The purpose of this study was to examine whether TSA is able to restore normal ROM in ADLs in patients with degenerative osteoarthritis of the glenohumeral joint over the course of 3 years. We analyzed shoulder motion pre- and postoperatively by three-dimensional video motion analysis using the HUX model as described previously
 and showed that TSA improves ROM in performing ADLs in patients with degenerative glenohumeral osteoarthritis. Comparison of maximum available ROM and maximum used ROM in performing the tested tasks of ADL in the control group showed that they used their full available ROM in abduction and adduction while in flexion tasks, they didn’t use their fully available ROM to perform the four ADLs (Table
2). In extension, the controls, and the patients’ pre- and postoperative follow-up showed the same phenomenon of having a higher ROM in performing ADLs than in maximum value task. We think that this could be explained by different initial positions. Performing the maximum value task, the subject started from a neutral position while in performing ADLs the reached maximum was independent from a neutral position. In adduction this phenomenon appeared only in the preoperative patient group.
Motion analysis of the upper extremity is challenging because of the great glenohumeral ROM. As the gold standard, maximum ROM is measured with a goniometer which, however, gives inaccuracies of about 5° to 10° degrees and fails to distinguish glenohumeral joint motion from trunk, spine, or scapulothoracical motion
[6, 18–27]. Patients are asked about their ability to perform ADLs for the CS
[5, 28, 29] but there is no objective assessment of ADLs in clinical routine. The novel HUX model using a 3 D motion analysis system allows for an exact and dynamic capture the movement in the calculated shoulder joint center in relation to the torso without impairment of the motion by e.g. heavy equipment. This means that the success of surgery in ADL can be evaluated better. In the present study, patients who received TSA were examined with respect to maximum glenohumeral ROM plus the ROM in performing ADLs in order to address the following questions: Does the maximum glenohumeral ROM change after TSA? Can shoulder arthroplasty restore a normal ROM in patients with omarthrosis? Are there changes in the ROM used when performing ADLs after shoulder replacement?
The results show that TSA improves the ROM in ADLs but when performing ADLs, TSA patients do not use the maximum available ROM in flexion, abduction and adduction. How can that be explained?
In flexion and adduction the controls also didn’t use more ROM in performing the four ADLs. Therefore we conclud, that it is not necessary to use more flexion and adduction ROM to perfrom the four ADLs. In abduction, however, controls needed 118° of their maximum available 113° to perform the ADLs, while the patients only use 76° of their maximum available 101°. Maybe they used compensatory motion patterns? In the literature, a study of Magermans et al.
 measured the shoulder motion of healthy subjects with an electromagnetic tracking device while performing ADLs. Their goal was to find the minimal requirements to perform ADL and how these ADL are performed in healthy subjects. For combing the hair, their subjects used at least 73° of glenohumeral elevation comparable to our control group. During arm elevation when combing hair, a large (20–100°) glenohumeral elevation motion was observed by Magermans et al. This shows that the results of the present study are comparable to published data. Veeger et al.
 hypothesized that the ability to perform an ADL is related to a compensatory movement implementation by means of clavicular retraction. They examined patients after TEP and HEP implantation and divided the patients into two groups according to their ability to comb their hair. Comparable to our results, both patient groups showed limitations in glenohumeral ROM postoperatively as compared to controls. Among patient groups, only axial rotation ROM was different: the ‘Able’ group had a larger external rotational ROM, but less internal rotation. During the ADL ‘combing hair’ the “Able” group appeared to successfully perform the task by applying a larger degree of clavicular retraction. They concluded that functional outcome after arthroplasty is limited due to a lack of glenohumeral ROM but that it is possible to compensate for this restriction by mechanisms such as greater clavicular retraction. The clavicular retraction might be related to a more efficient scapulothoracic motion. In comparison with other published studies
, our TSA patients showed less postoperative maximum active ROM. The comparative values in the literature were determined by using a conventional goniometer or an electric goniometer, with which it is difficult to adequately distinguish shoulder joint motion from compensatory movements of the trunk. Our measurement method seems to better illustrate pure thoracohumeral joint motion. Under this assumption, we showed that the improved ROM at early follow-up (6 months) after shoulder arthroplasty increased again significantly in the further course (3-year follow-up).
Postoperatively, TSA usually allows each patient to perform all required ADLs painlessly. Considering the maximum ROM in the TSA group, TSA patients were not using their maximum available abduction ROM to perform the ADLs 3 years postoperatively. This is not related to limitations in active abduction ROM, but rather may be caused by impaired proprioception. As the study group showed in a previous investigation
, proprioception that was measured by an active angle reproduction tended to deteriorate after shoulder arthroplasty. This might be related to the deltopectoral approach that includes divison of the subscapularis muscle and the glenohumeral ligaments and cause alterations in movement patterns. Another explanation could be that the osteoarthritis patients develop impaired motion patterns in ADL that could not be reversed by the sheer possibility to move the arm further because the patients were not sufficiently trained to use the new ROM. A limitation of the study is the relatively low number of subjects. This might disguise further significant changes. We did not examine the state of rotator cuff at the last follow-up. This parameter could also influence the result of the study. The study setting is, however, very elaborate and time consuming, and cannot be used in a routine follow-up.