Effects of Hypoglycemic Agents on The Musculoskeletal System: A Comprehensive Review

Introduction

Diabetes is a metabolic disease caused by genetic and environmental factors, which can lead to insulin insensitivity, insulin deficiency and impaired biological function. According to the Global Burden of Diseases, Injuries, and Risk Factors Study, diabetes is the eighth leading combined cause of death and disability worldwide.¹ The prevalence of diabetes is rising globally, and its prevention and control remains an ongoing challenge. A growing body of research suggests that impaired musculoskeletal health is a common complication of diabetes, associated with poor prognosis and increased mortality.

Elevated fracture propensity in diabetic patients cannot be predicted by bone mineral density (BMD) measurements. Instead, elevated fracture risk is attributed to secondary defects in microarchitecture² and bone tissue material properties.³ The results of skeletal biopsies from diabetic patients demonstrate low bone turnover, reduced osteoblast counts, and a significant reduction in the mineralized surface area of the bone cortex.⁴ In patients with diabetes, here is a reduction in serum concentrations of both osteocalcin and N-terminal propeptide of type 1 procollagen (P1NP). Concurrently, the volume of bone tissue mineralization is reduced and the rate of bone formation is decreased.⁵ Furthermore, it has been demonstrated that a state of prolonged hyperglycemia can result in the accumulation of advanced glycation end products (AGEs) within the bone tissue. This, in turn, has the potential to affect the processes of osteoblast differentiation and function, to reduce bone formation, and to increase the risk of fracture.⁶ Furthermore, it is widely acknowledged that dysregulated bioavailability of insulin-like growth factor 1 (IGF-1),^7,8 insulin resistance, enhanced oxidative stress, altered bone mineral and vitamin D homeostasis,⁹ and regional diabetic microvascular disease have a detrimental effect on bone quality in diabetic patients.¹⁰

The mechanism of diabetic skeletal muscle injury is primarily the result of a combination of factors, including elevated glucose levels, insulin resistance, and abnormal release of inflammatory factors. Imbalanced glucose metabolism and excess circulating free fatty acids of diabetic patients have been demonstrated to impair phosphatidylinositol 3-kinase (PI3K)/Akt activity.¹¹ The release of pro-inflammatory cytokines from diabetic patients increases the activity of inhibitor of nuclear factor kappa-light-chain-enhancer of activated B cells (NF-κB)/NF-κB. At the downstream level, the inhibition of protein synthesis-promoting pathways such as mechanistic target of rapamycin (mTOR), and protein degradation-promoting pathways such as forkhead box O (FoxO)¹² and ubiquitin-proteasome system,^13,14 which ultimately result in muscle damage. Furthermore, the reduced oxidative capacity of diabetic muscle is closely related to impaired mitochondrial function. The sustained hypermetabolic condition has been demonstrated to augment the intracellular concentration of reactive oxygen species (ROS), resulting in mitochondrial dysfunction and the subsequent development of insulin resistance within skeletal muscle tissue, which further impairs glucolipid metabolic homeostasis.¹⁵

Apart from the impact of diabetes itself on the musculoskeletal system, the use of anti-diabetic medications also affects multiple aspects in diabetic patients, including BMD, muscle mass, fracture risk, and motility. In recent years, several guidelines have recommended that the heterogeneous characteristics of patients should be taken into account when developing a treatment plan for patients to individualize treatment. Given the close relationship between diabetes, osteoporosis, and sarcopenia, the effects of different anti-diabetic drugs on musculoskeletal health should be taken into account when developing treatment plans for patients. The aim of this review is to provide a comprehensive overview of the effects of diabetes medications on the musculoskeletal system, integrating current research to guide clinical practice and future studies.

Skeletal Effects of Hypoglycemic Agents

Insulin Sensitizers

Metformin

Clinical Studies

Zinman et al¹⁶ found reduced levels of alkaline phosphatase (ALP), P1NP and parathyroid hormone (PTH) in type 2 diabetes mellitus (T2DM) patients treated with metformin (1000mg twice daily) for 12 months. In addition, mildly elevated levels of C-terminal telopeptide of type 1 collagen (CTX) were also seen in female participants.¹⁶ In contrast, a study of male patients with T2DM revealed that treatment with metformin at a dosage of 1000 mg twice a day over a period of 24 weeks did not result in alterations in P1NP, CTX and sclerostin levels.¹⁷ Bilezikian et al¹⁸ found an increase in BMD in both the femoral neck and lumbar spine in postmenopausal women with T2DM treated with metformin (2000 mg total daily dose) for 52 weeks. In a case-control study by Vestergaard et al, metformin treatment was associated with a reduced risk of hip fracture in patients with T1DM and T2DM.¹⁹ Another historical cohort study involving 1964 patients with T2DM showed that metformin treatment reduced the risk of fracture after adjustment for other risk factors.²⁰ However, some studies have reported a neutral effect of metformin on fracture risk. The clinical study conducted by Monami et al²¹ found no effect of metformin on fracture risk in T2DM patients who exposed to metformin for at least 36 months. A Scottish cohort study that included 200000 T2DM patients did not find a significant association between cumulative exposure to metformin and hip fracture incidence during a 9-year follow-up.²²

Preclinical Studies

Kanazawa et al²³ found that metformin induced the differentiation and maturation of osteoblasts through activation of the adenosine 5ʹ-monophosphate-activated protein kinase (AMPK) signaling pathway. In vitro studies have shown that metformin protected osteoblast function by increasing cell proliferation, ALP activity and calcium deposition, as well as inhibiting ROS formation and apoptosis.²⁴ In addition to affecting bone formation, metformin can regulate bone resorption through the receptor activator of nuclear factor κB (RANK)/receptor activator of nuclear factor κB ligand (RANKL)/osteoprotegerin (OPG) pathway.²⁵ In the ovariectomized (OVX) rat model, administration of 100 mg/kg metformin per day by oral intubation reduced the number of tartrate-resistant acid phosphatase (TRAP)-positive cells, elevated OPG and reduced RANKL, thereby increasing BMD.²⁵ In diabetic male rats, the addition of metformin 100 mg/kg/day to drinking water for 2 weeks reduced TRAP activity, resulting in modest improvements in bone microarchitecture.²⁶ On the contrary, some studies showed a neutral effect of metformin on bone metabolism. Wu et al reported that metformin did not affect the osteogenic differentiation of either human or rabbit bone marrow progenitor cells (BMPCs).²⁷ Jeyabalan et al reported that, in female Wistar rats, there was no difference in bone structure or fracture healing ability between the rats with or without metformin (2 mg/mL) added to drinking water.²⁸ Felice et al²⁹ found no significant changes in bone mineral content (BMC), BMD or other trabecular or cortical bone parameters after 3 weeks of metformin (100 mg/kg/day) treatment in male rats.

Thiazolidinediones

Clinical Studies

Measurements of bone turnover markers in T2DM patients treated with thiazolidinediones (TZDs) revealed no differences in P1NP levels, decreases or increases in ALP, no differences or decreases in osteocalcin, and increases in sclerostin.^16,18 However, the findings on the effect of TZDs on BMD were more consistent. A study by Schwartz et al reported that the use of TZDs (troglitazone, pioglitazone, and/or rosiglitazone) was associated with significant bone loss at multiple sites in older women aged 70–79 years with T2DM.³⁰ In patients with T2DM, the fixed-dose combination if rosiglitazone/metformin significantly reduced BMD in the lumbar spine and total hip at week 80 of follow-up, compared with metformin monotherapy.³¹ The study by Home et al included 4447 T2DM patients who receiving metformin or sulfonylurea monotherapy and randomly assigned these patients to addition of rosiglitazone (8 mg per day) or to a combination of metformin and sulfonylurea.³² They found that the fracture rate was 57% higher in the rosiglitazone group than in the control group, with the fracture rate being approximately 82% higher in women and 23% higher in men.³² Kanazawa et al³³ also reported that TZDs treatment was associated with vertebral fracture in postmenopausal women with T2DM, but had no effect on fracture risk in men with T2DM. Schwartz et al³⁴ conducted a prospective study of 6868 patients treated with TZDs in the ACCORD study (rosiglitazone was used by 74% and pioglitazone by 13% of participants) and suggested that the effect of TZDs on fracture risk was reversible in female patients. The fracture risk of female patients taking TZDs for 1~2 years or >2 years was 2.32 and 2.01 times higher than in patients not taking TZDs, respectively.³⁴ And female patients who discontinued TZDs for 1~2 years or >2 years had a 43% lower fracture risk than those who continued to take TZDs.³⁴ In contrast, a meta-analysis by Billington et al suggested that bone loss due to TZDs may be irreversible up to 1 year after discontinuation.³⁵

Preclinical Studies

The mechanism of action of TZDs on bone may be through the activation of peroxisome proliferator-activated receptor γ (PPARγ), and PPARγ can affect osteoblast-specific signaling pathways such as wingless-type MMTV integration site family (Wnt) and IGF-1.³⁶ A recent study showed that rosiglitazone (0.14 mg of rosiglitazone per 1000 mg of food) impaired endochondral fracture healing in C57BL/6J mice by increasing adipogenesis and decreasing osteogenesis of both bone marrow- and periosteum-derived skeletal progenitor cells.³⁷ Lecka-Czernik et al observed rosiglitazone affecting IGF-1 expression in vitro, in mice and in postmenopausal women.³⁸ Several studies have shown TZDs could also affect the bone absorption, the number of osteoclasts and OPG expression.³⁹ In OVX rats, 12 weeks of rosiglitazone treatment (10 mg/kg) was associated with an increase in adipose bone marrow volume, bone loss, and an increase in bone resorption parameters.⁴⁰ Lazarenko et al observed a decrease in the BMD in adult (6 month) and old (24 month) C57BL/6 mice after administration of rosiglitazone (0.14 mg of rosiglitazone per 1000 mg of food), whereas there was no change in young mice (1 month).⁴¹ On the contrary, it has been shown that no significant negative effects on bone turnover markers or BMD were found after 8 weeks of low-dose pioglitazone treatment (3 mg/kg) in a diabetic rat model.⁴² The neutral effect of pioglitazone on bone may be due to the lower affinity of pioglitazone for PPARγ binding than rosiglitazone.

Insulin Secretagogues

Sulfonylureas

Clinical Studies

Kanazawa et al reported that sulfonylureas significantly reduced the incidence of vertebral fracture in postmenopausal women with T2DM.³³ Vestergaard et al reported that sulfonylureas treatment was associated with a significantly lower risk of hip fracture in patients with diabetes compared to untreated patients.¹⁹ However, several other studies have reported neutral or negative effects of sulfonylureas on bone health. In a study that included 1945 patients with T2DM, there was no significant association between sulfonylurea treatment (including glibenclamide, gliclazide, glimepiride and other sulfonylureas) and fracture incidence.²¹ Moreover, a study by Colhoun et al found that hip fracture risk was not associated with long-term sulfonylureas therapy in both male and female T2DM patients.²² The data from the MrOS study showed that the use of sulfonylureas increased the risk of non-vertebral fractures by 66% in older men aged ≥65 years with T2DM.⁴³ In a retrospective cohort study, Rajpathak et al reported that T2DM patients ≥65 years treated with sulfonylureas had a fracture risk that was approximately 46% higher than patients in the non-sulfonylureas group.⁴⁴ Most studies concluded that sulfonylureas increase fracture risk primarily through elevated risk of falls, and not related to direct effects of the drugs on bone metabolism or BMD.⁴⁵

Preclinical Studies

There are relatively few preclinical studies of the effects of insulin secretagogues agents on bone health. Glimepiride has been found to significantly enhance osteoblast proliferation and differentiation through activation of the PI3K/Akt pathway.⁴⁶ Ma et al reported that glimepiride treatment (10 μmol/L) of primary osteoblasts enhanced the mRNA expression of runt-related transcription factor 2 (Runx2), osteocalcin and ALP, and significantly improve osteoblasts differentiation.⁴⁷ Furthermore, in the OVX rats, glimepiride (0.8 mg/kg via stomach tube) promoted bone formation and inhibited estrogen deficiency-induced alterations in skeletal development.⁴⁸ In the presence of estrogen (in un-OVX rats), glimepiride also promoted bone formation, but to a lesser extent.⁴⁸

Glinides

Clinical Studies

Chen et al conducted a retrospective analysis of T2DM patients in Taiwan’s National Health Insurance claims database in from 2000 to 2010. They found that repaglinide increased the risk of fracture, particularly among older female patients aged 65 to 74 years, whereas nateglinide did not increase the risk of fracture in T2DM patients.⁴⁹ In contrast, a network meta-analysis found that nateglinide was associated with a 35% increased risk of fracture in patients with T2DM.⁵⁰ Another study of elderly T2DM patients aged 65 years or over found that glinides treatment was not associated with an increased risk of vertebral or hip fractures compared to metformin.⁵¹ Given that falls due to hypoglycemia are often cited as a cause of fractures with insulin secretagogues, we examined the risk of hypoglycemia with the glinides. Overall, glinides were associated with a lower probability and severity of hypoglycemia than sulfonylureas.⁵² Also, the risk of hypoglycemia was higher with repaglinide than with nateglinide. The probability of hypoglycemic symptoms was higher in the repaglinide group than in the nateglinide group in both studies (7% vs 0% and 17.2% vs 6.1%).^53,54

Preclinical Studies

Few preclinical studies have been conducted on the effects of glinides on bone. A study by Viertel et al revealed when female rats were exposed to repaglinide (30 and 80 mg/kg) during late gestation and/or lactation, skeletal development of the limbs of offspring would be affected, with malformations of the scapula, proximal epiphysis of the humerus and femur.⁵⁵

Incretins

Dipeptidyl Peptidase-4 Inhibitors

Clinical Studies

Dipeptidyl peptidase-4 (DPP-4) inhibitors have beneficial or neutral effects on bone health in diabetic patients. Sitagliptin (100 mg per day, for 12 weeks) was found to reduce ALP and urinary deoxypyridinoline levels in postmenopausal women with diabetes.⁵⁶ In contrast, there were no significant changes in CTX or ALP levels after 1 year of treatment with vildagliptin (100 mg per day) in patients with T2DM.⁵⁷ Gamble et al⁵⁸ found that DPP-4 inhibitors did not increase the risk of fracture compared with sulfonylureas or insulin. A Korean study revealed no significant differences in the incidences of osteoporosis and osteoporotic fractures between DPP-4 inhibitors users and non-users in a T2DM population older than 50 years.⁵⁹ In addition, subgroup analyses by gender and age showed consistent results. An analysis of patients aged 60 years and older found that the use of DPP-4 inhibitors (including sitagliptin, linagliptin, and vildagliptin) was negatively associated with the incidence of fractures.⁶⁰ Another meta-analysis showed that DPP-4 inhibitors not only are safe for fracture in T2DM patients but also that the administration of 100 mg of sitagliptin per day may have protective effects on bone.⁶¹ Tsai et al showed that DPP-4 inhibitors had a neutral effect on fracture risk compared to placebo, but increased fracture risk compared to GLP-1 receptor agonists and sulfonylureas.⁶²

Preclinical Studies

Studies on the effects of DPP-4 inhibitors on bone have yielded mixed results. Gallagher et al found that DPP-4 inhibitor MK-0626 (4 g/kg, supplemented chow) had a neutral effect on osteoblast differentiation, as well as trabecular and cortical bone mass in a male diabetic mouse model.⁶³ Glorie et al reported that sitagliptin (2 g/L, dissolved in drinking water) reduced bone resorption markers, improved the reduction in trabecular number and increase in trabecular spacing in streptozotocin-induced diabetic male Wistar rats.⁶⁴ Wang et al found that sitagliptin (10 mg/kg and 25mg/kg, daily intragastric administration for 4 weeks) applied to OVX mice could effectively reduce bone loss by inhibiting osteoclastogenesis and affecting osteoclast-specific transcription factors.⁶⁵ DPP-4 inhibitors may increase the serum concentration of 25-hydroxyvitamin D3, thereby affecting fracture risk and BMD through vitamin D-related and other related signaling pathways.⁶⁶ The accumulation of crosslinks to AGEs is thought to play an important role in diabetes-related osteoporosis, and vildagliptin was found to inhibit the formation and accumulation of AGEs.⁶⁰ In the diabetic rat, vildagliptin (10 mg/kg/day for 5 weeks) reversed pioglitazone-induced bone loss, increased trabecular volume and BMD, and returned bone turnover markers to normal levels.⁶⁷ However, a study by Sbaraglini et al found that saxagliptin (2 mg/kg/day, administered in drinking water for 3 weeks) reduced the number of metaphyseal osteoblasts and the mean height of the proximal cartilage growth plate in young male non-diabetic Sprague Dawley rats.⁶⁸ Meanwhile, saxagliptin inhibited the osteogenic potential of bone marrow mesenchymal stem cells (BMSCs) in vitro.⁶⁸

Glucagon-Like Peptide-1 Receptor Agonists

Clinical Studies

A clinical trial including 62 newly diagnosed T2DM patients found no significant change in BMD or bone turnover markers after 24 weeks of exenatide (10 μg injection, twice daily) treatment.⁶⁹ In a group of healthy obese women with a body mass index of 34±0.5 kg/m², treatment with liraglutide (1.2 mg/d for 52 weeks) increased levels of the bone formation marker P1NP and prevented bone loss.⁷⁰ A cohort study including 216816 patients with T2DM revealed that no reduction in fracture risk in patients receiving glucagon-like peptide-1 receptor agonists (GLP-1RAs) compared with non-GLP-1RAs users.⁷¹ Su et al analyzed 16 randomized controlled trials including a total of 11206 patients and found on association between the use of GLP-1RAs and risk of fracture.⁷² However, a subgroup analysis showed that liraglutide significantly reduced fracture risk, whereas exenatide increased fracture risk.⁷² A meta-analysis of 39795 T2DM patients showed that treatment with liraglutide and lixisenatide significantly reduced the risk of fracture, whereas other GLP-1 RAs did not show superiority to placebo or other anti-diabetic drugs.⁷³ In addition, these beneficial depend on the duration of GLP-1RAs treatment, with GLP-1RAs treatment for more than 52 weeks significantly reducing fracture risk in T2DM patients.⁷³ The different molecular structures and pharmacokinetic characteristics of exenatide and liraglutide contribute be their differences in action. Liraglutide has 97% homology with GLP-1, and exenatide has only 50% homology with GLP-1.⁶²

Preclinical Studies

It has been shown that GLP-1 analogs increase the expression of type 1 collagen, Runx2, ALP and osteocalcin in MC3T3E1 cells and prevent the differentiation of human BMPCs into adipocytes.⁷⁴ In a study of type 2 diabetic male rats, the administration of subcutaneous injections of exenatide (5 nmol/kg) twice a day for a period of three weeks resulted in a reduction in sclerostin levels and an enhancement in femoral BMD. These findings suggest that GLP-1RAs may have the capacity to stimulate osteogenesis by modulating the Wnt signaling pathway.⁷⁵ Eminov et al reported that treatment with exenatide (3μg/kg/day for 12 weeks) reversed significant decreases in BMD, trabecular number, trabecular thickness, and trabecular area in OVX rats.⁷⁶ Recent studies have shown that liraglutide reduced RANKL-induced osteoclast differentiation, and this effect is mediated by the inhibition of the NF-κB and mitogen-activated protein kinase (MAPK) pathways via GLP-1 receptor.⁷⁷ Treatment of female diabetic mice with liraglutide (0.6 mg/kg/day for 8 weeks) alone or in combination with insulin (liraglutide 0.6 mg/kg/day + insulin 10 units/kg/day for 8 weeks) significantly restored trabecular volumetric BMD and cortical volumetric BMD, and partially restored the trabecular microstructure of the tibia.⁷⁸ In high-fat diet (HFD)-induced diabetic mice, administration of exenatide (25 nmol/kg twice daily for 52 days) significantly improved bone mechanical properties at the organ and tissue levels by altering cortical microarchitecture and bone compositional parameters.⁷⁹ Meanwhile, in vitro studies have shown that liraglutide can activate PI3K/Akt and cyclic AMP/protein kinase A (cAMP/PKA) signaling, thereby promoting osteogenic differentiation and bone formation.⁸⁰

Sodium-Glucose Cotransporter-2 Inhibitors

Clinical Studies

A study involving patients with inadequately controlled T2DM on metformin found that dapagliflozin (10 mg daily for 102 weeks) had no statistically significant effect on bone turnover markers as well as BMD.⁸¹ Bilezikian et al found that canagliflozin (100 mg or 300 mg daily) treatment reduced total hip BMD marginally, but had no effect on BMD of the femoral neck, spine, or radius in T2DM patients aged 55 to 80 years.⁸² Observational studies on sodium-glucose cotransporter-2 (SGLT-2) inhibitors-related fractures showed that the canagliflozin group (51.10%) had the highest incidence of fractures, followed by the dapagliflozin group (24.60%) and the empagliflozin group (23.66%).⁸³ A meta-analysis evaluating the safety of canagliflozin, dapagliflozin, empagliflozin, and ertugliflozin found that all four SGLT-2 inhibitors increased the risk of fractures.⁸⁴ Ueda et al conducted a study on a matched cohort of new users of SGLT-2 inhibitors (dapagliflozin, 61%; empagliflozin, 38%; canagliflozin, 1%) and new users of GLP-1RAs. The study found that SGLT-2 inhibitors treatment did not increase fracture risk in comparison with GLP-1RAs treatment.⁸⁵ A retrospective study of patients treated with SGLT-2 inhibitors (including canagliflozin, dapagliflozin, or empagliflozin) revealed a significant increase in fracture risk during the first 14 days of SGLT-2 inhibitors therapy in comparison with DPP-4 inhibitors treatment, and beyond this initial stage, SGLT-2 inhibitors had no significant effect on fracture incidence.⁸⁶ Another retrospective cohort study by van Dalem et al demonstrated that high cumulative dose (≥6.30 g dapagliflozin equivalents) of SGLT-2 inhibitors increased the risk of major osteoporotic fracture compared with sulfonylureas, suggesting that the duration of SGLT-2 inhibitors treatment may be associated with fracture risk.⁸⁷

Preclinical Studies

Some studies suggested that SGLT-2 inhibitors act on bone primarily through systemic metabolism and mineralization dysregulation, rather than directly at cellular and molecular levels.⁸⁸ Taylor et al suggested that SGLT-2 inhibitors increased phosphate reabsorption in the proximal tubule, induced secretion of PTH and fibroblast growth factor 23, and ultimately stimulated bone resorption.⁸⁹ In the study conducted by Yokono et al, male rats were administered ipragliflozin (10 mg/kg once daily) for a period of 4 weeks in an attempt to treat the obesity that was induced by HFD. The results of this study demonstrated that there was no reduction in lean body mass or BMD in the rats exposed to this treatment.⁹⁰ The administration of canagliflozin (20 mg/kg/day for 10 weeks) partially, but not completely, ameliorated defects in trabecular bone structure and reduction in bone strength in TallyHO T2DM mice.⁹¹ Dapagliflozin (1 mg/kg/day for 28 days) improved bone tissue material properties, bone matrix strength, and matrix biomechanics (maximum load, indentation modulus, and hardness) in male diabetic mice.⁹² Song et al⁹³ showed that canagliflozin, but not dapagliflozin or empagliflozin, increased BMD in mice with T2DM and improved bone microarchitecture. In addition, at high glucose concentrations, 5 mM canagliflozin promoted osteoblast differentiation; however, it is an inhibitor of MC3T3E1 proliferation at concentrations up to 10 mM.⁹³

Muscle Effects of Hypoglycemic Agents

Insulin Sensitizers

Metformin

Clinical Studies

Weight loss is common in diabetic patients during metformin treatment, especially the reduction of fat mass (FM), but no significant change in fat free mass (FFM).⁹⁴ A study of patients with newly diagnosed T2DM found an increase in lean-to-fat ratio in patients treated with metformin (1000 mg twice daily) for 6 months.⁹⁴ A clinical study investigating the effects of metformin (1700 mg per day) in combination with a progressive resistance training (PRT) program in healthy older participants (aged ≥65 years) showed that individuals treated with metformin exhibited muscle hypertrophy, strength gains and maintenance of functional independence.⁹⁵ However, another study showed that metformin administration (1700 mg per day) negatively impacts the hypertrophic response to PRT in healthy older individuals.⁹⁶ In another similar study, treatment with metformin (1000mg twice daily) in healthy older adults who underwent 12 weeks of aerobic exercise training attenuated the expected improvement in maximal oxygen uptake and skeletal muscle mitochondrial respiration from aerobic training.⁹⁷ In terms of physical function, a study including non-diabetic participants aged 60 years and older showed that walking speed was significantly higher in the metformin-treated group (1500 mg daily) than in the control group, but there were no differences in grip strength or myostatin level.⁹⁸

Preclinical Studies

Lyu et al showed that metformin treatment for 5 months could prevent sarcopenia by inhibiting ectopic lipid deposition and the inflammatory response in skeletal muscle in an AMPK-dependent manner in sarcopenic mice.⁹⁹ Oral metformin treatment (320 mg/day) ameliorated HFD-induced muscle atrophy in obese rats. Specifically, metformin treatment resulted in elevated peroxisome proliferator-activated receptor-γ coactivator 1α (PGC-1α) and decreased FoxO3, atrogin-1/muscle atrophy F-box (MAFbx) and muscle RING finger 1 (MuRF-1) in the soleus muscle of HFD-fed rats, which may be involved in PGC-1α-FoxO3 pathway regulation.¹⁰⁰ In premature aging mice model, metformin (1 g/kg dry food for 10 weeks) ameliorated frailty in senescent mice, including muscle mass and physical function.¹⁰¹ However, some studies have shown that metformin is ineffective at restoring muscle mass and strength. One study showed that long-term administration (up to 8 weeks) of metformin did not alter the loss of muscle mass caused by overactivation of mechanistic target of rapamycin complex 1(mTORC1) in the fasted state in aged male mice (23 months of age).¹⁰² Consistent with the findings of the above study, another study showed that metformin administration for 6 weeks had a limited effect on restoring normal metabolism and growth signaling in adipose tissue and muscle in older mice (22 months of age).¹⁰³ Wessels et al showed that both in healthy and diabetic rats, metformin (30, 100 or 300 mg/kg/day) treatment for 2 weeks impairs muscle oxidation in a dose-dependent manner.¹⁰⁴

Thiazolidinediones

Clinical Studies

Sporadic case reports have reported the occurrence of pioglitazone or troglitazone induced acute rhabdomyolysis in T2DM patients.¹⁰⁵ However, results on the effects of TZDs on muscle health are inconsistent. Lee et al have shown that treatment with TZDs significantly reduced muscle mass but did not affect walking speed in older women with diabetes.¹⁰⁶ Yokota et al discovered that male patients suffering from metabolic syndrome experienced a substantial enhancement in skeletal muscle energy metabolism and whole-body aerobic capacity following a 4-month course of oral pioglitazone at a dosage of 15 mg per day.¹⁰⁷ A clinical trial by Shea et al demonstrated that pioglitazone application (30mg/day) to an older (aged 65–79 years) nondiabetic obese male population resulted in subjects losing more thigh muscle volume.¹⁰⁸ Marsh et al revealed that resistance training improves muscle strength in overweight elderly adults (aged 65–79 years), and pioglitazone (30mg/day for 13 weeks) enhances this effect in the women but not in men.¹⁰⁹ Another study found that pioglitazone did not prevent skeletal muscle loss in older non-diabetic overweight or obese individuals on a weight loss program compared to resistance training.¹⁰⁸ Bastien et al found that treatment with rosiglitazone (titrated to a maximal dose of 8 mg/day) significantly increased subcutaneous FM and decreased aerobic exercise capacity in male with T2DM and stable coronary artery disease.¹¹⁰

Preclinical Studies

Administration of rosiglitazone (8mg/kg/day) to male mice with spontaneous T2DM improved the insulin resistance index and abnormal PI3K/Akt signaling and inhibited FoxO3 and downstream atrogin-1/MAFbx and MuRF-1.¹¹¹ It also reduced the activity of caspase-3 in muscle, thereby inhibiting protein degradation. Pioglitazone could inhibit oxidative stress, NF-κB signalling activation and inflammation in skeletal muscle, there by alleviating muscle atrophy in male mice with spinal cord and bulbar muscular atrophy.¹¹² Furthermore, in vitro studies have shown that rosiglitazone reduced skeletal muscle cell apoptosis through a PPARγ-dependent mechanism.¹¹³ It has been reported that rosiglitazone can effectively inhibit the transcription of NF-κB induced by inflammatory mediators, thereby reducing protein degradation in the muscle ducts of cultured skeletal muscle.¹¹⁴ Moreover, TZDs can promote the processing and oxidation of fatty acids in skeletal muscle,¹¹⁵ reduce the triglyceride content in muscle cells, and prevent the development of fat cells in muscle fibers.¹¹⁶ However, pioglitazone (15 mg/kg/day for 15 days) alone or as an adjunct to exercise training failed to improve maximal aerobic speed, endurance and grip strength, and mitochondrial function in young (3 months old) non-diabetic mice.¹¹⁷

Insulin Secretagogues

Sulfonylureas

Clinical Studies

In many studies, sulfonylureas are comparative drugs, and it is difficult to analyze whether sulfonylureas are harmful or beneficial to muscle function. Several studies found patients taking sulfonylureas had a higher frequency of falls than those using other medications,¹¹⁸ but the above studies could not determine what proportion of the increase in fall frequency could be attributed to skeletal muscle dysfunction. According to the United States Food and Drug Administration Adverse Event Reporting System database, muscle atrophic events were significantly associated with the use of glibenclamide, but not with other sulfonylureas.¹¹⁹ Another study found that patients with T2DM who were treated with glimepiride (0.5 mg per day, titrated to 1.0 mg per day over 24 weeks) did not have any significant changes in FM, free bone mass (FBFM) or FBFM/FM.¹²⁰

Preclinical Studies

Given the close relationship between ATP-sensitive K+ (KATP) channels and skeletal muscle function, the inhibitory effects of sulfonylureas and glinides on KATP channels increase the possibility that these drugs may have adverse effects on muscle health.¹²¹ Studies showed that the improvement of glucose uptake in skeletal muscle by gliclazide is associated with the increase of membrane content of glucose transporter 1.¹²² Glibenclamide (150 μM) has also been shown to increase fatigue slow muscle tone in mice and chickens in vitro, through direct effect on skeletal muscle KATP channels.¹²³ Tricarico et al investigated the involvement of KATP channels and glibenclamide in the atrophy of slow-twitch and fast-twitch muscles in vitro in drug-induced muscle atrophy and in vivo in 14-day-hindlimb-unloaded rats. They revealed that glibenclamide enhanced caspase-3 activity in slow-twitch muscle and decreased the ratio of protein concentration to muscle weight.¹²⁴

Glinides

Clinical Studies

According to a study by Mele et al in the United States Food and Drug Administration Adverse Event Reporting System database, muscle atrophy was not observed with repaglinide or nateglinide during the 8-month observation period.¹¹⁹ This difference between glibenclamide and glinides suggested that the drug-induced muscle atrophy may be associated with blockade of KATP channels and increased mitochondrial succinate dehydrogenase activity. However, there have been other studies suggesting a potential atrophic effect of the glinides (repaglinide and nateglinide), and should therefore be used with caution in diabetic patients.¹²⁵

Preclinical Studies

KATP channels are also affected by the inhibitory effect of glinides, which may have a negative impact on muscle health.¹²¹ Mele et al assessed the effects of the insulin secretagogues on the protein content/muscle weight, fibers viability, mitochondrial succinic dehydrogenases activity and channel currents in different types of mice muscle fibers.¹¹⁹ Compared with those in the control group, the protein content/muscle weight were significantly decreased after incubated with sulfonylureas and glinides for 24 hours. Among them, repaglinide and glibenclamide enhanced mitochondrial succinate dehydrogenase activity in atrophic flexor digitorum brevis muscles, indicating that mitochondrial targets may be involved in the atrophy process.

Incretins

Dipeptidyl Peptidase-4 Inhibitors

Clinical Studies

An observational study showed that there were no significant differences in visceral fat area, subcutaneous fat area, and hepatic attenuation index between patients receiving DPP-4 inhibitors and non-DPP-4 inhibitors for T2DM, whereas the skeletal muscle index (SMI) was higher in patients in the DPP-4 inhibitors group.¹²⁶ In a study that included elderly T2DM patients aged 65 years or older, those treated with DPP-4 inhibitors (vildagliptin 100 mg/day or sitagliptin 100 mg/day or saxagliptin 5 mg/day) had better muscle-generating parameters, including FFM, skeletal muscle mass (SMM), gait speed and muscle strength than those treated with sulfonylureas.¹²⁷ A retrospective study showed that adding DPP-4 inhibitors to T2DM therapy improved muscle strength in older patients (mean age 72.6 ± 7.1).¹²⁸ Ishii et al reported that after 24 weeks of sitagliptin treatment (25 mg titrated to 50 mg daily), SMM was significantly increased in overweight/obese T2DM patients.¹²⁰ However, several studies have found that DPP-4 inhibitors could cause musculoskeletal adverse events such as myalgia, limb pain, muscle weakness, and joint pain.¹²⁹ In addition, published case reports indicate that sitagliptin (100 mg/day), when used in combination with statins, is associated with rhabdomyolysis.¹³⁰

Preclinical Studies

Sitagliptin (10 mg/kg/day) has been shown to increase glucose transporter 4 displacement and expression in the skeletal muscle in a rat model of T2DM.¹³¹ In the Klotho-/- mouse model of premature aging, treatment with linagliptin (0.083 g/kg diet) for 4 weeks slowed the progression of aging and improved weight loss and muscle mass loss.¹³² Bianchi et al reported that the vildagliptin analogue PKF275-055 (10 mg/kg for 5 weeks) partially ameliorated muscle damage in streptozotocin-induced diabetic rats, leading to an increase in body and muscle weight.¹³³ In diabetic rats, sitagliptin dose-dependently upregulated circulating irisin and PGC-1α.¹³⁴ In addition, DPP-4 inhibitors may inhibit apoptosis in muscle tissue by increasing glucose utilization and inhibiting the apoptosis of islet cells.¹³⁵ However, high doses of DPP-4 inhibitors can cause acute toxicity in monkeys, particularly in crab-eating macaques, the use of high doses (160 mg/kg) of vildagliptin, could result in extremely high creatine kinase activity and limb skeletal muscle necrosis and intramuscular bleeding.¹³⁶ The researchers noted, although, that this a phenomenon appears to be unique to monkeys, and it is uncertain whether the acute toxicity caused by vildagliptin will occur in humans.¹³⁷

Glucagon-Like Peptide-1 Receptor Agonists

Clinical Studies

In overweight and obese elderly T2DM patients (mean age 68.2 ± 3.9 years), there was no muscle loss after 24 weeks of treatment with liraglutide (up to 3 mg per day).¹³⁸ Li et al showed that in patients with T2DM, after 12 weeks of liraglutide treatment (1.2 mg once daily), there was a significant weight loss but a 2.3% increase in relative total body lean mass (LM), and this change was significantly associated with elevated plasma atrial and brain natriuretic peptide levels.¹³⁹ Exenatide treatment (5 μg twice daily for 4 weeks then 10 μg twice daily for 8 weeks) significantly increased irisin in obese patients with T2DM, and the upregulation of irisin might be a novel mechanism for the beneficial effects of exenatide in T2DM patients.¹⁴⁰ However, there are also studies that have shown that GLP-1RAs did not have a significant effect on skeletal muscle mass. Rondanelli et al found that the change in SMM after 24 weeks of liraglutide (up to 3 mg/d) application in overweight and obese T2DM patients was not significant.¹⁴¹ A randomized clinical trial that included patients with heart failure found no difference in 6-minute walk distance between the liraglutide group (1.8 mg/d) and the placebo group after 6 months of treatment.¹⁴² Another study showed that application of liraglutide in patients with T2DM did not lead to a significant improvement in maximal oxygen uptake, cycle ergometry time and 6-minute walk distance.¹⁴³

Preclinical Studies

GLP-1RAs improves muscle mass and muscle function in a variety of skeletal muscle atrophy model animals.^88,144 In vitro studies have shown that exenatide and liraglutide could promote glucose uptake in skeletal muscle in L6 myotubes in an insulin-independent manner by activating AMPK.¹⁴⁵ Hong et al reported that exenatide inhibited the expression of myostatin and muscular atrophic factors (atrogin-1 and MuRF-1) and promoted the expression of myogenic factors (MyoG and MyoD).¹⁴⁶ This effect is associated with the activation of the PKA and Akt signaling pathways after GLP-1 receptor activation and the inhibition of phosphorylated NF-κB protein expression.¹⁴⁶ In the HFD rat model, liraglutide (200 μg/kg twice daily for 4 weeks) could improve skeletal muscle capillary density and endothelial function by increasing expression of vascular endothelial growth factor and phosphorylation of AMPK.¹⁴⁷ Ren et al revealed that treatment of HFD-fed mice with semaglutide (30 nmol/kg/day) resulted in a significant increase in relative muscle mass, muscle protein synthesis, and muscle function, possibly by altering the metabolism of muscle lipids and organic acids.¹⁴⁸ A similar study also demonstrated that liraglutide (400 μg/kg/d) and semaglutide (60 μg/kg/d) reduced weight gain, excess lipid accumulation, and muscle atrophy in HFD-fed mice, eliminated increases in muscle atrophy markers in skeletal muscle and C2C12 cells, and found that the protective effect of these drugs on skeletal muscle is by affecting the sirtuin1 pathway.¹⁴⁹

Sodium-Glucose Cotransporter-2 Inhibitors

Clinical Studies

A study by Nagai et al demonstrated a significant decrease in both FM and LM in T2DM patients treated with 24 weeks of ipragliflozin (50 mg/day).¹⁵⁰ Similarly, overweight T2DM patients treated with luseogliflozin (up to 5 mg daily) for 12 weeks showed a reduction in FM and SMM.¹⁵¹ A study conducted in Japan showed that obese patients with T2DM treated with tofogliflozin (20 mg/day) for 12 weeks had significantly lower SMI for both men and women.¹⁵² Yamakage et al showed that compared with T2DM patients received non-SGLT-2 inhibitors treatment, those treated with dapagliflozin (5 mg/day for 24 weeks) had lower myostatin levels, with no difference in SMM.¹⁵³ There are few studies on the relationship between SGLT-2 and muscle strength. Sano et al showed that T2DM patients treated with SGLT-2 inhibitors (ipragliflozin 50 mg, luseogliflozin 2.5 mg, or dapagliflozin 5 or 10 mg daily) for 10 weeks had an increase in grip strength in both hands.¹⁵⁴ A clinical study of elderly T2DM patients (mean age 74.1 years) revealed empagliflozin (10 mg daily) treatment for 52 weeks had no significant effect on muscle mass or grip strength.¹⁵⁵

Preclinical Studies

Preclinical studies have shown complex effects of SGLT-2 inhibitors on muscle metabolism. Okamura et al reported that luseogliflozin (added 0.01% w/w in chow) could improve muscle atrophy in db/db mice and inhibit FoxO1 expression.¹⁵⁶ Hata et al showed that db/db mice with dietary luseogliflozin (0.01% w/w in chow) had lower levels of gene expression of FoxO1, tripartite motif containing 63, atrogin-1, and histone deacetylase 4 than mice on a low-carbon diet, suggesting luseogliflozin had beneficial effects on muscle.¹⁵⁷ Otsuka et al showed that canagliflozin (30 mg/kg) affected slow and fast muscles differently in non-diabetic mice, with FoxO1, atrogin-1, and MuRF-1 increased in slow muscles, while fast muscles were not affected.¹⁵⁸ In addition, Tanaka et al showed that canagliflozin (0.03% of diet) combined with exercise training for 4 weeks enhanced the PI3K/Akt pathway in skeletal muscle and maintained skeletal muscle mass in HFD-induced obese mice.¹⁵⁹ Another study similarly showed that hyperglycemia-induced overactivation of the c-Jun N-terminal kinase (JNK)/Smad2 signaling pathway could lead to poor exercise response in mice, whereas treatment with canagliflozin (30 mg/kg/day) could normalize this pathway.¹⁶⁰ Nambu et al investigated the effects of empagliflozin (300 mg/kg of food) in mice with heart failure after myocardial infarction and found that the drug increased ketone body levels while improving their exercise tolerance, but did not improve spontaneous physical activity, skeletal muscle mass, or skeletal muscle strength.¹⁶¹ However, Lv et al showed that muscle regeneration and myoblast differentiation required increased mitochondrial leucyl tRNA synthetase (LARS2) expression, but canagliflozin reduced LARS2 expression in muscle stem cells, which negatively affected the recovery of ischemic skeletal muscle in mice.¹⁶²

Conclusions

In this review, we point out how hypoglycemic drugs play an important role in influencing musculoskeletal health (Table 1 and Table 2) and their mechanisms of action (Figures 1 and 2). The musculoskeletal health of patients with diabetes is a frequently overlooked aspect of their health, and these patients are more susceptible to a range of adverse outcomes such as fractures, falls and reduced mobility due to the disease itself. At the same time, the use of antidiabetic medications has a differential impact on the presence of musculoskeletal health. Consequently, musculoskeletal effects need to be considered in addition to glycemic status when developing an individualized diabetes treatment plan. Drugs that further negatively affect musculoskeletal health (eg TZDs, SGLT-2 inhibitors) should be avoided in patients at high risk or in patients with osteoporosis/sarcopenia. At the same time, pharmaceuticals that can benefit musculoskeletal health in diabetic patients (eg metformin) can reduce the risk of osteoporosis or sarcopenia while controlling blood glucose. In the future, as combination therapies become more common in the management of diabetes, the potential of these approaches to affect musculoskeletal health in diabetes management will require careful and ongoing evaluation. Nevertheless, there is still a lack of definitive clinical data to guide the rational choice of antidiabetic drugs, with conflicting results from different studies, and a lack of research on muscle strength and physical function. Therefore, large cohort studies are needed to clarify the effects of hypoglycemic drugs on musculoskeletal health in order to guide clinical use.

Table 1 Summary of Skeletal Effects of Hypoglycemic Agents: Clinical Evidences

Table 2 Summary of Muscle Effects of Hypoglycemic Agents: Clinical Evidences

Figure 1 Mechanisms by which hypoglycemic agents might impact skeleton health. ↑: increase; ↓: decrease. Green color indicates benefits for skeleton health; red color indicates harmful to skeleton health. Abbreviations: TZDs, thiazolidinediones; DPP-4 inhibitors, dipeptidyl peptidase-4 inhibitors; GLP-1RAs, glucagon-like peptide-1 receptor agonists; SGLT-2 inhibitors, sodium-glucose cotransporter-2 inhibitors; AMPK, adenosine 5-monophosphate-activated protein kinase; BMPCs, bone marrow progenitor cells; ROS, reactive oxygen species; RANKL, receptor activator of nuclear factor κB ligand; OPG, osteoprotegerin; GLP-1, glucagon-like peptide-1; GLP-2, glucagon-like peptide-2; GIP, glucose-dependent insulinotropic polypeptide; AGEs, advanced glycation end products; PTH, parathyroid hormone; FGF23, fibroblast growth factor 23; PI3K, phosphatidylinositol 3-kinase; BMSCs, bone marrow mesenchymal stem cells; PPARγ, peroxisome proliferator-activated receptor γ; IGF-1, insulin-like growth factor-1.

Figure 2 Mechanisms by which hypoglycemic agents might impact muscle health. ↑: increase; ↓: decrease. Green color indicates benefits for skeleton health; red color indicates harmful to muscle health. Abbreviations: TZDs, thiazolidinediones; DPP-4 inhibitors, dipeptidyl peptidase-4 inhibitors; GLP-1RAs, glucagon-like peptide-1 receptor agonists; SGLT-2 inhibitors, sodium-glucose cotransporter-2 inhibitors; ROS, reactive oxygen species; AMPK, adenosine 5-monophosphate-activated protein kinase; mTOR, mammalian target of rapamycin complex; PCG-1α, peroxisome proliferator-activated receptor-γ coactivator 1α; FoxO3, forkhead box O3; MuRF-1, muscle RING finger 1 MuRF-1; MAFbx, muscle atrophy F-box; GLUT4, glucose transporter 4; LARS2, mitochondrial leucyl tRNA synthetase; FoxO1, forkhead box O1; PI3K, phosphatidylinositol 3-kinase; KATP, ATP-sensitive K+; SIRT1, sirtuin1.

Furthermore, the potential for some hypoglycemic drugs and the mechanisms by which they interact with the musculoskeletal system could be used as potential targets to develop new treatments for osteoporosis/sarcopenia is worthy of consideration (eg metformin, GLP-1RAs). There have been a number of preclinical studies using metformin in the treatment of osteoarthritis,¹⁶³ diabetic periodontitis,¹⁶⁴ and sarcopenia,¹⁶⁵ demonstrating the potential of anti-diabetic medication in the treatment of musculoskeletal diseases. The majority of the extant studies relevant to this field have focused on patients with diabetes, and there is a need for future studies to target those without diabetes as well, including those with impaired glucose tolerance and those with normal glucose metabolism. Given the close links between musculoskeletal function, glucose and fat metabolism in health and disease, these marketed medications for diabetes deserve a more thorough review as potential avenues for the treatment of osteoporosis and sarcopenia.

With the increasing life expectancy of the population, more and more elderly people are suffering from a variety of age-related diseases at the same time, and the development of new uses of these existing drugs is expected to be a new treatment for the geriatric syndrome. This review will help clinicians select appropriate hypoglycemic medication to avoid further damage to the patient’s musculoskeletal system and even protect bone and muscle quality.