Shoulder Surgery Postoperative Immobilization: An International Survey of Shoulder Surgeons

Simple Summary No consensus currently exists on immobilization protocols following shoulder surgery. The aim of this study was to identify patterns and types of sling used by surgeons from the United States and Europe for a variety of shoulder surgical procedures and further to identify factors associated with the variations. Four-hundred and ninety-nine surgeons with a median 15 years of experience responded, with 54.7% from the United States and 45.3% from Europe. United States surgeons reported higher abduction pillow sling use than European surgeons, whereas European surgeons reported more simple sling utilization. Increasing experience was negatively correlated with sling duration, meaning more experienced respondents tended to recommend shorter durations of sling use. Considerable variation exists in the immobilization patterns after a variety of shoulder surgical procedures advocated by surgeons with apparent influence from both geographic location and years of clinical experience. Future work is required to establish the most clinically beneficial protocols for immobilization following shoulder surgery. Abstract Background: There is currently no consensus on immobilization protocols following shoulder surgery. The aim of this study was to establish patterns and types of sling use for various surgical procedures in the United States (US) and Europe, and to identify factors associated with the variations. Methods: An online survey was sent to all members of the American Shoulder and Elbow Society (ASES) and European Society for Surgery of the Shoulder and Elbow (ESSSE). The survey gathered member data, including practice location and years in practice. It also obtained preferences for the type and duration of sling use after the following surgical procedures: arthroscopic Bankart repair, Latarjet, arthroscopic superior/posterosuperior rotator cuff repair (ARCR) of tears <3 cm and >3 cm, anatomic total shoulder arthroplasty (aTSA) and reverse TSA (rTSA), and isolated biceps tenodesis (BT). Relationships between physician location and sling type for each procedure were analyzed using Fisher’s exact tests and post-hoc tests using Bonferroni-adjusted p-values. Relationships looking at years in practice and sling duration preferred were analyzed using Spearman’s correlation tests. Results: In total, 499 surgeons with a median of 15 years of experience (IQR = 9–25) responded, with 54.7% from the US and 45.3% from Europe. US respondents reported higher abduction pillow sling use than European respondents for the following: Bankart repair (62% vs. 15%, p < 0.0001), Latarjet (53% vs. 12%, p < 0.001), ARCR < 3 cm (80% vs. 42%, p < 0.001) and >3 cm (84% vs. 61%, p < 0.001), aTSA (50% vs. 21%, p < 0.001) and rTSA with subscapularis repair (61% vs. 22%, p < 0.001) and without subscapularis repair (57% vs. 17%, p < 0.001), and isolated BT (18% vs. 7%, p = 0.006). European respondents reported higher simple sling use than US respondents for the following: Bankart repair (74% vs. 31%, p < 0.001), Latarjet (78% vs. 44%, p < 0.001), ARCR < 3 cm (50% vs. 17%, p < 0.001) and >3 cm (34% vs. 13%, p < 0.001), and aTSA (69% vs. 41%, p < 0.001) and rTSA with subscapularis repair (70% vs. 35%, p < 0.001) and without subscapularis repair (73% vs. 39%, p < 0.001). Increasing years of experience demonstrated a negative correlation with the duration of sling use after Bankart repair (r = −0.20, p < 0.001), Latarjet (r = −0.25, p < 0.001), ARCR < 3 cm (r = −0.14, p = 0.014) and >3 cm (r = −0.20, p < 0.002), and aTSA (r = −0.37, p < 0.001), and rTSA with subscapularis repair (r = −0.10, p = 0.049) and without subscapularis repair (r = −0.19, p = 0.022. Thus, the more experienced surgeons tended to recommend shorter durations of post-operative sling use. US surgeons reported longer post-operative sling durations for Bankart repair (4.8 vs. 4.1 weeks, p < 0.001), Latarjet (4.6 vs. 3.6 weeks, p < 0.001), ARCR < 3 cm (5.2 vs. 4.5 weeks p < 0.001) and >3 cm (5.9 vs. 5.1 weeks, p < 0.001), aTSA (4.9 vs. 4.3 weeks, p < 0.001), rTSR without subscapularis repair (4.0 vs. 3.6 weeks, p = 0.031), and isolated BT (3.7 vs. 3.3 weeks, p = 0.012) than Europe respondents. No significant differences between regions within the US and Europe were demonstrated. Conclusions: There is considerable variation in the immobilization advocated by surgeons, with geographic location and years of clinical experience influencing patterns of sling use. Future work is required to establish the most clinically beneficial protocols for immobilization following shoulder surgery. Level of Evidence: Level IV.

A wide spectrum of postoperative rehabilitation protocols has been described for patients following open and arthroscopic procedures [18][19][20]. Surgeons may potentially improve outcomes after rotator cuff repair by controlling and optimizing the mechanical environment following surgery, with a period of immobilization being a widely adopted strategy [21][22][23][24][25]. Immobilization protects the shoulder from excessive forces that may damage the tissues or repair constructs and lead to early failure [26]. However, this must be balanced with the increased risk of postoperative shoulder stiffness and decreased shoulder function [26]. At present, the optimum duration of immobilization and its basic utility for a range of shoulder procedures remain unproven [27], with differing data and recommendations [21][22][23][24][25].
There is a wide range of shoulder immobilization products and protocols used by surgeons following shoulder surgery [25]. However, patterns of sling use among surgeons are not known and, furthermore, there is no consensus on the optimum position and duration of shoulder immobilization following a range of surgical procedures. The aim of this study was to establish patterns and types of sling use for various surgical procedures in the US and Europe and to identify factors associated with the variations. The hypothesis was that the types and durations of postoperative immobilization would vary depending on years of experience, procedures, and surgeons' geographic locations.

Materials and Methods
An 18-question internet survey was developed and emailed to the membership of the American Shoulder and Elbow Society (ASES) and the European Society for Surgery of the Shoulder and Elbow (SECEC-ESSSE) in 20 April 2020 (Table 1). A reminder email was sent out on 6 May 2020. The data was downloaded and stored in a secure location at the Johns Hopkins University School of Medicine Department of Orthopaedic Surgery. Three of the 18 questions targeted demographic information, including country and region of practice and years of independent clinical practice. The type of immobilization was determined, including simple sling, shoulder immobilizer, abduction pillow sling, neutral rotation sling, abduction pillow sling in neutral rotation, and abduction pillow (no sling) ( Figure 1). The remaining questions related to the technique and duration of immobilization following a variety of shoulder surgeries, namely arthroscopic Bankart repair, Latarjet procedure, arthroscopic superior/posterosuperior rotator cuff repair (ARCR) of tears <3 cm and >3 cm, anatomic total shoulder arthroplasty (aTSA) and reverse TSA (rTSA), and isolated biceps tenodesis.
Since this study did not rely on patient data but solely on doctors' answers to an online survey, an a priori approval by an ethical committee or written informed consent was not required. However, the participants (doctors) agreed by answering the survey to use their answers for research purposes.

Statistical Analysis
The responses were collected and tabulated using Microsoft Excel software version 16.69.1 (Microsoft Corporation, Redmond, WA, USA). Relationships between surgeon location and sling type were analyzed using Fisher's exact tests and post-hoc tests using Bonferroni-adjusted p-values. Relationships between experience and sling duration were analyzed using Spearman's correlation tests. The analysis was performed in RStudio (version 2022.12.0+353) (Posit Software, Boston, MA, USA) using a two-sided level of significance of 0.05.  Since this study did not rely on patient data but solely on doctors' answers to an online survey, an a priori approval by an ethical committee or written informed consent was not required. However, the participants (doctors) agreed by answering the survey to use their answers for research purposes.

Statistical Analysis
The responses were collected and tabulated using Microsoft Excel software version 16.69.1 (Microsoft Corporation, Redmond, WA, USA). Relationships between surgeon location and sling type were analyzed using Fisher's exact tests and post-hoc tests using Bonferroni-adjusted p-values. Relationships between experience and sling duration were analyzed using Spearman's correlation tests. The analysis was performed in RStudio (version 2022.12.0+353) (Posit Software, Boston, MA, USA) using a two-sided level of significance of 0.05.

Results
In total, 499 surveys were returned and completed. Some respondents did not answer some questions, and to maintain consistency, 499 was used as the denominator for all percentage calculations for the responses.

Results
In total, 499 surveys were returned and completed. Some respondents did not answer some questions, and to maintain consistency, 499 was used as the denominator for all percentage calculations for the responses.

Arthroscopic Bankart Repair (ABR)
The variation in the use of sling based on surgeon location is outlined in Table 3. Surgeons based in the US reported higher abduction pillow sling use than Europeans (62% vs. 15%, p < 0.0001), while Europeans reported higher simple sling use than Americans (74% vs. 31%, p < 0.001) ( Table 3). For each additional decade of experience, the average sling duration in weeks decreased by 0.2 weeks (Figure 2A). On average, respondents from the US reported one week longer sling durations than Europeans (mean 4.8 weeks vs. 4 weeks, p < 0.001). The interaction between location and experience was significant, indicating that the relationship between experience and sling duration was significantly different for US and European surgeons. Experience did not influence sling duration for European surgeons, but it did for US surgeons ( Figure 2B). There were no significant differences between regions within the US and Europe. Western 118 (52%)

Arthroscopic Bankart Repair (ABR)
The variation in the use of sling based on surgeon location is outlined in Table 3. Surgeons based in the US reported higher abduction pillow sling use than Europeans (62% vs. 15%, p < 0.0001), while Europeans reported higher simple sling use than Americans (74% vs. 31%, p < 0.001) ( Table 3). For each additional decade of experience, the average sling duration in weeks decreased by 0.2 weeks (Figure 2A). On average, respondents from the US reported one week longer sling durations than Europeans (mean 4.8 weeks vs. 4 weeks, p < 0.001). The interaction between location and experience was significant, indicating that the relationship between experience and sling duration was significantly different for US and European surgeons. Experience did not influence sling duration for European surgeons, but it did for US surgeons ( Figure 2B). There were no significant differences between regions within the US and Europe.

Latarjet
The variation in use of sling based on surgeon location is outlined in Table 4. Surgeons based in the US reported higher abduction pillow sling use than Europeans (53% vs. 12%, p < 0.0001), while Europeans reported higher simple sling use than Americans (78% vs. 44%, p < 0.001) ( Table 4). For each additional decade of experience, the average sling duration in weeks decreased by 0.25 weeks ( Figure 3A). On average, respondents from the US reported 1 week longer sling durations than Europeans (mean 4.6 vs. 3.6 weeks, p < 0.001). The relationship between experience and sling duration was significantly different for US and European surgeons: experience influenced sling duration more for US surgeons than it did for European surgeons ( Figure 3B). There were no significant differences between regions within the US and Europe. (78% vs. 44%, p < 0.001) ( Table 4). For each additional decade of experience, the aver sling duration in weeks decreased by 0.25 weeks ( Figure 3A). On average, responde from the US reported 1 week longer sling durations than Europeans (mean 4.6 vs. weeks, p < 0.001). The relationship between experience and sling duration was sign cantly different for US and European surgeons: experience influenced sling duration m for US surgeons than it did for European surgeons ( Figure 3B). There were no signific differences between regions within the US and Europe.

ARCR (Tears < 3 cm)
The variation in use of sling based on surgeon location is outlined in Table 5. S geons based in the US reported higher abduction pillow sling use than Europeans (8 vs. 42%, p < 0.0001), while Europeans reported higher simple sling use than Americ (50% vs. 17%, p < 0.001) ( Table 5). For each additional decade of experience, the aver sling duration in weeks decreased by 0.14 weeks ( Figure 4A). On average, responde from the US reported 1 week longer sling durations than Europeans (mean 5.2 weeks 4.5 weeks, p < 0.001). There were no significant differences between regions within the and Europe. The interaction between location and experience was significant, indicat

ARCR (Tears < 3 cm)
The variation in use of sling based on surgeon location is outlined in Table 5. Surgeons based in the US reported higher abduction pillow sling use than Europeans (80% vs. 42%, p < 0.0001), while Europeans reported higher simple sling use than Americans (50% vs. 17%, p < 0.001) ( Table 5). For each additional decade of experience, the average sling duration in weeks decreased by 0.14 weeks ( Figure 4A). On average, respondents from the US reported 1 week longer sling durations than Europeans (mean 5.2 weeks vs. 4.5 weeks, p < 0.001). There were no significant differences between regions within the US and Europe. The interaction between location and experience was significant, indicating that the relationship between experience and sling duration was significantly different for US and European surgeons. Experience did not influence sling duration for European surgeons, but it did for US surgeons ( Figure 4B).

ARCR Cuff (Tears > 3 cm)
The variation in use of sling based on surgeon location is outlined in Table 6. Surgeons based in the US reported higher abduction pillow sling use than Europeans (84% vs. 61%, p < 0.0001), while Europeans reported higher simple sling use than Americans (34% vs. 13%, p < 0.001) ( Table 6). For each additional decade of experience, the average sling duration in weeks decreased by 0.2 weeks ( Figure 5A); however, experience did not influence sling duration for US and European surgeons when considered separately (Figure 5B). On average, respondents from the US reported 1 week longer sling durations than Europeans (mean 5.9 vs. 5.1 weeks, p < 0.001). Experience influences sling duration more for US-based surgeons than for Europeans ( Figure 5B). There were no significant differences between regions within the US and Europe.

ARCR Cuff (Tears > 3 cm)
The variation in use of sling based on surgeon location is outlined in Table 6. Surgeons based in the US reported higher abduction pillow sling use than Europeans (84% vs. 61%, p < 0.0001), while Europeans reported higher simple sling use than Americans (34% vs. 13%, p < 0.001) ( Table 6). For each additional decade of experience, the average sling duration in weeks decreased by 0.2 weeks ( Figure 5A); however, experience did not influence sling duration for US and European surgeons when considered separately ( Figure 5B). On average, respondents from the US reported 1 week longer sling durations than Europeans (mean 5.9 vs. 5.1 weeks, p < 0.001). Experience influences sling duration more for US-based surgeons than for Europeans ( Figure 5B). There were no significant differences between regions within the US and Europe.

aTSA
The variation in use of sling based on surgeon location is outlined in Table 7. Surgeons based in the US reported higher abduction pillow sling use than Europeans (50% vs. 21%, p < 0.0001), while Europeans reported higher simple sling use than Americans

aTSA
The variation in use of sling based on surgeon location is outlined in Table 7. Surgeons based in the US reported higher abduction pillow sling use than Europeans (50% vs. 21%, p < 0.0001), while Europeans reported higher simple sling use than Americans (69% vs. 41%, p < 0.001) ( Table 6). On average, respondents from the US reported 1 week longer sling durations than Europeans (mean 4.9 weeks vs. 4.3 weeks, p < 0.001). For each additional decade of experience, the average sling duration in weeks decreased by 0.4 weeks ( Figure 6). However, the relationship between experience and sling duration was not significantly different between European and US-based surgeons. There were no significant differences between regions within the US and Europe.

rTSA with Subscapularis Repair
The variation in use of sling based on surgeon location is outlined in Table 8. Surgeons based in the US reported higher abduction pillow sling use than Europeans (61% vs. 22%, p < 0.001), while Europeans reported higher simple sling use than Americans (69% vs. 35%, p < 0.001) ( Table 8). There was no difference in sling durations between US and European respondents (mean 4.4 vs. 4.2 weeks, p = 0.414) and there were no significant differences between regions within the US and Europe. For each additional decade of experience, the average sling duration in weeks decreased by 0.1 weeks (p = 0.049) (Figure 7A). Experience did not influence sling duration differently for US and European surgeons ( Figure 7B).

rTSA with Subscapularis Repair
The variation in use of sling based on surgeon location is outlined in Table 8. Surgeons based in the US reported higher abduction pillow sling use than Europeans (61% vs. 22%, p < 0.001), while Europeans reported higher simple sling use than Americans (69% vs. 35%, p < 0.001) ( Table 8). There was no difference in sling durations between US and European respondents (mean 4.4 vs. 4.2 weeks, p = 0.414) and there were no significant differences between regions within the US and Europe. For each additional decade of experience, the average sling duration in weeks decreased by 0.1 weeks (p = 0.049) ( Figure 7A). Experience did not influence sling duration differently for US and European surgeons ( Figure 7B).

rTSA without Subscapularis Repair
Variation in use of sling based on surgeon location is outlined in Table 9. Surgeons based in the US reported higher abduction pillow sling use than Europeans (57% vs. 18%, p < 0.001), while Europeans reported higher simple sling use than Americans (73% vs. 39%, p < 0.001) ( Table 9). For each additional decade of experience, the average sling duration in weeks decreased by 0.19 weeks (p = 0.022) ( Figure 8A). Experience did not influence sling duration differently for US and European surgeons ( Figure 7B). On average, respondents from the US reported 1 week longer sling durations than Europeans (mean 4.0 vs. 3.6 weeks, p = 0.031). There were no significant differences between regions within the US and Europe.

rTSA without Subscapularis Repair
Variation in use of sling based on surgeon location is outlined in Table 9. Surgeons based in the US reported higher abduction pillow sling use than Europeans (57% vs. 18%, p < 0.001), while Europeans reported higher simple sling use than Americans (73% vs. 39%, p < 0.001) ( Table 9). For each additional decade of experience, the average sling duration in weeks decreased by 0.19 weeks (p = 0.022) ( Figure 8A). Experience did not influence sling duration differently for US and European surgeons ( Figure 7B). On average, respondents from the US reported 1 week longer sling durations than Europeans (mean 4.0 vs. 3.6 weeks, p = 0.031). There were no significant differences between regions within the US and Europe.

Isolated Biceps Tenodesis
The majority of respondents based in the US and Europe recommended simple sling immobilization following isolated biceps tenodesis procedures (77% and 82%, respectively, p = 0.899). However, respondents in the US reported higher abduction pillow sling use than Europeans (18% vs. 7%, p < 0.001) ( Table 10). Experience did not influence the duration of immobilization within the US or Europe. (Figure 9)There were no significant differences between regions within the US and Europe. US-based surgeons reported longer sling durations than their European counterparts. (3.7 vs. 3.3 weeks, p = 0.012). However, experience did not influence sling duration in either region.

Isolated Biceps Tenodesis
The majority of respondents based in the US and Europe recommended simple sling immobilization following isolated biceps tenodesis procedures (77% and 82%, respectively, p = 0.899). However, respondents in the US reported higher abduction pillow sling use than Europeans (18% vs. 7%, p < 0.001) ( Table 10). Experience did not influence the duration of immobilization within the US or Europe (Figure 9). There were no significant differences between regions within the US and Europe. US-based surgeons reported longer sling durations than their European counterparts. (3.7 vs. 3.3 weeks, p = 0.012). However, experience did not influence sling duration in either region.

Discussion
The most important finding of this study was the differences in preferred immobilization methods between US-based surgeons and their European counterparts. An abduction pillow sling was the preferred method of immobilization for the majority of US-based surgeons, following Bankart repair, Latarjet procedure, ARCR of tears < 3 cm and > 3 cm, aTSA, and rTSA. This is in contrast to European surgeons who preferred a simple sling for all procedures except for ARCR of > 3 cm in size.
Availability and cost play an important role in postoperative immobilization. For instance, some abduction braces are either not available in Europe, simply not reimbursed

Discussion
The most important finding of this study was the differences in preferred immobilization methods between US-based surgeons and their European counterparts. An abduction pillow sling was the preferred method of immobilization for the majority of US-based surgeons, following Bankart repair, Latarjet procedure, ARCR of tears <3 cm and >3 cm, aTSA, and rTSA. This is in contrast to European surgeons who preferred a simple sling for all procedures except for ARCR of >3 cm in size.
Availability and cost play an important role in postoperative immobilization. For instance, some abduction braces are either not available in Europe, simply not reimbursed by healthcare systems, or are sold at a prohibitive cost. Furthermore, regarding compliance with postoperative immobilization and abduction brace wearing, a publication revealed that about 50% of European patients actually do not adhere to the suggested immobilization regimen [28]. Moreover, some European publications advise the abolition of postoperative immobilization after Latarjet procedures, small to medium rotator cuff repairs, and rTSAs [23,29,30]. Consequently, surgeon selection of immobilization strategy may be influenced not only by the availability and cost of equipment and historical use of one method over another, but also by patient preference and compliance and recent publications, explaining differences found between the two continents.
There was also a considerable variation in sling durations recommended by respondents in both the US and Europe. Respondents recommended from 0-8 weeks of sling immobilization following Bankart repair, Latarjet procedure, ARCR of tears <3 cm and >3 cm, aTSA, and rTSA, and 0-6 weeks of immobilization following isolated biceps tenodesis. More experienced clinicians were more likely to recommend shorter periods of immobilization across all procedures surveyed, with the exception of isolated biceps tenodesis. For the majority of procedures, US-based surgeons reported longer sling durations than Europeans. This study highlights the considerable variation in practice among shoulder surgeons in postoperative immobilization. Currently, the optimum duration of immobilization and its basic utility for a range of shoulder procedures remains unproven [27] with differing data and recommendations [21][22][23][24][25]. Future work is required to establish the most clinically beneficial protocols for immobilization following shoulder surgery. This is challenging due to the range of immobilization strategies and durations currently being used.
This study has several strengths. This represented the largest study of shoulder surgeon specialists that the authors are aware of, with representation across all regions of the US and Europe. The surgeons who responded represent a comprehensive mix of experienced surgeons and surgeons just starting their practice. This study has several limitations, including the cross-sectional nature of the study, with responses based on surgeon opinion at a single point in time. As such, the study is subject to selection and recall bias. Additionally, although the number of responses was high, the response rate was low, which may introduce a potential source of bias pending the responses of those who did not participate, as well as those who chose to participate. Further, not all questions were answered by all respondents, likely due to survey fatigue, so a correction was needed in some cases to determine the proper percentages for the responses. In addition, using an online-only response system may have introduced selection bias by omitting potential responders who are less 'technologically adept'.

Conclusions
There is considerable variation in the immobilization advocated by surgeons with geographic location and years of clinical experience influencing patterns of sling use. Future work is required to establish the most clinically beneficial protocols for immobilization following various types of shoulder surgery. This is challenging due to the range of immobilization strategies and durations currently being used.

Data Availability Statement:
The following was the link sent out to all members of the American Shoulder and Elbow Society and the European Society for Surgery of the Shoulder and Elbow in April of 2020. https://docs.google.com/spreadsheets/d/1iCKrhYmPvbhTiYyXcAdd5-Obiw1E6 NUPVMXymYQKEAI/edit?usp=sharing (accessed on 21 January 2023). The initial survey was sent out on 20 April 2020 and a reminder email was sent out on 6 May 2020. The data was downloaded and stored in a secure location at the Johns Hopkins University School of Medicine Department of Orthopaedic Surgery.