The Longitudinal Association of Adipose-to-Lean Ratio with Incident Cardiometabolic Morbidity: The CARDIA Study

Abstract

Aims:

To assess the association of adipose-to-lean ratio (ALR) with incident type 2 diabetes mellitus (T2DM), hypertension, and dyslipidemia in middle adulthood.

Methods:

Black and White Coronary Artery Risk Development in Young Adults participants without T2DM, hypertension, or dyslipidemia in 2005-06 (baseline) were included. Baseline adipose and lean mass were assessed via dual-energy X-ray absorptiometry. ALR was calculated as adipose divided by lean mass and then standardized within sex strata. Single time-point incident morbidity was assessed every five years from baseline through 2016. Cox proportional hazards regression was used to estimate hazard ratios (HR) for morbidity over 10 years per 1-SD increment in ALR adjusted for cardiovascular risk factors.

Results:

The cumulative incidence of T2DM was 7.9% (129 events/N=1,643; 16,301 person-years), 26.7% (485 events/N=1,819; 17,895 person-years) for hypertension, and 49.1% (435 events/N=855; 8,089 person-years) for dyslipidemia. In the adjusted models, ALR was positively associated with a risk of T2DM (HR [95% CI]; 1.69 [1.31, 2.19]) and hypertension (1.23 [1.08, 1.40]). There was no significant interaction between ALR and sex for any morbidity.

Conclusions:

ALR in middle adulthood is associated with incident T2DM and hypertension. The extent to which localized body composition measures might inform morbidity risk merits further investigation.

1. Introduction

Body mass index (BMI; kg·m⁻²) is a statistical construct that was designed as a proxy for adiposity in screening and surveillance of weight at the population level. However, since BMI reflects total mass rather than composition and distribution, BMI is a non-specific instrument that does not measure details of body composition or clinical phenotype necessary to assess individual risk.^1-6 The distribution of tissue types such as adipose and lean mass may provide more comprehensive information that may better explain the risk of cardiometabolic disease than BMI.⁷ The American Medical Association (AMA) recently adopted a new policy that no longer uses BMI in isolation for assessing the health of an individual based on weight.⁸

Both lean and adipose mass have been associated with cardiometabolic disease. Lower lean mass and higher adipose mass are jointly associated with a greater risk for type 2 diabetes mellitus (T2DM),^{9, 10} atherosclerotic vascular diseases,¹⁰ and all-cause mortality.^{11, 12} However, examining lean and adipose mass as independent factors ignores their biological interdependence and influence on one another through muscle-adipose tissue cross-talk.^13-16 Therefore, recent research has used lean-adipose mass ratios to evaluate the joint association of lean and adipose mass on measures of cardiometabolic health including metabolic syndrome,^17-19 insulin resistance,^{17, 20} and sarcopenic obesity.²¹

Healthy ranges of adipose tissue differ by sex, as women need more adipose tissue to maintain normal physiological processes.²² Despite the need for larger amounts of adipose tissue and increased adipogenesis,²³ women have a lower prevalence of T2DM, hypertension, and dyslipidemia.²⁴ BMI is a proxy for adiposity but cut-points do not account for sex differences in vital adiposity for physiological functions or location of adipose deposition. Therefore, sex may modify the association between measures of body composition and T2DM, hypertension, and dyslipidemia. The advantages of BMI are the low cost and ease of measurement. These advantages are not without limitations, which include a lack of distinguishing tissue types contributing to overall body composition and adiposity, particularly differences by sex, and subsequent health risk. Therefore, our goal is to assess the potential non-clinical value of a sex-standardized measure of adipose-to-lean ratio (ALR) in middle adulthood with incident T2DM, hypertension, and dyslipidemia over 10 years of follow-up. We hypothesized that ALR would be positively associated with risk for T2DM, hypertension, and dyslipidemia and these associations would not differ by sex.

2. Participants and Methods

2.1. Study Population

The Coronary Artery Risk Development in Young Adults (CARDIA) study is a longitudinal investigation of the etiology of cardiovascular disease in a Black and White cohort of 5,115 adults aged 18–30 years. Participants were recruited in 1985-1986 from Birmingham, AL, Chicago, IL, Minneapolis, MN, and Oakland, CA. Details of study recruitment and design have been previously reported.²⁵ Participants were reexamined 2, 5, 7, 10, 15, 20, 25, and 30 years after baseline for sociodemographic factors, anthropometric and chemistries. Written informed consent was obtained from all participants and the study design, data collection, and analyses were performed per ethical standards of supervising institutional review boards of all the centers involved. The present analyses are based on data collected at the 20-year (2005–2006), 25-year (2010-2011), and 30-year (2015-2016) follow-up examinations.

2.2. Adipose-to-Lean Ratio

Body composition was measured at all study sites by dual x-ray absorptiometry (DXA; Hologic QDR 4500W, Delphi 11.2, Discovery XP 12.1, Discovery XP 2002; Hologic, Bedford, MA). DXA measured at Y20 (2005 – 2006) was used for the present study. Individuals were ineligible to complete DXA if any foreign objects were present (e.g., jewelry, hairpins, joint replacements) or if pregnant. Body composition was operationally divided into two mutually exclusive categories: lean mass (kg) and adipose mass (kg). Adipose mass included both subcutaneous and visceral adipose tissue. To calculate ALR, adipose mass was divided by lean mass, then standardized to have a mean of zero and a standard deviation (SD) of one within sex-specific strata. We used sex-specific strata to address sex differences in the association between body composition and the risk of these morbidities by using a sex-standardized measure of ALR. We used a sex-standardized ALR to account for the difference in the physiological need for higher amounts of adipose tissue among women thus we present sex-stratified results. We tested for an ALR-sex interaction, but due to using a sex-standardized ALR, we expected a non-significant interaction. Self-reported sex at baseline was used to define sex-specific strata.

2.3. Body Mass Index

Height (centimeters) and weight (pounds) were measured by trained research staff with participants in light examination clothes and no shoes. These measurements were used to calculate BMI (kg·m⁻²). BMI was then standardized to have a mean of zero and a standard deviation (SD) of one within sex-specific strata.

2.4. Morbidities

Morbidity outcomes were defined by serum glucose and plasma lipid levels ascertained via fasting blood samples at each study visit. Participants were asked to fast for 12 hours and avoid smoking and heavy physical activity for at least two hours before each study examination. Blood was drawn by venipuncture according to standard procedures across all field centers. The serum was separated by centrifugation, transferred into airtight vials, and stored at −70 °C until shipped on dry ice to a central laboratory for processing.

The concentration of glucose in the stored samples was determined with a Cobas Mira Plus chemistry analyzer (Roche Diagnostics, Indianapolis, IN) using the hexokinase ultraviolet method. Fasting blood glucose ≥126 mg·dL⁻¹, post-load blood glucose ≥200 mg·dL⁻¹, HbA1c ≥6.5%, or reported use of antidiabetic medication was used to classify T2DM. Incident T2DM was defined as the first study visit at which the T2DM criteria were met among participants without prevalent T2DM at the Year 20 visit.

Total cholesterol, triglycerides, and high-density lipoprotein cholesterol (HDL-C) were directly measured using standardized assays. Low-density lipoprotein cholesterol (LDL-C) was calculated by the Friedewald equation.²⁶ Dyslipidemia was classified as total cholesterol concentration ≥200 mg·dL⁻¹, LDL-C ≥130 mg·dL⁻¹, triglycerides >150 mg·dL⁻¹, HDL-C <50 mg·dL⁻¹for females or <40 mg·dL⁻¹for males or reported use of lipid-lowering medication. Incident dyslipidemia was defined as the first study visit at which the dyslipidemia criteria were met among participants without prevalent dyslipidemia at the Year 20 visit.

Blood pressure was ascertained using standardized protocols across all study examinations and field centers using automated oscillometric blood pressure monitors. After five minutes of rest, trained staff obtained three readings from the brachial artery. Systolic blood pressure was defined as the average systolic blood pressure from the second and third readings. Diastolic blood pressure was defined analogously. Hypertension was classified as systolic blood pressure ≥140 mmHg, diastolic blood pressure ≥90 mmHg, or reported use of antihypertensive medication. Incident hypertension was defined as the first study visit at which the hypertension criteria were met among participants without prevalent hypertension at the Year 20 visit.

2.5. Covariates

Age (years), race (Black/White), education (high school or less/some college/college graduate or more), and cigarette smoking status (never/former/current) were ascertained by interview at all examinations.

2.6. Statistical Analysis

Continuous variables are reported as mean (SD) and categorical variables are reported as frequency (proportion). Participants with missing DXA or missing covariate data were excluded from all analyses. Participants with prevalent T2DM at Year 20 were excluded from analyses where incident T2DM was the outcome. Similarly, prevalent hypertension and dyslipidemia cases at Y20 were excluded from hypertension and dyslipidemia analyses respectively.

Correlations between ALR, BMI, and waist circumference for each morbidity were calculated (Supplementary Table 1.). Cox proportional hazard models were used to estimate hazard ratios (HR) for each morbidity over ten years of follow-up according to one-SD increment in baseline ALR. Models were adjusted in three sequential steps: unadjusted; minimally adjusted for sex, center, age, education, and smoking status; and fully adjusted which included all covariates from the minimally adjusted model with the addition of an ALR-sex interaction term. All adjusted models were stratified by sex. To assess whether associations differed when measuring body composition by ALR compared to the standard BMI, secondary analyses examined the relation between BMI (instead of ALR) and outcomes using the same modeling approach described above. A sensitivity analysis looking at quartiles of ALR for each morbidity was also conducted to distinguish possible threshold values of ALR associated with risk. The proportional hazard assumption was visually inspected via Schoenfeld residuals for each model and no violations were noted. A two-tailed p-value < 0.05 was considered statistically significant such that a 95% confidence interval (CI) that did not contain one denotes statistical significance. All analyses were conducted using R (version 4.2.1, R Foundation for Statistical Computing, Vienna, Austria).

3. Results

Among the participants in the incident T2DM analysis (Table 1.; N=1,643), the baseline age was 45.2 (3.49) years (61.5% White; 55.6% female) with a mean (unstandardized) ALR of 0.46 (0.20) and a homeostatic model assessment for insulin resistance (HOMA-IR) of 3.09 (2.03). The 10-year cumulative incidence of T2DM was 7.9% (129 events; 16,301 person-years). The incidence of T2DM among men was 9.1% (66 events; 7,215 person-years) and among women was 6.9% (63 events; 9,086 person-years). ALR was positively associated with the risk for T2DM in the unadjusted (1.53 [1.30, 1.81]; Table 2.), minimally adjusted (1.52 [1.28, 1.81]), and fully adjusted Cox models (1.69 [1.31, 2.19]). ALR was positively associated with the risk of T2DM among both females (1.39 [1.10, 1.76]) and males (1.55 [1.30, 1.85]). The interaction between ALR and sex for the T2DM analysis was non-significant (Figure 1.; p=0.270).

Table 1.

Participant characteristics – Analysis for incident morbidity over 10 years

	Overall	Female	Male
T2DM
	N=1,643	n=914	n=729
Adipose Mass (kg)	24.60 (10.39)	27.60 (11.25)	20.82 (7.69)
Lean Mass (kg)	5.63 (11.78)	46.71 (7.17)	64.55 (8.42)
Adipose-to-Lean Ratioa	0.46 (0.20)	0.58 (0.18)	0.32 (0.10)
Body Mass Index (kg·m−2)a	28.20 (5.64)	28.38 (6.54)	27.98 (4.25)
HOMA-IR	3.09 (2.03)	2.93 (1.90)	3.29 (2.17)
Hypertension
	N=1,819	n=1,027	n=792
Adipose Mass (kg)	24.08 (10.33)	26.79 (11.24)	20.58 (7.73)
Lean Mass (kg)	54.29 (11.92)	46.33 (7.00)	64.61 (8.60)
Adipose-to-Lean Ratioa	0.46 (0.20)	0.57 (0.18)	0.32 (0.10)
Body Mass Index (kg·m−2)a	27.86 (6.14)	27.77 (6.39)	27.98 (5.50)
HOMA-IR	3.08 (2.65)	2.88 (2.67)	3.35 (2.60)
Dyslipidemia
	N=855	n=546	n=309
Adipose Mass (kg)	22.56 (10.66)	25.19 (11.30)	17.90 (7.41)
Lean Mass (kg)	52.00 (11.38)	45.56 (6.88)	63.38 (8.48)
Adipose-to-Lean Ratioa	0.48 (0.21)	0.58 (0.19)	0.30 (0.10)
Body Mass Index (kg·m−2)a	26.73 (5.58)	26.77 (6.21)	26.65 (4.26)
HOMA-IR	2.65 (2.11)	2.55 (1.69)	2.82 (2.69)

T2DM = Type 2 Diabetes Mellitus; HOMA-IR = Homeostatic Model Assessment for Insulin Resistance

Continuous variables are reported with means (standard deviations) while discrete variables are reported with frequencies (percentage of sample)

Incident T2DM was fasting blood glucose ≥126 mg·dL⁻¹, post-load blood glucose ≥200 mg·dL⁻¹, HbA1c ≥6.5%, or use of antidiabetic medication

Incident hypertension was systolic blood pressure ≥130 mmHg, diastolic blood pressure ≥80 mmHg, or use of antihypertensive medication

Incident dyslipidemia was total cholesterol concentration ≥200 mg·dL⁻¹, LDL-C ≥130 mg·dL⁻¹, triglycerides >150 mg·dL⁻¹, HDL-C <50 mg·dL⁻¹ for females or <40 mg·dL⁻¹ for males, or use of lipid-lowering medication

^aUnstandardized

Table 2.

Overall and Sex-specific Multivariable Cox Regressions by Morbidity

	Overall		Female		Male
Crude Incidence
T2DM	129		63		66
Hypertension	485		263		222
Dyslipidemia	435		292		143
Incidence Ratea
T2DM	7.9		6.9		9.1
Hypertension	27.1		26.0		28.6
Dyslipidemia	53.8		55.8		39.6
	ALR	BMI	ALR	BMI	ALR	BMI
Model 1b
T2DM	1.53 (1.30, 1.81)	1.66 (1.48, 1.86)
Hypertension	1.22 (1.12, 1.33)	1.17 (1.12, 1.23)
Dyslipidemia	1.09 (1.00, 1.19)	1.08 (0.99, 1.19)
Model 2c
T2DM	1.52 (1.28, 1.81)	1.60 (1.41, 1.81)
Hypertension	1.18 (1.08, 1.28)	1.16 (1.10, 1.22)
Dyslipidemia	1.05 (0.96, 1.16)	1.07 (0.97, 1.17)
Model 3d
T2DM	1.69 (1.31, 2.19)	1.55 (1.33, 1.81)	1.39 (1.10, 1.76)	1.69 (1.36, 2.10)	1.55 (1.30, 1.85)	1.61 (1.41, 1.84)
Hypertension	1.19 (1.05, 1.34)	1.22 (1.10, 1.35)	1.16 (1.02, 1.32)	1.14 (1.06, 1.22)	1.18 (1.08, 1.29)	1.18 (1.11, 1.26)
Dyslipidemia	1.05 (0.93, 1.18)	1.06 (0.95, 1.19)	1.07 (0.91, 1.25)	1.07 (0.92, 1.26)	1.05 (0.96, 1.16)	1.07 (0.97, 1.17)
LDL-C	1.08 (1.00, 1.21)	1.15 (1.03, 1.27)	1.07 (0.92, 1.24)	1.05 (0.90, 1.22)	1.08 (0.98, 1.18)	1.11 (1.02, 1.21)
HDL-C	1.40 (1.22, 1.62)	1.48 (1.30, 1.69)	1.06 (0.88, 1.27)	1.05 (0.88, 1.26)	1.25 (1.12, 1.40)	1.29 (1.16, 1.44)
Triglycerides	1.20 (1.06, 1.36)	1.24 (1.12, 1.38)	1.19 (1.03, 1.38)	1.14 (0.99, 1.31)	1.20 (1.09, 1.32)	1.20 (1.11, 1.31)

ALR = Adipose-to-Lean Ratio; T2DM = Type 2 Diabetes Mellitus; LDL-C = Low-density lipoprotein cholesterol; HDL-C = High-density lipoprotein cholesterol

Model results presented as HR (95% CI) per 1-SD increase

^aPer 1,000 person-years

^bUnadjusted

^cAdjusted for center, sex, age, education, and smoking

^dAdjusted for Model 2 covariates plus sex-ALR or sex-BMI interaction

Figure 1.

Kaplan Meier Survival Analysis for 10-Year Risk of Incident T2DM Stratified by Sex (p=0.270)

Among the participants in the incident hypertension analysis (N=1,819), the baseline age was 45.1 (3.55) years (62.3% White; 56.5% female) with a mean (unstandardized) ALR of 0.46 (0.20) and a HOMA-IR of 3.08 (2.65). The 10-year cumulative incidence of hypertension was 26.7% (485 events; 17,895 person-years). The incidence of hypertension among men was 28.0% (222 events; 7,761 person-years) and among women was 25.6% (263 events; 10,134 person-years). ALR was positively associated with the risk for hypertension in the unadjusted (1.22 [1.12, 1.33]), minimally adjusted (1.18 [1.08, 1.28]), and fully adjusted Cox models (1.19 [1.05, 1.34]). ALR was positively associated with the risk of hypertension among both females (1.16 [1.02, 1.32]) and males (1.18 [1.08, 1.29]). The interaction between ALR and sex for the hypertension analysis was non-significant (Figure 2.; p=0.825).

Figure 2.

Kaplan Meier Survival Analysis for 10-Year Risk of Incident Hypertension Stratified by Sex (p=0.825)

Among the participants in dyslipidemia analysis (N=855), the baseline age was 45.1 (3.56) years (56.7% White; 63.9% female) with a mean (unstandardized) ALR of 0.48 (0.21) and HOMA-IR of 2.65 (2.11). The cumulative incidence of dyslipidemia was 50.9% (435 events; 8,089 person-years). The incidence of dyslipidemia among men was 46.3% (143 events; 3,611 person-years) and among women was 53.5% (292 events; 5,231 person-years). ALR was not associated with the risk for dyslipidemia in any of the Cox models. The interaction between ALR and sex for the dyslipidemia analysis was non-significant (Figure 3.; p=0.824). When only examining the association between ALR and the risk of individual lipids, ALR was positively associated with low HDL-C (1.40 [1.22, 1.62]) and high triglycerides (1.20 [1.06, 1.36]). ALR was positively associated with low HDL-C among males (1.25 [1.12, 1.40]) and high triglycerides among females (1.19 [1.03, 1.38]) and males (1.20 [1.09, 1.32]).

Figure 3.

Kaplan Meier Survival Analysis for 10-Year Risk of Incident Dyslipidemia Stratified by Sex (p=0.824)

For the secondary analysis, we repeated the analysis using BMI rather than ALR (Table 2.). BMI was positively associated with the risk for T2DM in the unadjusted (1.66 [1.48, 1.86]), minimally adjusted (1.60 [1.41, 1.81]), and fully adjusted Cox models overall (1.55 [1.3, 1.81]). BMI was positively associated with the risk of T2DM for both females (1.69 [1.36, 2.10]) and males (1.61 [1.41, 1.84]). BMI was associated with the risk for hypertension in the unadjusted (1.17 [1.12, 1.23]), minimally adjusted (1.16 [1.10, 1.22]), and fully adjusted Cox models overall (1.22 [1.10, 1.35]). BMI was associated with the risk of hypertension for both females (1.14 [1.06, 1.22]) and males (1.18 [1.11, 1.26]). BMI was not associated with the risk for dyslipidemia in any Cox models. When only examining the risk of the individual lipids, in the overall sample BMI was associated with LDL-C (1.15 [1.03, 1.27]), HDL-C (1.48 [1.30, 1.99]), and triglycerides (1.24 [1.12, 1.38]) for the overall sample. BMI was associated with LDL-C (1.11 [1.02, 1.21]), HDL-C (1.29 [1.16, 1.44]), and triglycerides (1.20 [1.11, 1.31]) among males. The interaction between BMI and sex for the T2DM, hypertension, and dyslipidemia analyses were all non-significant (p=0.540; p=0.291; and p=0.923, respectively). There was also no BMI-sex interaction when examining LDL-C (p=0.608) and triglycerides (p=0.915). There was a significant BMI-sex interaction when examining HDL-C (p=0.027).

The sensitivity analysis examining quartiles of ALR, and each morbidity is found in Supplementary Table 2. Regarding T2DM, both the third and fourth ALR quartiles, compared with quartile one, were associated with the risk of T2DM (4.24 [1.42, 12.66] and 7.22 [2.51, 20.78], respectively). Only the fourth ALR quartile, compared with quartile 1, was associated with the risk of hypertension (1.76 [1.22, 2.55]). There was no association between ALR quartiles and the risk of dyslipidemia. Further, in Supplementary Table 3., ALR is presented by the cumulative incidence of T2DM, hypertension, and dyslipidemia through Year 30. ALR increased along with the cumulative incidence of morbidity, regardless of type, through Year 30.

4. Discussion

In this longitudinal cohort study of Black and White adults, we assessed whether body composition, operationally defined as a sex-standardized ratio of adipose-to-lean mass, measured in middle age was associated with a 10-year risk for developing T2DM, hypertension, and dyslipidemia. We observed that ALR was associated with a higher 10-year risk of incident T2DM and hypertension and was not associated with the incident risk of dyslipidemia. Secondarily, we found BMI to be associated with T2DM, hypertension, low HDL-C, and high triglycerides.

The results from the present study build upon earlier research utilizing ratios of adipose and lean mass to emphasize the importance of examining the association of adipose and lean mass on markers of cardiometabolic diseases. The positive association between ALR and risk of incident T2DM is not surprising given the role of lean mass, specifically muscle, as the most abundant insulin-sensitive tissue.²⁷ Work from the NAGALA (NAdlf in the Gifu Area, Longitudinal Analysis) cohort study examining the association between adipose and lean body mass indicated that adipose mass was positively and lean body mass negatively associated with risk of diabetes.²⁸ Among participants of the Korean Genome Epidemiology Study, a low ratio of muscle mass relative to body weight was associated with the risk of T2DM.²⁹ The present results are consistent with prior research showing the individual tissues (e.g., adipose, lean) to have a stronger association than BMI with T2DM. Combining the Health Professionals Follow-Up Study and the Nurses’ Health Study cohorts researchers demonstrated that predicted adipose mass compared with BMI has been shown to regularly exhibit a stronger association with risk of T2DM.³⁰

The present work on hypertension is consistent with prior research from the CARDIA Study which demonstrated a greater risk of incident hypertension among middle-aged adults with higher BMI.³¹ Both predicted body adipose and skeletal muscle percentages have been associated with hypertension in a pooled cohort of Scottish and English adults.³² Individuals with a higher body adipose or a lower muscle percentage were both found to have greater odds of having hypertension.³² An earlier analysis of non-hypertensive adults from the Korean Genome Epidemiology Study found that increased body adipose percentage was associated with a risk of hypertension across 10 years of follow-up.³³ Additionally, the greater risk of hypertension associated with high body adipose percentage was observed when comparing high and low BMI groups among Korean adults.³³ As with T2DM, ALR was positively associated with the incidence of hypertension and the magnitude of association was stronger than the association observed between BMI and the risk of incident hypertension.

Contrary to our hypothesis, ALR was not associated with dyslipidemia risk; however, ALR was associated with the risk of low HDL-C and triglycerides. Evidence supports that lower lean mass at a given weight is positively associated with dyslipidemia in older adults.³⁴ Researchers observed after a two-year follow-up, a muscle-to-adipose ratio to be negatively associated with elevated triglycerides.³⁵ Another team of researchers examining adults with T2DM found that an adipose-to-muscle ratio is correlated positively with triglycerides and negatively with HDL-C.³⁶ A possible explanation for the difference in results with the present study may be that lean mass is not comprised solely of muscle tissue. Lean body mass encompasses all tissues that are not adipose mass (e.g., muscle, fluid). Examining the joint and independent effects of muscle and adipose mass, as opposed to lean and adipose mass, may further elucidate the relation between body composition and the 10-year risk of dyslipidemia.

There were no observed sex differences in the association between sex-standardized ALR and the risk of incident cardiometabolic disease in the present study. Other indices of adiposity including relative adipose mass have been found to be associated with incident heart failure covering 13.5 years of follow-up in The Prevention of Renal and Vascular End-stage Disease (PREVEND) Study. However, there were no sex differences across adiposity indices; similar to ALR in the present study, relative adipose mass considered sex differences when being calculated.³⁷ These results suggest the ratio between lean and adipose tissues is an influential factor in the development of cardiometabolic disease. Further, the ratio between lean and adipose tissues is sex-specific. However, the positive relation between adiposity and incident cardiometabolic disease risk irrespective of sex may be more complex as the location of adiposity depots within the body does influence the risk of cardiometabolic disease.³⁸ Visceral adiposity is known to be positively associated with the risk of cardiometabolic disease, especially in postmenopausal women.^38-41 The relative distribution of adipose tissue to lean tissue (e.g., appendicular vs. truncal) should be elucidated to understand if there are intrapersonal ratios of body composition that are stronger measures of cardiometabolic risk assessment.

Both ALR and BMI were similarly associated with the risk of 10-year risk for developing T2DM, hypertension, and dyslipidemia. The results from the present study support that ALR may be an alternative indicator for identifying persons at risk for future T2DM, hypertension, and certain dyslipidemias. Bioelectrical impedance scales, a validated device for estimating adipose and lean mass, are becoming more cost-effective and could be introduced into clinical settings.⁴² Despite its limitations, BMI is the proxy of adiposity most used in clinical settings and ALR does not provide additional information.

Our study has limitations. ALR does not detail the quality or distribution of each tissue. However, body composition for the present study was DXA measured, which is a leading option for whole-body measurement of body composition.⁴² We did not assess the location within the body of each tissue type. Tissue deposit location has been observed to be associated with each of our outcomes.^43-45 Timing of the body composition measures during the life course and over what period of follow-up may alter the results of the examined associations. The mean BMI in this study was 28.20 kg·m⁻² limiting the generalized of the present results to only Black and White middle-aged adults with an overweight BMI. We did not assess the risk of each morbidity by the various metabolic phenotypes (e.g., metabolically healthy obesity, unhealthy metabolic normal weight) which may influence the observed associations and should be addressed in future studies.

In conclusion, in a cohort of Black and White adults, ALR in middle age is associated with incident T2DM and hypertension over 10 years of follow-up. These results are in line with extant evidence; however, we accounted for both lean and adipose mass at once which more closely reflects how these tissues interact within the body. The extent to which cardiometabolic disease risk is informed by measures of body composition and how sex differences affect those associations merits further investigation.

• BMI is associated with type 2 diabetes, hypertension, and dyslipidemia.

• BMI does not measure details of body composition necessary to assess individual risk.

• Both lean and adipose mass have been associated with cardiometabolic disease.

• Adipose-to-lean ratio is comparable to BMI for assessing risk of incident morbidity.