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
To determine the prevalence of systemic laboratory abnormalities among patients undergoing rotator cuff repair (RCR).
Methods
Patients who underwent RCR at the authors’ institution for 1 year between October 2021 to September 2022 were retrospectively identified. Preoperative laboratory values, including serum sex hormones, vitamin D, hemoglobin A1C, and a lipid panel, were obtained as part of our routine practice during the study period. Demographics and tear characteristics were compared in patients with laboratory data and those without. For included patients with laboratory data, mean laboratory values and percentage of patients with abnormal laboratory values were recorded.
Results
During a 1-year period of time, 135 RCRs were performed, of which preoperative labs were obtained on 105. Of these, 67% were sex hormone deficient, 36% were vitamin D deficient, 45% had an abnormal hemoglobin A1C, and 64% had an abnormal lipid panel. In total 4% had “normal” labs.
Conclusion
In this retrospective study, sex hormone deficiency is highly prevalent among patients undergoing RCR. Nearly all patients undergoing RCR have systemic laboratory abnormalities involving either sex hormone deficiency, vitamin D deficiency, dyslipidemia, and/or prediabetes.
Introduction
Rotator cuff pathology is among the most common musculoskeletal disorders.
Rotator cuff tears (RCTs) account for ∼4.5 million visits to physicians each year and have a lifetime incidence between 25% and 40% in the United States.
Initial medical management of rotator cuff tears: a demographic analysis of surgical and nonsurgical treatment in the United States Medicare population.
Although RCTs have classically been associated with etiologies such as traumatic injury and progressive degeneration, there are data suggesting that rotator cuff tendinopathy is due, in part, to systemic causes.
For example, hypertension, diabetes, smoking, osteoporosis, dyslipidemia, hyperuricemia, depression, and genetic predisposition have all been demonstrated to be risk factors associated with the development of RCTs.
Additionally, a recent analysis of a large insurance claims database demonstrated that sex hormone deficiency (SHD; i.e., hypogonadism) positively associated with the incidence of RCR.
This most recent study from Smith and colleagues suggests that SHD, or a testosterone-deficient state in males and an estrogen-deficient state in females, may also contribute to the biologic milieu in the development of RCTs. SHD is a clinical syndrome associated with impaired functional activity of the gonads, which can be classified as primary, secondary, or mixed.
Primary SHD is due to dysfunction at the level of the gonads, where testosterone production in the testes and estrogen production in the ovaries is insufficient in males and females, respectively. Secondary SHD is due to dysfunction at the level of the hypothalamus or pituitary, where gonadotropin-releasing hormone and/or gonadotropin release (i.e., follicle-stimulating hormone, luteinizing hormone) is insufficient. Mixed SHD is due to dual defects of the gonads and pituitary-hypothalamic axis. SHD is often due to congenital (e.g., Klinefelter syndrome in males, Turner syndrome in females) or acquired (e.g., secondary to irradiation, infection, trauma, kidney disease, liver disease, etc.) causes.
Sex hormone physiology is complex and involves numerous organ systems; however, this article will focus on SHD and its relationship to the musculoskeletal system, specifically that of the rotator cuff. For example, in the in vitro and animal literature, androgens have been shown to prevent supraspinatus muscle atrophy and fatty infiltration and, thus, may improve outcomes after rotator cuff repair (RCR).
In a murine study, the effect of an estrogen-deficient state on tendon-to-bone healing after RCR found that estrogen deficiency (ED) was associated with less development of chondroid tissue, decreased bone mineral density at the enthesis, and poorer repair biomechanics.
In human studies, polymorphisms in the estrogen-related receptor β gene have been shown to be significantly associated with rotator cuff disease and rotator cuff repair failure.
However, while SHD has been shown to be associated with RCR, the prevalence of systemic laboratory abnormalities among individuals currently undergoing RCR is unknown. Understanding how common SHD is in a typical shoulder and elbow clinical practice, and how SHD manifests via various endocrine markers will further help providers characterize this comorbidity and will provide guidance for future systemic and locally targeted intervention strategies.
The purpose of this study was to determine the prevalence of systemic laboratory abnormalities among patients undergoing RCR. We hypothesized that most patients would have laboratory evidence of either hypogonadism, vitamin D deficiency, dyslipidemia, or prediabetes.
Methods
Protocol Approval
Each institution approved the human protocol for this investigation, and all investigations were conducted in conformity with ethical principles of research. Informed consent for participation in the study was not required by our Institutional Review Board (IRB). This study was performed under the University of Utah IRB as approved protocol #144900.
Patients who underwent RCR at the authors’ institution for 1 year between October 2021 and September 2022 were retrospectively identified. Exclusion criteria were patients under the age of 18 years, those not undergoing RCR, and patients with incomplete laboratory data. In all patients undergoing RCR at the University of Utah Hospital, we obtained the following serologic laboratories preoperatively: 25-OH Vitamin D, Hgb A1C, and a lipid panel that includes cholesterol, triglycerides, high-density lipoprotein (HDL) cholesterol, non-HDL cholesterol, low-density lipoprotein (LDL) cholesterol, and very low-density lipoprotein (VLDL) cholesterol. In women, we obtained estradiol levels. In men, we obtained testosterone levels, free-testosterone levels, sex hormone-binding globulin levels, and bioavailable testosterone levels. Additionally, calculations from these data allowed us to measure the percentage of free testosterone. We retrospectively reviewed these patients to determine the prevalence of abnormalities within these laboratory examinations. Sex hormone levels were recorded as the primary outcome measure, while the Vitamin D, Hgb A1C, and lipid panel were recorded as the secondary outcome measures. Ranges provided by our laboratory were used to define “normal” (Table 1). Our laboratory limits for total testosterone, free testosterone, estradiol, and luteinizing hormone were comparable to existing definitions for sex hormone deficiency.
Table 1Normal Ranges (Low and High Limits) for Laboratory Values
Laboratory Test
Low
High
Means ± SD
% (N) Abnormal
Thyroid stimulating hormone (mU/L)
0.34
4.94
1.9 ± 1.3 (0.1 to 6.2)
5% (2/40)
Luteinizing hormone (IU/L)
1.7
8.6
12 ± 11 (0 to 35)
44% (11/25)
Estradiol (pg/mL)
27
433
59 ± 65 (0 to 280)
25% (13/25)
Testosterone (ng/dL)
By LC-MS/MS
300
1080
348 ± 260 (8 to 1158)
46% (16/35)
By ELISA
300
890
467 ± 330 (0 to 1500)
33% (23/69)
Bioavailable testosterone (ng/dL)
By LC-MS/MS
130
680
166 ± 151 (1 to 657)
71% (24/34)
By ELISA
131
682
201 ± 186 (0 to 1325)
64% (42/66)
Free testosterone (pg/mL)
By LC-MS/MS
47
244
58 ± 51 (0 to 221)
49% (17/25)
Calculated
47
244
77 ± 74 (0 to 446)
29% (20/70)
Sex hormone binding globulin (nmol/L)
11
80
49 ± 28 (10 to 182)
7% (5/72)
Percent of free testosterone (%)
1.6
2.9
1.6 ± 0.5 (0.6 to 3)
63% (43/68)
Vitamin D, 25-Hydroxy (ng/mL)
30
80
37 ± 18 (12 to 109)
36% (34/96)
Alkaline phosphatase (U/L)
40
120
79 ± 23 (22 to 145)
2% (1/47)
Bone alkaline phosphatase (U/L)
0
55
35 ± 14 (16 to 67)
13% (3/24)
Calcium (mg/dL)
8.4
10.5
9.4 ± 0.4 (8.4 to 10.3)
0% (0/56)
Hemoglobin A1C (%)
5.6 ± 0.6 (4.6 to 8.6)
44% (46/103)
Normal
0
5.6
Prediabetes
5.7
6.4
Diabetes
6.5
100
Cholesterol (mg/dL)
0
239
187 ± 37 (93 to 258)
5% (5/102)
Low-density lipoprotein (mg/dL)
0
129
108 ± 31 (11 to 163)
28% (28/101)
Very-low-density lipoprotein (mg/dL)
0
30
27 ± 14 (8 to 79)
34% (63/98)
High-density lipoprotein (mg/dL)
40
59
50 ± 14 (28 to 92)
21% (22/102)
Normal ranges (low and high limits) for laboratory values provided by our institution’s laboratory compared to laboratory values of the included patients, as well as the percent and number that were abnormal for each of those values. ELISA, enzyme-linked immunosorbent assay; LC-MS/MS, liquid chromatography/mass spectrometry.
Within these patients, we also reviewed their charts to collect the following preoperative demographic data: age, gender, shoulder laterality, hand dominance, American Society of Anesthesiologist (ASA) score, body mass index, work status, medical comorbidities sufficient for calculation of the Charlson Comorbidity Index, smoking status, and indication for surgery. On preoperative magnetic resonance imaging (MRI), we measured tear width, tear retraction, supraspinatus Goutallier
classification, infraspinatus Goutallier classification, subscapularis Goutallier classification, and teres minor Goutallier classification. Moreover, within this cohort, we collected preoperative values for an American Shoulder and Elbow Surgeons (ASES) Score,
Simple Shoulder Test (SST) scores, the visual analogue scores (VAS) for pain and function, and range of motion. On the basis of the intraoperative records, we collected the number of anchors, the repair construct, the biceps treatment, and whether a concomitant subscapularis repair was performed.
Statistical Analysis
All analyses were conducted in Excel 16 (Microsoft, Redmond, WA) and SPSS 28 (IBM, Armonk, NY). Continuous data were evaluated for normality using the Komolgorov-Smirnov test. Data with a Gaussian distribution were compared between groups using Student’s t-tests. Data with a non-Gaussian distribution were compared using Mann-Whitney U-tests. Categorical data will be compared between groups using Chi-square tests or Fischer’s Exact tests, as appropriate depending upon cell populations.
Results
Study Cohort
During the study period, 135 rotator cuff tears were performed. When the 105 included patients were compared to the 30 excluded patients, there were no differences in age, gender, whether the surgery was on the dominant side, smoking status, body mass index, ASA scores, Charlson comorbidity index, ASES score, range of motion, MRI characteristics, or intraoperative variables. The preoperative VAS pain scores were statistically, but not clinically, significantly higher in the excluded group (5.9 ± 2.5 vs 4.8 ± 2.4; P = .038), and the included group had more anchors used (2 ± 1 vs 3 ± 1; P = .011). The characteristics of the included patients are shown in Table 2. Patients were included if they had any of the preoperative labs available; only 85/105 (81%) of the included patients had a complete set of preoperative laboratory values. Of these 85 patients, 80 (94%) had either abnormal sex hormone levels, were already on sex hormone supplementation, had low vitamin D levels, had an abnormal hemoglobin A1C, or had an abnormal lipid panel. Among these 85, 5 patients (6%) had completely normal laboratory values, 24% (20) had at least one of the above categories abnormal, 17% (14) had two, 28% (24) had three, 18% (15) had four, and 8% (7) had five or more. Thus, most patients undergoing rotator cuff repair have systemic preoperative laboratory abnormalities (Fig 1).
Table 2Comparison of Included and Excluded Patients
Variable
Excluded (n = 30)
Included (n = 105)
P Value
Age (years)
58 ± 11
60 ± 10
.463
Female sex
40% (12/30)
34% (36/105)
.564
Dominant side
96% (25/26)
95% (69/73)
1.000
Current smokers
10% (3/30)
5% (5/103)
.178
Body mass index
30 ± 5
29 ± 6
.706
ASA score
2 ± 1
2 ± 1
.467
Charlson comorbidity index
2 ± 1
2 ± 2
.444
VAS pain
5.9 ± 2.5
4.8 ± 2.4
.038
ASES
41 ± 18
48 ± 19
.124
Active forward elevation (°)
148 ± 35
135 ± 45
.180
Adducted external rotation (°)
61 ± 18
53 ± 20
.088
Preop tear width (mm)
15 ± 9
19 ± 11
.092
Preop tear retraction (mm)
21 ± 10
18 ± 12
.424
Supraspinatus atrophy >1
20% (4/20)
28% (25/88)
.881
Infraspinatus atrophy >1
20% (2/20)
9.1% (8/88)
.260
Subscapularis atrophy >1
10% (2/20)
15.9% (14/88)
.373
Teres minor atrophy >1
0% (0/20)
1% (1/88)
.661
Revision rotator cuff repair
10% (3/30)
9% (9/105)
.729
Adducted internal rotation
Lateral thigh
4% (1/23)
5% (5/95)
Buttock
9% (2/23)
14% (13/95)
Lumbosacral junction
13% (3/23)
15% (14/95)
L3
17% (4/23)
29% (28/95)
T12
30% (7/23)
25% (24/95)
T7
26% (6/23)
12% (11/95)
.605
Number of anchors
2 ± 1
3 ± 1
.011
Repair construct
Single row
53% (16/30)
38% (40/105)
Double row
47% (14/30)
62% (65/105)
.135
Concomitant biceps tenodesis
83% (25/30)
85% (89/105)
.633
Subscapularis repair
33% (10/30)
31% (32/105)
.766
Continuous data are displayed as means ± SD, and discrete data are displayed as % (N). Muscular atrophy data reference Goutallier stage >1. ASA, American Society of Anesthesiologists; ASES, American Shoulder and Elbow Surgeons Score; VAS, visual analog scale for pain.
Fig 1This Venn diagram shows percent of patients undergoing rotator cuff repair that had normal or abnormal systemic lab abnormalities. Green, normal; Red, vitamin D deficient; Yellow, abnormal lipids and/or abnormal hemoglobin A1C; Blue, hypogonadal or taking hormone supplementation. Created using DeepVenn.
Within the 105 included patients, 69 (66%) were male, of whom 11 (16%) were taking hormone replacement therapy. Of the 69 males, 41 (59%) had either a low testosterone or a low bioavailable testosterone level (hypogonadal; Table 1). Of these 41, 11 had concomitant luteinizing hormone levels drawn, of which 2 (18%) were high, suggesting primary gonadal failure as the etiology of hypogonadism. The other 9 (82%) had normal or low luteinizing hormone levels, suggesting hypothalamic-pituitary disorders, or secondary hypogonadism. In combination, 71% (49/69) either had a low testosterone (i.e., were hypogonadal) or were already being supplemented. Within the 105 included patients, 36 (34%) were female, of whom 1 (3%) was taking hormone replacement therapy. Of the 25 females with a preoperative estradiol level, 13 (52%) were low (i.e., hypogonadal). In combination, 56% (14/25) of females either had a low estradiol level or were already being supplemented. Within the overall cohort with preoperative labs, 67% (63/94) of included patients were either hypogonadal or already taking hormone replacement therapy.
Vitamin D Deficiency and Thyroid
Of the 95 patients with preoperative vitamin D levels, 34 (36%) were abnormally low, with 3 having severe deficiency (<12 ng/mL), 6 having deficiency (12-19 ng/mL), and 25 having insufficiency (20-29 ng/mL). Only a single patient had a high alkaline phosphatase level, and this patient had a normal vitamin D level. There were no patients with abnormal calcium levels. Only 2 patients had abnormally high thyroid-stimulating hormone levels.
Diabetes and Prediabetes
Within our cohort, 103 patients had preoperative hemoglobin A1C levels. Of these, 57 (55%) were normal, 38 (37%) showed prediabetes, and 8 (8%) showed diabetes. Of those with hemoglobin A1C levels considered normal, 4% had a known diagnosis or diabetes or prediabetes. Of those with hemoglobin A1C levels considered prediabetic, 34% had a known diagnosis of diabetes or prediabetes. Of those with hemoglobin A1C levels consistent with diabetes, 75% had a known diagnosis of diabetes or prediabetes. Overall, these preoperative labs uncovered a previously unknown diagnosis of diabetes or prediabetes in 26% (27/103) of patients.
Dyslipidemia
Examining lipid panels, 5% (5/102) had high cholesterol, 28% (28/101) had high low-density lipoprotein, 34% (36/98) had high very-low-density lipoprotein, and 21% (22/102) had low high-density lipoprotein. In combination, 64% (63/98) of patients had an abnormality in at least 1 of the above lipid levels. When considering patients with a complete lipid panel and hemoglobin A1C, only 21/97 (20%) were normal, with 80% having at least one metabolic abnormality.
Discussion
In this retrospective case series, nearly all (94%) patients undergoing rotator cuff repair were either hypogonadal, vitamin D deficient, diabetic/prediabetic, or had dyslipidemia. Among males, 71% either were hypogonadal or were already being supplemented, among females 56% either were hypogonadal or were already being supplemented, and among the combined cohort, 67% either were hypogonadal or already taking hormone replacement therapy. Thirty-six percent had abnormally low vitamin D levels. Metabolic syndromes were common—37% had prediabetes, 8% had diabetes, and 26% of patients received a new diagnosis of either prediabetes or diabetes. In combination, 64% (63/98) of patients had an abnormality in at least one of their lipid levels. These results demonstrate that systemic laboratory abnormalities are common among patients undergoing rotator cuff repair.
Within our study, the prevalence of sex hormone deficiency, as defined based upon a low testosterone/bioavailable testosterone level or based upon the patient already receiving hormone replacement therapy, was 67%. Multiple prior studies have been conducted to determine the prevalence of testosterone deficiency using similar criteria in normal populations. Fink et al. conducted a cross-sectional study of men over the age of 65, and only 17% of men had testosterone levels <300 nL/dL, which was the same level used within our laboratory to define deficiency.
In another cross-sectional study, Harman et al. found hypogonadism in 12% of men under 50, 19% of men aged 50-60, and 28% of men aged 60-70, again using similar laboratory cutoffs.
These are historical controls and comparison to historical controls has limitations in that the populations are different, and the criteria used to define hypogonadism may be different. However, these rates are numerically much lower than the prevalence observed in our study population, confirming the findings of another recent study suggesting that hypogonadism associated with cuff tears.
Certainly, this definition alone does not account for whether these patients were symptomatic from the hormone deficiency, and future studies are planned in this regard. Additionally, given the findings of this study, in addition to other in vitro and animal studies, suggesting sex hormone deficiency is associated with poorer tendon-to-bone healing following RCR,
future research should determine whether normalizing preoperative sex hormone levels (i.e., estrogen in females and testosterone in males) would improve postoperative outcomes in patients undergoing RCR.
Within our study, the prevalence of vitamin D deficiency was 36%. A prior Korean study demonstrated that 44% of patients undergoing rotator cuff repair are vitamin D deficient, especially younger patients.
future studies should determine whether vitamin D supplementation may improve outcomes in patients undergoing rotator cuff repair with vitamin D deficiency.
Within our study, the prevalence of dyslipidemia was 64%, and prediabetes was 37%. Multiple prior studies have demonstrated a connection between dyslipidemia and rotator cuff disease.
The effect of diabetes, hyperlipidemia, and statins on the development of rotator cuff disease: A nationwide, 11-year, longitudinal, population-based follow-up study.
In combination with our own findings, these studies provide further support for the theory that rotator cuff disease is, in part, due to systemic abnormalities. Twenty-seven percent of patients within our study received a new diagnosis of either prediabetes or diabetes as part of preoperative testing. These results are strongly suggestive that these metabolic syndromes are common among patients undergoing rotator cuff repair and that this patient population may be underdiagnosed with these disorders. Given the importance of these disorders for overall health, the high prevalence in this setting suggests that routine screening could be considered.
Finally, further research will be necessary to determine whether these systemic biologic abnormalities are modifiable risk factors—i.e., whether supplementation with vitamin D for those patients with vitamin D deficiency improves outcomes.
Limitations
This study has several limitations. The sample size is limited. This dataset reflects the practices of the two senior authors at the University of Utah Hospital and may not be generalizable to other settings. Within this study, sex hormone deficiency was defined by laboratory analysis alone, and these findings do not consider clinical symptomatology. We did not measure sex hormone levels in those taking hormone replacement therapy. Additionally, it is common for laboratory values to be outside the upper or lower limits of normal reference ranges, but the clinical relevance of this is unknown. We included those patients with some, but not all, of the included laboratory values to increase study power, which could create selection bias. However, given that most patients (78%) were included and that there were no clinically significant differences between the included and the excluded groups (Table 2), there are no signs of selection bias. Postoperative functional scores or imaging was not performed on these patients; therefore, the impact of sex hormone deficiency on rotator cuff healing was not investigated. The findings of this study alone cannot be used to demonstrate that these abnormalities should be treated.
Conclusion
In this retrospective study, sex hormone deficiency is highly prevalent among patients undergoing RCR. Nearly all patients undergoing RCR have systemic laboratory abnormalities involving either sex hormone deficiency, vitamin D deficiency, dyslipidemia, and/or prediabetes.
Initial medical management of rotator cuff tears: a demographic analysis of surgical and nonsurgical treatment in the United States Medicare population.
The effect of diabetes, hyperlipidemia, and statins on the development of rotator cuff disease: A nationwide, 11-year, longitudinal, population-based follow-up study.
The authors report the following potential conflicts of interest or sources of funding: J.M.H. reports personal fees from StreamDx, Paterna Bio, Firm Tech, Turtle Health, Carrot, Maximus, and Inherent Biosciences; and grants from Boston Scientific, Coloplast, Endo Pharmaceuticals, outside the submitted work. P.N.C. reports IP royalties from DePuy and Responsive Arthroscopy; publishing royalties and financial or material support from Journal of Shoulder and Elbow Surgery; consulting fees from DePuy, DJ Orthopaedics, Smith & Nephew, speaker fees from DePuy; and stock or stock options from TitinKM Biomedical, outside the submitted work. R.T. reports editorial or board membership in Journal of Orthopaedic Trauma; and personal fees from Cayenne Medical, Conextions, INTRAFUSE, Journal of Bone and Joint Surgery - American, KATOR, Mitek, Zimmer, outside the submitted work. Full ICMJE author disclosure forms are available for this article online, as supplementary material.
00053-6&id=gr1.jpg){kind=link}