hiero enteral teamprano vs tardio en infantes prematuros pediatrics

8/6/2019 Hiero Enteral Teamprano vs Tardio en Infantes Prematuros Pediatrics

http://slidepdf.com/reader/full/hiero-enteral-teamprano-vs-tardio-en-infantes-prematuros-pediatrics 1/11

DOI: 10.1542/peds.2007-04952007;120;538Pediatrics

Axel R. FranzJochen Steinmacher, Frank Pohlandt, Harald Bode, Silvia Sander, Martina Kron and

5.3 Years' Corrected AgeWith a Birth Weight of Less Than 1301 Grams: Neurocognitive Development atRandomized Trial of Early Versus Late Enteral Iron Supplementation in Infants

http://pediatrics.aappublications.org/content/120/3/538.full.htmllocated on the World Wide Web at:

The online version of this article, along with updated information and services, is

of Pediatrics. All rights reserved. Print ISSN: 0031-4005. Online ISSN: 1098-4275.Boulevard, Elk Grove Village, Illinois, 60007. Copyright © 2007 by the American Academypublished, and trademarked by the American Academy of Pediatrics, 141 Northwest Point

publication, it has been published continuously since 1948. PEDIATRICS is owned,PEDIATRICS is the official journal of the American Academy of Pediatrics. A monthly

by guest on July 11, 2011pediatrics.aappublications.orgDownloaded from

http://pediatrics.aappublications.org/content/120/3/538.full.html


http://pediatrics.aappublications.org/







ARTICLE

Randomized Trial of Early Versus Late Enteral IronSupplementation in Infants With a Birth Weight ofLess Than 1301 Grams: Neurocognitive Developmentat 5.3 Years’ Corrected AgeJochen Steinmacher, MD a,b , Frank Pohlandt,MD, MS a , Harald Bode, MD b , SilviaSander, MS c , Martina Kron, PhD c , Axel R.Franz, MD a,d

Divisions of aNeonatology and Pediatric Critical Care andb Pediatric Neurology, Department of Pediatrics, andcInstitute of Biometrics, University of Ulm, Ulm, Germany;d Department of Neonatology, Center for Pediatrics, University of Bonn, Bonn, Germany

The authors have indicated they have no nancial relationships relevant to this article to disclose.

ABSTRACT

BACKGROUND.Iron deciency in early childhood may impair neurodevelopment. In amasked, randomized, controlled trial of early versus late enteral iron supplemen-tation in preterm infants with birth weights of 1301 g, early iron supplementa-tion reduced the incidence of iron deciency and the number of blood transfu-sions.

OBJECTIVE.We sought to examine whether early enteral iron supplementation im-proves neurocognitive and motor development in these infants.

METHODS.Children who participated in the above mentioned trial were evaluated byapplying the Kaufmann Assessment Battery for Children and the Gross MotorFunction Classication Scale at the age of school entry.

RESULTS.Of the 204 infants initially randomized, 10 died and 30 were lost tofollow-up. A total of 164 (85% of the survivors) were evaluated at a mediancorrected age of 5.3 years. In this population ( n 164), the mean ( SD) mentalprocessing composite in the early iron group was 92 ( 17) versus 89 ( 16) in thelate iron group. An abnormal neurologic examination was found in 17 of 90 versus

26 of 74, and a Gross Motor Function Classication Scale score of 1 was foundin 2 of 90 versus 5 of 74, respectively. Fifty-nine of 90 children in the early irongroup were without disability, compared with 40 of 74 in the late iron group.Severe disability was found in 5 of 90 versus 6 of 74 children and 67 of 90 versus49 of 74 qualied for regular schooling, respectively.

CONCLUSIONS.Early enteral iron supplementation showed a trend toward a benecialeffect on long-term neurocognitive and psychomotor development and showed noevidence for any adverse effect. Because the initial study was not designed toevaluate effects on neurocognitive development, the power was insufcient todetect small but potentially clinically relevant improvements. Additional studiesare required to conrm the trend towards a better outcome observed in the earlyiron group.

www.pediatrics.org/cgi/doi/10.1542/peds.2007-0495

doi:10.1542/peds.2007-0495

This trial has been registered atwww.clinicaltrials.gov (identierNCT00457990).

Key Wordsneurocognitive outcome, preterm infant,very low birth weight, iron deciency, ironsupplementation

AbbreviationsGMFCS—Gross Motor FunctionClassication ScaleLOS-KF 18—Lincoln-Oseretzky Scale(Short Form) 18MPC—Mental Processing CompositeKABC—Kaufmann Assessment Battery forChildrenOR—odds ratio

CI—condence intervalAccepted for publication Apr 25, 2007

Address correspondence to Axel R. Franz, MD,Department of Neonatology, Center forPediatrics, University Women’s Hospital,Sigmund-Freud-Strasse 25, 53105 Bonn,Germany. E-mail: [email protected]

PEDIATRICS (ISSNNumbers:Print, 0031-4005;Online, 1098-4275). Copyright© 2007by theAmericanAcademy of Pediatrics

538 STEINMACHER et al by guest on July 11, 2011pediatrics.aappublications.orgDownloaded from







IRON DEFICIENCY IN infancy is associated with short-term 1,2 and long-term 3–8 neurodevelopmental decits,

delayed maturation of the auditory brainstem re-sponse, 9,10 and abnormalities of memory 11 and behav-ior, 6,12 despite adequate correction of the initial irondecit. The long-lasting neural and behavioral effects ofiron deciency in infancy have recently been reviewed. 13

Iron supplementation of term infants at risk of irondeciency is associated with improved neurodevelop-mental outcome at 12 and 24 months of age. 14–17 Becausesmall preterm infants are susceptible to iron deciency because of their small iron store at birth, 18,19 their highgrowth velocity, and the iron losses caused by frequent blood sampling, we hypothesized that iron supplemen-tation of preterm infants may also be associated withimproved neurodevelopmental outcomes.

We previously showed in a randomized trial thatearly enteral iron supplementation in infants with a birth weight 1301 g is feasible and safe, and may

reduce the incidence of iron deciency and the numberof late blood transfusions. 20 However, the effect of earlyenteral iron supplementation on neurodevelopment isunknown in these very preterm infants. Because pre-term infants may be particularly vulnerable to oxygenradical injury, 21,22 iron supplementation may theoreti-cally even be harmful for the premature brain.

The aim of this study, therefore, was to examinewhether early enteral iron supplementation improveslong-term neurocognitive and motor development invery preterm infants with a birth weight of 1301 g.

The initial randomized trial was designed to evaluatethe effects of early enteral iron supplementation on ironstatus and iron deciency and not on long-term neuro-cognitive development. This follow-up study was, there-fore, not sufciently powered to detect small, but poten-tially clinically relevant, differences between thetreatment groups.

METHODS

The initial randomized trial and our follow-up studywere both approved by the institutional review board ofthe University of Ulm, and written parental consent wasobtained for both studies.

Study SubjectsEligible were all inborn infants with a birth weight of

1301 g admitted to the University of Ulm level 3 NICU between June 1996 and June 1999. Exclusion criteriawere major anomalies, hemolytic disease, twin-to-twintransfusion syndrome, and missing parental consent. Forthe trial prole and details of the trial interventions andstatistical analyses please refer to Franz et al. 20 In theinitial randomized trial, of 380 eligible infants with a birth weight of 1301 g, 108 were excluded becauseparents refused consent, 68 were excluded on the basisof predened exclusion criteria, 105 were randomly as-

signed to receive early enteral iron supplementation,and 99 were randomly assigned to receive late enteraliron supplementation. 20

RandomizationInfants were assigned to 1 of 2 strata according to theneed for blood transfusion within the rst 7 days of life

(stratum 1: no transfusion; stratum 2: 1 transfusionwithin the rst 7 days of life). At day 7 of life, the infantswere randomly allocated in blocks of 10 within eachstratum to early or late enteral iron supplementation byusing computer-generated randomization lists.

Treatment GroupsEarly enteral iron supplementation was started at a doseof 2 mg/kg per day of ferrous sulfate as soon as 100mL/kg per day of enteral feedings were reached. 23 In 39of 105 infants randomly assigned to early iron supple-mentation, the dose was increased to 4 mg/kg per day

when the hematocrit fell below 0.30. Late enteral ironsupplementation was started at 61 days of life at a doseof 2 mg/kg per day. 24 If iron deciency was diagnosed atany time throughout the study, iron was started at 4mg/kg per day.

Enteral iron supplementation was started at a medianage of 14 days (range: 7–61 days) in the early-iron groupversus 61 days (range: 12–74 days) in the late-irongroup. 20 In both groups, iron was administered with themilk feeds, 25 erythropoietin was not administered, andrestrictive red cell transfusion guidelines were fol-lowed. 26 Of a total of 306 red cell transfusions, only 10 (6in the early-iron group and 4 in the late-iron group)were not in agreement with the transfusion guidelines.

Primary End Points, Sample Size, and Blinding of the InitialRandomized TrialPrimary outcome variables were (1) ferritin level at 61days of life and (2) the number of infants who fullledthe criteria of iron deciency at any time throughout thestudy.

A sample-size calculation revealed that 63 patientswere required per group. 20 Because of death, early dis-charge, or referral to afliated hospitals, 204 patients hadto be randomly assigned to evaluate 133 infants whocompleted the initial protocol.

The study was performed as a masked study: thelaboratory personnel who handled the blood sampleswere unaware of treatment-group assignment, and theprimary outcome variables were objective, predenedlaboratory criteria. Double-blinding for enteral iron sup-plementation was impossible because of the effect onstool color.

Follow-upAll surviving infants were followed. At 5 years of age,the infants were contacted by mail (twice) and tele-

PEDIATRICS Volume 120, Number 3, September 2007 539 by guest on July 11, 2011pediatrics.aappublications.orgDownloaded from







phone (if necessary) to arrange for a follow-up exami-nation.

Of the initially randomly assigned 204 infants, 6 diedduring their initial hospitalization, and 4 died after dis-charge from the hospital. Of the surviving 194 infants,164 (85%) were evaluated. Thirty infants were not eval-uated because 5 (3%) refused participation, and 12 (6%)

were unable to attend because the family had moved,had no means of transportation, or did not come toseveral (at least 3) appointments for various reasons, and13 (7%) were lost to follow-up.

Standardized Follow-up AssessmentThe neurologic examination and all neurophysiologicaltesting was performed by an experienced pediatric neu-rologist (Dr Steinmacher) who was blinded to the in-fants’ treatment-group assignment.

The neurologic examination was rated as normal,mildly abnormal (in the presence of minor neurologic

signs, such as broad gait, dysdiadochokinesis, or dysme-tria) and severely abnormal (in the presence of anyparesis with or without spasticity, any cerebral nervepalsy, or any ataxia).

For evaluation of mobility, that is, the motor func-tioning, the Gross Motor Functioning Classication Scale(GMFCS) 27 was performed. A score of 0 represents nor-mal mobility, 1 represents mild abnormality (ie, walking,running, jumping are possible but somewhat reduced inprecision and velocity). A score of 2, 3, or 4 representobviously and severely impaired mobility and the lack ofindividual mobility, respectively.

To assess motor coordination, the Lincoln-OseretzkyScale (Short Form) 18 (LOS-KF 18) 28 was applied. Thetest consists of 18 tasks including walking backward,standing on 1 foot, touching one’s nose, jumping,throwing, catching, clapping, balancing tiptoe, and soon. Different reference values are available for this testfor children with a mental processing composite (MPC)of 71, 71 to 85, and 85. Abnormal motor coordina-tion was dened as an LOS-KF 18 score in the 3rdpercentile.

For evaluation of the cognitive function the Kauf-mann Assessment Battery for Children (KABC) was per-formed. The KABC comprises 2 summative scales: (1)the MPC, a global measure of cognitive ability in 2subscales, sequential processing, and simultaneous pro-cessing, and (2) the achievement scale, an assessment ofknowledge of facts, language concepts, and school-re-lated skills. The range of possible scores for both scales is40 to 150. The test was last standardized in 1992 to amean of 100 and an SD of 15 in a German referencepopulation. 29 The MPC can be interpreted similarly to anIQ test. Children whose severe cognitive impairment ordisability precluded the use of this assessment tool wereassigned a score of 30 if minimal speech and the abilityfor minimal communication with the parents were

present, and a score of 20 if no speech was present but atleast minimal sensory or motor achievements were elic-ited.

Assessment of visual impairment was based on oph-thalmologic records, and visual acuity was evaluated bystandard visual acuity charts. Severe visual impairmentwas dened as a refractory error in 1 eye of more than

10 diopter or any amblyopia dened as a best-cor-rected visual acuity of 20/40. A visual acuity after best-possible correction for ametropia 20/200 was de-ned as blindness.

Visual perception was assessed in children whoachieved a MPC on KABC of 85 by applying the scalesof visual perception of the Tubinger Lurija Christensenneurophysiological examination for children. 30 This testincludes evaluation of eye-motor coordination, spatialorientation (position in space and spatial relationship),and pattern recognition. Raw scores are obtained andthen converted to age-adjusted percentiles. Visual per-

ception was dened as impaired if the child scored in the15th percentile of a previously published cohort ofnormal children 31 in all subscales.

To assess developmental abnormalities in childhood behavior, the parents were asked to complete the ChildBehavior Check List for 4- to 18-year-old children in itsGerman adaptation of 1994. 32 Parents completed a ques-tionnaire regarding their child’s performance in games,activities, chores, and the quality of relationships withfriends and family. A total of 113 items related to be-havior had to be scored on a 3-point scale ranging fromnot true (0), somewhat true (1), to often true (2). A totalproblem score was obtained by summing all items. Rawscores were converted to age-standardized scores ( t scores[mean: 50; SD: 10]). Higher scores indicate more behav-ioral problems: t scores of 70 and 63 dened abnor-mal behavior in each subscale and in the summary scalesof total problems and internalizing and externalizing behavior, respectively.

Composite Outcome CriteriaSevere disability was dened as any of the following: anabnormal neurologic examination resulting in severelyimpaired mobility (GMFCS: 1), severe cognitive im-pairment (MPC: 51), hearing loss requiring amplica-tion, or blindness. Provided that none of the previouslymentioned criteria for classication as severe disabilitywere met, moderate disability was dened as any of thefollowing: any abnormal neurologic examination asso-ciated with a detectable impairment of mobility (GM-FCS: 1), cognitive impairment (MPC: 51–70), or anysevere impairment of vision. Provided that none of theabove mentioned criteria for classication as severe ormoderate disability were met, mild disability was denedas any abnormal neurologic examination result withnormal mobility (GMFCS: 0) and/or an MPC of 71 to 85.

The absence of signicant impairment (without dis-








ability) was dened as normal neurologic examination,normal mobility (GMFCS: 0), and normal cognitive de-velopment (MPC: 85), and the absence of severe hear-ing and visual impairment.

Recommendations for school assignment were basedon the previously described evaluations and the abilityto compensate impairments; that is, everyday function-ing of the children. In general, an MPC of 85 andabsence of any severe impairment were required. If im-pairments were present, these had to be compensated.

Statistical AnalysesFor qualitative data, counts and percentages were calcu-lated; for quantitative data, the mean, SD, median, min-imum, and maximum were calculated.

Categorical variables were compared between the pa-tient group with early and the patient group with lateiron supplementation with 2 test, or in case of smallnumbers with Fisher’s exact test, continuous variableswere compared with the Wilcoxon test.

Associations between risk factors and poor neurode-velopmental outcome were described by odds ratios(ORs) adjusted for treatment-group assignment with a95% condence interval (CI) and a P value of the 2 testfrom a logistic regression. The following risk factors were

evaluated: gestational age, gender, multiple birth, severeintraventricular or periventricular hemorrhage ( 3°),periventricular leukomalacia, severe retinopathy of pre-maturity ( 3°), need for mechanical ventilation, intra-uterine growth retardation, language, and maternal ed-ucation.

Multiple logistic regression with forward selection(selection level: 5%) was performed to identify impor-tant risk factors with simultaneous value for predictionamong the previously mentioned risk factors for poorneurodevelopmental outcome adjusted for assignmentto late versus early iron supplementation. ORs with 95%CIs and P values were calculated. The value for predic-tion of outcome of the important risk factors identiedwas assessed by the area under the receiver operatingcharacteristic curve.

All analyses were performed with SAS 9.0 (SAS In-stitute, Cary, NC).

RESULTS

A total of 164 (85%) of 194 surviving initially randomlyassigned children completed the follow-up assessment ata median corrected age of 5.3 years (range: 4.7–6.6years). The demographic data and neonatal morbiditiesof these 164 infants and of the 30 infants who were lost

TABLE 1 Demographic and Neonatal Morbidity Data

Early Iron Late Iron

Follow-up(n 90)

Lost to Follow-up(n 10)

Follow-up(n 74)

Lost to Follow-up(n 20)

Gestational ageMean SD, wk 27.6 2.2 26.7 1.7 27.4 2.5 26.9 1.7Median (minimum–maximum), wk 27.7(23.9–35.4) 26.6(23.9–28.9) 27.1(23.6–35.4) 26.9(23.7–29.9)

Birth weightMean SD, g 884 222 801 237 863 215 916 249Median (minimum–maximum), g 870 (380–1310) 810 (390–1140) 865 (370–1300) 890 (580–1300)

Birth weight 3rd percentile,n (%) 13 (14) 1 (10) 12 (16) 1 (5)Female gender, n (%) 47 (52) 4 (40) 40 (54) 12 (60)Any antenatal steroids,n (%) 82 (91) 9 (90) 64 (86) 19 (95)Critical Risk Index for Babies score33

Mean SD, wk 4.8 3.5 6.3 3.4 5.4 4.0 5.4 3.7Median (minimum–maximum), wk 4 (1–16) 5.5 (3–15) 4 (1–15) 5 (1–13)

3, n (%) 32 (36) 0 (0) 23 (31) 6 (30)3–7, n (%) 34 (38) 8 (80) 32 (43) 7 (35)8–12, n (%) 21 (23) 1 (10) 14 (19) 5 (25)

12, n (%) 2 (2) 1 (10) 5 (7) 1 (5)Not documented, n (%) 1 (1) 0 (0) 0 (0) 1 (5)

Intraventricular hemorrhage 3°,n (%) 5 (6) 0 (0) 6 (8) 3 (15)Periventricular leukomalacia,n (%) 2 (2) 0 (0) 1 (1) 0 (0)Retinopathy of prematurity 3°,n (%) 5 (6) 0 (0) 10 (14) 0 (0)Necrotizing enterocolitis Bell stage 2,

n (%)4 (4) 2 (20) 7 (9) 1 (5)

Chronic lung disease,n (%)a 16 (18) 4 (40) 22 (30) 4 (20)Duration of stay in hospital,n (%)

51 d 12 (13) 3 (30) 13 (18) 6 (30)51–100 d 56 (62) 2 (20) 37 (50) 9 (45)

100 d 22 (24) 5 (50) 24 (32) 5 (25)a Chroniclung diseaseis dened asa minimumfractionof inspiredoxygenof 0.21to achievean pulseoxygensaturationof 90%at36 weeks’postmenstrual age.








to follow-up are summarized according to treatment-group assignment in Table 1. Infants lost to follow-upwere similar to infants with complete follow-up. In theinfants with complete follow-up, there were no differ-ences in any prenatal or perinatal variable between the2 treatment groups.

In the late-iron group, 8 infants with complete fol-low-up received the rst dose of enteral iron “early” (ie,at days of life 12, 32, 40, 41, 41, 42, 43, and 46) accord-ing to the study protocol on the basis of predenedcriteria of iron deciency. 20 In the early-iron group, 11infants with complete follow-up received the rst doseof enteral iron 1 week too late (ie, at the days of life 16,

18, 21, 21, 26, 27, 28, 31, 37, 51, and 61). In theseinfants, iron was started late because the staff forgot toinitiate supplementation as soon as enteral nutrition of

100 mL/kg per day was reached.Table 2 summarizes the main outcome data. First,

there were more infants with an abnormal result onstandardized neurologic examination in the group re-ceiving late enteral iron supplementation. This differ-ence was mainly because of an increased number ofinfants with mild neurologic abnormalities (19 of 74 inthe late-iron and 11 of 90 in the early-iron group; P .04). Second, there were suggestive trends toward im-proved cognitive development with early iron supple-

TABLE 2 Outcome Data

Long-term Survivors With Full Follow-up Early Iron(n 90)

Late Iron(n 74)

P

Neurologic examination,n (%)Normal 73 (81) 48 (65)Mildly or severely abnormal 17 (19) 26 (35) .02a

Growth at a median corrected age of 5.3 y(range: 4.7–6.6 y),n (%)

Weight 3rd percentile 15 (17) 18 (24) .22a

Length 3rd percentile 9 (10) 7 (9) .91a

Head circumference 3rd percentile 15 (17) 19 (26) .16a

Hearing and vision,n (%)Requiring hearing aid 1 (1) 0Bilateral blindness 0 1 (1)Severe visual impairment 4 (4) 2 (3)

Mobility,n (%)GMFCS 0 81 (90) 61 (82)GMFCS 1 7 (8) 8 (11)GMFCS 1 2 (2) 5 (7) .25b,c

Cognitive developmentMPC

Mean SD 92 17 89 16

Median (minimum–maximum) 96 (30–124) 91 (30–116) .10d

Score of 51, n (%) 3 (3) 2 (3)Score of 51–70,n (%) 5 (6) 7 (9)Score of 71–85,n (%) 16 (18) 16 (22)Score of 85, n (%) 66 (73) 49 (66)

Sequential processing scoreMean SD 95 12 92 15Median (minimum–maximum) 94 (48–125) 93 (46–128) .16d

Simultaneous processing scoreMean SD 95 16 90 16Median (minimum–maximum) 97 (55–137) 93 (50–115) .10d

Achievement scale scoreMean SD 96 15 91 17Median (minimum–maximum) 97 (50–133) 93 (50–129) .08d

Composite outcome variablesAny severe disability 5 (6) 6 (8)Any moderate disability 10 (11) 10 (14)Any mild disability 16 (18) 18 (24)Without signicant disability 59 (66) 40 (54) .13a,e

School recommendationQualied for regular schooling 67 (74) 49 (66) .25a

a 2 test.b Fisher’s exact test.c Comparison of GMFCS scores of 1 vs 1.d Wilcoxon test.e Comparison of with versus without signicant disability.








mentation in the overall MPC, the subscale of simulta-neous processing, and the achievement scale of theKABC.

There was no difference in the proportion of childrenwith impaired motor coordination; the incidence of anLOS-KF 18 score in the 3rd percentile was 25 (28%) inthe early-iron group and 20 (27%) in the late-iron group

(P .93). The number of infants with abnormal behav-ior according to the Child Behavior Checklist was alsosimilar in both groups: 17 (20%) of 87 vs 13 (19%) of 67(P .98).

In children with normal MPC ( 85; n 115), theincidence of impaired visual perception was 7 (11%) of66 in the early-iron group versus 5 (10%) of 49 in thelate-iron group ( P .97).

A total of 133 infants completed the initial random-ized trial per protocol. One of these infants died aftercompletion of the trial, 115 (87% of the survivors) werereevaluated with similar results as in the intention-to-

treat population described earlier (data available on re-quest).Table 3 presents the crude OR of late versus early iron

supplementation as a risk factor for moderate or severedisability and for cognitive impairment along with theadjusted ORs for several prenatal and perinatal risk fac-tors for poor neurodevelopmental outcome.

The results of multiple logistic regression analysesevaluating the effect of early versus late enteral ironsupplementation with variable selection of the risk fac-tors listed in Table 3 are summarized in Table 4. In thenal model, late versus early iron supplementation wasnot an important risk factor with additional value forprediction for impaired outcome. The areas under the

receiver operating characteristic curve for the logisticmodels of mildly or severely abnormal neurologic exam-ination, of moderate or severe disability, and of cognitiveimpairment (MPC: 71) were 0.739, 0.861, and 0.869,respectively.

DISCUSSION

Preterm infants are at risk of impaired neurodevelop-mental outcome, 34–36 and additional efforts to identifyeffective interventions to improve outcome in these vul-nerable children are required, although it is unlikely thata single therapeutic approach will improve long-termoutcome to a major degree.

Thus far, iron supplementation of preterm infants hasonly been shown to be an effective component to pre-vent and/or treat anemia of prematurity. 20,23 We hypoth-esized that iron supplementation administered early toprevent iron deciency may also be a successful strategyto improve long-term neurodevelopmental outcome in

preterm infants on the basis of the following observa-tions.Iron deciency is common in unsupplemented pre-

term infants as outlined previously, and iron deciencyis associated with impaired neurodevelopmental out-come. The negative effects of iron deciency duringinfancy and childhood on neurodevelopment may atleast in part be long-lasting or even irreversible despiteadequate treatment of iron deciency. 3–12 Similarly, irondeciency during intrauterine development seems tohave long-lasting negative effects on neurodevelopment(reviewed by Lozoff and Georgieff 37), because decreasedumbilical cord serum ferritin levels were associated withpoorer neurobehavioral status at 37 weeks’ postmen-

TABLE 3 ORsof Risk Factors for Mildlyor Severely Abnormal Neurologic Examination, for Moderate or SevereDisability,and for CognitiveImpairment (MPC < 71) Adjusted for Assignment to Late Versus EarlyIron Supplementation

Mildly or Severely AbnormalNeurologic Examination

Moderate or Severe Disability Cognitive Impairment(MPC 71)

OR (95% CI) P OR (95% CI) P OR (95% CI) P

Assignment to treatment groupIron supplementation, late vs early 2.3 (1.1–4.7) .02 1.4 (0.6–3.0) .42 1.4 (0.5–3.9) .50

Risk factorGestational age, wk 0.8 (0.7–1.0) .02 0.6 (0.5–0.8) .01 0.6 (0.4–0.8) .01

Gender, male vs female 1.7 (0.8–3.4) .16 1.3 (0.6–2.8) .55 0.6 (0.2–1.7) .32Multiple birth, yes vs no 1.0 (0.5–2.2) .90 0.5 (0.2–1.1) .09 0.3 (0.1–1.3) .11SGA, yes vs no 1.1 (0.4–2.9) .87 1.1 (0.4–3.1) .89 1.2 (0.3–4.5) .78IVH/PVH 3° vs 3° 16.1 (3.2–80.2) .01 14.8 (3.6–60.2) .01 9.6 ( 2.5–36.2) .01Periventricular leukomalacia, yes vs no 1.6 (0.1–18.7) .72 2.2 (0.2–25.9) .52 4.7 (0.4–55.8) .22ROP, 3° vs 3° 4.5 (1.5–13.7) .01 8.5 ( 2.7–26.5) .01 5.5 ( 1.6–19.2) .01MV, yes vs no 3.5 (1.6–8.1) .01 31.0 (4.1–233.9) .01 a

Duration of MV, 7 vs 7 d 5.4 (2.5–11.6) .01 20.3 (7.0–58.9) .01 21.3 (4.6–98.8) .01Language, other vs German 1.4 (0.4–4.3) .60 2.3 (0.7–7.3) .16 3.7 (1.0–13.3) .05Highest academic degree of mother, none or

lowest school degree vs higher degreesb2.6 (1.2–5.5) .01 2.2 (1.0–4.9) .06 4.2 (1.4–12.9) .01

SGAindicatessmallfor gestationalage (ie,birthweight 3rdpercentilefor gestationalage); IVH/PVH,intraventricular/periventricularhemorrhage; ROP,retinopathyof prematurity; MV,mechanicalventilation.a Not estimable because of zero cells.b Lowest level of the 3-level German school system that qualies for nonacademic professional education.








strual age 38 and impaired mental and psychomotor de-

velopment at 5 years of age.39

Furthermore, infants ofdiabetic mothers who have decreased iron stores, de-ned as umbilical cord serum ferritin levels of 35 g/L,had abnormal auditory recognition memory and a lowerphysical developmental index at 1 year of age. 11,40 Fi-nally, low levels of neonatal hemoglobin, serum iron,and or ferritin in cord blood were associated with higherlevels of negative emotionality, and lower levels of alert-ness and soothability. 41

Our hypothesis, that early iron supplementation ofpreterm infants may improve neurodevelopmental out-come, was further supported by the fact that iron sup-plementation of term infants at risk of iron deciencyimproves neurodevelopmental outcome at 12 and 24months of age. 14–17

In our study, early enteral iron supplementation wasassociated with fewer long-term neurologic abnormali-ties on standardized physical examination and trendstoward improved outcomes on formal assessment ofmental processing (Table 2). These ndings may reectlong-term consequences of perinatal iron deciency pre-viously observed by others 38–41 and may be related tonegative effects of perinatal iron deciency on myelina-tion, brain iron content, striatal and hippocampal orga-nization, and neurotransmitter metabolism observed inrodent models of perinatal iron deciency (recently re-viewed in 37). However, taking into account that multiplecomparisons were made, the ndings of our study haveto be interpreted with caution.

The lack of more obvious benecial results of earlyiron supplementation may have several explanations.First, there may be no benet (and the previously men-tioned results are chance ndings). Second, our initialrandomized trial was designed to examine effects ofearly iron supplementation on iron stores, the incidenceof iron deciency, and the need for blood transfusionsand not to evaluate long-term neurodevelopmental out-

come. The study was, therefore, underpowered to detect

small but clinically important differences in long-termoutcome data between groups. To have 80% power todemonstrate at a 2-sided signicance level of 5% in aconrmatory trial that early iron supplementation doesindeed improve neurodevelopment assuming a degreesuggested in the present sample (eg, effect size for theMPC 0.2 [ie, difference of the means divided by SD])a sample size of n 394 infants per treatment groupwould be required. One may argue that an effect size of0.2 is negligible and not worth pursuing. On the otherhand, it is unlikely that any single intervention in neo-natal care will result in a higher standardized meandifference in the MPC, because the population and theassociated problems are so heterogeneous.

Third, an oral dose of 4 to 6 mg/kg per day may have been insufcient to meet the iron needs of all growingpreterm infants. In fact, despite increasing iron supple-mentation up to 4 mg/kg per day (in formula-fed infantsup to 6 mg/kg per day) as soon as the hematocrit fell below 0.30, iron deciency occurred in 15% of infantsin the early-iron group during the initial randomizedtrial. 20 This is not surprising, considering that (1) only

25% (range: 10%–50%) of enterally administered ironis absorbed, 25,42 (2) the daily needs of growing preterminfants are thought to be 0.5 to 1 mg/kg per day, 43 and(3) that every milliliter of blood loss represents a loss of0.35 to 0.5 mg of iron (assuming hemoglobin concen-trations of 10–15 g/dL). Consequently, higher doses ofenteral iron supplementation may be required to meetthe needs of at least some preterm infants and to opti-mize neurodevelopmental outcome in this vulnerablepopulation.

One of the concerns with iron supplementation isthat free ferrous iron is thought to increase the produc-tion of free radicals (known as Fenton reaction) andthereby to increase oxidative stress especially in the pre-mature infant who has a limited capacity to assimilate

TABLE 4 Results of Multiple Logistic Regression With Forward Selection of Risk Factors for Mildlyor SeverelyAbnormal NeurologicExamination, for Moderate or Severe Disability, and for Cognitive Impairment (MPC < 71) Adjusted for Assignment to Late VersusEarlyIron Supplementation

Mildly or Severely AbnormalNeurologic Examination

Moderate or SevereDisability

Cognitive Impairment(MPC 71)

OR (95% CI) P OR (95% CI) P OR (95% CI) P

Assignment to treatment groupIron supplementation, late vs early 1.8 (0.8–4.1) .15 0.7 (0.3–2.0) .50 0.9 (0.3–3.1) .92

Risk factor —IVH/PVH 3° vs 3° 7.4 (1.4–38.4) .02 5.9 (1.2–29.0) .03 —ROP of 3° vs 3° — 5.8 (1.4–23.6) .01 —Duration of MV of 7 vs 7 d 4.1 (1.8–9.3) .01 11.5 (3.7–35.6) .01 19.9 (4.1–95.4) .01Highest academic degree of mother, none or

lowest school degree vs higher degreesa— — 4.4 (1.3–14.9) .02

— indicates not an important risk factor with additional value for prediction in this model; IVH/PVH, intraventricular/periventricular hemorrhage; ROP, retinopathy of prematurity; MV, mechanicalventilation.a Lowest level of the 3-level German school system that qualies for nonacademic professional education.








free iron and to degrade free radicals. 21,22,44 Early ironsupplementation of premature infants may thereforetheoretically cause oxidative injury to the developing brain and consequently may result in impaired long-term neurodevelopment. However, there was no evi-dence of such an adverse effect after early enteral ironsupplementation in our sample. This is in agreement

with the fact that, to our knowledge, enteral iron sup-plementation has not yet been shown to increase surro-gate markers of oxidative injury in preterm infants. 45,46

However, the sample size of our study was insufcient todetect infrequent adverse effects of early enteral ironsupplementation.

The main limitations of this study were the insuf-cient sample size as mentioned earlier and the fact thatthe evaluation of the effects on neurodevelopmentaloutcome was not considered when the initial random-ized trial was designed. The results of this study, there-fore, do not conrm but may support the hypothesis that

early enteral iron supplementation improves neurocog-nitive and motor development in very preterm infants.Furthermore, because the study was conducted at a

single institution, the results may not have general va-lidity and require verication by a larger multicentertrial.

On the basis of the results reported and discussedearlier, such a multicenter trial of iron supplementationin preterm infants should aim to evaluate measures ofmyelination (eg, latency in brainstem-evoked responseaudiometry 9,10 ), hippocampal function, that is, memory(eg, auditory recognition memory in response to moth-er’s voice 11 ), frontal lobe function, that is, emotionalityand behavior (as described in detail in 12,17,47,48 ) in addi-tion to standardized neurologic examination, and globalevaluation of motor and neurocognitive development.

On the basis of the results of our study, the incidenceof an abnormal result on standardized neurologic exam-ination should be chosen as primary outcome variable.Assuming an OR of 1.8 (Table 4), a 2-sided signicancelevel of .05, a power of 80%, and an incidence of ab-normal neurologic examination of 19% in the early-irongroup (Table 2), a sample-size calculation revealed that252 infants would be required per group for such a trial.

CONCLUSIONS

This long-term follow-up study after a randomized trialof early versus late enteral iron supplementation in verypreterm infants with a birth weight 1301 g showedtrends toward improved neurodevelopmental outcomein the early-iron group. These trends and the reductionof late blood transfusion observed during the initial tri-al 20 suggest that early enteral iron supplementation maydo more good than harm. However, additional suf-ciently powered trials on the long-term effects of earlyenteral iron supplementation in very preterm infants arerequired to test this hypothesis.

ACKNOWLEDGMENTS

This study was supported by the Else-Kroner-FreseniusFoundation (Bad Homburg, Germany) and the Rudolfand Clothilde Eberhardt Foundation (Ulm, Germany).

REFERENCES

1. Lozoff B, Brittenham GM, Wolf AW, et al. Iron deciencyanemia and iron therapy effects on infant developmental testperformance [published correction appears in Pediatrics . 1988;81:683]. Pediatrics. 1987;79:981–995

2. Walter T, De Andraca I, Chadud P, Perales CG. Iron deciencyanemia: adverse effects on infant psychomotor development.Pediatrics. 1989;84:7–17

3. Palti H, Meijer A, Adler B. Learning achievement and behaviorat school of anemic and non-anemic infants. Early Hum Dev.1985;10:217–223

4. Lozoff B, Jimenez E, Wolf AW. Long-term developmental out-come of infants with iron deciency. N Engl J Med. 1991;325:687–694

5. Hurtado EK, Claussen AH, Scott KG. Early childhood anemiaand mild or moderate mental retardation. Am J Clin Nutr.

1999;69:115–1196. Lozoff B, Jimenez E, Hagen J, Mollen E, Wolf AW. Poorer

behavioral and developmental outcome more than 10 yearsafter treatment for iron deciency in infancy. Pediatrics. 2000;105(4). Available at: www.pediatrics.org/cgi/content/full/105/4/e51

7. Lozoff B, Jimenez E, Smith JB. Double burden of iron de-ciency in infancy and low socioeconomic status: a longitudinalanalysis of cognitive test scores to age 19 years. Arch Pediatr Adolesc Med. 2006;160:1108–1113

8. Shar T, Angulo-Barroso R, Calatroni A, Jimenez E, Lozoff B.Effects of iron deciency in infancy on patterns of motor de-velopment over time. Hum Mov Sci. 2006;25:821–838

9. Roncagliolo M, Garrido M, Walter T, Peirano P, Lozoff B.Evidence of altered central nervous system development ininfants with iron deciency anemia at 6 mo: delayed matura-tion of auditory brainstem responses. Am J Clin Nutr. 1998;68:683–690

10. Algarin C, Peirano P, Garrido M, Pizarro F, Lozoff B. Irondeciency anemia in infancy: long-lasting effects on auditoryand visual system functioning. Pediatr Res. 2003;53:217–223

11. Siddappa AM, Georgieff MK, Wewerka S, Worwa C, NelsonCA, Deregnier RA. Iron deciency alters auditory recognitionmemory in newborn infants of diabetic mothers. Pediatr Res.2004;55:1034–1041

12. Corapci F, Radan AE, Lozoff B. Iron deciency in infancy andmother–child interaction at 5 years. J Dev Behav Pediatr. 2006;27:371–378

13. Lozoff B, Beard J, Connor J, Barbara F, Georgieff M, Schallert

T. Long-lasting neural and behavioral effects of iron deciencyin infancy. Nutr Rev. 2006;64:S34–S4314. Moffatt ME, Longstaffe S, Besant J, Dureski C. Prevention of

iron deciency and psychomotor decline in high-risk infantsthrough use of iron-fortied infant formula: a randomizedclinical trial. J Pediatr. 1994;125:527–534

15. Williams J, Wolff A, Daly A, MacDonald A, Aukett A, BoothIW. Iron supplemented formula milk related to reduction inpsychomotor decline in infants from inner city areas: random-ised study [published correction appears in BMJ . 2000;321:23].BMJ. 1999;318:693–697

16. Friel JK, Aziz K, Andrews WL, Harding SV, Courage ML,Adams RJ. A double-masked, randomized control trial of ironsupplementation in early infancy in healthy term breast-fedinfants. J Pediatr. 2003;143:582–586








17. Lozoff B, De Andraca I, Castillo M, Smith JB, Walter T, Pino P.Behavioral and developmental effects of preventing iron-deciency anemia in healthy full-term infants [published cor-rection appears in Pediatrics . 2004;113:1853]. Pediatrics. 2003;112:846–854

18. Widdowson EM, Spray CM. Chemical development in utero. Arch Dis Child. 1951;26:205–215

19. Hågå P. Plasma ferritin concentrations in preterm infants incord blood and during the early anaemia of prematurity. ActaPaediatr Scand. 1980;69:637–641

20. Franz AR, Mihatsch WA, Sander S, Kron M, Pohlandt F. Pro-spective randomized trial of early versus late enteral iron sup-plementation in infants with a birth weight of less than 1301grams. Pediatrics. 2000;106:700–706

21. Lackmann GM, Schnieder C, Bohner J. Gestational age-dependent reference values for iron and selected proteins ofiron metabolism in serum of premature human neonates. Biol Neonate. 1998;74:208–213

22. Sullivan JL Iron, plasma antioxidants, and the “oxygen radicaldisease of prematurity. ” Am J Dis Child. 1988;142:1341–1344

23. Lundstrom U, Siimes MA, Dallman PR. At what age does ironsupplementation become necessary in low-birth-weight in-fants? J Pediatr. 1977;91:878–883

24. American Academy of Pediatrics, Committee on Nutrition. Nu-tritional needs of low-birth-weight infants. Pediatrics. 1985;75:976–986

25. McDonald MC, Abrams SA, Schanler RJ. Iron absorption andred blood cell incorporation in premature infants fed an iron-fortied infant formula. Pediatr Res. 1998;44:507–511

26. Franz AR, Pohlandt F. Red blood cell transfusions in very andextremely low birthweight infants under restrictive transfusionguidelines: is exogenous erythropoietin necessary. Arch DisChild Fetal Neonatal Ed. 2001;84:F96–F100

27. Palisano R, Rosenbaum P, Walter S, Russell D, Wood E, Ga-luppi B. Development and reliability of a system to classifygross motor function in children with cerebral palsy. Dev Med Child Neurol. 1997;39:214–223

28. Luer G, Cohen R, Eggert D. Determination of motor ability inhandicapped children by use of a Hamburg version of theLincoln-Oseretzky motor development scale [in French]. Prax Kinderpsychol Kinderpsychiatr. 1970;19:18–25

29. Melchers P, Preuss U. Adaptation of the Kaufman AssessmentBattery for Children for German- speaking areas. Part 1: Intro-duction of the battery [in German]. Z Kinder Jugendpsychiatr.1992;20:85–93

30. Deegener B. Neuropsychological Diagnostics in Children andAdolescents [in German]. Weinheim, Germany: Beltz; 1992

31. Steinmacher J, Storck M, Fingerle J, Gra le A, KupferschmidC, Bode H. Diagnosis of partial cognitive decits in 5-year-oldchildren [in German]. Kinder Jugendarzt. 2002;33:638–642

32. Dopfner M, Schmeck K, Berner W, Lehmkuhl G, Poustka F.Reliability and factorial validity of the Child Behavior

Checklist: an analysis of a clinical and eld sample [in Ger-man]. Z Kinder Jugendpsychiatr. 1994;22:189–205

33. International Neonatal Network. The CRIB (Clinical Risk Indexfor Babies) score: a tool for assessing initial neonatal risk andcomparing performance of neonatal intensive care units. Lan-cet. 1993;342:193–198

34. Hack M, Taylor HG, Klein N, Eiben R, Schatschneider C, Mer-curi Minich N. School-age outcomes in children with birthweights under 750 g. N Engl J Med. 1994;331:753–759

35. Vohr BR, Allan WC, Westerveld M, et al. School-age outcomesof very low birth weight infants in the indomethacin intraven-tricular hemorrhage prevention trial. Pediatrics. 2003;111(4).Available at: www.pediatrics.org/cgi/content/full/111/4/e340

36. Marlow N, Wolke D, Bracewell MA, Samara M. Neurologic anddevelopmental disability at six years of age after extremelypreterm birth. N Engl J Med. 2005;352:9–19

37. Lozoff B, Georgieff MK. Iron deciency and brain develop-ment. Semin Pediatr Neurol. 2006;13:158–165

38. Armony-Sivan R, Eidelman AI, Lanir A, Sredni D, Yehuda S.Iron status and neurobehavioral development of prematureinfants. J Perinatol. 2004;24:757–762

39. Tamura T, Goldenberg RL, Hou J, et al. Cord serum ferritinconcentrations and mental and psychomotor development ofchildren at ve years of age. J Pediatr. 2002;140:165–170

40. Nelson CA, Wewerka SS, Borscheid AJ, Deregnier RA, Geor-

gieff MK. Electrophysiologic evidence of impaired cross-modalrecognition memory in 8-month-old infants of diabetic moth-ers. J Pediatr. 2003;142:575–582

41. Wachs TD, Pollitt E, Cueto S, Jacoby E, Creed-Kanashiro H.Relation of neonatal iron status to individual variability inneonatal temperament. Dev Psychobiol. 2005;46:141–153

42. Widness JA, Serfass RE, Haiden N, Nelson SE, Lombard KA,Pollak. A Erythrocyte iron incorporation but not absorption isincreased by intravenous iron administration in erythropoie-tin-treated premature infants. J Nutr. 2006;136:1868–1873

43. Friel JK, Andrews WL, Hall MS, et al. Intravenous iron admin-istration to very-low-birth-weight newborns receiving totaland partial parenteral nutrition. JPEN J Parenter Enteral Nutr.1995;19:114–118

44. Inder TE, Clemett RS, Austin NC, Graham P, Darlow BA. Highiron status in very low birth weight infants is associated withan increased risk of retinopathy of prematurity. J Pediatr. 1997;131:541–544

45. van Zoeren Grobben D, Moison RMW, Haasnoot AA. Ironcontaining feeding formula and oxidative stress in preterm babies. Pediatr Res. 1998;43:270A

46. Friel JK, Aziz K, Andrews WL, Serfass RE. Iron absorption andoxidant stress during erythropoietin therapy in very low birthweight premature infants: a cohort study. BMC Pediatr. 2005;5:29

47. Perez EM, Hendricks MK, Beard JL, et al. Mother-infant inter-actions and infant development are altered by maternal irondeciency anemia. J Nutr. 2005;135:850–855

48. Lozoff B, Corapci F, Burden MJ, et al. Preschool-aged children

with iron deciency anemia show altered affect and behavior.J Nutr. 2007;137:683–689








DOI: 10.1542/peds.2007-04952007;120;538Pediatrics

Axel R. FranzJochen Steinmacher, Frank Pohlandt, Harald Bode, Silvia Sander, Martina Kron and

5.3 Years' Corrected AgeWith a Birth Weight of Less Than 1301 Grams: Neurocognitive Development atRandomized Trial of Early Versus Late Enteral Iron Supplementation in Infants

ServicesUpdated Information &

mlhttp://pediatrics.aappublications.org/content/120/3/538.full.htincluding high resolution figures, can be found at:

References

ml#ref-list-1http://pediatr ics.aappublications.org/content/120/3/538.full.htat:This article cites 45 articles, 16 of which can be accessed free

Citations

ml#related-urlshttp://pediatrics .aappublications.org/content/120/3/538.full.ht

This article has been cited by 2 HighWire-hosted articles:

Subspecialty Collections

http://pediatrics.aappublications.org/cgi/collection/bloodBloodthe following collection(s):This article, along with others on similar topics, appears in

Permissions & Licensing

mlhttp://pediatrics.aappublications.org/site/misc/Permissions.xhttables) or in its entirety can be found online at:Information about reproducing this article in parts (figures,

Reprints http://pediatrics.aappublications.org/site/misc/reprints.xhtml

Information about ordering reprints can be found online:

rights reserved. Print ISSN: 0031-4005. Online ISSN: 1098-4275.Grove Village, Illinois, 60007. Copyright © 2007 by the American Academy of Pediatrics. Alland trademarked by the American Academy of Pediatrics, 141 Northwest Point Boulevard, Elk publication, it has been published continuously since 1948. PEDIATRICS is owned, published,PEDIATRICS is the official journal of the American Academy of Pediatrics. A monthly

by guest on July 11 2011pediatrics aappublications orgDownloaded from




http://pediatrics.aappublications.org/content/120/3/538.full.html#ref-list-1



http://pediatrics.aappublications.org/content/120/3/538.full.html#related-urls



http://pediatrics.aappublications.org/cgi/collection/blood



http://pediatrics.aappublications.org/site/misc/Permissions.xhtml



http://pediatrics.aappublications.org/site/misc/reprints.xhtml













hiero enteral teamprano vs tardio en infantes prematuros pediatrics

Documents