The Intriguing Intersection of Type 2 Diabetes, Obesity-Related Insulin Resistance, and Osteoarthritis.

Osteoarthritis (OA) is characterized by degeneration of hyaline joint cartilage, hypertrophic changes to the joint capsule, and damage to underlying bone with subchondral remodeling and osteophyte formation. OA is a prominent cause of chronic disability secondary to pain and altered joint function, which result from distinct pathologic changes in the context of patient-centered psychosocial factors. While the underlying pathogenesis of OA is likely multifactorial, the intersection between obesity-related insulin resistance, diabetes mellitus, and OA is a major focus of current research. Overall, insulin promotes type II collagen and proteoglycan production, and inhibits cartilage breakdown. And although mechanical forces play a prominent role in OA (to varying degrees depending on the joint involved), inflammatory mediators are also key drivers of joint tissue destruction. However, classic cellular inflammation (as observed in other forms of inflammatory arthritis) is not typically seen in OA. Instead, a macrophage-predominant, innate immune response in the joint synovium occurs, leading to the release of proinflammatory cytokines and chemokines and activation of proteolytic enzymes that damage the joint extracellular matrix (1). In addition to this inflammatory component, epigenetic modifications (including downregulation of specific microRNA molecules) occur in normal compared with OA cartilage (2).

Type 2 diabetes mellitus (T2DM) is also, in part, driven by a low-grade, chronic inflammatory state. Systemic and tissue-specific inflammation strongly contribute to insulin resistance, one of the two major pathogenic factors leading to T2DM. Approximately 1 of 10 adults in the United States have diabetes, with the vast majority (90% to 95%) classified as having obesity-associated T2DM, and over two-thirds are overweight or obese (3). Particularly concerning, rates of overweight/obesity are increasing in younger individuals, with one-third of American youths being classified as overweight. In accordance with this increase, there has been a striking concomitant increase in T2DM in this age demographic (4). This is an especially salient point given the risk of multiple comorbidities that rise with higher adiposity and, importantly, its duration. Of note, OA incidence is up to 5-fold higher in those who are obese, and its onset and severity is closely linked with weight status and duration of obesity (5). Therefore, understanding the relationship between T2DM and insulin resistance with OA is oftentimes confounded by the degree of adiposity, making it difficult to discern independent, causal effects on OA pathology.

Previous longitudinal studies demonstrate that both T2DM and insulin resistance (6) have independent effects on OA risk, in part due to the aforementioned convergence of proinflammatory cytokine expression and downregulated synthesis of structural proteins in articular cartilage. In one study, serum insulin concentrations were elevated in 82% of patients with OA; and paradoxically, high insulin concentrations led to articular cartilage degeneration (7). In the accompanying study by Zaharia et al (8), the authors investigate whether insulin sensitivity differs between those with T2DM versus controls, with or without OA, and further delineate whether any features of musculoskeletal impairment are associated with insulin resistance and microRNA expression changes. They employ comprehensive measurements of insulin sensitivity (intravenous glucose tolerance testing and a hyperinsulinemic-euglycemic clamp procedure), body composition (bioelectrical impedance), and detailed assessments of joint mobility (goniometer), muscle strength (handgrip dynamometer), and balance, among others. The authors conclude that patients with normal glucose tolerance and OA still exhibit lower insulin sensitivity, nerve conduction velocity, and balance compared with those without OA. They make similar observations in subjects with T2DM, in which those with OA had lower insulin sensitivity and more advanced musculoskeletal impairment (range of motion) versus those without OA. Overall, OA-related symptoms were associated significantly with the degree of insulin resistance.

The impact of T2DM on circulating microRNA levels in the presence of OA is largely unknown. A growing body of literature provides evidence for a large number of epigenetic changes in OA, including microRNAs that affect gene expression of inflammatory cytokines and matrix proteinases. T2DM also leads to a myriad of epigenetic changes, many of which relate to expression of proinflammatory mediators. This intersection of epigenetics occurring in 2 inflammatory conditions is intriguing and warrants further investigation. Although the study by Zaharia et al (8) finds an overall upregulation of microRNA in patients with T2DM and downregulation in patients with OA, in accordance with prior studies, there is no difference in T2DM patients based on the presence or absence of OA, suggesting for the first time that diabetes may mask arthritis-related microRNA expression profiles.

The above findings not only add to the existing body of literature, but also expand upon our current knowledge base with substantially improved clinical phenotyping of insulin sensitivity and musculoskeletal function and detailed microRNA analyses. They demonstrate not only insulin resistance in OA patients without frank T2DM, but that T2DM is an independent factor associated with musculoskeletal impairment, even after adjustment for age, sex, and body mass index (BMI). These results imply a strong, independent metabolic connection between insulin resistance and joint function. In addition, selected inflammatory mediators were related to insulin resistance and OA (most prominently leptin and interleukin-1 receptor antagonist), strengthening evidence for a connection between an overall, systemic proinflammatory environment in both T2DM and OA. Although the findings are noteworthy, the study design and its findings are not without limitations. The main limitations include the relatively small sample size (likely due to conducting extensive, detailed phenotyping in all subjects), inclusion of only well-controlled T2DM patients, exclusion of patients on immunosuppressive agents, and the lack of quantification of OA by radiographic methodology. These issues may contribute to the observed overall lower prevalence of self-reported OA compared with prior studies. Although corrected through statistical means, the BMIs were significantly different between the control and T2DM groups; therefore, the contributing role of obesity in mediating the effects of insulin resistance and hyperglycemia on OA and OA-related functional impairment cannot definitively be discounted. In addition, patients with type 1 diabetes mellitus (T1DM) were not studied, pertinent as radioglenohumeral OA occurs in upwards of 35% of T1DM patients (9), a manifestation of complete insulin deficiency and hyperglycemia. Finally, the results presented are predominantly correlative, and thus lack significant mechanistic insight as to how insulin resistance and T2DM may affect immune function and other key functional measures in OA.

Nonetheless, the study is an advancement in our understanding of the link between insulin resistance, T2DM, and OA, and provides further impetus to determine the role of T2DM in OA pathogenesis and OA-related functional impairments. The potential masking effects of T2DM on microRNA expression are novel and worthy of future study in a larger group of subjects. Although the study is not definitive, due to several important limitations, including general applicability and its reliance on associative factors (which limit the mechanistic scope), the reported findings are an important lead-in to future studies on the underlying processes, including specific immune cell abundance/activation and inflammation, linking these highly prevalent and costly conditions. This information will be critical for subsequent therapeutic considerations.