Stroke is now the second leading cause of death and disability worldwide (1). Data indicates a concerning trend of stroke incidence shifting toward younger demographics (2). Moreover, a staggering 70% of stroke-related fatalities and 87% of resulting disabilities are concentrated in low- and middle-income nations (3). This alarming situation not only leads to a significant loss of labor force but also imposes a substantial financial and healthcare burden on these countries (4). Fracture represents a perilous complication arising from stroke (5). The occurrence of fractures in the stroke population can hinder functional recovery by leading to prolonged hospitalization, reduced independence and delays in rehabilitation (6). Therefore, prevention of fractures is important for the recovery of stroke patients.

Depression affects approximately 71% of stroke survivors within the critical three-month post-stroke period (7). This mental health condition is a significant contributor to the exacerbation of physical decline and the decline in cognitive-psychological performance among individuals who have experienced a stroke (8–10). Notably, a negative correlation exists between the depression severity and bone density, with severe depression contributing to osteoporosis, and increased fracture risk (11, 12). The mechanisms underlying these phenomena are likely closely related to the regulation of the “brain-neuro-musculoskeletal” axis, which involves the bidirectional interaction between the brain’s neural control over the musculoskeletal system and the feedback regulation of brain function by the musculoskeletal system (13). For instance, the hypothalamic–pituitary–adrenal (HPA) axis, a critical pathway through which the nervous and endocrine systems interact, leads to the secretion of cortisol and other hormones upon activation (14). Depression often activates the HPA axis, increasing cortisol levels. Elevated cortisol not only exacerbates mood disorders like depression but also impairs musculoskeletal health by inhibiting osteoblast activity, reducing calcium absorption, and weakening bone structure. Cortisol plays a key role in stress responses, including fear, anxiety, and depression, while also inhibiting osteoblast activity, bone calcium absorption, and vascularization (15, 16). These effects indirectly disrupt bone microstructure and impedes bone turnover and metabolism (17). Therefore, depression, as a prevalent mood disorder, may significantly influence the link between stroke and fracture.

However, current research on the association between neuropsychiatric disorders and musculoskeletal diseases primarily focuses on adolescent and elderly populations (18–20). The relationship between depression and fracture incidence in stroke patients has not been well-defined. While previous studies have reported significant associations between depression and fall risk in stroke patients, most have been limited by small sample sizes and insufficient analytical depth, leaving the precise role of depression in the relationship between stroke and fracture poorly understood (21). To address these limitations, this study employs robust statistical methods, such as Sobel’s test and bootstrap analysis, which enhance the reliability and validity of the mediation findings. This provides a solid foundation for future research and clinical practice, supporting the development of more effective strategies (22, 23).

In summary, to delve deeper into the association between stroke and fracture incidence risk, we used data from the National Health and Nutrition Examination Survey (NHANES) 2017–2020 to (1) identify the influencing factors affecting fracture incidence in stroke and non-stroke populations; (2) investigate the relationship between stroke and fracture incidence and analyze the mediating role and extent of the influence of depression between stroke and fracture; (3) analyze stroke and fracture incidence-related Potential Mechanisms.

Using data from 2017 to 2020 in the National Health and Nutrition Examination Survey (NHANES), the study found a significant association between stroke and fractures, with depressive symptoms mediating the risk of fractures in stroke patients (32). Sobel and Bootstrap methods supported the results of fundamental analysis, demonstrating the robustness of the results (33).

The relationship between stroke and fracture is influenced by multiple factors. While stroke survivors are prone to dizziness, often leading to falls and fractures (34, 35). However, the present study found no significant association between dizziness and fracture risk in this population. This suggests that dizziness may not be a primary factor in stroke-related fractures (36). As is well known, osteoporosis is the primary factor leading to fractures (37). The study identified a significant association between osteoporosis and fractures in stroke patients, likely due to limited mobility, reduced mechanical load from paralysis (38), and oxidative stress from cerebral ischemia–reperfusion injury. This stress disrupts the dynamic balance of osteoblast differentiation, apoptosis, and metabolism, contributing to osteoporosis (39). At the same time, gender emerges as an important influencing factor (40). Table 2 shows that in the stroke population, women in the stroke population are more prone to fractures than men. Yao et al. (41) reveal that estrogen fluctuations during menopause disrupt neurotransmitter secretion, affecting osteoblast and osteoclast balance and leading to osteoporosis. Furthermore, estrogen may contribute to osteoporosis progression by inducing oxidative stress, triggering inflammatory responses, and modulating microRNA expression, further increasing fracture susceptibility (42).

Existing research confirms that depression plays a pivotal mediating role in stroke and its associated complications (43), with a significant association between stroke and depression (44). In addition, osteoporosis can be triggered by increased life stress and depression (45). Hence, we cautiously posit that depression could serve as a critical intermediary in the relationship between stroke and fractures. Table 3 shows that the stroke and unadjusted fracture models showed significant associations. Nonetheless, following the partial adjustment for confounding factors, the association observed between stroke and fracture models became non-significant. This indicates that the association between stroke and fractures is likely due to a complex interplay of various contributing elements (46). However, stratified regression analyses of the stroke population yielded a significant association between depressive status and the fracture model in the stroke population, whether or not confounders were excluded (47). Therefore, depression plays an important role in the relationship between stroke and fracture. Table 4 indicates that while the mediated association between depression and the occurrence of stroke-related fractures is modest (7.57%, p < 0.05) but statistically significant (48). The Sobel test results, as illustrated in Figure 2, confirm the statistical significance of the indirect effects (Z = 2.80, p < 0.01), direct effects (Z = 3.61, p < 0.001), and total effects (Z = 3.92, p < 0.001) of stroke on depressive symptoms and subsequent fracture risk. Bootstrap analysis provides a more nuanced view, pinpointing the direct effect of stroke on fractures at 0.079 (95% CI: 0.036–0.121) and the total effect at 0.085 (95% CI: 0.043–0.128). The indirect effect, mediated by depression, is a modest but significant 0.007 (95% CI: 0.002–0.011). These insights highlight the pivotal role of depression as a mediator in the relationship between stroke and fracture risk, underscoring the need for integrated approaches in managing these conditions (49).

In addition to exploring depression as a mediator between stroke and fracture risk, the underlying neurophysiological mechanisms warrant further investigation. Stroke induces structural and functional impairments in critical brain regions, which can precipitate negative emotional states such as depression (50). This emotional disturbance is associated with dysregulation of neurotransmitter secretions (51). Specifically, the HPA axis increases cortisol levels while decreasing serotonin levels (52). These imbalances impair osteoblast function, hinder calcium absorption in bones, and impede skeletal angiogenesis (53). Consequently, bone microarchitecture is compromised, and normal bone turnover and metabolism are disrupted, predisposing individuals to osteoporosis and an increased risk of fractures. Conversely, fractures also impact stroke recovery (47). Research indicates that fractures can enhance inflammation in the peri-infarct area, thereby exacerbating ischemic stroke (54). Furthermore, inadequate post-fracture care leading to lower extremity deep vein thrombosis also increases the risk of stroke (55). Thus, stroke and fractures may have a reciprocal causal relationship, with regulatory mechanisms based on the “brain-neuro-musculoskeletal” axis.

During the analysis, we searched for papers published on PubMed before September 2024 using the keywords “stroke,” “fracture,” and “depression.” A total of 199 papers were retrieved, of which three were relevant to our study. Yeh et al. (56) found that post-stroke depression significantly contributes to hip fractures among stroke survivors, exerting a disproportionately negative effect on younger individuals, irrespective of gender and the presence of comorbid conditions. Kelly et al. (57) consider that the brain-bone axis plays a crucial role in the regulation of skeletal metabolism, sensory innervation, and endocrine crosstalk among these organs. Furthermore, a previous study identified a notable association between stroke and fractures in elderly patients undergoing rehabilitation post-stroke (58). However, unlike our findings, no significant disparities in depressive symptoms and functional status were observed upon admission and discharge among the compared patient cohorts. This discrepancy may be related to the failure to exclude confounding factors (59). Additionally, the previous study did not quantify depression’s influence on the stroke-fracture relationship. Moreover, it lacked an analysis of the underlying mechanisms at play.

Our research has to some extent overcome these shortcomings and has the following advantages. Firstly, we established a distinct association between stroke and fractures, particularly noting a significant association between fracture incidence and depression in stroke patients. Secondly, we employed a weighted regression model to scrutinize the relationship between depression and fractures in this patient population, effectively addressing the challenge of a limited sample size. Thirdly, through an in-depth analysis with depression as a mediating variable, we discovered that while depression’s contribution is not substantial, its statistical significance is marked. Lastly, we propose an interactive relationship exists between brain neurotransmitters and bone cells, potentially regulated by the ‘brain-bone’ axis.

However, our study has some limitations. The sample size is relatively small and the data is restricted to the American demographic, primarily composed of Hispanic Americans, the elderly, and children, which may affect the generalizability of the findings. Additionally, the stroke situation is obtained through self-report, which could easily cause recall bias. Moreover, NHANES only records fracture data for individuals over 50 years old, preventing us from observing stroke and fracture incidence in younger populations.

The cross-sectional design of our study limits the ability to establish a temporal sequence among stroke, depression, and fracture occurrence. Future studies should incorporate multicenter and demographically diverse cohorts to further validate the intricate relationships among these variables. A prospective study design would be beneficial, as it would allow for a more accurate assessment of the temporal relationships between stroke, depression, and fracture risk (60). This approach would help in understanding the causal pathways and potential differences in outcomes across various stroke subtypes, such as ischemic and hemorrhagic strokes. Additionally, intervention studies could be recommended to evaluate the impact of targeted mental health interventions on reducing fracture risk in stroke patients. Such studies could include randomized controlled trials to assess the efficacy of depression management programs in improving bone health outcomes and reducing fall-related injuries. Moreover, animal experiments exploring the mechanisms of the brain-bone axis could provide valuable insights into the underlying physiological processes. Studies using animal models can help elucidate the bidirectional communication between the brain and bone, involving factors such as neuroendocrine signals and extracellular vesicles. Such research could advance our understanding of neuroendocrine signals and extracellular vesicles, ultimately contributing to the development of therapeutic strategies for treating both neurological and musculoskeletal disorders. By exploring these mechanisms, researchers may pave the way for developing novel therapeutic strategies that target both neurological and musculoskeletal disorders.

This study indicates a positive association between stroke and fractures, highlighting depression as a substantial mediating factor. It suggests that clinicians should consider both conditions concurrently in their assessments.