If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Department of Orthopaedic Surgery and Traumatology, Shoulder, Elbow and Orthopaedic Sports Medicine, Inselspital, Bern University Hospital, University of Bern, Bern, Switzerland
Department of Orthopaedic Surgery and Traumatology, Shoulder, Elbow and Orthopaedic Sports Medicine, Inselspital, Bern University Hospital, University of Bern, Bern, Switzerland
Correspondence to Dr Eiji Itoi, Department of Orthopaedic Surgery, Tohoku University School of Medicine, 1-1 Seiryo-machi, Aoba-ku, Sendai 980-8574, Japan
The rotator cuff tendon undergoes degeneration and tear quite commonly in the elderly population and was rarely observed in patients under 50 years of age. In addition to ageing, smoking, diabetes mellitus and other comorbidities are known to be the risk factors of tendon tear. Bony anatomy of the acromion relative to the glenoid plays a role on the onset of degenerative cuff tears. The biological responses of chronic tendinopathy, degenerative tear and the following muscle atrophy and fatty infiltration are regulated by macrophages. Age, tear size and chronicity have direct influence on healing after rotator cuff repair. Systematic reviews have shown that there is better healing in larger tears with a double row repair compared with a single row repair, but no differences in clinical outcome. The tendon healing is characterised by a fibrovascular scar response rather than by regenerating normal tendon tissue. As a result, the material and structural properties are much weaker than the normal tendon-to-bone interface. With this knowledge, better repair techniques and repair methods are expected to be developed for better healing of the tendon.
The shoulder joint is the one with the greatest range of motion in our body. As a result, the rotator cuff muscles and tendons, the mover and stabiliser of the shoulder joint, are often times involved in various pathological conditions, such as rotator cuff tear, calcified tendinitis, frozen shoulder, labral tear, bicipital tendinitis and so on. In this article, we focus on the rotator cuff tear, the most common pathology of the shoulder joint. We would like to introduce how common this pathology is observed as well as risk factors related to rotator cuff tears in the first chapter of epidemiology. In the second chapter, we are discussing the pathology of rotator cuff tendons and muscles, especially contemplating the biological responses to tendinopathy and tendon degeneration as a potential result of skeletal morphological changes. In the last chapter, we highlight the healing process of tendon-to-bone junction after the tendon repair, which is far different from the normal structure of tendon-to-bone junction.
Epidemiology of rotator cuff tear
Prevalence of rotator cuff tear
Rotator cuff tear is one of the most common causes of shoulder pain and disability of shoulder function. Regarding the prevalence, classical studies using cadaveric specimens have reported it to range from 3% to 39% for full-thickness rotator cuff tear.
Since advanced radiological techniques enable to assess the condition of rotator cuff tendons non-invasively in live subjects, there have been several studies that demonstrate the epidemiology of the rotator cuff tear and its age-related, degenerative characteristics. Sher et al
investigated MRI obtained from 96 asymptomatic shoulders and demonstrated that a rotator cuff tear was present in 33 shoulders (34%); among these, 14 shoulders had the full-thickness rotator cuff tear and 19 shoulders had the partial tear. In addition, they showed the prevalence of full-thickness and partial-thickness tears increased significantly with subject's age (p<0.001 and 0.05, respectively).
For larger scale analysis to reveal the prevalence of full-thickness rotator cuff tears, several studies applied sonographic examination to detect the rotator cuff tears in asymptomatic shoulders. Tempelhof et al
assessed 411 asymptomatic shoulders using ultrasound. They found 23% of the patients who had a rotator cuff tear and an age-dependent increase of the prevalence; 13% in 50–59 years, 20% in 60–69 years, 31% in 70–79 years and 51% in >80 years. Moosmayer et al
also reported sonographic observation in 420 asymptomatic volunteers. They found overall 32 subjects (7.6%) with full-thickness rotator cuff tear. The prevalence was reported to 2.1% for 50–59 years, 5.7% for 60–69 years and 15% for 70–79 years. Additional two sonographic studies also showed the prevalence of full-thickness rotator cuff tears in both symptomatic and asymptomatic subjects.
reported 22.1% (147 out of 664 subjects) in one village had full-thickness rotator cuff tear. They found that the prevalence of the tear was 0% in those under 50 years, 11% in the 50–59 years, 15% in the 60–69 years, 27% in the 70–79 years and 37% in the 80–87 years. They also demonstrated that symptomatic tears accounted for 35% of all tears and asymptomatic tears for 65%. The prevalence of tear was significantly greater in male subjects than in female subjects for 50s (p<0.001) and 60s (p=0.01), but not for over 70 years. Yamamoto et al
Factors involved in the presence of symptoms associated with rotator cuff tears: a comparison of asymptomatic and symptomatic rotator cuff tears in the general population.
reported the rotator cuff tear were present in 20.7% (283 out of 1366 shoulders) of residents of a mountain village. In addition, they investigated the risk factors for the occurrence of rotator cuff tear using multivariable analysis and demonstrated a history of trauma (OR 2.46, 95% CI 1.33 to 4.53), dominant arm (OR 1.66, 95% CI 1.25 to 2.22), in addition to increased age (OR 1.08 (95% CI 1.07 to 1.10) as risk factors. Most reports agree in the point that degenerative rotator cuff tears are only observed above the age of 50 years.
observed the residents in a village for an average of 3.5 years. There were 464 residents without a full-thickness rotator cuff tear at initial examination by ultrasonography. In 3.5 years, 30 of them developed a new full-thickness tear. The incidence of rotator cuff tear is calculated to be 30/464/3.5 × 1000=18/1000 person-years. To our knowledge, this is the only study that has shown an incidence of rotator cuff tear in a general population.
Risk factors of rotator cuff tear
Daily smoking habit
Although the aetiology of rotator cuff disease is multifactorial, the daily consumption of favourite items as a cause of rotator cuff tear has been investigated. Among these, smoking tobacco has been focused to cause and/or develop the rotator cuff tear. Itoi et al
first reported a positive correlation between the size of rotator cuff tears and the smoking index of the patients. They indicated larger size of the tear in association with greater smoking index. Kane et al
investigated 72 cadaveric shoulders with focus of a history of smoking. They found macroscopic rotator cuff tears and microscopic rotator cuff degeneration were more frequently seen in shoulders with smoking history than in those without any smoking history. Baumgarten et al
also reported that smoking increased the incidence of rotator cuff tears in a dose-dependent and time-dependent manner from a cohort study. Carbone et al
analysed the correlation between smoking habit and size of rotator cuff tear in patients who underwent arthroscopic rotator cuff repair. They concluded that the total number of cigarettes smoked in life had a positive relationship with the size of rotator cuff tear. More recently, Jeong et al
investigated 486 asymptomatic subjects using ultrasonography. They revealed smoking history significantly affected the prevalence of the full-thickness rotator cuff tear; 9.4% of smokers had a rotator cuff tear, whereas 3.7% of non-smokers had a tear (p=0.002).
Diabetes mellitus
Regarding the comorbidity that might affect the development of the rotator cuff tear, Jeong et al
reported the prevalence of full-thickness rotator cuff tear significantly increases with the presence of diabetes mellitus. Among 486 asymptomatic subjects, full-thickness rotator cuff tear was present in 7.7% of those with diabetes mellitus in contrast to 4.5% of those without the history (p=0.0418).
Other factors
Moreover, several factors have been reported to be potentially associated with the occurrence of rotator cuff tears. A systematic review
demonstrated that individuals with an adverse lipid profile might have an increased risk to cause tendon pathologies including rotator cuff tear. The meta-analysis revealed higher serum levels of total cholesterol (TC), low-density lipoprotein cholesterol and triglycerides as well as lower levels of high-density lipoprotein cholesterol in those with the Achilles tendinopathy or rotator cuff tears.
reported excessive alcohol consumption as a risk factor of rotator cuff tears in both male with OR 3.0 (95% CI 1.5 to 6.0) and female with OR 3.6 (95% CI 1.7 to 7.8). There is a report showing a negative effect of non-steroidal anti-inflammatory drugs (NSAIDs) on tendon healing.
They investigated the effect of licofelone, an inhibitor of 5-lipoxygenase and cyclooxygenase (COX)-1 and COX-2 enzymes on the tendon–bone healing process in rat rotator cuff repair models and revealed increased fibrocartilage formation at the enthesis with biomechanical improvement. The effect of NSAIDs on rotator cuff tear is yet to be determined (table 1).
Factors involved in the presence of symptoms associated with rotator cuff tears: a comparison of asymptomatic and symptomatic rotator cuff tears in the general population.
Factors involved in the presence of symptoms associated with rotator cuff tears: a comparison of asymptomatic and symptomatic rotator cuff tears in the general population.
Factors involved in the presence of symptoms associated with rotator cuff tears: a comparison of asymptomatic and symptomatic rotator cuff tears in the general population.
Anatomical variances predisposing to rotator cuff tear
With the large humeral head sitting on the small glenoid, the shoulder joint has a unique anatomy requiring dynamic stabilisation of the glenohumeral joint. This dynamic stabilisation depends on a distinct interplay of the rotator cuff muscles.
When the anatomy of the joint varies from the ‘ideal’ anatomical shape, the well-orchestrated interplay between the muscles is disturbed causing an imbalance of the forces. This imbalance sets the muscles and tendon of the rotator cuff prone to degenerate through repetitive trauma and thereby leading to degenerative rotator cuff tear. Anatomical variances associated with rotator cuff tear were found in the acromial arch as well as in the vertical glenoid orientation. Nevertheless, it is important to note that these variances are mostly seen in patients with degenerative rotator cuff tear and not in traumatic rotator cuff tears.
The classification of the acromial morphology and the measurements that determine its relation to the scapular anatomy (ie, acromial slope and acromial angle) focus on the shape of the acromion and thus neglect the dimensions of the acromion and its relation to the glenoid and humeral head.
The influence of bony anatomy
The lateral extent of the acromion, the so-called acromion index (AI),
measures the distance from the glenoid cavity to the tip of the acromion relative to the distance from the glenoid cavity to the most lateral part of the humeral head. The AI and thereby the lateral extension of the acromion is increased in patients with rotator cuff tear and also with rotator cuff retears.
It is hypothesised that the increase of lateral extension causes a higher upward pulling force of the deltoid muscle during active abduction of the arm and thereby pressing the humeral head into the acromial arch causing the supraspinatus muscle to impinge.
Most of the subsequent studies investigating AI with rotator cuff tear support the association between a high AI and the occurrence of full thickness rotator cuff tear
Relationship of radiographic acromial characteristics and rotator cuff disease: a prospective investigation of clinical, radiographic, and sonographic findings.
These controversial results indicate that the shape of the acromion has an influence on the genesis of rotator cuff tear and the orientation of the glenoid. Compared with healthy controls, patients with rotator cuff tear have an increased inclination of the glenoid fossa. In addition, glenoid anteversion was associated with tears of the posterior muscles and retroversion with tears of anterior muscles.
Is there an association between the individual anatomy of the scapula and the development of rotator cuff tears or osteoarthritis of the glenohumeral joint?: A radiological study of the critical shoulder angle.
developed the concept of the CSA. The CSA measures the angle between a line connecting the superior and inferior glenoid margins and a line connecting the inferior margin with the most lateral border of the acromion on true anteposterior radiographs of the shoulder (figure 1). Deviations from A CSA> 35° (grade 3) indicates an increased inclination and/or lateral extension of the acromion that is associated with the occurrence of rotator cuff tears. Normal shoulders have a CSA that lies between 30° and 35° (grade 2) and a CSA <30° (grade 1) is associated with a higher rate of osteoarthritis.
Is there an association between the individual anatomy of the scapula and the development of rotator cuff tears or osteoarthritis of the glenohumeral joint?: A radiological study of the critical shoulder angle.
Recent studies confirmed the significantly higher CSA in patients with rotator cuff tear (RCT), although they questioned the classification of CSA in RCT to be >35°.
Figure 1Measurement of the critical shoulder angle (CSA) in the antero-posterior radiography of the shoulder. The CSA is the angle between the line connecting the superior glenoid rim with the inferior glenoid rim and the line connecting the inferior glenoid rim with the most lateral acromion edge.
A possible biomechanical explanation why patients with a high CSA are more likely to have rotator cuff tear and patients with a low CSA osteoarthritis goes back on the concept of glenohumeral joint compressing and shearing forces.
A CSA >35° destabilises the glenohumeral joint and increases the superior shear forces from deltoid activity during the glenohumeral motion to 80° abduction to in a biomechanical model.
This increase in superior shear forces require up to 44% higher supraspinatus muscle activity to compensate and maintain the glenohumeral joint alignment. This causes chronic overload to the supraspinatus muscle with a peak of the superior shear forces between 40° and 90° thoracohumeral abduction.
This interesting yet surprising finding indicates that not overhead motions increase stress to the rotator cuff tendon but daily activities where the arm does not exceed 90° abduction.
When thoracohumeral abduction exceeds 90°, a higher CSA leads to inferior translation of the humeral head in normal shoulders as shown in an in-vivo model.
This counterintuitive effect may be explained by a downward directed lever arm through the surrounding soft tissue that is pressed into the acromial arch during deltoid activity in abduction above 90°. With the increase in lateral extension of the acromion, the deltoid moment arm is higher producing a larger downward directed force and causing the evasive movement of the humerus in the inferior direction.
In patients with rotator cuff tear, no such correlation was observed indicating that the joint is destabilised with the loss of supraspinatus muscle function.
For clinical routine use, it is important that the radiographs are accurately taken, because if the viewing angle on the radiographs differs from the true anteroposterior direction, CSA measurement may not be accurate.
Together with age and trauma, the CSA accurately predicts rotator cuff tears. The risk to develop a rotator cuff tear can be calculated with the input of the three variables in the rotator cuff tear score formula. A score >10 predicts posterosuperior rotator cuff tear with a sensitivity 84% and specificity of 81%.
With the CSA, a reliable measurement for anatomical variety that may lead to rotator cuff tear was found, and it also explains why rotator cuff tear and osteoarthritis are usually mutually exclusive.
Biological changes in rotator cuff tear
The previously discussed anatomical varieties in the glenohumeral joint cause repetitive trauma to the rotator cuff muscles through subacromial impingement.
Together with chronic overuse, this may lead to degeneration of the musculotendinous unit. Due to its insertion on the superior part of the humeral head, the supraspinatus muscle is more exposed to the previously described superiorly directed shearing forces than the other rotator cuff muscles and therefore the most commonly injured. Generally, degenerative changes in the musculotendinous unit are induced by a combination of external and internal triggers that lead to chronic tendinopathy. The complex interplay of these factors was reviewed in extend by Seitz et al.
Chronic tendinopathy is the inflammatory response of the hypocellular tendon tissue with infiltration of inflammatory cells that induce the degradation of the extracellular matrix (ECM) and drive the resident tenocytes to undergo apoptosis.
Once the tendon is completely torn, subsequent degenerative processes in the muscle are initiated. Inflammatory cells infiltrate into the muscle and initiate the degradation of muscle fibres.
This results in atrophy of the muscle with loss of sarcomeres in series and leads to both decrease of muscle fibre cross section and muscle fibre length. The loss of muscle volume is associated with an increase in pennation angle that causes an enlargement of the intermyofibrillar and intramyofibrillar spaces. The free space is then filled with connective tissue and the degenerated muscle fibres are replaced by fat cells. An advanced fatty infiltration marks the irreversible end point of the degeneration of the musculotendinous unit.
The cellular and molecular processes in tendon and muscle degeneration share some common mechanisms, which are indeed to date only partially understood.
Inflammation is the first response to injury
Recent research has revealed the important role of inflammatory processes during tendinopathy and muscle degeneration.
Commonly, this response is divided into three stages: mechanical stress induces the inflammatory response, which increases the permeability of the tissue's vasculature. Immediately neutrophils invade into the site of injury. These cells are the first-line response and stimulate M1 macrophages that enter the site of injury to remove cell debris and to regulate the subsequent proliferative phase. During the proliferative phase, a temporary scar tissue that is rich in unstructured collagen type III is formed by invading fibroblasts. This temporary scar acts as a scaffold for the formation of highly structured type I collagen tendon tissue during the following remodelling phase.
M1 macrophages are proinflammatory and release inflammatory cytokines (interleukin1-ß (IL1-ß) and tumour necrosis factor α (TNFα)) that sustain inflammatory processes. The cytokines activate specific intracellular cascades in tendon and muscle cells and attract fibroblasts that produce a disorganised collagen scaffold. Subsequently, M1 macrophages switch their phenotype and become alternatively activated M2 macrophages.
M2 macrophages have a wide spectrum of functions and so far it remains controversial if the different subsets are only functionally different or distinct subpopulations.
They have anti-inflammatory properties and are thought to regulate the tissue's regeneration and remodelling of the scar tissue. During early remodelling, regulatory M2 macrophages (Mreg) promote reorganisation of the disorganised collagen scaffold. They are usually beneficial in re-establishing tissue integrity and promote homeostasis.
M2a macrophages release excessive amounts of growth factors (ie, transforming growth factor beta 1 (TGFß1)) that lead to specific reactions in the tendon and muscle cells discussed later. Generally, these cytokines cause the fibroblasts to excessively deposit ECM components and thus cause fibrosis.
On the intracellular level, the inflammatory cytokines mobilise intracellular protein cascades that have common pathways but differ in their end product depending on the tissue type. M1 macrophages release IL1-ß and TNFα and their binding upregulates the expression of nuclear factor kappa B (NF-κB). NF-κB expression is modulated by poly [adenosine diphosphate ribose] polymerase 1 (PARP-1), which in turn has distinct influences on other degenerative cascades in the tendon and muscle after injury. The disturbance of this interaction by deletion or downregulation of PARP-1 leads to a lower inflammatory reaction, which protects muscle from damage secondary to unloading in a mouse model of RCT.
Neer award 2016: reduced muscle degeneration and decreased fatty infiltration after rotator cuff tear in a poly(ADP-ribose) polymerase 1 (PARP-1) knock-out mouse model.
The upregulation of NF-κB is a crucial step but potentially detrimental in the response of tenocytes to proinflammatory cytokines (IL1-ß and TNFα). NF-κB upregulation leads to apoptosis of tenocytes and the upregulation of metalloproteinases that degrade the ECM and downregulates the collagen producing proteins.
During the proliferative phase, the anti-inflammatory M2reg macrophages increase scar tissue formation rather than promote the regeneration of normal tendon tissue. They attract fibroblasts from the epitenon/peritenon to infiltrate into the tendon, which have increased transcription of genes encoding for collagen and thereby form a scar tissue that mainly consist of the collagen type III, which is mechanically unstable. M2reg macrophages resolve inflammation and promote subsequent tissue regeneration.
During the following remodelling phase, the ECM is reorganised by replacing the mechanically inferior collagen type III matrix by a mechanically stronger collagen type I matrix. This process requires the tendon to be mechanically loaded. The mechanism how mechanical forces are transduced into biomechanical signals are not fully understood, but integrins—transmembrane proteins linking the ECM with the cytoskeleton—play an important role in a process termed mechanosensing. With chronical overload or unloading, the processes are redirected into chronic tendinopathy and the macrophages switch into the profibrotic M2a subtype and overexpress growth factors that promote neovascularisation (vascular endothelial growth factor (VEGF)) and fibrosis (TGF-ß) in the attempt to promote healing.
TGF-ß expression is highly variable during the regenerative process depending on the injury mechanism, and its role during degenerative and regenerative processes is not entirely understood.
TGF-ß is activated during regeneration after tendon injury and is an important mediator during normal tendon development and homeostasis. However, overexpression of TGF-ß induces fibrosis and tenocyte apoptosis.
Regeneration and degeneration are in balance during normal muscle homeostasis due to mutual interactions between atrophy and regeneration. Overloading of the muscle leads hypertrophy of the muscle fibres and conversely unloading of the muscle disturbs the balance and induces pathways that ultimately lead to the degradation of the fibres.
Similarly to the reaction in tendons, inflammatory cells enter the muscle during the initial inflammatory phase. With the release of IL-1ß and TNFα, they trigger the intracellular activation of NFκB (figure 2). In muscle fibres, NF-κB has an important influence on the degeneration after muscle injury. It regulates the expression of inflammatory cytokines that lead to rapid apoptosis or necrosis of muscle fibres. The cell debris of these fibres are phagocytised by macrophages. Not all muscle fibres undergo immediate destruction by apoptosis or necrosis. Most of the fibres degenerate through a process that allows controlled degradation of the muscle fibres by the ubiquitin-proteasome system. NF-κB promotes muscular atrophy and fibre degradation directly by activating the effectors of the ubiquitin-proteasome system. For this process, activation of intermediary cytokines is not necessary. The polyubiquinated fibre components are then degraded through proteolysis by the proteasome. This degenerative process removes excessive muscle fibres, and sustained activation of NF-κB inhibits the onset of regenerative processes by downregulating the expression of the myogenic regulatory factors (MRFs). This inhibition of myogenic differentiation and regeneration is the third major effect of NF-κB in muscle degeneration. Although inhibited by NF-κB, the muscle retains its remarkable ability to regenerate. Soon after injury, the M1 macrophages switch to M2reg macrophages that secrete anti-inflammatory IL-10 and promote myogenesis. During the proliferative phase, satellite cells and mesenchymal stem cells (MSCs) are activated and undergo proliferation and differentiation. This process is orchestrated by the MRFs which, in cooperation with other endocrine growth factors, instigate the development mature myocytes from precursor cells.
Figure 2Schematic drawing of the extracellular and intracellular processes involved in the degeneration of muscle cells. As shown, NF-kB plays a central role in the inflammatory processes and is activated by extracellular inflammatory cytokines. Transcriptional activation of NF-kB is promoted by the poly-(ADP-ribose) polymerase 1 (PARP-1), which plays thereby a crucial role in NF-kB regulated processes. IL-1ß, Interleukin 1 beta; MyoD, myogenenic factor 3; NF-kB, nuclear factor kappa B; TGF-ß, tumour growth factor ß; TNF-α, tumour necrosis factor alpha.
and M2reg macrophages switch to become profibrotic M2a macrophages. These cells trigger the secretion of TGF-ß and myostatin from the fibroblasts in the ECM.
Myostatin belongs to the TGF-ß superfamily and is a potent inhibitor of muscle hypertrophy and thus muscle regeneration. The inhibition of regenerative pathways directs the postinjury reaction to atrophy and fibrosis.
With this induction of degenerative pathways, M2a macrophages influence the balance of muscular homeostasis towards degeneration. This overpowers the regenerative pathways and leads to the progression of atrophy, retraction and fibrosis. Ultimately, the muscle fibres degenerate, and the free space is filled with fat vacuoles. Additionally, adipocytes infiltrate from the vasculature into the free intermyofibrillar and intramyofibrillar space. The regulation of this fatty infiltration is not entirely understood, but mutual interaction of proadipogenic pathways and regenerative MRFs were described.
In addition, myostatin and TGF-ß inhibit proadipogenic pathways and thereby delay the onset of fatty infiltration during the presence of M2a macrophages.
Taken together, the regulation of muscle homeostasis and reaction to injury is a complex interplay between degenerative and regenerative processes. Muscles exhibit a remarkable ability to regenerate injured muscle fibres to remain its structural integrity after injury under normal circumstances with continued loading of the tendon. However, with sustained unloading, the degenerative processes overpower regeneration and ultimately lead to fatty infiltration. A pathological such as a rotator cuff tear marks a potential for the irreversible end-point of the muscle's degenerative processes
The effects of myostatin on adipogenic differentiation of human bone marrow-derived mesenchymal stem cells are mediated through cross-communication between Smad3 and Wnt/beta-catenin signaling pathways.
Rotator cuff tendon healing has received increased attention over the past several years. While rotator cuff repair enjoys a historical record of good results, studies that examine the antatomic integrity of the tendon after repair reveal a much higher rate of retear than expected.
Several factors are associated with failure of tendon healing including, age, medical comorbidities, tear size, presence of degenerative changes in the tendon and muscle (chronicity) and environmental factors. To add to the complexity, whether this is actual retear or failure of healing, is debatable. Clinical studies document this healing phenomenon and a body of basic science research has evaluated this process on a more in-depth level.
Clinical healing
Failure of healing has been documented for over two decades. One of the original manuscripts to include this information by Harryman et al,
evaluated the anatomic integrity of rotator cuff tears with ultrasound. After open rotator cuff repair, they found higher rates of recurrent tears in larger tears. They reported better outcomes in patients with intact cuffs. With data extrapolation, age seems a significant factor. A few years later, Liu and Baker
found that 17 of 18 repairs of massive rotator cuff tears had retears after single row repair. Importantly, these were massive tears and all but one patient were over 60 years of age. In spite of this anatomic result, the average American Shoulder and Elbow Surgeons score improved from 44 to 88. Bishop, et al evaluated the early results after arthroscopic and open repair.
Their healing rates were similar after both techniques; however, in larger tears over 3 cm, there were more defects in the arthroscopic group (open 62% vs 24% arthroscopic intact). Boileau et al
evaluated results after arthroscopic repair. Seventy-one per cent of patients had healed repairs, but age was a significant factor in healing. In fact, if patients were over age 65, there was only a 43% chance of healing a single tendon tear. This is one of the first manuscripts to evaluate age as an independent variable. Lichtenberg et al
Influence of tendon healing after arthroscopic rotator cuff repair on clinical outcome using single-row Mason-Allen suture technique: a prospective, MRI controlled study.
analysed 177 rotator cuff repairs. With multivariate analysis, age was not an independent factor, but rather fatty degeneration and retraction were strongly associated with failure of healing. Importantly, age, retear and degenerative changes in the muscle are colinear variables, so large, well-powered studies are necessary to investigate this. Other studies show very high rates of healing. Keener et al found an over 90% healing rate after arthroscopic repair.
found a high healing rate after arthroscopic repair, and these repairs were in younger, healthy patients. Clearly, there are trends that coincide with epidemiology of rotator cuff tears, and most agree that age, tear size and chronicity have direct influence on healing after rotator cuff repair.
Iannotti et al evaluated the timing of failure of cuff healing after rotator cuff repair.
In this multicentre study of 113 patients with rotator cuff repairs of 1–4 cm in size, 17% had recurrent defects seen on serial MRIs. The majority of the tears occurred 12–26 weeks after arthroscopic cuff repair. There was a linear increase in retears over this time period.
A retrospective, cohort study of 1600 patients showed a 13% retear rate. Although similar improvements were seen in range of motion and pain levels with overhead activity, supraspinatus and external rotation power were greater in the shoulders with intact repairs.
A study by Tham et al showed an increase in vascularity and bursal thickness after repair, and the tendon thickness remained constant and similar to the opposite side in shoulders evaluated by ultrasound after repair.
introduced the idea of restoring the footprint of the rotator cuff. The anatomy of the cuff insertion is such that the enthesis is several millimetres to a centimetre in width. Following that, new repair techniques focused on restoring the footprint. Cadaveric studies
documented increased pull out strength with the new double row repairs in terms of pull out strength, maintenance of integrity with cyclic loading and contact pressure. While this is the case in cadaveric studies, the clinical results are more mixed.
compared single versus double row repair in 80 shoulders with postoperative MRIs. Twenty-two out of 39 repairs had defects after single row repairs. Eleven out of 41 repairs had defects after double row repair. This was one of the first studies to demonstrate improvement in structural outcome with double row repair. Franceschi et al
showed better healing with fewer defects after double row repair; however, the mechanical advantage did not translate to a better clinical result as there were no differences in functional outcomes between the groups. Charousset et al
Can a double-row anchorage technique improve tendon healing in arthroscopic rotator cuff repair?: a prospective, nonrandomized, comparative study of double-row and single-row anchorage techniques with computed tomographic arthrography tendon healing assessment.
evaluated single versus double row repair. Overall, there was better healing in larger tears with a double row repair, but no differences in clinical outcome.
Another study evaluating single row repairs prospectively found 90% of the repairs intact for single tendon tears and 83% intact in two tendon tears. Outcome results were good. These findings suggest that patient biology may have the greatest impact on healing potential. While the reasons are not definitively elucidated, differences in patients' age, activity, pain thresholds and expectations may cloud the findings. Indications for double row may ultimately vary according to these certain variables.
Basic science of tendon healing
Tendon healing occurs by the formation of hypervascular poorly organised ECM that slowly remodels with time to a construct that can bear load. Animal studies have shown that healing response is characterised by a fibrovascular scar response rather than by regenerating normal tendon tissue (figure 3).
Structural properties represent the strength (ie, pull out strength) of the tissue and achieve approximately ½ those of normal tissue. The material properties represent the tissue quality or the viscoelastic properties of the repair. Material properties achieve only about 1/5 to 1/10 of normal tissue. Thus, in animal studies, the repaired insertion sites accrue strength by laying down a larger volume of tissue with inferior viscoelastic properties relative to normal tissue.
The tissue is more vascular and less organised. The tendon often heals in a more elongated or stretched out condition, due to altered viscoelastic properties (table 3).
Figure 3A rat supraspinatus tendon stained for collagen type I. (A) The uninjured tendon has tightly organised construct with a clear four-zone insertion site. (B) The injured and repaired specimen show a disorganised, hypercellular tendon without a definitive four-zone insertion site 8 weeks from the time of surgery.
Both skeletal morphological changes as well as biological changes may influence the development of chronic rotator cuff tears. By knowing the mechanisms of tendon and muscle degeneration, we might be able to develop various means to slow down this degenerative process for the purpose of prevention and treatment of degenerative rotator cuff tears. Although the most commonly performed surgery, rotator cuff repair achieves fibrovascular scar formation at the tendon–bone junction rather than the regeneration of normal tendon–bone junction. Hopefully, we will be able to find a method to regenerate the normal tendon-to-bone junction in the future (box 1).
Sher JS, Uribe JW, Posada A, et al. Abnormal findings on magnetic resonance images of asymptomatic shoulders. J Bone Joint Surg Am 1995;77:10–5.
2.
Minagawa H, Yamamoto N, Abe H, et al. Prevalence of symptomatic and asymptomatic rotator cuff tears in the general population: from mass-screening in one village. Journal of Orthopaedics 2013;10:8–12.
3.
Yamamoto A, Takagishi K, Kobayashi T, et al. Factors involved in the presence of symptoms associated with rotator cuff tears: a comparison of asymptomatic and symptomatic rotator cuff tears in the general population. J Shoulder Elbow Surg 2011;20:1133–7.
4.
Moor BK, Wieser K, Slankamenac K, et al. Relationship of individual scapular anatomy and degenerative rotator cuff tears. Journal of shoulder and elbow surgery / American Shoulder and Elbow Surgeons (et al.) 2014;23:536–41.
5.
Millar NL, Hueber AJ, Reilly JH, et al. Inflammation is present in early human tendinopathy. The American Journal of Sports Medicine 2010;38:2085–91.
6.
Lichtnekert J, Kawakami T, Parks WC, et al. Changes in macrophage phenotype as the immune response evolves. Current Opinion in Pharmacology 2013;13:555–64.
7.
Killian ML, Lim CT, Thomopoulos S, et al. The effect of unloading on gene expression of healthy and injured rotator cuffs. Journal of Orthopaedic Research : Official Publication of the Orthopaedic Research Society 2013;31:1240–8.
8.
Schmutz S, Fuchs T, Regenfelder F, et al. Expression of atrophy mRNA relates to tendon tear size in supraspinatus muscle. Clinical Orthopaedics and Related Research 2009;467:457–64.
9.
Galatz LM, Sandell LJ, Rothermich SY, et al. Characteristics of the rat supraspinatus tendon during tendon-to-bone healing after acute injury. Journal of Orthopaedic Research : Official Publication of the Orthopaedic Research Society 2006;24:541–50.
10.
Thomopoulos S, Hattersley G, Mertens L RM, et al The localised expression of extracellular matrix components in healing tendon insertion sites: an in situ hybridization study. Journal of Orthopaedic Research 2002;20:454–63.
References
Codman EA
Akerson IB
The pathology associated with rupture of the supraspinatus tendon.
Factors involved in the presence of symptoms associated with rotator cuff tears: a comparison of asymptomatic and symptomatic rotator cuff tears in the general population.
Relationship of radiographic acromial characteristics and rotator cuff disease: a prospective investigation of clinical, radiographic, and sonographic findings.
Is there an association between the individual anatomy of the scapula and the development of rotator cuff tears or osteoarthritis of the glenohumeral joint?: A radiological study of the critical shoulder angle.
Neer award 2016: reduced muscle degeneration and decreased fatty infiltration after rotator cuff tear in a poly(ADP-ribose) polymerase 1 (PARP-1) knock-out mouse model.
The effects of myostatin on adipogenic differentiation of human bone marrow-derived mesenchymal stem cells are mediated through cross-communication between Smad3 and Wnt/beta-catenin signaling pathways.
Influence of tendon healing after arthroscopic rotator cuff repair on clinical outcome using single-row Mason-Allen suture technique: a prospective, MRI controlled study.
Can a double-row anchorage technique improve tendon healing in arthroscopic rotator cuff repair?: a prospective, nonrandomized, comparative study of double-row and single-row anchorage techniques with computed tomographic arthrography tendon healing assessment.
Contributors MAZ: author of the tendon pathology section. MK: author of the tendon pathology section. TH: author of the epidemiology section. LMG: author of the tendon healing section. EI: author of the epidemiology section, entire management and planning.
Competing interests None declared.
Provenance and peer review Commissioned; externally peer reviewed.
00244-3&id=gr1.jpg){kind=link}
00244-3&id=gr2.jpg){kind=link}
00244-3&id=gr3.jpg){kind=link}