Albuminuria According to Status ofAutoimmunity and Insulin SensitivityAmong Youth With Type 1 and Type 2DiabetesAMY K. MOTTL, MD, MPH

1

ABIGAIL LAUER, MS2

DANA DABELEA, MD, PHD3

DAVID M. MAAHS, MD, PHD4

RALPH B. D’AGOSTINO JR., PHD2

LARRY M. DOLAN, MD5

LISA K. GILLIAM, MD6

JEAN M. LAWRENCE, SCD, MPH, MSSA7

BEATRIZ RODRIGUEZ, MD, PHD8

SANTICA M. MARCOVINA, PHD9

GIUSEPPINA IMPERATORE, MD, PHD10

ROOPA KANAKATTI SHANKAR, MBBS5

MARYAM AFKARIAN, MD, PHD11

KRISTI REYNOLDS, PHD, MPH7

ANGELA D. LIESE, PHD12

MICHAEL MAUER, MD13

ELIZABETH J. MAYER-DAVIS, PHD14

FOR THE SEARCH FOR DIABETES IN YOUTH

STUDY GROUP

OBJECTIVEdTo evaluate whether etiologic diabetes type is associated with the degree ofalbuminuria in children with diabetes.

RESEARCH DESIGN AND METHODSdSEARCH is an observational, longitudinalstudy of children with diabetes. Youth with newly diagnosed diabetes were classified accordingto diabetes autoantibody (DAA) status and presence of insulin resistance. We defined insulinresistance as an insulin sensitivity score ,25th percentile for the United States general youthpopulation. DAA status was based on positivity for the 65-kD isoform of glutamate decarboxylaseand insulinoma-associated protein 2 antigens. The four etiologic diabetes type groups were asfollows: DAA+/insulin-sensitive (IS) (n = 1,351); DAA+/insulin-resistant (IR) (n = 438); DAA2/IR(n = 379); andDAA2/IS (n = 233). Urinary albumin:creatinine ratio (UACR)wasmeasured from arandom urine specimen. Multivariable regression analyses assessed the independent relationshipbetween the four diabetes type groups and magnitude of UACR.

RESULTSdAdjusted UACR means across the four groups were as follows: DAA+/IS = 154 mg/mg; DAA+/IR = 137 mg/mg; DAA2/IR = 257 mg/mg; and DAA2/IS = 131 mg/mg (P , 0.005).Only DAA2/IR was significantly different. We performed post hoc multivariable regressionanalysis restricted to the two IR groups to explore the contribution of DAA status and insulinsensitivity (continuous) to the difference inUACR between the IR groups. Only insulin sensitivitywas significantly associated with UACR (b = 20.54; P , 0.0001).

CONCLUSIONSdIn youth with diabetes, the DAA2/IR group had a greater UACR than allother groups, possibly because of the greatermagnitude of insulin resistance. Further explorationof the relationships between severity of insulin resistance, autoimmunity, and albuminuria inyouth with diabetes is warranted.

P revious reports suggested that theclinical course and factors contrib-uting to the development and pro-

gression of diabetic nephropathy did notdiffer by diabetes type and that diabeticnephropathy was preceded by albumin-uria that worsened over time (1,2). Morerecent data have further elucidated thenatural history of diabetic kidney disease.In the absence of albuminuria, a signifi-cant number of people with diabetes, es-pecially type 2, still develop a decline inglomerular filtration rate (3,4). Thus,there may be identifiable differences inthe natural history of nephropathy inher-ent to the underlying diabetes type. Mul-tiple pediatric diabetes cohorts havefound a higher prevalence of albuminuriaand progressive kidney failure in youthwith a clinical diagnosis of type 2 diabetesthan with type 1 diabetes (5–7). Such datasuggest that insulin resistance, a key com-ponent of the pathophysiology of type 2diabetes, may be an important contribu-tor to diabetic nephropathy in youth withdiabetes.

The epidemic of overweight and obe-sity has made it increasingly difficult toclinically diagnose diabetes type, becauseinsulin resistance and autoimmunity of-ten coexist (8,9). Cohort studies of youthwith type 1 diabetes have found a sig-nificant increase in microvascular andmacrovascular diseases in those with con-current insulin resistance (10–12). The

c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c

From the 1University of North Carolina KidneyCenter, University of North Carolina School ofMedicine, Chapel Hill, North Carolina; the 2De-partment of Biostatistical Sciences, Wake ForestSchool of Medicine, Winston Salem, North Car-olina; the 3Department of Epidemiology, Schoolof Public Health, University of Colorado Denver,Aurora, Colorado; the 4Barbara Davis Centerfor Childhood Diabetes, University of ColoradoDenver, Aurora, Colorado; the 5Division of En-docrinology, Cincinnati Children’s Hospital Medi-cal Center, Cincinnati, Ohio; the 6Department ofMedicine, University of Washington, Seattle,Washington; the 7Department of Research andEvaluation, Kaiser Permanente Southern California,Pasadena, California; the 8Kuakini MedicalCenter, Honolulu, Hawaii; the 9Department of

Medicine, Northwest Lipid Metabolism and Di-abetes Research Laboratories, University ofWashington, Seattle, Washington; the 10Di-vision of Diabetes Translation, National Centerfor Chronic Disease Prevention and HealthPromotion, Centers for Disease Control andPrevention, Atlanta, Georgia; the 11Renal Re-search Institute, University of Washington,Seattle, Washington; the 12Department of Ep-idemiology and Biostatistics, Center for Re-search in Nutrition and Health Disparities,Arnold School of Public Health, University ofSouth Carolina, Columbia, South Carolina; the13Departments of Pediatrics and Medicine,University of Minnesota Medical School, Min-neapolis, Minnesota; and the 14Departments ofNutrition and Medicine, University of North

Carolina Schools of Public Health and Medicine,Chapel Hill, North Carolina.

Corresponding author: Amy K. Mottl, amy_mottl@med.unc.edu.

Received 7 March 2013 and accepted 1 May 2013.DOI: 10.2337/dc13-0568The contents of this paper are solely the responsibility

of the authors and do not necessarily represent theofficial position of the Centers for Disease Controland Prevention or the National Institute of Diabetesand Digestive and Kidney Diseases.

© 2013 by the American Diabetes Association.Readers may use this article as long as the work isproperly cited, the use is educational and not forprofit, and the work is not altered. See http://creativecommons.org/licenses/by-nc-nd/3.0/ fordetails.

care.diabetesjournals.org DIABETES CARE 1

P a t h o p h y s i o l o g y / C o m p l i c a t i o n sO R I G I N A L A R T I C L E

Diabetes Care Publish Ahead of Print, published online July 11, 2013

prevalence of albuminuria in insulin-resistant (IR) individuals with type 1diabetes has not been compared with indi-viduals with type 2 diabetes. Thus, the roleof autoimmunity and insulin resistanceacross the spectrum of diabetes types andthe risk for microvascular complicationswarrant investigation. Herein, we investi-gate the magnitude of albuminuria accord-ing to the status of autoimmunity andinsulin resistance in youth with newly di-agnosed type 1 and type 2 diabetes.

RESEARCH DESIGN ANDMETHODS

Study populationThe SEARCH for Diabetes in Youth Studyis a multicenter, observational study ofyouth with incident diabetes who arefollowed-up longitudinally. A detaileddescription of study methods has beenpublished previously (13). In brief,SEARCH is an ongoing study that beganin 2001 to conduct population-based as-certainment of cases of diabetes in youthyounger than 20 years of age. The studyprotocol was reviewed and approved bylocal institutional review boards that hadjurisdiction over the local study popula-tions. Cases were ascertained from geo-graphically defined populations in Ohio,Colorado, SouthCarolina, andWashington,among enrollees in several health plansin California and Hawaii, and IndianHealth Service beneficiaries from fourAmerican Indian populations. Youth iden-tified with incident nonsecondary diabeteswere invited to a baseline study visit. Self-reported race and ethnicity were collectedusing the 2000 United States Censusquestions (14).

Sample selection, design, andmeasurementsThis is a cross-sectional analysis explor-ing the association between diabetes eti-ologic group and the magnitude ofalbuminuria. Inclusion criteria includedhaving a baseline visit during whichfasting blood was drawn, urine was col-lected, and anthropometric measure-ments were taken. These were used tomeasure diabetes autoantibodies (DAAs)and covariates used to calculate the in-sulin sensitivity score (waist circumfer-ence, triglycerides [TG], hemoglobinA1c [HbA1c]) and to assess urine albu-min:creatinine ratio (UACR). Individualsusing ACE inhibitors or angiotensin re-ceptor blockers were excluded from theseanalyses (n = 38).

The study visit included measure-ment of waist circumference using theNational Health and Nutrition Examina-tion Survey protocol (15), systolic bloodpressure, diastolic blood pressure, height,and weight, as previously described (16).Height and weight were measured to cal-culate BMI (kg/m2), which was then con-verted to z-scores using the standardCenters for Disease Control and Preven-tion approach (17).

Blood and urine samples were ob-tained under conditions of metabolicstability, defined as no episodes of di-abetic ketoacidosis in the precedingmonth and the absence of fever and acuteinfections. Urine was not collected fromgirls who were menstruating. Participantsexcluded from analysis because of miss-ing urine sample (n = 1,064) had similarsociodemographic characteristics as theanalyzed cohort (data not shown). Speci-mens were processed locally and shippedwithin 24 h to the central laboratory(Northwest Lipid Metabolism and Dia-betes Research Laboratories, Seattle,WA). Measurements of serum choles-terol, TG, and HDL cholesterol were per-formed using Roche reagent on a RocheModule P autoanalyzer (Roche Diagnostics,Indianapolis, IN). HbA1c wasmeasured by adedicated ion-exchange high-performanceliquid chromatography instrument (TOSOHBioscience).

Random spot urine samples werecollected. Urinary creatinine was measuredby the Jaffemethod using Roche reagent onthe Roche Modular P autoanalyzer. Twoquality-control samples were analyzed ineach run, and the interassay coefficient ofvariation was consistently ,2%. Urinealbumin was measured immunochemi-cally using Siemens reagent on a SiemensBNII nephelometer. The sensitivity of theassay was also 0.2 mg/dL. The interassaycoefficient of variation is ,5% for thehigh-level and ,6.5% for the low-levelquality-control sample. Albuminuriawas defined as a UACR $30 mg/mg asrecommended by the American Diabe-tes Association guidelines (18) and Na-tional Kidney Disease Outcomes QualityInitiative (19).

Definitions of DAAs and insulinsensitivity or insulin resistanceBlood samples taken at the baseline visitwere analyzed for the 65-kD isoformof glutamate decarboxylase antibodies(GADA) and insulinoma-associated pro-tein 2 antibodies (IA-2A) using the Na-tional Institute of Diabetes and

Digestive and Kidney Diseases standard-ized method (20). The cutoff values forpositivity were 33 units/mL for GADAand 5 units/mL for IA-2A. The specificityand sensitivity were 97 and 76%, respec-tively, for GADA and 99 and 64%, respec-tively, for IA-2A (20). DAA positivity(DAA+) was defined by positive titers foreither GADA or IA-2A. Because many par-ticipants were treated with insulin, analy-sis of insulin autoantibodies was notperformed.

The insulin sensitivity score was cal-culated from variables measured at thestudy visit using the following equation:

Expf4:6472522 ð0:02032 � ½waist; cm�Þ2 ð0:002350 � ½TG;mg=dL�Þ2 ð0:09779 � ½HbA1c;%�Þg

This equation was developed and vali-dated using direct measurements ofglucose disposal rate from euglycemic-hyperinsulinemic clamps conductedamong 85 of the 2,401 SEARCH partici-pants included in this report and 22matched nondiabetic control subjects(21). As previously reported, we definedinsulin resistance among SEARCH partic-ipants in this study as an insulin sensitivityscore value,25th percentile for the UnitedStates general youth population (insulinsensitivity,8.15) (22).

Participants were assigned to one offour diabetes etiologic groups, accordingto the status of autoimmunity and insulinresistance at their baseline visit. Thesefour groups were as follows: DAA+/insulin-sensitive (IS); DAA+/IR; DAA2/IR; and DAA2/IS.

Statistical analysesStatistical analyses were performed usingSAS software version 9.1 (SAS Institute,Cary, NC) and S-PLUS software version6.0 (Insightful, Seattle,WA). Eachminoritygroup was limited in sample size; hence,for the present report, all racial/ethnicgroups other than non-Hispanic whitewere combined into a single “ethnic mi-nority” category. The distribution of eachpotential covariate was evaluated and,when necessary, logarithmically trans-formed for normalization of the distri-bution. The means and percents ofcovariates were compared across thefour etiologic groups using x2 and ANOVAtests when appropriate. Multivariable re-gression analyses assessed the relation-ship between the four etiologic groupsand the magnitude of UACR. Both the

2 DIABETES CARE care.diabetesjournals.org

Albuminuria among youth

Shapiro–Wilk test and Kolmogorov–Smirnov test indicated that the residualsdid not deviate significantly from a nor-mal distribution. A plot of residualsagainst the predicted values of the out-come variable found no evidence thatthe variance of the residuals changedacross the range of predicted values. Co-variates included in the model were age atvisit, sex, race/ethnicity, parental educa-tion and insurance type, clinic site, dia-betes duration, HbA1c, systolic bloodpressure z-score, and BMI z-score. Re-sults were considered significant if P ,0.05.

RESULTSdThe sociodemographic andclinical characteristics of the 2,401 par-ticipants, according to the four etiologicgroups, are depicted in Table 1. The eth-nic minority group comprised of 323 His-panics, 312 non-Hispanic blacks, 99Asians/Pacific Islanders, and 23 NativeAmericans/Alaska Natives. There weresignificant differences across the four eti-ologic groups for all covariates. The larg-est differences were in the DAA2 /IRgroup, which, in comparison with theother three groups, demonstrated a pre-ponderance of ethnic minorities and ele-vated systolic blood pressure, diastolicblood pressure, and TG levels. ElevatedUACR ($30 mg/mg) was prevalent in16% of the DAA2/IR group, which wassignificantly higher than that of all othergroups (P = 0.0007).

Multivariable analysis suggested thatthe etiologic groups significantly con-tributed to the variability of UACR (P =0.004). The adjusted mean UACR for theDAA2/IR group was significantly higherthan those of the other three groups (Ta-ble 2). All other pairwise comparisonswere nonsignificant (data not shown).

To explore reasons for the differencein UACR between the two IR groups, weperformed a post hoc t test on the meansof the insulin sensitivity scores andfound them to be significantly different(P, 0.0001). We then assessed the con-tribution of DAA status and insulin sen-sitivity to the difference in UACRbetween the two IR groups by per-forming a post hoc multivariable analysisrestricted to the IR participants. The re-gression equation used the originalmodel but incorporated DAA status andinsulin sensitivity (continuous) in placeof the four etiologic diabetes type groups.DAA status was not statistically signifi-cant (b = 0.18; P = 0.08), whereas insulinsensitivity was significantly and inverselyassociated with UACR (b = 20.54; P ,0.0001).

CONCLUSIONSdThis is the firststudy to compare the magnitude of albu-minuria in youth with diabetes classifiedaccording to markers of the underlyingetiology of diabetes using measures ofautoimmunity and insulin resistance. Wefound that in youthwith recently diagnosed

autoimmune-mediated diabetes, there wasno difference in UACR between those whowere IS compared with IR. There was,however, a significantly higher UACR inyouth without autoimmunity but with IRover all other subgroups. There were sig-nificant difference in covariates that couldbe confounders or mediators of the effectof etiologic subgroup; however, we sta-tistically controlled for this issue in ourmultivariable analysis. We hypothesizedthat the difference in albuminuria be-tween the two IR groups could be at-tributable to a greater severity of insulinresistance in the DAA2/IR group. Posthoc analyses showed insulin sensitivityto be significantly associated with UACRin the IR groups.

Ourfinding that therewas no differencein UACR between youth with autoimmune-mediated diabetes who were IS comparedwith IR was unexpected. The hypothesisthat insulin resistance in addition to auto-immunity could increase the risk of micro-vascular complications of diabetes wasproposed 20 years ago (23). Several studieshave since identified increases in bothmicrovascular and macrovascular com-plications in persons with type 1 diabeteswith versus without insulin resistance(11,12,24,25). It is difficult to comparethese studies with ours because of differ-ences in study population and methodol-ogies, especially our pediatric cohort withnewly diagnosed diabetes and estimationof insulin resistance.

Table 1dSociodemographic and clinical characteristics of 2,401 youth with type 1 or type 2 diabetes according to etiologic group:SEARCH for Diabetes in Youth Study

Variable*DAA+/IS DAA+/IR DAA2/IR DAA2/IS

P†n = 1,351 n = 438 n = 379 n = 233

Age at diagnosis, years 9.3 (3.7) 12.9 (2.8) 13.8 (2.6) 9.3 (4.0) ,0.0001Age at visit, years 10.3(3.7) 14.1 (2.9) 14.9 (2.7) 10.4 (4.0) ,0.0001Male, n (%) 729 (54) 221 (50) 145 (38) 137 (59) ,0.0001Ethnicity, NHW n (%) 1,044 (77) 310 (71) 105 (28) 177 (76) ,0.0001Diabetes duration, months 9.1 (5.8) 11.2 (6.7) 11.1 (6.9) 9.7 (6.2) ,0.0001‡HbA1c, % 7.4 (1.2) 8.6 (2.1) 7.7 (2.3) 7.2 (1.4) ,0.0001‡BMI z-score 0.3 (0.9) 1.2 (0.9) 2.0 (0.8) 0.2 (0.9) ,0.0001Insulin sensitivity score 11.7 (1.2) 5.9(1.4) 3.8 (1.9) 11.9 (1.3) ,0.0001‡SBP z-score 20.6 (0.9) 20.4 (0.9) 0.4 (1.1) 20.5 (0.9) ,0.0001DBP z-score 0.1 (0.8) 0.2 (0.8) 0.5 (0.9) 0.1 (0.7) ,0.0001Total cholesterol, mg/dL 158 (27) 168 (34) 175 (43) 157 (27) ,0.0001‡TG, mg/dL (median [25th, 75th]) 52 (41, 66) 82 (60, 113) 122 (83, 189) 53 (43, 69) ,0.0001‡HDL, mg/dL 56 (13) 51 (13) 42 (11) 55 (13) ,0.0001LDL, mg/dL 90 (23) 98 (26) 104 (36) 90 (22) ,0.0001‡UACR, mg/mg (median [25th, 75th]) 7 (5, 13) 6 (4, 13) 8 (5, 17) 7 (5, 12) 0.0146‡UACR .30 mg/mg, n (%) 123 (9) 53 (12) 59 (16) 17 (7) 0.0007

DBP, diastolic blood pressure; NHW, non-Hispanic white; SBP, systolic blood pressure. *Mean (SD) unless otherwise noted. †P value across all four categories usingx2 test or ANOVA when appropriate. ‡Using log distribution for association tests.

care.diabetesjournals.org DIABETES CARE 3

Mottl and Associates

The significant difference in UACRbetween the DAA+/IR and DAA2 /IRgroups also was unexpected. Post hocanalysis restricted to the two IR groupsindicated that differences in the degreeof IS between the two groups was associ-ated with the higher UACR in the DAA2/IR group. This finding, along with thelack of difference in UACR between theDAA+/IR and the two IS groups, suggest athreshold effect of IR on the magnitude ofalbuminuria. Our study design imposes alogical yet arbitrary cutpoint for IR to de-fine the four diabetes etiologic groups.Future studies exploring the shape ofthe relationship between insulin sensi-tivity and UACR are needed to clarifytheir relationship.

We are not the first to find an asso-ciation between the degree of insulinresistance and the degree of albuminuria(26–28). The components of metabolicsyndrome have long been identified asrisk factors for albuminuria, even in theabsence of diabetes (27,29). Recently,several theories have been proposed re-garding the mechanisms with which in-sulin resistance could lead to albuminuriaas well as other microvascular and macro-vascular complications (30–32). Visceraladipose tissue is unique in its ability tofunction as a metabolic and endocrino-logic organ (32). Macrophage infiltrationand adipokine production result in an in-flammatory and hormonal cascade thathas been found to have direct effects onendothelial and podocyte functionswithin the glomerulus (30,31,33). More-over, in the presence of adipokines, po-docytes have been noted to developaltered insulin signaling that rendersthem more susceptible to apoptosis, lead-ing to albuminuria (30,31). It is also im-portant to consider that the relationshipsof diabetic nephropathy lesions and albu-minuria are more complex in type 2 ver-sus type 1 diabetes. A substantial subset ofmicroalbuminuric type 2 patients may

have glomerular structural parameters inthe normal range, whereas this is rareamong type 1 patients (34). Thus, theprognostic significance of the albumin-uria findings in the current study will re-quire long-term follow-up.

There are several limitations to ourstudy. Most importantly, we collected onlya single random urine specimen. Ortho-static proteinuria is fairly common inchildren, as is intraindividual variation inUACR (35,36). However, both of these sit-uationswould result in nondifferentialmis-classification, which would bias the resultstoward the null hypothesis. Another limi-tation is the use of only two antibodies,IA-2A and GADA, for identification ofautoimmune-mediated diabetes. Recently,autoantibodies to zinc transporter 8 havebeen identified in the pathogenesis oftype 1 diabetes (37). Although there areplans to measure this autoantibody inSEARCH, it is not yet available. This couldhave resulted in misclassification in theDAA2/IR group; however, again, thiswould bias our results toward the nullhypothesis. In addition, we estimated in-sulin sensitivity rather than measuring itwith the euglycemic-hyperinsulinemicclamp. Euglycemic-hyperinsulinemicclamp studies are impractical for usein a large cohort of children because ofthe invasiveness and high cost. Our esti-mating equation was developed andvalidated in a subset of 85 SEARCH partic-ipants and 22 nondiabetic controls whounderwent euglycemic-hyperinsulinemicclamp study and was found to explain74% of the variance of glucose disposalrate (21). This is superior to a previouslypublished estimating equation that usedonly 24 participants and found their equa-tion to explain 57% of the variance in glu-cose disposal rate (38). We defined insulinresistance using an arbitrary cutpoint of the25th percentile in the insulin sensitivityscore in nondiabetic children. Previously,we performed sensitivity analyses using

higher and lower cutpoints of the IS scoreand found the groups to be extremely sim-ilar (22).

Strengths of our study include thelarge sample size and ethnic and geo-graphic diversity of the SEARCH cohort,which make our findings very general-izable. Moreover, SEARCH is a unique andvaluable resource because it includes youthwith newly diagnosed type 1 and type 2diabetes. No other study has had the capa-bility to explore the impact of insulin re-sistance on microvascular complicationsacross the diabetes etiologic spectrum in apediatric population.

In summary, using a novel approachto etiologic classification of diabetes type,we have been able to explore the associ-ation of diabetes etiologic subgroups withthe severity of albuminuria in newly di-agnosed youth with diabetes. Our resultssuggest that rather than the presence ofdiabetes autoimmunity, it may be the se-verity of insulin resistance that morestrongly associates with albuminuria. Thisfinding increases the urgency for investiga-tors to determine the pathogenesis andprognostic significance of albuminuria inthese young diabetic patients, and forclinicians and public health experts toaddress the overweight and obesity epi-demic in youth. Further analyses of thespectrum of insulin resistance and thresh-olds that may increase the risk for diabetickidney disease are warranted.

AcknowledgmentsdSEARCH for Diabetes inYouth is funded by the Centers for DiseaseControl and Prevention (PA numbers 00097,DP-05-069, and DP-10-001) and supported bythe National Institute of Diabetes and Digestiveand Kidney Diseases. Site contract numbers forthe study are as follows: Kaiser PermanenteSouthern California (U48/CCU919219, U01DP000246, and U18DP002714); University ofColorado Denver (U48/CCU819241-3, U01DP000247, andU18DP000247-06A1); KuakiniMedical Center (U58CCU919256 and U01DP000245); Children’s Hospital MedicalCenter, Cincinnati (U48/CCU519239, U01DP000248, and 1U18DP002709); University ofNorth Carolina at Chapel Hill (U48/CCU419249,U01 DP000254, and U18DP002708-01); Uni-versity of Washington School of Medicine(U58/CCU019235-4, U01 DP000244, andU18DP002710-01); and Wake Forest Univer-sity School ofMedicine (U48/CCU919219, U01DP000250, and 200-2010-35171). The authorsalso received support from involvement ofGeneral Clinical Research Centers (GCRC) atthe South Carolina Clinical and TranslationalResearch (SCTR) Institute at the Medical Uni-versity of South Carolina (National Institutes of

Table 2dUACR least-square means and pairwise comparisons with DAA2/IR groupin the SEARCH population*,†

Diabetes subcategory LS mean for UACR P

DAA+/IS 154 mg/mg 0.01DAA+/IR 137 mg/mg 0.0005DAA2/IR 257 mg/mg ReferentDAA2/IS 130 mg/mg 0.005

LS, least square. *All other pairwise comparisons were nonsignificant. †Covariates included in the multi-variable analysis were age at visit, sex, race/ethnicity, parental education and insurance type, clinic site, di-abetes duration, HbA1c, systolic blood pressure z-score, and BMI z-score.

4 DIABETES CARE care.diabetesjournals.org

Albuminuria among youth

Health/National Center for Research Resourcesgrant number UL1RR029882); Children’s Hos-pital and Regional Medical Center (grant num-ber M01RR00037); Colorado Pediatric GeneralClinical Research Center (grant number M01RR00069) and the Barbara Davis Center at theUniversity of Colorado at Denver (DERC Na-tional Institutes of Health P30 DK57516); andthe Institutional Clinical and Translational Sci-ence Award (CTSA) and National Institutes ofHealth/National Center for Research Resourcesat the University of Cincinnati (grant number1UL1RR026314-01).No potential conflicts of interest relevant to

this article were reported.A.K.M. designed the study, wrote the man-

uscript, interpreted the data, and made sub-stantial contributions to the manuscript. A.L.processed and analyzed the data. D.D. col-lected the data, designed the study, in-terpreted the data, and made substantialcontributions to the manuscript. D.M.M. in-terpreted the data and made substantial con-tributions to the manuscript. R.B.D. processedand analyzed data. L.M.D. collected the data,interpreted the data, and made substantialcontributions to the manuscript. L.K.G. in-terpreted the data and made substantial con-tributions to the manuscript. J.M.L. collecteddata, interpreted the data, and made sub-stantial contributions to the manuscript. B.R.collected the data. S.M.M. collected the data,interpreted the data, and made substantialcontributions to the manuscript. G.I. col-lected the data, interpreted the data, andmadesubstantial contributions to the manuscript.R.K.S. interpreted the data and made sub-stantial contributions to the manuscript. M.A.interpreted the data and made substantialcontributions to the manuscript. K.R. in-terpreted the data and made substantial con-tributions to the manuscript. A.D.L. collectedthe data. M.M. interpreted the data and madesubstantial contributions to the manuscript.E.J.M.-D. collected the data, designed thestudy, interpreted the data, and made sub-stantial contributions to the manuscript. A.K.M.is the guarantor of this work and, as such, hadfull access to all of the data in the study and takesresponsibility for the integrity of the data andthe accuracy of the data analysis.Parts of this studywere presented in abstract

form at the 72nd Scientific Sessions of theAmerican Diabetes Association, Philadelphia,Pennsylvania, 8–12 June 2012.The SEARCH for Diabetes in Youth Study is

indebted to the many youth and their families,and their health care providers, whose partic-ipation made this study possible.

References1. Pugh JA, Medina R, Ramirez M. Com-

parison of the course to end-stage renaldisease of type 1 (insulin-dependent) andtype 2 (non-insulin-dependent) diabeticnephropathy. Diabetologia 1993;36:1094–1098

2. Hasslacher C, Ritz E, Wahl P, Michael C.Similar risks of nephropathy in patientswith type I or type II diabetes mellitus.Nephrol Dial Transplant 1989;4:859–863

3. Kramer HJ, Nguyen QD, Curhan G, HsuCY. Renal insufficiency in the absence ofalbuminuria and retinopathy amongadults with type 2 diabetesmellitus. JAMA2003;289:3273–3277

4. Molitch ME, Steffes M, Sun W, et al.;Epidemiology of Diabetes Interventionsand Complications Study Group. De-velopment and progression of renal in-sufficiency with and without albuminuriain adults with type 1 diabetes in the di-abetes control and complications trial andthe epidemiology of diabetes interventionsand complications study. Diabetes Care2010;33:1536–1543

5. Maahs DM, Snively BM, Bell RA, et al.Higher prevalence of elevated albuminexcretion in youth with type 2 than type 1diabetes: the SEARCH for Diabetes inYouth study. Diabetes Care 2007;30:2593–2598

6. Dart AB, Sellers EA, Martens PJ, Rigatto C,Brownell MD, Dean HJ. High burden ofkidney disease in youth-onset type 2 di-abetes. Diabetes Care 2012;35:1265–1271

7. Eppens MC, Craig ME, Cusumano J, et al.Prevalence of diabetes complications inadolescents with type 2 compared withtype 1 diabetes. Diabetes Care 2006;29:1300–1306

8. Tripathi A, Rizvi AA, Knight LM, JerrellJM. Prevalence and impact of initial mis-classification of pediatric type 1 diabetesmellitus. South Med J 2012;105:513–517

9. Liu LL, Lawrence JM, Davis C, et al.;SEARCH for Diabetes in Youth StudyGroup. Prevalence of overweight andobesity in youth with diabetes in USA: theSEARCH for Diabetes in Youth study.Pediatr Diabetes 2010;11:4–11

10. Nadeau KJ, Regensteiner JG, Bauer TA,et al. Insulin resistance in adolescents withtype 1 diabetes and its relationship tocardiovascular function. J Clin EndocrinolMetab 2010;95:513–521

11. Kilpatrick ES, Rigby AS, Atkin SL. Insulinresistance, the metabolic syndrome, andcomplication risk in type 1 diabetes:“double diabetes” in the Diabetes Controland Complications Trial. Diabetes Care2007;30:707–712

12. Orchard TJ, Chang YF, Ferrell RE,Petro N, Ellis DE. Nephropathy in type1 diabetes: a manifestation of insulinresistance and multiple genetic suscep-tibilities? Further evidence from thePittsburgh Epidemiology of DiabetesComplication Study. Kidney Int 2002;62:963–970

13. SEARCH Study Group. SEARCH for Di-abetes in Youth: a multicenter study of the

prevalence, incidence and classification ofdiabetes mellitus in youth. Control ClinTrials 2004;25:458–471

14. Ingram DD, Parker JD, Schenker N, et al.United States Census 2000 populationwith bridged race categories. Vital HealthStat 2 2003 (135):1–55

15. Fern�andez JR, Redden DT, Pietrobelli A,Allison DB. Waist circumference percen-tiles in nationally representative samplesof African-American, European-American,and Mexican-American children and ado-lescents. J Pediatr 2004;145:439–444

16. Rodriguez BL, FujimotoWY,Mayer-DavisEJ, et al. Prevalence of cardiovasculardisease risk factors in U.S. children andadolescents with diabetes: the SEARCH fordiabetes in youth study. Diabetes Care2006;29:1891–1896

17. Ogden CL, Kuczmarski RJ, Flegal KM,et al. Centers for Disease Control andPrevention 2000 growth charts for theUnited States: improvements to the 1977National Center for Health Statistics ver-sion. Pediatrics 2002;109:45–60

18. American Diabetes Association. Standardsof medical care in diabetesd2011. Di-abetes Care 2011;34(Suppl. 1):S11–S61

19. KDOQI. KDOQI Clinical Practice Guide-lines andClinical PracticeRecommendationsfor Diabetes and Chronic Kidney Disease.Am J Kidney Dis 2007;49(Suppl. 2):S12–S154

20. Bonifacio E, Yu L, Williams AK, et al.Harmonization of glutamic acid decar-boxylase and islet antigen-2 autoantibodyassays for national institute of diabetesand digestive and kidney diseases con-sortia. J Clin Endocrinol Metab 2010;95:3360–3367

21. Dabelea D, D’Agostino RB Jr, Mason CC,et al. Development, validation and use ofan insulin sensitivity score in youths withdiabetes: the SEARCH for Diabetes inYouth study. Diabetologia 2011;54:78–86

22. Dabelea D, Pihoker C, Talton JW, et al.;SEARCH for Diabetes in Youth Study.Etiological approach to characterizationof diabetes type: the SEARCH for Diabetesin Youth Study. Diabetes Care 2011;34:1628–1633

23. Teupe B, Bergis K. Epidemiological evi-dence for “double diabetes.” Lancet 1991;337:361–362

24. Chillarón JJ, Goday A, Flores-Le-Roux JA,et al. Estimated glucose disposal rate inassessment of the metabolic syndromeand microvascular complications in pa-tients with type 1 diabetes. J Clin Endo-crinol Metab 2009;94:3530–3534

25. Olson JC, Erbey JR, Williams KV, et al.Subclinical atherosclerosis and estimatedglucose disposal rate as predictors ofmortality in type 1 diabetes. Ann Epi-demiol 2002;12:331–337

26. Rowley KG, Iser DM, Best JD, O’Dea K,Leonard D, McDermott R. Albuminuria in

care.diabetesjournals.org DIABETES CARE 5

Mottl and Associates

Australian Aboriginal people: prevalenceand associations with components of themetabolic syndrome. Diabetologia 2000;43:1397–1403

27. Mykkänen L, Zaccaro DJ, WagenknechtLE, Robbins DC, Gabriel M, Haffner SM.Microalbuminuria is associatedwith insulinresistance in nondiabetic subjects: the in-sulin resistance atherosclerosis study. Di-abetes 1998;47:793–800

28. Nosadini R, Cipollina MR, Solini A,et al. Close relationship between mi-croalbuminuria and insulin resistancein essential hypertension and non-insulindependent diabetes mellitus. J Am SocNephrol 1992;3(Suppl.):S56–S63

29. Bonnet F, Marre M, Halimi JM, et al.;DESIR Study Group. Waist circumferenceand the metabolic syndrome predict thedevelopment of elevated albuminuria in

non-diabetic subjects: the DESIR Study. JHypertens 2006;24:1157–1163

30. Diez-Sampedro A, Lenz O, Fornoni A.Podocytopathy in diabetes: a metabolicand endocrine disorder. Am J Kidney Dis2011;58:637–646

31. Welsh GI, Coward RJ. Podocytes, glucoseand insulin. Curr Opin Nephrol Hyper-tens 2010;19:379–384

32. Antuna-Puente B, Feve B, Fellahi S,Bastard JP. Adipokines: the missing linkbetween insulin resistance and obesity.Diabetes Metab 2008;34:2–11

33. Kang YS, Song HK, Lee MH, et al. Visfatinis upregulated in type-2 diabetic rats andtargets renal cells. Kidney Int 2010;78:170–181

34. Fioretto P, Stehouwer CD, Mauer M, et al.Heterogeneous nature of microalbuminuriain NIDDM: studies of endothelial function

and renal structure. Diabetologia 1998;41:233–236

35. Brandt JR, Jacobs A, Raissy HH, et al.Orthostatic proteinuria and the spectrumof diurnal variability of urinary proteinexcretion in healthy children. PediatrNephrol 2010;25:1131–1137

36. McHardy KC, Gann ME, Ross IS, PearsonDW. A simple approach to screeningfor microalbuminuria in a type 1 (insulin-dependent) diabetic population. Ann ClinBiochem 1991;28:450–455

37. Énée E, Kratzer R, Arnoux JB, et al. ZnT8is a major CD8+ T cell-recognized auto-antigen in pediatric type 1 diabetes. Di-abetes 2012;61:1779–1784

38. WilliamsKV,Erbey JR, BeckerD,ArslanianS,Orchard TJ. Can clinical factors estimateinsulin resistance in type 1 diabetes? Di-abetes 2000;49:626–632

6 DIABETES CARE care.diabetesjournals.org

Albuminuria among youth