Published online June 1, 2006
PEDIATRICS Vol. 117 No. 6 June 2006, pp. 1930-1938 (doi:10.1542/peds.2005-1926)
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow P3Rs: Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when P3Rs are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow E-mail this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My File Cabinet
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via CrossRef
Right arrow Citing Articles via ISI Web of Science (6)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Löfqvist, C.
Right arrow Articles by Hellström, A.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Löfqvist, C.
Right arrow Articles by Hellström, A.
Related Collections
Right arrow Premature & Newborn

Postnatal Head Growth Deficit Among Premature Infants Parallels Retinopathy of Prematurity and Insulin-like Growth Factor-1 Deficit

Chatarina Löfqvist, PhDa,b, Eva Engström, MD, PhDa, Jon Sigurdsson, MDa, Anna-Lena Hård, MD, PhDc, Aimon Niklasson, MD, PhDa, Uwe Ewald, MDd, Gerd Holmström, MDe, Lois E. H. Smith, MD, PhDb and Ann Hellström, MD, PhDa,c

a Göteborg Pediatric Growth Research Center, Institute for Clinical Sciences, Department of Pediatrics, Sahlgrenska Academy of Göteborg University, Göteborg, Sweden
b Department of Ophthalmology, Children's Hospital, Harvard Medical School, Boston, Massachusetts
c Department of Clinical Neurosciences, Section of Ophthalmology, Sahlgrenska Academy at Göteborg University, Göteborg, Sweden
d Departments of Women's and Children's Health
e Ophthalmology, Uppsala University, Uppsala, Sweden


   ABSTRACT
TOP
 ABSTRACT
METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
BACKGROUND. We hypothesized that in premature infants, retinal vascular growth retardation between birth and postmenstrual age of ~30 to 32 weeks that initiates retinopathy of prematurity is paralleled by brain growth retardation.

METHODS. In a prospective longitudinal study, we measured postnatal head growth, retinopathy of prematurity stage, protein and energy intake, severity of illness and serum insulin-like growth factor-1 levels in 58 preterm infants (mean gestational age at birth: 27.6 weeks) from birth until postmenstrual age of ~40 weeks.

RESULTS. Premature infant head growth decelerates dramatically after birth until postmenstrual age of ~30 weeks. Head growth retardation coincides with retinal vascular growth suppression. Accelerated growth follows between post menstrual ages of ~30 to 32 weeks and ~40 weeks. The degree of head growth retardation up to postmenstrual age of 31 weeks corresponds to the degree of retinopathy of prematurity and to the degree of suppression of serum levels of insulin-like growth factor-1. At postmenstrual age of 31 weeks, if a child’s head circumference SD is below –2.5, then the probability of also developing at least stage 3 retinopathy of prematurity increases fivefold compared with head circumference above –2.5 SD (32% vs 6%) suggesting parallel processes in brain and retina. Serum insulin-like growth factor-1 levels correlate positively with head circumference SD score and with the degree of retinopathy of prematurity.

CONCLUSIONS. The correlation between head and retinal growth is consistent with insulin growth factor-1 being one of the postnatal growth factors involved in this multifactorial process and also suggests that factors that contribute to retinopathy of prematurity during this critical period may also affect neurological dysfunction. Additional studies are required to establish this connection.

Key Words: retinopathy • prematurity • head growth

Abbreviations: IGF-1—insulin-like growth factor-1 • ROP—retinopathy of prematurity • PMA—postmenstrual age • CNS—central nervous system • GA—gestational age • HC—head circumference • IVH—intraventricular hemorrhage • NEC—necrotizing enterocolitis • BPD—bronchopulmonary dysplasia • BW—birth weight • SDS—SD score

The majority of very low birth weight infants have some neurologic impairment measurable by 6 to 8 years of age.1–3 Often they also develop retinopathy of prematurity (ROP), a disease that is initiated by retinal vascular growth retardation.4 The degree of early retinal growth delay is correlated with the severity of ROP.4–7 The severity of ROP among premature infants is also associated with the severity of neurologic impairments,8,9 which suggests a possible common mechanism of early growth abnormalities of the central nervous system (CNS) during a vulnerable period. Our understanding of ROP may help to elucidate brain dysfunction among preterm infants.

ROP is a disease of vasoproliferation that is a response to retinal dysfunction and hypoxia resulting from lack of blood vessels. After preterm birth, retinal vascular and neural development slows or ceases.4–6,10

The degree of vascular growth retardation before the onset of vasoproliferation at postmenstrual age (PMA) of ~30 weeks is correlated with the degree of ROP,7 a disease of destructive rapid vasoproliferation that leads to retinal detachment and blindness. We hypothesized that a similar postnatal growth pattern of inhibition followed by proliferation would be found in head circumference (HC) measurements, reflecting brain growth changes,11–13 determined at least in part by postnatal growth factors. In particular, postnatal CNS growth retardation among premature infants might correspond to low postnatal insulin-like growth factor-1 (IGF-1) levels.14

IGF-1 has a pleotrophic effect on neurons, supporting survival, neurite outgrowth, and myelination.14 In transgenic mice, lack of IGF-1 retards brain growth and overexpression of IGF-1 causes increased brain growth and maturation.15–17 In earlier studies, we found an association between low postnatal serum IGF-1 levels and the ultimate degree of ROP.18–20

We posed the following questions. Does inhibition of retinal vascular growth correspond to inhibition of brain growth? Does the onset of vasoproliferative ROP at PMA of ~30 to 32 weeks correspond to the onset of increased brain growth, suggesting a possible common stimulus for retina and brain? Do changes in serum IGF-1 levels correspond to changes in retinal and brain growth? To study these questions, we examined the associations between early postnatal brain growth (as reflected in HC changes),11,12,21 the degree of ROP, and the degree of early postnatal IGF-1 deficiency. We also examined caloric and protein intakes before the onset of ROP development at PMA of ~30 to 32 weeks in this clinical study.


   METHODS
TOP
 ABSTRACT
 METHODS
RESULTS
 DISCUSSION
 REFERENCES
 
Characteristics and Selection of Study Group
Consecutive infants born at a gestational age (GA) of ≤31.3 weeks at the Queen Silvia Children's Hospital (Göteborg, Sweden) between December 1999 and April 2002 or at the Uppsala University Hospital (Uppsala, Sweden) between February 2001 and April 2002 were recruited for the study. Exclusion criteria were inability to complete postnatal clinical follow-up evaluations until PMA of 36 to 40 weeks and any conspicuous congenital anomaly or hydrocephalus. Informed consent for participation was obtained from the parents of 70 children at the Queen Silvia Children's Hospital and from the parents of 14 infants at the Uppsala University Hospital. A total of 21 infants were excluded because the time of HC measurements and the time of blood sampling for IGF-I measurements exceeded 3 days during the first weeks. Three infants developed hydrocephalus and were excluded. Two infants were moved to another hospital during data collection, leaving a total of 58 study subjects (35 girls and 23 boys). The median GA at birth (based on results of fetal ultrasonography performed at PMA of 16–18 weeks) was 27.6 weeks (range: 23.6–31.3 weeks). The median birth weight (BW) was 935 g (range: 530–2015 g). The median BW expressed in SD score (SDS) [SDS = (actual value – mean value)/SD for age and gender] was –1.57 (range: –5.0 to 1.9 SDS).22 Gender, BW, and morbidity (presence of necrotizing enterocolitis [NEC], bronchopulmonary dysplasia [BPD], or intraventricular hemorrhage [IVH]) were recorded (Table 1). All infants were included in an earlier study of the role of IGF-1 in ROP.18,20 The group included 14 twins (median GA: 28.9 weeks; median BW: 1078 g). The Committee for Ethics at the Sahlgrenska Academy at Göteborg University approved the study (protocol Ö 594-00).


View this table:
[in this window]
[in a new window]
  		TABLE 1 Neonatal and Morbidity Variables for 58 Preterm Children

 
ROP Examination
The infants were examined according to a routine protocol that consisted of dilated eye fundus examinations once or twice each week, depending on the severity of the disease, from the chronologic age of 5 to 6 weeks until the retinas were fully vascularized or until the condition was considered stable. After pupillary dilation with 0.2% cyclopentolate and 1% phenylephrine, the retinas were examined through indirect ophthalmoscopy by a trained pediatric ophthalmologist, who had no knowledge of IGF-1 status. Care was taken to minimize pain and stress during the examinations. The eyelids were held apart gently with cotton-tipped applicators, and the child's head was tilted for observation of the periphery. ROP was classified according to the international classification23 and subdivided into stage 1 (demarcation line), stage 2 (ridge), stage 3 (ridge with extraretinal fibrovascular proliferations), stage 4 (subtotal retinal detachment), or stage 5 (total retinal detachment). In all gestational weeks, each child was classified according to the most advanced ROP stage observed. Proliferative ROP was defined as stage 3 and moderate ROP as stage 1 or stage 2. Twenty-six children had no ROP (stage 0), 23 had stage 1 or 2 ROP, and 9 had stage 3 ROP, of whom 5 were treated. In this study group, there were no infants with Rush disease and 3 of the infants with stage 3 ROP had plus disease. No patients with ROP of lower stages had plus disease. The zone (zone 1–3, according to the international classification) was noted when ROP was found.

Postnatal Measurements of HC, IGF-1 Levels, and Nutrition
A trained nurse without knowledge of ROP or IGF-1 status recorded the mean of 2 HC measurements weekly, to the nearest 1 mm, in a standard way with a measuring tape, from birth until discharge from the hospital. Thereafter, the infants came weekly as outpatients for blood sampling and recording of growth by the study nurse. HC SDS values were calculated from a recent Swedish reference standard of mean HC at birth at different GAs, which correlates very closely with fetal measurements.24 Venous blood samples (0.5 mL) were obtained weekly, and serum was stored at –80°C until assayed. All samples from an individual were analyzed with the same assay. Serum was diluted 1:50, and IGF-1 levels were measured in duplicate with an IGF-binding protein-blocked radioimmunoassay, without extraction and in the presence of a ~250-fold excess of IGF-II (Mediagnost GmbH, Tübingen, Germany). The intraassay coefficients of variation at 10.2 µg/L and 34.5 µg/L were 15.7% and 9.6%, respectively. The interassay coefficients of variation at 10.2 µg/L and 34.5 µg/L were 23.9% and 12.1%, respectively. All infants were nourished according to the routines for premature infants in the neonatal units. Enteral feeding with increasing amounts of breast milk was introduced early (2–48 hours after birth). If full enteral feeding was not achieved, then supplementary parenteral nutrition with glucose, amino acids, and fat was provided. The analyzed breast milk25 given to infants with BW of <1500 g was fortified with 0.8 g of protein per 100 mL (gradually introduced over 1 week) from 10 days of age until the infant weighed 2000 g. The nutritional intake (protein and energy) of each child was recorded daily. The goal was to follow the recommendations for infants with BW of <1500 g, to achieve a daily intake of 3.5 g of protein per kg and 500 to 550 kJ/kg.26

Morbidity
Sixteen infants (29%) had BPD, on the basis of chest radiographic examinations and a need for oxygen supplementation at 36 weeks' PMA. Three infants (5%) had NEC, with gut perforation leading to surgery, and 2 infants had IVH, grade 2 to 4 (Table 1). A majority of the study group (38 of 58 infants) received prenatal treatment with betamethasone. A dichotomized morbidity variable was defined, which took the value of 1 if a child suffered from ≥1 of BPD, NEC, or IVH and the value of 0 if the child had none of these complications.

Statistical Methods
The data are longitudinal, consisting of sequential measurements for each infant. For comparisons between different subgroups, the Kruskal-Wallis test was used with the mean value for each child. For investigation of the degree of relationship between IGF-1 levels and HC SDS, the mean values for each child, at different time points, were used (mean IGF-1 levels from birth to week 31 versus mean HC SDS during week 31 and the value at week 35). To analyze the risk of developing at least stage 3 ROP in relation to the early HC SDS, a logistic regression analysis was performed. The response variable was dichotomized as whether the child did develop at least stage 3 ROP or not. The explanatory variables were mean HC SDS at week 31, GA at birth, and BW (grams and SDS). P values of <.05 were considered significant. For investigation of which variables were associated with HC SDS, regression analysis was used, with HC SDS in week 31 as the response variable and BW, GA, morbidity score, ROP, and protein intake as possible explanatory variables.

A test was performed on the possible differences in protein and energy intake between ROP and no ROP. All of the longitudinal data on protein and energy intake for each child (up to week 40) were used. A multiple regression analysis allowing for repeated values for each individual, in which protein or energy was assumed to be a function of the PMA week, was performed (with PROC MIXED in SAS release 8.12; SDS Institute, Cary, NC). In our model, the intercept a is individual, allowing each infant to have his or her own starting level. A test was performed to test the possible differences in levels of protein or energy intake between children with ROP and without ROP.


   RESULTS
TOP
 ABSTRACT
 METHODS
 RESULTS
DISCUSSION
 REFERENCES
 
Head Growth Retardation Coincides With Retinal Growth Retardation
It is well established that retinal vascular growth retardation is associated with the degree of ROP.4–6,27 As expected, in our cohort the degree of ROP when disease first presented (after PMA of 30 weeks) was associated with the degree of vascular growth retardation, as measured by zone. Vascularization of the retina is normally complete to the nasal periphery by PMA of 30 to 32 weeks. All infants with ROP stage 1 had initial disease in zone 3 (mean zone: 3 ± 0), which indicates almost-complete vascularization, with minimal ROP. Infants with ROP stage 2 had an initial mean zone presentation of 2.45 ± 0.51, and infants with ROP stage 3 had an initial mean zone presentation of 1.69 ± 0.48, which indicates minimal vascularization at the time of first presentation of ROP (ROP stage 1 versus ROP stage 3, P = .0000001; ROP stage 1 versus ROP stage 2, P = .0001; ROP stage 2 versus ROP stage 3, P = .0001).

The retinal growth retardation phase of ROP from birth to PMA of ~30 to 31 weeks PMA was mirrored by retardation of brain growth, measured as HC. The mean HC SDS (SD of HC from well-established normative values at birth for infants of varying PMA)28 decreased significantly for all premature infants with all stages of ROP and infants without ROP from birth to PMA of ~30 to 31 weeks (Fig 1A). The decline in head growth coincided with the growth inhibition phase of ROP (birth to onset of the proliferative phase of ROP) (Fig 1B, arrow).


Figure 1
View larger version (16K):
[in this window]
[in a new window]
  		FIGURE 1 A indicates longitudinal mean HC SDS from the historical normative value at birth for Swedish infants from week 25 to week 40 GA, for infants with no ROP (stage 0) (n = 26), ROP stage 1 or 2 (n = 23), or ROP stage 3 (n = 9). The onset of increased head growth (arrow) at PMA week ~30 coincides with the onset of ROP. The SD from the normative value decreases initially for all premature infants. B indicates the timing of ROP development (any ROP stage) among 32 preterm children. These results are consistent with earlier findings.43 The onset of ROP coincides with increased head growth (arrow).

 
The Degree of Head Growth Retardation Among Premature Infants From Birth to PMA of 31 Weeks Corresponds to the Degree of ROP
Patients who did not develop ROP had the least inhibition of head growth between birth and PMA of 31 weeks, those with ROP stage 3 had the most severe inhibition of head growth, and those with intermediate stages of ROP had intermediate inhibition of head growth (Fig 1A). All infants (including those with no ROP) exhibited an inhibition of head growth, which is most likely related to inadequate nutrition among these premature infants after birth. It should be noted that, when we analyzed the decrease from birth up to week 31, the children had different birth weeks, which cannot be seen in Fig 1. To adjust for the difference in birth week, we analyzed the decrease per day. For patients with no ROP (stage 0), the mean HC decreased from –0.77 SD from the normative value at birth to –1.56 SD at PMA of 31 weeks (mean difference: –0.79 SD; mean difference per day: 0.065 ± 1.96 x 0.025; P = .01). In ROP stage 1 or 2, the mean difference in HC SDS from birth to PMA of 31 weeks was –1.13 SD (mean per day: 0.039 ± 1.96 x 0.0082; P < .005); in ROP stage 3, the mean difference was –1.90 SD (mean per day: 0.045 ± 1.96 x 0.012; P < .005).

There was a considerable increase in HC for all patients, with and without ROP, starting at PMA of ~30 to 32 weeks (Fig 1A, arrow). However, infants who began the head growth spurt with the largest head growth deficit also developed the most severe ROP. At PMA of 31 weeks, there was a very significant difference in mean HC SDS for patients with no ROP and those with proliferative ROP (stage 3) (P = .001) (Fig 2).


Figure 2
View larger version (17K):
[in this window]
[in a new window]
  		FIGURE 2 Box-plot of mean HC SDS at PMA of 31 weeks. Mean HC SDS was calculated for no ROP (stage 0) (n = 26), ROP stage 1 or 2 (n = 23), and ROP stage 3 (n = 9). There is a significant difference in HC SDS from the normative value between ROP stage 0 and ROP stage 3 (P = .001) and between ROP stage 1 or 2 and ROP stage 3 (P = .015). The 3 horizontal lines in each box mark the 25th, 50th, and 75th percentiles. Horizontal lines outside the box mark the minimal and maximal values.

 
There was also a significant difference in mean HC SDS between patients who developed ROP stage 1 or 2 and those who developed ROP stage 3 (P = .015) (Fig 2). At PMA of 31 weeks, the probability of also developing severe ROP (at least stage 3) was 5 times greater if mean HC was below or equal to –2.5 SD than if mean HC was above –2.5 SD (32% vs 6%; P = .007). At PMA of 31 weeks, the risk of also developing ROP stage 3 almost doubled (increased 97%) for each decrease of 0.5 SD of HC, after adjustment for GA and BW.

The Risk of ROP Is Associated With HC SDS at Week 31 and GA
To assess whether HC SDS merely reflects lower BW and lower GA at birth, the association between ROP and HC SDS was analyzed (with logistic regression), after adjustment for GA at birth. BW (and BW SDS) was tried in the model but was found not to give any additional information regarding the risk of stage 3 or proliferative ROP. The results of the multiple logistic regression analysis, with HC SDS during week 31 and GA as explanatory variables, was logit(proliferative ROP) = 20.05 – 1.35(mean HC SDS during week 31) – 0.91(GA week). The R2 value was 0.66 in the multiple logistic regression analysis, performed with the statistical program SPSS (SPSS, Chicago, IL). The relative risk of proliferative ROP associated with a 0.5 increase of HC SDS during PMA week 31 was 0.5 x –1.354 = –0.677 when adjusted for GA. Therefore, if a child at week 31 has a HC SDS 0.5 lower than that of another child with the same GA at birth, then the risk of also having proliferative ROP increases by 97%. Including BW in the model resulted in the same R2 value (0.66), and this factor was therefore excluded.

Postnatal (Not Prenatal) Head Growth Correlates With the Degree of ROP
The association between poor head growth (low HC SDS) and later development of proliferative ROP (at least stage 3) among infants was not present at birth (or within 2 weeks after birth) (P = .969) (data not shown). However, there was an association at PMA of 31 weeks between retarded head growth (HC SDS) and proliferative ROP (P = .005) (Fig 2). This association between retarded head growth (HC SDS) and proliferative ROP persisted at 35 weeks' PMA (P = .002) (data not shown). By PMA of 40 weeks, there was a very pronounced catch-up in head growth for all patients with all stages of ROP. However, even at PMA of 40 weeks, mean head growth was less with worse ROP (Fig 1A).

Mean Serum IGF-1 Levels (From Birth to PMA of 31 Weeks) Correlate With Head Growth
We showed previously that there was a strong correlation between low IGF-1 levels and retinal vascular growth retardation before ~PMA of 30 to 31 weeks and the later development of ROP.18–20 Here we found the same correlation for head growth. Mean serum levels of IGF-1 from birth to PMA of 31 weeks correlated with HC SDS (r = 0.43; P = .001) (Fig 3). Head growth correlated with serum IGF-1 levels to a lesser degree at week 35 PMA (r = 0.28; P = .02; data not shown).


Figure 3
View larger version (11K):
[in this window]
[in a new window]
  		FIGURE 3 HC SDS at PMA of 31 weeks versus mean serum IGF-1 levels from birth to PMA of 31 weeks among 58 preterm children (r = 0.43; P = .001). The curve is the least-squares curve when IGF-I levels are modeled as a second-degree polynomial of HC SDS.31

 
Infants With and Without ROP Had Similar Protein and Energy Intakes
To determine whether in this cohort HC SDS simply reflected poorer nutrition among infants with ROP, we assessed HC SDS at PMA of 31 weeks with respect to protein and energy intake. HC SDS did not correlate with mean protein intake (mean grams per day from birth to PMA of 31 weeks) (r = 0.023; R2 = 0.00053; P = .88). Also, infants with ROP had protein intake similar to that of infants with no ROP (P = .83) (Fig 4). Similarly, there was no correlation between ROP and daily mean energy intake and no correlation between HC SDS and daily mean energy intake (mean joules per kilogram per day from birth to PMA of 31 weeks) (P = .82; data not shown).


Figure 4
View larger version (16K):
[in this window]
[in a new window]
  		FIGURE 4 ROP (any stage) or no ROP versus mean protein intake (mean grams per day) from birth to PMA of 43 weeks (P = .83). There is no significant difference between any stage of ROP versus no ROP in protein intake.

 
Morbidity Significantly Affects ROP and HC SDS
We reported previously that ROP severity was associated with other comorbidities (NEC, IVH, and BPD).20 To evaluate the simultaneous effect of several variables on HC SDS, a multiple regression analysis was performed, with HC SDS as the response variable and the following explanatory variables: BW, PMA, morbidity (see above), mean protein intake up to week 31, ROP, and HC SDS at birth (within 7 days after birth). The protein variable was not significant (P = .98). The morbidity variable was significant (P = .016), as was the ROP variable (P = .029) and HC SDS at birth (0.013). Therefore, after adjustment for morbidity and HC SDS at birth, there was still a significant negative association between severe ROP and HC SDS (ie, lower HC SDS among children with severe ROP; P = .029).


   DISCUSSION
TOP
 ABSTRACT
 METHODS
 RESULTS
 DISCUSSION
REFERENCES
 
Approximately 80% of very preterm infants have neurologic dysfunction at 6 years of age,2 but the causes are not well understood. Because the severity of these impairments correlates with the severity of ROP, which is also a disease of the CNS, we postulated a common mechanism, namely, growth retardation during a critical developmental period.

Because of the rapid brain growth velocity in fetuses in utero of the same GA as the group studied, serial HC measurements (corrected for gender and age with the use of SDS values) reflect deviations in brain growth sensitively. HC was shown in a number of studies to be an accurate indicator of overall brain size.11,12,21 With simple longitudinal measures of HC SDS, we found that, among premature infants, there is early transient head growth retardation from birth to PMA of ~30 to 31 weeks that corresponds closely to the vascular growth retardation in the retina. In the retina, the more profound the retinal vascular growth delay during this period, the worse is the degree of later ROP4–6,27 and the worse is neural retinal function.10,29 We also found that infants with ROP have abnormal retinal vascular growth (as measured with zones), compared with that of intrauterine life, and the severity of ROP was correlated with the severity of vascular retardation.

This suggests that a similar lack of growth in this critical period could possibly predispose infants to neurologic dysfunction. The correlation between retinal and brain growth also suggests that brain dysfunction may have a component of vascular suppression during this period. It also suggests that the same factors that are known to contribute to retinal vascular and retinal neural growth abnormalities, such as IGF-1 and oxygen,30 may contribute to CNS growth retardation. Inadequate nutrition almost certainly contributes to low levels of IGF-1 and the morbidity of prematurity, including ROP and brain growth.26

Starting at PMA of ~31 weeks, there is a period of brain growth proliferation, such that the mean HC is much closer to normal by PMA of 40 weeks. This period of head growth corresponds closely to the vasoproliferative phase of ROP. Despite significant growth in HC by PMA of 40 weeks, very preterm infants are at high risk for neurologic dysfunction.2,8,9 Persistent brain growth deficits at 8 months were associated with greater neurologic dysfunction.31

ROP is usually viewed as a disease of abnormal retinal vascular development but it is increasingly apparent that it is also a disease of neural retinal function. Vascular development takes place in the context of neural retinal development, which determines the need for oxygen and nutrients and determines the need for vascular growth. There is substantial evidence that neural retinal development is interrupted in infants who develop ROP, because they do not develop normal visual acuity and have attenuated visual fields even with regression of vascular disease.7 There is also a strong correlation between the degree of ROP and the degree of abnormal retinal physiologic function.10,29 Because of the link between neural and vascular growth and because the degree of ROP is correlated with the degree of brain dysfunction among premature infants,8,9 it could be hypothesized that brain growth abnormalities are associated with vascular abnormalities in both growth phases.

These growth abnormalities in the retina and head are associated with the smallest and sickest infants. The causes are almost certainly multifactorial. One such factor, IGF-1 (levels of which are low after premature birth), is associated with growth abnormalities of retina and brain. Low serum levels of IGF-1 are associated with abnormal retinal vessel growth.18,20,32 IGF-1 is also important for postnatal brain growth in transgenic mice.15–17,33–36 IGF-1 is neuroprotective in hypoxic injury and suppresses apoptosis.37 We confirmed in our study that brain growth suppression is correlated with low serum IGF-1 levels. It is difficult to distinguish between the effect of IGF-1 on vessels and that on neural development. It is possible that our earlier findings of vascular growth retardation with low IGF-1 levels were a manifestation of an IGF-1 effect on neural retinal development, with a subsequent effect on retinal vessels, or that inhibition of brain growth with low IGF-1 levels is secondary to suppression of vascular growth. IGF-1 is known to be required for vessel remodeling in the adult brain.38 IGF-1 also has a role in recovery from neural injury34 and increased neural proliferation in the embryo and inhibits apoptosis postnatally.33

It is likely that growth factors other than IGF-1 are also involved. IGF-1 is associated with the growth phase of ROP, and suppressing IGF-1 in this second, proliferative phase of the disease can suppress neovascularization,39,40 as does endothelial cell-specific suppression of the insulin receptor. Mice with vascular endothelial cell-specific absence of the insulin receptor have less retinopathy than do mice with endothelial cell-specific knockout of the IGF receptor.41 It is possible that low levels of insulin or insulin resistance could be associated with poor retinal and head growth among preterm infants. We were unable to measure insulin levels in our study because of blood volume restrictions. Others noted a correlation between elevated blood glucose levels in the first 4 weeks of life among preterm infants and increased incidence of ROP.42 Although our results do not prove that low IGF-1 levels cause brain growth retardation, they suggest that IGF-1 is likely to be one of the growth factors involved in this neural and vascular process. It is also possible that low IGF-1 levels reflect inadequate nutritional intake.26,43,44 Although we found that, in this cohort of patients, there was no significant difference in protein or energy intake for patients who did or did not have ROP, it is likely that inadequate nutrition plays a role in this disease process.

The timing of the onset of vasoproliferative ROP at ~31 weeks' PMA, independent of GA at birth, has remained unexplained because it was first observed in 1991.44 Our results suggest that the onset of proliferative ROP is timed to the onset of head and neural retinal growth and that this process has its own internal clock. This suggests that the retina, as part of the CNS, undergoes a proliferation switch at this time. There are other possible explanations for this observation. There may be a genetically determined susceptibility to CNS growth retardation among patients who develop ROP. Nutrition may be involved. Even infants without ROP have a lag in head growth until PMA of ~30, which is likely to be secondary to inadequate nutrition. It is possible that infants who develop ROP and have poor brain development have inadequate liver or other organ development to produce critical growth factors.

The lack of predictive value of HC SDS at birth for later ROP suggests that postnatal brain and retinal growth retardation, rather that prenatal growth, influences most significantly the degree of ROP and perhaps the degree of brain abnormalities. These results also suggest that, despite catch-up head growth by PMA of ~40 weeks, the early head growth deficit between birth and PMA of 30 to 31 weeks has lasting effects, as it does in retina.

These findings that head and retinal growth patterns are parallel have a number of implications. First, normalization of head growth and retinal growth between birth and PMA of ~30 to 32 weeks may help prevent ROP and other brain-associated sequelae of premature birth. Second, the decline in head growth precedes the occurrence of proliferative ROP. Therefore, simple longitudinal HC measurements may be helpful in predicting ROP (both onset and degree) and may also predict which infants will later show neurologic impairments. Follow-up data are needed to prove this. Changes in IGF-1 levels are consistent with a role in postnatal brain and retinal growth. We speculate that restoration of IGF-1 and other growth factors to normal in utero levels may help restore head and retinal growth and may help prevent ROP and other brain abnormalities associated with premature birth.


   ACKNOWLEDGMENTS
 
This work was supported by grants from the Göteborg Medical Society, the Medical Research Council (grants 7509, 10863, and 13515), the Swedish Research Council (C.L.), the V. Kann Rasmussen Foundation, and the National Eye Institute (grants EY008670 and EY14811, to L.E.H.S.). L.E.H.S. is the recipient of the Research to Prevent Blindness Lew R. Wasserman Merit Award.

We are grateful to the staff members of the Department of Ophthalmology for their practical help with the patients, to Lisbeth Larsson for analyzing the IGF-1 samples, and to Eva Andersson for the statistical work.


   FOOTNOTES
 
Accepted Nov 23, 2005.

Address correspondence to Ann Hellström, MD, PhD, Section of Pediatric Opthamology, Queen Silvia Children’s Hospital, S-416 85 Göteborg, Sweden. E-mail: ann.hellstrom{at}medfak.gv.se; or Lois E. H. Smith, MD, PhD, Department of Ophthalmology, Children's Hospital, 300 Longwood Ave, Boston, MA 02115. E-mail: lois.smith{at}childrens.harvard.edu

Drs Smith and Hellström contributed equally to this work.

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


   REFERENCES
TOP
 ABSTRACT
 METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 

  1. Hack M, Flannery DJ, Schluchter M, Cartar L, Borawski E, Klein N. Outcomes in young adulthood for very-low-birth-weight infants. N Engl J Med. 2002;346 :149 –157[Abstract/Free Full Text]
  2. Marlow N, Wolke D, Bracewell MA, Samara M. Neurologic and developmental disability at six years of age after extremely preterm birth. N Engl J Med. 2005;352 :9 –19[Abstract/Free Full Text]
  3. Cooke RW. Are there critical periods for brain growth in children born preterm? Arch Dis Child Fetal Neonatal Ed. 2006;91 :F17 –F20[Abstract/Free Full Text]
  4. Schaffer DB, Palmer EA, Plotsky DF, et al. Prognostic factors in the natural course of retinopathy of prematurity: the Cryotherapy for Retinopathy of Prematurity Cooperative Group. Ophthalmology. 1993;100 :230 –237[ISI][Medline]
  5. Fielder AR, Shaw DE, Robinson J, Ng YK. Natural history of retinopathy of prematurity: a prospective study. Eye. 1992;6 :233 –242[Medline]
  6. Cryotherapy for Retinopathy of Prematurity Cooperative Group. Multicenter trial of cryotherapy for retinopathy of prematurity: natural history ROP: ocular outcome at 5 years in premature infants with birth weights less than 1251 g. Arch Ophthalmol. 2002;120 :595 –599[Abstract/Free Full Text]
  7. Dobson V, Quinn GE, Summers CG, et al. Effect of acute-phase retinopathy of prematurity on grating acuity development in the very low birth weight infant: the Cryotherapy for Retinopathy of Prematurity Cooperative Group. Invest Ophthalmol Vis Sci. 1994;35 :4236 –4244[Abstract/Free Full Text]
  8. Msall ME, Phelps DL, DiGaudio KM, et al. Severity of neonatal retinopathy of prematurity is predictive of neurodevelopmental functional outcome at age 5.5 years. Pediatrics. 2000;106 :998 –1005[Abstract/Free Full Text]
  9. Msall ME, Phelps DL, Hardy RJ, et al. Educational and social competencies at 8 years in children with threshold retinopathy of prematurity in the CRYO-ROP multicenter study. Pediatrics. 2004;113 :790 –799[Abstract/Free Full Text]
  10. Fulton AB, Hansen RM, Petersen RA, Vanderveen DK. The rod photoreceptors in retinopathy of prematurity: an electroretinographic study. Arch Ophthalmol. 2001;119 :499 –505[Abstract/Free Full Text]
  11. Frisk V, Amsel R, Whyte HE. The importance of head growth patterns in predicting the cognitive abilities and literacy skills of small-for-gestational-age children. Dev Neuropsychol. 2002;22 :565 –593[CrossRef][ISI][Medline]
  12. Bartholomeusz HH, Courchesne E, Karns CM. Relationship between head circumference and brain volume in healthy normal toddlers, children, and adults. Neuropediatrics. 2002;33 :239 –241[CrossRef][ISI][Medline]
  13. Carver JA. Probing the structure and interactions of crystallin proteins by NMR spectroscopy. Prog Retin Eye Res. 1999;18 :431 –462[CrossRef][ISI][Medline]
  14. Leinninger GM, Feldman EL. Insulin-like growth factors in the treatment of neurological disease. Endocr Dev. 2005;9 :135 –159[Medline]
  15. Doublier S, Duyckaerts C, Seurin D, Binoux M. Impaired brain development and hydrocephalus in a line of transgenic mice with liver-specific expression of human insulin-like growth factor binding protein-1. Growth Horm IGF Res. 2000;10 :267 –274[CrossRef][ISI][Medline]
  16. Fukudome Y, Tabata T, Miyoshi T, et al. Insulin-like growth factor-I as a promoting factor for cerebellar Purkinje cell development. Eur J Neurosci. 2003;17 :2006 –2016[Medline]
  17. Lee KH, Calikoglu AS, Ye P, D'Ercole AJ. Insulin-like growth factor-I (IGF-I) ameliorates and IGF binding protein-1 (IGFBP-1) exacerbates the effects of undernutrition on brain growth during early postnatal life: studies in IGF-I and IGFBP-1 transgenic mice. Pediatr Res. 1999;45 :331 –336[ISI][Medline]
  18. Hellstrom A, Perruzzi C, Ju M, et al. Low IGF-I suppresses VEGF-survival signaling in retinal endothelial cells: direct correlation with clinical retinopathy of prematurity. Proc Natl Acad Sci USA. 2001;98 :5804 –5808[Abstract/Free Full Text]
  19. Smith LE, Hellstrom A. RP-forme fruste. J Perinatol. 2002;22 :257[CrossRef][Medline]
  20. Hellstrom A, Engstrom E, Hard AL, et al. Postnatal serum insulin-like growth factor I deficiency is associated with retinopathy of prematurity and other complications of premature birth. Pediatrics. 2003;112 :1016 –1020[Abstract/Free Full Text]
  21. Lindley AA, Benson JE, Grimes C, Cole TM III, Herman AA. The relationship in neonates between clinically measured head circumference and brain volume estimated from head CT-scans. Early Hum Dev. 1999;56 :17 –29[CrossRef][ISI][Medline]
  22. Marsal K, Persson PH, Larsen T, Lilja H, Selbing A, Sultan B. Intrauterine growth curves based on ultrasonically estimated foetal weights. Acta Paediatr. 1996;85 :843 –848[ISI][Medline]
  23. Flynn JT. An international classification of retinopathy of prematurity: development of the classification of the late stages of retinopathy of prematurity. Birth Defects Orig Artic Ser. 1988;24 :175 –183, C2–C3[Medline]
  24. Hellstrom A, Svensson E, Carlsson B, Niklasson A, Albertsson-Wikland K. Reduced retinal vascularization in children with growth hormone deficiency. J Clin Endocrinol Metab. 1999;84 :795 –798[Abstract/Free Full Text]
  25. Michaelsen K, Pedersen S, Skafte L, Jaeger P, Peitersen B. Infrared analysis for determining macronutrients in human milk. J Pediatr Gastroenterol Nutr. 1998;7 :229 –235[CrossRef]
  26. Engstrom E, Niklasson A, Wikland KA, Ewald U, Hellstrom A. The role of maternal factors, postnatal nutrition, weight gain, and gender in regulation of serum IGF-I among preterm infants. Pediatr Res. 2005;57 :605 –610[CrossRef][Medline]
  27. Penn JS, Tolman BL, Henry MM. Oxygen-induced retinopathy in the rat: relationship of retinal nonperfusion to subsequent neovascularization. Invest Ophthalmol Vis Sci. 1994;35 :3429 –3435[Abstract/Free Full Text]
  28. Niklasson A, Karlberg P. New reference for Swedish infants' size at birth. Horm Res. 1999;51(suppl 2) :28
  29. Fulton AB, Hansen RM. Photoreceptor function in infants and children with a history of mild retinopathy of prematurity. J Opt Soc Am A Opt Image Sci Vis. 1996;13 :566 –571[ISI][Medline]
  30. Felderhoff-Mueser U, Bittigau P, Sifringer M, et al. Oxygen causes cell death in the developing brain. Neurobiol Dis. 2004;17 :273 –282[CrossRef][ISI][Medline]
  31. Hack M, Breslau N, Weissman B, Aram D, Klein N, Borawski E. Effect of very low birth weight and subnormal head size on cognitive abilities at school age. N Engl J Med. 1991;325 :231 –237[Abstract]
  32. Hellstrom A, Carlsson B, Niklasson A, et al. IGF-I is critical for normal vascularization of the human retina. J Clin Endocrinol Metab. 2002;87 :3413 –3416[Abstract/Free Full Text]
  33. Popken GJ, Hodge RD, Ye P, et al. In vivo effects of insulin-like growth factor-I (IGF-I) on prenatal and early postnatal development of the central nervous system. Eur J Neurosci. 2004;19 :2056 –2068[CrossRef][ISI][Medline]
  34. D'Ercole AJ, Ye P, O'Kusky JR. Mutant mouse models of insulin-like growth factor actions in the central nervous system. Neuropeptides. 2002;36 :209 –220[CrossRef][ISI][Medline]
  35. Murphy LJ, Rajkumar K, Molnar P. Phenotypic manifestations of insulin-like growth factor binding protein-1 (IGFBP-1) and IGFBP-3 overexpression in transgenic mice. Prog Growth Factor Res. 1995;6 :425 –432[CrossRef][Medline]
  36. D'Ercole AJ, Dai Z, Xing Y, et al. Brain growth retardation due to the expression of human insulin-like growth factor binding protein-1 in transgenic mice: an in vivo model for the analysis of IGF function in the brain. Brain Res Dev Brain Res. 1994;82 :213 –222[Medline]
  37. Seigel GM, Chiu L, Paxhia A. Inhibition of neuroretinal cell death by insulin-like growth factor-1 and its analogs. Mol Vis. 2000;6 :157 –163[ISI][Medline]
  38. Lopez-Lopez C, LeRoith D, Torres-Aleman I. Insulin-like growth factor I is required for vessel remodeling in the adult brain. Proc Natl Acad Sci USA. 2004;101 :9833 –9838[Abstract/Free Full Text]
  39. Smith LE, Kopchick JJ, Chen W, et al. Essential role of growth hormone in ischemia-induced retinal neovascularization. Science. 1997;276 :1706 –1709[Abstract/Free Full Text]
  40. Smith LE, Shen W, Perruzzi C, et al. Regulation of vascular endothelial growth factor-dependent retinal neovascularization by insulin-like growth factor-1 receptor. Nat Med. 1999;5 :1390 –1395[CrossRef][ISI][Medline]
  41. Kondo T, Vicent D, Suzuma K, et al. Knockout of insulin and IGF-1 receptors on vascular endothelial cells protects against retinal neovascularization. J Clin Invest. 2003;111 :1835 –1842[CrossRef][ISI][Medline]
  42. Garg R, Agthe AG, Donohue PK, Lehmann CU. Hyperglycemia and retinopathy of prematurity in very low birth weight infants. J Perinatol. 2003;23 :186 –194[Medline]
  43. Smith WJ, Underwood LE, Clemmons DR. Effects of caloric or protein restriction on insulin-like growth factor-I (IGF-I) and IGF-binding proteins in children and adults. J Clin Endocrinol Metab. 1995;80 :443 –449[Abstract]
  44. Palmer EA, Flynn JT, Hardy RJ, et al. Incidence and early course of retinopathy of prematurity: the Cryotherapy for Retinopathy of Prematurity Cooperative Group. Ophthalmology. 1991;98 :1628 –1640[ISI][Medline]
PEDIATRICS (ISSN 1098-4275). ©2006 by the American Academy of Pediatrics



This article has been cited by other articles:


Home page
 J Trop PediatrHome page
M. G. Oliveira, R. C. Silveira, and R. S. Procianoy
Growth of Very Low Birth Weight Infants at 12 Months Corrected Age in Southern Brazil
J Trop Pediatr, February 1, 2008; 54(1): 36 - 42.
[Abstract] [Full Text] [PDF]

Home page
 Proc. Natl. Acad. Sci. USAHome page
C. Lofqvist, J. Chen, K. M. Connor, A. C. H. Smith, C. M. Aderman, N. Liu, J. E. Pintar, T. Ludwig, A. Hellstrom, and L. E. H. Smith
From the Cover: IGFBP3 suppresses retinopathy through suppression of oxygen-induced vessel loss and promotion of vascular regrowth
PNAS, June 19, 2007; 104(25): 10589 - 10594.
[Abstract] [Full Text] [PDF]

Home page
 Proc. Natl. Acad. Sci. USAHome page
K.-H. Chang, T. Chan-Ling, E. L. McFarland, A. Afzal, H. Pan, L. C. Baxter, L. C. Shaw, S. Caballero, N. Sengupta, S. L. Calzi, et al.
IGF binding protein-3 regulates hematopoietic stem cell and endothelial precursor cell function during vascular development
PNAS, June 19, 2007; 104(25): 10595 - 10600.
[Abstract] [Full Text] [PDF]

Home page
 PediatricsHome page
M. E. Msall
The retina as a window to the brain in vulnerable neonates.
Pediatrics, June 1, 2006; 117(6): 2287 - 2289.
[Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow P3Rs: Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when P3Rs are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow E-mail this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My File Cabinet
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via CrossRef
Right arrow Citing Articles via ISI Web of Science (6)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Löfqvist, C.
Right arrow Articles by Hellström, A.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Löfqvist, C.
Right arrow Articles by Hellström, A.
Related Collections
Right arrow Premature & Newborn