Bone disease in diabetes mellitus

INTRODUCTION — The skeleton is increasingly recognized as a site of end-organ damage in diabetes. The effects of diabetes on bone are complex and an area of active investigation. While most studies demonstrate that fracture risk is increased in both type 1 and type 2 diabetes, bone mineral density (BMD), as measured by dual-energy x-ray absorptiometry (DXA), may not reflect bone fragility or accurately predict fracture risk. This is particularly true in individuals with type 2 diabetes, in whom BMD is often normal or elevated compared with those without the disease. Our understanding of the extent to which the metabolic abnormalities of diabetes impact bone metabolism, structure, quality, and BMD is evolving. This topic will review the effects of type 1 and type 2 diabetes on bone (table 1). The musculoskeletal complications of diabetes are discussed separately. (See "Overview of the musculoskeletal complications of diabetes mellitus".)

Derangements in bone metabolism that occur in diabetes-related kidney disease and other forms of chronic kidney disease are also discussed separately. (See "Overview of chronic kidney disease-mineral and bone disorder (CKD-MBD)".)

BONE FRACTURE

Increased risk of fracture — The majority of epidemiologic studies have demonstrated that both type 1 and type 2 diabetes contribute to bone fragility and increase the risk of fractures (table 1). As examples:

●Type 1 diabetes – In meta-analyses of observational studies in patients with and without type 1 diabetes, patients with type 1 diabetes had an increased risk of hip, lumbar spine, and all fractures across the lifespan [1-3]. In one meta-analysis, fractures occurred in 4.8 percent of adults aged 18 to 50 years with type 1 diabetes versus 2 percent without diabetes (pooled relative risk [RR] for all fracture 1.88, 95% CI 1.52-2.32) [2].

●Type 2 diabetes – In meta-analyses of observational studies in patients with and without type 2 diabetes, patients with type 2 diabetes had an increased risk of hip, vertebral, peripheral, and overall fracture (RRs ranging from 1.16 to 1.37) [4-6]. Although both men and women with type 2 diabetes appear to have increased risk of nonvertebral and hip fracture [7,8], data from a prospective cohort study (men aged ≥65 years) showed no difference in the prevalence and incidence of vertebral fractures between men with and without type 2 diabetes [9].

Type 1 and type 2 diabetes may not confer equivalent fracture risk. Although both are associated with a relatively increased risk of hip fracture compared with individuals without diabetes, meta-analyses and systematic reviews suggest that the magnitude of increase in hip fracture risk attributable to the presence of diabetes is greater in individuals with type 1 diabetes (RR 6.3-6.94) than in those with type 2 diabetes (RR 1.38-1.7) [8,10]. This differential magnitude of increased risk may, in part, reflect the higher prevalence of diabetes-related risk factors for fracture in type 1 compared with type 2 diabetes (eg, vascular complications, longer duration of disease). (See 'Diabetes-related risk factors' below.)

Prolonged fracture healing also has been described in both type 1 and type 2 diabetes [11,12]. In a systematic review of studies evaluating complications of fracture healing, patients with diabetes and surgically treated lower extremity fractures had increased rates of malunion, infection, and reoperation compared with patients without diabetes [12].

Diabetes-related risk factors — Population-based studies have found associations between risk of fractures (nonvertebral, vertebral, and hip) and longer diabetes duration, poor glycemic management, diabetic retinopathy, nephropathy, neuropathy, and need for insulin therapy [13-19]. For example, in a retrospective, population-based cohort study of almost three million veterans (900,402 with diabetes), the risk of any clinical fracture (RR 1.22, 95% CI 1.21-1.23) and hip fracture (RR 1.21, 95% CI 1.19-1.23) was increased in veterans with diabetes [20]. Peripheral neuropathy, cardiovascular disease, and congestive heart failure were associated with an increased risk of fracture. In a cross-sectional study in individuals with type 1 diabetes enrolled in the Epidemiology of Diabetes Interventions and Complications (EDIC) study, higher mean A1C and the presence of kidney disease were inversely associated with bone mineral density (BMD) at the total hip [21], although fracture data were not reported.

PATHOGENESIS — Increased bone fragility in diabetes is multifactorial. In broad terms, micro- and macrovascular complications of the disease, hormonal alterations, and diabetes treatments all contribute to diminished bone strength (table 1).

Type 1 versus type 2 diabetes — Some mechanisms underlying bone fragility are common to both type 1 and type 2 diabetes, whereas others are distinct (table 1). Type 1 diabetes is usually diagnosed prior to the age of peak bone accrual and is therefore more likely to affect peak bone mass. Further, as it generally develops at a younger age than type 2 diabetes, type 1 diabetes imparts longer exposure to dysregulated glucose homeostasis, with possible sequelae including earlier onset of microvascular complications, increased deposition of advanced glycation end products (AGEs), and greater glycation of collagen. Indeed, in individuals with type 1 diabetes, glycemia is inversely associated with bone mineral density (BMD) [21]; however, clinical studies have not yet established whether improved insulin delivery systems and attendant improvements in long-term glycemia reduce the risk of bone fragility and fracture.

Contributing factors

Diabetes-related complications — Micro- and macrovascular complications of diabetes can adversely affect bone health. Microvascular disease may involve the bone microvasculature and alter the bone marrow microenvironment where bone progenitor cells reside, with shifts in cellular differentiation to adipocytes rather than osteoblasts and consequent increase in bone marrow adiposity [22]. In some studies, marrow adiposity was associated with glycated hemoglobin (A1C) level or fractures [23-26].

In a study using high-resolution peripheral quantitative computed tomography (HR-pQCT) to evaluate patients with type 2 diabetes (with and without microvascular complications) and controls, only the group with microvascular complications had cortical bone deficits [27]. A similar study in patients with type 1 diabetes showed that the presence of microvascular complications was associated with deficits in both cortical and trabecular bone [28]. Another HR-pQCT study suggested that microvascular disease may play a role in increased cortical porosity in patients with type 2 diabetes [29]. (See 'Bone quantity and quality' below.)

Patients with diabetes are at increased risk for falls compared with those without diabetes, due in part to hypoglycemic events but also to autonomic and peripheral neuropathy. Retinopathy with visual impairment may further compromise patient safety and lead to gait instability. An increased risk of falling and fall-related fractures has been reported in older women with diabetes, although these factors do not fully explain the increased fracture risk observed [30,31]. Neuropathy may also cause localized bone loss, which may increase the risk of fracture at the foot and ankle. Diabetic kidney disease can lead to abnormal bone remodeling and/or mineralization defects, termed chronic kidney disease-mineral and bone disorder (CKD-MBD). CKD-MBD is discussed elsewhere. (See "Overview of chronic kidney disease-mineral and bone disorder (CKD-MBD)".)

Diabetes therapy — Fracture risk may also be related to diabetes pharmacotherapy. Thiazolidinediones and the sodium-glucose cotransporter 2 (SGLT2) inhibitor, canagliflozin, have been reported to increase the risk of fractures. The adverse effects of these drugs are reviewed separately. (See "Thiazolidinediones in the treatment of type 2 diabetes mellitus", section on 'Skeletal fractures' and "Sodium-glucose cotransporter 2 inhibitors for the treatment of hyperglycemia in type 2 diabetes mellitus", section on 'Skeletal fragility'.)

Insulin and other hormones — Insulin has anabolic effects on bone, but the effects of elevated endogenous levels in type 2 disease (due to peripheral insulin resistance) or the effects of exogenous insulin use on bone health are unclear. Insulin requirements in type 2 diabetes often mark longstanding or progressive diabetes that can be accompanied by vascular complications and other comorbidities that contribute to fracture risk; such comorbidities therefore may confound studies evaluating the effects of insulin treatment on fracture. The anabolic effects of insulin may be mediated through the insulin-like growth factor 1 (IGF-1) pathway, and in type 1 diabetes, low levels of insulin and IGF-1 may impair osteoblast function [32].

In postmenopausal women with type 2 diabetes, serum IGF-1 levels were inversely associated with the presence of vertebral fractures [33]. In men with diabetes, hypogonadism may also contribute to bone fragility. Central hypogonadism is common in men with both type 1 and type 2 diabetes, even in the absence of obesity [34]. Hypogonadism, in turn, may promote further adverse changes in both body composition (decreased lean mass, increased fat mass) and glucose homeostasis. Testosterone is important for bone remodeling by both directly promoting bone formation and indirectly preventing bone resorption through effects mediated after its aromatization to estradiol [35]. Thus, central hypogonadism in men with diabetes may contribute to impaired bone strength [36].

Finally, vitamin D deficiency is common in adults with both type 1 and type 2 diabetes [37]. Diabetes-related kidney disease can lead to impaired synthesis of bioactive vitamin D, with detrimental effects on bone metabolism. Skeletal consequences of vitamin D deficiency can include osteomalacia and secondary hyperparathyroidism. (See "Vitamin D and extraskeletal health", section on 'Diabetes' and "Epidemiology and etiology of osteomalacia", section on 'Vitamin D deficiency and resistance'.)

Hyperglycemia and advanced glycation end products — In chronic hyperglycemia, excess glucose combines with free amino acids on circulating or tissue proteins. This nonenzymatic process initially forms reversible, early glycation products and, later, irreversible AGEs. The accumulation of AGEs in bone matrix alters collagen structure; impacts osteoblast, osteoclast, and osteocyte number and function; and increases bone marrow adiposity and local cytokine production; all of these factors contribute to aberrant bone remodeling, bone embrittlement, and reduced bone quality [21,23,38-40].

Bone quantity and quality — Both bone quantity and quality help determine bone strength and fracture risk. Clinically, bone quantity can be measured by dual-energy x-ray absorptiometry (DXA) as areal BMD and, less commonly, from computed tomography (CT) as volumetric BMD. Parameters of bone quality are more challenging to quantify and usually measured only in research settings. Measures of bone quality include bone structure (microarchitecture), material properties, and remodeling.

●Bone density – Compared with the reference population, BMD is usually lower in patients with type 1 diabetes and normal or increased in patients with type 2 diabetes [23,41,42]. As examples:

•In a study of 60 adolescents with type 1 diabetes, total body and lumbar spine BMD Z-scores were significantly lower compared with the age- and sex-matched reference population [43]. The onset of diabetes in adolescence may result in a decreased peak bone mass and a decrease in BMD due to a failure to acquire endosteal bone during growth [44].

•In a study of 65 adults (mean age 62.6 years) with longstanding (over 50 years), well-managed type 1 diabetes from a specialist center, Z-scores at the lumbar spine, total hip, femoral neck, and radius were normal [45].

•In a study of older, well-functioning patients with type 2 diabetes, hip and total body BMD were higher in Black and White adults with diabetes compared with their appropriate controls [46].

In some individuals with diabetes, therefore, BMD does not appear to reflect fracture risk. Poor bone quality likely contributes to bone fragility and increased risk of fracture independent of measured BMD [10].

●Bone structure – Type 1 and type 2 diabetes may have different effects on bone microarchitecture depending on bone type (ie, trabecular versus cortical) [47]. Trabecular bone score (TBS) is an index of bone microarchitecture derived from lumbar spine DXA, and it serves as an indirect measurement of trabecular bone quality [48]. Adults with type 1 and type 2 diabetes have a lower TBS than individuals with normal glycemia [48-51]. In addition, TBS is a significant predictor of fracture risk independent of bone density [49,52]. Research using HR-pQCT suggests that a deficit in cortical bone, resulting in increased cortical porosity, may contribute to fracture risk in type 2 diabetes and in part explain the reduction in bone strength that is not detected by DXA [53,54].

●Bone material properties – Abnormal bone tissue composition (collagen and mineral) may contribute to abnormal bone mechanical properties and thus increased fragility. Compared with bone samples from healthy individuals, those collected from individuals with type 2 diabetes had increased total mineral content and increased bone stiffness yet lower resistance to nanoindentation [55]. Similarly, an analysis of bone samples from a small cohort of patients with type 1 diabetes showed that trabecular bone mineral content was greater in individuals who had experienced fracture compared with those who had not fractured [38]. In both types of diabetes, bone levels of AGEs are increased and may contribute to deficiencies in bone strength. (See 'Hyperglycemia and advanced glycation end products' above.)

●Bone remodeling – Low bone turnover, with a reduction in unmineralized bone matrix, and increased collagen glycation may contribute to increased fragility of diabetic bone. In histomorphometric and biochemical studies in patients with diabetes, bone turnover is low with a reduction in both bone formation and, to a lesser degree, bone resorption [23,41,56,57]. In a systematic review and meta-analysis of 66 studies evaluating bone metabolism in patients with diabetes, markers of both bone formation (osteocalcin) and resorption (C-telopeptide) were decreased in patients with type 1 and type 2 diabetes compared with controls [56]. (See "Bone physiology and biochemical markers of bone turnover".)

Bone turnover also may be affected by the complications of diabetes (eg, kidney failure, which can lead to a complex skeletal pathophysiology including low turnover or adynamic bone disease) [58-60]. However, decreased bone formation can be demonstrated in patients with type 1 diabetes before the onset of clinical kidney disease, when estimated glomerular filtration rate (eGFR) is >60 mL/min/1.73 m² [61-63].

APPROACH TO EVALUATION AND MANAGEMENT

Clinical evaluation — The evaluation of fracture risk in patients with diabetes mellitus is similar to patients without diabetes, with some caveats, and includes assessment of the following:

●Assessment of clinical risk factors for fracture (table 2).

●Bone mineral density (BMD), with cautious interpretation. The use of BMD to predict fracture risk in patients with diabetes (particularly type 2) is problematic because the risk of fracture may be higher than that predicted by BMD T-score alone [64-66]. (See 'Bone quantity and quality' above.)

●Fracture probability using a risk assessment tool, such as FRAX. For a given FRAX score, however, the risk of fracture may be higher in patients with diabetes than in those without diabetes [67,68]. The FRAX algorithm does not include diabetes as a risk factor; notably, one multivariate analysis suggested that the effect of type 2 diabetes on fracture risk is equivalent to adding 10 years to the patient's age using FRAX [69]. An alternative strategy to capture this increased fracture risk is entering "yes" for the presence of rheumatoid arthritis in the FRAX algorithm [70].

Whereas fracture risk conferred by type 1 diabetes may be estimated by answering "yes" to "secondary cause of osteoporosis" in FRAX, this strategy will not increase the predicted risk of fracture if femoral neck bone density is included in the model. Limited data exist for the predictive utility of FRAX in type 1 diabetes. In one study in adults with type 1 diabetes, FRAX score was associated with risk of major osteoporotic fracture only when "secondary cause of osteoporosis" was indicated and BMD omitted from the algorithm [71].

More research is required in additional population cohorts worldwide before diabetes as a risk factor can be included. Tools that may enhance estimation of fracture risk in patients with diabetes include the trabecular bone score (TBS), which is derived from lumbar spine dual-energy x-ray absorptiometry (DXA) images, and TBS-adjusted FRAX [65].

Fracture risk assessment, including a description of FRAX, TBS, and the clinical applications of fracture risk assessment, is reviewed in detail elsewhere. (See "Osteoporotic fracture risk assessment", section on 'Assessment of fracture risk' and "Osteoporotic fracture risk assessment", section on 'New and emerging technologies'.)

Treatment

Lifestyle measures and diabetes pharmacotherapy — Prevention of chronic hyperglycemia may benefit skeletal health by reducing advanced glycation end products (AGEs), glycation of collagen, and risk of microvascular complications. Glycemic targets should be individualized, balancing the demonstrated benefits for prevention and delay of microvascular complications with the risk of hypoglycemia. Both lifestyle modifications and pharmacotherapy are important interventions to optimize glycemic management as well as overall health in patients with type 1 and type 2 diabetes. (See "Overview of general medical care in nonpregnant adults with diabetes mellitus", section on 'Blood glucose monitoring and target A1C'.)

●Lifestyle measures – The general recommendations regarding healthy lifestyle (exercise to improve muscle strength and reduce the risk of falls, calcium and vitamin D supplementation) are applicable to older adults with diabetes. (See "Calcium and vitamin D supplementation in osteoporosis", section on 'Optimal intake'.)

●Diabetes pharmacotherapy – Thiazolidinediones and sodium-glucose cotransporter 2 (SGLT2) inhibitors (particularly canagliflozin) have been associated with an increased risk of fracture. In patients at high risk for fracture (eg, low bone density, previous fracture), thiazolidinediones should not be used, except in the rare circumstance that other glucose-lowering agents cannot be used, and SGLT2 inhibitors should be used with caution. Although bone fractures have been reported to occur more frequently only in patients taking canagliflozin, based on putative mechanisms, other SGLT2 inhibitors may also reduce bone mass and increase bone fractures. (See 'Diabetes therapy' above.)

Osteoporosis pharmacotherapy

●Patient selection for pharmacotherapy – In addition to optimization of glycemic management, patients at high risk for fracture should receive pharmacologic therapy to improve bone health. Similar to patients without diabetes, patients with type 1 or type 2 diabetes at high risk for fracture include those with a history of fragility fracture or osteoporosis based upon BMD measurement (T-score ≤-2.5). For patients with T-scores between -1.0 and -2.5, we treat patients with a 10-year probability of fracture (using FRAX calculations adjusted to capture diabetes-related fracture risk) that meets country-specific intervention thresholds. (See 'Clinical evaluation' above.)

In patients with diabetes who do not meet these criteria for pharmacotherapy, clinical judgment should be used if additional risk factors for fracture are present that are not captured in the FRAX model or bone density reporting. For example, patients at high fall risk, those actively losing bone, or those with immobility or debilitation may still merit bone pharmacotherapy [72,73]. Treatment intervention thresholds based on FRAX are reviewed separately. (See "Overview of the management of low bone mass and osteoporosis in postmenopausal women", section on 'Patient selection' and "Treatment of osteoporosis in men", section on 'Patient selection' and "Osteoporotic fracture risk assessment", section on 'Fracture risk assessment tool'.)

●Choice of pharmacologic therapy for fracture prevention – In the absence of studies specifically targeting osteoporosis therapy in individuals with diabetes, treatment should follow general guidelines for the treatment of osteoporosis. (See "Overview of the management of low bone mass and osteoporosis in postmenopausal women" and "Treatment of osteoporosis in men".)

Few studies specifically address the pharmacologic treatment of osteoporosis in individuals with diabetes. In general, individuals with diabetes constitute a small fraction of participants in clinical trials of osteoporosis treatments, necessitating analyses of pooled data. In one systematic review of nine studies (predominantly subgroup or post hoc analyses of trials evaluating alendronate, risedronate, raloxifene, and teriparatide for the treatment of osteoporosis), there were similar increases in bone density and reductions in vertebral (alendronate, raloxifene) and nonvertebral (teriparatide) fracture risk in patients with and without diabetes [74]. The majority of the patients in the trials had type 2 diabetes. In a post hoc analysis of the FREEDOM and FREEDOM extension studies comparing denosumab with placebo in patients with diabetes, denosumab significantly increased BMD and decreased vertebral fracture rate compared with placebo [75]. In a more recent analysis, pooled individual patient data (n = 96,385, 6.8 percent with type 2 diabetes) from trials of antiresorptive osteoporosis therapies (bisphosphonates, selective estrogen receptor modulators, denosumab, odanacatib) demonstrated that these medications were effective for reducing fracture risk in all participants, irrespective of diabetes status [76].

Serious adverse events from bisphosphonate treatment do not appear increased in patients with diabetes [77-79]. Nonetheless, bisphosphonate use may be limited by the presence of diabetic kidney disease, as bisphosphonates are generally not recommended for individuals with creatinine clearance below 30 to 35 mL/min. This topic is reviewed elsewhere. (See "Risks of bisphosphonate therapy in patients with osteoporosis" and "Bisphosphonate therapy for the treatment of osteoporosis", section on 'Use in chronic kidney disease'.)

SUMMARY

●Fracture risk – Fracture risk is increased in individuals with both type 1 and type 2 diabetes, possibly due to factors in addition to bone mineral density (BMD), such as duration of diabetes, glycemic management, diabetes-related complications, bone quality, treatment, and risk of falling (table 1). (See 'Bone fracture' above and 'Contributing factors' above.)

●Bone quantity and quality – BMD, as measured by dual-energy x-ray absorptiometry (DXA), is lower in patients with type 1 diabetes and normal or increased in patients with type 2 diabetes, although deficits in bone quality may contribute to increased fracture risk for all patients with diabetes. (See 'Bone quantity and quality' above.)

The trabecular and cortical microarchitecture of bone may be altered in patients with type 1 and type 2 diabetes and predispose patients to bone fragility. Abnormal bone tissue composition (collagen and mineral) may contribute to abnormal bone mechanical properties and thus increased fragility (table 1). (See 'Bone quantity and quality' above.)

●Treatment

•Diabetes pharmacotherapy – Prevention of chronic hyperglycemia may benefit skeletal health by reducing advanced glycation end products (AGEs), glycation of collagen, and risk of microvascular complications.

Thiazolidinediones and sodium-glucose cotransporter 2 (SGLT2) inhibitors (particularly canagliflozin) for the treatment of diabetes have been associated with an increased risk of fracture. In patients at high risk for fracture (eg, low bone density, previous fracture), thiazolidinediones should not be used (except in the rare circumstance that other glucose-lowering agents cannot be used), and SGLT2 inhibitors (particularly canagliflozin) should be used with caution. (See 'Diabetes therapy' above and 'Lifestyle measures and diabetes pharmacotherapy' above.)

•Osteoporosis pharmacotherapy – Osteoporosis pharmacotherapy (antiresorptive and anabolic) reduces fracture risk in patients with type 1 and type 2 diabetes. In the absence of studies specifically targeting osteoporosis management in individuals with diabetes, recommendations should follow general guidelines for the treatment of osteoporosis as well as patient-specific considerations (table 2). (See 'Osteoporosis pharmacotherapy' above and "Overview of the management of low bone mass and osteoporosis in postmenopausal women" and "Treatment of osteoporosis in men".)

ACKNOWLEDGMENT — The UpToDate editorial staff acknowledges Lesley D Hordon, MD, who contributed to earlier versions of this topic review.