The Impact Of Waist Circumference And Diabetes On Incident Of Cardiovascular Death During 9 Years Of Follow Up In General Population

Background: It has been reported that obesity and diabetes are both the risk factors for the development of cardiovascular diseases (CVD). However, recent articles reported that compared with BMI, waist circumference (WC) can better reect obesity, more closely related to visceral fat tissue which is positively associated with an increased risk of cardiovascular death. Moreover, few studies have both investigated the prognostic value of both WC and diabetes during a long-term follow up. We aimed to investigate whether higher level of WC measurements and diabetes were able to predict cardiovascular mortality in general population. Methods: In this prospective cohort study, a total of 1521 consecutive subjects free of clinical cardiovascular disease were included. The end point was cardiovascular death. The Kaplan-Meier method and Cox regression models were used to evaluate the cumulative risk of outcome at different WC levels with or without diabetes. Results: During a median follow up of 9.2 years, there were 265 patients had the occurrence of cardiovascular death. Kaplan-Meier survival estimates indicated that the patients with higher levels of WC (WC>94cm) coexist with diabetes had signicantly increased risk of cardiovascular death (log-rank p(cid:0) 0.05). After adjustment for potential confounders, multiple COX regression models showed that the incidence of cardiovascular death was signicantly higher when patients with high WC coexisted with DM (HR 3.78; 95% CI: 3.35–3.98; p(cid:0)0.001). Conclusion: Patients with high WC and diabetes represent a high-risk population for cardiovascular death. WC and diabetes may provide incremental prognostic value beyond traditional risks factors.


Background
Recent data from the United States showed that nearly a third of the world's population is now classi ed as overweight or obese [1]. It is recommended that obesity, which is usually de ned as excessive body fat and damage to health, can no longer be assessed only by body mass index (BMI) [2]. In epidemiological studies, BMI is often used to de ne overweight and obesity. However, BMI has a low sensitivity and there are large individual differences between body fat percentages, partly due to age, gender, and race [3]. individuals without type 2 diabetes mellitus [8]. These patients are all susceptible to cardiovascular abnormalities and cardiovascular diseases; their coexistence should further increase the risk of cardiovascular outcomes [9]. These data indicate that obesity patients with diabetes may be more common than large epidemiological studies have shown and requires more urgent attention. Relying solely on BMI to assess its prevalence may hinder future interventions for obesity prevention and control. Moreover, long-term research on WC and diabetes related to predictive value is still very limited [10].
Therefore, this study ought to investigate the prognostic value of WC on the cardiovascular death during a long-term follow up.

The study population
From January 2010 to December 2020, a total of 2162 people in Wan Shou Lu Street were included in the study. Among these participants, patients had a history of stroke or cardiovascular disease or cancer were excluded from the study. In the current study, patients were excluded if the patients could not give informed consent, had comorbidities lead to changes abdominal circumference measurement (secondary to chronic ascites, liver disease, cancer, intestinal obstruction, abdominal mass in the abdomen, preexisting abdominal stoma, incisional hernia). Participants were also excluded if their blood sample or detailed data were not available. Candidates lost follow up were also excluded from the study. Finally, a total of 1521 patients were included in the present study. The owchart of the study was listed in Figure   1.

Baseline measurement
The measures of the participants were collected at the time of registration. Anthropometric measurements include weight in kilograms, height in meters, hips circumference and WC. WC was measured on the mid-axillary line between the lowest border of the thoracic cage and the top of the iliac crest to the nearest 0.1 cm. Diabetes mellitus (DM) is de ned as the presence of diabetic symptoms and resting plasma glucose concentration ≥200mg/dL, fasting plasma glucose concentration ≥126mg/dL, or 2-hour plasma glucose concentration ≥200mg/dL, or the use of an oral hypoglycemic agent or insulin at the time of admission. BMI is calculated as weight/height 2 . After resting for 5 minutes, use the right arm in a sitting position and use standard or automated equipment to measure blood pressure, and use the average of the two blood pressure measurements as the nal blood pressure. The patient was considered to have hypertension with BP>140/90 mmHg or received antihypertensive medication. Hyperlipidemia is de ned as a known but untreated dyslipidemia or current treatment with lipid-lowering drugs. If a current smoker report smoking in the last 30 days, it is de ned as smoking.
Baseline data was collected at the time of enrollment. Blood samples were collected in the early morning. On the basis of protocol, the blood was obtained by the EDTA-anticoagulated plastic tubes. All the blood samples were centrifuged at 1000 g for 10 min and serum samples were stored at 80 ℃. Enzyme colorimetry was used to assess total cholesterol (TC), triglyceride (TG), high-density lipoprotein cholesterol (HDL-C) levels. A two-point kinetic assay kit (CH 9702, Randox Laboratories Ltd, UK) was used to determine the level of low-density lipoprotein cholesterol (LDL-C). The blood glucose level was measured using commercial reagents following standard procedures.

Outcome assessment
Patients were followed up until December 2020 or until the occurrence of cardiovascular death. All participants were followed up by analyses of clinical materials and telephone contact quarterly. The endpoint was cardiovascular death. Cardiovascular death was de ned as deaths caused by coronary heart disease or stroke, as well as deaths that cannot be classi ed without evidence of the source. Unless a clear non-cardiac cause is established, all deaths are considered cardiac deaths. Based on the International Classi cation of Diseases (ICD) codes for con rmation of each cause of death or routine death registration. Excluding the participants who were lost to follow-up, we obtained follow-up of all patients until the primary outcome or date of censoring. The follow-up time was calculated from the date of cardiac death onset to the date of mortality occurrence or the date of the last follow-up. Written informed content was obtained from all study participants, and the study was approved by the ethics committee of Chinses PLA General Hospital.

Statistical analysis
Patients were rstly divided into three groups according to the levels of WC. Then, the patients were further categorized into six groups as WC with DM and non-DM. Variables with a normal distribution are expressed as the mean ± SD, and in the case of non-normality, the medians are presented. Categorical data are expressed in counts or percentages. Differences in baseline characteristics between the three groups were evaluated by chi-square tests (categorical variables), analysis of variance as appropriate. The Kaplan-Meier method was used to evaluate the cumulative risk of outcome at different WC levels with or without DM, and compared by log-rank tests. Cox multivariate analyses were used to evaluate the association of WC levels and DM with the study endpoints. The results are presented as the hazard ratios (HRs) and 95% con dence intervals (CIs) according to levels of WC. Four multivariate proportional hazards models were tted. Model 1 contains the following variables: age, sex, BMI, current smokers, hypertension, TC, TG, HDL-C, LDL-C. Variables were input into the model according to their statistical signi cance in univariate analysis. Model 2 was based on model 1 with the addition of WC. Model 3 contains the Model 1 with the addition of diabetes. Model 4 contains Model 1 with the addition of WC and diabetes. The relationship between the WC levels and the outcome is represented by the COX proportional hazard model with WC as a continuous variable and WC as a categorical variable. Increase of (area under the curve) AUC in ROC (receiver operating characteristic) curve was used to compare the predictive power of the target parameter. In addition, when WC or diabetes was added to the established model, continuous net weight classi cation index (NRI) and comprehensive discrimination improvement (IDI) were generated to assess any improvement in prognostic prediction. SPSS 22.0 and R 4.0.0 (R Foundation for Statistical Computing) were used for descriptive data analysis. All statistical tests were 2tailed, and p values <0.05 were considered statistically signi cant.

Baseline characteristics
Baseline measurements of WC were available in 1521 patients. We divided the patients into three groups based on the levels of WC. The baseline clinical and laboratory characteristics of the study patients are presented in Table 1. The patients with a higher WC level were older, more likely to be female, more likely to be diabetes. Moreover, they had a higher BMI, hips circumference, fasting blood glucose, 2hPBG level, HbA1c level, uric acid level. Also, they had a higher systolic blood pressure (SBP) and diastolic blood pressure (DBP) level. However, they had a lower HDL-C as well as TC level.

Principal ndings
In this perspective, observational study on a large Chinese cohort with long-term follow-up, we examined the prognostic association of DM with outcomes in patients with different degree of WC. Our results suggested that higher WC and diabetes were both signi cant and independent predictors of cardiovascular death (Figure 4a). In addition, our results suggested that patients with higher levels of WC coexist with diabetes had the worst outcome, the association is still signi cant after adjustment of other clinical confounders (Figure 4b). In summary, our results suggested that the addition of WC and diabetes to established cardiovascular risk factors may further improve risk strati cation in general population. Our results provided updated information about the long-term prognostic role of WC and diabetes in general population.

Limitations of BMI
Adipose tissue is now considered to be a key organ for the fates of excessive dietary lipids, which may determine whether it will maintain body homeostasis (metabolic healthy obesity) or whether it will produce in ammation/insulin resistance, which can have harmful cardiovascular consequences. Obesity, especially visceral obesity, can also cause various structural adaptations/changes in the structure/function of CV. Adipose tissue can now be regarded as an endocrine organ that coordinates important interactions with vital organs and tissues (such as the brain, liver, skeletal muscle, heart, and blood vessels themselves).
The most commonly used anthropometric tool for assessing relative weight and classifying obesity is BMI, which is expressed as the ratio of total body weight to the square of height (kg/m 2 ). Individuals with a BMI <18.5 kg/m2 are considered underweight, while individuals with a BMI between 18.5 and 24.9 kg/m 2 are classi ed as normal or acceptable weight. Individuals with a BMI between 25 and 29.9 kg/m 2 are classi ed as overweight, while those with a BMI of ≥30 kg/m 2 are obese. BMI itself is associated with clinical outcomes and mortality in a U or J type [11]. This inverse relationship has triggered controversy in the literature, called the "obesity paradox" [12]. Compared with non-obese patients, patients with elevated BMI with chronic diseases have a higher survival rate and fewer cardiovascular (CV) events [13].
Moreover, previous research reported that patients with an increased BMI were found to show lower mortality [14]. In addition, BMI cannot distinguish between weight gain due to high levels of lean body mass and fat body mass. Generally, excess body fat (BF) is more often associated with metabolic abnormalities than high levels of lean body mass. Another explanation for this paradox is the lack of control over the main individual differences in the regional BF distribution. Therefore, more and more scholars believe that the BMI has its limitations to fully capture cardiometabolic risk. It is partially related to the fact that BMI in isolation is an insu cient biomarker of abdominal adiposity [5]. By using the BMI, one must rely on the assumption that adipose tissue is distributed evenly over the body, which does not take into account the heterogeneity of regional body fat deposition [15].
The level of obesity must be considered in the risk strati cation. In a recent meta-analysis of 2.88 million people, all levels of obesity combined were associated with increased mortality, with a hazard ratio of 1.18 (95% CI, 1.12-1.25). However, when analyzing separately, grade 1 obesity (table 1) is associated with a higher risk of death with a hazard ratio of 0.97 (95% CI, 0.90-1.04), compared with normal body weight.
In contrast, serious obesity (grades 2 and 3) and risk of death (hazard ratio 1.34 -95% CI, 1.21 -1.47) [16]. The prognostic value of BMI needs to pay attention to the length of follow-up time. There was a J-type association between BMI and sudden cardiac death, and the lowest risk was observed within the normal weight range. However, in studies with a longer follow-up period, the increased risk of low BMI was attenuated [17]. In other words, the obesity phenotype may change over time to re ect the increase in abdominal obesity. For example, Ian Janssen and colleagues studied the changes in waist circumference for a given BMI over a 30-year period in a Canadian sample35. It is worth noting that for a given BMI, Canadians had a larger waist circumference in 2007 than in 1981. Speci cally, the researchers observed that between 1981, men with a BMI of 25 kg/m2 increased their waist circumference by 1. The result of one study involving more than 58 000 elderly persons, during a 5-year-follow-up, showed that increased mortality risks for elderly people with an increased WC-even across BMI categories-and for those who were classi ed as 'underweight' using BMI. Part of the reason why BMI cannot fully capture cardiometabolic risks is that BMI alone is an insu cient biomarker for the whole body. More importantly, the central abdominal fat mass does not explain the extreme changes in intra-abdominal (visceral) fat mass. Fat distribution between individuals [18]. Compared with BMI, waist circumference has a higher predictive value for cardiovascular death [19].

Visceral adipose tissue and the underlying mechanisms
Visceral adipose tissue has been proved to be independently associated with elevated CVD (cardiovascular disease) risk [20]. Data from several past epidemiological studies 30 years of experience shows that VAT is an independent sign of morbidity and mortality [21]. In some populations, WC has been found to be more predictive of overall mortality, coronary heart disease (CHD), and CVD mortality than BMI, However, prospective data on the impact of abdominal obesity on CVD incidence is still scarce. Many experimental studies support the potential connection between VAT and biological pathways that are important in the pathogenesis of multiple disease outcomes. Adipokines are biologically active molecules secreted by adipose tissue and are key components of these pathways, including in ammatory cytokines, angiogenic factors, lipid metabolites, and extracellular matrix components [22]. The secretion of adipokines among speci c fat depots appears to be different [23], and compared with subcutaneous adipose tissue (SAT), VAT exhibits more pro-in ammatory and pro-angiogenic gene expression. In addition, compared with SAT, small arteries in VAT are more likely to exhibit endothelial dysfunction [24], indicating that VAT has a potentially toxic effect on the vasculature. Visceral adipocytes differ from subcutaneous adipocytes in that they release secreted proteins that are known or potential risk factors for CHD. In at least one study, visceral fat expressed and released more plasminogen activator inhibitor-1, a brinolysis inhibitor, than subcutaneous fat [25]. Angiotensinogen is a potential blood pressure regulator and is also highly expressed in visceral adipose tissue.
There are currently multiple methods to assess body fat distribution. The most accurate method is costly and time-consuming. It is not suitable for large-scale population research. Since routine access to CT, magnetic resonance imaging (MRI) may too expensive to be feasible for many clinicians, and the use of these methods to image visceral and ectopic fat has historically been reserved for research purposes, Perhaps the most widely used and these measurements are taken for waist circumference. Ashwell and colleagues were the rst to show that there is a correlation between visceral fat mass and waist-to-hip ratio.

Limitations of our study
The limitation of our study is the relatively small cohort patients; therefore, a larger sample of research is needed. Another limitation would be the lack of "gold standard" methods for abdominal obesity, such as computed tomography (CT) or MRI, may be another limitation of current research. However, the effectiveness of WC has previously been con rmed in cross-sectional studies and prospective studies.

Conclusion
In conclusion, our study demonstrated that an increased WC (WC≥94cm) was associated with an increased cardiovascular death in general population.

Consent for publication
Written informed consent for publication was obtained from each author and each patient.
Availability of data and material The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request. 30. Xing Z, Peng Z, Wang X, Zhu Z, Pei J, Hu X, Chai X: Waist circumference is associated with major adverse cardiovascular events in male but not female patients with type-2 diabetes mellitus. Cardiovasc Diabetol2020, 19(1):39. Figure 1 The owchart of the study