A Combined Index of Waist Circumference and Muscle Quality is Associated with Accumulation of Cardiovascular Disease Risk Factors in Obese Patients: A Cross-sectional Study

Background: Aging individuals are prone to sarcopenic obesity (SO). SO is associated with higher risks of low physical function, metabolic diseases, cardiovascular disease (CVD), and mortality. Worldwide diagnostic criteria for SO and its cutoff values have not yet been established. In this study, from among obesity- and sarcopenia-evaluated indices, we determined a combined index most associated with the accumulation of CVD risk factors in obese patients. Methods: In this cross-sectional study, we evaluated body composition using a multifrequency bioelectrical impedance analysis device and measured handgrip strength in 188 obese patients (73 men and 115 women). We selected each index most associated with the accumulation of CVD risk factors among obesity- (body mass index, percentage body fat, or waist circumference [WC]) and sarcopenia-evaluated indices (skeletal mass index, handgrip strength, or muscle quantity [MQ]), respectively. We then classied obese patients into four groups (group C, control; group O, obesity alone; group S, sarcopenia alone; and group SO, sarcopenic obesity) using the two selected indices with each median value and compared the CVD risk score, the number of concomitant CVD risk factors, among the four groups. Results: Based on the dichotomous comparison, WC and MQ were selected as the indices most associated with CVD risk factor accumulation. The CVD risk score was signicantly higher in groups S and SO as compared with group C (1.94 ± 0.80, 1.95 ± 0.91 vs. 1.41 ± 0.84, all p < 0.05). Odds ratios for CVD risk scores ≥ 2 were signicantly higher in groups S, O, and SO compared with group C (4.78 [95% condence interval {CI} 1.79-12.81], p = 0.002; 3.28 [95% CI 1.37-7.88], p = 0.008; 3.92 [95% CI 1.74-8.81], p = 0.001; respectively). Furthermore, odds ratios for CVD risk scores = 3 were signicantly higher only in group SO compared with group C (4.29


Page 3/12
This cross-sectional study included obese outpatients who regularly visited the Diabetes Center at the National Hospital Organization Kyoto Medical Center between January 2019 and July 2019. The diagnosis of obesity was based on the standards of the Japan Obesity Society, BMI ≥ 25 kg/m 2 [31]. We excluded participants with incomplete data, implantation of a cardiac pacemaker, and cancer from the study. This study was approved by the Ethics Committee for Human Research at National Hospital Organization Kyoto Medical Center (approval No.  and was conducted in accordance with the principles of the Declaration of Helsinki and the ethical guidelines for medical and health research involving human subjects.

Clinical examination
We measured height and body weight in increments of 0.1 cm and 0.1 kg, respectively. BMI was calculated as the body weight (kg) divided by the squared height (m 2 ). WC was measured at the umbilical level in a standing position. HGS was measured twice for each hand using the Smedley grip force system (Grip-D, Takei Equipment Company, Tokyo, Japan) in a standing position, and the maximum value was included in the analyses. The appendicular skeletal muscle mass (ASM) and PBF were measured using a multifrequency bioelectrical impedance analysis device (MC-780A-N, TANITA, Tokyo, Japan). A previous validation study demonstrated that body composition measured using this device was highly correlated with measurements obtained from of dual-energy Xray absorptiometry [32]. SMI was calculated as the ASM (kg) divided by the squared height (m 2 ). In this study, MQ was calculated as the HGS (kg) divided by muscle mass of the upper limbs (kg) according to previous reports [33][34][35]. Systolic and diastolic blood pressure was measured with an automatic electrical sphygmomanometer (BP-203RV ; Fukuda Colin, Kyoto, Japan). Blood was taken from the antecubital vein in the morning after an overnight fast, and we determined fasting plasma glucose, hemoglobin A1c (HbA1c), triglycerides (TG), high-density lipoprotein cholesterol (HDL-C), and low-density lipoprotein cholesterol (LDL-C).

Diagnosis of sarcopenia
Sarcopenia was diagnosed by low SMI and weak HGS [6]. The cutoff values for low SMI were <7.0 kg/m 2 for men and <5.7 kg/m 2 for women and those for weak HGS were <28 kg for men and <18 kg for women, respectively [10].

Diagnosis of hypertension, diabetes, and dyslipidemia
The diagnosis of hypertension, diabetes, and dyslipidemia was based on the criteria of each academic society; hypertension was de ned as systolic blood pressure ≥140 mmHg and/or diastolic blood pressure ≥90 mmHg or taking medications for hypertension [36]; diabetes was de ned by fasting plasma glucose ≥126 mg/dL, and/or random plasma glucose ≥200 mg/dL, and/or HbA1c (National Glycohemoglobin Standardization Program) ≥6.5%, or taking medications for diabetes [37]; dyslipidemia was de ned by LDL-C ≥140 mg/dL, and/or HDL-C <40 mg/dL, and/or TG ≥150 mg/dL, or taking medications for dyslipidemia [38].
De nition of CVD risk score In this study, the CVD risk score was de ned as the number of concomitant CVD risk factors (hypertension, diabetes, and dyslipidemia; 0-3 points), referring to a previous report [9].
Classi cation of obese patients using each obesity-and sarcopenia-evaluated index As shown in Figure 1, obese patients were classi ed into four groups using each obesity-and sarcopenia-evaluated index with the median value: group C, control (low obesity-evaluated index and high sarcopenia-evaluated index); group S, sarcopenia alone (low obesity-evaluated index and low sarcopeniaevaluated index); group O, obesity alone (high obesity-evaluated index and high sarcopenia-evaluated index); and group SO, sarcopenic obesity (high obesityevaluated index and low sarcopenia-evaluated index).

Statistical analysis
We performed statistical analyses using SPSS (version 25; IBM Corp, Armonk, NY, USA). Data are presented as mean ± standard deviation, median (interquartile range), or frequency percentage. In all cases, a probability (p) value of <0.05 was considered statistically signi cant.
Obese patients were dichotomized by the median value in each obesity-and sarcopenia-evaluated index. Those above the median value were classi ed as "high" and those below as "low." We conducted an analysis of covariance (ANCOVA) to compare the CVD risk scores between the low and high groups in each obesity-evaluated index (BMI, PBF, or WC) and sarcopenia-evaluated index (SMI, HGS, or MQ). The ANCOVA was constructed as follows: model 1 was unadjusted, model 2 was adjusted for sex, and model 3 was further adjusted for age. We conducted analysis of variance followed by Tukey's test or Kruskal-Wallis test followed by Bonferroni correction for continuous variables and chi-square test for categorical variables to compare the characteristics of the four groups (group C, group S, group O, and group SO). We used multiple logistic regression analysis to determine odds ratios (ORs) and 95% CIs for each CVD risk score (≥1, ≥2, and =3) as compared with group C, after adjusting for sex and age.

Page 4/12
Clinical characteristics of the study participants As shown in Figure 2, 196 patients (76 men and 120 women) were enrolled in this study. Six patients with incomplete data, one patient with a cardiac pacemaker, and one patient undergoing treatment for cancer were excluded. Finally, 188 obese patients (73 men and 115 women) were included in the study population.  34.3% ± 7.2% in men and 48.5% ± 8.0% in women, and WCs were 103.0 (98.5-111.0) cm in men and 102.0 (95.0-111.0) cm in women, respectively. There were 71 men (97.3%) with an abdominal circumference of ≥85 cm and 102 women (88.7%) with an abdominal circumference of ≥90 cm, suggesting that the participants had substantial visceral fat accumulation. On the other hand, as for the sarcopenia-evaluated indices, SMIs were 8.94 ± 1.04 kg/m 2 in men and 7.47 ± 0.83 kg/m 2 in women, HGS values were 39.6 ± 7.4 kg in men and 23.7 ± 4.9 kg in women, and MQs were 6.59 ± 1.12 kg/kg in men and 6.52 ± 1.18 kg/kg in women, respectively. Low SMI was observed in 3 men (4.1%) and weak HGS was noted in 5 men (6.8%) and 14 women (12.2%), respectively.
However, sarcopenia was diagnosed in only one male (1.4%). The prevalence of hypertension, diabetes, and dyslipidemia was high; therefore, patients had a high CVD risk score (1.92 ± 0.86 in men and 1.65 ± 0.90 in women).

Obesity-evaluated index most associated with the accumulation of CVD risk factors
For the obesity-evaluated indices, the median BMI was 30.6 kg/m 2 in men and 31.8 kg/m 2 in women, the median PBF were 34.1% in men and 47.7% in women, and the median WC was 103.0 cm in men and 102.0 cm in women, respectively. Table 2A shows the comparisons of CVD risk scores between the low and high groups in each obesity-evaluated index. The CVD risk score was signi cantly higher in the high group as compared with the low group only for WC (

Sarcopenia-evaluated index most associated with accumulation of CVD risk factors
For the sarcopenia-evaluated indices, the median SMI was 8.98 kg/m 2 in men and 7.37 kg/m 2 in women, the median HGS were 40.6 kg in men and 23.5 kg in women, and the median MQ was 6.63 kg/kg in men and 6.58 kg/kg in women, respectively. Table 2B shows the comparisons of CVD risk scores between the low and high groups in each sarcopenia-evaluated index. The CVD risk score was signi cantly higher in the low group compared with the high group only for MQ ( Clinical characteristics of the obese patients classi ed by a combined index of WC and MQ Based on the above results, we selected WC and MQ as the indices most associated with accumulation of CVD risk factors, respectively. Then, we classi ed obese patients into four groups using WC and MQ with each median value ( Figure 1 Table 3). As for other obesity-and sarcopenia-evaluated indices, both BMI and PBF were also signi cantly higher in group O and group SO compared with group C (all p < 0.05), and HGS was also lower in group S and group SO compared with group C (all p < 0.05; Table 3).
Interestingly, in contrast to MQ, the SMI in group SO was signi cantly higher than that in group C (8.32 ± 1.27 vs. 7.82 ± 1.03 kg/m 2 , p < 0.05).
Although there was no signi cant difference in the prevalence of hypertension and dyslipidemia among the four groups, that of diabetes was signi cantly higher in group S, group O, and group SO compared with group C (48.6, 41.5, 44.8 vs. 16.7%, all p < 0.05; Table 3). The CVD risk score was also signi cantly higher in group S and group SO compared with group C (1.94 ± 0.80, 1.95 ± 0.91 vs. 1.41 ± 0.84, all p < 0.05; Table 3).
ORs for accumulation of CVD risk factors in obese patients classi ed by a combined index of WC and MQ Finally, we investigated the effect of a combined index of WC and MQ on the CVD risk score (Figure 3). All ORs were adjusted for sex and age. We found no difference in ORs for CVD risk scores ≥1 among the four groups. However, ORs for CVD risk scores ≥2 were signi cantly higher in group S, group O, and group

Discussion
To the best of our knowledge, this is the rst study to determine a combined index associated with the accumulation of CVD risk factors in obese patients among each of the obesity-(BMI, PBF, or WC) and sarcopenia-evaluated indices (SMI, HGS, or MQ), respectively. Among obese patients, sarcopenia was diagnosed in only one man (1.4%). This result is consistent with previous studies reporting that few obese individuals diagnosed with high BMI meet the conventional diagnostic criteria for sarcopenia [6,10,39,40]. However, the prevalence of hypertension, diabetes, and dyslipidemia in these patients was high (Table 1), indicating that they are at a high risk for CVD. Therefore, we attempted to establish a diagnostic method for SO that re ects CVD risk. In this study, we provided evidence that classi cation using the combined index of WC and MQ re ects the accumulation of CVD risk factors in obese patients, regardless sex and age.
Many indices have been proposed for evaluating obesity. However, controversy remains on the obesity-evaluated index that best re ects the accumulation of CVD risk factors [41]. In this study, WC was most highly associated with the accumulation of CVD risk factors as compared with BMI and PBF, regardless of sex and age (Table 2A). BMI has been widely used to determine the prevalence of obesity and various risks in populations. In addition, PBF is often used as a criterion for evaluating the magnitude of accumulation of adipose tissue. However, in recent years, the indices of abdominal obesity, mainly WC, have been shown to be more closely related to CVD and mortality than BMI and PBF are [42,43], which is consistent with the ndings of our study.
Many indices have been proposed to evaluate sarcopenia, including muscle mass, muscle strength, physical performance, and MQ, but it is also unclear which sarcopenia-evaluated index best re ects CVD risk accumulation. Cao et al reported that in patients with metabolic syndrome, low SMI may be an independent risk factor for atherosclerosis [44]. In addition, low HGS has been reported to be associated with coronary artery calci cation, CVD, and all-cause mortality [45][46][47] Our study demonstrated that MQ was most associated with the accumulation of CVD risk factors compared with SMI and HGS, regardless of sex and age (Table 2B). MQ was reported to be negatively associated with insulin resistance after adjusting for age, body fat, highly sensitive C-reactive protein levels, and physical activity level in adult obese women [48] and further that insulin resistance contributes to the development of atherosclerosis [49,50], ndings that are consistent with our results.
In this study, HGS was lower in group SO compared with group C (27.9 ± 9.0 vs. 33.0 ± 9.7 kg/m 2 , p < 0.05), whereas SMI was signi cantly higher in group SO compared with group C (8.32 ± 1.27 vs. 7.82 ± 1.03 kg/m 2 , p < 0.05; Table 3). Therefore, MQ calculated as the HGS (kg) divided by the muscle mass of the upper limbs was signi cantly lower in group SO compared with group C (5.71 ± 0.72 vs. 7.50 ± 0.71 kg/kg, p < 0.05; Table 3). Mesinovic et al reported that overweight and obese older adults with metabolic syndrome have larger muscle size but poor MQ [51]. The lower MQ in group SO might be attributed to increased fat accumulation in the muscles [52,53]; however, in this study, we did not measure MQ with imaging analysis, such as ultrasonography or computed tomography. Recently, the usefulness of phase angle as an indicator for MQ has been suggested [6,54], and further studies on the increased fat accumulation in muscles are required in the future.
The combined index of WC and MQ was well associated with the accumulation of CVD risk factors in obese patients (Figure 3). To the best of our knowledge, there have been no reports on the diagnostic method for SO using a combined index of WC and MQ. Murai et al reported that patients with type 2 diabetes who had both visceral fat accumulation and low MQ were more affected with CVD [34]. The effectiveness of the combined index of WC and MQ as a relevant indicator of accumulation of CVD risk factors in obese patients requires further investigation.
The use of multiple medications per day is common with aging. Many drugs taken regularly for diseases may interact with some mechanisms that can alter the balance between protein synthesis and degradation [55], and researchers have reported that polypharmacy is associated with sarcopenia [56]. In this study, diuretics and biguanide (BG) were used signi cantly more often in group SO compared with group S and group C, respectively (p < 0.05, p < 0.05; Supplementary Table S1). The use of diuretics, particularly loop diuretics, has been suggested as a risk factor of sarcopenia [57]. On the other hand, BG improves insulin resistance and may inhibit the progression of sarcopenia [55]. Unfortunately, we did not assess insulin resistance in this study, but it has been reported that patients with SO are in an insulin-resistant state [2,4]. It is likely that BG was provided for insulin resistance in group SO in this study. The effects of BG on muscle remain to be elucidated.
The present study had several limitations that warrant mention. First, our study used a cross-sectional design. Thus, we could investigate only the associations between the combined index of WC and MQ and the accumulation of CVD risk factors. Second, because of the hospital-based design of the study, there may be bias among the study participants, which could limit the generalization of the study results. Third, the cutoff values for both WC and MQ were not clear. However, we were able to diagnose SO using the median of WC and MQ indices as cutoff values, which re ected well the accumulation of CVD risk factors. To resolve these limitations, larger cohort and prospective studies including various populations are needed in the future.
In conclusion, our study demonstrated that the obesity-evaluated index, WC, and the sarcopenia-evaluated index, MQ, were most closely associated with the accumulation of CVD risk factors in obese patients, respectively. Furthermore, classi cation by the combined index of WC and MQ re ects the accumulation of CVD risk factors in obese patients, regardless of sex and age. inhibitor; GLI, glinide; αGI, alpha glucosidase inhibitor; CVD: cardiovascular disease. Table 2 Comparisons of CVD risk scores between the low and high groups in each index BMI, body mass index; PBF, percentage body fat; WC, waist circumference; SMI, skeletal muscle mass index; HGS, handgrip strength; MQ, muscle quality; SBP, systolic blood pressure; DBP, diastolic blood pressure; FPG, fasting plasma glucose; HbA1c, hemoglobin A1c; TG, triglyceride; HDL-C, high-density lipoprotein cholesterol; LDL-C, low-density lipoprotein cholesterol; CVD, cardiovascular disease. § p value for difference among the four groups in means (ANOVA followed by Tukey's test), or medians (Kruskal-Wallis test followed by Bonferroni correction), or percentages (Chi-square test).
* p < 0.05 vs. group C; † p < 0.05 vs. group S; # p < 0.05 vs. group O. Figure 1 Classi cation using each obesity-and sarcopenia-evaluated index. Obese patients were classi ed into four groups, group C, group O, group S, and group SO, using each obesity-and sarcopenia-evaluated index.

Figure 2
Study ow chart of participants ORs for the CVD risk scores in obese patients of each group Multivariate sex-and age-adjusted ORs in obese patients classi ed by a combined index of WC and MQ (group C, group S, group O, and group SO). Squares, ORs for CVD risk score ≥1; triangles, ORs for CVD risk score ≥2; circles, ORs for CVD risk score = 3. Solid horizontal lines, 95% CI.

Supplementary Files
This is a list of supplementary les associated with this preprint. Click to download.