Androgen therapy and musculoskeletal health in older adults: An umbrella review

Abstract

Background

Testosterone and androgen supplementation have been increasingly prescribed in older adults to counteract age-related declines in lean body mass and strength, however, there is uncertainty about their benefits. We synthesized the evidence concerning androgen therapy, particularly testosterone, on muscle-related outcomes in older adults.

Methods

Following prospective registration (OSF; osf.io/8wm3k), we conducted an umbrella review of systematic reviews (SRs) and meta-analyses of randomized clinical trials (RCTs) involving adults aged ≥60 years receiving androgen therapy. PubMed/MEDLINE, Scopus, and WOS were searched from inception to October 2025. Outcomes included lean body mass, muscle strength, and adverse events. Methodological quality was assessed with the AMSTAR-2 tool. The overlap of primary RCTs across reviews was assessed by calculating covered area (CCA). Pooled effect sizes with 95% confidence intervals (CIs) were summarized.

Results

There were 11 eligible SRs, comprising 121 unique RCTs, with 20,846 male participants, none were women (of whom 8447 received testosterone therapy). Methodological quality was rated in 3 SRs as high, in 5 SRs as moderate, and in 3 SRs as low or critically low. The overlap was moderate (CCA = 9.92%). Lean body mass changes in response to testosterone reported in 5 SRs of moderate to high quality ranged from 1.6 kg (95% CI 0.6–2.6; I² = 0.0%) to 3.59 kg (95% CI 2.38–4.81; I² = 98%). Muscle strength changes in response to testosterone reported in 6 SRs of moderate to low quality showed maximal lower-limb effect in leg press (91.23 Newtons; 95% CI 0.23–182.22; I² = 62%), and leg extension (144.10 Newtons; 95% CI 44.21–244.00; I² = 0%). There was no effect demonstrated on major cardiovascular events or all-cause mortality.

Conclusion

Testosterone therapy improved lean muscle mass and strength without adverse effect in older men. There is a need for well-designed trials, including both elderly men and women, to guide primary care and public health policy and practice in the future.

Resumen

Antecedentes

La suplementación con testosterona y otros andrógenos se ha prescrito cada vez más en adultos mayores para contrarrestar los descensos relacionados con la edad en la masa magra y la fuerza; sin embargo, existe incertidumbre sobre sus beneficios. Se sintetizó la evidencia sobre los efectos de la terapia androgénica, en particular la testosterona, en resultados relacionados con el músculo en los adultos mayores.

Métodos

Siguiendo un registro prospectivo (OSF; osf.io/8wm3k), se llevó a cabo una umbrella review de revisiones sistemáticas (RS) y metaanálisis (MA) de ensayos clínicos aleatorizados (ECA) que incluían adultos ≥60 años que recibían terapia androgénica. Se realizaron búsquedas en PubMed/MEDLINE, Scopus y WOS desde su inicio hasta octubre de 2025. Los resultados incluyeron masa corporal magra, fuerza muscular y eventos adversos. La calidad metodológica se evaluó con la herramienta AMSTAR-2. El solapamiento de los ECA primarios entre revisiones se evaluó mediante el cálculo del área cubierta (CCA). Se resumieron los tamaños del efecto agrupados con sus intervalos de confianza (IC) del 95%.

Resultados

Se incluyeron 11 RS, que comprendían 121 ECA únicos, con 20.846 participantes varones; no se incluyeron mujeres (de los cuales 8.447 recibieron terapia con testosterona). La calidad metodológica se clasificó como alta en 3 RS, moderada en 5 RS y baja o críticamente baja en 3 RS. El solapamiento fue moderado (CCA = 9,92%). Los cambios en la masa corporal magra en respuesta a la testosterona, reportados en 5 RS de calidad moderada a alta, oscilaron entre 1,6 kg (IC 95%: 0,6-2,6; I² = 0,0%) y 3,59 kg (IC 95%: 2,38-4,81; I² = 98%). Los cambios en la fuerza muscular, reportados en 6 RS de calidad moderada a baja, mostraron el mayor efecto en la extremidad inferior, con mejoras en el leg press (91,23 Newtons; IC 95%: 0,23-182,22; I² = 62%) y en la extensión de rodilla (144,10 Newtons; IC 95%: 44,21-244,00; I² = 0%). No se demostró efecto sobre eventos cardiovasculares mayores ni sobre la mortalidad por cualquier causa.

Conclusión

La terapia con testosterona mejora la masa muscular magra y la fuerza sin efectos adversos sobre el riesgo cardiovascular mayor en los varones mayores. Se necesitan ensayos bien diseñados, que incluyan tanto a varones como a mujeres de edad avanzada, para orientar la atención primaria y las políticas de salud pública en el futuro.

.

Introduction

Ageing is accompanied by a progressive decline in skeletal muscle mass and strength, which are key determinants of physical function and independence in older adults.¹ Globally, the number of people aged ≥60 years is expected to reach approximately 1.4 billion by 2030 and 2.1 billion by 2050, reflecting a rapid demographic shift toward older populations.² The deterioration of these parameters contributes substantially to frailty, disability, falls, and loss of autonomy, representing a major public health challenge in aging populations. This process underlies sarcopenia, a condition defined as the coexistence of reduced lean body mass with impaired strength and/or physical performance.³ The global burden of sarcopenia is expected to exceed 200 million cases in the coming decades as population ageing accelerates.⁴

Among the biological mechanisms contributing to age-related muscle loss, declining androgen levels, particularly testosterone, play a central role. Testosterone preserves muscle mass and strength by enhancing protein synthesis and reducing protein breakdown.⁵ However, serum testosterone decreases by about 1.6% per year after age 30, and hypogonadism affects 2.1–12.8% of older men.⁶ By the age of 60, 25–30% of men present low serum testosterone levels, shifting physiological balance toward catabolism,⁷ and increasing vulnerability to falls, fractures, loss of independence, and greater long-term healthcare needs.⁸ Against this biological backdrop, hormone replacement therapy and androgen supplementation have gained considerable attention not only in the research field, but also in clinical settings.⁹ In the US alone, testosterone prescriptions have increased more than 500% over the past decade, driven by the low cost of these therapies and their well-documented anabolic effects on lean body mass, fat, and bone.¹⁰ Although androgen therapy may also be prescribed in women in selected clinical contexts (i.e., hypoactive sexual desire disorder), the evidence remains more limited compared to men.¹¹

Despite growing interest and widespread clinical use, the effects of testosterone therapy in older adults remain uncertain. Heterogeneity in dosing, intervention duration, population characteristics, and interactions with physical activity adds complexity to the interpretation of existing evidence collected in systematic reviews (SRs) and meta-analyses (MAs).⁷ In a search for overviews of evidence, there was one potentially relevant umbrella review⁹; however, it assessed multiple treatments without specifically focusing on androgen therapy and it did not assess the overlap of primary studies across reviews (CCA). This study aimed to collate randomized evidence syntheses on the effects of androgen therapy, particularly testosterone, on muscle-related outcomes in older adults.

Methods

This umbrella review followed a prospective protocol registered in the Open Science Framework (OSF; osf.io/8wm3k) and is reported according to the PRISMA 2020 statement.¹²

Search strategy and selection criteria

A comprehensive literature search was conducted without language or date restrictions across three databases: PubMed/MEDLINE, Scopus, and WOS, from inception to October 2025. We used a combination of terms and free-text keywords related to “testosterone,” “androgen therapy,” “hormone replacement therapy,” “sarcopenia,” “muscle strength,” “muscle mass,” “physical performance,” and “older adults.” Complete search strategies for each database are presented in Supplementary Table 1. In addition to database searching, manual (hand) searches were performed to ensure comprehensive coverage. Reference lists of all included SRs and MA were screened for additional relevant studies. Citation tracking of key articles was conducted using Google Scholar to identify potentially eligible SRs and/or MAs not captured through the database searches. Relevant grey literature was explored by screening conference proceedings and checking the OSF and PROSPERO registries for completed or ongoing reviews related to androgen therapy in older adults.

All retrieved records were imported into EndNote X9 for deduplication. Two reviewers (PL and VRGN) independently screened titles and abstracts using Rayyan software. Eligible studies were SRs with or without MA of RCTs evaluating androgen therapy by any route and dosage in adults with a mean age of ≥60 years, and reporting at least one outcome related to lean body mass, muscle strength, and physical performance and/or body composition, functional independence, quality of life, and adverse events (e.g., cardiovascular risk, prostate health). Studies focusing on other populations, study designs, or outcomes were excluded. Detailed reasons for exclusion for each study are provided in Supplementary Table 2. Any disagreements were resolved by consensus or consultation with a third reviewer (NCI).

Data extraction, quality assessment, and evidence grading

Two reviewers (PL and VRGN) independently extracted data on study characteristics, including citation details (first author, year), intervention type and dosage, comparator, number of included RCTs, number and characteristics of participants, and pooled effect sizes with 95% confidence intervals (CIs) for the outcomes. Methodological quality of the included SRs and MAs was independently assessed by two reviewers (PL and VRGN) using the 16-item AMSTAR-2 tool.¹³ Each review received an overall rating of high, moderate, low, or critically low, based on the number and severity of 16 potential critical and non-critical weaknesses. High quality indicated no or only one non-critical weakness, whereas moderate and low indicated one or more than one critical weakness, respectively. Detailed evaluations, including reasons for downgrading, are provided in Supplementary Table 3. Disagreements were resolved via consultation with a third reviewer (N.C-I). The certainty of evidence for each outcome was evaluated,¹⁴ using the GRADE framework,¹⁵ reflecting the levels of risk of bias, indirectness, inconsistency, imprecision, and potential publication bias.¹⁶ Certainty was rated as high, moderate, low, or very low. Inconsistency was assessed using I² statistic, with values >50% indicating a serious limitation and values >80% very serious limitation.¹⁷ Publication bias was evaluated based on funnel plot asymmetry and Egger's test results reported within each meta-analysis.¹⁶

Data synthesis and overlap analysis

Data from included SRs and MAs were summarized descriptively and organized according to intervention type and outcome domain. For each androgen outcome, statistically significant findings were summarized alongside their corresponding GRADE certainty ratings. To quantify the degree of overlap among primary trials included across SRs and MAs, the corrected covered area (CCA) was calculated using the formula: CCA = (N − r)/(r × c − r) where N represents the total number of included publications, r is the number of unique rows (primary studies), and c is the number of columns (SRs). Overlap was interpreted as: slight (<5%), moderate (5–10%), high (10–15%), and very high (>15%).¹⁸

Results

Study selection and overview

A total of 547 records were identified, with 450 records screened after duplicate removal. Of 33 full-text articles assessed for eligibility, 11 SRs were included in the umbrella review (Fig. 1). These SRs comprised 245 RCT inclusions, corresponding to 121 unique RCTs, involving 28,046 male participants, none were women of whom 10,301 (37%) received testosterone therapy.

Figure 1.

Flowchart of the selection process for systematic reviews included in the umbrella review.

Characteristics of the included SRs

Included SRs were published between 2004 and 2024 and reported 30 MA assessing muscle mass, muscle strength, physical performance, body composition, and safety outcomes (Table 1). Four SRs (36%) exclusively included men with biochemically confirmed hypogonadism,^8,19–21 whereas six SRs (55%) included populations with predominantly low-to-borderline testosterone levels.^{1,6,7,22–24} One review (9%) included adult men without clear stratification by baseline testosterone levels.²² While these reviews did not strictly exclude eugonadal participants, their weighted baseline means (e.g., 302.8–314.3 ng/dL) indicate a study population largely concentrated at or below the clinical threshold for deficiency.^{1,6,7,22–24} All SRs assessed mixed testosterone replacement therapy formulations, including injectable, transdermal, and oral preparations. Follow-up duration ranged from 2 weeks to 36 months, although most musculoskeletal outcomes were assessed within 6–12 months.^6,19,20

Table 1.

Main characteristics of the systematic reviews included in the umbrella review of androgen therapy for muscles in the elderly.

Author, year	Country	No. of RCTs included	Sample size	No. of patients(intervention/control)	Population characteristics	Intervention	Follow-up duration	Main outcomes/risk factors
Krause Neto et al., 2015	Brazil	8 RCTs	651	356/295	Elderly men ≥60 y; T versus placebo	Testosterone therapy with variable doses, formulations, and administration routes across included studies.5 oral capsules (OC), 4 intramuscular (IM), and 1 gel.	12 weeks–36 months	Lean body mass (LBM) and fat mass (FM)
Parahiba et al., 2023	Brazil	11 RCTs	773	392/381	Middle-aged and elderly men with a mean age ranging from 66.0 ± 7.93 to 77.9 ± 7.3 years in the intervention groups and 65.0 ± 3.0 to 76.3 ± 8.0 years in the placebo groups.	Testosterone therapy with variable doses, formulations, and administration routes across included studies.7 intramuscular (IM), 3 gels, and 1 transdermal patch.	12 weeks–36 months	Muscle mass, strength, physical performance
Ottenbacher et al., 2006	EE.UU	11 RCTs	474	NR/NR	Healthy men aged ≥65 years,	Testosterone or DHT replacement therapy versus placebo6 intramuscular (IM), 3 transdermal patches, 1 oral capsule, and 1 gel.	4 weeks–156 weeks	Muscle strength (upper and lower extremities)
Buratto et al., 2023	EE.UU	16 RCTs	1,728	922/806	Adult men (mean age 77.1 ± 7.6 years).	Testosterone therapy versus placebo6 gels, 4 intramuscular (IM), 3 oral capsules (OC), and 2 transdermal patches.	6 months–36 months	Bone density, lean mass, fat mass, handgrip strength, adverse events.
De Spiegeleer et al., 2018	Umbrella review	NR	NR	NR/NR	Older adults ≥65 years	Testosterone therapy versus placebo	NR	Pharmacological interventions for sarcopenia
Isidori et al., 2005	Italy	29 RCTs	1,083	625/437	Middle age men (mean age: 64.5)	Testosterone therapy versus placebo19 intramuscular (IM), 9 patches, and 2 gels.	1 month–36 months	Body composition, bone metabolism, lipids
Skinner et al., 2018	Australia	41 RCTs	FFM: 31 RCTs, n = 2,381Strength: 17 RCTs, n = 5,095	31 RCTs for FFM (n = 1,213 TRT, n = 1,168 placebo); 17 RCTs for strength (n = 2,572 TRT, n = 2,523 placebo).	Middle aged and older men receiving T by various routes versus placebo	15 intramuscular (IM), 12 gels, and 6 transdermal patches.	2.5 months–36 months	Fat-free mass (FFM), total body strength, lower-extremity strength, upper-extremity strength.
Nam et al., 2018	South Korea	18 RCTs	≈2,600	NR/NR	Elderly men aged ≥60 years.	Testosterone therapy versus placebo11 transdermal (gel/patch), 5 intramuscular (IM) and 3 oral capsules.	2 months–36 months	Muscle strength, physical function
Jaiswal et al., 2024	EE.UU	30 RCTs	11,502	5,843/5,659	Adult men (mean age 61.6–61.8 years)	Testosterone replacement therapy (TRT) versus placebo	2 weeks–3 years	CVD events, stroke, myocardial infarction, all-cause and CVD mortality
Haddad et al., 2007	EE.UU	30 RCTs	1,642	808/834	Men with low or low-normal testosterone;	TRT versus placebo18 intramuscular (IM), 7 transdermal patches, and 5 oral capsules (OC).	2 weeks–3 years	BP, glycemia, TC, HDL, LDL, TG, cardiovascular events
Fernández-Balsells et al., 2010	EE.UU	51 RCTs	2,701	≈1,355/≈1,346 hombres	Adult men (median age: 61 years)	TRT versus placebo22 intramuscular (IM) administrations, 11 transdermal patches, 6 oral capsules (PO), and 5 gels.	3 months–36 months	Adverse effects of T

Methodological quality assessment and overlap of evidence

The distribution of evidence strength across outcomes, based on GRADE assessment (Supplementary Table S3). For lean body mass, 2/6 (33.3%) MAs were rated as high certainty,^19,20 1/6 (16.7%) as moderate certainty,⁶ and 2/6 (33.3%) as low certainty.^7,8 For fat mass, 1/3 (33.3%) MAs was rated as high certainty,¹⁹ 1/3 (33.3%) as moderate certainty,⁶ and 1/3 (33.3%) as low certainty.⁷ For muscle strength outcomes, 4/11 (36.4%) MAs were rated as high certainty,^19,20 1/11 (9.1%) as moderate certainty,¹ 2/11 (18.2%) as low certainty,⁸ and 1/11 (9.1%) as very low certainty,²³ reflecting substantial variability across reviews. For physical performance outcomes, most of the evidence was of low certainty, with 3/4 (75%) MAs rated as very low²³ and 1/4 (25%) as low.⁸ Regarding safety outcomes, 2/3 (66.7%) MAs were rated as high certainty,^21,24 while 1/3 (33.3%) was rated as moderate certainty.²²

Using the AMSTAR 2 tool (Supplementary Table S4 and Fig. 2). Three SRs (27%) were rated as high methodological quality.^19,21,23 Five (45%) as moderate quality,^6,22 one (9%) as low quality,²³ and two (18%) as critically low quality.^1,20 The high-quality reviews mainly assessed the effects of testosterone therapy on metabolic, musculoskeletal, and cardiovascular outcomes.^19,21,24 Reviews with moderate quality primarily focused on specific subpopulations or limited evidence bases, such as early testosterone trials.^6,7 The critically low-quality reviews corresponded to SRs with substantial methodological limitations, including lack of protocol registration, incomplete reporting of inclusion criteria, and absence of formal bias assessment.^1,20 The degree of overlap among primary trials across SRs was moderate (CCA) was 9.92% (Supplementary Table S5).

Figure 2.

Methodological quality of 11 systematic reviews included in the umbrella review evaluating the effects of testosterone replacement therapy (TRT) on lean mass, muscle strength, body composition, physical performance, and cardiovascular safety in middle-aged and older men. Quality assessment was conducted using the 16 items of the AMSTAR-2 tool. Critical items: 2, 4, 7, 9, 11, 13 and 15; non-critical items: 1, 3, 5, 6, 8, 10, 12, 14 and 16.

Effects of intervention

Lean body mass

Five SRs consistently reported that testosterone therapy increased lean body mass compared with placebo or no treatment. Reported pooled mean differences ranged from +1.6 kg (95% CI 0.6–2.6, P < 0.001, I² = 0.0%) to +3.59 kg (95% CI 2.38–4.81, I² = 98%).^19,7 Although most analyses demonstrated statistically significant effects, heterogeneity varied substantially, with I² values reaching 98% in one.⁷ Subgroup analyses indicated that intramuscular formulations were associated with larger increases in lean body mass than transdermal preparations (approximately +5.7% vs +1.7%, respectively),²⁰ being more pronounced within the first 6 months.⁶ Certainty of evidence for lean body mass outcomes ranged from moderate to high (Table 2).

Table 2.

Summary of effects, methodological quality, and certainty of evidence included in the umbrella review of androgen therapy for muscles in the elderly.

Author, year	No. of systematic reviews/meta analyses	Results/findings	AMSTAR-2 rating	GRADE certainty of evidence
Krause Neto et al., 2015	1 SR + 2 MAs	Mean lean mass increased, ranging from 1.65 kg (95% CI 1.61–1.69) to 6.20 kg (95% CI 5.22–7.18), although heterogeneity was high (I^2 = 98%). The pooled effect estimate was 3.59 kg [95% CI 2.38–4.81].Androgen therapy reduced fat mass, with a pooled effect estimate of −1.78 kg [95% CI −2.57 to −0.99], also showing substantial heterogeneity (I^2 = 81%).	Moderate	Low
Parahiba et al., 2023	1 SR + 4 MA	Testosterone supplementation significantly increased lean body mass (LBM), with a mean effect of +2.54 kg (95% CI: 1.27–3.80; P < 0.001).For muscle strength, there was a significant improvement in handgrip strength (HGS) of +1.58 kg (95% CI: 0.17–3.00; P = 0.03).Regarding lower-limb strength, testosterone increased leg press strength by +91.23 N (95% CI: 0.23–182.22; P = 0.05) and leg extension strength by +144.10 N (95% CI: 44.21–244.00; P < 0.01).In physical performance, four studies evaluated this outcome; three reported positive effects, but the meta-analysis – including only two studies using the physical performance test – showed no significant overall effect of testosterone supplementation.	Moderate	Low
Ottenbacher et al., 2006	1 SR + 1 Mas	The mean g-index (effect size) adjusted for sample size was 0.53 (95% CI: 0.21–0.86).Sub-analyses showed greater effects in lower-extremity strength (g = 0.63; 95% CI: 0.31–1.28) compared to upper-extremity strength (g = 0.47; 95% CI: 0.12–0.84).Larger effect sizes were observed for injected testosterone/DHT (g = 0.95; 95% CI: 0.30–1.58) compared to topical (g = 0.26; 95% CI: 0.08–0.42) or oral administration (g = −0.21; 95% CI: −1.40–1.02).Effect magnitude varied according to study quality and attrition rate. Sensitivity analysis showed that excluding one outlier study reduced the overall effect to g = 0.23.	Critically low	Moderate
Buratto et al., 2023	1 SR + 5 MAs	Lean mass: 6 mo: ↑ 1.44 kg (95% CI 0.71–2.18; P = 0.001; I^2 = 0%); 12 mo: NS (MD 1.27 kg; 95% CI −0.53–3.06; P = 0.17; I^2 = 0%).Fat mass: 6 mo: NS (−0.62 kg; 95% CI −1.78–0.53; P = 0.29; I^2 = 0%).Hip BMD: 6 mo: ↑ 0.018 g/cm^2 (95% CI 0.006–0.029; P = 0.001; I^2 = 0%).Lumbar BMD: 6 mo: NS (0.004 g/cm^2; 95% CI −0.012–0.019; P = 0.65; I^2 = 0%); 12 mo: NS (0.029 g/cm^2; 95% CI −0.024–0.082; P = 0.28; I^2 = 0.01%).Handgrip strength: 6 mo: NS (Hedges g = 0.03; 95% CI −0.16–0.23; P = 0.74; I^2 = 0%); 12 mo: NS (Hedges g = 0.22; 95% CI −0.71–0.28; P = 0.39; I^2 = 0%).	Moderate	Moderate
De Spiegeleer et al., 2018	Umbrella review 7 SRs	NR	Moderate	NR
Isidori et al., 2005	1 SR + 3 MAs	Body composition: ↓ total fat mass −1.6 kg (95% CI −2.5 to –0.6), −6.2% (95% CI −9.2 to –3.3); ↑ fat-free mass +1.6 kg (95% CI 0.6–2.6), +2.7% (95% CI 1.1–4.4); no change in body weight.Muscle strength: heterogeneous; tendency toward improvement in leg/knee extension and handgrip (SMD = 0.3; 95% CI −0.0–0.6).BMD: ↑ lumbar spine BMD +3.7% (95% CI 1.0–6.4); no change at femoral neck; ↓ bone resorption markers (SMD = −0.6; 95% CI −1.0 to −0.2).Lipids: ↓ total cholesterol −0.23 mmol/L (95% CI −0.37 to −0.10); no change in LDL; ↓ HDL only in studies with higher baseline T (−0.085 mmol/L; 95% CI −0.017 to −0.003).Heterogeneity: Explained partly by dose/type of T and aromatization potential.	High	High
Skinner et al., 2018	1 SR + 3 MAs	Overall TRT effect (all routes combined):• ↑ FFM: ES = 1.20 ± 0.15 (95% CI 0.91–1.49)• ↑ Total body strength: ES = 0.90 ± 0.12 (0.67–1.14)• ↑ Lower-extremity strength: ES = 0.77 ± 0.16 (0.45–1.08)• ↑ Upper-extremity strength: ES = 1.13 ± 0.18 (0.78–1.47)All P < 0.001.By administration route:• Intramuscular TRT: ↑ FFM + 5.7% [ES = 1.49 ± 0.18 (1.13–1.84)]↑ Total body strength + 10–13% [ES = 1.39 ± 0.12 (1.15–1.63)]↑ Lower-extremity strength [ES = 1.39 ± 0.17 (1.07–1.72)]↑ Upper-extremity strength [ES = 1.37 ± 0.17 (1.03–1.70)]All P < 0.001.• Transdermal TRT: ↑ FFM + 1.7% [ES = 0.98 ± 0.21 (0.58–1.39)]↑ Total strength [ES = 0.55 ± 0.17 (0.22–0.88)]↑ Upper-extremity strength [ES = 0.97 ± 0.24 (0.50–1.45)]No effect on lower-extremity strength [ES = 0.26 ± 0.23 (−0.19–0.70), P = 0.26].Subanalyses in men ≥60 years showed similar results.	Critically low	High
Nam et al., 2018	1 SR + 5 MAs	TRT produced small but consistent functional improvements across outcomes. Significant gains were observed in physical activity (PASE, WMD = +18.22 [95% CI: 1.27–35.18]) and physical performance (PPT, SMD = 0.27 [95% CI: 0.12–0.41]), as well as a modest increase in 6-min walking distance (WMD = +9.35 m [95% CI: 0.64–18.07]). In contrast, the effects on handgrip strength (WMD = 0.27 [95% CI: −0.66–1.20]) and knee flexor strength (SMD = 0.23 [95% CI: −0.25–0.72]) were variable and did not reach statistical significance	Low	Very low
Jaiswal et al., 2024	1 SR + 5 Mas	Any CVD events: OR = 1.12 (0.77–1.62); stroke: 1.01 (0.68–1.51); myocardial infarction: 1.05 (0.76–1.45); all-cause mortality: 0.94 (0.76–1.17); CVD mortality: 0.87 (0.65–1.15); all P > 0.3	High	High
Haddad et al., 2007	1 SR + 5 Mas	No significant effect on BP or lipids: TC OR = −0.22 (−0.71 to 0.27); HDL OR = –0.04 (−0.39 to 0.30); LDL OR = 0.06 (−0.30 to 0.42); TG OR = −0.27 (−0.61 to 0.08). Cardiovascular events: OR = 1.82 (0.78–4.23).	Moderate	Moderate
Fernández-Balsells et al., 2010	SR + 7 MAs	Testosterone led to a notable rise in hemoglobin levels (WMD = 0.80 g/dL; 95% CI 0.45–1.14) and hematocrit (WMD = 3.18%; 95% CI 1.35–5.01). Conversely, it produced a significant reduction in HDL cholesterol (WMD = −0.49 mg/dL; 95% CI −0.85 to −0.13). No meaningful differences were observed regarding overall mortality, prostate-related outcomes, or cardiovascular events.	High	High

Muscle strength

Muscle strength outcomes were evaluated across 6 SRs. Compared with placebo or no treatment, testosterone therapy was associated with improvements in lower-limb strength (+91.23 Newton (N) (95% CI 0.23–182.22)) and leg extension strength of +144.10 N (95% CI 44.21–244.00).⁸ Improvements in handgrip strength were modest (+1.58 kg; 95% CI 0.17–3.00), and several analyses showed non-significant findings for upper-extremity measures.⁸ Intramuscular administration was associated with greater improvements in total and lower-limb strength (10–13%) compared with transdermal formulations (2–5%), while oral preparations showed no consistent benefits.²⁰ Overall certainty of evidence for muscle strength outcomes ranged from low to moderate, largely due to heterogeneity and imprecision (Table 2).

Fat mass

Changes in fat mass were evaluated across 3 SRs.^6,19 Compared with placebo or no treatment, testosterone therapy was associated with modest reductions in total fat mass in some analyses, with pooled effect estimates ranging from −1.78 kg (95% CI −2.57 to −0.99) to −1.6 kg (95% CI −2.5 to −0.6).^7,19 Heterogeneity was high (I² up to 81%).⁷ Overall certainty of evidence for fat mass outcomes ranged from low to moderate (Table 2).

Physical performance

Physical performance outcomes were assessed across two SRs. Significant improvements were observed in physical activity scores (PASE +18.22 points; 95% CI 1.27–35.18),²³ physical performance test scores (PPT SMD 0.27; 95% CI 0.12–0.41),²³ and walking distance test (6-min walk test +9.35 m; 95% CI 0.64–18.07).²³ In contrast, the overall pooled results for the physical performance test remained non-significant; 1.46; 95% CI [−0.69, 3.61] and were characterized by high heterogeneity (I² to 90%).⁸ Overall certainty of evidence for physical performance outcomes was low to moderate (Table 2).

Safety and adverse effects

Safety outcomes were assessed across 5 SRs. Testosterone therapy was not associated with an increased risk of major cardiovascular events or all-cause mortality (OR 1.12; 95% CI 0.77–1.62) when administered at physiological doses.²¹ In contrast, testosterone therapy was associated with significant increases in hemoglobin (+0.80 g/dL; 95% CI 0.45–1.14) and hematocrit (+3.18%; 95% CI 1.35–5.01). No increased risk of prostate cancer or clinically relevant urological outcomes was observed.²⁴ Overall certainty of evidence was moderate to high (Table 2).

Discussion

This umbrella review synthesized evidence from SRs and MAs evaluating the effects of TRT on musculoskeletal outcomes predominantly in older men. Across the included reviews, TRT consistently demonstrated anabolic effects on lean body mass. In contrast, improvements in muscle strength and functional performance were modest and heterogeneous, suggesting that increases in muscle mass do not necessarily translate into clinically meaningful improvements in physical function. This discordance between structural and functional outcomes has been observed in previous pharmacological interventions targeting sarcopenia and may reflect the complex interplay between muscle quantity, neuromuscular activation, and physical performance. Importantly, current evidence does not indicate a significant increase in major cardiovascular events or all-cause mortality when TRT is administered within physiological ranges, although consistent haematological changes such as increases in haemoglobin and haematocrit highlight the need for careful clinical monitoring.

Several methodological considerations should be considered when interpreting these findings.²⁵ As an overview of SRs, this umbrella review inherently depends on the methodological quality, scope, and analytical decisions of the included reviews and their underlying randomized trials. Although methodological quality was formally evaluated using the AMSTAR-2 tool,¹³ several reviews presented critical or non-critical weaknesses that may affect the certainty of the synthesized evidence. In addition, moderate overlap of primary randomized trials across reviews (CCA = 9.9%) indicates partial redundancy within the evidence base, meaning that frequently cited trials may disproportionately influence overall conclusions.¹⁸ Differences in population characteristics, intervention protocols, comparator definitions, follow-up duration, and outcome measurements further constrained direct comparability across reviews. Women were not included in the studies included in the umbrella review because the selected SRs exclusively focused on male populations. Additionally, the absence of women participants reflects the existing evidence base rather than a predefined restriction of this review, and therefore the findings cannot be generalized to female populations. This reflects the lack of primary evidence in women, where existing studies rarely assess musculoskeletal outcomes.¹¹ The findings in men may not be considered generalizable to women, as the evidence stands. Nevertheless, the present umbrella review provides a methodologically rigorous synthesis of the literature, strengthened by the systematic evaluation of review quality (AMSTAR-2),¹³ assessment of evidence certainty using the GRADE framework,¹⁶ and explicit quantification of overlap among primary trials.¹⁸

The observed anabolic effects of TRT on body composition are consistent with previous MAs. For instance, Parahiba et al. reported significant improvements in lean body mass and modest gains in muscle strength following testosterone supplementation in middle-aged and older men.⁸ Similarly, Isidori et al. demonstrated increases in fat-free mass and reductions in fat mass, although the magnitude of functional improvements remained limited.¹⁹ These findings are biologically plausible given the well-established role of testosterone in stimulating protein synthesis, satellite cell activation, and muscle fiber hypertrophy.⁵ Furthermore, evidence suggests that individuals with lower baseline testosterone concentrations, particularly those with clinically confirmed hypogonadism, may derive greater anabolic benefit from TRT.⁹ Dose and route of administration may also influence treatment effects. Meta-analytic evidence indicates that intramuscular TRT may produce larger increases in lean mass and muscle strength compared with transdermal formulations, likely reflecting differences in pharmacokinetics and achieved serum testosterone concentrations.²⁰

Despite consistent improvements in body composition, the translation of anabolic effects into functional benefits remains uncertain. Several methodological and clinical factors may contribute to the heterogeneity observed across reviews. First, considerable variability exists in study populations, including differences in baseline testosterone status, age, comorbidities, and physical activity levels, all of which may modify treatment responsiveness. Second, substantial heterogeneity is present in intervention characteristics, including formulation, dosage, treatment duration, and adherence. Third, outcome definitions and measurement methods vary widely across trials, particularly for muscle strength and physical performance, limiting comparability and potentially diluting pooled estimates. Follow-up duration also differed markedly across studies, ranging from 2 weeks to 36 months, although most musculoskeletal outcomes were assessed within the first 6–12 months, which may be insufficient to detect longer-term functional adaptations.

Conclusions

TRT consistently increases lean body mass in older men, but its effects on muscle strength and physical performance remain modest and heterogeneous. Current evidence does not indicate an increased risk of major cardiovascular events when TRT is administered within physiological ranges and under appropriate monitoring. Clinically, TRT may represent a therapeutic option for carefully selected older men with confirmed hypogonadism, particularly to improve body composition. However, the variability in treatment protocols and outcome measures highlights the need for well-designed, long-term RCTs to clarify its impact on functional outcomes and guide clinical decision-making.

Authors’ contributions

All authors have contributed equally to this work and have read and accepted the publisher's published version.

Statement from the institutional review board

Not applicable.

Informed consent

Not applicable.

Funding

This manuscript did not receive external funding.

Conflict of interest

The authors declare no conflicts of interest.