Relationship Between Testosterone and Rotator Cuff Tears

In This Article

Abstract and Introduction

Abstract

Introduction: The purpose of this study was to evaluate rates of rotator cuff tears (RCTs), repairs (RCRs), and revision RCR in patients who were prescribed testosterone replacement therapy (TRT) and compare these patients with a control group.

Methods: The PearlDiver database was queried for patients who were prescribed testosterone for at least 90 days between 2011 and 2018 to evaluate the incidence of RCTs in this population. A second analysis evaluated patients who sustained RCTs using International Classification of Diseases, 9th/10th codes to evaluate these patients for rates of RCR and revision RCR. Chi square analysis and multivariate regression analyses were used to compare rates of RCTs, RCR, and subsequent or revision RCR between the testosterone and control groups, with a P-value of 0.05 representing statistical significance.

Results: A total of 673,862 patients with RCT were included for analysis, and 9,168 of these patients were prescribed testosterone for at least 90 days before their RCT. The TRT group had a 3.6 times greater risk of sustaining an RCT (1.14% versus 0.19%; adjusted odds ratio (OR) 3.57; 95% confidence interval (CI) 3.57 to 3.96). A 1.6 times greater rate of RCR was observed in the TRT cohort (TRT, 46.4% RCR rate and control, 34.0% RCR rate; adjusted OR 1.60; 95% CI 1.54 to 1.67). The TRT cohort had a 26.7 times greater risk of undergoing a subsequent RCR, irrespective of laterality, within 1 year of undergoing a primary RCR when compared with the control group (TRT, 47.1% and control, 4.0%; adjusted OR 26.4; 95% CI 25.0 to 27.9, P < 0.001).

Conclusions: There is increased risk of RCTs, RCRs, and subsequent RCRs in patients prescribed testosterone. This finding may represent a musculoskeletal consequence of TRT and is important for patients and clinicians to understand. Additional research into the science of tendon injury in the setting of exogenous anabolic steroids remains of interest.

Level of Evidence: Level III, retrospective cohort study.

Introduction

Rotator cuff tears (RCTs) are a common cause of pain and decreased upper extremity function in patients. Reports of shoulder pain account for more than 4.5 million physician visits annually in the United States, with rotator cuff disease being the most common reason for presentation.^[1,2] RCTs, whether full-thickness or partial, have been estimated to be present in up to 34% of asymptomatic individuals, with high rates of progression to symptomatic tears.^[3,4] Tear development and progression are strongly associated with age, as well as smoking and family history.^[5] Approximately one-third of patients presenting with symptomatic RCTs will undergo rotator cuff repair (RCR), which has proven to be a highly effective and cost-efficient technique for successfully managing RCTs.^[6]

Hypogonadism is defined as the clinical presentation of low sex hormone levels, which can include symptomatic decreases in muscle mass, strength, bone mass, libido, concentration, and mood.^[7,8] In cases of symptomatic hypogonadism, with quantifiably low testosterone levels, exogenous androgen supplementation can be prescribed as part of a testosterone replacement therapy (TRT) plan. The use of TRT increased by 359% in the United States between 2001 and 2011 and has continued to grow in popularity internationally despite a poorly characterized safety profile and controversy regarding treatment indications.^[9,10] TRT has been associated with increased risks of cardiovascular events, stroke, hematologic pathology, and mortality.^[11] Despite consequent efforts to limit the prescription of TRT, it remains a very popular therapeutic modality capable of providing tangible benefits to affected patients.

The relationship between anabolic sex hormones and rotator cuff pathology remains the subject of ongoing research. Supraphysiological dosing of exogenous testosterone has demonstrated rapid hypertrophy of muscle at a rate that exceeds tendon adaption, therefore increasing the risk of tendon injury.^[12–14] However, there is a paucity of literature examining rates of RCT and RCR in the general population being prescribed testosterone, particularly given the notable correlations that both RCT and hypogonadism demonstrate with aging. The purpose of this study was to examine any association between prescription testosterone usage with rates of RCTs, RCR, and ipsilateral retear injuries after RCR. Our hypothesis was that patients who had previously undergone testosterone administration would have markedly higher rates of RCT and RCR when controlling for age and demographic variables.

Table 1. Comparison of Patient Demographics of Unmatched and Matched Testosterone and Control Cohorts
Characteristic	Prior Testosterone Use	Control	P
Characteristic	Prior Testosterone Use	N = 200,000	N = 544,702
Sex, male (%)	171,762 (85.9)	240, 316 (44.1)	<0.001
Age, mean ± SD	54.4 ± 12.2	53.8 ± 11.6	<0.001
CCI, mean ± SD	0.9 ± 1.5	0.1 ± 0.5	<0.001
Comorbidities (%)
Hypogonadism	141,878 (70.0)	64,769 (11.9)	<0.001
Diabetes mellitus	78,363 (39.2)	153,304 (28.1)	<0.001
Tobacco use	62,826 (31.4)	140,315 (25.8)	<0.001
Osteoarthritis	63,295 (31.6)	95,079 (17.5)	<0.001
Osteoporosis	5,733 (2.9)	33,945 (6.2)	<0.001
Chronic kidney disease	28,308 (14.2)	47,626 (8.7)	<0.001
Alcohol use	12,358 (6.2)	28,909 (5.3)	<0.001
Obesity	72,637 (36.2)	135,757 (24.9)	<0.001
Class III obesity	29,588 (14.8)	53,407 (9.8)	<0.001
Lung disease	49,251 (24.6)	87,059 (16.0)	<0.001
Congestive heart failure	11,569 (5.8)	22,288 (4.1)	<0.001
Dementia	5,779 (2.9)	24,342 (4.5)	<0.001

CCI, Charlson Comorbidity Index

Table 2. Comparison of the Rates of Rotator Cuff Tear Over a 1-Year Period From the Completion of a 90-Day Continuous Prescription of Testosterone Replacement Therapy
Comparison of the Rates of Rotator Cuff Tear
	No. of Tears		Unadjusted Analysis		Adjusted Analysis^a
	Prior Testosterone Use	Control	OR (95% CI)	P	OR (95% CI)	P
Overall	N = 200,000	N = 544,702	—	—	—	—
n (%)	2,277 (1.14)	1,051 (0.19)	5.96 (5.54–6.41)	<0.001	3.57 (3.23–3.95)	<0.001
Male	N = 171,762	N = 240,316	—	—	—	—
n (%)	2,067 (1.20)	496 (0.21)	5.89 (5.34–6.50)	<0.001	4.73 (4.24–5.29)	<0.001
Female	N = 28,238	N = 304,324	—	—	—	—
n (%)	210 (0.74)	555 (0.18)	4.10 (3.50–4.81)	<0.001	3.49 (2.89–4.19)	<0.001

CI = confidence interval, OR = odds ratio
^aAdjusted by controlling for age, sex, Charlson Comorbidity Index, hypogonadism (testosterone or estrogen deficiency), tobacco use, diabetes, chronic kidney disease, osteoporosis, osteoarthritis, obesity, class III obesity, alcohol use, lung disease, congestive heart failure, and dementia.

Table 3. Comparison of Rates of Rotator Cuff Repair, Any Subsequent Cuff Repair After Surgery, and Ipsilateral Cuff Repair After Surgery
Rates of Cuff Repair and Subsequent Retear
Outcome	Group	Prior Testosterone Use	Control	Unadjusted Analysis		Adjusted^a Analysis
Outcome	Group	Prior Testosterone Use	Control	OR (95% CI)	P	OR (95% CI)	P
RCR, n (%)	Overall	N = 9,168	N = 664,694	—	—	—	—
		4,251 (46.37)	226,112 (34.02)	1.68 (1.61–1.75)	<0.001	1.60 (1.54–1.67)	<0.001
	Male	N = 7,757	N = 331,165	—	—	—	—
		3,547 (45.73)	116,570 (35.20)	1.55 (1.48–1.62)	<0.001	1.61 (1.54–1.69)	<0.001
	Female	N = 1,411	N = 333,529	—	—	—	—
		704 (49.89)	109,542 (32.84)	2.04 (1.83–2.26)	<0.001	2.00 (1.80–2.22)	<0.001
Subsequent RCR within 1 yr, n (%)	Overall	N = 6,980	N = 568,044	—	—	—	—
		3,289 (47.12)	22,893 (4.03)	26.69 (25.42–28.03)	<0.001	26.44 (25.05–27.90)	<0.001
	Male	N = 5,761	N = 292,697	—	—	—	—
		3,127 (54.28)	12,055 (4.12)	27.64 (26.16–29.20)	<0.001	28.37 (26.49–30.38)	<0.001
	Female	N = 1,219	N = 275,347	—	—	—	—
		562 (46.1)	10,838 (3.94)	20.88 (18.62–23.40)	<0.001	21.07 (18.76–23.64)	<0.001
Subsequent ipsilateral RCR within 1 yr, n (%)	Overall	N = 2,992	N = 164,755	—	—	—	—
		227 (7.59)	6,129 (3.72)	2.14 (1.85–2.44)	<0.001	1.90 (1.63–2.20)	<0.001
	Male	N = 2,572	N = 83,172	—	—	—	—
		212 (8.24)	3,172 (3.81)	2.27 (1.96–2.62)	<0.001	1.99 (1.68–2.35)	<0.001
	Female	N = 420	N = 81,583	—	—	—	—
		15 (3.57)	2,957 (3.62)	0.99 (0.59–1.65)	1	1.01 (0.60–1.69)	0.972

CI = confidence interval; OR = odds ratio, RCRs = rotator cuff repairs
^aAdjusted by controlling for age, sex, Charlson Comorbidity Index, hypogonadism (testosterone or estrogen deficiency), tobacco use, diabetes, chronic kidney disease, osteoporosis, osteoarthritis, obesity, class III obesity, alcohol use, lung disease, congestive heart failure, and dementia.