In a nationally representative population-based cross-sectional study, we showed for the first time that a novel anthropometric indicator of obesity, WWI, was significantly and positively associated with the prevalence of major depression in NAFLD. Each unit increase in WWI was associated with a 72.5% increased odds ratio among NAFLD participants, with a significant dose-response association. Gender was identified as a significant effect modifier, where the association between WWI and depression in NAFLD was greater in men. Importantly, our findings suggest that the association of WWI with depression in NAFLD is independent of BMI, whereas the association of BMI with depression disappears after additional adjustment for WWI. Interestingly, BMI was identified as another effect modifier, with the association between WWI and depression attenuating as BMI increased. These results highlight that WWI may be a novel predictor of depression in NAFLD independent of BMI, whereas the traditional association of BMI with depression is dependent on WWI. Central obesity as reflected by the WWI has an important role in the development of depression in NAFLD, and it may be used for risk stratification and related prevention of depression in adult individuals with NAFLD.
WWI is a recently proposed novel indicator of obesity that was first shown to be strongly associated with cardiometabolic comorbidities and mortality [21]. In contrast to BMI, the traditional reflective indicator of overall obesity, cumulative evidence suggests that WWI accurately reflects body composition including adipose tissue and muscle mass, suggesting that WWI is not only a marker of central obesity as distinguished from BMI, but also reflects abdominal adipose tissue accumulation to some extent [22–24]. A recent study showed that WWI compared to BMI, WC and waist-to-height ratio had the highest AUC for predicting sarcopenic obesity in men (0.781) [36]. Substantial large-sample observational clinical studies have linked WWI to the development and comorbidities of a variety of clinical conditions, including NAFLD. A cross-sectional analysis similarly using NHANES 2017–2020 suggested that WWI was associated with hepatic steatosis, independently of BMI [37]. Another cross-sectional study, also using NHANES 2017–2020, showed that WWI was significantly and positively associated with NAFLD and liver fibrosis [38].
Several cross-sectional analyses using NHANES have shown that WWI is positively associated with the prevalence of depression in the general population. Li et al. [25] included 34,528 participants from NHANES 2005–2018, in which the prevalence of depression was 7.61%, comparable to our study. WWI was positively associated with the prevalence of depression (OR = 1.21, p < 0.0001) and showed a non-linear association [25]. Our findings, on the other hand, suggest that WWI exhibits a linear association with depression in NAFLD as determined by VCTE, with a significant dose-response association. However, this study did not additionally adjust for BMI, and our study suggested that the association of WWI with depression in NAFLD was independent of BMI and attenuated with increasing BMI. Liu et al. [26] similarly used NHANES 2005–2018 and showed that WWI was significantly associated with depression in the general population (OR = 1.18) and similarly exhibited a nonlinear association. Furthermore, although this study did not additionally adjust for BMI, it indicated that the OR for the association of WWI with depression was higher than for BMI and WC. Shen et al. [39] similarly employed NHANES 2005–2018 to show that WWI was positively associated with the odds of depression in the general population and had a stronger association compared to BMI and WC. Interestingly, this study similarly showed that the association of WWI with depression was influenced by gender and BMI. However, this study indicated that this association was more pronounced in women and strengthened with increasing BMI, contrary to our findings. These findings may indicate an inconsistent pattern of association between WWI and depression in NAFLD compared to the general population.
NAFLD has been recognized as a sexually dimorphic disease. Overall, men have a higher prevalence of NAFLD compared to premenopausal women, which may be attributed to the protective effect of estrogen on liver metabolism [40, 41]. Female individuals with NAFLD may have a higher risk of developing depression than males, although inconsistencies remain [16]. Compared to premenopausal women, men have more abdominal fat accumulation (Android pattern), whereas women tend to have more subcutaneous fat (Gynoid pattern), although fat distribution in postmenopausal women is similar to that of men [42]. Thus, while the mechanisms of how gender affects this relationship remain unclear, gender variability in NAFLD, depression, and adiposity distribution may be underlying factors, suggesting that an emphasis needs to be placed on the association of WWI with depression in men with NAFLD.
Another important finding was that the association between WWI and depression in NAFLD was independent of BMI, whereas the association between BMI and depression disappeared after adjusting for WWI, suggesting that WWI may better reflect the association between obesity and depression than BMI. However, BMI was also identified as an effect modifier, and the association of WWI with depression was attenuated at higher BMI and was more pronounced in those with normal BMI. A prospective cohort study conducted from the China Health and Retirement Longitudinal Study demonstrated that the cumulative burden of WWI was positively associated with the risk of developing depression and was independent of BMI [43]. Sarcopenic obesity is suggested to represent a better association with depression than obesity defined with BMI [44, 45]. A Mendelian randomization analysis demonstrated that low muscle mass was independently associated with risk of depression, even after adjusting for BMI, suggesting that sarcopenia may be the real cause of the increased risk of depression [46]. Thus, WWI, as an indicator of obesity that accurately assesses body composition, better reflects the association between obesity and depression. In addition, BMI-defined lean/normal weight NAFLD represents a distinct subpopulation [47]. BMI-defined obesity may lead to misclassification of obesity phenotypes and may ignore the adverse effects of abnormal body composition, i.e., the "obesity paradox" [48]. We speculate that elevated WWI in patients with normal BMI may indicate lower muscle mass and therefore WWI is more pronounced in association with depression, whereas in those with overweight or obesity, BMI may override the effect of WWI. Thus, our findings emphasize that elevated WWI in NAFLD populations with normal BMI is associated with more significant odds of depression and that screening and prevention strategies need to be developed based on BMI.
Obesity-related hypothalamic-pituitary-adrenocortical axis dysfunction, chronic systemic inflammation and insulin resistance are thought to be involved in the pathogenesis of depression [11, 49]. Chronic systemic inflammation and the release of inflammatory cytokines may cause depressive-like behavior by triggering neuroinflammation and affecting neuroendocrine pathways [50]. Metabolic dysregulation and immune-inflammatory activation associated with sarcopenic obesity may be potential mechanisms for the development of depression [45]. In addition, multiple lifestyle and behavioral patterns associated with obesity may also be common risk factors for the development of depression, such as sedentary lifestyle, unhealthy diet, short sleep duration, and life stress [51]. Gut microbiome dysbiosis in obesity may also be an important contributor to depressive symptoms [52]. Given that the mechanisms linking obesity, especially central/sarcopenic obesity, to depression remain understudied, further exploration is needed in the future.
Our study has some important strengths. First, this is a nationally representative large population-based study with good generalizability and replicability. The diagnosis of NAFLD is based on the well-established VCTE with excellent diagnostic accuracy. In addition, the PHQ-9 has good consistency as a widely validated depression screening tool. We fully considered the effect of confounders, including BMI. However, our study has limitations. It was a cross-sectional study, and therefore causality could not be derived and could be influenced by residual confounders. In addition, NHANES lacks external validation of other tools for diagnosing depression. More well-designed cohort studies are needed in the future to confirm our findings.