Does immobilization position after arthroscopic rotator cuff repair impact work quality or comfort?
- First Online:
- Cite this article as:
- Gumina, S., Candela, V., Passaretti, D. et al. Musculoskelet Surg (2014) 98: 55. doi:10.1007/s12306-014-0327-y
- 97 Views
The purpose of this study was to quantify subjective discomfort and decrease in working performance in patients submitted to arthroscopic rotator cuff repair.
We enrolled 101 asymptomatic administrative employees (mean age 55). Subjects were asked to write a text using Microsoft Word and to make a table using Microsoft Excel, with and without shoulder braces which kept the right shoulder in neutral (brace A) and internal rotation (brace B). Total time needed to complete the tests and number of mistakes committed were annotated. Furthermore, a questionnaire to assess the subjective and interpersonal discomfort caused by the braces was compiled. Data were submitted to statistical analysis.
When any brace is worn, both times and mistakes are higher than those registered without brace (p < 0.02). Both times and mistakes are higher for brace B in comparison with brace A (p < 0.02). Subjects wearing brace B had a severe/very severe discomfort degree three times higher than that registered in subjects wearing brace A. Finally, 91 % of subjects preferred brace A to brace B.
If the choice of the brace is not supported by biological or clinical advantages, we recommend to use the brace that keeps the arm at the side, since it ensures better working performance and lower discomfort. It also received the greatest satisfaction of the respondents.
Level of evidence
KeywordsRotator cuff repairShoulder bracePost-operative work performancePost-operative subjective discomfortShoulder slingPost-operative shoulder immobilization
Rotator cuff tears are present in 20.7 % of the general population and prevalence increases with age [1, 3]. Ultrasound studies, performed on asymptomatic and shoulder pain subjects aged between 50 and 59 years, showed that the prevalence of rotator cuff tear is, respectively, 2 %  and 13 % . The percentage rose to 14 and 26 % in the group of subjects aged between 60 and 69 years.
Arthroscopic rotator cuff repair generally provides satisfactory clinical results, in terms of shoulder pain decrease and improvement of range of motion. After the operation, immobilization of the involved shoulder with a brace is recommended [4–7]. Length of immobilization ranges between 3 and 6 weeks, during which passive rehabilitation program may be started.
In our country, retirement age depends on several parameters (chronological age, years of contribution, type of work, etc.). Usually, administrative staff concludes work activity around 65 years of age; therefore, many patients with rotator cuff tear might be still involved in administrative tasks. Before surgery, many patients wish to be informed on the period of inactivity subsequent to tendon repair. Many of these (private workers or personnel involved in jobs of high responsibility, particularly), ask this question: “Doc, do you know if I can write with the computer wearing the shoulder sling and how much my writing skills will be slowed down?”
The purpose of this study was to verify the presence of the decrease in working performance and the subjective discomfort during work activities in the immediate post-operative period in patients submitted to arthroscopic rotator cuff repair employed in administrative activities.
From November 2011 to March 2012, we enrolled 101 right-handed subjects (23 males and 78 females). They were all aged between 50 and 65 years (average age 55); all of them were administrative employees of our university and particularly skilled with the use of writing and data collection programs (like Microsoft Word and Microsoft Excel—almost 5 h per day). None of the subjects suffered by diseases that could compromise their working ability.
Initially, the subjects were instructed about the test run. Each subject was asked to undergo two tests: (1) a written test (which had to be made without [first text] and with the braces in neutral [second text] and internal position [third text]) (first test) and (2) compilation of a questionnaire (second test). The first test consisted in three parts.
First part: to write a text (first text), using Microsoft Word, and make an Excel table. Both the written proofs were dictated from one of us. Another author wrote down all the writing mistakes, committed by the subject, and the total time needed to write text and table. No mistakes were corrected during the test to avoid to increase the running time.
The Word tests and the Excel tables were chosen by specialists in spoken and written language disorders who guaranteed on the same difficulty degree. A previous analysis conducted on 20 administrative right-handed volunteers (aged 48–65 years) showed that the times needed to write the three texts without the braces were almost superimposable. The three texts consisted of 300 words, easy to understand. In addition, to reduce the differences between the three parts, the texts were extrapolated from the same novel. Each table consisted of 50 boxes arranged in 5 columns and 10 lines. Each box contained two numbers and one letter.
Furthermore, the order in which the three parts of the test were performed was randomized to avoid the possible influence of deconcentration and fatigue on the results.
At the end of the second part, the subject was asked to indicate the preferred brace to work with.
We calculated the arithmetic mean of the time of writing and of the average number of mistakes in the first, second and third part of the first test. Moreover, to verify the existence of a statistically significant difference (defined as no overlap of data) between the average times and number of mistakes in any part of the first test, we set the p value of 0.02 (CI 98 %).
Finally, we used the Pearson coefficient to verify the statistical correlation between the writing times of the first, second and third part of the first test.
As like as the first one, we calculated the arithmetic mean of the scores of the answers of the questionnaire (second test). Then, we divided the patients into four groups according to their DI.
Finally, we set the p value at 0.02 to verify the existence of a statistically significant difference (defined as no overlap of data) between the average scores for each answer.
Time (s) word
No. of mistakes word
Time (s) excel
No. of mistakes excel
With brace A
With brace B
p < 0.02
Time for word text
Time for excel table
Without brace—brace A
Without brace—brace B
Brace B–brace A
The average discomfort index for subjects wearing brace A is between 31.7 and 38.6 (CI 98 %; p < 0.02), while those wearing brace B have an average DI between 45.6 and 55.7 (CI 98 %; p < 0.02).
Finally, 92 subjects (91 %) out of 101 preferred brace A to brace B to work.
Literature data indicate that after surgical rotator cuff repair, a period of 3–6 weeks of immobilization of the involved shoulder may facilitate tendon healing [8, 9]. Other studies have been performed to determine whether the position assumed by the upper limb, after a surgical procedure carried out to repair capsuloligamentous or tendon structures of the shoulder, may influence the healing process [10, 11]. Ours is the first study that evaluates the decrease in job performance on a computer and the interpersonal work environment discomfort consequent to the presence of a shoulder brace immediately after the surgical repair of rotator cuff tear.
Our data suggest that any shoulder brace obviously limits the ability to write on a computer. However, the brace in internal rotation leads to an increase in the execution times of computer work compared with those recorded for the same work performance done wearing the brace that keeps the arm at the side with an abduction of 15°. In addition, the brace in internal rotation leads subjects to commit a greater number of typos. Objective deterioration of work performance caused by the brace in internal rotation has to be summed to the greater discomfort (related to job performance and interpersonal relationships) that it causes in comparison with the other brace. Our results were not affected by deconcentration or fatigue because the three parts of the first test were randomly carried out. Moreover, this interpretation is corroborated by the fact that times needed to write the text with the brace at the side were constantly lower than those registered with the arm in internal rotation even when it was worn at the end of the test.
Therefore, if the choice of the brace is not conditioned by beliefs based on the greater effectiveness of a brace, on modality and time of the cuff healing, we recommend to use the one which keep the arm at the side. This brace not only ensure better working performance and lower discomfort, but it also received the greatest satisfaction of the respondents.
Our study has some limitations: one of them is only apparent and the others are actual. The first is that for this study, patients with rotator cuff tear were not employed. However, if we used real patients, they would have had difficulty in writing the text, even in the absence of the brace, because of shoulder pain due to the lesion and this would have affected data interpretation. The second is that post-operative pain was not considered and it might further compromise both work performance and discomfort degree. Finally, the last limitation is that our study does not consider the benefits that a brace might make, in comparison with the other, in terms of cuff healing and shoulder pain.
Conflict of interest
S. Gumina, V. Candela, D. Passaretti, L. Mariani, L. Orsina declare that they have no conflict of interest.
All procedures followed were in accordance with the ethical standards of the responsible committee on human experimentation (institutional and national) and with the Helsinki Declaration of 1975, as revised in 2000 (5). All patients provided written informed consent to enrolment in the study and to the inclusion in this article of information that could potentially lead to their identification.