Association between sarcopenic obesity, obesity, sarcopenia and quality of life in middle-aged and older Chinese: the Guangzhou Biobank Cohort Study

Sarcopenic obesity (SO) is complex phenotype that combines excess adiposity with a significant reduction in muscle mass and strength [1, 2]. The clinical significance of SO is underscored by its association with more severe adverse health outcomes compared to obesity or sarcopenia alone, including higher risks of all-cause mortality, cardiovascular disease, metabolic syndrome, and type 2 diabetes [3‐7]. A recent systematic review and meta-analysis showed that over a tenth of older adults globally were affected by SO [8]. With the global population aging, the prevalence of SO is anticipated to increase, highlighting the need for healthcare systems and policymakers to understand and address far-reaching implications of SO.

The coexistence of obesity and sarcopenia in SO complicates the clinical management of affected individuals, often exacerbating declines in physical function and overall well-being. These deteriorations are closely tied to quality of life (QoL), a multifaceted construct including physical, psychological, and social domains of health. However, the association between SO and QoL remains inconclusive. Some studies reported that individuals with SO experienced poorer QoL compared to those without obesity or sarcopenia, or those with obesity alone [9‐12]. However, some studies found no significant differences or suggested that SO may have better QoL than those with sarcopenia alone [10, 13]. Given the conflicting evidence and the importance of this issue, our study aimed to analyze the association between SO and QoL in middle-aged and older Chinese adults, using data from the Guangzhou Biobank Cohort Study (GBCS).

Of the 9,801 participants who completed the SF-12v2 questionnaire, 3,469 were excluded due to incomplete information on TUGT, grip strength, BMI and WC, or extreme values of TUGT and grip strength, leaving 6,332 participants (75.3% women) in the present study. Of them, 3,842 (60.7%) were identified as probable sarcopenia: 578 (9.1%) based on grip strength only, 2,222 (35.1%) based on TUGT only, and 1,042 (16.5%) meeting both criteria. Additionally, 680 (10.7%) participants had a BMI ≥ 28 kg/m², 248 (3.9%) had a BMI ≥ 30 kg/m², and 1,985 (31.4%) were abdominal obesity (Supplementary Table 1). Based on the classification criteria 1, 1,819 (28.7%) participants were in the NONS group, 671 (10.6%) in the IO group, 2,453 (38.7%) in the IS group, and 1,389 (21.9%) in the SO group. The mean age of participants was 60.45 (standard deviation (SD) = 7.7) years (Table 1).

Table 1 shows that, compared to participants with NONS, those with SO were older and had a higher proportion of women (all P < 0.001). The SO group also had lower education, a higher prevalence of manual occupation, lower family income, lower physical activity levels, lower adherence to the DASH diet pattern, and lower rates of smoking and alcohol consumption (all P < 0.05). Moreover, compared to the NONS group, those with SO had longer sleep duration, a greater number of co-morbidities, more falls in the past 6 months, higher GDS-15 scores, and higher BMI (all P < 0.01). The mean (SD) scores for the PCS and MCS, as well as the eight domains of the SF-12v2 were as follows: PCS 48.6 (5.5), MCS 55.8 (6.9), GH 32.5 (7.9), PF 53.8 (5.8), RP 55.0 (5.5), BP 53.9 (7.3), RE 53.2 (6.3), MH 55.7 (7.9), VT 58.6 (7.3) and SF 54.4 (5.8). Compared to participants with NONS, those with SO had significantly lower scores in PCS, GH, PF, RP, BP, RE, VT, and SF (all P < 0.01) (Table 2).

In Fig. 1, after adjusting for sex, age, family income, education, occupation, smoking status, alcohol use, DASH diet adherence, snoring, physical activity, number of co-morbidities, number of falls in past 6 months, and GDS-15 score, compared to participants with NONS, those with SO had significantly lower PCS, GH, PF, RP, BP, and SF scores. The adjusted βs (95% CIs) were as follows: PCS -1.23 (-1.68, -0.79), GH -0.86 (-1.51, -0.21), PF -1.28 (-1.77, -0.80), RP -0.51 (-0.95, -0.07), BP -0.77 (-1.37, -0.18), and SF -0.74 (-1.21, -0.26). Moreover, when compared to participants with IO or IS, those with SO had significantly lower PCS, PF, RP, and BP scores (all P < 0.05). No significant association of SO with MCS, RE, MH, or VT score was found. Participants with IS had significantly lower PCS (adjusted β -0.52, 95% CI -0.89 to -0.15) and GH (-1.09, -1.62 to -0.55) scores, and significantly higher MH scores (0.65, 0.13 to 1.17) (Fig. 1).

No significant sex-interaction was found on QoL (Table 3, P for interaction from 0.21 to 0.99). In sex-stratified analyses, the negative association between SO and PCS scores remained significant for both women and men. However, in women, the negative association of SO with GH (adjusted β -0.77, 95% CI -1.50 to -0.04), PF (-1.40, -1.98 to -0.82), and SF (-0.77, -1.33 to -0.21) scores remained significant (Table 3), while in men, the association between SO and RP remained significant (-1.09, -1.97 to -0.22) (Table 3).

Sensitivity analyses using four alternative SO criteria showed similar results for the association of SO with QoL (Supplementary Fig. 1–4), with no significant sex interaction observed (Supplementary Tables 2–5).

In this large population-based study, we found that SO was significantly associated with lower QoL in middle-aged to older adults, particularly in domains related to physical functioning, general health and bodily pain. However, no significant association was found between SO and mental health-related QoL, indicating that the detrimental effects of SO may be more confined to physical rather than psychological domains. These findings contribute to the growing body of evidence highlighting the distinct and severe burden of SO on physical health, particularly among middle-aged and older adults.

Previous studies examining the association between SO and QoL showed generally consistent results when using participants with NONS or those without sarcopenia as reference groups [9‐11]. For example, a cross-sectional study on 95 Spanish women aged over 70 years showed significantly lower QoL scores (measured by SarQol, a sarcopenia-specific QoL scale) in women with SO than those without sarcopenia (mean [SD] SarQol overall score: 45.09 [7.9] vs. 66.51 [16.17], P = 0.001) [9]. In this study, SO was defined using a combination of muscle strength criteria (grip strength < 16 kg or chair stand test ≥ 17 s), and body composition measures (total skeletal muscle mass adjusted by weight ≤ 27.6% or fat mass > 40.9%). However, this study did not adjust for important potential confounders such as smoking status, alcohol use, and sleep duration, which may lead to biased estimates [9]. Similarly, another cross-sectional study involving 423 Turkish adults aged over 65 years found that SO was negatively associated VT score, as measured by the SF-36 (OR_{SO vs. NONS} = 0.96, 95% CI = 0.93 to 0.99) [10]. In this study, SO was defined using criteria that included BMI (BMI ≥ 30 kg/m²), skeletal muscle mass index (< 9.2 kg/m² for men, < 7.4 kg/m² for women), and indicators of physical function such as walking speed (< 0.8 m/s) and/or muscle strength (< 32 kg for men, < 22 kg for women) [10]. Furthermore, a large cross-sectional study on 11,521 Koreans aged over 20 years showed that SO was associated with lower scores of QoL on the EQ-5D scale, particularly in domains related to self-management, daily activity, and exercise ability, when compared to NONS [11]. SO in this study was defined based on the ratio of appendicular skeletal muscle to weight (%) (< 30.3 for men, < 23.8 for women) and WC (≥ 90 cm for men, ≥ 85 cm for women) [11]. These studies above used various definitions of SO, yet consistently demonstrated its detrimental impact on QoL. Our results align with these studies and add to the literature by providing robust estimates using a larger, well-characterized sample of middle-aged and older Chinese adults.

We also found that participants with SO had significantly lower physical health-related QoL than IO or IS. Evidence regarding the association of SO with QoL using obesity or sarcopenia alone as the reference is sparse. Only one cross-sectional study on 130 Lebanese aged over 18 years reported that SO was significantly associated with poorer QoL compared obesity alone [12]. Additionally, two studies that used sarcopenia alone as the reference reported mixed findings [10, 13]. A cross-sectional study in Turkey found no difference in QoL between participants with SO and those with sarcopenia alone [10]. However, a longitudinal study involving 261 hemodialysis patients reported that SO was associated with better QoL than sarcopenia alone [13]. However, these studies were limited by small sample sizes and involved samples that differ significantly from the general community [10, 12, 13].

Despite the current inconsistency in the definition of SO, we found similar results for the association of SO with QoL in sensitivity analyses using four alternative SO criteria (Supplementary Table 1). However, all the criteria of obesity were based on BMI or WC. Given the potentially nuanced and more complex changes in body composition in the aging population, future studies using other measures such as dual-energy X-ray absorptiometry (DXA), or bioelectrical impedance analysis (BIA) are warranted. Additionally, given the lack of clarity in categorizing severity, we did not assess the heterogeneity within the SO phenotype.

The mechanisms underlying the association of SO with QoL remain unclear, although several hypotheses have been proposed. First, SO is associated with increased fat deposition in muscle cells, which reduces the number of mitochondria and increases the production of reactive oxygen species. This process impairs muscle function and hinders the growth of new muscle tissue [1]. Furthermore, the inflammatory response and insulin resistance associated with SO may exacerbate loss of muscle mass and increase fat accumulation, further impairing muscle function and reducing health-related QoL [2, 28]. Among older adults, these factors also lead to difficulties in exercise, further deteriorating muscle function and mass, and reducing physical health-related QoL [29]. Moreover, the loss of muscle mass and strength in SO may increase the risk of osteoarthritis, which can further limit physical activity and negatively impact both physical and mental health [30, 31]. Inflammation associated with obesity or sarcopenia may also disrupt the hypothalamic-pituitary-adrenal (HPA) axis, potentially leading to depression [32]. However, this inflammation pathway appears to be more pronounced in younger individuals, who are more susceptible to metabolic disturbance such as inflammation and insulin resistance [32]. The lack of significantly association between SO and mental aspects of QoL in our study of middle-aged and older Chinese aligns with this observation.

The strengths of our study included the population-based design, the use of different definitions of SO, and the comprehensive adjustment for potential confounders. However, there were several limitations in our study. First, the causal relationship between SO and QoL could not be established in this cross-sectional study. Second, the overrepresentation of women in our sample may limit the generalizability of our findings to the broader population. However, given that no significant sex interaction was observed on QoL and adjustments for sex were made, this imbalanced is unlikely to have substantially affected our conclusions. Third, although we adjusted for multiple potential confounders in linear regression models, residual confounding could not be completely ruled out. In particular, while we accounted for the number of chronic conditions, the absence of a validated comorbidity index, such as the Charlson Comorbidity Index, may limit our ability to fully capture the impact of disease burden on QoL. Additionally, complex interactions among confounders that may nonlinearly affect QoL were not explicitly modeled. Further randomized controlled trials could provide more definitive evidence regarding causality and the effectiveness of specific interventions aimed at reducing SO and improving QoL.

In conclusion, we found that SO was associated with lower physical health-related QoL compared to obesity or sarcopenia alone, but non-significantly associated with mental health-related QoL.