Urine Albumin-to-Creatinine Ratio: A Marker of Early Endothelial Dysfunction in Youth.

ORIGINAL

ARTICLE

Urine Albumin-to-Creatinine Ratio: A Marker of Early Endothelial Dysfunction in Youth Sara Klinepeter Bartz, Maria C. Caldas, Anca Tomsa, Ramkumar Krishnamurthy, and Fida Bacha Children’s Nutrition Research Center (S.K.B., M.C.C., A.T., F.B.), Baylor College of Medicine, Houston, Texas 77030; and Division of Pediatric Diabetes and Endocrinology (S.K.B., M.C.C., A.T., F.B.) and Department of Radiology, (R.K.), Texas Children’s Hospital, Houston, Texas 77030

Context: The urine albumin-to-creatinine ratio (UACR) is a useful predictor of cardiovascular (CV) events in adults. Its relationship to vascular function in children is not clear. Objective: We investigated whether UACR was related to insulin resistance and endothelial function, a marker of subclinical atherosclerosis, in adolescents across the spectrum of glucose regulation. Participants: Participants were 58 adolescents: 13 normal weight (NW), 25 overweight with normal glucose tolerance (OW-NGT), and 20 overweight with prediabetes (OW-PreD). Interventions: Interventions included oral glucose tolerance test, hyperinsulinemic-euglycemic clamp with determination of insulin sensitivity (IS), endothelial function assessment by peripheral arterial tonometry determination of the reactive hyperemia index (RHI), body composition (dualenergy x-ray absorptiometry), and abdominal fat distribution (magnetic resonance imaging). Primary Outcome Measure: Fasting UACR was determined. Results: The 3 groups did not differ with respect to age, sex, or Tanner stage. The NW group had significantly lower percent body fat, higher IS (10.4 ⫾ 0.9, 3.5 ⫾ 0.6, and 2.1 ⫾ 0.2 mg/kg/min per ␮U/mL; P ⬍ .001), and higher RHI (1.84 ⫾ 0.1, 1.56 ⫾ 0.1, and 1.56 ⫾ 0.1, P ⫽ .04) than the OW-NGT and OW-PreD groups, respectively. lnUACR was related to percent body fat (r ⫽ 0.4, P ⫽ .001), RHI (r ⫽ ⫺0.33, p ⫽ .01), and IS (r ⫽ ⫺0.27, P ⫽ .043). In multiple regression analysis with lnUACR as the dependent variable and RHI, percent body fat, age, sex, race, systolic blood pressure, cholesterol, glycated hemoglobin, and IS as independent variables, RHI (␤ ⫽ ⫺0.3, P ⫽ .045) and sex (␤ ⫽ 0.31, P ⫽ .06) contributed to the variance in UACR (R2 ⫽ 0.35, P ⫽ .02). Conclusions: UACR is an early marker of endothelial dysfunction in youth, independent of glycemia. Endothelial dysfunction may mediate the link between obesity-related insulin resistance and early microalbuminuria. (J Clin Endocrinol Metab 100: 3393–3399, 2015)

T

he urine albumin-to-creatinine ratio (UACR) has been associated with vascular inflammation and is a useful predictor of cardiovascular (CV) events in adults (1–3). In adult studies of type 1 and type 2 diabetes, urine mi-

croalbuminuria (UACR of ⱖ30 mg/g and ⬍300 mg/g) and macroalbuminuria (UACR of ⱖ300 mg/g) are recognized as indicative of nephropathy and are predictive of macrovascular disease risk (4 – 8). Progression of microalbu-

ISSN Print 0021-972X ISSN Online 1945-7197 Printed in USA Copyright © 2015 by the Endocrine Society Received May 8, 2015. Accepted July 8, 2015. First Published Online July 15, 2015

Abbreviations: AdDIT, Adolescent Type 1 Diabetes Cardio-Renal Intervention Trial; BMI, body mass index; BP, blood pressure; CV, cardiovascular; CVD, cardiovascular disease; eGFR, estimated glomerular filtration rate; FFM, fat-free mass; HbA1c, glycated hemoglobin; HDL, high-density lipoprotein; hs-CRP, high sensitivity-C-reactive protein; IFG, impaired fasting glucose; IGT, impaired glucose tolerance; IS, insulin sensitivity; LDL, lowdensity lipoprotein; NGT, normal glucose tolerance; NHANES, National Health and Nutrition Examination Survey; NW, normal weight; OGTT, oral glucose tolerance test; OW-NGT, overweight with normal glucose tolerance; OW-PreD, overweight with prediabetes; PAT, peripheral arterial tonometry; RHI, reactive hyperemia index; UACR, urine albumin to creatinine ratio.

doi: 10.1210/JC.2015-2230

J Clin Endocrinol Metab, September 2015, 100(9):3393–3399

press.endocrine.org/journal/jcem

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 17 September 2015. at 05:07 For personal use only. No other uses without permission. . All rights reserved.

3393

3394

Bartz et al

UACR and Endothelial Function in Youth

minuria further increases the risk of a CV event, and importantly, reduction decreases risk (4, 9). This is consistent with the notion that urinary albumin excretion is a result of increased vascular permeability related to endothelial damage. It is being increasingly recognized that the association between UACR and cardiovascular disease risk (CDV) in adults may occur at levels of urine albumin excretion that are below the traditional threshold for the definition of microalbuminuria (3, 9). This readily available measure has been proposed to be added to the evaluations of the CVD risk profile, in addition to lipids and blood pressure (BP) (10, 11). In addition to diabetes, obesity itself has a negative impact on renal function, which can lead to progression to renal disease (12). In children, elevated UACR has been associated with obesity (13), with features of the metabolic syndrome (13, 14), and with impaired glucose metabolism in obese children (15). More recently, microalbuminuria has been linked to insulin resistance in adolescents with type 2 diabetes (16). The relationship of albuminuria to CVD risk factors in children is not as well studied as that in adults. The mechanisms underlying albuminuria in childhood and the role of glucose toxicity vs insulin resistance in its pathogenesis are also unclear. We hypothesized that endothelial dysfunction is a significant determinant of UACR in youth and that this relationship is evident along the spectrum of glycemia. Specifically, we aimed to (1)evaluate endothelial function, using peripheral arterial tonometry and vascular markers and UACR in adolescents with and without dysglycemia and (2) assess the relationship of endothelial function to UACR, accounting for differences in body composition, insulin sensitivity (IS), and glycemia.

Subjects and Methods A total of 58 adolescents were evaluated. Glucose tolerance status was defined according to the American Diabetes Association criteria after a 2-hour oral glucose tolerance test (OGTT) (17). Normal glucose tolerance (NGT) was thus defined as glycated hemoglobin (HbA1c) of ⬍5.7%, fasting glucose of ⬍100 mg/dL, and a 2-hour value of ⬍140 mg/dL during the OGTT, impaired fasting glucose (IFG) as a fasting glucose value of ⱖ100 mg/dL, and impaired glucose tolerance (IGT) as a 2-hour value of ⱖ 140 mg/dL but ⬍200 mg/dL. Prediabetes indicated the presence of IFG, IGT, or combined IFG/IGT. Thirteen participants were of normal weight (NW) (age- and sex-specific body mass index [BMI] of ⬍85th percentile) (18) with NGT, 25 were overweight (BMI of ⱖ85th percentile) with normal glucose tolerance (OWNGT), and 20 were overweight with prediabetes (OW-PreD). Participants were recruited through advertisement in the community and at the medical center. Patients were excluded if they were engaged in scheduled diet or physical activity programs, had chronic medical conditions, were taking any medication, or


were smoking. Menstrual history and a urine pregnancy test were obtained for female participants. Females were studied in the follicular phase of the cycle when feasible. Estradiol levels in females were 84.7 ⫾ 13.8 pg/mL, consistent with the follicular phase of the cycle. Studies were conducted at the Children’s Nutrition Research Center at Texas Children’s Hospital and approved by the institutional review board of Baylor College of Medicine. Parental consent and child assent were obtained before any research procedures.

Anthropometric measures and body composition Participants underwent a physical examination with determination of pubertal Tanner stage by a pediatric endocrinologist according to the Tanner method (19, 20). Height was measured using a wall-mounted Harpenden stadiometer (Holtin Limited). The average of 3 measurements was recorded. Weight was obtained by a digital scale (Health-O-Meter). BMI was calculated, and BMI percentile and Z-score were defined based on age and sex norms. Waist circumference was measured in the midaxillary line at the midpoint between the lower edge of the ribs and the top of the iliac crest. Hip circumference was measured at the level of the greater trochanters. An average of 3 waist and hip measurements were taken and then were divided to produce the waist-to-hip ratio. BP was measured with an appropriate size cuff using an electronic sphygmomanometer (Welch-Allyn) while the subject was resting in the morning, and the average of 7 measurements taken 10 minutes apart was recorded.

Body composition Body composition was determined by dual-energy x-ray absorptiometry to determine measurements of total fat mass, fatfree mass (FFM), and percentage of body fat using a Hologic absorptiometer.

Abdominal fat Abdominal fat distribution was measured using a 1.5-T magnetic resonance scanner (Philips Achieva 1.5-T magnet) and abdominal visceral and subcutaneous fat were quantified at the L4 –L5 level using a multiecho dual flip angle magnetic resonance imaging sequence.

Endothelial function assessment Assessment of endothelial function was performed by peripheral arterial tonometry (PAT) using an EndoPAT device (Itamar Medical Ltd) (21, 22). The test was performed on resting subjects in a quiet, dimly lit room, in a fasting state, before any other study procedures. In brief, the index fingers are placed into pneumatic probes, and a BP cuff is placed around the nondominant arm. Pulse waves are recorded from both fingers. After a 5-minute equilibration period, the cuff in the test arm is inflated to suprasystolic pressure for 5 minutes (the nonoccluded arm serving as a control). Thereafter the cuff is rapidly deflated to allow for reactive flow-mediated hyperemia, and the pulse wave is recorded for another 5 minutes (23). The reactive hyperemia index (RHI) is calculated as the ratio of the average of the PAT signal starting 1 minute after cuff deflation divided by the average amplitude of the PAT signal of the 3.5-minute period before cuff inflation, by computer software (Itamar Medical Ltd) in an operator-independent manner. The ratio is normalized to the signal from the contralateral unoccluded digit (21).


doi: 10.1210/JC.2015-2230


3395

Urinary tests

Statistical analysis

Two fasting first morning void urine specimens were obtained on 2 different mornings for determination of urinary albumin (micrograms per milliliter) and creatinine (milligrams per deciliter). The UACR was calculated as the ratio of urine albumin to creatinine, expressed in milligrams per gram (24). Of the females included in the study, none were menstruating at the time of the urine collection. The participants were asked not to engage in physical activity in the preceding 48 hours. The average of the 2 measurements was used except in 6 patients who had only 1 sample.

The distribution of the different variables was examined, and the appropriate parametric or nonparametric test was applied. Differences in continuous variables among the 3 groups were determined by ANOVA. Pearson or Spearman correlation tests were used to examine bivariate relationships as appropriate. Multiple regression analysis was used to examine multivariate relationships. The natural log of UACR was used to normalize the data. All analyses were performed using SPSS (version 21; IBM). Figures were created using GraphPad (GraphPad Software Inc). Data are presented as means ⫾ SE, and a P value of ⱕ.05 was considered statistically significant.

OGTT A 2-hour OGTT (1.75 g/kg glucose, maximum 75 g) was performed after a 10- to 12-hour overnight fast. Blood samples were obtained at ⫺15, 0, 15, 30, 60, 90, and 120 minutes for determination of glucose and insulin.

Hyperinsulinemic-euglycemic clamp After a 10- to 12-hour overnight fast, insulin-stimulated glucose metabolism was investigated with a 3-hour hyperinsulinemic-euglycemic clamp. After baseline fasting, blood samples were collected (4 samples every 10 minutes), and intravenous crystalline insulin (Humulin) was infused at a constant rate of 40 mU/m2/min in participants with a BMI of ⬍85th percentile and 80 mU/m2/min for participants with a BMI of ⱖ85th percentile, as described previously (25). Plasma glucose was measured every 5 minutes using a YSI glucose analyzer (Yellow Springs Instruments), and a variable rate of 20% dextrose was administered to maintain plasma glucose at 100 mg/dL. Blood was sampled every 10 to 15 minutes for determination of insulin concentrations.

Biochemical analysis Plasma glucose was measured by the glucose oxidase method with the use of a YSI glucose analyzer. Plasma insulin and estradiol were measured by an electrochemiluminescence assay on the Elecsys 2010 (Roche Diagnostics Corporation). HbA1c by Roche Tina Quant turbidometric immunoinhibition and a lipid panel by enzymatic methods were measured at LabCorp. High sensitivity-C-reactive protein (hs-CRP) was measured by nephelometry at Esoterix Inc. Urine albumin was measured using an immunoturbidimetric assay, and urine creatinine was measured using a kinetic assay with the Jaffe creatinine method; both were processed at LabCorp.

Calculations The rate of insulin-stimulated glucose disposal (Rd) is calculated under steady-state conditions during the last 30 minutes of the hyperinsulinemic-euglycemic clamp as equivalent to the rate of glucose infusion. IS is calculated as Rd divided by the average insulin concentration during the last 30 minutes of the clamp and expressed per total body weight (kilograms) and per metabolically active FFM. Non– high-density lipoprotein (HDL) cholesterol was calculated by subtracting HDL from total cholesterol as described by Frost and Havel (26). Glomerular filtration was calculated using the modified Schwartz equation: estimated glomerular filtration rate (eGFR) (milliliters per minute per 1.73 m2) ⫽ (0.413 ⫻ height in centimeters)/serum creatinine in milligrams per deciliter (27, 28).

Results Subjects’ physical and metabolic characteristics A total of 58 children (23 male and 35 female; 29 Hispanic, 16 African-American, and 13 Caucasian), 15.7 ⫾ 0.2 years in age, underwent study evaluations. Subjects’ characteristics are detailed in Table 1. Of the children with prediabetes, 2 had IFG, 10 had IGT, and 8 had combined IFG/IGT. The 3 groups did not differ in terms of age, sex, or Tanner stage distribution. As expected, the NW group had significantly lower BMI, BMI Z-score, percent body fat, visceral and subcutaneous fat, and waist to hip ratio, compared with those for the OW-NGT and OW-PreD groups, respectively. As expected, the NW group had lower average fasting insulin and higher IS than the OWNGT and OW-PreD groups, respectively. Female subjects had lower serum creatinine (0.64 ⫾ 0.02 vs 0.75 ⫾ 0.04 mg/dL) than male subjects. When the NW and OW-NGT and OW-PreD groups were analyzed separately, there were no significant sex differences in BMI Z-score, RHI, or IS. CV risk markers RHI was significantly higher in the NW group than in the other 2 groups and was not significantly different between the obese groups with NGT and prediabetes. Leptin and hs-CRP increased, and adiponectin decreased from the NW to OW-NGT to OW-PreD groups. Triglyceride and total, low-density lipoprotein (LDL) cholesterol, and non-HDL cholesterol levels were not significantly different among the NW, OW-NGT, and OW-PreD groups. HDL was significantly higher in the NW group than in the other 2 groups. Systolic BP and diastolic BP were not significantly different among the 3 groups. UACR did not reach the threshold of microalbuminuria (ⱖ30 mg/g) and was not significantly different among the 3 groups. UACR was significantly higher in females than in males (6.51 ⫾ 0.68 vs 3.98 ⫾ 0.52, P ⫽ .01), with no significant differences noted by race. There was no sex difference in eGFR. eGFR was significantly higher in the OW-NGT and OW-


3396

Bartz et al

Table 1.



Participants’ Anthropometric and Metabolic Characteristics

Anthropometric measures Age, y Sex (male/female) Tanner stage (II-III/IV-V) Ethnicity (Hispanic/African American/ Caucasian) BMI, kg/m2 BMI Z-score Body fat, % Waist-to-hip ratio Visceral fat (cm2) Subcutaneous fat (cm2) Metabolic markers HbA1c, % Fasting glucose, mg/dL Fasting insulin, ␮U/mL IS, mg/kg/min per ␮U/mL IS per FFM, mg/kgFFM/min per ␮U/mL Vascular markers RHI hs-CRP, mg/L Leptin, ng/mL Adiponectin, ␮g/mL Cholesterol, mg/dL Non-HDL cholesterol, mg/dL Triglycerides, mg/dL HDL, mg/dL LDL, mg/dL Systolic BP, mm Hg Diastolic BP, mm Hg UACR, mg/g eGFR, mL/min/1.73 m2

NW (n ⴝ 13)

OW-NGT (n ⴝ 25)

OW-PreD (n ⴝ 20)

ANOVA P Value

16.3 ⫾ 0.4 6/7 0/13 2/5/6

15.5 ⫾ 0.4 8/17 3/22 16/4/5

15.6 ⫾ 0.4 9/11 0/20 11/7/2

NS NS NS .026

21.4 ⫾ 0.6 0.23 ⫾ 0.12 20.6 ⫾ 2.2 0.80 ⫾ 0.02 33.8 ⫾ 6.3 130.4 ⫾ 25.7

30.8 ⫾ 0.9 1.87 ⫾ 0.08 35.5 ⫾ 1.6 0.89 ⫾ 0.01 70.1 ⫾ 7.7 381.8 ⫾ 27.4

34.7 ⫾ 1.2 2.17 ⫾ 0.1 36.5 ⫾ 1.6 0.96 ⫾ 0.02 78.9 ⫾ 6.7 394.3 ⫾ 31.8

⬍.001a,b,c ⬍.001a,b,c ⬍.001a,b ⬍.001a,b,c .001a,b ⬍.001a,b

5.4 ⫾ 0.1 90.4 ⫾ 2.1 9.2 ⫾ 1.0 10.4 ⫾ 0.9 13.8 ⫾ 1.3

5.5 ⫾ 0.1 89.0 ⫾ 1.0 23.9 ⫾ 2.4 3.5 ⫾ 0.6 5.4 ⫾ 0.7

5.7 ⫾ 0.1 100.4 ⫾ 2.2 32.8 ⫾ 4.1 2.1 ⫾ 0.2 3.3 ⫾ 0.2

.048 ⬍.001b,c ⬍.001a,b ⬍.001a,b ⬍.001a,b

1.84 ⫾ 0.1 0.59 ⫾ 0.14 7.5 ⫾ 1.5 23.4 ⫾ 3.4 160.3 ⫾ 11.4 106.6 ⫾ 8.7 67.2 ⫾ 7.0 53.7 ⫾ 4.8 93.2 ⫾ 8.5 103 ⫾ 2 62 ⫾ 2 4.15 ⫾ 0.53 94.0 ⫾ 4.6

1.56 ⫾ 0.1 2.33 ⫾ 0.47 20.9 ⫾ 2.4 15.2 ⫾ 2.3 145.1 ⫾ 5.8 101.4 ⫾ 6.4 103.9 ⫾ 12.6 43.7 ⫾ 1.9 80.6 ⫾ 5.2 106 ⫾ 2 66 ⫾ 1 6.03 ⫾ 0.75 106.0 ⫾ 3.9

1.56 ⫾ 0.1 3.33 ⫾ 0.47 27.7 ⫾ 4.6 11.2 ⫾ 2.4 162.3 ⫾ 6.5 118.2 ⫾ 5.8 99.2 ⫾ 12.1 44.1 ⫾ 2.7 98.4 ⫾ 4.8 110 ⫾ 2 67 ⫾ 2 5.74 ⫾ 0.98 108.3 ⫾ 4.0

.039a,b .02a,b .001a,b .018b NS NS NS .05 a NS NS NS NS .049b

Abbreviation: NS, not significant. Results are reported as means ⫾ SE. To convert from mg/g of creatinine to mg/mmol, multiply by 0.113 (44). Post hoc significant differences in aNW vs OW-NGT, bNW vs OW-PreD, and cOW-NGT vs OW-PreD.

PreD groups than in the NW group, although below the hyperfiltration range defined as eGFR ⱖ 135 mL/min/ 1.73 m2 (29). eGFR was significantly different across race/ ethnicity (92.7 ⫾ 4.5 in African Americans, 101.2 ⫾ 4.7 in Caucasians, and 110.9 ⫾ 3.2 in Hispanics, P ⫽ .006). This race difference persisted after adjustment for group effect (P ⫽ .02). Determinants of UACR and eGFR UACR and eGFR were positively related to percent body fat (r ⫽ 0.41, P ⫽ .001 and r ⫽ 0.57, P ⬍ .001, respectively). UACR did not correlate with age, HbA1c, triglycerides, total, LDL, and HDL cholesterol, BP, hsCRP, leptin, or adiponectin. eGFR was inversely related to age (r ⫽ ⫺0.454, P ⫽ .001) and positively related to total abdominal fat (r ⫽ 0.36, P ⫽ .025), subcutaneous abdominal fat (r ⫽ 0.4, P ⫽ .01), diastolic BP (r ⫽ 0.31, P ⫽ .024), leptin (r ⫽ 0.27, P ⫽ .06), and hs-CRP (r ⫽ 0.28, P ⫽ .049) but not to HbA1c (r ⫽ 0.04, P ⫽ .8) or visceral abdominal fat (r ⫽ 0.23, P ⫽ .1). lnUACR was inversely related to Rd (r ⫽ ⫺0.29, P ⫽ .03), IS (r ⫽ ⫺0.276, P ⫽ .043), and

IS/FFM (r ⫽ ⫺0.25, P ⫽ .07). Similarly, eGFR was inversely related to Rd (r ⫽ ⫺0.52, P ⬍ .001), IS (r ⫽ ⫺0.43, P ⫽ .002), and IS/FFM (r ⫽ ⫺0.38, P ⫽ .007). lnUACR (r ⫽ ⫺0.33, P ⫽ .011) (Figure 1A) and eGFR (r ⫽ ⫺0.38, P ⫽ .005) (Figure 1B) were both inversely related to RHI. When the 2 sex groups were examined separately, the overall relationships of lnUACR to RHI (r ⫽ ⫺0.386, P ⫽ .02 in females and r ⫽ ⫺0.319, P ⫽ .1 in males, respectively) and eGFR to RHI (r ⫽ ⫺0.310, P ⫽ .09 in females, and r ⫽ ⫺0.498, P ⫽ .02 in males, respectively), were consistent with the findings in the total study population. In a multiple regression analysis with lnUACR as the dependent variable and RHI, percent body fat, age, sex, race, systolic BP, cholesterol, HbA1c, and IS per FFM as independent variables, RHI (␤ ⫽ ⫺0.3, P ⫽ .045) and sex (␤ ⫽ 0.31, P ⫽ .06) contributed to the variance in UACR (R2 ⫽ 0.35, P ⫽ .02). With eGFR as the dependent variable in the model, age (␤ ⫽ ⫺0.31, P ⫽ .01), race (␤ ⫽ 0.26, P ⫽ .04), and percent body fat (␤ ⫽ 0.40, P ⫽ .03) contributed to the variance in eGFR (R2 ⫽ 0.60, P ⬍ .001).


doi: 10.1210/JC.2015-2230


3397

the association of microalbuminuria with CVD risk factors (IFG, insulin resistance, hypertension, and smoking) was present in overweight adolescents only (39). In a recent study, obese children were found to have a prevalence of microalbuminuria and hyperfiltration similar to that of agematched lean peers with type 1 diabetes and higher than that of lean controls, further supporting a role of obesity in early nephropathy (40). This finding is consistent with our Figure 1. Relationship of RHI to the lnUACR (A) and eGFR (B). findings of a relationship between total body adiposity with urinary alDiscussion bumin and between total body, abdominal adiposity, and inflammatory markers with hyperfiltration as manifested In this study, we investigated the relationship of endotheby higher eGFR in the overweight/obese children. lial function, IS, and glycemia to the UACR. Our findings The relationship between in vivo IS and UACR in our indicate that endothelial dysfunction is a major determistudy is consistent with the findings of Burgert et al (15) of nant of UACR in children. Importantly, the relationship a relationship between UACR and OGTT-derived insulin between endothelial dysfunction and UACR persisted afresistance indices in a population of obese children. In ter accounting for adiposity, IS, and glycemia. To our addition, Bjornstad et al (16) recently reported a relationknowledge, this is the first pediatric study demonstrating ship between microalbuminuria and higher eGFR with a link between UACR with a direct measure of endothelial insulin resistance in adolescents with type 2 diabetes (16), dysfunction, a surrogate marker of early subclinical athconsistent with our results. Our finding that endothelial erosclerosis (30), in nondiabetic children. This relationfunction is the significant determinant of UACR after acship was independent of other traditional CVD risk factors, such as BP, lipids, or glycemia. In addition, our data counting for adiposity, IS, and glycemia further extends support the role of obesity and insulin resistance in the these observations and suggests that the effect of insulin resistance on the vasculature mediates the relationship beearly increase in the glomerular filtration rate. The relationship between UACR and CV risk in the tween obesity, insulin resistance, and early renal damage. Moreover, the continuous relationship between UACR, adult population has been demonstrated by several studies eGFR, and IS and endothelial function observed in this (4, 11, 31–36). Recent studies in adults show an indepenstudy at levels of urinary albumin within the currently dent association between microalbuminuria and arterial stiffness, vascular inflammation, and hypertension, which defined “normal range” are in line with recent findings are known predictors of CVD (3, 32, 37). In a large study from the Adolescent Type 1 Diabetes Cardio-Renal Interof ⬎9000 adults with and without diabetes, Gerstein et al vention Trial (AdDIT). In that study of adolescents with (11) demonstrated that UACR was a strong, independent, type 1 diabetes, raised UACR (within the normal range) and continuous risk factor for future CVD risk. Some in- was associated with arterial stiffness (measured by pulse vestigators even suggested lowering the threshold for the wave velocity) and evidence of renal damage (such as definition of microalbuminuria, given the strong evidence lower cystatin C levels and higher eGFR levels) (41). Our of UACR as a determinant of CV risk at levels of urinary finding that the relationship of UACR to RHI was indealbumin below the traditional definition of microalbu- pendent of measures of glycemia is also consistent with findings of the AdDIT study which showed that the abminuria (31). The pediatric literature has less of a consensus on the normalities in the CV and renal profiles in type 1 diabetic relationship between UACR and markers of metabolic children with increased UACR were independent of glydysfunction and CV risk. One study in healthy NW ado- cemic control (41). However, unlike our finding of a relescents did not find a relationship between UACR and lationship between RHI and UACR, this was not evident BMI, insulin resistance, or CVD risk factors such as lipid in the AdDIT study, probably related to differences in the abnormalities or BP (38). Yet, albumin excretion correlated population in the 2 studies with the AdDIT population with fasting insulin in that study. In a review of the National consisting of 10- to 16-year-old adolescents with type 1 Health and Nutrition Examination Survey (NHANES) data, diabetes with average HbA1c of 8%. In another report from


3398

Bartz et al


AdDIT, UACR correlated with aortic intima-media thickness, demonstrating a relationship between early evidence of vascular dysfunction and microalbuminuria (42). Our current findings in NW and overweight youth with and without impaired glucose regulation and the AdDIT results in children with type 1 diabetes of a relationship between vascular function and UACR in the relatively normal range are consistent with endothelial dysfunction as the underlying mechanism for early albumin leakage and are supportive of the observation that only trivial amounts of albumin excretion should be considered normal (11, 31, 33). Furthermore, these data provide support for the link between UACR and endothelial dysfunction, a marker of early subclinical atherosclerosis. Obese children in our study did not have higher UACRs than lean children. This result has also been observed in 12- to 19-year-old adolescents from NHANES (39); the causes remain unclear and are attributed to an element of orthostatic proteinuria in the NW children who are likely to be more active. We obtained first morning urinary samples to avoid this occurrence. UACR levels were higher in females than in males in our study population. This has also been previously noted in the literature, with females having been reported to have higher UACR (39, 43), attributed to differences in muscle mass and lower urinary creatinine excretion (4). Age and race were found to be significant contributors to the variance in eGFR, consistent with NHANES data showing lower eGFR values with increasing age and higher eGFR in Hispanics compared with Caucasians and African-Americans, similar to our findings (43). These observations merit further investigation. Our main limitation is the cross-sectional nature of this study, which precludes us from implying cause/effect relationships. However, a strength is the use of a direct measure of endothelial function and evaluation of IS using the gold standard hyperinsulinemic-euglycemic clamp methodology. This allowed us, despite the relatively small sample size and UACR with the “normal range,” to identify the important relationship of UACR and eGFR to IS and endothelial dysfunction. Our observations suggest the need to conduct longitudinal studies to better understand the effect of obesity and insulin resistance on early nephropathy and atherosclerosis. In conclusion, our findings demonstrate the usefulness of UACR as a marker of endothelial function and suggest that in obese youth, the adverse effects on renal function are mediated by the effect of adiposity-related insulin resistance on endothelial function, independent of the effect of glycemia. UACR is an easily obtainable measure, suitable for clinical studies aiming to assess CVD risk in children and may constitute a biomarker for intervention


studies that target changes in vascular function and CVD risk.

Acknowledgments We thank the nurses and staff of the Metabolic Research Unit at the Children’s Nutrition Research Center, Baylor College of Medicine; Elizabeth Johnson, RN, for her research-coordinating efforts; and Susan Sharma for the laboratory expertise. We are thankful to the study volunteers and their parents. Address all correspondence and requests for reprints to: Fida Bacha, MD, Children’s Nutrition Research Center, Baylor College of Medicine, 1100 Bates Street, Houston, TX 77030. E-mail: [email protected]. This work was supported by the US Department of Agriculture Agricultural Research Service (Grant 6250-5100-054 to F.B.) and the Thrasher Research Fund (to F.B.). S.K.B. contributed to data analysis and wrote the first draft of the manuscript. M.C.C. and A.T. contributed to data collection and reviewed the manuscript. R.K. contributed to the data and reviewed the manuscript. F.B. obtained funding, conceived and carried out the study, analyzed and interpreted the data, reviewed and edited the manuscript, and is the guarantor of this work. All authors approved the submitted and published versions. Disclosure Summary: The authors have nothing to disclose.

References 1. Widlansky ME, Gokce N, Keaney JF Jr, Vita JA. The clinical implications of endothelial dysfunction. J Am Coll Cardiol. 2003;42: 1149 –1160. 2. Cao JJ, Barzilay JI, Peterson D, et al. The association of microalbuminuria with clinical cardiovascular disease and subclinical atherosclerosis in the elderly: the Cardiovascular Health Study. Atherosclerosis. 2006;187:372–377. 3. Dutta D, Choudhuri S, Mondal SA, Mukherjee S, Chowdhury S. Urinary albumin:creatinine ratio predicts prediabetes progression to diabetes and reversal to normoglycemia: role of associated insulin resistance, inflammatory cytokines and low vitamin D. J Diabetes. 2014;6:316 –322. 4. Schmieder RE, Schrader J, Zidek W, et al. Low-grade albuminuria and cardiovascular risk: what is the evidence? Clin Res Cardiol. 2007;96:247–257. 5. Rossing P, Hougaard P, Borch-Johnsen K, Parving HH. Predictors of mortality in insulin dependent diabetes: 10 year observational follow up study. BMJ. 1996;313:779 –784. 6. Retnakaran R, Cull CA, Thorne KI, Adler AI, Holman RR, UKPDS Study Group. Risk factors for renal dysfunction in type 2 diabetes: U.K. Prospective Diabetes Study 74. Diabetes. 2006;55:1832–1839. 7. Molitch ME, DeFronzo RA, Franz MJ, et al. Nephropathy in diabetes. Diabetes Care. 2004;27(suppl 1):S79 –S83. 8. Hogg RJ, Portman RJ, Milliner D, Lemley KV, Eddy A, Ingelfinger J. Evaluation and management of proteinuria and nephrotic syndrome in children: recommendations from a pediatric nephrology panel established at the National Kidney Foundation conference on proteinuria, albuminuria, risk, assessment, detection, and elimination (PARADE). Pediatrics. 2000;105:1242–1249. 9. Stehouwer CD, Smulders YM. Microalbuminuria and risk for cardiovascular disease: analysis of potential mechanisms. J Am Soc Nephrol. 2006;17:2106 –2111.


doi: 10.1210/JC.2015-2230

10. Pedrinelli R, Dell’Omo G, Penno G, Mariani M. Non-diabetic microalbuminuria, endothelial dysfunction and cardiovascular disease. Vasc Med. 2001;6:257–264. 11. Gerstein HC, Mann JF, Yi Q, et al. Albuminuria and risk of cardiovascular events, death, and heart failure in diabetic and nondiabetic individuals. JAMA. 2001;286:421– 426. 12. Xiao N, Jenkins TM, Nehus E, et al. Kidney function in severely obese adolescents undergoing bariatric surgery. Obesity. 2014;22: 2319 –2325. 13. Csernus K, Lanyi E, Erhardt E, Molnar D. Effect of childhood obesity and obesity-related cardiovascular risk factors on glomerular and tubular protein excretion. Eur J Pediatrics. 2005;164:44 – 49. 14. Invitti C, Maffeis C, Gilardini L, et al. Metabolic syndrome in obese Caucasian children: prevalence using WHO-derived criteria and association with nontraditional cardiovascular risk factors. Int J Obes (Lond). 2006;30:627– 633. 15. Burgert TS, Dziura J, Yeckel C, et al. Microalbuminuria in pediatric obesity: prevalence and relation to other cardiovascular risk factors. Int J Obes (Lond). 2006;30:273–280. 16. Bjornstad P, Maahs DM, Cherney DZ, et al. Insulin sensitivity is an important determinant of renal health in adolescents with type 2 diabetes. Diabetes Care. 2014;37:3033–3039. 17. American Diabetes Association. Standards of medical care in diabetes—2010. Diabetes Care. 2010;33(suppl 1):S11–S61. 18. Rosner B, Prineas R, Loggie J, Daniels SR. Percentiles for body mass index in U.S. children 5 to 17 years of age. J Pediatr. 1998;132: 211–222. 19. Marshall WA, Tanner JM. Variations in pattern of pubertal changes in girls. Arch Dis Child. 1969;44:291–303. 20. Marshall WA, Tanner JM. Variations in the pattern of pubertal changes in boys. Arch Dis Child. 1970;45:13–23. 21. Selamet Tierney ES, Newburger JW, Gauvreau K, et al. Endothelial pulse amplitude testing: feasibility and reproducibility in adolescents. J Pediatr. 2009;154:901–905. 22. Hamburg NM, Benjamin EJ. Assessment of endothelial function using digital pulse amplitude tonometry. Trends Cardiovasc Med. 2009;19:6 –11. 23. Axtell AL, Gomari FA, Cooke JP. Assessing endothelial vasodilator function with the Endo-PAT 2000. J Vis Exp. 2010;44. 24. Keane WF, Eknoyan G. Proteinuria, albuminuria, risk, assessment, detection, elimination (PARADE): a position paper of the National Kidney Foundation. Am J Kidney Dis. 1999;33:1004 –1010. 25. Bacha F, Saad R, Gungor N, Arslanian SA. Adiponectin in youth: relationship to visceral adiposity, insulin sensitivity, and ␤-cell function. Diabetes Care. 2004;27:547–552. 26. Frost PH, Havel RJ. Rationale for use of non-high-density lipoprotein cholesterol rather than low-density lipoprotein cholesterol as a tool for lipoprotein cholesterol screening and assessment of risk and therapy. Am J Cardiol. 1998;81:26B–31B. 27. Schwartz GJ, Muñoz A, Schneider MF, et al. New equations to estimate GFR in children with CKD. J Am Soc Nephrol. 2009;20: 629 – 637. 28. Selistre L, De Souza V, Cochat P, et al. GFR estimation in adolescents and young adults. J Am Soc Nephrol. 2012;23:989 –996. 29. Sharma AP, Yasin A, Garg AX, Filler G. Diagnostic accuracy of cystatin C-based eGFR equations at different GFR levels in children. Clin J Am Soc Nephrol. 2011;6:1599 –1608.


3399

30. Bonetti PO, Pumper GM, Higano ST, Holmes DR Jr, Kuvin JT, Lerman A. Noninvasive identification of patients with early coronary atherosclerosis by assessment of digital reactive hyperemia. J Am Coll Cardiol. 2004;44:2137–2141. 31. Klausen K, Borch-Johnsen K, Feldt-Rasmussen B, et al. Very low levels of microalbuminuria are associated with increased risk of coronary heart disease and death independently of renal function, hypertension, and diabetes. Circulation. 2004;110:32–35. 32. Nah DY, Lee CG, Bae JH, et al. Subclinical renal insufficiency range of estimated glomerular filtration rate and microalbuminuria are independently associated with increased arterial stiffness in never treated hypertensives. Korean Circ J. 2013;43:255–260. 33. Nicholl DD, Hemmelgarn BR, Turin TC, et al. Increased urinary protein excretion in the “normal” range is associated with increased renin-angiotensin system activity. Am J Physiol Renal Physiol. 2012;302:F526 –F532. 34. Chugh A, Bakris GL. Microalbuminuria: what is it? Why is it important? What should be done about it? An update. J Clin Hypertens (Greenwich). 2007;9:196 –200. 35. Yuyun MF, Khaw KT, Luben R, et al. A prospective study of microalbuminuria and incident coronary heart disease and its prognostic significance in a British population: the EPIC-Norfolk study. Am J Epidemiol. 2004;159:284 –293. 36. Yudkin JS, Forrest RD, Jackson CA. Microalbuminuria as predictor of vascular disease in non-diabetic subjects. Islington Diabetes Survey. Lancet. 1988;2:530 –533. 37. Liu JJ, Tavintharan S, Yeoh LY, et al. High normal albuminuria is independently associated with aortic stiffness in patients with type 2 diabetes. Diabet Med. 2014;31:1199 –1204. 38. Rademacher E, Mauer M, Jacobs DR Jr, Chavers B, Steinke J, Sinaiko A. Albumin excretion rate in normal adolescents: relation to insulin resistance and cardiovascular risk factors and comparisons to type 1 diabetes mellitus patients. Clin J Am Soc Nephrol. 2008; 3:998 –1005. 39. Nguyen S, McCulloch C, Brakeman P, Portale A, Hsu CY. Being overweight modifies the association between cardiovascular risk factors and microalbuminuria in adolescents. Pediatrics. 2008;121: 37– 45. 40. Franchini S, Savino A, Marcovecchio ML, Tumini S, Chiarelli F, Mohn A. The effect of obesity and type 1 diabetes on renal function in children and adolescents [published online ahead of print August 11, 2014]. Pediatr Diabetes. doi:10.1111/pedi.12196. 41. Marcovecchio ML, Woodside J, Jones T, et al. Adolescent Type 1 Diabetes Cardio-Renal Intervention Trial (AdDIT): urinary screening and baseline biochemical and cardiovascular assessments. Diabetes Care. 2014;37:805– 813. 42. Maftei O, Pena AS, Sullivan T, et al. Early atherosclerosis relates to urinary albumin excretion and cardiovascular risk factors in adolescents with type 1 diabetes: Adolescent Type 1 Diabetes Cardiorenal Intervention Trial (AdDIT). Diabetes Care. 2014;37:3069 – 3075. 43. Chavers BM, Rheault MN, Foley RN. Kidney function reference values in US adolescents: National Health and Nutrition Examination Survey 1999 –2008. Clin J Am Soc Nephrol. 2011;6:1956 – 1962. 44. Wheeler DC, Becker GJ. Summary of KDIGO guideline. What do we really know about management of blood pressure in patients with chronic kidney disease? Kidney Int. 2013;83:377–383.


L9. NGAL, an early marker of endothelial dysfunction in preeclampsia.

Triglycerides-to-HDL ratio as a new marker of endothelial dysfunction in obese prepubertal children.

The Relation Between Neutrophil-Lymphocyte Ratio and Endothelial Dysfunction.

Endothelial dysfunction in patients with pulmonary thromboembolism: neutrophil to lymphocyte ratio and platelet to lymphocyte ratio.

YKL-40, a marker of cardiovascular disease and endothelial dysfunction, in kidney transplant recipients.

Asymmetric Dimethylarginin (ADMA) as a Marker of Endothelial Dysfunction in Primary Aldosteronism.

Allelic ratio: a marker of clonal dominance.

Plasma Neutrophil Gelatinase Associated Lipocalin Ratio Is a Sensitive and Specific Marker of Subclinical Acute Kidney Injury in Mice.

Neutrophil-lymphocyte ratio is a reliable predictive marker for early-stage diabetic nephropathy.

Corrigendum: A reduced lymphocyte ratio as an early marker for predicting acute pancreatitis.

Urine melatonin in alcoholic patients: a marker of alcohol abuse?

A high-performance liquid chromatography-tandem mass spectrometry-based targeted metabolomics kidney dysfunction marker panel in human urine.

Urine alpha1-microglobulin is a better marker for early tubular dysfunction than beta2-microglobulin among tenofovir-exposed human immunodeficiency virus-infected men who have sex with men.

CYFRA 21-1 in urine: a diagnostic marker for endometriosis?

Pentraxin 3 as a novel bio-marker of inflammation and endothelial dysfunction in autosomal dominant polycystic kidney disease.

Oxidative stress in kidney transplantation: malondialdehyde is an early predictive marker of graft dysfunction.

Endothelial dysfunction in joint disease.

Endothelial dysfunction in pulmonary hypertension.

Endothelial dysfunction as a nexus for endothelial cell-cardiomyocyte miscommunication.

Endothelial dysfunction in systemic sclerosis.

Increased plasma advanced oxidation protein products is an early marker of endothelial dysfunction in type 2 diabetes patients without albuminuria 2.

Endothelial dysfunction in vascular disease.

Left Ventricular Ejection Index as a Marker of Early Myocardial Dysfunction in Primary Mitral Regurgitation: Novel or Old in Disguise?

Bubble-Induced Endothelial Microparticles Promote Endothelial Dysfunction.