The CASPIAN-V Study

Arch Metab. Endocrinol 2020;64/2 ratesofMetSranged between0and19.2% prevalence MetS isnotlimitedtodeveloped countries(3). andlifestylechanges. ofnutritional disorders, burden factors includingepidemiologic transition,double and adolescents;becauseofvarious inchildren problem During lastdecades,MetShasbecomeamajorhealth earlylife(2). (1).MetS canoccurfrom blood pressure obesity, glucoseintolerance,dyslipidemiaandraised M INTRODUCTION Metabolic syndrome;bodymassindex;tri-ponderalpediatric Keywords Endocrinol Metab. 2020;64(2):171-8 especiallyinagegroupsof11-14than BMIinbothgenders, and15-19 forgirlsandboys. years valueforidentifyingMetS.However,BMI hadmoderatepredictive TMI was abetter predictorofMetS (OR =1.33 vs1.22) and15-18 (1.16 years vs1.15) ingirlsandboys,respectively. our findingsshowedthatoddsratioofMetsfor TMI was greaterthanBMIinagegroupsof 11-14 years also greaterthanBMIforagegroupof11-14 (AUC years =0.74; 95%CI(0.67, 0.81))ingirls.Furthermore, In boys,theareaundercurve(AUC) of TMI was greaterthanBMIforallagegroups. AUC of TMI was mean (standarddeviations)ageforboysandgirlswere12.62 respectively. (3.02)and12.25 (3.05)years, assess therelationshipbetweentheseindiceswithMetS. TMI forMetSweredeterminedusingReceiver-operator curves.Logistic regressionanalysiswas usedto Prevention of Adult Non-communicableDisease’ (CASPIAN-V) study. valueofBMI and The predictive aged 7-18 obtainedfromthefifth years surveyof‘Childhoodand AdolescenceSurveillanceand Subjects andmethods: value of TMI andBMIformetabolicsyndrome(Mets)inchildren andadolescentsofbothgenders. evaluate bodyadiposityinthevariousagegroups. The presentstudy aims tocomparethepredictive Objective: ABSTRACT https://orcid.org/0000-0001-7455-1495 Roya Kelishadi https://orcid.org/0000-0002-8134-7940 Ramin Heshmat https://orcid.org/0000-0002-7421-0196 Hasan Ziaodini https://orcid.org/0000-0002-6101-548X Mohammad EsmaeilMotlagh https://orcid.org/0000-0001-9465-7588 Mostafa Qorbani https://orcid.org/0000-0001-5543-5536 Motahar Heidari-Beni https://orcid.org/0000-0002-2883-8652 Khoshhali Mehri the CASPIAN-V study metabolicpediatric syndrome: body massindex inprediction of Tri-ponderal mass index and A systematic review determined the worldwide determined A systematic review of cardio metabolicriskfactors,includingcentral of cardio (MetS)isdefinedasacluster etabolic syndrome Body mass index (BMI) and tri-ponderal mass index (TMI) are anthropometric measures to Body massindex(BMI)andtri-ponderal(TMI)areanthropometricmeasuresto 8 6 7 1 3,4 2 A cross-sectional study conducted on3731Iranian children andadolescents 5 Results: 52.6% ofparticipants were boys. The period (11). Several studies determined that the body period (11). Several studies determined thegrowing withheightacross but uncorrelated withweight closelycorrelated height indices,whichare theoptimalweight- todetermine (10). Itisimportant inthe21 publichealthconcerns important adolescence have emerged as one of the most of 1-22%inIran(9). ratesof MetS intherange prevalence reported review (5-8).Asystematic amongIranianchildren disorder studies indicatedthatMetSisacommonmetabolic andadolescentsin2012(4).Previous among children Obesity and overweight inchildhoodand Obesity andoverweight

Conclusion: TMI and TMI and Arch Arch

6 Medical Sciences, Ahvaz, Iran Ahvaz Jundishapur Universityof 5 Medical Sciences, Tehran, Iran Institute, Tehran Universityof PopulationMetabolism Sciences Research Center, Endocrinology, and Non-Communicable Diseases 4 of MedicalSciences, Tehran, Iran Sciences Institute, Tehran University Clinical Endocrinology andMetabolism Researchand Metabolism Center, Sciences, Karaj,Iran,Endocrinology Center, University ofMedical Alborz 3 Medical Sciences,Isfahan, Iran Disease, Isfahan University of Prevention ofNon-Communicable Center, Research Institute for Primordial Growth andDevelopment Research 2 Medical Sciences,Isfahan Disease, Isfahan University of Prevention ofNon-Communicable Center, Research Institute for Primordial Growth andDevelopment Research 1 DOI: 10.20945/2359-3997000000206 DOI: 10.20945/2359-3997000000206 Accepted onMar/29/2019 Received onJuly/30/2018 [email protected] Isfahan, Iran ofMedicalSciences, University Disease, Isfahan Prevention ofNon-communicable Research Institute for Primordial Research Center, Child Growth andDevelopment Department ofPediatrics, Roya Kelishadi C Medical Sciences,Isfahan, Iran Disease, Isfahan University of Prevention ofNon-Communicable Center, Research Institute for Primordial Growth andDevelopment Research 8 of MedicalSciences, Tehran, Iran Sciences Institute, Tehran University PopulationEndocrinology andMetabolism 7 Education Ministry, Tehran, Iran ChronicDiseasesResearch Center, HealthPsychology Research Center, DepartmentofPediatrics, DepartmentofEpidemiology, Non-CommunicableDiseasesResearch DepartmentofPediatrics, Child DepartmentofPediatrics, Child DepartmentofPediatrics, Child orrespondence to: original article original st century century 171

Copyright© AE&M all rights reserved. Copyright© AE&M all rights reserved. 172 itinbrief. wereport (27)andhere before havebeenexplained Details onthestudyprotocol inI wasconducted in30provinces health survey (CASPIAN-V) study. Thisschool-basednationwide and entitled system oftheschool-basedsurveillance fifth survey student highriskbehaviors”(2014-2015),asthe ofschool ofthe“Nationalsurvey collected asapart studywere The dataofthisnationwidecross-sectional Study designandsample METHODS ANDMATERIALS values forBMIandTMIbygenderagegroups. theoptimalcutoff adolescents and(b)todetermine and among alarge populationofIranianchildren ofMets ability ofTMIandBMIfortheprediction predictive (a)tocompare purposes ofthisstudywere andyouth (26).The population ofColombianchildren FMI ((fat mass)/height ofgender-specific thresholds that explored TMIand However, isonlyonestudy toourknowledge,there andadolescents (22-25). its componentsinchildren withMetSand measures between anthropometric training (21).Manystudieshaveindicatedassociation withminimal highly technologicequipmentbystaff without cost, non-invasiveandcanbeperformed (1995-2017) (20). study longitudinalgrowth intheBarcelona children non-obesemillennial age forhealthynon-underweight, values of BMI and TMI for reference study reported BMI inadolescentsaged8to17years(19).Another accuratelythan more estimated bodyfatpercentage studies (14).Recently, PetersonetalshowedthatTMI in the still not clear due topopulationdifferences p are p byageandgenderwasspecified.However, of trends adiposityininfants(17,18). determine showed thattheaccuracyofTMIishigherthanBMIto used asanindicatoroffetalnutrition.Severalstudies originally suggestedasacorpulenceindexandhasbeen andcols.(16).Itwas was devisedin1921byRohrer (15). Traditionally,for newborns the ponderal index body weight(kg)/height tri-ponderalmassindex(TMI: (12-14). Furthermore, andadult children indexinpreschool is apreferable mass TMI andBMIinpredicting pediatricMetS The anthropometric measures are reliable, low reliable, are measures The anthropometric of Power (p) inWeight/HeightThe general trend P index (BMI:bodyweight(kg)/height revent C hildhood and I on of A dult 3 ) for the prediction of MetS in ) for the prediction N 3 (m A on-communicable Disease dolescence 3 )) maybeapplicable S urveillance urveillance 2 (m ran. 2 )) measured without shoes to the nearest 0.1cm (28). withoutshoestothenearest measured while subjectswearingalightcloth,andheightwere 0.1 kg tothe nearest on ascaleplacedflatground Weightby usingcalibratedinstruments. wasmeasured protocols theexaminationsunderstandard performed checklists;they basedonapproved information recorded experts A team of trained health care Physical measurements blood sampling. randomlyselectedfor sex ratio.4200studentswere to size method and with equal using the proportional andsecondary) andlevelofeducation(primary rural) (urbanor tothestudent’splaceofresidence according wasconducted method. Samplingwithineachprovince selected using multistage stratified cluster sampling thecountry. across areas Theywere urban and rural schools in andsecondary 7-18 years in primary for the pediatric age group (33). for thepediatricagegroup Adult Treatment PanelIII(ATP III)criteriamodified tothe of thefollowingcriteriaaccording least three classifiedashavingMetS,iftheyhadat Subject were MetS components (Tokyo, Japan)(31,32). enzymatically by Hitachi auto148 analyzer measured (HDL-C)were high-density lipoprotein-cholesterol (LDL-C)and low-density lipoprotein-cholesterol (TC), (FBG), triglycerides (TG), total cholesterol fast.Fastingbloodglucose after 12–14hofovernight students obtainedfrom Fasting bloodsampleswere Blood sampling (30). andtheaveragewas registered recorded were systolicanddiastolicpressures at 5-minintervals; 2 times size. It was measured cuff with an appropriate sphygmomanometer usingamercury on therightarm cm (29). 0.1 expirationtothenearest attheendofnormal crest of the rib cage and the iliac between the lower border usinganon-elastictapeat a pointmidway was measured (kg) toheightcubed(m (m squared BMI wascalculatedbydividingweight(kg)toheight The studypopulationconsistedofstudentsaged 1) inthesittingposition wasmeasured Blood pressure

height ratio equal to or more than0.5(34). height ratioequal toormore Abdominal obesity was defined as waist-to- 2 ) andTMIwascalculateddividingweight 3 ). Waist (WC) circumference Arch Metab. Endocrinol 2020;64/2 Arch Metab. Endocrinol 2020;64/2 BMI: bodymassindex; TMI: tri-ponderalmassindex. Table 1. Youden index(sensitivity+specificity−1)bygender. obtainedusing the were BMI, TMIandtheirz-score accuracy. valuesfor theoptimalcutoff Furthermore, under theROC(AUC)valueequalto1meansperfect Area operatingcharacteristic(ROC) curve. receiver under the forMetSwascalculatedbyareas z-scores weight andheightalsoMetScomponents. associationsbetweenBMIandTMIwith determine independent Studentttest. using performed in boys and girls were measures (SD). Comparisons between means of anthropometric deviation asmean±standard represented are The results Statistical analysis TMI (kg/m BMI (kg/m Height (cm) Weight (kg) Gender The discriminatory powerofBMI,TMIandtheir The discriminatory usedto were coefficients Pearson correlation 5) 4) 3) 2) Characteristics ofparticipantsbyageandgender: theCASPIAN-VStudy for age,sexandheight)(30). (DBP)(≥90 diastolic bloodpressure forage,sexandheight)orhigh percentile (SBP)(≥90 high systolicbloodpressure (35). was<45mg/dL) 15-18 yinwhomthecut-off Elevated blood pressure wasdefinedaseither Elevated bloodpressure HDL-C < 40 mg/dL (except in boy Low serum TG≥100mg/dL. High serum Elevated FBG≥100mg/dL. 3 2 ) ) Age group(yr) 15-18 11-14 15-18 11-14 15-18 11-14 15-18 11-14 7-10 7-10 7-10 7-10 Total Total Total Total 2012 2012 2012 2012 602 845 565 602 845 565 602 845 565 602 845 565 N th percentile percentile Boys 148.44 ±18.56 164.80 ±16.74 148.56 ±12.13 130.78 ±10.62 42.27 ±17.05 56.85 ±17.85 41.07 ±11.89 12.50 ±2.90 12.46 ±2.75 12.37 ±2.57 12.74 ±3.46 18.46 ±4.18 20.45 ±4.40 18.33 ±3.64 16.54 ±3.74 28.53 ±8.25 Mean ±SD th

of weight, height and BMI in boys were more thangirls more of weight,heightandBMIinboyswere by age and gender.BMI and TMI of participants Mean girls, respectively. Table 1showstheweight,height, was 12.62(3.02)and12.25(3.05)yearsforboys deviation)age boys.Themean(standard them were ratewas91.5%.Intotal,52.4%of The participation RESULTS asstatisticallysignificant. considered were analysismethod.P-valueslessthan0.05 using survey Texas, USA).Allstatisticalanalysiswasperformed STATAsoftware 12.0(STATA Corp,CollegeStation, (CI). confidence interval asoddsratio(OR)and95% presented are regression of logistic model by gender and age. Theresults with riskofMetSwasevaluatedusinglogisticregression The associationbetweenTMI,BMIandtheirz-score in girls was significantly greater thanboys ( in girlswassignificantlygreater than girls( in boys significantly greater weight andheightwere ( meanofweightandheight thanboys greater of11-14years,girlshad significantly in theagegroup of 7-10 years ( in theage group p < 0.05). In the age group of15-18years, meanof <0.05).Intheagegroup The analyses data were performed usingstatistical performed The analysesdatawere p < 0.05) whereas meanofBMIandTMI <0.05)whereas 1831 1831 1831 1831 440 809 582 440 809 582 440 809 582 440 809 582 N Girls TMI andBMIinpredicting pediatricMetS 145.52 ±16.12 159.86 ±10.41 149.86 ±11.37 128.85 ±9.82 40.45 ±15.46 54.93 ±13.42 42.49 ±12.33 12.66 ±2.70 13.31 ±2.49 12.43 ±2.41 12.48 ±3.13 18.42 ±4.28 21.27 ±4.15 18.62 ±3.82 16.01 ±3.51 26.85 ±7.59 Mean ±SD p < 0.05). However, p <0.05). < 0.001 < 0.001 < 0.001 < 0.001 0.665 0.189 0.001 0.003 0.117 0.015 0.096 0.029 0.002 0.757 0.063 0.019 P 173

Copyright© AE&M all rights reserved. Copyright© AE&M all rights reserved. 174 AUC: BMI: areaunder thereceiveroperatingcurve; bodymassindex; TMI: tri-ponderalmassindex. Table 3. * P-value<0.05. density lipoprotein;BMI: bodymassindex; TMI: tri-ponderalmassindex. WC: waistcircumference;SBP: systolicbloodpressure;DBP: diastolicbloodpressure;FBG: fastingbloodglucose; TG: triglycerides; TC: totalcholesterol;HDL: high densitylipoprotein;LDL: low Table 2. for coefficients correlation and TMI,withthegreatest withBMI hadsignificantcorrelation diastolic pressures WCandsystolic with weight.Furthermore, correlation BMIhadthegreatest boys in all age groups, showedinTablecomponents are 2.Forbothgirlsand of BMIandTMIwiththeweight,heightMetS TMI andBMIinpredicting pediatricMetS Girls Boys Gender/Indices Girls Boys TMI BMI T BMI The correlation coefficients between the indices coefficients The correlation M I Correlation betweenindicesofBMIand TMI withweight, heightandcomponentsofmetabolicsyndrome: theCASPIAN-VStudy ROC curve analysisforindicesofBMIand ROC curve TMI aspredictorsofmetabolicsyndromebygenderandage: theCASPIAN-VStudy BMI BMI TMI TMI group (yr) 15-18 11-14 15-18 11-14 15-18 11-14 15-18 11-14 7-10 7-10 7-10 7-10 Age Age groups(yr) 15-18 11-14 15-18 11-14 15-18 11-14 15-18 11-14 7-10 7-10 7-10 7-10 Total Total Total Total Weight 0.72* 0.64* 0.47* 0.90* 0.88* 0.77* 0.44* 0.47* 0.41* 0.84* 0.82* 0.74* Height -0.26* -0.17* -0.20* -0.36* 0.27* 0.34* 0.10* 0.34* 0.26* -0.05 -0.03 0.02 cutoffs 13.26 19.65 23.05 17.71 13.43 15.35 14.34 19.61 20.95 19.73 18.09 13.17 13.26 12.19 14.8 20.7 0.64* 0.47* 0.20* 0.74* 0.65* 0.44* 0.44* 0.43* 0.19* 0.73* 0.67* 0.44* WC 0.27* 0.12* 0.16* 0.15* 0.28* 0.26* 0.21* 0.23* 0.13* 0.16* 0.31* 0.24* SBP Sensitivity 45.57% 76.92% 51.28% 44.44% 60.76% 76.92% 69.23% 55.56% 63.21% 64.71% 84.44% 66.67% 63.21% 61.76% 60.00% 59.26% TMI z-score. In all age groups forboys,TMIand Inallagegroups TMI z-score. sensitivity, and specificity andAUCofBMIz- score TMI. Table corresponding 4showsoptimalcutoffs, sensitivity andspecificity, AUCforBMIand alsosignificantinboysaged15-18years. and LDLwere betweenBMI and TMIwithTG BMI. The correlation Table 3 shows the optimal cutoffs, corresponding Table corresponding 3showstheoptimalcutoffs, 0.22* 0.09* 0.16* 0.16* 0.23* 0.22* 0.22* 0.18* 0.08* 0.15* 0.23* 0.15* DBP Specificity -0.001 0.014 0.003 86.18% 57.88% 89.92% 87.69% 68.27% 54.19% 68.68% 84.14% 72.75% 71.69% 58.33% 72.99% 69.72% 81.31% 81.54% 75.67% -0.01 -0.01 -0.04 -0.06 -0.04 -0.02 -0.02 -0.01 -0.01 FBG -0.0001 0.12* 0.10* -0.02 -0.02 0.02 0.04 0.03 0.02 0.01 0.02 0.03 TG Arch Metab. Endocrinol 2020;64/2 0.68 0.64 0.74 0.67 0.67 0.64 0.74 0.71 0.72 0.70 0.74 0.73 0.69 0.67 0.74 0.69 AUC -0.003 0.005 0.003 0.003 -0.02 -0.06 -0.03 -0.03 0.05 0.03 0.02 0.06 TC -0.08* -0.05 -0.01 -0.01 -0.02 -0.04 -0.06 -0.06 -0.06 0.02 0.01 0.01 HDL (0.61, 0.74) (0.47, 0.81) (0.67, 0.81) (0.56, 0.79) (0.61, 0.73) (0.48, 0.81) (0.66, 0.81) (0.60, 0.81) (0.67, 0.78) (0.59, 0.80) (0.67, 0.81) (0.63, 0.83) (0.63, 0.75) (0.55, 0.79) (0.66, 0.81) (0.58, 0.80) 95% CI -0.0003 -0.003 0.10* 0.10* -0.01 0.01 0.03 0.02 0.01 0.01 0.04 0.02 LDL Arch Metab. Endocrinol 2020;64/2 AUC: BMI: areaunder thereceiveroperatingcurve; bodymassindex; TMI: tri-ponderalmassindex. Table 4. BMI: bodymassindex; TMI: tri-ponderalmassindex. Figure 1. MetS inthetotaldata. of indexesfor prediction ofanthropometric curves ROC 1represents using thesamemethod.Figure obtained were agegroups BMI andTMIindifferent for 7-10 and11-15yearsingirls.Theoptimalcutoffs of forgroups wasalsogreater the AUCofBMIz-score of7-10yearsand thanTMIforagegroup was greater ofMets.However,for theprediction theAUCofBMI AUCthanBMIandz-score hadgreater z-score Girls Boys Gender/Indices Sensitivity TMI z-score BMI z-score TMI z-score BMI z-score 0.00 0.25 0.50 0.75 1.00 0.00 ROC curve analysisforindicesofBMIz-scoreand ROC curve TMI z-scoreaspredictorsofmetabolicsyndromebygenderandage: theCASPIAN-VStudy Roc curve ofanthropometricindicesforpredictingmetabolicsyndrome inchildrenandadolescents:Roc curve The CASPIAN-Vstudy. 0.25 M O ra .0 BMIROCarea:0.68 Reference TMI ROCarea:0.70 Age groups(yr) 1- Specificity 15-18 11-14 15-18 11-14 15-18 11-14 15-18 11-14 7-10 7-10 7-10 7-10 Total Total Total Total 0.50 0.75 cutoffs 0.91 0.25 0.61 0.88 0.68 0.16 0.36 0.75 0.33 0.82 0.16 0.42 0.56 0.78 0.21 0.88 1.00 Sensitivity 46.15% 69.23% 57.89% 48.15% 53.13% 69.23% 68.42% 55.56% 62.38% 56.25% 70.45% 64.00% 57.43% 59.38% 70.45% 48.45%

Sensitivity of15-18yearsinboys. age group of11-14and15-18yearsingirls for agegroups thanBMIandz-score greater were TMI z-score of15-18years.ThevaluesORforTMIand group exceptingirlswithage and boysforallagegroups foundbetweenBMIandTMIMetSingirls were MetS by age and gender. The significant associations with association between BMI, TMI and theirz-scores 0.00 0.25 0.50 0.75 1.00 0.00 Table the ORs and 95%CIfor the 5 presents Reference TMI z-scoreROCarea:0.72BMI0.70 Specificity 82.099% 79. 81% 85.26% 62.87% 76.80% 84.66% 61.39% 70.56% 82.95% 72.32% 68.71% 77.82% 78.65% 83.40% 67.01% 83.46% 0.25 1- Specificity TMI andBMIinpredicting pediatricMetS 0.50 0.69 0.65 0.71 0.69 0.71 0.65 0.73 0.71 0.72 0.70 0.74 0.72 0.70 0.68 0.73 0.69 AUC 0.75 (0.63, 0.76) (0.47, 0.82) (0.62, 0.80) (0.58, 0.80) (0.65, 0.77) (0.48, 0.81) (0.64, 0.82) (0.60, 0.82) (0.67, 0.78) (0.60, 0.81) (0.67, 0.82) (0.61, 0.83) (0.65, 0.76) (0.56, 0.80) (0.65, 0.81) (0.57, 0.80) 95% CI 1.00 175

Copyright© AE&M all rights reserved. Copyright© AE&M all rights reserved. 176 study.present (36),whichwassimilartothefindingsof puberty that TMIisabetteradiposity indexthanBMIduring studiesshowed (13). Thefindingsofprevious terms inpercentage thantwicetheheightspurt ismore spurts statistically significant. than BMIinbothgirlsandboysbutwasnot of15-18years,TMIwasgreater girls; andinagegroup in thanBMI(BMIz-score) wasgreater (TMI z-score) of11-14years,TMI girls andboys;intheagegroup inboth washigherthanTMI(TMIz-score) z-score) of7-10years,theriskMetsforBMI(BMI group ourfindingsshowedthatinthe age Furthermore, 19.61 (0.68)and14.8(0.91)ingirls,respectively. 19.65(0.56)and13.17(0.33) inboysand MetS were forpredicting andTMI(TMIz-score) (BMI z-score) forBMI of 11-14yearsforgirls.Theoptimalcutoffs andin theagegroup of Metsinboysforallagegroups power ofTMIwashigherthanBMIintheprediction population. Thisstudyindicatedthatthepredictive accuracy ofTMIandBMIindicesforMetsinpediatric our knowledge,thisisthefirststudycomparing on adiposi BMIand TMI studiesassessedandcompared Previous DISCUS OR: oddsratio;CI: BMI: confidenceinterval; bodymassindex; TMI: tri-ponderalmassindex. CASPIAN-V study Table 5. TMI andBMIinpredicting pediatricMetS Total 15-18 11-14 7-10 Total 15-18 11-14 7-10 Age groups(yr) According to findings, weight spurt during puberty during puberty tofindings,weightspurt According Odds ratio(95%CI)fortheassociationbetweenanthropometricindiceswithriskofmetabolicsyndromebyageandgender: The SION ty (13)andbodyfat(19).To thebestof TMI z-score TMI TMI z-score TMI TMI z-score TMI TMI z-score TMI BMI z-score BMI BMI z-score BMI BMI z-score BMI BMI z-score BMI 1.84 (1.40, 2. 41) 1.63 (1.42, 1.88) 1.15 (1.10, 1.21) 1.54 (1.19, 2.00) 1.16 (1.06, 1.28) 1.65 (1.32, 2.06) 1.19 (1.09, 1.29) 1.70 (1.33, 2.19) 1.14 (1.06, 1.21) 1.89 (1.62, 2.20) 1.16 (1.11, 1.20) 1.89 (1.42, 2.53) 1.15 (1.07, 1.22) 1.94 (1.51, 2.47) 1.21 (1.13, 1.29) 1.16 (1.08, 1.24) OR (95%CI) Boys and or different definitions of pediatric Mets (35). definitions ofpediatricMets(35). and ordifferent ofthesample populations described bytheheterogeneity could be inresults and adolescents.Thisdiscrepancy accuracy foridentifyingMetS amongIranianchildren study,present TMIandBMIhad moderatediagnostic lessthan0.755)(26).Inthe both girlsandboyswere of9-12and13-17yearsin (all AUCsforagegroups powertodetectMetS had amoderatediscriminatory and adolescents,theROCanalysisshowedthatTMI of bothgenders(24).InstudyonColombianchildren (AUC =0.98)forMetSamongSpanishadolescents value thatBMIhadhighpredictive study reported in 5th-gradegirlstudents(25).Anothercross-sectional and privateschoolsshowedthatBMIhadAUCat0.754 public from study on Brazilian children cross-sectional of BMI and TMI for MetS in various populations. A ofthesestudies. with results concordant components (37,38).Ourfindingswere between fatdistribution(BF%and/orFMI)MetS (19).Severalstudieshavefoundthe associations scores accuratelythanBMI adolescentsmore overweight theyindicatedthatTMIcoulddiagnose Furthermore, in non-Hispanicwhiteadolescentsaged8to17years. accurately than BMI estimated body fat levels more There are differences between the predictive values values betweenthe predictive differences are There Recently, Petersonandcols.showedthatTMI < 0.001 < 0.001 < 0.001 < 0.001 < 0.001 < 0.001 < 0.001 < 0.001 < 0.001 < 0.001 < 0.001 < 0.001 < 0.001 < 0.001 0.001 0.002 P 1.53 (0.94, 2.49) 1.18 (0.97, 1.43) 2.00 (1.52, 2.62) 1.33 (1.19, 1.49) 1.36 (1.08, 1.71) 1.08 (1.01, 1.16) 1.34 (1.20, 1.50) 1.12 (1.07, 1.17) 1.32 (0.98, 1.77) 1.11 (0.99, 1.25) 1.62 (1.36, 1.93) 1.22 (1.14, 1.32) 1.41 (1.13, 1.76) 1.11 (1.04, 1.19) 1.58 (1.34, 1.87) 1.14 (1.08, 1.21) OR (95%CI) Arch Metab. Endocrinol 2020;64/2 Girls < 0.001 < 0.001 < 0.001 < 0.001 < 0.001 < 0.001 0.102 0.102 0.009 0.033 0.024 0.024 0.065 0.071 0.002 0.003 P z Arch Metab. Endocrinol 2020;64/2 the Isfahan University of Medical Sciences approved andEthicscouncilof organizations. TheResearch nationalregulatory committees andotherrelevant by ethical andapproved reviewed were protocols pediatric agegroup. index forlarge population-based studies inthe anthropometric as an appropriate can be considered ofMetSthanBMIinbothgenders.TMIpredictor adolescents. However, generallyTMIwasabetter andaccuracy foridentifyingMetSinchildren that bothBMIandTMIhadmoderatediagnostic large population-basedsample. andconsideringa novelty inthepediatricagegroup its ofourstudyare in adolescents.Themainstrengths age onpuberty limitation isthelackofinformation thesefindings.Another from cannot beinferred relationships thecause–effect therefore sectional nature, weight(42-44). normal andadolescents with among Asianadults,children CVDriskfactorsandmetabolicsyndrome reported in Iran (42). Moreover,weight children several studies levelsamongnormal- TG andlowHDLcholesterol ofhigh finding maybebecauseofthehighprevalence orobesity. overweight than thevaluestopredict This less MetSwere topredict andTMIz-score z-score study,In thecurrent valuesforBMI theoptimalcutoff inlifestylebetweenvariouspopulation(41). differences inthesevalues couldbedueto girls. Thedifferences inbothboysand studiesforallagegroups previous thanvaluesin greater studywere values inthepresent and 12.48forgirls,respectively. However, TMIcutoff 12.10and11.9yearsforboys12.13 years were of 9-12 and13-17 and adolescents for age groups MetSinColombianchildren valuestopredict cutoff = 27forbothgender)(24).Moreover, theoptimalTMI valuesinSpanishadolescents(cutoff from which differs 19.65kg/m were MetSinourstudy valuestopredict optimal BMIcutoff population (41). The suggested in different studies are forBMIandTMIbyagegender.cutoffs More forTMI. age particularly with riskofMetsinpuberty TMI and between BMI studyalsoshowedrelationship Mets (40).Thepresent ondistributionsofadiposity(39)andthe puberty Ethics approval and consent to participate: study andconsenttoparticipate: Ethics approval studyshowed ofpresent In conclusion,theresults study iscross- The mainlimitationofthepresent optimal studyprovide Findings ofthepresent of effects studieshaveidentifiedstrong Previous 2 forboysand19.61kg/m 2 forgirls, 4. 3. 2. 1. REFERENCES was reported. relevant tothisarticle nopotentialconflictofinterest Disclosure: program. surveillance offunding: Source andstudents,respectively. theparents obtained from consentand verbal consentwere written informed explanation ofthestudyobjectivesandprotocols, stu 15. 14. 13. 12. 11. 10. 9. 8. 7. 6. 5.

dy (Project number: 194049). After complete dy (Project Abarca-Gómez L, Abdeen ZA,Hamid Abu-Rmeileh NM, Friend A, CraigL, Turner S. The Prevalence ofMetabolicSyndrome Kelishadi R.Childhoodoverweight, obesity, andthemetabolic Kelishadi R,Poursafa P. A reviewonthegenetic,environmental, Wu Y-E, ZhangC-L,ZhenQ. Metabolic syndromeinchildren Lehingue Y, Remontet L,MunozF, MamelleN.Birth ponderalindex Hattori K,Hirohara T. Age change ofpowerinweight/heightp Cole TJ. Weight/heightp compared to weight/height2 for assessing Hattori K. Age Changeof Allometry. J Anthr Soc Nippon. NCD RiskFactor Collaboration(NCD-RisC). A centuryoftrendsin Kelishadi R, Hovsepian S, Djalalinia S, Jamshidi F, Qorbani M. A Ostovaneh MR,ZamaniF, SharafkhahM, Ansari-Moghaddam A, Mirhosseini N-Z, Yusoff NAM,ShaharS,Parizadeh SMR, M,Kelishadi Qorbani M,Mehrkash R,MohammadianS, Khashayar P, HeshmatR,QorbaniM,MotlaghME, Aminaee in Children: A Systematic Review of the Literature. Metab Syndr syndrome indevelopingcountries.EpidemiolRev. 2007;29:62-76. disease. Curr Probl Pediatr Adolesc HealthCare.2014;44:54-72. and lifestyleaspectsoftheearly-life originsofcardiovascular (Review). Exp Ther Med.2016;12:2390-4. J HumBiol.1998; and bodymassindexreferencecurves inalargepopulation. Am Biol. 2002;14:275-9. ofadiposityinJapanese.indices usedasindicators Am JHum puberty. Ann HumBiol.1986;13:433-51. adiposity inchildhood: influenceofageandboneonpduring 1975;83:29-38. adult humanheight.Elife.2016;26;5:1-29. 2017;390:2627-42. studies in1289 millionchildren, adolescents,andadults.Lancet. 2016: apooledanalysisof2416 population-basedmeasurement mass index,underweight,overweight, andobesityfrom1975 to Acosta-Cazares B, Acuin C, et al. Worldwide trends in body- Iranian children andadolescents.JRes MedSci. 2016;21:1-9. systematic reviewontheprevalenceofmetabolicsyndromein Arch IranMed.2014;17:477-82. syndrome in Amol and Zahedan, Iran: A population based study. Akhavan KhaleghiN,Saeedian FS,etal.Prevalence ofmetabolic Mashhad, Iran. Asia Pac JClinNutr. 2009;18:131-6. amongadolescentgirlsin syndrome anditsinfluencingfactors MT.Mobarhen MG,Shakery Prevalence ofthemetabolic Southeast Asian J Trop Med Public Health.2012;43:756-63. Among aRepresentative Sample ofIranian Adolescents. Mousavinasab F, EsmaeilM,etal.ObesityandMetabolicSyndrome Endocrinol. 2013:1-8. the MiddleEastandNorth Africa Risk FactorsinaNationalSample of Adolescent Population in T, Ardalan G,etal.Metabolic Syndrome andCardiovascular Relat Disord.2013;11:71-80.

this study was conducted as part of a national ofanational this studywasconductedaspart 10:327-40. TMI andBMIinpredicting pediatricMetS : The CASPIAN IIIStudy. IntJ 177

Knowles KM, Paiva LL, Sanchez SE, Revilla L, Lopez T, Yasuda MB, WHO. Physical status:the use and interpretation of anthropometry. Motlagh M,ZiaodiniH,Qorbani Taheri M, Aminaei T, Goodarzi Ramírez-Vélez R, Correa-Bautista JE, Carrillo HA, González- Andaki ACR, Tinôco ALA, MendesEL, Andaki Júnior R,Hills AP, Kelishadi R,Heidari-Beni M, QorbaniMotamed-GorjiN, Wicklow BA,Becker A, ChateauD,Palmer K,Kozyrskij A, Sellers Carrascosa A, Yeste D,Moreno-Galdó A, Gussinyé M,Ferrández Peterson CM,SuH, Thomas DM,HeoM,Golnabi AH, Pietrobelli A, Howe LD, K,BenfieldL, Logue Tilling J, Sattar N, Ness AR, etal. Freedman DS, Wang J, Thornton JC,MeiZ,Sopher AB, Pierson Cooley SM,DonnellyJC, Walsh T, KirkhamC,GillanJ, Geary Perona JS,Schmidt-RioValle J, Rueda-Medina B,Correa-Rodríguez Vogt BP, Ponce D,CaramoriJCT. PredictAnthropometric Indicators et al. Waist Circumference,BodyMassIndex,andOtherMeasures Ser. 1995;854:1-452. Report ofa WHO Expert Committee. World HealthOrgan Tech Rep 2017;8:1-9. noncommunicable disease: The caspian-vstudy. IntJPrev Med. childhood and adolescence surveillance and prevention of adult A, etal.Methodologyandearlyfindingsofthefifth surveyof Young People. Nutrients.2018;10(4). pii:E412. for MetabolicSyndrome Prediction inColombianChildrenand Ponderal MassIndexvs.Fat Mass/Height3asaScreening Tool Jiménez E,Schmidt-RioValle J, Correa-Rodríguez M,etal. Tri- 2014;17: 2287-94. prediction ofmetabolicsyndromeinchildren. Public HealthNutr. Amorim PRS. Anthropometry andphysicallevelinthe activity its components,inSpanishadolescents.NutrRes. 2017;45:38-45. valueformetabolicsyndrome,andwaist-to-hip predictive ratiofor M, González-JiménezE. Waist circumference showsthehighest adolescents: The CASPIAN-V study. Nutrition.2017;43:32-8. wrist circumferences and cardiometabolic risk in children and Motlagh ME, ZiaodiniH,et al. Association betweenneck and cohortprospective data.IntJObes.2015;39:1070-8. to predictmetabolicsyndromeinadolescence: Analysis of EAC. Comparisonofanthropometric measurementsinchildren Patients. NutrClinPract. 2016;31:368-74. Metabolic Syndrome Diagnosis inMaintenanceHemodialysis Pediatría. 2018. 89(3):137-43. generación del milenio. Estudio longitudinal de Barcelona. An triponderal de1.453 niñosnoobesosnimalnutridosdela A, ClementeM,etal.Índicedemasacorporaleíndice Body Fat During Adolescence. JAMA Pediatr. 2017;35294:1-8. et al. Tri-Ponderal Mass IndexvsBodyMassinEstimating atage15. Hernandez factors AV, editor. PLoS One.2010;5:1-13. and theirassociationswithfatmasscardiovascular risk Changes inponderalindexandbodymassacrosschildhood 2009;163:805-11. for-age categoriesamongchildren. Arch Pediatr Adolesc Med. RN, etal.Classificationofbodyfatnessbymassindex- wellbeing? JObstetGynaecol.2012;32:439-43. primigravid population:which isthebetter predictoroffetal MP. Ponderal index(PI)vsbirth weightcentilesinthelow-risk 36. 35. 34. 33. 32. 31. 30. 44. 43. 42. 41. 40. 39. 38. 37.

Cole TJ. A methodforassessingage-standardized weight- Zimmet P, Alberti KGM,Kaufman F, Tajima N,SilinkM, Arslanian Li C,Ford ES,Mokdad AH, CookS.Recent Trends in Waist National CholesterolEducationProgram (NCEP)Expert Panel on McNamara JR, Schaefer EJ. Automated enzymatic standardized Friedewald WT, Levy RI,Fredrickson DS.Estimationofthe Adolescents N. The Fourth Report ontheDiagnosis,Evaluation, irmN, adyR, ir , hraR Dv R Khanna DeviJR, SharmaR, Misra A, Pandey RM, Vikram NK, etal. ZhangJ, Zhao WH, PiaoJH, ZhaiFY, Li YangYP, XG, Kelishadi R,CookSR,Motlagh ME,GouyaMM, Ardalan G, Chen X, Liu Y, Sun X, Yin Z, Li H, Deng K, et al. Comparison of Reinehr T, Wolters B,KnopC,LassN,HollRW. Strongeffect Loomba-Albrecht LA,StyneDM.Effect ofpuberty onbody Ramírez-Vélez R,Correa-Bautista JE,Sanders-Tordecilla A, McCarthy HD.Bodyfatmeasurements inchildren aspredictors IDF consensusreport. Pediatr Diabetes.2007;8:299-306. S, et al. The metabolic syndrome in children and adolescents – an Adolescents. Pediatrics. 2006;118:1390-8. Circumference and Waist-Height Ratio Among US Children and Treatment Panel III)finalreport. Circulation.2002;106:3143-421. Evaluation, and Treatment ofHighBloodCholesterolin Adults (Adult Cholesterol EducationProgram (NCEP)Expert Panel onDetection, in Adults (Adult Treatment Panel III). Third Report oftheNational Detection. Evaluation,and Treatment ofHighBloodCholesterol Acta. 1987;166:1-8. lipid analysesforplasmaandlipoproteinfractions.ClinChim 1972;18:499-502. ultracentrifuge.ClinChem. without useofthepreparative concentration oflow-density lipoproteincholesterol inplasma, Adolescents. Pediatrics. 2004;114:555-76. and Treatment ofHighBloodPressure inChildrenand amo of Adiposity inPredicting Cardiovascular DiseaseRiskFactors Nutrition. 2003;19:503-9. normal waist circumferencehavehighcardiovascular risk. N. Non-obese(bodymassindex<25kg/m 2005;18:401-10. Disease risksofchildhood obesityinChina.BiomedEnviron Sci. Study. J Am Diet Assoc. 2008;108:82-90. Phenotypically ObeseMetabolicallyNormal Youths: The CASPIAN Motaghian M,etal.MetabolicallyObeseNormal Weight and Chinese cohort study. JHumHypertens. 2018;32(3):228-35. to-height ratioforpredictingincidenceofhyperten body massindex,waist circumference, conicityindex,andwaist- longitudinal study. JClinEndocrinolMetab.2015;100:301-8. of pubertal statusonmetabolichealthinobesechildren: a composition. Curr OpinEndocrinolDiabetesObes.2009;16:10-5. students. Nutrients.2017;9(9). pii:E1009. for metabolicsyndromepredictioninColombianuniversity Percentage ofbodyfatandmassindexasascreeningtool Ojeda-Pardo ML,Cobo-MejíaEA,Castellanos-Vega RDP, etal. Nutr Soc. 2006;65:385-92. for themetabolicsyndrome:focusonwaist circumference.Proc 1979;6:249-68. for-height inchildren seen cross-sectionally. Ann HumBiol. ng Peruvian Adults. IntJHypertens. 2011;2011:931402. Arch Metab. Endocrinol 2020;64/2 2 ) Asian Indianswith sion: therural