Effects of intrauterine growth retardation in premature infants on early childhood growth

Effects of intrauterine growth retardation in premature infants on early childhood growth Richard S. Strauss, MD, a n d William H. Dietz, MD, PhD From the Division of Pediatric Gastroenterologyand Nutrition, Floating Hospitalfor Children at New England Medical Center and Tufts University School of Medicine, Boston, Massachuseffs

Objective: To evaluate in preterm infants the role of intrauterine growth retardation and infant body proportionality on subsequent childhood growth. Methods: Preterm infants (818) prospectively enrolled in the Infant Health and Development Program were studied from birth to 36 months of corrected age. Weights and lengths were recorded at eight intervals. Growth parameters were compared in preterm infants with differing body symmetry at birth, defined by length-for-age and weight-for-age. Infants with both Iow length-for-age and Iow weight-for-age at birth were categorized as symmetrically growth retarded, and infants with normal length-for-age and Iow weight-for-age were categorized as asymmetrically growth retarded. Results: Infants born with Iow length-for-age demonstrated increased growth veIocity until 8 months of corrected age (p <0.001). However, infants born with Iow weight-for-age demonstrated decreased weight-gain velocity compared with preterm infants with appropriate weight for gestational age (AGA) unti140 weeks of corrected age (p <0.001). Heights and weights of infants with either symmetric or asymmetric intrauterine growth retardation remained significantly retarded compared with AGA preterm patients and the National ChUd Health Survey (NCHS) reference population (p <0.00 I). Infants born short but with normal weight did not significantly differ from AGA preterm infants in either weight or length at 36 months of corrected age. Preterm infants with both symmetric and asymmetric growth retardation demonstrated limited catch-up growth in weight untU age 4 months and then paralleled the AGA preterm patients and the NCHS reference population. Very Iow birth weight (-~ 1250 gm), gestational age, and a broadbased day-care intervention did not influence growth outcome after control for the presence of intrauterine growth retardation. Conclusions: Preterm infants with both symmetric and asymmetric intrauterine growth retardation demonstrate limited catch-up growth. Intrauterine growth deficits persist into early childhood. (J Pediatr 1997;130:95-102)

Supported by grants DK 07471 and DK 46200 from the National Institutes of Health. The data used in this manuscript were made available by the InterUniversity Consortium for Political and Social Science Research. The data were originally collected by the Infant Health and Development Program. Neither the original source nor collectors of the data bear any responsibility for the analyses or interpretations presented here.

Submitted for publication Dec. 27, 1995; accepted July 9, 1996. Reprint requests: Richard S. Strauss, MD, Division of Pediatric Gastroenterology and Nutrition, University of Medicine and Dentristry of New Jersey-Robert Wood Johnson Medical School, One Robert Wood Johnson Place, New Brunswick, NJ 08903. Copyright © 1997 by Mosby-Year Book Inc. 0022-3476/97/$5.00 + 0 9/21176362

95

96

Strauss and Dietz

The long-term growth potential of premature infants remains unclear. Although some studies showed normal childhood growth in preterm infants, 1-3 others showed significant weight and height deficits in early childhood.4-9 Growth of preterm infants bom in the early or mid-1970s 2-6 may have been impaired by poor perinatal nutritional support. Other smdies are difficult to interpret because of limited follow-up or small numbers 1' 7 or because they did not identify whether the infants were also intrauterine growth retarded.1' 3, 7-9 Preterm infants who are small for gestational age are at particular risk for poor long-term growth. 1°-~4However, the causes of intrauterine growth retardation vary, and it is likely that a simple classification based on birth weight or birth length may not sufficiently describe this population. Historically, growth-retarded infants have been divided into ' symmetric" (low weight and length) and "asymmetric" (low weight-for-length) groups. 15-18 Animal experiments suggest that symmetric growth retardation indicates stressAGA CCA HAZ IHDP ~GR NCHS SGA TPN WAZ

Appropriate in size for gestational age Corrected for chronologic age Height-for-age z score Infant Health and Development Program Intrauterine growth retardation National Center for Health Statistics Small for gestational age total parenteral nutrition Weight-for-age z score

ful events that began in early pregnancy. Other studies suggested that symmetrically growth-retarded infants most likely have limited growth potential. 19'2° Third-trimester placental insufficiency, like that accompanying preeclampsia, predisposes infants to asymmetric growth retardation 2~-23 who may be more likely to recover from their growth deficits. Previous investigators showed that term infants with asymmetric IUGR are more likely to demonstrate increased growth velocity in the neonatal period, whereas term infants with symmetric 1UGR are likely to remain small.24-26 Because large numbers of growth-retarded preterIn infants can rarely be studied in a single institution, the long-term growth outcome of preterm infants with different types of IUGR has remained unclear. To explore this issue, we analyzed the data from the Infant Health and Development Program, an eight-center collaborative study of the growth and development of low birth weight, preterm infants. Previous studies using the IHDP data s' 27 reported poorer growth in preterm infants with lower birth weight, but they did not evaluate the effects of IUGR on subsequent weight and length.

The Journal of Pediatrics Bnuary 1997

METHODS Sample. The IHDP cohort consisted of preterm infants who weighed less than 2500 gm and who were bom between January and October 1985 after fewer than 37 weeks' gestation. Approximately one third of patients were randomly assigned to an intervention group, which consisted of home visits, parental education and support, and a high-quality educaüonal day-care program. The other two thirds received high-quality follow-up only. Other details of the cohort population are available elsewhere. 8, 28, 29 In brief, gestational age was measured according to the physical cfitefia in the Ballard assessment pefformed within 48 hours of birth. Obstetric data were obtained by review of the infant's chart. Prepregnancy weight was determined by matemal recall. Infants were excluded if they died within the first 48 hours; received oxygen for more than 90 days; were hospitalized longer than 60 days after 40 weeks of corrected gestational age; had a severe neurologie abnormality, neural robe defect, severe sensory defect, or chromosome multiple anomaly syndrome; or were a sibling of an ineligible patient. Patients also were exeluded for maternal ding or alcohol abuse, previous psychiatrie hospitalization, or inability to commanicate adequately in English. We also excluded patients with a gestational age greater than 35 weeks (n = 183), patients with evidence of cerebral palsy at either 30 or 36 months' corrected chronologic age (n = 52), and patients without both length and weight measurements at birth (n = 35). After all exclusions, 818 patients were analyzed. Measures. All infants were initially measured by trained study staff or staff clinicians at the time of either the Ballard assessment or admission to the nursery. Birth weight was measured to the nearest 10 gm, and birth length was measured with standard measufing boards. 2s Infants were assessed at 40 weeks' postconceptional age, and at 4, 8, 12, 18, 24, 30, and 36 months' CCA. Infants were weighed undressed on a calibrated infant balance seale, and length was measured in the supine position on a standardized infant measuring board. Ninety-seven percent of patients had weight and length determJned at 40 weeks CCA, 93% at 4 months CCA, 92% at 8 months CCA, 91% at 12 months CCA, 89% at 18 months CCA, 88% at 24 months CCA, 87% at 30 months CCA, and 88% at 36 months CCA. Individual growth veloeities for weight and length were calculated for growth between eight age intervals (birth to 40 weeks CCA and birth to 4, 4 to 8, 8 to 12, 12 to 18, 18 to 24, 24 to 30, and 30 to 36 months of eorreeted age) by dividing the difference between suecessive weights (of lengths) by the time between measurements. Outcomes. Patients were classified as SGA if they were at less than or equal~to the 10th percentile for gestational age by weight or as "short" if they were at less than or equal to

The Joumal of Pediatrics Volume 130, Number 1

Strauss and Dietz

WEIGHT

L E N G T

I-I

_< lOth Percentile

_< lOth Percemile

> lOth Percentile

Symmetric IUGR (8%)

Short for Age (6%)

i200 / 1801160~140~120 I100B

ß

97

Weight(gms/week)

60 40

> 10th Percentile

Asymmetric IUGR

Appropriate for Age (82%)

(4%)

Length(cms/week) 0

1.2 0.9

Fig,, |. Classification of patients with IUGR.

0.6 0.3

the 10th percentile by length, by the criteria set forth by Lubchenco.30, 31 Three IUGR groups and one control group were further defined (Fig. 1): (1) appropriate for gestational age: normal weight and length; (2) symmetric IUGR: short and SGA; (3) short for age: short with normal weight for age; and (4) asymmetric IUGR: SGA and normal length for age. A similar classification of IUGR has been used by Karlberg and Albertsson-Wikland32 in term infants. SGA infants were additionally classified as either in the low or the normal ponderal index by the criteria set foith by Lubchenko. 3° The z scores, which indicate the number of standard deviations by which an infant's length and weight deviate from a reference population, were calculated by using the Centers for Disease Control and Prevention anthropometry software based on the Fels Research Institute (Wright Stare University, Yellowsprings, Ohio) longitudinal smdy and the National Health and Nutrition Examination surveys 33, 34 for all growth points collected. All patients with measurements performed at the 40-week postconception visit had z scores calculated on the basis of an average age of 0 weeks. All other z scores were calculated at the CCA. We preferred the use of z scores to individualized weights and lengths because z scores allow comparisons across genders and minimize sample variation if all measurements are not recorded at identical intervals. Statisücs. Data were analyzed by using the SPSS-X program (SPSS Inc., Chicago, Ill.). Differences in proportions were compared by corrected chi-square tests. Differences in continuous variables were determined by analysis of variance with least significant differences for multiple corrections. Data that were not normally distributed (e.g., number of days of ventilator use, number of days of total parenteral nutrition) were analyzed by using the Mann-Whitney U test

o

, I

o

I

~

I

8

I

I

~:

[

9 12 1; 1; ;1 2', 2~ 3'o 3; MONTHS

Fig. 2. Mean rate of weight gain _+ SEM (top panel) and length gain _+ SEM (bottom panel) from birth to 36 months of corrected age by 1UGR subtype: (O- -©) AGA preterm; (A - -A) syrranetric IUGR; (= --) short-for-age; and (A A) asymmetric 1UGR. Mean rate of weight gain was significantly decreased in symmetric and asymmetfic IUGR patients compared with AGA and short-for-age patients from birth until 40 weeks CCA (p <0.005). Mean rate of length gain was increased for patients with low birth length until 8 months CCA (p <0.005). with a Bonferroni correction for multiple comparisons. The Cochran C test for the homogeneity of variance was used to compare variance among the institutions. RESULTS

Characteristics of mothers. More than 95% of mothers had at least some prenatal care, and more than 60% were high school graduates. There was a slight predominance of black patients in the asymmetric IUGR group (p <0.01). Matemal prepregnancy weight and age did not differ significantly between any of the four infant groups (see Table I). No significant differences were found in matemal education, heavy matemal smoking (->20 cigarettes per day), lack of prenatal care, matemal diabetes, or placenta previa/abruptio placentae between any of the four infant groups (data not shown). The prevalence of preeclampsia/eclampsia in the patients with symmetric and asymmetric IUGR was 1.5 to 2 times the rate in A G A preterm patients, although this achieved statistical significance for only the symmetrically growth-retarded group (p <0.01). Characteristics of newborn infants. Gestational age or

98

Strauss and Dietz

The Journal of Pediatrics January 1997

LU 0

0.5

100

o

~ ._... 80

LU (9 -0.5

~~00

~-1.0

~~.

(a ù,, -1.5 1~ -2.0

20 I

I

I

I

I

I

I

I

LU O

40

0

4

8

12 18 24 MONTHS

30

36

Fig. 4. Percentage ofpatients with weight-for-age at the 10th percentile or less: (O- -O) AGA preterm; (& - -lk) symmetric IUGR; (« =) short-for-age; and (A A) asymmetric IUGR. The percentage of symmetric and asymmetric children with weight-forage at the 10th percentile or less decreased from 100% to 30% until 4 months of corrected chronologic age and then remained relatively unchanged.

0.5

tU -0.5

B-l.0 t~ -1.5

ù,~"

~

-2.0 [

I

0

5

10

I

I

I

I

I

15

20

25

30

35

MONTHS

Fig. 3. Mean weight-for-age and height-for-age z score _+SEM from 40 weeks of corrected age to 36 months of corrected age by IUGR type: (O- -O) AGA preterm; (A - -Ik) symmetric 1UGR; (• =) short-for-age; and (A A) asymmetric IUGR. Mean WAZ was significandy decreased in symmetric and asymmetric IUGR patients compared with AGA patients at all time points (p <0.001). Mean HAZ was significantly decreäsed in symmetric IUGR patients (p < 0.001) and asymmetric IUGR patients (p <0.05) compared with AGA patients at all time points. gender did not differ significantly between any of the groups. There were no significant differences in the mean or variance of birth length among the eight institutions. As expected from the selection criteria, birth weight was comparable in the patients with symmetric and asymmetric IUGR, and both groups had significantly lower birth weight than the AGA (p <0.001) and SGA patients (p <0.001). Birth length was similar in the symmetric IUGR patients and SGA patients, and both were significantly shorter than the patients with asymmetric IUGR (p <0.01). All three IUGR groups were significantly shorter than the AGA patients (p <0.001). Patients with symmetric and asymmetric IUGR received TPN for a significantly longer period than did AGA preterm patients (Table I; p <0.005). Patients who were asymmetric at birth also underwent mechanical ventilation significantly longer than AGA preterm patients (p <0.005). Although statistically significant, these differences were minimal. The lungs of 80% of patients with symmetric and asymmetric

IUGR were mechanically ventilated for less than 1 week, and 80% received TPN for less than 2 weeks. There was no significant difference in the prevalence of moderate or severe respiratory distress syndrome between any of the IUGR groups. The prevalence of low 5-minute Apgar scores was Significantly increased only in the infants with asymmetric IUGR (p <0.01). Growth velocity. Between birth and 40 weeks CCA, both groups of infants with low birth weight had significantly decreased rate of weight gain compared with the AGA preterm infants (p <0.005; Fig. 2). After 40 weeks CCA, rates of weight gain did not differ significantly between groups. Linear growth velocity was significantly greater in the SGA patients (p <0.005) than in the AGA preterm group from birth to age 4 months. Linear growth velocity in proportionally small patients was significantly greater than in AGA preterm infants from 40 weeks CCA to 8 months CCA (p <0.005). As a group, patients who were short at birth showed increased rates of length gain from birth until age 8 months (p <0.005). After age 8 months, there were no significant differences in length-gain velocities between groups. Weight and length growth. Patients with symmetric and asymmetric IUGR demonstrated limited catch-up growth. Weights and lengths of both groups remained significantly lower than those of the AGA preterm patients (p <0.001) at each interval from 40 weeks CCA to 36 months CCA (Fig. 3). Weights and lengths of both groups remained between -0.8 and 1.2 SD below the National Center for Health Statistics reference population from 4 months CCA to 36 months CCA. In patients with normal birth weight for gestational age (AGA and SGA patients), weight-for-age z scores remained between -0.2 and -0.6 SD below those of

The Journal of Pediatrics Volume 130, Number 1

Strauss and Dietz

99

T a b l e I. Maternal and infant characteristics

IUGR A G A preterm (n = 675)

Short-for-age (n = 46)

Symmetric (n = 65)

Asymmetric (n = 32)

24.7 _+ 0,2 51 59.1 -+ 0.5 50 32.5 _+ 0.1 1835 - 16 43.0 _+ 0.1 2. 9 2. 1 27 12

25.4 _+ 1.0 63 59.5 _+ 1.8 50 32.9 ± 0.3 1694 ± 52 38.3 -+ 0.5* 3. 3 3. 0 30 17

23.8 _+ 0.7 62 59.5 _+ 1.8 45 32.5 ± 0.3 1203 _+ 39* 37.4 +_ 0.5* 8.4* 4. 2 31 15

25.8 __+1.1 75 65.0 _+ 3.6 41 32.0 ___0.4 1252 ± 61" 40.0 __+0.6* 5.9* 5.7* 41 28

Matemal age Matemal race (% black) Matemal prepregnancy weight (kg) Sex (% male) Gestational age (wk) Birth weight (gm) Birth length (cm) Mean days of TPN Mean days of mechanical ventilation % with RDS (moderate/severe) % with 5-minute Apgar <7

Values are expressed as mean ± SEM. *p <0.005 compared with AGA preterm patients (see text for details).

T a b l e II. Weight-for-age and height-for-age z-scores at 36 months

Birth weight (gm)

WAZ (36 mo) AGA preterm SGA preterm

HAZ (36 mo) AGA preterm SGA preterm

Gestational age (wk)

Interventional group

~1250

>1250

-<30

>30

Intervention

No intervention

ù0.40 _+ 0.13 (n = 60) -0.93 ± 0.15 (n = 48) p <0.008

-0.31 + 0.04 (n = 575) -0.94 _+0.18 (n = 35) p <0.001

ù0.34 -+ 0.08 (n = 125) -0.94 + 0.20 (n = 23) p <0.006

-0.32 --- 0.05 (n = 510) -0.94 + 0.14 (n = 60) p <0.001

-0.34 -+ 0.06 (n = 268) -0.63 _ 0.19 (n = 25) p <0.01

-0.31 --- 0.05 (n = 367) -1.07 ___0.14 (n = 58) p <0.001

0.21 _+ 0.12 (n = 60) -0.26 _+0.16 (n = 48) p <0.02

0.24 _+ 0.04 (n = 575) -0.34 _+ 0.17 (n = 35) p <0.001

0.36 ± 0.09 (n = 125) -0.21 _+ 0.19 (n = 23) p <0.02

0.20 ± 0.05 (n = 510) -0.33 ± 0.14 (n : 60) p <0.001

0.21 4- 0.07 (n = 268) -0.02 + 0.20 (n : 25) p <0.05

0.25 ± 0.05 (n = 367) -0.41 ± 0.t4 (n = 58) p <0.001

Values expressed as mean _+SEM.

the NCHS reference population between ages 8 and 36 months. Length-for-age z score was significantly lower in the patients with symmetric I U G R (p <0.001) and the patients with asymmetric I U G R (p <0.05) at each time point from birth to 36 months C C A compared with the A G A patients. Although S G A patients had lower H A Z compared with A G A preterm patients from 8 months C C A to 36 months CCA, the difference was not significant (Fig. 3). After 18 months, the H A Z s may overestimate differences between study patients and the NCHS reference population because heights of children in the NCHS study were measured, whereas length continued to be measured in this study until 36 months. Initially all patients with asymmetric and symmetric I U G R had weights below the 10th percentile; by age 4 months, 30% of preterm patients with I U G R remained be-

low the 10th percentile (Fig. 4). F r o m ages 4 to 36 months, there were no significant changes in the percentage of patients with weights below the 10th percentile. Ponderal index as a m e a s u r e of b o d y s y m m e t r y . S G A patients with low ponderal index were longer at birth than S G A patients with normal ponderal index (40.1 vs 37.4 cm; p <0.001). However, both groups did not have significanfly different weights at birth (1.23 vs 1.22 kg; p = 0.89). F r o m 40 weeks C C A to 36 months CCA, there were no significant differences in weight and length between infants b o m with low ponderal index and infants horn with normal ponderal index (data not shown).

Effects of birth weight, gestational age, and interventional status. S G A infants remained significantly lighter and shorter at 36 months than did A G A infants, even after stratification by birth weight, gestational age, or interventional

10 0

Strauss and Dietz

status (Table II), although the sample size for the intervention offered limited power, particularly for the SGA infants. Furthermore, at 36 months in patients with birth weights less than 1250 gm, weights and lengths did not differ significantly from those of the heavier newbom infants, in both the SGA group (weight-for-age z score:p = 0.96; HAZ:p = 0.75) and the AGA group (WAZ: p = 0.37; HAZ: p = 0.83). Similarly, the weights and lengths of those infants born before 30 weeks' gestation did not significantly differ from those bom afterward, in both the SGA group (WAZ: p = 0.99; HAZ: p = 0.60) and the AGA group (WAZ: p = 0.62; HAZ: p = 0.18). Finally, the Weights and lengths of infants receiving the broad-based day-care intervention did not significantly differ from those who did not receive the intervention in either the SGA group (WAZ: p = 0.08; HAZ: p = 0.12) or the AGA group (WAZ: p = 0.78; HAZ: p = 0.94). DISCUSSION From 40 weeks of corrected gestational age to 36 months, infants with weights at less than the 10th percentile (SGA infants) remained significantly growth retarded compared with the AGA preterm infants and the NCHS reference population, regardless of body symmetry at birth. The mean WAZ of low-birth-weight infants at 36 months was 0.93, which corresponds to the 17th percentile on the NCHS growth charts. Birth length did not predict weight or length at age 36 months in preterm infants. By 36 months of corrected age, weight in AGA preterm infants and the SGA infants was minimally decreased compared with that of the NCHS reference population. The mean WAZ in this group was -0.32, which corresponds to the 38th percentile of the NCHS reference population. No significant difference in growth occurred between preterm infants with symmetric and asymmetric growth retardation as measured by either weight-for-length or ponderal index. Several possibilities could explain the lack of difference. Length in preterm infants is subject to error, and some infants may have been misclassified because of measurement errors. However, SGA infants were 4.7 cm shorter than AGA infants, and infants with symmetric IUGR were 2.6 cm shorter than infants with asymmetric IUGR. A1though a few of the infants may have been misclassified, it is unlikely that measurement error would contribute to the large differences observed between groups. The fact that there was no difference in the variability of length measured at the eight institutions implies also that there were no grave systematic errors in length measurement. It is therefore possible that weight and length may be regulated differently in utero. Factors that predispose infants to decreased length may be reversible, whereas factors affecting their weight may not. Our observation that infants who were short at birth had an increased linear growth velocity from birth to 8

The Journal of Pediatrics Bnuary 1997

months CCA supports the second hypothesis, becanse we would not expect to see persistent differences in either length or linear growth velocity if the classification scheine were not accurate. The IHDP cohort represents a sample of relatively healthy preterm infants, because it included no infants with congenital abnormalities, severe neurologic abnormalities, oxygen dependence more than 3 months, or extended hospitalization. Out analysis excluded patients with evidence of cerebral palsy at 30 or 36 months because growth in these patients may differ significantly from that of other preterm infants. 35,36 Because data were collected only on infants weighing less than 2500 gm at birth, the birth weights of infants with gestational ages greater than 35 weeks were highly skewed toward the lower weight percentiles, making interpretation of the results difficult. These exclusion criteria eliminated some of the confounding factors that may have contributed to poor growth in preterm infants with IUGR, observed in other studies. 1H3 However, even in this relatively healthy preterm population, significant long-term growth deficits remained in preterm infants with IUGR. Casey et al.8, 27 used the IHDP data to present growth charts for preterm infants stratified by birth weight. Their data, as well as data from Georgieff et al., 7 suggested that very low birth weight infants (birth weight <1250 to 1500 gm) had lower growth potential than larger preterm infants. However, the major limitation of both of these studies is that they failed to adjust for the high incidence of IUGR in the very low birth weight infants. Our data suggest that after correction for IUGR status, no added growth deficit is cansed by very low birth weight. Our data indicate also that gestational age and interventional status did not independently predict long-term childhood growth. At best, a comprehensive interventional program improves but does not eliminate the growth deficit. "'Catch-up" growth is used to describe "the acceleration in growth that occurs when a period of growth retardation ends and favorable conditions are restored. ''37 Catch-up growth has been well described after treatment of hypothyroidism, growth hormone deficiency, celiac disease, Crohn disease, and chronic malnutrition (reviewed by Ashworth and Millward 37 and Tanner) 8 In the IHDP cohort, patients with low birth weight for gestational age initially had a decreased rate of weight gain compared with that of AGA preterm patients. This observation contradicts the observations in term patients with IUGR who demonstrated increased growth velocity from birth to age 1 to 3 months, as-z5 However, in these studies, catch-up growth was described only in asymmetrically small infants. Our results demonstrated limited catch-up growth in infants with both symmetric and asymmetric IUGR. Although average weight-gain velocity was not increased

The Jouma! of Pediatrics Volume 130, Number 1

in the infants with symmetric and asymmetric IUGR, they nevertheless crossed percentiles upward by maintaining proportionately higher growth velocity in comparison to their weight. Limited catch-up growth is similarly observed in studies of older children recovering from growth stunting resulting from chronic malnutrition. Such children often remain with permanent stunting eren after supplementation with adequate calories or ad lib feeding (reviewed by Martorell et al. 39 and Golden. 4° However, unlike older malnourished children, low weight and not short length is most predictive for poor long-term growth in premature SGA infants. The absence of a powerful growth effect of a broad-based interventional program that included one or two meals a day further supports the hypothesis that pregnancy is a critical period for the development of long-term growth potential. Other data supporting the existence of this critical period are increasingly common. Discordant monozygous twin pairs followed longitudinally demonstrate that the smaller twin will usually remain significantly lighter and shorter than the larger cotwin, despite having idenfical genetic backgrounds. 41-43 In addition, exposure of the fetus to either ovemutrition44, 45 or undernutrition46 as early as the first trimester of pregnancy can result in altered growth during childhood. The mechanisms that account for the effects of the intrauterine environment on growth remain unclear. We gratefully acknowledge the assistance of Rahel Berhane and Richard J. Grand for their reviews of early drafts of the manuscript. We also acknowledge the physicians and nurses involved in the Infalt Health and Development Program. REFERENCES

1. Altigani M, Murphy JF, Newcombe RG, Gray OP. Catch-up growth in preterm infants. Acta Paediatr Scand 1989;357(Suppl):3-19. 2. Brandt I. Growth dynamics of low-b~h-weight infants. Acta Paediatr Scand 1985;319(Suppl):38-47. 3. Beck GJ, Van den Berg BJ. The relationship of the rate of intrauterine growth of low-birth-weight infants to later growth. J Pediatr 1985;86:504-11. 4. Cruise MO. A longitudinal study of the growth of low birth weight infants: velocity and distance growth, birth to 3 years. Pediatrics 1973;51:620-8. 5. Kimble KJ, Ariagno RL, Stevenson DK, Sunshine P. Growth to age 3 years among very-low-birth-weight sequelae-free survivors of modern neonatal intensive care. J Pediatr 1980; 100:622-4. O. Babson SG. Growth of low-birth-weight infants. J Pediatr 1970;77:11-8. 7. Georgieff MK, MilIs MM, Zempel CE, Chang PN. Catch-up growth, muscle and fat accretion, and body symmetry of infants one year after newbom intensive care. J Pediatr 1989; 114:28892. 8. Casey PH, Kraemer HC, Bembaum J, Yogman MW, Seils JC. Growth status and growth rates of a varied sample of low birth

Strauss and Dieß

9. 10, 11.

12.

13.

14.

15.

16.

17. 18. 19.

20. 21. 22. 23.

24.

25.

26.

27.

28.

10 1

weight preterm infants: a longitudinalcohort from birth to three years of age. J Pediatr 1991;119:599-605. Kitchen WH, Ford GW, Doyle LS. Growth and very low birth weight. Arch Dis Child 1989;64:379-82. Vohr BR, Oh W. Growth and development in preterm infants small for gestational age. J Pediatr 1983;103:941-5. Sung IK, Vohr B, Oh W. Growth and neurodevelopmental outcome of very low birth weight infants with intrauterine growth retardation: comparison with control subjects matched by birth weight and ges tational age. J Pediatr 1993;123:618-24. Morley R, Brooke OG, Cole TJ, Powell R, Lucas A. Birthweight ratio and outcome in preterm infants. Arch Dis Child 1989;65:30-4. Hack M, Merkatz IR, McGrath SK, Jones PK, Fanaroff AA. Catch-up growth in very-low-birth-weight infants. Am J Dis Child 1984;138:370-5. Pena IC, Teberg AJ, Finello KM. The premature small-forgestational-age infant during the first year of fife: comparison by birth weight and gestational age. J Pediatr 1988; 113:106673. Rosso P, Winick M. Intrauterine growth retardation: a new systemic approach based on the clinical and biochemical characteristics of this condition. J Perinat Med 1974;2:147-60. Villar J, Belizan JM. The timing factor in the pathophysiology of the intrauterine growth retardation syndrome. Obstet Gynecol Surv 1982;37:499-506. MillerHC. Intrauterinegrowthretardation: anunmetchallenge. Am J Dis Child 1981;135:944-8. Rudolph AJ. Failure to thrive in the perinatal period. Acta Paediatr Scand 1985;319 Suppl:55-61. Widdowson EM, McCance RA. The effects of finite periods of undemutrition at different ages on the composition and subsequent development of the rat. Proc R Soc Lond (Biol) 1963; 158:329-42. Winick M, Noble A. Cellular response in rats during malnutrition at various ages. J Nutr 1966;89:300-6. Naeye RL. Abnormalities in infants of mothers with toxemia of pregnancy. Am J Obstet Gynecol 1966;95:276-83. Naeye RL. Malnutrition: probable cause of fetal growth retardation. Arch Pathol 1965;79:284-91. Ounsted M, Moar VA, Scott A. Proportionality of small-forgestafional age babies at birth: pefinatal associations and postnatal sequelae. Early Hure Dev 1986;14:77-88. Davies DP, Platts P, Pritchard JM, Wilkinson PW. Nutritional status of light-for-dates infants at birth and its influence on early postnatal growth. Arch Dis Child 1979;54:703-6. Holmes GI, Miller HC, Hassanein K, Lanskey SB, Goggin JE. Postnatal somatic growth in infants with atypical fetal growth patterns. Am J Dis Child 1977;131:1078-83. Villar J, Smeriglio V, Martorell R, Brown CH, Klein RE. Heterogeneous growth and mental development of intrauterine growth-retarded infants during the first 3 years of life. Pediatrics 1984;74:783-91. Casey PH, Kraemer HC, Bembaum J, Tyson JE, Sells JC, Yogman M, et al. Growth pattems oflow birth weight preterm infants: a longitudinal analysis of a large, varied sample. J Pediatr 1990;117:298-307. Gross RT, Director. Infant Health and Development Program (IHDP): enhancing the outcomes of low birth weight, premarare infants in the United States, 1985-88. Ann Arbor (MI): Inter-university Consortium for Polifical and Social Research, 1992.

102

Strauss and Dietz

The Journal of Pediatrics January 1997

29. Infants Health and Development Program. Enhancing the outcomes of low-birth-weight, premature infants. JAMA 1990; 263:3035-42. 30. Lubchenco LO, Hansman C, Dressler M, Boyd E. Intrauterine growth as estimated from livebom birth-weight data at 24 to 42 weeks of gestation. Pediatrics 1963;32:793-800. 31. Lubchenco LO, Hansman C, Boyd E. Intranterine growth in length and head circumference as estimated from live births at gestational ages from 26 to 42 weeks. Pediatrics 1966;37: 403-8. 32. Karlberg J, Albertsson-Wildand K. Growth in full-term smallfor-gestational-age infants: from birth to final height. Pediatr Res 1995;38:733-9. 33. Dibley MJ, Goldsly JB, Staehling NW, Trowbridge FL. Development of normalized curves for the international growth reference: historical and technical considerations. Am J Clin Nutr 1987;46:736-48. 34. Dibley MJ, Staehling W, Nieburg P, Trowbridge H. Interpretation of Z-score anthropometric indicators from the international growth reference. Am J Clin Nutr 1987;46:746-62. 35. Stallings VA, Charney EB, Davies JC, Cronk CE. Nutritional status and growth of children with diplegic or hemiplegic cerebral palsy. Dev Med Child Neurol 1993;35:997-1006. 36. Thommessen M, Heiberg A, Kase BF, Larsen S, Riis G. Feeding problems, height and weight in different groups of disabled children. Acta Paediatr Scand 1991;80:527-33. 37. Ashworth A, Millward DJ. Catch-up growth in children. Nulr Rer 1986;44:157-63.

38. Tanner JM. Growth as a target-seeking function: catch-up and catch-down growth in man. In: Falkner F, Tanner JM, editors. Human growth: a comprehensive treatise; vol 1.2nd ed. New York: Plenum Press, 1986:167-79. 39. Martorell R, Khan LK, Schroeder DG. Reversibility of stunting: epidemiological findings in children from developing countries. Eur J Clin Nutr 1994;48 Suppl 1:$45-57. 40. Golden MHN. Is complete catch-up possible for stunted malnourished children? Eur J Clin Nutr 1994;48 Suppl 1:$5871. 41. Babson SG, Phillips DS. Growth and development in twins dissimilar in size at birth. N Engl J Med 1973;289:937-40. 42. Wilson RS. Growth and development of human twins. In: Falkner F, Tanner JM, editors. Human growth: a comprehensive treatise; vol 1.2nd ed. New York: Plenum Press, 1986: 197-211. 43. Allison DB, Panltre F, Heymsfield SB, Pi-Sunyer FX. Is the intra-uterine period really a critical period for the development of adiposity? Int J Obes 1995;19:397-402. 44. Pettit DJ, Baird HR, Aleck KA, Bennett PA, Knowler WC. Excessive obesity in offspring of Pima Indian women with diabetes during pregnancy. N Engl J Med 1983;308:242-5. 45. Vohr BR, Lipsitt LP, Oh W. Somatic growth of children of diabetic mothers with reference to birth size. J Pediatr 1980; 97:196-9. 46. Ravell GP, Stin ZP, Suffer MW. Incidence of obesity in young men after famine exposure in utero and early infancy. N Engl J Med 1976;295:349-53.

BOUND VOLUMES AVAILABLE TO SUBSCRIBERS Bound volumes of the 1997 issues of The Journal of Pediatrics are available to subscribers (only) from the Publisher, at a cost of $83.00 for domestic, $105.93 for Canadian, and $99.00 for international subscribers, for Vol. 130 (January-June) and Vol. 131 (July-December), shipping charges included. Each bound volume contains subject and author indexes, and all advertising is removed. Copies are shipped within 60 days after publication of the last issue in the volume. The binding is durable buckram, with the Journal name, volume number, and year stamped in gold on the spine. Payment taust accompany all orders. Contact M o s b y - Y e a r Book, Inc., Subscription Services, 11830 Westline Industrial Dr., St. Louis, M O 63146-3318, USA/800-453-4351, or 314-4534351. Subscriptions taust be in force to qualify. Bound volumes are not available in place of a regular Journal subscription.