13 patients were female and the average patient age was 76 (range, 64–84) years. In 66% the right side was operated on. RTSA was performed with prosthesis by Lima (proximal humerus fractures) and Tornier/Wright, Aequalis/Ascend II (cuff arthropathy). The follow up including the ultrasound examination was carried out on average 15 months postoperatively (range, 4–48).
With an average BMI of 25.9 (range, 22.1–30.1) the cohort was slightly obese. Two patients were still active smokers. In average every patient had at least two comorbidities, more than a third of the patients had multi medication (> 5 daily). In this small group is was not able to correlate the comorbidities with the functional outcome.
In total the recorded functional outcome was satisfactory. All patients after RSA due to osteoarthritis claimed they would undergo surgery again. In average 18 points were achieved in the VAS for pain (range, 5–47) and 64 points in the VAS for function (range, 64–97). The Constant Murley Score showed an average of 66 points (range, 35–89). In general patients with RTSA after cuff arthropathy showed better functional results than patients suffering from proximal humeral fractures and also showed a higher satisfaction rate. The average range of motion consisted of 150° of abduction, 25° of internal rotation, 140° of elevation and 25° of external rotation.
In our collective there were no cases of infection and no necessary revisions or periprosthetic fractures.
The force measurement was performed on all patients. As expected patients were not able to create the same amount of force on the RTSA side in comparison to the healthy side. In average 48.8N were achieved after RTSA in comparison to 58.3 N on the healthy side as the maximum capable strength. This difference was only merely significant (p > 0.07). We were not able to find any connection between muscle elasticity and the postoperative functional outcome shown in the scores.
Radiological assessment was performed on the most recent available shoulder x rays. In average the prosthesis was positioned at an offset of 39.07 mm with a tilt of 3.2°. The glenosphere was size 36 in all but two cases in which the size was 42 and was implanted in a slight retroversion of 3°. The average acromio-humeral distance was 28.5 mm (range 18.2–33.9 mm).
In all of the examined areas of the deltoid muscle under relaxed circumstances the shear wave elastography SWE was able to show a higher muscle tension of the deltoid in patients after RSA compared to the contralateral non-operated side. The differences were particularly visible in the pars clavicularis PC and pars acromialis PA. There was a statistical significant difference, p < 0.05 which did not show in the posterior part, pars spinalis PS (Fig. 3).
Measurements were repeated on both sides under isometric load. Under the force all regions of the deltoid muscle adapted to new condition independent of RSA with a significant increase of elasticity, p < 0.001 (Fig. 4). This change of elasticity underlines SWE as a capable method of detecting tissue properties in the deltoid.
Taking a closer look at the elasticity increase under isometric loading there is a difference in increase of RSA and the healthy side. As seen, all deltoid areas showed a higher tension during the load but the amount of increase of elasticity is again to be significantly higher after implantation of RSA (Fig. 5). This increase can only be see in the pars clavicularis and the pars acromialis and is only significantly higher in the pars clavicularis (RSA: 238.96 ± 160; Healthy side: 154.57 ± 114; p > 0.04) (Fig. 6,7).
Again we could not correlate the clinical and functional outcome with the elastography findings.
Limitations of this study are to be taken into account a.e. the small amount of patients as well as the heterogenic cohort (fractures, osteoarthritis etc). Further consideration has to go to the isometric loading which was not performed under fully controlled conditions.