You’ve been told you have a rotator cuff tear, and you are scared that your shoulder is going to hurt forever. You’re afraid you’ve done damage that you’re never going to come back from. Maybe you’re wondering if there’s a way you can avoid getting surgery?
Let’s look at the ins and outs of rotator cuff tears so you understand what you’re dealing with.
1) What are rotator cuff tears?
2) Do rotator cuff tears cause pain?
3) Does surgery for rotator cuff tears get you back to doing things you love?
Along the way, we’ll address some other related questions as well, and figure out the best way for YOU to help yourself.
Before you can make a well-informed decision about how you want to manage your shoulder, you need to know what the rotator cuff is and what a tear means.
Unfortunately, shoulder pain is a common problem. The shoulder joint has many directions of movements and is involved in many daily functional activities, making it more prone to injury. The rotator cuff itself is made of four muscle tendons that help give your shoulder stability in spite of its inherent instability. The supraspinatus that abducts your arm, the infraspinatus and teres minor that externally rotate your arm, and the subscapularis that internally rotates your arm. Partial or full tears, to one or more of these muscles, can occur due to trauma, degenerative changes, overuse, anatomical changes causing impingement, or glenohumeral dislocation.
Surgery is not for everyone who has a rotator cuff tear, and it's important to thoroughly examine the literature says about treatment options. Considering it is a complex injury with a prolonged course, we first need to examine a little more background information.
A systematic review looked at the use of MRI, MRA, and ultrasound in diagnosing rotator cuff tears, including a total of 20 studies and 1147 shoulders (Lenza 2013). They conclude that any of these methods could have good diagnostic accuracy for full thickness tears. However, evidence is limited in the reliability of detecting partial thickness tears by both ultrasound and MRI, with ultrasound being less sensitive than MRI.
Some smaller studies looked at these diagnostic tests separately. A study that looked specifically at MRI (Bhatnagar et al 2016) examined MRI results versus what the shoulder joint looked at during arthroscopic shoulder surgery. Of 39 patients, a high sensitivity and specificity was found for the diagnosis of rotator cuff tears using MRI.
Now let’s get more specific into what the tears may mean for your symptoms and for your clinical presentation.
This is where things get tricky. While it is believed that rotator cuff tears are definitely the cause of pain and disability, this relationship is not as clear as previously believed. People may have rotator cuff tears and a lot of pain and disability. But a rotator cuff tear, even if severe, may not result in any pain at all.
A study by Dunn et al (2014) looked the relationship between the severity of the rotator cuff tear and the symptoms. They looked at 393 subjects who had symptomatic (meaning they had pain), full-thickness rotator cuff tears that were nontraumatic. Interestingly, the severity of the rotator cuff injury did not correlate with actual pain.
And, believe it or not, some patients have no symptoms, despite a considerable tear. According to Yamamoto et al in 2011, it is unknown why certain symptoms develop in rotator cuff tears, so their study attempted to determine what kind of patients don’t have symptoms. 211 people were examined who had a full-thickness tear as determined by ultrasound, and an impressive 65.4% had NO symptoms at their involved shoulder. The patients who did have symptoms, tended to have the tears in their dominant arm, weakness in external rotation, and a positive impingement sign. Asymptomatic patients tended to have tears in their nondominant arm, higher active flexion, intact strength of abduction and external rotation, and a negative impingement sign.
What we’re seeing here is a lack of connection between tears and actual symptoms. Rotator cuff tears may be blamed for the weakness in external rotation and may be blamed for the positive impingement sign, but rotator cuff tears do not actually seem to have a causal relationship to those symptoms.
According to the American Academy of Orthopaedic Surgeons (AAOS), rotator cuff repair is recommended for “pain that does not improve with nonsurgical methods.” Additionally, they state that surgery can be considered if you have one of the following:
They do, however, acknowledge the worse chance for a successful surgery for patients who have any of the following:
Let’s see what the research supports.
One study looked at psychosocial factors and shoulder exam to see if either of these could predict your outcome. Woollard et al (2016) followed-up 46 patients who were scheduled for surgery. At a 6-month follow-up, those who had already needed job modifications or had a worker’s compensation claim had a poorer outcome.
They also looked at those who have fear-avoidance beliefs using a questionnaire, which is an attempt to quantify the impact of your thoughts and beliefs about pain which can translate to your perceptions of relief. In this study, those who had a higher fear-avoidance behavior had a lesser chance of improving with surgery. Those who had a surgery on their dominant shoulder had a better chance for success with surgery.
An attempt to determine what factors determe a positive outcome was completed with a systematic review by Lambers et al (2014), with 12 studies. Moderate evidence was found to support that surgery is less successful for those who are older, have a larger tear, have had or need additional procedures, and those who have worker’s compensation claims.
Considering a large cause of rotator cuff tears is degenerative changes it is important to consider the results of this specific group. A study by Carr et al (2015) acknowledged the uncertainty in their treatment, so they followed up on 273 patients over age 50 following surgery at 8, 12, and 24 months’ post operation.
At eight months, 77% of participants reported shoulder problems ‘much or slightly better’ and by 24 months, this increased to 85%. Using the Oxford Shoulder Score and an MRI to assess for integrity of the repair, those who had tears that were impossible to repair had the poorest outcomes, followed by those who had re-tears. We will talk more about re-tears in a little bit, but this does support those who have ‘impossible to repair’ outcomes and those who have re-tears will not have good outcomes, especially in the older adult category.
Another group that is prone to shoulder pain or injury is those who have jobs that require heavy lifting or repetitive use of the arms and shoulders. A retrospective study by Namdari (2014) examined what contributed to failures in their patients, and as you can see, those workers were less likely to have a positive outcome. Patients were considered ‘successful’ if they had a score of 80 points or greater on the American Shoulder and Elbow Surgeons (ASES) score, and unsuccessful if they had less than 80 points. Those placed in the unsuccessful group more likely had labor-intensive occupation, a worse preoperative scores, and worse preoperative external rotation. Of those who had a failed rotator cuff repair, only 54% had successful outcomes. For those who have a labor-intensive occupation, they will definitely be an even higher risk of a poor outcome and risk of rotator cuff repair failure.
So, in conclusion from the research we completed, the following will fall into the less successful surgical outcomes category:
Several studies examine patient satisfaction following rotator cuff repair.
In an attempt to determine the what contributes to patient satisfaction, Tashijan et al (2007) assessed 112 patients at an average of 54 months after surgery. An impressive 95% of patients were satisfied, and this satisfaction correlated with general health status and improvement in function. Patients who were married, currently working and nondisabled were more likely to be satisfied. Their recommendation for improving patient satisfaction is to concentrate on improving functional outcome through rehabilitation. Interesting that function through rehabilitation is mostly what impacts your satisfaction, right?
Another group found patients who had less satisfaction were patients who had pain with their daily activities or who had difficulty using the bathroom (Chen 2006).
Even another group looked at the specifics of tears in relation to satisfaction. A year after surgery, of 254 patients, those who were less satisfied were patients who had massive or irreparable cuff tears, subscapularis tears, and larger supraspinatus or infraspinatus tears. In terms of pain and movement, decreased flexion of the shoulder, impingement signs, and acromioclavicular joint pain and tenderness were associated with decreased satisfaction after surgery. Those who continued to have pain, difficulty with function, or disability in regards to work, had decreased satisfaction with their surgery, as you might expect (O’Holleran 2005).
Patient satisfaction depends on restoring shoulder function (i.e. getting back to your life without pain. Let’s see what the research says about what how many patients are getting back to this after surgery.
As with most surgeries, your results may vary wildly.
Gore et al (1986) examined the range of motion and strength as well as the ability to perform daily activities in 58 patients who had their full-thickness rotator cuff tears repaired. Range of motion was about 75% of normal and strength of abduction was 86% of normal. While most reported that they had considerable pain relief and reported mild to no deficits, the patients who had longer tears had poorer functional outcomes.
Limited evidence exists regarding early rehabilitation and the influence of tear size. Harris et al (2013) looked at 274 patients and grouped them according to tear size: 58% had small tears, 25% had medium and 17% had large. Shoulder range of motion was measured preoperatively and at 2 and 6 weeks, 3 and 6 months, and 1 year postoperatively. Regardless of the time before or after surgery, large tears were stiffer than small tears in external rotation and forward elevation. The study found it takes 1 year to fully regain external rotation after small and medium tears, whereas mild residual stiffness remains after large tears. Full forward elevation is restored by 3 months for small tears, but takes 6 months for medium and large tears.
While surgery is usually a quick procedure, it is crucial to consider how many months of recovery and therapy you will need before you return to using that shoulder in a functional way. A retrospective evaluation of 201 patients looked at recovery periods for function following surgery (Manaka et al 2010). According to the results 31% of patients took less than 3 months, 40% took between 3 and 6 months, and 28% took greater than 6 months to achieve a score greater than 80% in their functional measurements. Take note that number is 80%, not 100%, and only 72% obtained functional recovery within 6 months. Those who did better were either younger with less stiffness, or those with smaller tears.
Many occupations involve some movement of the shoulders, whether it’s a desk job or a job that involves lifting, pushing, pulling, or driving. A rotator cuff repair can limit your ability to do your job or drive, and often this may still be limited even after a surgical intervention.
The Gore study from 1986 found that 79% of patients, from their small sample of 19, returned to work, but points out that they often returned to a different type of work due to the shoulder problems.
Nové-Josserand et al (2011) examined 254 patients with work-related injury or disease resulting in surgery and assessed to see if return to work was possible for this specific group of patients. Sadly, return to work occurred in only 59.5% of the cases. The inability to return to work was due to retirement, starting at age 55 (14%), an unrelated medical condition (10%), but most likely the return to work was impacted by the outcome on the operated shoulder (16%). While the type of work did not affect the ability to return, it did increase the time that it took to return to work.
The return to driving after surgery varied vastly (Gholson et al 2015), from the same day to four months following surgery, and two out of 54 patients were not driving by that time. The reasons for not driving was due to continued narcotic use in 12% of patients and due to continued sling use in 33% of patients. Some who did return to driving reported feeling unsafe or had difficulty with maneuvering due to pain and weakness.
For many pro and recreational athletes, this is the most important question to ask. You want a solution to shoulder pain that can get you performing at a high level again.
A literature review in Reuter et al in 2016 found 12 studies and 314 athletes to review the rate of return to sport after surgery. The average time to return to sports was 8.3 months after surgery. When reviewing all athletes, 79% of the patients returned to sports, but from this group only 60.5% were able to return to their pre-injury level. When looking at professional athletes in overhead sports the numbers drop considerably. Only 48% were able to return to their pre-injury level, and another 20.3% returned, but at a lower level.
Another systematic review by Klouche et al (2015) also reviewed the return to sport, including 25 studies and 859 patients and assessing if athletes not only return to playing their sport but at what level. The majority of patients were baseball and tennis plays with a smaller group represented by golfers. The rate of return to sport was 84.7%, finding 65.9% were able to play at a similar level of play, after a range of 4 to 17 months. Of the professional and competitive athletes, only 49.9% returned to the same level of play.
Both of these articles support that your chances of returning to support are fair if you are a recreational athlete, but significantly smaller if you are a higher level or competitive athlete, and that return to sport could happen anywhere from 4 to 17 months. Depending on what sport(s) you play, this could be a season or two out, and when you do return, it does not guarantee a return to the same level of play.
Rotator cuff surgery is generally thought to have minimal complication rate, but potential complications can occur either during or after a rotator cuff surgery.
Let’s take a look at these, based on their symptoms and clinical presentation. As Parada et al outlines (2015), complications can be divided between complications that are treated nonoperatively and those that require additional operative treatment.
Management of complications after rotator cuff surgery. Parada et al 2015
While it is concerning to think about, a good chance of failure exists after rotator cuff surgery due to the potential of a re-tear of the muscle or muscles that were repaired.
According to Le et al in 2014, the rate of re-tear varies widely from 11% to 94%. So, what factors can increase your chance of a re-tear?
They retrospectively looked at 1000 patients who had undergone a primary surgery and received an ultrasound six months later to examine the integrity of the surgical repair. From this group, 17% of patients had a re-tear within six months. Of full-thickness tears, 27% of patients had a chance of re-tears, so therefore concluded a larger tear could increase your likelihood of re-tear.
And, while it seems common sense, if you have a re-tear you will likely have worse function. Kim et al (2014)looked at 180 surgical patients. 26% had a re-tear. The re-tear group, not surprisingly, performed significantly worse in functional assessments compared to those who had their repair intact. In the re-tear group, those who were younger, had a lower education level, and had worker’s compensation claims had even poorer outcomes.
So know you know the possibility of failure exists, and could possibly lead to a second surgery to try to repair this same tear again. Does that second surgery guarantee a good outcome? DeOrio and Cofield in 1984 looked at results of 27 patients. It described the initial failures as a result of massive or large tendon tears, damage to the deltoid during the original surgery, and potentially inadequate postoperative support. An additional 7 patients underwent at THIRD surgery, due to continued pain or weakness. Patients were examined for about 4 years after the second repair and found 63% still had moderate or severe pain, but 76% did report an improvement from before surgery. In regards to abduction, patients, on average increased active abduction by only 8 degrees and only 7 patients had greater than 30 degrees gain, and 19 had moderate or marked weakness. In terms of overall success of the surgery, 17% had good results, 25% had fair results, and 58% had poor results, definitely not a resounding endorsement for a second surgery.
Another study examined this as well. Ma et al (2003) identified 20 patients who required a second surgery for continued pain and functional impairment. Similar reasons were provided for failure, including: large to massive tear initially (80%), continued subacromial impingement (75%), poor condition of the deltoid (35%), rupture of the biceps tendon (25%), poor quality of cuff tissue (25%), severe subacromial or intraarticular adhesions (20%).
As you can see, many patients had more than one cause for failure. The second surgery occurred, on average, 11.8 months after the initial surgery, and they completed follow-up at about 60.7 months. This follow-up showed 75% of patients had pain free or occasional aching soreness. They looked at function through active flexion, which increased from 80 degrees to 127 degrees post-op. 60% of patients had no functional impairment or were left with some restriction. Satisfactory results were noted in 55% of patients. This further supports that a second surgery will not likely improve function, but may decrease some of your pain. To expect a completely pain-free outcome is less likely.
Another study decided to follow-up after a longer period of time, 7.6 years, on how patients did that did not elect for the second surgery after their failed first surgery. Without an additional surgery, interestingly, the 20 re-ruptures had not increased in size, and reports of pain, function, and strength had not changed from the initial 3.2-year follow-up. Impressively, 8 out of 20 had healed on their own, without intervention (Jost 2006).
Imagine the frustration of going through an initial surgery and rehabilitation process, only to realize you have to go through the entire process again. And starting over does not guarantee you any better results than the first time, and in fact, it could be even worse. It could create doubt about going down the route of surgery in the first place.
The simple answer is conservative treatment should almost always be used as the initial treatment. Let’s get more into the details.
According to Thorpe et al (2016) rates of rotator cuff surgery are increasing despite a lack of an associated increase in the prevalence of shoulder pain. These trends were evaluated and did indeed show surgical rates (and associated healthcare costs) had increased from 2001 to 2013. Interestingly they stated that ‘in spite of evidence that surgical outcomes are no different to exercise interventions’, the surgical rates were rising. They recommend that conservative treatment should always be recommended as the initial treatment.
A systematic review by Seida et al (2010) attempted to compare nonoperative and operative treatments, and found 137 studies that met their criteria. Overall the limited evidence available was low in quality, and all trials had high risk for bias, so it was truly difficult to compare treatment methods. They say “Evidence on the comparative effectiveness and harms of various operative and nonoperative treatments for rotator cuff tears is limited and inconclusive.”
Another meta-analysis was completed to compare surgery versus conservative measures (Ryosa et al 2016). After analysis of three randomized control trials, they too found “limited evidence that surgery is not more effective in treating rotator cuff tear than conservative treatment alone” and reinforce what we’ve seen before: conservative approach should be the initial treatment method. They reiterate that conservative treatment is both less costly and less risky, and both studies imply that this should be tried first.
This can be easily seen by having a placebo group do nothing and comparing it to groups that receive treatment. Both groups in a study that received either surgery and exercise demonstrated higher success rate than placebo (Brox 1999). This shows clearly that exercise is most definitely better than doing nothing!
What about exercises at home? Krischak et al (2013) included 38 adults with full thickness tears and completed two months of conservative treatment, in one group with independent home-based exercise and in the second group, to occupational therapy. Two-thirds of the patients improved shoulder testing, regardless of the group. Ironically, the only significant difference was that patients with the home-based exercise improved more on the health-related quality of life measure.
Lombardi et al looked at patients with shoulder impingement syndrome, and placed half of the 60 patients in a progressive resistance training program, twice a week for two months, and the other group remained on a waiting list. Those in the training program improved in shoulder function and reported decreased pain at rest and with movement.
Well, 389 patients were placed in a structured physical therapy program to see what strengthening could do to improve pain and function (Harris 2011). Interestingly, improvement occurred in scapulothoracic kinematics, active abduction, and resisted abduction and flexion, which all helped improve patients’ pain and function without surgery.
And now we can examine the group of patients who receive rehabilitation following surgery.
One study looked at if early or delayed active movement or range of motion can affect healing after surgery, through a systematic review and meta-analysis. Kluczynski et al (2015) found 37 studies. Depending on the surgery, it appeared that for small and large rotator cuff tears could be at a higher risk for structural defect after surgery if the patient completes early active range of motion.
Another study be Lee et al in 2011 examined two different rehabilitation protocols after surgery. The first protocol included aggressive early passive rehabilitation and unlimited self-passive stretching exercise and the second group included limited early passive rehabilitation. The first group improved in range of motion more quickly immediately, but at 1-year follow-up no difference existed between groups, except for internal rotation when the shoulder was at 90 degrees of abduction. They also completed an MRI to assess the integrity of the repair on average of 7.6 months after surgery. While not statistically significant, the group that had the more aggressive early rehabilitation had re-tear in 23.3% of cases compared with 8.8% in the more conservative group. Regardless of the protocol, all improved in pain, range of motion, strength, and function.
The overarching theme is that properly structured exercise makes a difference to shoulder function. If you’re trying improve shoulder function, proper exercise is mandatory and can work whether or not you have surgery.
Of course choosing surgery or conservative therapy is a difficult decision that each person needs to make to choose the most effective method for them. But what if patients have a skewed perception of which one is more effective? A 2016 study (Dunn et al) found that the patient’s strongest predictor of choosing to undergo surgery is not because of pain or anatomic features of the rotator cuff, but it is because they have low expectations regarding the effectiveness of physical therapy. This definitely points to the importance of motivation toward your rehabilitation program, and is especially ironic in light of the underwhelming research on surgical results.
Rotator cuff tears are complicated. It’s unclear that rotator cuff tears actually cause the symptoms of pain and disability because so many people can have them with absolutely zero symptoms. The chances for complications or re-tears following surgery compel you to consider your other options.
Interestingly, the American Academy of Orthopaedic Surgeons reviewed what evidence was available in the research literature in a 2010 published report titled “Optimizing the Management of Rotator Cuff Problems, Guideline and Evidence Report”. Below, you’ll find a chart summarizing their results and recommendations.
Another review of the effectiveness of surgery was looked at by Coghlan et al in 2008, including 14 randomized controlled trials, and involving a total of 829 patients. Based on this thorough review, they were unable to fully support the effectiveness of surgery and the safety of surgery.
So both a large surgical organization and a high level literature review acknowledge the considerable lack of strong evidence for the treatment of rotator cuff repair.
That doesn’t mean your situation is hopeless, though. Quite the contrary. The takeaway is that it may actually be possible to make your shoulder feel better EVEN IF you have a rotator cuff and EVEN IF you don’t get surgery.
There is little risk in attempting to gradually improve your strength, coordination, and posture. At the very worst, if your shoulder does not improve, you may always elect a surgery in the future, and go into the surgery with a stronger shoulder and thus recover better. And if you succeed in retraining your shoulder pain by improving the control of muscles around the shoulder, you may be able to reduce your pain and discomfort. If you can do that with just exercise, which is doable, you can avoid the expensive and lengthy process of undergoing and recovering from surgery.
Lenza et al 2013: http://onlinelibrary.wiley.com/doi/10.1002/14651858.CD009020.pub2/full
Bhatnagar et al 2016: https://www.ncbi.nlm.nih.gov/pubmed/27042543
Kurz et al 2016: https://www.ncbi.nlm.nih.gov/pubmed/27424254
Dunn et al 2014: https://www.ncbi.nlm.nih.gov/pubmed/24875019
Yamamoto et al 2011: https://www.ncbi.nlm.nih.gov/pubmed/21454096
American Academy of Orthopaedic Surgeons: http://orthoinfo.aaos.org/topic.cfm?topic=a00406
Woollard et al 2016: https://www.ncbi.nlm.nih.gov/pubmed/27548366
Lambers et al 2014: https://www.ncbi.nlm.nih.gov/labs/articles/24725900/
Carr et al 2015: https://www.ncbi.nlm.nih.gov/pubmed/26463717
Tashijan et al 2007: https://www.ncbi.nlm.nih.gov/pubmed/17964816
Chen et al 2006: https://www.ncbi.nlm.nih.gov/pubmed/17097315
O’Holleran et al 2005: https://www.ncbi.nlm.nih.gov/pubmed/15634822
Gore et al 1986: https://www.ncbi.nlm.nih.gov/pubmed/3944164
Harris et al 2013: https://www.ncbi.nlm.nih.gov/pubmed/23380011
Manaka et al 2010: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3094637/
Nové-Josserand L et al 2011: https://www.ncbi.nlm.nih.gov/pubmed/21511555
Gholson et al 2015 https://www.ncbi.nlm.nih.gov/pubmed/25988695
Reuter et al 2016: https://www.ncbi.nlm.nih.gov/labs/articles/27763580/
Klouche et al 2015: https://www.ncbi.nlm.nih.gov/pubmed/26316611
Parada et al 2015: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4596187/
Le et al 2014 https://www.ncbi.nlm.nih.gov/pubmed/24748610
Kim et al 2014 https://www.ncbi.nlm.nih.gov/pubmed/24430409
DeOrio JK et al 1984: https://www.ncbi.nlm.nih.gov/pubmed/6707035
Ma et al. 2003: https://www.ncbi.nlm.nih.gov/pubmed/12716007
Jost et al 2006: https://www.ncbi.nlm.nih.gov/pubmed/16510810
Thorpe et al 2016: https://www.ncbi.nlm.nih.gov/pubmed/27490156
Seida et al 2010: https://www.ncbi.nlm.nih.gov/pubmed/20621893
Ryosa et al 2016: https://www.ncbi.nlm.nih.gov/pubmed/27385156
Brox et al 1999: https://www.ncbi.nlm.nih.gov/pubmed/10226960
Krischak et al 2013: https://www.ncbi.nlm.nih.gov/pubmed/23523073
Lombardi et al, 2008: https://www.ncbi.nlm.nih.gov/pubmed/18438933
Harris et al, 2011: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3632074/
Kluczynski et al 2015: https://www.ncbi.nlm.nih.gov/pubmed/25943112
Lee et al 2011: https://www.ncbi.nlm.nih.gov/pubmed/22014477
Dunn et al 2016: http://www.jshoulderelbow.org/article/S1058-2746(16)30119-7/abstract