In this extensive longitudinal cohort study involving a large representative sample of the general US population, we observed a significant positive association between SADHtR and the risk of all-cause mortality. Subgroup analysis revealed that this association remained consistent and appeared to be particularly pronounced among individuals aged 65 years or above.

After reviewing relevant literature, we identified studies that yielded comparable findings to ours. For instance, a cohort study consisted of 635 individuals with type 2 diabetes but no prior history of myocardial infarction or stroke at baseline [27]. Following a mean follow-up period of 7.1 years, their analysis demonstrated that SAD > 25 cm, as opposed to BMI or WC, remained significantly associated with major cardiovascular events, even after adjusting for covariates (HR 2.81, 95% CI 1.37–5.76). The outcome encompassed fatal or non-fatal cardiovascular events. In our investigation, we similarly observed a positive association between SADHtR and overall mortality. However, unlike the previous study limited to diabetes patients, our analysis included a broader population, and our outcome focused on all-cause mortality. Moreover, their study dichotomized SAD using a 25 cm cut-off point. In contrast, our analysis treated SADHtR as both a continuous variable (to evaluate the effect of each SD increase) and a categorical variable (by dividing it into tertiles). This approach enabled a more comprehensive assessment of its association with mortality, capturing both gradual changes and distinct groupings. Similarly, another study enrolled 30 intensive care unit patients diagnosed with severe sepsis [28]. Baseline measurements of both SAD and BMI were obtained, and they found that SAD, rather than BMI, exhibited an association with mortality up to day 60 post-admission. In contrast, our study revealed a similar positive association among participants in an epidemiological survey, rather than solely hospitalized patients. Furthermore, compared to their study, we implemented more rigorous adjustments for covariates in our multivariable regression models, enhancing the stability of our results. In yet another study, researchers included 418 incident peritoneal dialysis patients [29]. Through adjustment for pertinent risk factors, they independently linked SAD measured via lateral abdominal X-ray with all-cause and cardiovascular mortality during a mean follow-up period of 39.4 months. In our study, which involved NHANES participants with varying kidney functions, SAD was assessed using an abdominal caliper rather than X-ray. Even after adjusting for covariates, including eGFR, our analysis demonstrated a persistent positive association between SADHtR and all-cause mortality. Additionally, a study [30] encompassing 403 patients in the intensive care unit revealed that compared to the control group, the abdominally obese group (SAD ≥ 75th percentile) exhibited an increased risk of mortality in the intensive care unit up to day 60 post-admission (adjusted odds ratio 2.12, 95% CI 1.25–3.60). Although our findings were similar, it is important to note that our study was not conducted within a hospital setting, and our surveyed population differed significantly. Furthermore, our study featured a considerably larger sample size.

Our study differed from another study that examined 82 in-hospital patients with acute respiratory distress syndrome [31]. The participants were categorized as abdominally obese patients (SAD ≥ 26 cm) and the control group (SAD < 26 cm). The aforementioned study showed no significant difference in ICU mortality between the two groups from admission to day 7 (p = 1.00), however, it did reveal an elevated risk of mortality specifically among abdominally obese patients subjected to prolonged cumulative prone positioning. In contrast, our study revealed a positive association between higher SADHtR and increased all-cause mortality risk in the general US adult population. This positive association was consistently observed across all subgroups.

In Supplementary Table S4, SADHtR exhibited a stronger association with all-cause mortality compared to BMI, WC, and VAT in US adults. These findings suggest that SADHtR may offer a more accurate reflection of the association between visceral adiposity and mortality than conventional anthropometric and fat measures. Nonetheless, these conclusions warrant further validation in diverse populations and over extended follow-up periods.

We acknowledge that the findings in subgroup analyses might appear intriguing and seemingly contradictory, suggesting that diabetes and hyperlipidemia are protective for mortality. However, we believe these results can be explained through several plausible reasons. Firstly, it is important to distinguish between absolute risk and relative risk. Diabetes and hyperlipidemia are well-established major risk factors for mortality, meaning individuals with these conditions are already at a higher absolute risk of death. In this context, SADHtR might appear to have a smaller relative risk due to this higher baseline risk. This phenomenon, where relative risks are attenuated in high-risk groups, is not uncommon in epidemiology. Secondly, individuals with diabetes and hyperlipidemia typically receive more intensive medical monitoring and treatment, including medications, dietary modifications, and lifestyle interventions. These interventions might mitigate the impact of other risk factors, potentially reducing the observed relative risk in these populations. Thirdly, survival bias could also be a factor. Individuals with diabetes or hyperlipidemia who have survived long enough to be included in our study might represent a subgroup with better overall health management or resilience, which could skew the relative risk calculations. In addition, given that the p for interaction in the subgroups of diabetes and hyperlipidemia are both greater than 0.05, it suggests that the effect of the SADHtR on mortality does not significantly differ by the presence or absence of diabetes and hyperlipidemia. To further elucidate this phenomenon, future research should incorporate additional confounding variables, extend follow-up durations, and ensure sufficient subgroup-specific events to provide robust estimates.

We outlined several potential biological mechanisms that underlie the association between SADHtR and all-cause mortality. First, SADHtR is often linked to metabolic syndrome, including insulin resistance, elevated glucose, insulin levels, hypertension, and high triglycerides [12,13,14]. These metabolic abnormalities can contribute to the development of cardiovascular diseases, diabetes, and other chronic conditions, thereby increasing the risk of mortality. Second, visceral adiposity is associated with increased systemic inflammation. Excessive fat accumulation leads to the release of inflammatory mediators from adipocytes, such as tumor necrosis factor-alpha (TNF-α) and interleukin-6 (IL-6), triggering chronic inflammatory responses [32, 33]. This chronic inflammation may promote the formation of atherosclerosis, increasing the risk of cardiovascular disease and related mortality. Third, visceral adiposity may further increase overall mortality risk by impacting the risk of conditions such as sleep apnea, osteoporosis, and certain cancers [5, 34, 35].

Our study has several notable strengths. Firstly, we employed a large sample size and utilized a weighted analysis method, resulting in our findings being highly representative and applicable to the broader US population. Secondly, the stability and robustness of our results were evident through the implementation of multivariable regression models and subgroup analysis. Thirdly, the measurement of SADHtR is straightforward to implement in clinical practice and epidemiological surveys, and it also offers a cost-effective alternative to more expensive imaging techniques such as computed tomography and magnetic resonance imaging.

Several limitations should be acknowledged in our study. Firstly, it is important to recognize that this study adopts an observational design, which precludes the establishment of causal relationships. Secondly, we identified a positive association between SADHtR and all-cause mortality in US adults. However, the generalizability of these findings to individuals in other countries or regions remains uncertain. Further research is needed to elucidate this association comprehensively. Thirdly, it is worth noting that the incidence of mortality, particularly in relation to cause-specific mortality, appears to be relatively low. However, it is important to mention that data on SAD prior to the year 2011 are unavailable.