Introduction

We all know that osteoarthritis (OA) is common in elderly individuals and it puts a large burden on their day-to-day life. In some people with end-stage glenohumeral OA, shoulder arthroplasty is opted for. Three different types of surgeries are being performed: hemiarthroplasty, anatomical total shoulder arthroplasty and reverse total shoulder arthroplasty, each having its own (dis)advantages and clinical indications. Of course, as in every pathology, there is a need for evidence-based rehabilitation, but this is further pushed by a steep increase in the number of these surgeries performed. This review therefore looks at the factors that lead to better outcomes after shoulder arthroplasty.

Methods

The aim of this research was to summarize the evidence on factors associated with better outcomes after shoulder arthroplasty. With more knowledge of the associations between (non-)modifiable factors after physiotherapy rehab, the aim is to increase the success rate of these shoulder arthroplasties.

The PICO was defined as:

• P: Shoulder arthroplasty
• I: Post-operative physiotherapy
• C: Therapy influencing factors
• O: Treatment outcome

Treatment outcomes of interest were shoulder functionality, pain, ROM, ADL activities, muscle strength; satisfaction and quality of life.

Results

The review included 14 studies of which 4 were RCTs, 1 was a non-randomized controlled study and 9 studies were observational (2 prospective, 4 retrospective, and 3 combination of cohort and case-control studies). Most of the studies were at high risk of bias (86%), and the other 2 studies were at moderate risk of bias.

Considering hemiarthroplasty,non-modifiable factors were soft tissue integrity of the rotator cuff and the type of implant. Preliminary evidence indicates that those with an intact rotator cuff prior to hemiarthroplasty had larger improvements in active flexion and abduction than those with a torn rotator cuff at 6 months. Patients with hemiarthroplasty surgery had less ROM in forward flexion and internal rotation than patients who got an anatomical total shoulder arthroplasty after 8.7 years. There was no difference in the outcome of external rotation. The type of implant was also associated with strength, patients receiving hemiarthroplasty were less strong after 8.7 years than patients after anatomic total shoulder arthroplasty.

Modifiable factors in patients after hemiarthroplasty were preoperative function and ROM and the use of telemedicine. Preliminary evidence indicates that those with lower preoperative function showed larger improvements in shoulder function. Similarly, those with less preoperative active external rotation ROM showed more improvement in shoulder function postoperatively. The use of telemedicine resulted in more improvements in function, external ROM, pain, and quality of life after 8 weeks.

Non-modifiable factors in patients after receiving anatomic total shoulder arthroplasty included gender, soft tissue integrity of the rotator cuff, healing osteotomy of the subscapularis, and implant type. Men had lower improvements in internal rotation ROM at 3 years after surgery. Soft tissue status of the rotator cuff influenced postoperative ROM. Similar to what was seen in the hemiarthroplasty group, larger improvements in active flexion and abduction were seen in those with an intact rotator cuff at the time of the anatomic total shoulder arthroplasty surgery. A healed osteotomy of the subscapularis resulted in greater improvements in shoulder function after 1 year. Patients with anatomic total shoulder arthroplasty had better function in internal rotation-based activities at 3 years and forward flexion at 8.7 years. All these factors were supported by preliminary evidence.

Modifiable factors after an anatomic total shoulder arthroplasty included BMI, preoperative ROM, immediate ROM exercises, and the position of the sling.

• Preliminary evidence: A lower BMI resulted in better active and passive external rotation, internal rotation, flexion, and abduction after 3 years with an anatomic shoulder replacement. Those with better preoperative ROM had less active ROM in flexion and abduction after 6 months. The initiation of ROM exercises immediately after surgery resulted in larger improvements in shoulder function at 4 and 8 weeks and a larger pain reduction at 4 weeks.
• Moderate evidence: The positioning of the sling in a neutral resulted in less night pain at 2 weeks, greater pain relief with passive adduction after 1 year, more improvements in active and passive external rotation and passive adduction ROM after 1 year

Non-modifiable factors after a reverse total shoulder replacement include gender and subscapularis repair. For gender, preliminary evidence found that men had lower improvements in internal ROM at 3 years postoperative. Conflicting evidence however showed that those with subscapularis repair had higher internal ROM and better improvements in internal ROM at 3 years postoperative.

Modifiable factors after a reverse total shoulder replacement were BMI, immediate ROM exercises, and accelerated rehab. A lower BMI resulted in better internal rotation and abduction ROM at 3 years. Performing immediate ROM exercises resulted in larger improvements in shoulder function at 6 months. Accelerating the rehab by reducing the immobilization period resulted in better abduction external ROM at 1 year postoperative. These factors were supported by preliminary evidence.

Questions and thoughts

You may wonder if these non-modifiable factors are useful to consider in your rehabilitation. I think they are very useful, in fact. They may give you an idea of what to expect during the rehab of your patient. For example, when you know that the rotator cuff was intact prior to the arthroplasty surgery, you may expect better outcomes in movements of the shoulder in flexion, abduction, and external rotation. This may not only guide you in your prognostication but may also help you in explaining why someone might need a little more time to improve in case his rotator cuff wasn’t intact prior to their shoulder replacement surgery.

Of course, modifiable factors are factors we do have a spell on. Knowing these factors allows us to adjust and adapt them accordingly. BMI, however, is a factor we cannot change in a couple of days. But in spite of this, you can use this to educate your patient about changing their lifestyle towards a more healthy one as this may have an impact on their treatment outcome. It should be noted, however, that only the positioning of the sling was supported by moderate evidence. The factors supported by preliminary evidence should be tested further.

Talk nerdy to me

Not many concerns arise when looking at the methodological aspects. Even more, a very strict quality assessment was conducted. A study received a low risk of bias label when all domains were of low risk of bias. On the other hand, the presence of only 1 domain at moderate risk led to an overall moderate risk of bias and consequently, the presence of just one domain with a high bias score was enough to rate the whole study as a high risk of bias study. This resulted in an overall high risk of bias in the study. Consequently, the level of evidence for this study is rather low.

A potential limitation of this study is that only a few studies had the primary aim of investigating the associations between the factors and better outcomes after shoulder arthroplasty. Frequently, they were added as a subanalysis. Another remark may be the use of the same search strategy for all databases. Normally the search string should be appropriately adapted based on the type of database. The date limitation for the inclusion of studies after January 2000 is justified because it was to limit the use of outdated implant types.

Take home messages

Better outcomes after shoulder arthroplasty may be expected when considering following the modifiable factors: BMI, preoperative ROM, and immediate ROM exercises, accelerated rehab by reducing the immobilization period, preoperative function and ROM, and the position of the sling.

Modifiable factors that were associated with better treatment outcomes include:

• BMI: A lower BMI was associated with better outcomes after anatomic or reversed shoulder replacement
• Preoperative ROM and immediate ROM exercises were associated with better outcomes in those after hemiarthroplasty or an anatomic total shoulder replacement. A greater preoperative ROM resulted in less active ROM towards flexion and abduction postoperatively. Those performing immediate ROM exercises had larger improvements in shoulder function and a greater decrease in pain.
• Accelerated rehab: In patients undergoing reverse total shoulder arthroplasty, a shorter duration of immobilization (0 to 3 weeks) resulted in better function and ROM towards abduction and elevation compared with 6 weeks of immobilization.
• Preoperative function and ROM were found to be associated with larger improvements in postoperative outcomes. Lower preoperative shoulder function and lower preoperative external rotation ROM showed larger improvements in postoperative function, while more preoperative ROM led to less ROM to flexion and abduction postoperatively in those receiving hemiarthroplasty.
• The position of the sling was also associated. In those with an anatomic shoulder replacement, less night pain was seen when the sling was worn in a neutral position.

Reference

Claes A, Mertens MG, Verborgt O, Baert I, Struyf F. Factors associated with better treatment outcome of physical therapy interventions after shoulder arthroplasty: A systematic review. Clin Rehabil. 2022 Oct;36(10):1369-1399. doi: 10.1177/02692155221106627. Epub 2022 Jun 13. PMID: 35698750. 

Learn more

Listen: https://www.physiotutors.com/podcasts/episode-034-orthopaedics-and-physio/