid-term survivorship and clinical outcome. The surgical and cementing techniques were related to the decrease in radiolucent lines around the glenoid compared with earlier studies. One concern, however, was the fact that radiolucent lines increased over time and there was a rate of glenoid loosening of 9% after five years and 33% after nine years. This suggests that the design of the glenoid component, and the implantation and cementing techniques may need further improvement.

[Comparison of short-term results after CUP prosthesis with cemented glenoid components and total shoulder arthroplasty: a matched-pair analysis]

AIM

The purpose of this clinical study (matched-pair design) was to compare the functional short-term results obtained in patients with surface replacement of the humeral head and cemented glenoid prosthesis with those obtained after total shoulder arthroplasty.

METHOD

20 patients (average age 67.7 [43-85] years, 17 women, three men) who received surface replacement of the humeral head with cemented glenoid prosthesis were matched to a control group of 20 patients (average age 67.55 [42-85] years, 17 women, three men) with a conventional total shoulder arthroplasty. Six patients were treated for osteoarthritis, two for post-traumatic arthritis, 1 each for osteonecrosis and rheumatoid arthritis. Preoperative status, perioperative results and postoperative status (Constant score, subjective assessment, range of motion, radiographic evaluation) were compared in all patients and controls.

RESULTS

The adjusted Constant score improved from a mean of 37.25% to a mean of 87.75% in the hybrid group and from a mean of 30.8% to a value of 87.1% in the TSA group. Regarding the relative improvement at 12 months compared to baseline, patients treated with hybrid prostheses showed a comparable benefit in the Constant score, pain reduction and range of motion. Only the criterion "strength" revealed a significantly better result in the TSA group (p = 0.025). There was one irreversible injury of the brachial plexus in one case and neural injuries with a full recovery in two cases of Hybrid prosthesis.

CONCLUSIONS

The combination of humeral surface replacement with cemented glenoid component offers a relatively new option for the treatment of different pathologies at the shoulder joint which need a total joint substitute.The short-term results are comparable with those of conventional total shoulder arthroplasty. Surface replacement of the shoulder facilitates later revision because of less loss of bone stock. It must be considered that surface replacement with implantation of cemented glenoid prosthesis is a difficult procedure because of the exploration of the glenoid.

Total shoulder replacement in young and middle-aged patients with glenohumeral osteoarthritis

Our aim in this prospective study was to evaluate the outcome of total shoulder replacement in the treatment of young and middle-aged active patients with primary glenohumeral osteoarthritis. We reviewed 21 patients (21 shoulders) with a mean age of 55 years (37 to 60). The mean follow-up was seven years (5 to 9). The same anatomical, third-generation, cemented implant had been used in all patients. All the patients were evaluated radiologically and clinically using the Constant and Murley score.

No patients required revision. In one a tear of the supraspinatus tendon occurred. Overall, 20 patients (95%) were either very satisfied (n = 18) or satisfied (n = 2) with the outcome. Significant differences (p < 0.0001) were found for all categories of the Constant and Murley score pre- and post-operatively. The mean Constant and Murley score increased from 24.1 points (10 to 45) to 64.5 points (39 to 93), and the relative score from 30.4% (11% to 50%) to 83% (54% to 116%). No clinical or radiological signs of loosening of the implant were seen.

For young and middle-aged patients with osteoarthritis, third-generation total shoulder replacement is a viable method of treatment with a low rate of complications and excellent results in the mid-term.

[Hemiarthroplasty or total shoulder replacement in glenohumeral osteoarthritis?]

Shoulder arthroplasty has proven to be a successful option in the treatment of glenohumeral osteoarthritis. Although introduced in the 1970s by Ch. Neer, resurfacing of the glenoid remains controversial. With regard to some single prospective studies total shoulder replacement (TSR) seems to be superior to humeral head replacement (HHR) in the treatment of concentric osteoarthritis. The current literature shows that TSR leads to a better shoulder function and less pain. The decision on whether to perform a TSR or HHR should be guided by the individual pathology. The final treatment should be reliable in terms of a better shoulder function, reduced pain, and good strength. The role of glenoid loosening in TSR and its impact on shoulder function in the long term remains uncertain and has to be clarified in the future.

Humeral head resurfacing for fixed anterior glenohumeral dislocation

The purpose of this prospective study was to describe cementless humeral surface replacement arthroplasty (CHSRA) as a bone preserving treatment option for patients with fixed anterior glenohumeral dislocation. Ten patients with post-traumatic fixed anterior glenohumeral dislocation underwent CHSRA with a mean follow-up of 24 months. All patients were evaluated clinically using the Constant score and with radiographs in two planes. There were two reoperations: one patient developed glenoid erosion and was revised and in another case redislocation occurred. Clinical or radiographical signs of implant loosening were not found. The humeral head centred in the glenoid in nine out of ten cases radiographically. The Constant score increased from 20 points preoperatively to 61 points postoperatively (p < 0.007). CHSRA is a viable treatment option for elderly patients with fixed anterior glenohumeral dislocation and bone defects of the humeral head. Good clinical results and a moderate complication rate were found in the short term.

Analytical-form model observers for decompressed images

We report a method for evaluating the performance of model observers for decompressed images in analytical form using compression noise statistics. It derives test statistics and detectabilities for the ideal observer, the nonprewhitening observer, the Hotelling observer, and the channelized Hotelling observer (CHO) on decompressed images. The derived CHO performance is validated using the Joint Photographic Experts Group (JPEG) compression algorithm. The validation results show that the derived CHO receiver operating characteristics (ROCs) and areas under ROC curves predict accurately their corresponding estimated values. These analytical-form quality measures of decompressed images provide a way to optimize compression algorithms analytically, subject to a model-observer performance criterion. They also provide a theoretical foundation for efforts to create a model observer for decompressed images.

Das Bewegungsausmaß der Schulter und des Ellenbogens bei Alltagsbewegungen in der 3D-Bewegungsanalyse

AIM

Three-dimensional motion analysis of the lower limb has been an approved method of diagnosis and therapy planning for several years. In this study we observed the range of motion of the shoulder and elbow in 10 activities of daily life (ADL) with a marker-based biomechanical model for the upper extremity. With this database we hope to improve the evaluation of different handicaps of the upper limb.

METHOD

The used biomechanical model is based on 14 infrared light-reflecting markers. The ranges of motion in ADL for shoulder and elbow were measured in a standardised case setting in seven test persons with a mean age of 25 years (SD 15 years). The 10 observed ADL were eating with a spoon, combing hair, genital hygiene, using a telephone, typing on a keyboard, drinking from a glass, turning a key, turning a page, pouring water in a glass and drawing.

RESULTS

For the ten explored ADL, the test persons needed a range of motion in the shoulder of 91 - 0 - 9 degrees (total 100 degrees) flexion/extension, 112 - 23 - 0 degrees (total 89 degrees) abduction/adduction, and 91 - 0 - 114 degrees (total 205 degrees) external/internal rotation. Most of the ADL were performed in external rotation and, excluding the motion genital hygiene, the test persons only needed an internal rotation of 10 degrees. Maximal shoulder flexion was used with opening a door, the minimum was reached with genital hygiene. The maximum angles of abduction and rotation were reached with combing hair and the minimum values were reached with genital hygiene. To perform the ADL, an elbow extension/flexion of 0 - 36 - 146 degrees (total 110 degrees), and 55 - 0 - 72 degrees (total 127 degrees) pro-/supination was needed. Maximal pronation was reached with "pour from a pitcher". Maximal supination was present with genital hygiene.

CONCLUSIONS

The decisive benefit of 3D motion analysis is the exact capturing of complex and dynamic movements at any time. Therefore, not only static joint positions can be recorded, but also the dynamic course of a movement can be traced. By using our model on every day movements, we were able to collect data that can serve as the basis for the desired range of motion of the upper extremities in patients.

Endoprothetik bei der Rotatorenmanschetten

The treatment of cuff tear arthropathy (CTA) of the shoulder is based on the patient's subjective complaints, functional deficits, and the treatment goals that result from both. Only in the second line is the radiologic morphology of joint destruction important for the surgeon's decision. Based on these aspects it is possible to discriminate three types of CTA that obviously have similar functional deficits and radiological characteristics. This symptom-based classification can offer decision guidelines for the choice of an appropriate joint replacement.For the most frequent, stable, and osteoarthritic type with pain during motion and a tolerably restricted active and passive range of motion (ROM) surface replacement or hemiarthroplasty can lead to reasonable results. In the unstable type with only minimal radiological changes of the humeral head and the chief complaint of severe restriction of active ROM and in some cases recurrent shoulder dislocations, hemiarthroplasty with a glenoid reconstruction shell or reversed prosthesis are promising. A reverse prosthesis is the implant of choice for the necrotic type of CTA with extensive destruction of the humeral head, the glenoid, and the adjacent bony structures.

Differenzialtherapie massiver Rotatorenmanschettenläsionen

The treatment of massive rotator cuff tears must be adapted to the patient's individual needs and preoperative parameters to achieve the best outcome. First, the shoulder surgeon has to determine whether a direct transosseous repair is possible. If there is not enough remaining tissue, the tissue is atrophic, and the tendon stump can be reduced only with great tension, one can use a margin convergence technique for partial closure, perform a biceps tendoplasty, or perform local tendon transfers with the subscapularis or infraspinatus muscle. If the defect cannot be sufficiently closed, elderly patients with low demands can be treated with tubercleplasty/subacromial decompression, whereas patients younger than 60 years with higher demands should receive muscle and tendon transfers. A balanced posterosuperior defect can be reconstructed by a deltoid muscle transfer, in contrast to an unbalanced one, which is best treated with an active transfer of the latissimus dorsi muscle and tendon. Anterosuperior defects can be addressed by a pectoralis muscle transfer. If the humeral head is superiorly migrated, if signs of osteoarthritis are present, and if the patient is older than 70 years, a reverse prosthesis can be implanted as a salvage procedure.

Direct magnetic resonance detection of neuronal electrical activity

Present noninvasive neuroimaging methods measure neuronal activity indirectly, via either cerebrovascular changes or extracranial measurements of electrical/magnetic signals. Recent studies have shown evidence that MRI may be used to directly and noninvasively map electrical activity associated with human brain activation, but results are inconclusive. Here, we show that MRI can detect cortical electrical activity directly. We use organotypic rat-brain cultures in vitro that are spontaneously active in the absence of a cerebrovascular system. Single-voxel magnetic resonance (MR) measurements obtained at 7 T were highly correlated with multisite extracellular local field potential recordings of the same cultures before and after blockade of neuronal activity with tetrodotoxin. Similarly, for MR images obtained at 3 T, the MR signal changed solely in voxels containing the culture, thus allowing the spatial localization of the active neuronal tissue.

Incidence of rotator cuff tears in shoulder dislocations and results of therapy in older patients

INTRODUCTION

In older patients we documented the incidence of additional injuries in shoulder dislocations in a non-randomised, prospective study and compared the results of conservative and of operative therapy of rotator cuff tears.

MATERIAL AND METHODS

Between 1993 and 1999 a total of 87 patients were prospectively enrolled. In the case of documented tears of the rotator cuff in combination with symptoms persisting after conservative therapy patients were free to decide between surgical and conservative treatment. Patients were monitored for function, subjective stability, and satisfaction over a minimum follow-up period of 1 year.

RESULTS

In 54% of the patients enrolled a cuff tear was documented; the frequency increased with advancing age to 100% in patients over the age of 70. Surgical treatment of the cuff tears resulted in better function and satisfaction. There were three recurrences in the conservative treatment group, while no recurrences were documented in surgically treated patients. The significant benefit of cuff repair was confirmed by regression analysis.

CONCLUSION

In this selected patient group, we believe surgical repair of the symptomatic rotator cuff tear should be discussed with the patient.

Influence of the design of the prosthesis on the outcome after hemiarthroplasty of the shoulder in displaced fractures of the head of the humerus

We reviewed 39 patients with displaced three- and four-part fractures of the humerus. In 21 patients (group A) we had used an anatomical prosthesis for the humeral head and in 18 (group B) an implant designed for fractures.

When followed up at a mean of 29.3 months after surgery the overall Constant score was 51.9 points; in group A it was 51.5 and in group B 52.4 points. The subjective satisfaction of the patients was assessed using a numerical rating scale and was similar in both groups. In group A complete healing of the tuberosities was found in 29% and 50% in group B. Partial integration was seen in 29% of group A and in only one patient in group B, while resorption was noted in 43% of group A and 44% of group B. The functional outcome was significantly better in patients with complete or partial healing of the tuberosities (p = 0.022). The specific trauma prosthesis did not lead to better healing of the tuberosities. The difference in clinical outcome obtained by the two designs did not reach statistical significance.

[Efficiency of a postoperative treatment after rotator cuff repair with a continuous passive motion device (CPM)]

AIM

The main objective of this study was to prove that a postoperative combined continuous passive motion (CPM) and physiotherapy treatment protocol (CPM group) can achieve 90 degrees active abduction in the shoulder joint earlier than physiotherapy alone (PT group). The indication was a complete tear of the rotator cuff.

METHOD

The study was conducted under in-patient and out-patient conditions. 55 patients were included in this study. The prospective, randomized multicenter study design complies with DIN EN 540. The primary endpoint was the time span until 90 degrees active abduction was achieved by the patients.

RESULTS

Patients in the CPM group reached the primary endpoint on average 12 days earlier than the control group. This difference was statistically significant (p = 0.0292). Analyzing the secondary endpoints, e. g., pain and disablement, the results in the CPM group showed again advantages of the combined treatment protocol (CPM + physiotherapy).

CONCLUSION

The postoperative treatment of a total tear of the rotator cuff with a combined continuous passive motion and physiotherapy protocol provided a significantly earlier range of motion in the shoulder joint than physiotherapy alone. There was no report of CPM-related adverse effects.

Discrimination of photon from proton irradiation using glow curve feature extraction and vector analysis

Two types of thermoluminescence dosemeters (TLDs), the Harshaw LiF:Mg,Ti (TLD-100) and CaF(2):Tm (TLD-300) were investigated for their glow curve response to separate photon and proton irradiations. The TLDs were exposed to gamma irradiation from a (137)Cs source and proton irradiation using a positive ion accelerator. The glow curve peak structure for each individual TLD exposure was deconvolved to obtain peak height, width, and position. Simulated mixed-field glow curves were obtained by superposition of the experimentally obtained single field exposures. Feature vectors were composed of two kinds of features: those from deconvolution and those taken in the neighbourhood of several glow curve peaks. The inner product of the feature vectors was used to discriminate among the pure photon, pure proton and simulated mixed-field irradiations. In the pure cases, identification of radiation types is both straightforward and effective. Mixed-field discrimination did not succeed using deconvolution features, but the peak-neighbourhood features proved to discriminate reliably.

Intramuskuläre Längenänderungen und Bewegungsausmaße nach lokalen Sehnentransfers zur Deckung eines retrahierten Supraspinatussehnendefektes

BACKGROUND

The current cadaver study was performed to evaluate and compare infraspinatus (ISP) vs. subscapularis (SSC) tendon transfer used for the repair of a large supraspinatus (SSP) rotator cuff defect. The main outcome measures were the range of motion and intramuscular length changes during motion.

METHODS

Thirteen fresh-frozen cadaver shoulders were used in the study. A supraspinatus defect extending to the apex of the humeral head (Patte size II) was created. Transosseous repair was then attempted with the ISP and with the SSC in all cases; intramuscular length changes were measured by Hall effect transducers during motion of the arm in steps of 30 degrees.

RESULTS

Repair was successful in all cases when the ISP was used, while use of the SSC resulted in a successful repair in only 8 of the 13 (61.5%). Passive range of motion did not differ between ISP and SSC. ISP was shortened during flexion and abduction in contrast to the SSC that was lengthened during flexion (p<0.05) and shortened to a lesser degree (p<0.05). Length changes during rotation were not significantly different.

CONCLUSIONS

In this cadaver model the ISP proved more favorable than the SSC for covering a Patte size II SSP defect if the assumption is true that lengthening of a muscle is less favorable than shortening because of the vascularization of the tendon.

Die Dezentrierung des Schultergelenkes als Spätfolge nach hemi- und totalendoprothetischem Gelenkersatz

BACKGROUND

Migration after shoulder arthroplasty can induce muscular dysbalance and instability of the shoulder joint. The aim of this study was to analyse the frequency, causes and consequences of secondary migration after shoulder arthroplasty.

METHODS

107 shoulder arthroplasties in patients with primary osteoarthritis (OA, n = 83) or avascular necrosis of the humeral head (AVN, n = 24), were followed clinically and radiographically over a period of 1 to 10 years.

RESULTS

17 Patients (16 %) developed secondary migration of the prosthesis. These were determined after a mean period of 33 months after shoulder arthroplasty. Patients with secondary migration of the prosthesis showed a worse function with a mean Constant score (CS) of 52.3 points in comparison to the patient group without migrated shoulder arthroplasties (62 points) (p = 0.038). In patients with total shoulder arthroplasty (TSA, n = 75), migration was observed in 13 % of cases during follow-up. Lucent lines at the glenoid were found in radiographs in 55 % of migrated shoulders and in 52 % of non-migrated shoulders (p = 0.633). The Constant score showed no significant difference between the groups (non-migrated shoulders 63 points, migrated shoulders 61 points, p > 0.5). Migration was seen in 22 % (7/32 patients) after hemiarthroplasty, however, no significant difference could be found between the respective Constant scores for migrated (47.7 points) or non-migrated (55.1 points) shoulders (p = 0.447). In patients with OA there was no significant difference in Constant score between migrated and non-migrated shoulders (p = 0.331), whereas patients with AVN showed significantly worse function after migrated shoulder arthroplasty (p = 0.007).

CONCLUSION

Migration after total shoulder arthroplasty leads in the medium term neither to an increased frequency of lucent lines at the glenoid nor to a worsening of the functional score in comparison to non-migrated prosthesis. In patients with AVN, migration is accompanied by a worsening of the functional score and can consequently be regarded as a negative prognostic factor.

Posttraumatische Schulterinstabilit�t im Wachstumsalter

BACKGROUND

Post-traumatic shoulder instability in adolescence represents a very rare condition. However, most of these patients develop recurrent shoulder dislocations over time.

METHODS

We report about a retrospective study, which included 32 patients younger than 16 years of age. The mean follow-up was 5 years. Our evaluation focussed on patients' age at the time of first shoulder dislocation, type of injury, and the results after conservative and operative treatment.

RESULTS

After the first dislocation, all shoulders were immobilized for 3 weeks and were treated with physiotherapy afterwards. Of these 32 patients, 30 (94%) developed recurrent shoulder dislocations. Eight patients remained in a conservative regimen (age <15 years) and 21 patients were operated for persistent shoulder instability (age >15 years). The recurrence rate was 8 of 21 patients (36.5%) in the operative group and 4 of 8 patients in the conservative group. At the time of follow-up, 4 of 8 conservatively treated patients had returned to their former sports activities and 15 of 21 (71%) in the operated group.

CONCLUSIONS

Our results on post-traumatic shoulder instability in adolescence show high recurrence rates in both conservative and operative groups. After adolescence, including ossification of the glenoid, operative treatment is able to decrease the recurrence rate and to increase the sports ability of these young patients.