In China, approximately 6% of individuals aged 60 and above are affected by dementia, and it is estimated that more than 15 million people have dementia (Jia et al., 2020). Dementia imposes substantial economic costs on patients, their families, caregivers and society as a whole, with a national study showing that the annual cost for treating patients with AD in China was $167.74 billion in 2015 and is projected to reach $507.49 billion by 2030 (Aranda et al., 2021; Jia et al., 2018). With the rapidly aging Chinese population, the incidence and economic burden of dementia are expected to increase. Therefore, identifying risk factors for cognitive decline is crucial for slowing the onset and progression of cognitive impairment.

A high prevalence of sarcopenia has been reported in patients with dementia (Chou et al., 2022). Individuals with sarcopenia may be at a higher risk of cognitive decline (Chen et al., 2022; Chang et al., 2016; Beeri et al., 2021). Sarcopenia is closely associated with an increased risk of dementia, and its underlying mechanisms involve multiple processes, including myokines, inflammation, oxidative stress, and insulin resistance. Sarcopenia is often accompanied by a chronic low-grade inflammatory state, and inflammation has been confirmed as a significant risk factor for cognitive decline (Zhao et al., 2024). Individuals with sarcopenia often experience heightened oxidative stress, which can detrimentally impact mitochondrial function, resulting in neuronal loss and consequent cognitive decline (Yu et al., 2024). Furthermore, patients with sarcopenia frequently exhibit insulin resistance, which can impair cognitive function through various mechanisms, including damaging synaptic integrity, increasing Aβ deposition, and promoting tau protein phosphorylation (Kellar and Craft, 2020). Additionally, muscle tissues secrete a variety of myokines that act on the central nervous system to promote neuronal growth and synaptic plasticity, thereby improving cognitive function (Pedersen, 2019; Scisciola et al., 2021). It is noteworthy that physical exercise can enhance cognitive function and decrease the risk of dementia by intervening in the aforementioned pathophysiological processes. Regular exercise has been documented to mitigate chronic inflammation by suppressing the secretion of pro-inflammatory cytokines, boost antioxidant defense mechanisms to mitigate oxidative stress, enhance insulin sensitivity, elevate levels of brain-derived neurotrophic factor (BDNF), irisin, and other muscle factors, thereby enhancing cognitive function (Fernández-Rodríguez et al., 2022; Mahalakshmi et al., 2020; Vints et al., 2022).

The 2019 Asian Sarcopenia Working Group suggests using calf circumference (CC) as a screening method for sarcopenia due to its strong sensitivity and specificity in Asian populations (Chen et al., 2020; Kim S. et al., 2018; Kawakami et al., 2015). CC has been used as a surrogate for assessing skeletal muscle mass (Abdalla et al., 2021; Kiss et al., 2024; Asai et al., 2019). In addition, measuring CC is very simple and non-invasive and is readily available in the community.

Several epidemiological studies have investigated the association between CC and cognitive function, yielding inconclusive results. Some studies have reported a positive correlation between CC and cognitive performance (Won et al., 2017; Kim et al., 2019); indicating that lower CC is linked to a higher risk of cognitive impairment (Liu et al., 2022; Wu et al., 2021; Tai et al., 2021). However, a recent study found no significant correlation between CC and overall cognitive performance (Sampaio et al., 2023). Sex disparities in the relationship between CC and cognitive function have also been documented. For instance, in a study involving Chinese older adults, a stronger association between CC and cognitive decline was observed in female participants (Wu et al., 2021). Conversely, another study reported that reduced CC was only associated with cognitive decline in males, with no such correlation found in females (Kim M. et al., 2018). Sex-specific differences in CC and cognitive function warrant further investigation. Given the inconsistent findings and the limited sample sizes and geographic representation in existing studies, there is a need for comprehensive studies with larger sample sizes encompassing diverse Chinese populations to elucidate the relationship between CC and overall cognitive function. In conclusion, numerous studies in China have examined the relationship between CC and cognitive function in small samples and specific regions, resulting in conflicting conclusions. Therefore, we conducted a cross-sectional study. We hypothesized a non-linear relationship between CC and global cognition, considering sex differences.

Our study has the largest sample size to investigate the link between CC and overall cognitive abilities in older Chinese population. In the entire population, the CC showed a non-linear relationship with global cognition. CC was positively associated with global cognition when the CC was < 31 cm. Conversely, this relationship disappeared when the CC was > 31 cm. Furthermore, older women exhibited a non-linear relationship between global cognitive function and CC, whereas older men did not show the same correlation.

CC is positively correlated with calf muscle thickness and is a good indicator of muscle mass (Kinoshita et al., 2022; Real et al., 2018; Asai et al., 2019). Our results suggest that CC is positively associated with global cognition when CC is less than 31 cm, whereas it is not related to overall cognitive function when CC is greater than 31 cm. Several studies have reported the relationship between CC and cognitive function. A study of 2,322 older adults in Malaysia found that CC was associated with improved cognitive function (Won et al., 2017). A Japanese study of 1,192 older adults found that individuals with lower CC exhibited poorer cognitive performance, while those with larger CC demonstrated a reduced risk of cognitive decline (Kim et al., 2019). Currently, most studies have found that older adults with lower CC have poorer cognitive function, while higher CC is a protective factor for cognitive impairment (Tai et al., 2021; Ma et al., 2021; Liu et al., 2022; Wu et al., 2021). Most studies agree that CC is positively associated with cognitive function, which is consistent with the findings of our study (when CC is less than 31 cm). In contrast, a study conducted in Brazil involving 263 older adults with an average age of 78 years reported no significant association between CC and cognitive impairment, diverging from our research findings (Sampaio et al., 2023). The variation in the results could stem from the diversity in age and sample size of the study participants. In addition, these studies did not incorporate factors such as nutritional supplements, living alone, or social activities as covariates. In contrast, our study incorporated more covariates and had a larger sample size with a geographically broad study population that was more representative of the Chinese population.

Several factors, including myokines, systemic inflammation, the muscle-gut-brain axis, insulin resistance, and oxidative stress, are implicated in the association between cognitive function and CC (Oudbier et al., 2022; Schlegel et al., 2019). The first mechanism of association between CC and cognitive impairment may be inflammation. Higher systemic markers of inflammation (e.g., CRP, IL-6, TNFα) are associated with lower muscle mass (Tuttle et al., 2020; Can et al., 2017). Persistent inflammation increases the likelihood of developing sarcopenia and dementia (Pan et al., 2021; Xie et al., 2021). In addition, skeletal muscle atrophy may mediate neuroinflammation to affect cognitive function adversely. For instance, muscle atrophy has been shown to impair memory function in young mice, potentially through neuroinflammation induced by muscle-derived hemoglobin (Nagase and Tohda, 2021). Skeletal muscle is capable of secreting various myokines, such as BDNF and irisin, which play crucial roles in the communication between muscle and brain. Recent evidence suggests that myokines produced and released by skeletal muscle can modulate brain function, reducing oxidative stress and neuroinflammation (Scisciola et al., 2021; Rai and Demontis, 2022). Irisin, a myokine released from skeletal muscle, has been shown to enhance BDNF expression, promote synaptic plasticity, increase neuronal survival, and confer anti-inflammatory and antioxidant effects, ultimately enhancing cognition (Peng and Wu, 2022). However, muscle wasting causes postoperative cognitive impairment in rats by reducing BDNF (Nemoto et al., 2022). Moreover, oxidative stress may underlie the association between CC and cognitive function, as evidenced by elevated oxidative stress levels in patients with sarcopenia, which can detrimentally impact neuronal and synaptic function, thereby contributing to cognitive decline (Bellanti et al., 2018). Recent research has highlighted the role of the muscle-gut-brain axis in modulating cognition, potentially linking CC to cognitive function. Muscle factors like irisin and BDNF influence gut microbial composition, promote the growth of beneficial bacteria, support intestinal barrier function, enhance neuroplasticity, mitigate neuroinflammation, and enhance cognitive performance (Cutuli et al., 2023). Additionally, insulin resistance may play a role in the relationship between CC and cognitive function, as it is associated with reduced skeletal muscle mass and represents a risk factor for cognitive decline (Merz and Thurmond, 2020; Ekblad et al., 2017). Peripheral insulin resistance has been implicated in stimulating tau phosphorylation, increasing amyloid-beta toxicity, exacerbating oxidative stress, and compromising synaptic integrity, ultimately contributing to cognitive decline (Kellar and Craft, 2020). It is worth noting that exercise can interfere with pathological factors related to muscle and cognitive function. Physical activity has been shown to decrease inflammatory markers and mitigate age-related oxidative stress (El Assar et al., 2022; Ihalainen et al., 2018). Moreover, regular exercise enhances muscle strength and insulin sensitivity (Grevendonk et al., 2021). Notably, exercise-induced muscle factors facilitate communication between the brain and muscles. For instance, the release of Cathepsin B during exercise can traverse the blood-brain barrier, stimulating BDNF expression in the hippocampus and thereby enhancing spatial memory (Moon et al., 2016). Additionally, exercise influences the gut microbiota and cognitive function through the secretion of muscle factors. Exercise has been proposed to synergistically enhance cognitive function by promoting hippocampal neurogenesis through the upregulation of irisin and BDNF, modulation of inflammation, and promotion of beneficial bacteria and short-chain fatty acids (Schlegel et al., 2019; Cutuli et al., 2023; Morella et al., 2023). Therefore, based on the available evidence, it is advisable for older adults to engage in consistent physical activity and uphold adequate calf circumference to mitigate the likelihood of cognitive decline.

Sex differences in the association between CC and global cognition remain understudied. Our investigation revealed notable sex distinctions in this relationship, particularly in older women. Unlike older men, a non-linear correlation between CC and global cognition was observed in older women. Consistent with our findings, a study by Bing Wu et al. found that the relationship between CC and cognitive impairment was more robust in women (Wu et al., 2021). However, several studies have reported opposite results. A study of older Korean women suggests that CC is not associated with cognitive function (Kim M. et al., 2018). Another large epidemiological study also pointed out that there was no correlation between decreased muscle mass and cognitive decline in older women aged (Abellan van Kan et al., 2013). Our findings do not align with these conclusions. These disparities underscore the need for further well-designed prospective cohort studies across multiple sites to elucidate this issue. The underlying mechanism for these disparities remains unclear. However, the observed variances may stem from fluctuations in sex hormones. Notably, estrogen confers neuroprotective benefits to female nerves by augmenting antioxidant capabilities and synaptic plasticity (Bustamante-Barrientos et al., 2021). Nonetheless, the substantial decrease in estrogen levels post-menopause among women could heighten the susceptibility to cognitive decline (Karamitrou et al., 2023). Moreover, a deficiency in estradiol leads to escalated inflammation (Geraci et al., 2021), which is a contributing factor to cognitive function deterioration. Additionally, there was a general lack of regular exercise in women compared to men in our study (Supplementary Table 2). Movement and exercise are widely acknowledged as protective elements for cognitive function, whereas lack of exercise poses a risk to cognitive health (Northey et al., 2018; Karssemeijer et al., 2017). The lower engagement in physical activity among women could elucidate the sex-based contrast in CC and cognitive function. Furthermore, persistent low-grade inflammation could underlie the sex-specific differences in the association between CC and cognitive function. Women typically manifest more robust immune responses and a heightened susceptibility to autoimmune diseases compared to men (Wilkinson et al., 2022), potentially contributing to a higher incidence of chronic low-grade inflammation in women. Prior research has indicated that increased levels of inflammatory markers influence the correlation between cognitive function and muscle mass in women but not in men (Canon and Crimmins, 2011). Muscle growth inhibitors, proteins produced and released by skeletal muscle, have been identified as promoters of cognitive decline in AD mice (Lin et al., 2019). Previous clinical research revealed a notable increase in myostatin levels in older women compared to older men, suggesting a potential role in sex-related variations in cognitive function and CC (Bergen et al., 2015). Future prospective studies are warranted to confirm this sex disparity, and additional mechanistic investigations are essential to substantiate these findings.

A strength of our study is our study boasts the most extensive sample size among surveys in China focusing on CC and global cognition. Furthermore, in contrast to prior research, our study accounted for a broader range of covariates that had not been fully addressed in previous investigations. These covariates include dyslipidemia, living arrangements, history of cancer, use of nutritional supplements, engagement in social activities, and history of stroke, thereby substantially enhancing the reliability of our findings. Additionally, we conducted curve fitting analyses to assess non-linear associations and performed sensitivity analyses treating CC both as a continuous and categorical variable to bolster the robustness of our results. On the other hand, this work has limitations. First, due to the nature of this cross-sectional study, we were unable to establish a causal link between CC and overall cognitive function. Second, a short epidemiological screening tool (MMSE) was utilized to identify cognitive impairment instead of a comprehensive clinical evaluation. Third, certain variables in our research, like hypertension and diabetes, relied on self-reported information, potentially leading to memory bias. Fourth, other unknown confounders could have affected the results; therefore, more extensive, rigorously designed studies are still needed.

To the best of our knowledge, this is the first study to discover a non-linear correlation between CC and global cognition. CC showed a strong positive correlation with global cognition below 31 cm, which disappeared above 31 cm. In addition, there were sex-related differences in the non-linear association between CC and global functioning. It is advisable to conduct widespread cognitive screening during routine check-ups for older adults to identify potential cognitive decline and recommend preventative measures early. Additionally, older women with low CC are advised to moderately increase their CC to maintain muscle mass and strength, potentially slowing cognitive impairment through these pathways.