This study examined the role of vitamin D as a factor accounting for fatty degeneration and muscle function in the rotator cuff. There were 366 patients with disorders of the shoulder. A total of 228 patients had a full-thickness tear (group 1) and 138 patients had no tear (group 2). All underwent magnetic resonance arthrography and an isokinetic muscle performance test. The serum concentrations of vitamin D (25(OH)D3) were measured.
In general, a lower serum level of vitamin D was related to higher fatty degeneration in the muscles of the cuff. Spearman’s correlation coefficients were 0.173 (p = 0.001), −0.181 (p = 0.001), and −0.117 (p = 0.026) for supraspinatus, infraspinatus and subscapularis, respectively. In group 1, multivariate linear regression analysis revealed that the serum level of vitamin D was an independent variable for fatty degeneration of the supraspinatus and infraspinatus.
The serum vitamin D level has a significant negative correlation with the fatty degeneration of the cuff muscle and a positive correlation with isokinetic muscle torque.
The main function of vitamin D is to maintain calcium and phosphate homeostasis. In the skin, 7-dehydrocholesterol is converted to pre-vitamin D3 under the influence of ultraviolet radiation, and is then sequentially changed to vitamin D3 (cholecalciferol). Vitamin D3 is also present in the diet and is absorbed in the intestine. It is then transported with a binding protein to the liver, where it is hydroxylated to 25-hydroxyvitamin D3 (25(OH)D3, calcifediol). In the kidneys this is further hydroxylated to 1,25-dihydroxyvitamin D3 (1,25(OH)D3, calcitriol), which is the biologically active form of vitamin D. Although calcitriol is the active form, the serum concentration of calcifediol is 1000 times that of calcitriol, so it is generally agreed that the level of calcifediol in the serum accurately reflects the status of vitamin D.1 Various target organs of vitamin D have been discovered, such as the intestine, kidney, parathyroid and bone. Since the work of Birge and Haddad,2 the relation between vitamin D and muscle function is also well recognised. They noted that calcifediol directly influences the intracellular accumulation of phosphate by muscle cells, and plays an important role in the maintenance of muscle metabolism and function. The active metabolite of vitamin D binds to a highly specific nuclear receptor in muscle tissue3,4 and leads to protein synthesis and growth of the muscle cell.5,6 There are numerous studies that indicate that vitamin D is directly related to muscle mass, strength and function.2,5–9
A tear of the rotator cuff is one of the most common conditions affecting muscles around the shoulder girdle, and the incidence is known to increase with age.10,11 Many structural changes have been seen to occur after the detachment of rotator cuff muscle in animal models.12,13 Among these, fatty degeneration of the muscle of the cuff has been seen as a predictor of repairability and outcome.14–16 It is closely related to the strength of contraction of these muscles.14 Some authors have stated that the change is irreversible, even in patients with successful cuff repair. Fatty degeneration of muscle in a tear of the cuff is known to be related to various factors, such as the duration and size of the tear and dysfunction of the suprascapular nerve.12,17,18 However, in the clinical situation, there are patients who do not follow this general rule, as some have a higher grade of fatty degeneration despite a small tear of short duration, and others with a massive tear may have low grade degeneration. Furthermore, some patients exhibit moderate fatty degeneration of the infraspinatus muscle in the presence of a supraspinatus tear alone. Fatty degeneration of the muscles of the rotator cuff may also be seen in ageing, disuse atrophy, diabetes mellitus, muscular dystrophy and poliomyelitis.19–24
There have been no reports of the level of vitamin D in patients with a tear of the rotator cuff and no demonstration of a relationship between vitamin D and fatty degeneration of the muscles of the cuff as a prognostic indicator.
This study aimed to evaluate the correlation between fatty degeneration of these muscles and the serum level of vitamin D.
Patients and Methods
Between March 2007 and July 2008, 366 consecutive patients with shoulder symptoms underwent magnetic resonance arthrography. The serum levels of calcifediol were measured at the same time. The study was approved by the Institutional Review Board, and all patients gave their informed consent. Exclusion criteria included patients who either did not want to undergo magnetic resonance arthrography or have the serum level of calcifediol measured, those with a history of previous shoulder surgery, and those with a medical condition affecting the serum vitamin D level, such as chronic renal failure. No patient had long-term steroid use or had received a steroid injection prior to the investigation. There were 228 patients with a full-thickness tear (group 1) and 138 patients with other conditions of the shoulder (group 2), including 56 partial-thickness tears, 29 superior labral anterior to posterior (SLAP) lesions, 19 stiff shoulders, 18 with anterior instability, ten with tendinopathy of the rotator cuff, four with calcific tendinitis, one with a rupture of the long head of biceps, and one with rheumatoid arthritis.
In group 1 there were 92 men and 136 women with a mean age of 61.3 years (43 to 80). In group 2 there were 67 men and 71 women with a mean age of 54.1 years (20 to 74). The mean duration of symptoms was 31.3 months (sd 55.4) and 22.1 months (sd 33.7) in groups 1 and 2, respectively. In terms of the demographic differences between the two groups, gender distribution was not significant (p = 0.125), but age difference was (p < 0.001).
Vitamin D (25(OH)D3) and iPTH levels in the serum.
All blood samples were taken fasting, and the calcifediol level was measured with a radioimmunoassay kit (25 OH-VIT.D3-RIA-CT, BIOSOURCE, Nivelles, Belgium). The reference interval for the assay given by the manual was 11 ng/ml to 70 ng/ml (2.5 to 97.5th percentile). The time between acquisition of the blood sample and magnetic resonance arthrography did not exceed ten days. All assays were performed at the department of clinical pathology at our institution, which was blind to the study. Vitamin D hypovitaminosis was defined as a serum level of calcifediol < 40 ng/ml. Vitamin D insufficiency was defined as < 20 ng/ml, and vitamin D deficiency as < 10 ng/ml.24
Fatty degeneration, the size of the tear and test of the isokinetic muscle.
The fatty degeneration of supraspinatus, infraspinatus and subscapularis was measured with the magnetic resonance arthrography (Philips Gyroscan Intera, Philips Medical Systems, Utrecht, The Netherlands) in the outpatient clinic, and evaluated with the grading system of Goutallier et al15 by radiologists who were blind to the current study. This classification system was originally described using axial CT scans, but we applied it to the T1-weighted oblique sagittal magnetic resonance arthrography image where the base of the coracoid, the spine and the body of the scapula form a Y shape, as described by Fuchs et al26 (Fig. 1⇓).
In group 1 the size and retraction of the cuff tear were measured with a probe during arthroscopic repair. After debridement of the torn end, the anteroposterior dimension of the tear was measured at the lateral edge of the footprint. Medial retraction of the cuff was estimated by the distance from the apex of the tear to the footprint. An isokinetic muscle performance test (Biodex system 3 pro, Biodex Medical System Inc, Shirley, New York) was performed to evaluate the torque of the abduction and external rotation of both the affected and the unaffected shoulders.
All statistical analyses were performed using the SPSS software package (version 12.0, SPSS Inc., Chicago, Illinois), and a p-value < 0.05 was determined as the level of statistical significance. Pearson’s correlation coefficients were analysed to determine the relationship between various parametric variables, including the serum concentrations of calcifediol, age, the duration of symptoms, the size and retraction of the tear, the abduction and external rotation torque of the affected and unaffected shoulders. Spearman’s correlation coefficients were calculated between the grade of fatty degeneration of each cuff muscle and the serum concentrations of calcifediol and these variables. Analysis of variance (ANOVA) was used for the analysis of the difference in concentration of calcifediol according to the grades of fatty degeneration. Student’s t-test and the Mann-Whitney U test were used to verify the differences in variables between groups and genders. Finally, stepwise and backward linear regression analysis was used to determine which was an independent factor for fatty degeneration of the cuff muscles in group 1.
Vitamin D levels in the serum.
The mean levels of calcifediol was 44.02 ng/ml (sd 20.26) in the all patients, 44.26 ng/ml (sd 21.54) in group 1 and 43.64 ng/ml (sd 18.02) in group II. There were 117 patients with vitamin D hypovitaminosis in group 1, and 72 in group 2. There were five patients with vitamin D insufficiency in group 1, and four in group 2. No patient in either group had a level of vitamin D < 10 ng/ml.
Correlation with fatty degeneration.
The mean grade of fatty degeneration of each cuff muscle in all the patients was 2.1 (sd 1.1), 1.2 (sd 0.9) and 1.0 (sd 0.8) for supraspinatus, infraspinatus and subscapularis, respectively. They were 2.5 (sd 1.0), 1.5 (sd 0.9) and 1.1 (sd 0.8) in group 1, and 1.4 (sd 0.8), 0.8 (sd 0.5) and 0.7 (sd 0.7) in group 2. There were significant differences in the mean grades of supraspinatus (p < 0.001), infraspinatus (p < 0.001) and subscapularis (p < 0.001) between groups 1 and 2. Between genders, there were also significant differences in the mean grades of supraspinatus (male:female = 1.9 (sd 1.1):2.3 (sd 1.1), p < 0.001), infraspinatus (1.0 (sd 0.8):1.3 (sd 0.9) p < 0.001) and subscapularis (0.8 (sd 0.8):1.1 (sd 0.8) p < 0.001). The differences in grades of fatty degeneration between genders were significant in group 1 with p = 0.041 for supraspinatus, p = 0.015 for infraspinatus and p = 0.016 for subscapularis. They were also significant in group 2, with p = 0.004 for supraspinatus, p = 0.006 for infraspinatus and p = 0.002 for subscapularis.
Full-thickness tears, female gender, larger size and retraction of the tear, older age, lower serum levels of calcifediol and lower abduction and external rotation torque of the affected and unaffected shoulders were related to higher grades of fatty degeneration. The duration of symptoms did not correlate with fatty degeneration (Table I⇓). In patients with a full-thickness tear (group I), higher grades of degeneration in the supraspinatus and infraspinatus were related to older age, a larger size and retraction of the tear, a lower level of calcifediol and a lower abduction and external rotation torque of the affected and unaffected shoulder. The grade in subscapularis was not correlated with the serum level of calcifediol (Table II⇓). In the overall population, ANOVA revealed a significantly different level of calcifediol among each grade of degeneration of supraspinatus (p = 0.031), and post hoc analysis showed that the level of calcifediol in grade IV degeneration in supraspinatus was significantly lower than that in the other grades. The serum level of calcifediol exhibited statistical significance on ANOVA (p = 0.016) with the grade I fatty degeneration in infraspinatus, but no statistical significance was identified in subscapularis (p = 0.621, Table III⇓). In group 2, a higher level of fatty degeneration was statistically related to female gender and older age, a lower level of calcifediol, and a lower abduction and external rotation torque of the affected and unaffected shoulder (Table IV⇓).
Correlation with calcifediol (25 (OH)D3).
The difference in the mean serum level of calcifediol was not significant between groups but was significantly lower in women (male:female = 48.85 (sd 23.43):40.32 (sd 16.56), p < 0.001). A lower serum level of calcifediol also related to a higher degree of fatty degeneration of each cuff muscle, and a lower abduction and external rotation torque of the affected and unaffected shoulder on the isokinetic muscle performance test (Table V⇓). In general, the serum level of calcifediol correlated with fatty degeneration of all the cuff muscles, and when analysed separately in each group with the supraspinatus and infraspinatus.
Linear regression analysis.
In group I, the grades of fatty degeneration in supraspinatus and infraspinatus were related to age, the size and retraction of the tear, the serum level of calcifediol and the abduction and external rotation torque of the affected and unaffected shoulders. Multivariate linear regression analysis showed that retraction of the tear (p < 0.001), the abduction torque of the unaffected shoulder (p = 0.002) and the serum level of calcifediol (p = 0.006) were the independent variables for fatty degeneration of the supraspinatus. For retraction of the infraspinatus tear (p < 0.001), the external rotation torque of the unaffected shoulder (p = 0.001), the serum level of calcifediol (p = 0.020) and the size of the tear (p = 0.042) were independent variables. However, for subscapularis, the abduction torque of the affected shoulder (p < 0.001), the size of the tear (p = 0.002), the external rotation torque of the affected shoulder (p = 0.015), and retraction of the tear (p = 0.022) showed significance in the analysis.
Fatty degeneration of the muscles of the rotator cuff is related to a tear, and the degree of these changes increases with the size of the tear and the duration of the rupture.12,27 Fatty degeneration has been shown in infraspinatus in patients with isolated tears of supraspinatus.19,28 Some authors have suggested that the dysfunction and atrophy of infraspinatus are related to suprascapular neuropathy in massive tears.17,29 It is also known that fatty degeneration of the cuff muscles due to age or disuse can develop without a tear.19–21 There are certain conditions, such as diabetes,22 muscular dystrophy23 and polymyositis,24 which can lead to fatty degeneration. However, the exact biomechanical or molecular biological causes of fatty degeneration of muscles of the rotator cuff remain unclear.
The relationship between vitamin D and muscle function is well known, and can be explained by the direct effects on protein synthesis and cell growth in the muscle tissue.3,5,6 Vitamin D deficiency is common in the general population, especially in the elderly.30,31 Muscle weakness due to vitamin D deficiency is predominantly of the proximal muscle groups, and muscle atrophy, particularly of type II fibres, has been described histopathologically,32,33 but these changes are reversible with vitamin D supplementation.33–35 There are randomised controlled studies describing this in the elderly.7,36,37 Sato et al,37 in a randomised controlled trial, showed that low doses of vitamin D increase the relative number and size of type II muscle fibres and improved muscle strength. Bischoff et al7 found that a course of vitamin D plus calcium over a period of three months reduced the risk of falling by 49% compared to calcium alone. The impact of vitamin D on falls might be explained by the improvement in musculoskeletal function. Hedstrom et al36 also observed that treatment with vitamin D and calcium had a beneficial effect on muscle volume, bone mineral density and clinical function.
We believe that these data on the relationship between fatty degeneration of muscles of the rotator cuff and the serum level of vitamin D are important in that tears of the cuff particularly afflict the elderly, and the prevalence of low vitamin D levels is considerably higher in that age group. The level of serum vitamin D and the grade of fatty degeneration of muscle tears of the rotator cuff bore a close relation and regression analysis verified that the vitamin D level was one of the independent variables for the degeneration of the supraspinatus and infraspinatus. The muscle torque of both the normal shoulder and the affected shoulder correlated with the level of vitamin D which supports previous studies indicating that the serum level of vitamin D is related to general muscle function.
Recent reports have emphasised that fatty degeneration of the cuff is an irreversible prognostic indicator.14–16 There is therefore a need to determine how progression of this process can be influenced, even in patients with a tear, by surgical intervention, where possible, or by supplementation with drugs. Further study is needed to determine whether the level of vitamin D in the tissue of the cuff muscle correlates with the grade of fatty degeneration and the serum level, and whether supplementation with vitamin D will significantly improve the outcome.
There are some limitations to this study. First, we could not demonstrate the relationship between fatty degeneration of the subscapularis and the vitamin D level when analysis was performed separately in groups 1 and 2. This could be due to lack of statistical power although correlation between the two was meaningful when analysed as a total. There were limitations in the grading of degeneration in this muscle and the intact cuff which were subtle, mostly falling into grades 0 and 1, compared to supraspinatus and infraspinatus which were group 1. The grading system is such that discriminating between grade 0 and grade 1 is rather subjective.26 The use of a quantitative grading system might overcome this limitation.38,39
Secondly, there was no normal control group, although group 2 included patients who did not have a full-thickness tear. We did not evaluate the fatty degeneration of other muscle groups in the same patients, although we measured the isokinetic performance of the unaffected shoulder. Further prospective randomised comparative studies with age-and gender-matched controls and other muscle groups would clarify the relationship between vitamin D levels and fatty degeneration of muscles of the rotator cuff. In addition, the age distribution of the two groups was different, those in group 2 being younger. It was not easy to recruit a population of similar age as group 1 who did not have many rotator cuff tears. However, it was not a comparative study and there was also correlation between fatty degeneration of the cuff, abduction and external rotation torque and vitamin D in group 2.
The third limitation of the study was that recruitment of the cohort extended over a year, and the serum vitamin D level is affected by the exposure to sunlight.40,41 However, there was no correlation between such exposure and the serum vitamin D level (correlation coefficient = −0.003 and p = 0.961) in this study, and no seasonal predilections related to the other variables. A single evaluation at a certain time may not represent the previous vitamin D status of the patients and a gradual cut-off value was used to evaluate vitamin D deficiency in our study.25 Currently, there is a trend toward upward readjustment in the cut-off value and the recommendation of supplementation. However, in any comparison of the 25(OH)D3 level in different studies, caution is needed because different assay techniques may critically affect the result.42
This study demonstrates that the serum vitamin D level has a significant negative correlation with fatty degeneration of the torn cuff muscles and a positive correlation with isokinetic muscle performance. These findings are true in shoulder disorders other than cuff tears. Future studies are needed to verify the local concentration of vitamin D in the cuff muscles, and whether supplementation with vitamin D affects the functional and anatomical outcomes as well as the improvement of fatty degeneration after rotator cuff repair.
We thank Ms H. R. Kim and Ms S. M. Shim for their support in data collection and patient recruitment. Pacific Edit reviewed the English manuscript prior to submission.
No benefits in any form have been received or will be received from a commercial party related directly or indirectly to the subject of this article.
- Received February 18, 2009.
- Accepted July 20, 2009.
- © 2009 British Editorial Society of Bone and Joint Surgery