Research Article Determinants and Regression …downloads.hindawi.com/journals/crp/2015/380729.pdfin a Healthy Italian Pediatric Population VeronicaFibbi, 1 PiercarloBallo, 1 SilviaFavilli,

Research ArticleDeterminants and Regression Equations for the Calculation of119911 Scores of Left Ventricular Tissue Doppler Longitudinal Indexesin a Healthy Italian Pediatric Population

Veronica Fibbi1 Piercarlo Ballo1 Silvia Favilli2 Gaia Spaziani2 Giovanni B Calabri2

Iva Pollini2 Alfredo Zuppiroli3 and Enrico Chiappa2

1Cardiology Unit S Maria Annunziata Hospital Florence 50012 Italy2Department of Pediatric Cardiology Meyer Hospital Florence 50139 Italy3Regional Health Agency of Tuscany Florence 50141 Italy

Correspondence should be addressed to Veronica Fibbi verof82virgilioit

Received 29 July 2015 Accepted 1 November 2015

Academic Editor Vicky A Cameron

Copyright copy 2015 Veronica Fibbi et al This is an open access article distributed under the Creative Commons Attribution Licensewhich permits unrestricted use distribution and reproduction in any medium provided the original work is properly cited

Aim We investigated the predictors of tissue Doppler left ventricular (LV) longitudinal indexes in a healthy Italian pediatricpopulation and established normative data and regression equations for the calculation of 119911 scoresMethods and Results A total of369 healthy subjects aged 1ndash17 years (age of 64plusmn 11 years 491 female) underwent echocardiography LVpeak longitudinal velocityat systole (1199041015840) early diastole (1198901015840) and late diastole (1198861015840) was determined by tissue Doppler The ratio of peak early diastolic LV fillingvelocity to 1198901015840 was calculated Age was the only independent determinant of 1199041015840 (120573 = 0491119901 lt 00001) and the strongest determinantof 1198901015840 (120573 = 0334 119901 lt 00001) and1198641198901015840 (120573 = minus0369 119901 lt 00001) Heart rate was themain determinant of 1198861015840 (120573 = 0265 119901 lt 00001)Male gender showed no effects except for a weak association with lateral 1199041015840 suggesting no need of gender-specific reference rangesAge-specific reference ranges regression equations and scatterplots for the calculation of 119911 scores were determined for each indexConclusion In a pediatric Italian population age was the strongest determinant of LV longitudinal dynamics The availability ofage-specific normality data for the calculation of 119911 scoresmay allow for correctly detecting LV dysfunction in pediatric pathologicalpopulations

1 Introduction

Tissue Doppler (TD) imaging is an established echocardio-graphic technique that provides reproducible sensitive andeasy-to-calculate indexes of longitudinal left ventricular (LV)function [1 2] Though the majority of TD studies havebeen performed in adults TD has also been widely used inpediatric populations as a tool to detect early LV longitudinaldysfunction in several pathophysiological conditions inmostcases by comparing patients with matched controls [3ndash7]However few data exist regarding the determinants and thereference ranges of LV longitudinal indexes in the pediatricage [8ndash10] From a clinical point of view investigating thedeterminants of longitudinal LV velocities in a healthy pedi-atric population might be helpful to better understand themechanisms underlying LV failure in patientswith congenital

heart disease Also obtaining normative data for LV longi-tudinal velocities particularly referred to as the possibilityof allowing calculation of 119911 scores may be important toprovide a tool to rapidly identify early LV dysfunction in dailypractice

The aim of this study was to explore the determinantsof LV longitudinal indexes in a healthy Italian pediatricpopulation and to provide normative data obtained in ahealthy Italian pediatric population Regression equationsfor each index were obtained to allow the calculation ofcorresponding 119911 scores

2 Materials and Methods

21 Study Population Consecutive healthy subjects aged 1 to17 years visited at the Department of Pediatric Cardiology

Hindawi Publishing CorporationCardiology Research and PracticeVolume 2015 Article ID 380729 8 pageshttpdxdoiorg1011552015380729

2 Cardiology Research and Practice

of Anna Meyer Children Hospital Florence Italy over a9-month enrolment period were prospectively enrolled inthis study Subjects were deemed as normal if they hadunremarkable clinical history normal findings at clinicalexamination ECG standard echocardiography and no fam-ily history of genetic cardiac disease (eg Marfanrsquos syndromeor cardiomyopathy) and if they were not on pharmacologicalagents Patients with technically inadequate image qualitywere excluded Main reasons for referral were cardiovascularassessment for nonagonistic sport activities (119899 = 209) inno-cent cardiac murmur (119899 = 70) vasovagal lipothymia (119899 =27) fever (119899 = 23) atypical chest pain (119899 = 22) and familyhistory of cardiac disease (119899 = 18)The study protocol agreedwith the 1964 Helsinki declaration and successive emenda-tions and was approved by the local Ethical Committee

22 Echocardiography Studies were performed with use ofhigh-quality commercially available ultrasound systems (IE33 Philips Medical Systems Andover MA) LV dimensionsand LV mass were measured using M-mode imaging fromthe parasternal long-axis view in accordance with currentASEEACVI recommendations [11] End-diastolic and end-systolic LV volumes were calculated from apical views usingthe biplane modified Simpsonrsquos rule Left atrial volume at endsystole was measured from apical views using the biplanemethod of discs LV mass LV volumes and left atrial volumewere considered in the analysis after indexation to bodysurface area [12 13] Pulsed wave Doppler analysis of LVinflow was performed in the apical 4-chamber view byplacing the cursor at the level of the mitral leaflet tips Earlyto late peak filling velocity ratio and deceleration time weremeasured Pulsed TD of LVmitral annulus was performed inthe apical 4-chamber views by placing the sample volume atthe septal and lateral annular sites in accordancewith currentguidelines [14] To optimize quality of pulsed TD filters andgains were adjusted at the minimal optimal level allowing thebest signal-to-noise ratio The peaks of myocardial systolic(1199041015840) early diastolic (1198901015840) and late diastolic (1198861015840) waves weredetermined Average 1199041015840 1198901015840 and 1198861015840 were obtained as the meanof septal and lateral values Septal lateral and average 1198641198901015840ratio were also calculated Right ventricular systolic functionwas assessed by measuring tricuspid annulus plane systolicexcursion by M-mode and peak systolic velocity of tricuspidannulus by pulsed TD All measurements were taken byaveraging values obtained in three consecutive cycles

Reproducibility was assessed in a subset of 40 randomlyselected subjects For intraobserver analysis one investigatorreviewed echocardiographic images andmeasured septal andlateral TD velocities with at least onemonth interval betweenmeasurements For interobserver analysis two experiencedinvestigators independently reviewed images Intraobservervariability coefficients for 1199041015840 1198901015840 1198861015840 and 1198641198901015840 were as followsaverage 40 32 44 and 31 septal 42 37 45and 40 and lateral 44 39 46 and 41 Corre-sponding intraclass correlation coefficients were as followsaverage 095 096 097 and 096 septal 089 092 093 and096 lateral 097 097 095 and 096 and 119901 lt 00001 for allInterobserver variability coefficients were as follows average43 36 48 and 33 septal 48 43 50 and

45 and lateral 47 41 58 and 42 Correspondingintraclass correlation coefficients were as follows average094 095 096 and 093 septal 084 090 092 and 095lateral 096 097 094 and 096 and 119901 lt 00001 for all

23 Statistical Analysis Data are presented as mean plusmn SDor number (percentages) Correlations were expressed usingPearsonrsquos coefficients Gender comparison was performedusing the Student 119905-test for unpaired data For the deter-mination of age-specific reference ranges the following ageintervals were predefined 1-2 years 3-4 years 5-6 years 7-8 years 9ndash11 years and 12ndash16 years Multivariable regressionanalysis was used to explore the independent determinantsof TD indexes Models were obtained by testing all variableswith univariable 119901 lt 010 in a stepwise multivariableanalysis A 119865-to-remove of gt010 and 119865-to-enter of lt005were used as criteria for the selection of variables in thestepwise procedure Collinearity diagnostics were performedto explore model stability setting a cut-off of lt020 fortolerance for the identification of significantmulticollinearityIn case of multivariable models with evidence of collinearityproblems the variable with the highest variance inflationfactor was removed from the model This procedure wasiterated until a stable model was obtained The goodness-of-fit was expressed using the adjusted 1198772

For the generation of regression equations for LV lon-gitudinal indexes the following procedure was used Firstto account for heterogeneous variances across the range ofage a natural logarithmic transformation was performed foreach index to stabilize variances Second nonlinear regres-sion analysis was performed by testing different polynomialequations and choosing the best fitting model according tothe Akaike Information Criterion For all LV longitudinalindexes a third-grade polynomial regression was found toprovide the best fitting A cubic regression model of the form119884= 1198873Age3 + 119887

2Age2 + 119887

1Age + 119887

0was then obtained for each

index where 119884 represents the expected value of logarithmic-transformed longitudinal index according to age Thirdvalues were converted to the original units by exponentiatingLog 119884 and then allowing calculation of the expected valueof the echocardiographic index according to age In additionto the curve showing the expected values other 6 curvescorresponding to the 119911 scores plusmn1 plusmn2 and plusmn3 were generatedfor each index by adding or subtracting the correspondingmean square error (MSE) to the regression equation Thesignificance level for all analyses was set at 005 All testswere two-tailed Analyses were performed using the SPSS(Statistical Package for Social Sciences Chicago Illinois) forWindows release 150

3 Results

31 General Characteristics A total of 393 subjects metthe inclusion criteria during the period of study Adequatemeasurement of TD velocities was not possible in 24 subjects(in one case because of poor acoustic windows secondary topectus excavatum in the other cases because of uncontrol-lable crying in subjects aged lt4 years) Main characteristicsof the remaining 369 subjects are shown in Table 1

Cardiology Research and Practice 3

Table 1 Main clinical and echocardiographic characteristics of thestudy population

Age (years) 64 plusmn 11Female gender () 181 (491)Weight (kg) 260 plusmn 146Height (m) 1176 plusmn 272Body mass index (kgm2) 175 plusmn 36Body surface area (m2) 091 plusmn 036Systolic blood pressure (mmHg) 990 plusmn 93Diastolic blood pressure (mmHg) 619 plusmn 67Heart rate (bpm) 940 plusmn 201Indexed LV end-diastolic volume (mLm2) 534 plusmn 132Indexed LV end-systolic volume (mLm2) 193 plusmn 60Ejection fraction () 639 plusmn 47Indexed LV mass (gm2) 620 plusmn 138119864 wave (cms) 942 plusmn 138119860 wave (cms) 510 plusmn 154119864119860 ratio 20 plusmn 06Deceleration time (ms) 1411 plusmn 336Indexed LA volume (mLm2) 194 plusmn 55Septal 1199041015840 (cms) 79 plusmn 11Septal 1198901015840 (cms) 131 plusmn 20Septal 1198861015840 (cms) 65 plusmn 15Septal 11989010158401198861015840 ratio 21 plusmn 06Septal 1198641198901015840 ratio 73 plusmn 16Lateral 1199041015840 (cms) 97 plusmn 23Lateral 1198901015840 (cms) 171 plusmn 31Lateral 1198861015840 (cms) 69 plusmn 17Lateral 11989010158401198861015840 ratio 26 plusmn 08Lateral 1198641198901015840 ratio 57 plusmn 14Average 1199041015840 (cms) 88 plusmn 15Average 1198901015840 (cms) 151 plusmn 21Average 1198861015840 (cms) 67 plusmn 14Average 11989010158401198861015840 ratio 23 plusmn 06Average 1198641198901015840 ratio 63 plusmn 13119860 peak late diastolic transmitral flow velocity 1198861015840 peak late diastolic mitralannulus velocity 119864 peak early diastolic transmitral flow velocity 1198901015840 peakearly diastolic mitral annulus velocity LV left ventricular 1199041015840 peak systolicmitral annulus velocity

Univariable correlations of LV longitudinal indexes withclinical and echocardiographic variables are shown inTable 2Most indexes showed significant relationships with age bodysurface area systolic and diastolic blood pressure heart rateLV end-diastolic and end-systolic volumes LV mass leftatrial volume LV inflow indexes and measures of rightventricular systolic function However the relations with LVvolumes and left atrial volume were no longer significantor become considerably weaker when indexed volumes wereconsideredThe strongest associations of 1199041015840 1198901015840 and 1198641198901015840 wereobserved with age whereas the strongest association of 1198861015840was observed with heart rate Gender comparison (Figure 1)showed no differences in average 1198901015840 1198861015840 and 1198641198901015840 (119901 = NS forall) whereas slightly higher average 1199041015840 velocitieswere found in

male as compared to female subjects (119901 = 0017) Referenceranges according to predefined age classes are shown inTable 3

32 RegressionAnalysis for Calculation of 119911 Scores Theresultsof regression analysis are shown inTable 4 For each LV longi-tudinal index coefficients for the cubic quadratic and linearterms are provided together with the intercept theMSE andthe 1198772 value as an overall measure of model goodness-of-fit The table allows an exact calculation of the 119911 score forany observed value 119910 of a LV longitudinal index in a patientwith given age This can be achieved by inserting patientrsquosage in the corresponding equation so that the expectedvalue 119884 of the logarithmic-transformed longitudinal indexis obtained Then the 119911 score can be calculated using thestandard formula [Log(observed 119910) minus119884]MSE For examplein a 12-year-old patient with an 1198901015840 of 126 cms the expectedLog 1198901015840 value is given by (26 sdot 10minus4) sdot 123minus (84 sdot 10minus3) sdot 122 +0088 sdot 12 + 245 = 2746 (which corresponds to an expected 1198901015840of 1198902746 = 156 cms) The 119911 score is obtained as [Log(126) minus2746]0141 = minus151

The scatterplots of LV longitudinal indexes obtainedagainst age are shown in Figures 2ndash5 In each plot the thicksolid line is the regression equation whereas the other 6curves represent the values corresponding to plusmn1 plusmn2 andplusmn3 119911 scores These figures allow an alternative less precisebut easier and more rapid estimation of 119911 scores which canbe useful for quick application in daily practice For anyobserved value 119910 of a LV longitudinal index in a patientwith given age it is sufficient to ideally draw a verticalline corresponding to patientrsquos age and a horizontal linecorresponding to 119910 and to look at where the two lines crossConsidering the example above it is easy to observe inFigure 2 that the crossing point corresponding to an 1198901015840 valueof 126 cms in a 12-year subject falls approximately midwaybetween the 119911 = minus1 and 119911 = minus2 curves which directly leadsto an estimated 119911 score of around minus15

33 Determinants of LV Longitudinal Indexes Multivariableanalysis (Table 5) confirmed that age was the main inde-pendent determinant of 1199041015840 1198901015840 and 1198641198901015840 ratio whereas heartrate was the strongest independent determinant of 1198861015840 Agewas the only predictor of 1199041015840 and accounted for 766 and967 of the total variability expressed by the 1198901015840 and 1198641198901015840models respectively Heart rate accounted for 731 of thetotal variability expressed by the 1198861015840model Notably as a resultof strong interrelations with age most of other variables asso-ciated with LV longitudinal indexes at univariable analysissuch as body surface area heart rate and measures of cardiacgrowth were removed because of violation of collinearitycriteria Collinearity diagnostic confirmed that forcing any ofthem into the models yielded unacceptable levels of varianceinflation factor and tolerance

4 Discussion

41 Main Findings Accurate assessment of LV function is akey aspect in the management of children with several typesof congenital and acquired cardiac disease In this study we


Table 2 Univariable relationships of tissue Doppler indexes with clinical and echocardiographic variables in the study population Valueswere computed by considering the average of velocities taken at the septal and lateral site

Average 1199041015840 Average 1198901015840 Average 1198861015840 Average 1198641198901015840

119877 119901 119877 119901 119877 119901 119877 119901

Age 049 lt00001 040 lt00001 minus020 lt00001 minus043 lt00001Body mass index 015 00044 004 050 001 089 minus005 037Body surface area 049 lt00001 037 lt00001 minus020 lt00001 minus041 lt00001Systolic blood pressure 038 lt00001 021 lt00001 004 047 minus023 lt00001Diastolic blood pressure 029 lt00001 019 00003 minus003 055 minus026 lt00001Heart rate minus017 00014 minus031 lt00001 035 lt00001 035 lt00001LV end-diastolic volume 043 lt00001 034 lt00001 minus019 00002 minus033 lt00001Indexed LV end-diastolic volume 011 00030 013 015 minus008 011 minus005 031LV end-systolic volume 037 lt00001 028 lt00001 028 lt00001 minus032 lt00001Indexed LV end-systolic volume 004 040 006 022 minus007 016 minus005 037LV mass 048 lt00001 033 lt00001 minus014 00075 minus033 lt00001Indexed LV mass 027 lt00001 014 00056 minus003 057 minus010 0066Ejection fraction 013 0011 011 0029 004 045 minus004 047Left atrial volume 041 lt00001 034 lt00001 minus013 0011 minus032 lt00001Indexed left atrial volume 004 045 008 013 002 065 minus003 061Mitral 119864119860 ratio 003 056 028 lt00001 minus030 lt00001 minus007 015Deceleration time 023 lt00001 022 lt00001 minus014 0010 minus019 00004TAPSE 031 lt00001 023 lt00001 minus001 077 minus010 0047Tricuspid 1199041015840 036 lt00001 019 00003 011 0036 minus006 024119860 peak late diastolic transmitral flow velocity 1198861015840 peak late diastolic mitral annulus velocity 119864 peak early diastolic transmitral flow velocity 1198901015840 peak earlydiastolic mitral annulus velocity LV left ventricular 1199041015840 peak systolic mitral annulus velocity TAPSE tricuspid annulus plane systolic excursion

Table 3 Reference ranges of left ventricular longitudinal indexes according to age class expressed as mean plusmn standard deviation

1-2 y (119899 = 73) 3-4 y (119899 = 77) 5-6 y (119899 = 53) 7-8 y (119899 = 61) 9ndash11 y (119899 = 49) 12ndash16 y (119899 = 56)Septal 1199041015840 (cms) 75 plusmn 09 78 plusmn 10 75 plusmn 10 79 plusmn 10 80 plusmn 09 86 plusmn 14Septal 1198901015840 (cms) 121 plusmn 19 129 plusmn 17 131 plusmn 22 134 plusmn 21 138 plusmn 17 138 plusmn 23Septal 1198861015840 (cms) 72 plusmn 16 66 plusmn 13 64 plusmn 16 63 plusmn 14 60 plusmn 11 64 plusmn 16Septal 11989010158401198861015840 ratio 18 plusmn 05 20 plusmn 05 22 plusmn 06 22 plusmn 06 24 plusmn 06 23 plusmn 05Septal 1198641198901015840 ratio 81 plusmn 14 77 plusmn 14 75 plusmn 16 69 plusmn 15 68 plusmn 14 67 plusmn 14Lateral 1199041015840 (cms) 81 plusmn 13 90 plusmn 15 95 plusmn 17 101 plusmn 25 106 plusmn 19 116 plusmn 28Lateral 1198901015840 (cms) 145 plusmn 22 164 plusmn 25 183 plusmn 29 182 plusmn 29 178 plusmn 31 181 plusmn 27Lateral 1198861015840 (cms) 72 plusmn 18 71 plusmn 15 71 plusmn 19 67 plusmn 17 67 plusmn 18 65 plusmn 16Lateral 11989010158401198861015840 ratio 21 plusmn 05 24 plusmn 07 27 plusmn 07 28 plusmn 07 28 plusmn 08 29 plusmn 07Lateral 1198641198901015840 ratio 68 plusmn 13 61 plusmn 14 54 plusmn 12 51 plusmn 11 53 plusmn 10 51 plusmn 10Average 1199041015840 (cms) 78 plusmn 08 84 plusmn 11 85 plusmn 11 90 plusmn 16 93 plusmn 12 101 plusmn 19Average 1198901015840 (cms) 133 plusmn 17 146 plusmn 18 157 plusmn 17 158 plusmn 20 158 plusmn 19 160 plusmn 21Average 1198861015840 (cms) 72 plusmn 14 68 plusmn 12 67 plusmn 15 65 plusmn 14 63 plusmn 12 65 plusmn 14Average 11989010158401198861015840 ratio 19 plusmn 04 22 plusmn 05 24 plusmn 06 25 plusmn 06 26 plusmn 05 26 plusmn 05Average 1198641198901015840 ratio 74 plusmn 12 68 plusmn 13 62 plusmn 12 58 plusmn 10 59 plusmn 09 58 plusmn 101198861015840 peak late diastolic mitral annulus velocity 119864 peak early diastolic transmitral flow velocity 1198901015840 peak early diastolic mitral annulus velocity 1199041015840 peak systolic

mitral annulus velocity

investigated the independent determinants of indexes of LVlongitudinal function in a normal Italian pediatric populationand sought to provide normative data for the calculation of 119911scores We found that age but not gender was the strongestdeterminant of most indexes We then derived age-specificreference ranges and regression equations for the calculationof 119911 scores for each index

42 Determinants of LV Longitudinal Indexes To our knowl-edge only two studies previously explored the determinantsof longitudinal LV function in healthy pediatric subjects Ina retrospective analysis of 325 healthy American childrenEidem et al [8] found that unindexed LV end-diastolicdiameter and age were the strongest predictors of mitralannulus systolic and early diastolic velocities and that weaker


Table 4 Coefficients for regression equations relating left ventricular longitudinal indexes and age For each index application of thecorresponding third-grade equation provides the expected logarithmic value according to age In a patient with given age this allowscalculation of the 119911 score that corresponds to the observed value of the index in that patient (see text)

Age3 Age2 Age Intercept MSE 1198772

1199041015840 83 sdot 10minus5 minus19 sdot 10minus3 0031 201 0161 0241198901015840 26 sdot 10minus4 minus84 sdot 10minus3 0088 245 0141 0241198861015840 28 sdot 10minus5 23 sdot 10minus3 minus0040 203 0200 0111198641198901015840 ratio minus19 sdot 10minus4 75 sdot 10minus3 minus0097 215 0198 025

119860 peak late diastolic transmitral flow velocity 1198861015840 peak late diastolic annulus velocity 119864 peak early diastolic transmitral flow velocity 1198901015840 peak early diastolicannulus velocity

79

100

8978

9486

0

4

8

12

16

Septal Lateral Average

130

174152

131

168

150

0

6

12

18

24


6570

6766 68 67

0

4

8

12

16


75

5664

7257 64

0

4

8

12

16


s998400

velo

city

(cm

s)

e998400

velo

city

(cm

s)

a998400

velo

city

(cm

s)

Ee998400

ratio

MalesFemales

MalesFemales

Figure 1 Comparison of left ventricular (LV) longitudinal indexes between male and female subjects in the study population 1198861015840 is peak latediastolic LV longitudinal velocity 1198901015840 is peak early diastolic LV longitudinal velocity 119864 is peak early diastolic left ventricular filling velocity 1199041015840is peak systolic LV longitudinal velocity

associations existed with heart rate Also unindexed LV end-diastolic diameter and LVmass were the only determinants ofthe 1198641198901015840 ratio In another study on 242 healthy Chinese chil-dren Liu et al [9] found that LV end-diastolic diameter wasthe most important predictor of all LV longitudinal indexesexcept for 1198861015840 and that weaker independent relationshipsexistedwith age body surface area and heart rate Our resultsconfirm the important independent role of age as a majordeterminant of systolic and early diastolic mitral annulusvelocities in the pediatric age but also showed that the asso-ciations with body surface area heart rate and measures ofcardiac growth such as LV dimensions and LV mass were no

longer significant inmultivariablemodelsWe also found thatage was the strongest determinant of the 1198641198901015840 ratio Severalfactors could explain these discrepancies across studies Firstthe different geographic locations where the studies wereperformed could have played a role Second the use of LVvol-umes in our study instead of LV diameters used in previousanalyses might have provided more reliable measures of car-diac growthThird the use of unindexed values of LV dimen-sions and mass in previous studies may have favoured theselection of these variables in multivariable models Accord-ingly in our population indexing LV volumes LV massand left atrial volume dramatically reduced the strength of


Table 5 Independent predictors of left ventricular longitudinalindexes in the study population calculated as the average of septaland lateral values

120573 coefficient 119901 value1199041015840 (model 1198772 = 241 119901 lt 00001)Age 0491 lt000011198901015840 (model 1198772 = 212 119901 lt 00001)Age 0334 lt00001119864119860 ratio 0207 lt00001Tricuspid 1199041015840 0108 00251198861015840 (model 1198772 = 160 119901 lt 00001)Heart rate 0265 lt00001119864119860 ratio minus0184 00007Tricuspid 1199041015840 0125 00101198641198901015840 ratio (model 1198772 = 199 119901 lt 00001)Age minus0369 lt00001Heart rate 0108 0077119860 peak late diastolic transmitral flow velocity 1198861015840 peak late diastolic mitralannulus velocity 119864 peak early diastolic transmitral flow velocity 1198901015840 peakearly diastolic mitral annulus velocity 1199041015840 peak systolic mitral annulusvelocity tricuspid 1199041015840 peak systolic tricuspid annulus velocity

Aver

ages

998400(c

ms

)

Age (years)181614121086420

20

18

16

14

12

10

8

6

4

2

0

z = 3

z = 2

z = 1

z = 0

z = minus1

z = minus2

z = minus3

Figure 2 Scatterplot of peak systolic left ventricular longitudinalvelocity (1199041015840) versus age In this figure and in the successive ones thesolid line represents the estimated regression equation (indicated as119911 = 0) The other lines represent the plusmn1 plusmn2 and plusmn1 119911 values aboveand below the regression curve

the associations with LV longitudinal indexes Fourth therestrictive collinearity criteria adopted in our analysis mayhave contributed to removal of highly interrelated variablessuch as body surface area heart rate and measures of cardiacgrowth

43 Normative Data and Regression Equations Few data alsoexist about the normal ranges of LV longitudinal indexes in

Age (years)181614121086420

z = 3

z = 2

z = 1

z = 0

z = minus1

z = minus2

z = minus3Aver

agee

998400(c

ms

)

30

28

26

24

22

20

18

16

14

12

10

8

6

4

2

0

Figure 3 Scatterplot of peak early diastolic left ventricular longitu-dinal velocity (1198901015840) versus age

Age (years)181614121086420

16

14

12

10

8

6

4

2

0

z = 3

z = 2

z = 1

z = 0

z = minus1

z = minus2

z = minus3

Aver

agea

998400(c

ms

)

Figure 4 Scatterplot of peak late diastolic left ventricular longitu-dinal velocity (1198861015840) versus age

children After early studies based on relatively small samplesizes and old TD software [15 16] the above-mentioned studyby Eidem et al provided reference ranges for LV longitudinalindexes in a large sample of normal American children [8]More recently van der Hulst et al reported reference rangesof LV longitudinal indexes in 123 healthy Dutch children [10]Our data may add to these findings by providing age-specificnormality ranges in an Italian pediatric population andreporting regression equations and plots for the calculation of119911 score In this regard because of the lack of significant effectof gender on LV longitudinal indexes we avoided calculation


Age (years)181614121086420

16

14

12

10

8

6

4

2

0

z = 3

z = 2

z = 1

z = 0

z = minus1

z = minus2

z = minus3

Aver

ageE

e998400

Figure 5 Scatterplot of the ratio between peak early diastolic leftventricular filling velocity (119864) and peak early diastolic longitudinalvelocity (1198901015840) versus age

of sex-specific reference ranges In addition to take intoaccount the expectable reduction of cardiac growth rate withincreasing age we predefined relatively narrow age classes forsubjects in the first years of life compared to those for olderchildren and teenagers This allowed us to obtain distinctnormative data for narrower age classes in comparison withprevious studies encompassing two- or three-year rangesuntil 11 years of age

44 Clinical Implications and Study Limitations The findingsof this study provide age-specific normality ranges andtwo different approaches for the calculation of 119911 scores forLV longitudinal indexes which could be used to estimatedeviation from normality in clinical practice The availabilityof adequate normative data for LV longitudinal indexes inchildren is important to allow correct identification of abnor-mal LV longitudinal function in pathological populationsparticularly in those with congenital heart disease [17] LVlongitudinal systolic indexes are well known to be early andsensitive markers of systolic dysfunction and can be used todetect subtle contractile fiber impairment even in subjectswith preserved LV ejection fraction and normal indexes ofLV circumferential shortening [18] On the other hand subtleLV longitudinal systolic dysfunction in children has beenshown to predict adverse outcome in various congenital andacquired pathophysiological conditions [19ndash23] Moreoverknowledge of normal values for diastolic longitudinal veloc-ities in children is crucial to allow correct identification ofLV diastolic impairment which is a key feature of manycongenital heart diseases [24ndash26]

Some limitations should be considered in this study Alarger sample size could have allowed us to obtain even nar-rower age classes All measurements were taken using equip-ment from one vendor Since previous evidence showed that

different vendors cannot be used interchangeably for explo-ration of myocardial velocities and strains [27ndash29] cautionis required in generalizing our findings Speckle tracking isknown to be a technique that can overcome most of the lim-itations of TD particularly angle-dependency and sensitivityto passive motion [30 31] Further studies aimed at definingnormative data for LV longitudinal strain measures arewarranted

Conflict of Interests

The authors declare that there is no conflict of interestsregarding the publication of this paper

References

[1] L Pauliks ldquoTissue doppler myocardial velocity imaging ininfants and childrenmdasha window into developmental changesof myocardial mechanicsrdquo Echocardiography vol 30 no 4 pp439ndash446 2013

[2] S Ge ldquoPediatric echocardiography new developments andapplicationsrdquo Echocardiography vol 30 no 4 pp 426ndash4272013

[3] C De Felice S Maffei C Signorini et al ldquoSubclinical myocar-dial dysfunction in Rett syndromerdquo European Heart JournalmdashCardiovascular Imaging vol 13 no 4 pp 339ndash345 2012

[4] A E Kibar F A Pac S Ballı et al ldquoEarly subclinical left-ventricular dysfunction in obese nonhypertensive children atissue doppler imaging studyrdquo Pediatric Cardiology vol 34 no6 pp 1482ndash1490 2013

[5] J Zamorano V Serra L P De Isla et al ldquoUsefulness of tissueDoppler on early detection of cardiac disease in Fabry patientsand potential role of enzyme replacement therapy (ERT) foravoiding progression of diseaserdquo European Heart JournalmdashCardiovascular Imaging vol 12 no 9 pp 671ndash677 2011

[6] Y I Balci andDGurses ldquoDetection of early cardiac dysfunctionin patients with 120573-thalassemia major and thalassemia trait bytissue doppler echocardiographyrdquo Pediatric Hematology andOncology vol 28 no 6 pp 486ndash496 2011

[7] N Sasaki M Garcia I Lytrivi et al ldquoUtility of Doppler tissueimaging-derived indices in identifying subclinical systolic ven-tricular dysfunction in children with restrictive cardiomyopa-thyrdquo Pediatric Cardiology vol 32 no 5 pp 646ndash651 2011

[8] B W Eidem C J McMahon R R Cohen et al ldquoImpactof cardiac growth on doppler tissue imaging velocities astudy in healthy childrenrdquo Journal of the American Society ofEchocardiography vol 17 no 3 pp 212ndash221 2004

[9] X-Q Liu W-Z Li Y-L Wang and Y Ai ldquoA study onatrioventricular annularmovement in healthy children by tissueDoppler imagingrdquo Zhonghua Er Ke Za Zhi vol 44 no 10 pp738ndash742 2006

[10] A E van der Hulst V Delgado A D J Ten Harkel et alldquoTissue Doppler imaging in the left ventricle and right ventriclein healthy children normal age-related peak systolic veloci-ties timings and time differencesrdquo European Heart JournalmdashCardiovascular Imaging vol 12 no 12 pp 953ndash960 2011

[11] R M Lang L P Badano V Mor-Avi et al ldquoRecommendationsfor cardiac chamber quantification by echocardiography inadults an update from the American Society of Echocardiogra-phy and the European Association of Cardiovascular Imagingrdquo


Journal of the American Society of Echocardiography vol 28 no1 pp 1e14ndash39e14 2015

[12] R B Devereux D R Alonso E M Lutas et al ldquoEchocardio-graphic assessment of left ventricular hypertrophy comparisonto necropsy findingsrdquoThe American Journal of Cardiology vol57 no 6 pp 450ndash458 1986

[13] G de Simone S R Daniels R B Devereux et al ldquoLeft ventric-ular mass and body size in normotensive children and adultsassessment of allometric relations and impact of overweightrdquoJournal of the American College of Cardiology vol 20 no 5 pp1251ndash1260 1992

[14] M A Quinones C M Otto M Stoddard A Waggoner andWA Zoghbi ldquoRecommendations for quantification ofDopplerechocardiography a report from the Doppler QuantificationTask Force of the Nomenclature and Standards Committeeof the American Society of Echocardiographyrdquo Journal of theAmerican Society of Echocardiography vol 15 no 2 pp 167ndash1842002

[15] K Harada T Orino K Yasuoka M Tamura and G TakadaldquoTissue Doppler imaging of left and right ventricles in normalchildrenrdquoTheTohoku Journal of ExperimentalMedicine vol 191no 1 pp 21ndash29 2000

[16] L Kapusta J M Thijssen M H M Cuypers P G M Peerand O Daniels ldquoAssessment of myocardial velocities in healthychildren using tissue Doppler imagingrdquoUltrasound inMedicineamp Biology vol 26 no 2 pp 229ndash237 2000

[17] M K Friedberg and L Mertens ldquoTissue velocities strainand strain rate for echocardiographic assessment of ventricularfunction in congenital heart diseaserdquo European Heart JournalmdashCardiovascular Imaging vol 10 no 5 pp 585ndash593 2009

[18] P Ballo I Quatrini E Giacomin A Motto and S MondilloldquoCircumferential versus longitudinal systolic function inpatients with hypertension a nonlinear relationrdquo Journal ofthe American Society of Echocardiography vol 20 no 3 pp298ndash306 2007

[19] S Ouali I Bougmiza S Abroug et al ldquoRelationship of brainnatriuretic peptide concentrations to left ventricular functionand adverse outcomes in children with end-stage renal diseaseundergoing hemodialysisrdquo Pediatric Cardiology vol 32 no 5pp 568ndash577 2011

[20] S Takatsuki T Nakayama P-N Jone et al ldquoTissue Dopplerimaging predicts adverse outcome in children with idiopathicpulmonary arterial hypertensionrdquo Journal of Pediatrics vol 161no 6 pp 1126ndash1231 2012

[21] K G Friedman D B McElhinney S D Colan et al ldquoLeftventricular remodeling and improvement in diastolic functionafter balloon aortic valvuloplasty for congenital aortic stenosisrdquoCirculation Cardiovascular Interventions vol 5 no 4 pp 549ndash554 2012

[22] A Yildirim F S Tunaolu F G Pinarli et al ldquoEarly diagnosisof anthracycline toxicity in asymptomatic long-term survivorsdobutamine stress echocardiography and tissueDoppler veloci-ties in normal and abnormalmyocardial wallmotionrdquoEuropeanHeart JournalmdashCardiovascular Imaging vol 11 no 10 pp 814ndash822 2010

[23] F I Lunze S D Colan K Gauvreau et al ldquoTissue Dopplerimaging for rejection surveillance in pediatric heart transplantrecipientsrdquoThe Journal of Heart and Lung Transplantation vol32 no 10 pp 1027ndash1033 2013

[24] A Frigiola A Giardini A Taylor et al ldquoEchocardiographicassessment of diastolic biventricular properties in patientsoperated for severe pulmonary regurgitation and association

with exercise capacityrdquo European Heart Journal CardiovascularImaging vol 13 no 8 pp 697ndash702 2012

[25] L A Rocha E Araujo Junior L C Rolo et al ldquoPrenatal detec-tion of congenital heart diseases one-year survey performinga screening protocol in a single reference center in BrazilrdquoCardiology Research and Practice vol 2014 Article ID 1756355 pages 2014

[26] Y P Mi and H Abdul-Khaliq ldquoThe pulsed Doppler and tissueDoppler-derived septal Ee1015840 ratio is significantly related toinvasive measurement of ventricular end-diastolic pressure inbiventricular rather than univentricular physiology in patientswith congenital heart diseaserdquo Clinical Research in Cardiologyvol 102 no 8 pp 563ndash570 2013

[27] L P Koopman C Slorach W Hui et al ldquoComparison betweendifferent speckle tracking and color tissue Doppler techniquesto measure global and regional myocardial deformation inchildrenrdquo Journal of the American Society of Echocardiographyvol 23 no 9 pp 919ndash928 2010

[28] M Martensson A Bjallmark and L-A Brodin ldquoEvaluationof tissue Doppler-based velocity and deformation imaginga phantom study of ultrasound systemsrdquo European HeartJournalmdashCardiovascular Imaging vol 12 no 6 pp 467ndash4762011

[29] N Risum S Ali N T Olsen et al ldquoVariability of globalleft ventricular deformation analysis using vendor dependentand independent two-dimensional speckle-tracking software inadultsrdquo Journal of the American Society of Echocardiography vol25 no 11 pp 1195ndash1203 2012

[30] M C de Knegt T Biering-Sorensen P Sogaard J SivertsenJ S Jensen and R Mogelvang ldquoConcordance and repro-ducibility between M-mode tissue Doppler imaging and two-dimensional strain imaging in the assessment of mitral annulardisplacement and velocity in patients with various heart con-ditionsrdquo European Heart JournalmdashCardiovascular Imaging vol15 no 1 pp 62ndash69 2014

[31] J A A Barbosa C C C Mota A C S E Silva M DC P Nunes and M M Barbosa ldquoAssessing pre-clinical ven-tricular dysfunction in obese children and adolescents thevalue of speckle tracking imagingrdquo European Heart JournalmdashCardiovascular Imaging vol 14 no 9 pp 882ndash889 2013

Submit your manuscripts athttpwwwhindawicom

Stem CellsInternational

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014


MEDIATORSINFLAMMATION

of


Behavioural Neurology

EndocrinologyInternational Journal of



Disease Markers


BioMed Research International

OncologyJournal of



Oxidative Medicine and Cellular Longevity


PPAR Research

The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014

Immunology ResearchHindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Journal of

ObesityJournal of



Computational and Mathematical Methods in Medicine

OphthalmologyJournal of


Diabetes ResearchJournal of



Research and TreatmentAIDS


Gastroenterology Research and Practice


Parkinsonrsquos Disease

Evidence-Based Complementary and Alternative Medicine

Volume 2014Hindawi Publishing Corporationhttpwwwhindawicom


of Anna Meyer Children Hospital Florence Italy over a9-month enrolment period were prospectively enrolled inthis study Subjects were deemed as normal if they hadunremarkable clinical history normal findings at clinicalexamination ECG standard echocardiography and no fam-ily history of genetic cardiac disease (eg Marfanrsquos syndromeor cardiomyopathy) and if they were not on pharmacologicalagents Patients with technically inadequate image qualitywere excluded Main reasons for referral were cardiovascularassessment for nonagonistic sport activities (119899 = 209) inno-cent cardiac murmur (119899 = 70) vasovagal lipothymia (119899 =27) fever (119899 = 23) atypical chest pain (119899 = 22) and familyhistory of cardiac disease (119899 = 18)The study protocol agreedwith the 1964 Helsinki declaration and successive emenda-tions and was approved by the local Ethical Committee

22 Echocardiography Studies were performed with use ofhigh-quality commercially available ultrasound systems (IE33 Philips Medical Systems Andover MA) LV dimensionsand LV mass were measured using M-mode imaging fromthe parasternal long-axis view in accordance with currentASEEACVI recommendations [11] End-diastolic and end-systolic LV volumes were calculated from apical views usingthe biplane modified Simpsonrsquos rule Left atrial volume at endsystole was measured from apical views using the biplanemethod of discs LV mass LV volumes and left atrial volumewere considered in the analysis after indexation to bodysurface area [12 13] Pulsed wave Doppler analysis of LVinflow was performed in the apical 4-chamber view byplacing the cursor at the level of the mitral leaflet tips Earlyto late peak filling velocity ratio and deceleration time weremeasured Pulsed TD of LVmitral annulus was performed inthe apical 4-chamber views by placing the sample volume atthe septal and lateral annular sites in accordancewith currentguidelines [14] To optimize quality of pulsed TD filters andgains were adjusted at the minimal optimal level allowing thebest signal-to-noise ratio The peaks of myocardial systolic(1199041015840) early diastolic (1198901015840) and late diastolic (1198861015840) waves weredetermined Average 1199041015840 1198901015840 and 1198861015840 were obtained as the meanof septal and lateral values Septal lateral and average 1198641198901015840ratio were also calculated Right ventricular systolic functionwas assessed by measuring tricuspid annulus plane systolicexcursion by M-mode and peak systolic velocity of tricuspidannulus by pulsed TD All measurements were taken byaveraging values obtained in three consecutive cycles

Reproducibility was assessed in a subset of 40 randomlyselected subjects For intraobserver analysis one investigatorreviewed echocardiographic images andmeasured septal andlateral TD velocities with at least onemonth interval betweenmeasurements For interobserver analysis two experiencedinvestigators independently reviewed images Intraobservervariability coefficients for 1199041015840 1198901015840 1198861015840 and 1198641198901015840 were as followsaverage 40 32 44 and 31 septal 42 37 45and 40 and lateral 44 39 46 and 41 Corre-sponding intraclass correlation coefficients were as followsaverage 095 096 097 and 096 septal 089 092 093 and096 lateral 097 097 095 and 096 and 119901 lt 00001 for allInterobserver variability coefficients were as follows average43 36 48 and 33 septal 48 43 50 and

45 and lateral 47 41 58 and 42 Correspondingintraclass correlation coefficients were as follows average094 095 096 and 093 septal 084 090 092 and 095lateral 096 097 094 and 096 and 119901 lt 00001 for all

23 Statistical Analysis Data are presented as mean plusmn SDor number (percentages) Correlations were expressed usingPearsonrsquos coefficients Gender comparison was performedusing the Student 119905-test for unpaired data For the deter-mination of age-specific reference ranges the following ageintervals were predefined 1-2 years 3-4 years 5-6 years 7-8 years 9ndash11 years and 12ndash16 years Multivariable regressionanalysis was used to explore the independent determinantsof TD indexes Models were obtained by testing all variableswith univariable 119901 lt 010 in a stepwise multivariableanalysis A 119865-to-remove of gt010 and 119865-to-enter of lt005were used as criteria for the selection of variables in thestepwise procedure Collinearity diagnostics were performedto explore model stability setting a cut-off of lt020 fortolerance for the identification of significantmulticollinearityIn case of multivariable models with evidence of collinearityproblems the variable with the highest variance inflationfactor was removed from the model This procedure wasiterated until a stable model was obtained The goodness-of-fit was expressed using the adjusted 1198772

For the generation of regression equations for LV lon-gitudinal indexes the following procedure was used Firstto account for heterogeneous variances across the range ofage a natural logarithmic transformation was performed foreach index to stabilize variances Second nonlinear regres-sion analysis was performed by testing different polynomialequations and choosing the best fitting model according tothe Akaike Information Criterion For all LV longitudinalindexes a third-grade polynomial regression was found toprovide the best fitting A cubic regression model of the form119884= 1198873Age3 + 119887

2Age2 + 119887

1Age + 119887

0was then obtained for each

index where 119884 represents the expected value of logarithmic-transformed longitudinal index according to age Thirdvalues were converted to the original units by exponentiatingLog 119884 and then allowing calculation of the expected valueof the echocardiographic index according to age In additionto the curve showing the expected values other 6 curvescorresponding to the 119911 scores plusmn1 plusmn2 and plusmn3 were generatedfor each index by adding or subtracting the correspondingmean square error (MSE) to the regression equation Thesignificance level for all analyses was set at 005 All testswere two-tailed Analyses were performed using the SPSS(Statistical Package for Social Sciences Chicago Illinois) forWindows release 150

3 Results

31 General Characteristics A total of 393 subjects metthe inclusion criteria during the period of study Adequatemeasurement of TD velocities was not possible in 24 subjects(in one case because of poor acoustic windows secondary topectus excavatum in the other cases because of uncontrol-lable crying in subjects aged lt4 years) Main characteristicsof the remaining 369 subjects are shown in Table 1









4 Discussion





119877 119901 119877 119901 119877 119901 119877 119901












79

100

8978

9486

0

4

8

12

16


130

174152

131

168

150

0

6

12

18

24


6570

6766 68 67

0

4

8

12

16


75

5664

7257 64

0

4

8

12

16


s998400

velo

city

(cm

s)

e998400

velo

city

(cm

s)

a998400

velo

city

(cm

s)

Ee998400

ratio

MalesFemales

MalesFemales







Aver

ages

998400(c

ms

)

Age (years)181614121086420

20

18

16

14

12

10

8

6

4

2

0

z = 3

z = 2

z = 1

z = 0

z = minus1

z = minus2

z = minus3




Age (years)181614121086420

z = 3

z = 2

z = 1

z = 0

z = minus1

z = minus2

z = minus3Aver

agee

998400(c

ms

)

30

28

26

24

22

20

18

16

14

12

10

8

6

4

2

0


Age (years)181614121086420

16

14

12

10

8

6

4

2

0

z = 3

z = 2

z = 1

z = 0

z = minus1

z = minus2

z = minus3

Aver

agea

998400(c

ms

)




Age (years)181614121086420

16

14

12

10

8

6

4

2

0

z = 3

z = 2

z = 1

z = 0

z = minus1

z = minus2

z = minus3

Aver

ageE

e998400








References








































of






Disease Markers



OncologyJournal of





PPAR Research



Journal of

ObesityJournal of
























4 Discussion





119877 119901 119877 119901 119877 119901 119877 119901












79

100

8978

9486

0

4

8

12

16


130

174152

131

168

150

0

6

12

18

24


6570

6766 68 67

0

4

8

12

16


75

5664

7257 64

0

4

8

12

16


s998400

velo

city

(cm

s)

e998400

velo

city

(cm

s)

a998400

velo

city

(cm

s)

Ee998400

ratio

MalesFemales

MalesFemales







Aver

ages

998400(c

ms

)

Age (years)181614121086420

20

18

16

14

12

10

8

6

4

2

0

z = 3

z = 2

z = 1

z = 0

z = minus1

z = minus2

z = minus3




Age (years)181614121086420

z = 3

z = 2

z = 1

z = 0

z = minus1

z = minus2

z = minus3Aver

agee

998400(c

ms

)

30

28

26

24

22

20

18

16

14

12

10

8

6

4

2

0


Age (years)181614121086420

16

14

12

10

8

6

4

2

0

z = 3

z = 2

z = 1

z = 0

z = minus1

z = minus2

z = minus3

Aver

agea

998400(c

ms

)




Age (years)181614121086420

16

14

12

10

8

6

4

2

0

z = 3

z = 2

z = 1

z = 0

z = minus1

z = minus2

z = minus3

Aver

ageE

e998400








References








































of






Disease Markers



OncologyJournal of





PPAR Research



Journal of

ObesityJournal of



















119877 119901 119877 119901 119877 119901 119877 119901












79

100

8978

9486

0

4

8

12

16


130

174152

131

168

150

0

6

12

18

24


6570

6766 68 67

0

4

8

12

16


75

5664

7257 64

0

4

8

12

16


s998400

velo

city

(cm

s)

e998400

velo

city

(cm

s)

a998400

velo

city

(cm

s)

Ee998400

ratio

MalesFemales

MalesFemales







Aver

ages

998400(c

ms

)

Age (years)181614121086420

20

18

16

14

12

10

8

6

4

2

0

z = 3

z = 2

z = 1

z = 0

z = minus1

z = minus2

z = minus3




Age (years)181614121086420

z = 3

z = 2

z = 1

z = 0

z = minus1

z = minus2

z = minus3Aver

agee

998400(c

ms

)

30

28

26

24

22

20

18

16

14

12

10

8

6

4

2

0


Age (years)181614121086420

16

14

12

10

8

6

4

2

0

z = 3

z = 2

z = 1

z = 0

z = minus1

z = minus2

z = minus3

Aver

agea

998400(c

ms

)




Age (years)181614121086420

16

14

12

10

8

6

4

2

0

z = 3

z = 2

z = 1

z = 0

z = minus1

z = minus2

z = minus3

Aver

ageE

e998400








References








































of






Disease Markers



OncologyJournal of





PPAR Research



Journal of

ObesityJournal of





















79

100

8978

9486

0

4

8

12

16


130

174152

131

168

150

0

6

12

18

24


6570

6766 68 67

0

4

8

12

16


75

5664

7257 64

0

4

8

12

16


s998400

velo

city

(cm

s)

e998400

velo

city

(cm

s)

a998400

velo

city

(cm

s)

Ee998400

ratio

MalesFemales

MalesFemales







Aver

ages

998400(c

ms

)

Age (years)181614121086420

20

18

16

14

12

10

8

6

4

2

0

z = 3

z = 2

z = 1

z = 0

z = minus1

z = minus2

z = minus3




Age (years)181614121086420

z = 3

z = 2

z = 1

z = 0

z = minus1

z = minus2

z = minus3Aver

agee

998400(c

ms

)

30

28

26

24

22

20

18

16

14

12

10

8

6

4

2

0


Age (years)181614121086420

16

14

12

10

8

6

4

2

0

z = 3

z = 2

z = 1

z = 0

z = minus1

z = minus2

z = minus3

Aver

agea

998400(c

ms

)




Age (years)181614121086420

16

14

12

10

8

6

4

2

0

z = 3

z = 2

z = 1

z = 0

z = minus1

z = minus2

z = minus3

Aver

ageE

e998400








References








































of






Disease Markers



OncologyJournal of





PPAR Research



Journal of

ObesityJournal of



















Aver

ages

998400(c

ms

)

Age (years)181614121086420

20

18

16

14

12

10

8

6

4

2

0

z = 3

z = 2

z = 1

z = 0

z = minus1

z = minus2

z = minus3




Age (years)181614121086420

z = 3

z = 2

z = 1

z = 0

z = minus1

z = minus2

z = minus3Aver

agee

998400(c

ms

)

30

28

26

24

22

20

18

16

14

12

10

8

6

4

2

0


Age (years)181614121086420

16

14

12

10

8

6

4

2

0

z = 3

z = 2

z = 1

z = 0

z = minus1

z = minus2

z = minus3

Aver

agea

998400(c

ms

)




Age (years)181614121086420

16

14

12

10

8

6

4

2

0

z = 3

z = 2

z = 1

z = 0

z = minus1

z = minus2

z = minus3

Aver

ageE

e998400








References








































of






Disease Markers



OncologyJournal of





PPAR Research



Journal of

ObesityJournal of

















Age (years)181614121086420

16

14

12

10

8

6

4

2

0

z = 3

z = 2

z = 1

z = 0

z = minus1

z = minus2

z = minus3

Aver

ageE

e998400








References








































of






Disease Markers



OncologyJournal of





PPAR Research



Journal of

ObesityJournal of












































of






Disease Markers



OncologyJournal of





PPAR Research



Journal of

ObesityJournal of





















of






Disease Markers



OncologyJournal of





PPAR Research



Journal of

ObesityJournal of
















Documents

Research Article Determinants and Regression …downloads.hindawi.com/journals/crp/2015/380729.pdfin a Healthy Italian Pediatric Population VeronicaFibbi, 1 PiercarloBallo, 1 SilviaFavilli,