Acta Pædiatrica ISSN 0803-5253
REGULAR ARTICLE
Birth weight is the most important predictor of abnormal retinal vascularisation in moderately preterm infants K Allvin1, A Hellstr€om2, J Dahlgren1, M Andersson Gr€onlund ([email protected])2 1.Gothenburg Pediatric Growth Research Center (GP-GRC), Institute of Clinical Sciences, Sahlgrenska Academy at the University of Gothenburg and The Queen Silvia Children’s Hospital, Gothenburg, Sweden 2.Institute of Neuroscience and Physiology/Ophthalmology, Sahlgrenska Academy at the University of Gothenburg, Gothenburg, Sweden
Keywords Birth size, Insulin-like growth factor-I, Pre-eclampsia, Retinal vascularisation, Small for gestational age
ABSTRACT Aim: To find predictors of abnormal retinal vascularisation in moderately to late preterm
Correspondence €nlund, MD, PhD, Department Marita Andersson Gro of Paediatric Ophthalmology, The Queen Silvia Children’s Hospital, SE 416 85 Gothenburg, Sweden. Tel: +46-31-343-4688 | Fax: +46-31-845113 | Email: [email protected]
Methods: Seventy-eight infants (34 girls) were recruited from a longitudinal study of
Received 13 September 2013; revised 12 November 2013; accepted 10 February 2014. DOI:10.1111/apa.12599
newborn infants considered to have no risk of developing retinopathy of prematurity. otherwise healthy premature children born at a gestational age of 32 + 0–36 + 6 weeks. Retinal vessel morphology was evaluated at mean postnatal age 7 days. Insulin-like growth factor-I (IGF-I) levels were analysed in umbilical cord blood. Results: Of the 78 infants, 21 (27%) had abnormal retinal vessel morphology; they had significantly lower median (range) birth weight [1850 g, (1190–3260), vs. 2320, (1330– 3580), p < 0.0001], shorter birth length [43.0 cm, (38–49), vs. 46.0, (40–50), p < 0.0001] and smaller head circumference [31.0 cm, (27.7–34.0), vs. 32.0, (27.5– 36.5), p = 0.003]. They also had significantly lower gestational age [34 + 1 weeks, (32 + 2–35 + 3), vs. 34 + 6, (32 + 2–36 + 6), p = 0.004] and mean SD IGF-I levels (24.6 17.0 lg/L vs. 46.7 21.5, p < 0.0001). A higher percentage of these infants were small for gestational age (57.1% vs. 15.8%, p = 0.001), and maternal hypertension/ preeclampsia rates were also higher (47.6% vs. 19.3%, p = 0.03). Step-wise logistic regression showed that birth weight was the strongest predictor of abnormal retinal vascularisation (p < 0.0001, odds ratio 0.040, 95% confidence interval 0.007–0.216). Conclusion: In this population of moderately to late preterm newborns, birth weight appeared to affect the retinal vascular system.
INTRODUCTION Previous studies have shown that abnormal retinal vascularisation in very preterm children persists into adult life (1,2). This may also be the case in moderately to late preterm (MLP) infants, but has not previously been investigated. Moderately to late preterm infants are by definition born at ≥32 weeks’ of gestation, with moderately preterm infants born at 32 + 0–33 + 6 weeks’ gestational age (GA) and late preterm infants born at 34 + 0–36 + 6 weeks’ GA (3). More than 80% of all preterm babies are MLP (3,4). It has previously been reported that both moderately and late preterm infants have an increased risk of mortality and morbidity, such as hyperbilirubinaemia, temperature instability, apnoea and respiratory distress syndrome, as well as
Key notes
Abbreviations AGA, Appropriate for gestational age; GA, Gestational age; IGFII, Insulin-like growth factor II; IGF-I, Insulin-like growth factor I; IGF, Insulin-like growth factor; LBW, Low birth weight; MLP, Moderately to late preterm; RIA, Radioimmunoassay; ROP, Retinopathy of prematurity; SDS, Standard deviation score; SD, Standard deviation; SGA, Small for gestational age.
594
impaired long-term neurodevelopmental outcomes, compared with term infants (5,6). Some studies point out that, like children born extremely preterm, those born moderately preterm or small for gestational age (SGA) may have an increased risk of vascular dysfunction and hypertension later in life (7). Other studies have found low birth weight (LBW), due to preterm birth, intrauterine growth restriction, or being born SGA, to be associated with impaired microvascular development and with cardiovascular disease (8,9). Kistner
Children born extremely preterm are at risk of developing retinopathy of prematurity in the neonatal period, as well as cardiovascular disease later in life. In our study, 21/78 infants born moderately to late preterm (gestational age 32 + 0–36 + 6 weeks), showed abnormal retinal vascularisation in the first month of life. This finding may be a risk indicator for later development of cardiovascular disease.
©2014 Foundation Acta Pædiatrica. Published by John Wiley & Sons Ltd 2014 103, pp. 594–600
Allvin et al.
et al. found that women who had been born preterm, at a median GA of 30 weeks and with a median birth weight of 1250 g, had higher length index for retinal arterioles at 23–30-years-of-age, had higher length index for retinal arterioles compared with those born at term (either SGA or normal size). These women also had fewer retinal vascular branching points and high casual blood pressure (2). Retinal arteriolar narrowing may be an underlying marker of LBW and hypertension (10). Gopinath et al. (11) found narrower retinal arterioles in young adolescents born SGA and Mitchell et al. (12) reported that, in a population of 6-year-old children, both preterm and term, mean retinal arteriolar calibre narrowed by 2.3 lm per kg reduction in birth weight. The retina is not fully developed and vascularised until gestational week 37. Retinopathy of prematurity (ROP) is a proliferative vascular complication in the immature retina. In the developed part of the world, ROP is seen predominantly in very preterm children with very low birth weight. In other parts of the world, ROP is seen at a higher GA (13). Fully developed ROP can progress to retinal detachment and blindness if not detected in time. To prevent this, laser photocoagulation is performed at a certain stage of the disorder and a screening program for detecting and treating ROP has been established (14). In the clinical setting of our hospital, all children born with a GA < 32 weeks are screened for ROP. Insulin-like growth factor-I (IGF-I) is the most important growth factor during foetal and infant life. Low IGF-I levels are associated with intrauterine and extrauterine growth retardation, microcephaly and intellectual impairment (15). IGF-I is also important for the development of retinal vessels (16). The purpose of this study was to evaluate retinal vessel morphology of MLP infants during their first week of life to determine predictors of abnormal retinal vascularisation in newborns considered to have no risk of developing ROP. Our hypothesis was that foetal growth restriction inhibits optimal vascular growth as a result of decreased angiogenetic growth factors and hypoxia. To our knowledge, this has not been investigated previously, and the present study may give new insights into what effects the intrauterine environment might have on children born MLP.
PATIENTS AND METHODS Patients From an ongoing population-based, longitudinal study of otherwise healthy, preterm children born at GA 32 + 0– 36 + 6 weeks (n = 162), 80 infants (35 girls) born at the two delivery wards at Sahlgrenska University Hospital, Gothenburg, Sweden, were consecutively enrolled between October 2002 and October 2003. The following exclusion criteria were used: severe illness, asphyxia, malformations and chromosomal defects, as well as severe maternal illness, psychosocial problems or language difficulties. The parents of one twin boy were unwilling to let the boy be examined, and one girl with extreme growth
Birth weight and retinal vascularisation
retardation, who later died, was excluded. Consequently, 78 infants (34 girls, 44 boys) with a median GA of 34 + 4 weeks (range 32 + 2–36 + 6) were included in this study. In the 82 newborns not enrolled, no ophthalmological examination was performed due to early dropout in seven cases. The remaining 75 neonates (33 girls) with a median GA of 36 + 2 weeks (range 32 + 2–36 + 6) underwent no eye examination because they were discharged from the hospital before any examination could be performed. Methods Retinal vessel morphology during neonatal life was examined under cycloplegia (cyclopentolate 0.2% + phenylephrine 1%) by indirect ophthalmoscopy according to a specific, standardised study protocol. Retinal vessel morphological changes of the fundus were defined as: Tortuosity of the retinal vessels (moderately increased or significantly increased), number of vascular branching points (decreased or increased), or retinal vessel calibre (narrow or increased). These changes were subjectively assessed by a paediatric ophthalmologist specially trained in ocular fundus morphology (AH). The aim was to examine the infants during their first weeks of life, when they were still inpatients in the maternal or neonatal wards. The paediatric ophthalmologist was not given any information about gestational age, birth weight or growth retardation when examining the infants. Birth weight, using baby scales or electronic step scales, and birth length were measured with the infant in supine position by trained nurses in the study team. Genderspecific standard deviation scores (SDSs) were calculated according to the reference by Niklasson and AlbertssonWikland (17). It was recorded whether the infant was born SGA (birth weight and/or birth length < 2 SDSs) or was a twin. In addition, the following pregnancy complications were recorded if present: hypertension (blood pressure >140/90 mm Hg); pre-eclampsia (blood pressure >140/90 and proteinuria after 20 weeks’ gestation); and maternal diabetes type 1 or type 2 before pregnancy or pathological oral glucose tolerance test during pregnancy (the latter defined as plasma glucose >10.0 mmol/L at 2 h). Serum IGF-I concentrations were measured in umbilical cord blood using a specific radioimmunoassay (RIA) (Me€ bingen, Germany). The IGF-I samples diagnost GmbH, Tu were diluted 1:50. The intra-assay coefficients of variation were 18%, 11% and 7% at concentrations of 9, 33 and 179 lg/L, respectively. The methods have previously been described in detail (18). Pathological retinal vessel morphology was analysed to establish any correlation between the following infant variables: GA at birth, birth weight (g), birth length (cm), head circumference (cm), IGF-I (lg/L), gender, appropriate for gestational age (AGA)/SGA and twin/singleton. The maternal variables diabetes and pre-eclampsia/hypertension were also tested for any correlation with pathological retinal vessel morphology.
©2014 Foundation Acta Pædiatrica. Published by John Wiley & Sons Ltd 2014 103, pp. 594–600
595
Birth weight and retinal vascularisation
Allvin et al.
Statistical analyses Means, medians, standard deviations (SD), SDSs and ranges were calculated for descriptive purposes. For comparison between two groups, the Mann–Whitney’s U-test was used for continuous variables; for dichotomous variables, Fisher’s exact test was used. All tests were two-tailed and conducted at the 5% significance level. A step-wise logistic regression was performed to find the variable, or set of variables, that best predicted abnormal retinal vascularisation. Ethical consideration The study was approved by the Ethics Committee of the Medical Faculty of the University of Gothenburg. Informed consent was obtained from the parents of the participants.
RESULTS Table 1 shows auxological data at birth (GA, weight, length and head circumference), for all children born during the study period and included in the prospective longitudinal study (n = 153), divided into those infants who underwent a fundus examination neonatally (n = 78) and those who did not (n = 75). The table also shows infant variables (gender, twin vs. singleton, AGA vs. SGA) and maternal health variables (diabetes mellitus and pre-eclampsia/hypertension during pregnancy).
The infants who underwent an eye examination had lower GA at birth and lower birth weight, birth length and head circumference, with a higher percentage of SGA. There was no difference regarding gender, twin/singleton, pre-eclampsia/hypertension or maternal diabetes (Table 1). Serum mean SD levels of IGF-I were also lower in the umbilical cord in those newborns who underwent eye examination compared with those who did not (41 23 lg/L vs. 56 24 lg/L, p < 0.0001). In total, 21 (10 girls) of the 78 investigated neonates (27%) had abnormal retinal vessel morphology, such as increased vascular tortuosity (n = 3), decreased number of vascular branching points (n = 12), increased number of vascular branching points (n = 1) and narrow arterioles and/or venules (n = 13). None of them showed signs of ROP. The 78 neonates were examined at a mean postnatal age of 7 days (ranging from 1 to 28 days). Infants showing abnormal retinal vascularisation underwent an eye examination at the same postnatal age (mean 7 days; ranging from 3 to 19 days) as those with normal retinal vascularisation (mean 7 days; ranging from 1 to 28 days). Infants with abnormal retinal vascularisation (e.g. increased vascular tortuosity, decreased/increased number of vascular branching points and/or narrow arterioles and/or venules) had significantly lower birth weight, birth length and head circumference at birth, as well as lower GA at birth (Table 2). They had also a higher percentage of SGA
Table 1 Auxological data at birth, infant data and maternal data of 153 infants born moderately to late preterm (MLP) divided into those who had an ophthalmological examination performed (study group, n = 78) and those who did not (n = 75) Auxological data at birth
All newborns n = 153
Study group n = 78
Not examined n = 75
p-Value
Gestational age (week) Birth weight (g) Birth length (cm) Head circumference (cm) Infant data Gender Girl Boy Twin No Yes SGA/AGA AGA SGA Maternal data Pre-eclampsia/hypertension No Yes Pre-eclampsia Hypertension Diabetes mellitus No Yes
35 + 4 2500 47.0 32.5
34 + 4 2250 45.0 31.8
36 + 2 2660 47.0 33.0
0.0001 0.0001 0.0001 0.0001
(32 + 2; 36 + 6) (1190; 3575) (38.0; 51.0)* (27.5; 36.5)
(32 + 2; 36 + 6) (1190; 3575) (38.0; 50.0)** (27.5; 36.5)
(32 + 2; 36 + 6) (1585; 3420) (42.0; 51.0) (29.5; 35.0)
67 (43.8%) 86 (56.2%)
34 (43.6%) 44 (56.4%)
33 (44%) 42 (56%)
1.000
120 (78.4%) 33 (21.6%)
59 (75.6%) 19 (24.4%)
61 (81.3%) 14 (18.7%)
0.51
124 (81%) 29 (19%)
57 (73.1%) 21 (26.9%)
67 (89.3%) 8 (10.7%)
0.018
114 39 34 5
57 21 19 2
57 18 15 3
(74.5%) (25.5%) (22.2%) (3.3%)
146 (95.4%) 7 (4.6%)
(73.1%) (26.9%) (24.3%) (2.6%)
73 (93.6%) 5 (6.4%)
(76%) (24%) (20%) (4%)
73 (97.3%) 2 (2.7%)
0.820
0.476
For categorical variables, n (%) is presented. For continuous variables, median (min; max) are presented. *n = 152; **n = 77; AGA, Appropriate for gestational age; SGA, Small for gestational age.
596
©2014 Foundation Acta Pædiatrica. Published by John Wiley & Sons Ltd 2014 103, pp. 594–600
Allvin et al.
Birth weight and retinal vascularisation
Table 2 Auxological data at birth of 78 infants born moderately to late preterm (MLP) divided into those with abnormal (n = 21) and normal (n = 57) retinal vascularisation Auxological variables Gestational age (week) Birth weight (g) Birth length (cm) Head circumference (cm)
Abnormal retinal vascularisation (n = 21)
Normal retinal vascularisation (n = 57)
p-Value
34 + 1 (32 + 2; 34 + 6 (32 + 35 + 3) 2; 36 + 6) 1850 (1190; 3260) 2320 (1330; 3580)
0.0039
REGULAR ARTICLE
Birth weight is the most important predictor of abnormal retinal vascularisation in moderately preterm infants K Allvin1, A Hellstr€om2, J Dahlgren1, M Andersson Gr€onlund ([email protected])2 1.Gothenburg Pediatric Growth Research Center (GP-GRC), Institute of Clinical Sciences, Sahlgrenska Academy at the University of Gothenburg and The Queen Silvia Children’s Hospital, Gothenburg, Sweden 2.Institute of Neuroscience and Physiology/Ophthalmology, Sahlgrenska Academy at the University of Gothenburg, Gothenburg, Sweden
Keywords Birth size, Insulin-like growth factor-I, Pre-eclampsia, Retinal vascularisation, Small for gestational age
ABSTRACT Aim: To find predictors of abnormal retinal vascularisation in moderately to late preterm
Correspondence €nlund, MD, PhD, Department Marita Andersson Gro of Paediatric Ophthalmology, The Queen Silvia Children’s Hospital, SE 416 85 Gothenburg, Sweden. Tel: +46-31-343-4688 | Fax: +46-31-845113 | Email: [email protected]
Methods: Seventy-eight infants (34 girls) were recruited from a longitudinal study of
Received 13 September 2013; revised 12 November 2013; accepted 10 February 2014. DOI:10.1111/apa.12599
newborn infants considered to have no risk of developing retinopathy of prematurity. otherwise healthy premature children born at a gestational age of 32 + 0–36 + 6 weeks. Retinal vessel morphology was evaluated at mean postnatal age 7 days. Insulin-like growth factor-I (IGF-I) levels were analysed in umbilical cord blood. Results: Of the 78 infants, 21 (27%) had abnormal retinal vessel morphology; they had significantly lower median (range) birth weight [1850 g, (1190–3260), vs. 2320, (1330– 3580), p < 0.0001], shorter birth length [43.0 cm, (38–49), vs. 46.0, (40–50), p < 0.0001] and smaller head circumference [31.0 cm, (27.7–34.0), vs. 32.0, (27.5– 36.5), p = 0.003]. They also had significantly lower gestational age [34 + 1 weeks, (32 + 2–35 + 3), vs. 34 + 6, (32 + 2–36 + 6), p = 0.004] and mean SD IGF-I levels (24.6 17.0 lg/L vs. 46.7 21.5, p < 0.0001). A higher percentage of these infants were small for gestational age (57.1% vs. 15.8%, p = 0.001), and maternal hypertension/ preeclampsia rates were also higher (47.6% vs. 19.3%, p = 0.03). Step-wise logistic regression showed that birth weight was the strongest predictor of abnormal retinal vascularisation (p < 0.0001, odds ratio 0.040, 95% confidence interval 0.007–0.216). Conclusion: In this population of moderately to late preterm newborns, birth weight appeared to affect the retinal vascular system.
INTRODUCTION Previous studies have shown that abnormal retinal vascularisation in very preterm children persists into adult life (1,2). This may also be the case in moderately to late preterm (MLP) infants, but has not previously been investigated. Moderately to late preterm infants are by definition born at ≥32 weeks’ of gestation, with moderately preterm infants born at 32 + 0–33 + 6 weeks’ gestational age (GA) and late preterm infants born at 34 + 0–36 + 6 weeks’ GA (3). More than 80% of all preterm babies are MLP (3,4). It has previously been reported that both moderately and late preterm infants have an increased risk of mortality and morbidity, such as hyperbilirubinaemia, temperature instability, apnoea and respiratory distress syndrome, as well as
Key notes
Abbreviations AGA, Appropriate for gestational age; GA, Gestational age; IGFII, Insulin-like growth factor II; IGF-I, Insulin-like growth factor I; IGF, Insulin-like growth factor; LBW, Low birth weight; MLP, Moderately to late preterm; RIA, Radioimmunoassay; ROP, Retinopathy of prematurity; SDS, Standard deviation score; SD, Standard deviation; SGA, Small for gestational age.
594
impaired long-term neurodevelopmental outcomes, compared with term infants (5,6). Some studies point out that, like children born extremely preterm, those born moderately preterm or small for gestational age (SGA) may have an increased risk of vascular dysfunction and hypertension later in life (7). Other studies have found low birth weight (LBW), due to preterm birth, intrauterine growth restriction, or being born SGA, to be associated with impaired microvascular development and with cardiovascular disease (8,9). Kistner
Children born extremely preterm are at risk of developing retinopathy of prematurity in the neonatal period, as well as cardiovascular disease later in life. In our study, 21/78 infants born moderately to late preterm (gestational age 32 + 0–36 + 6 weeks), showed abnormal retinal vascularisation in the first month of life. This finding may be a risk indicator for later development of cardiovascular disease.
©2014 Foundation Acta Pædiatrica. Published by John Wiley & Sons Ltd 2014 103, pp. 594–600
Allvin et al.
et al. found that women who had been born preterm, at a median GA of 30 weeks and with a median birth weight of 1250 g, had higher length index for retinal arterioles at 23–30-years-of-age, had higher length index for retinal arterioles compared with those born at term (either SGA or normal size). These women also had fewer retinal vascular branching points and high casual blood pressure (2). Retinal arteriolar narrowing may be an underlying marker of LBW and hypertension (10). Gopinath et al. (11) found narrower retinal arterioles in young adolescents born SGA and Mitchell et al. (12) reported that, in a population of 6-year-old children, both preterm and term, mean retinal arteriolar calibre narrowed by 2.3 lm per kg reduction in birth weight. The retina is not fully developed and vascularised until gestational week 37. Retinopathy of prematurity (ROP) is a proliferative vascular complication in the immature retina. In the developed part of the world, ROP is seen predominantly in very preterm children with very low birth weight. In other parts of the world, ROP is seen at a higher GA (13). Fully developed ROP can progress to retinal detachment and blindness if not detected in time. To prevent this, laser photocoagulation is performed at a certain stage of the disorder and a screening program for detecting and treating ROP has been established (14). In the clinical setting of our hospital, all children born with a GA < 32 weeks are screened for ROP. Insulin-like growth factor-I (IGF-I) is the most important growth factor during foetal and infant life. Low IGF-I levels are associated with intrauterine and extrauterine growth retardation, microcephaly and intellectual impairment (15). IGF-I is also important for the development of retinal vessels (16). The purpose of this study was to evaluate retinal vessel morphology of MLP infants during their first week of life to determine predictors of abnormal retinal vascularisation in newborns considered to have no risk of developing ROP. Our hypothesis was that foetal growth restriction inhibits optimal vascular growth as a result of decreased angiogenetic growth factors and hypoxia. To our knowledge, this has not been investigated previously, and the present study may give new insights into what effects the intrauterine environment might have on children born MLP.
PATIENTS AND METHODS Patients From an ongoing population-based, longitudinal study of otherwise healthy, preterm children born at GA 32 + 0– 36 + 6 weeks (n = 162), 80 infants (35 girls) born at the two delivery wards at Sahlgrenska University Hospital, Gothenburg, Sweden, were consecutively enrolled between October 2002 and October 2003. The following exclusion criteria were used: severe illness, asphyxia, malformations and chromosomal defects, as well as severe maternal illness, psychosocial problems or language difficulties. The parents of one twin boy were unwilling to let the boy be examined, and one girl with extreme growth
Birth weight and retinal vascularisation
retardation, who later died, was excluded. Consequently, 78 infants (34 girls, 44 boys) with a median GA of 34 + 4 weeks (range 32 + 2–36 + 6) were included in this study. In the 82 newborns not enrolled, no ophthalmological examination was performed due to early dropout in seven cases. The remaining 75 neonates (33 girls) with a median GA of 36 + 2 weeks (range 32 + 2–36 + 6) underwent no eye examination because they were discharged from the hospital before any examination could be performed. Methods Retinal vessel morphology during neonatal life was examined under cycloplegia (cyclopentolate 0.2% + phenylephrine 1%) by indirect ophthalmoscopy according to a specific, standardised study protocol. Retinal vessel morphological changes of the fundus were defined as: Tortuosity of the retinal vessels (moderately increased or significantly increased), number of vascular branching points (decreased or increased), or retinal vessel calibre (narrow or increased). These changes were subjectively assessed by a paediatric ophthalmologist specially trained in ocular fundus morphology (AH). The aim was to examine the infants during their first weeks of life, when they were still inpatients in the maternal or neonatal wards. The paediatric ophthalmologist was not given any information about gestational age, birth weight or growth retardation when examining the infants. Birth weight, using baby scales or electronic step scales, and birth length were measured with the infant in supine position by trained nurses in the study team. Genderspecific standard deviation scores (SDSs) were calculated according to the reference by Niklasson and AlbertssonWikland (17). It was recorded whether the infant was born SGA (birth weight and/or birth length < 2 SDSs) or was a twin. In addition, the following pregnancy complications were recorded if present: hypertension (blood pressure >140/90 mm Hg); pre-eclampsia (blood pressure >140/90 and proteinuria after 20 weeks’ gestation); and maternal diabetes type 1 or type 2 before pregnancy or pathological oral glucose tolerance test during pregnancy (the latter defined as plasma glucose >10.0 mmol/L at 2 h). Serum IGF-I concentrations were measured in umbilical cord blood using a specific radioimmunoassay (RIA) (Me€ bingen, Germany). The IGF-I samples diagnost GmbH, Tu were diluted 1:50. The intra-assay coefficients of variation were 18%, 11% and 7% at concentrations of 9, 33 and 179 lg/L, respectively. The methods have previously been described in detail (18). Pathological retinal vessel morphology was analysed to establish any correlation between the following infant variables: GA at birth, birth weight (g), birth length (cm), head circumference (cm), IGF-I (lg/L), gender, appropriate for gestational age (AGA)/SGA and twin/singleton. The maternal variables diabetes and pre-eclampsia/hypertension were also tested for any correlation with pathological retinal vessel morphology.
©2014 Foundation Acta Pædiatrica. Published by John Wiley & Sons Ltd 2014 103, pp. 594–600
595
Birth weight and retinal vascularisation
Allvin et al.
Statistical analyses Means, medians, standard deviations (SD), SDSs and ranges were calculated for descriptive purposes. For comparison between two groups, the Mann–Whitney’s U-test was used for continuous variables; for dichotomous variables, Fisher’s exact test was used. All tests were two-tailed and conducted at the 5% significance level. A step-wise logistic regression was performed to find the variable, or set of variables, that best predicted abnormal retinal vascularisation. Ethical consideration The study was approved by the Ethics Committee of the Medical Faculty of the University of Gothenburg. Informed consent was obtained from the parents of the participants.
RESULTS Table 1 shows auxological data at birth (GA, weight, length and head circumference), for all children born during the study period and included in the prospective longitudinal study (n = 153), divided into those infants who underwent a fundus examination neonatally (n = 78) and those who did not (n = 75). The table also shows infant variables (gender, twin vs. singleton, AGA vs. SGA) and maternal health variables (diabetes mellitus and pre-eclampsia/hypertension during pregnancy).
The infants who underwent an eye examination had lower GA at birth and lower birth weight, birth length and head circumference, with a higher percentage of SGA. There was no difference regarding gender, twin/singleton, pre-eclampsia/hypertension or maternal diabetes (Table 1). Serum mean SD levels of IGF-I were also lower in the umbilical cord in those newborns who underwent eye examination compared with those who did not (41 23 lg/L vs. 56 24 lg/L, p < 0.0001). In total, 21 (10 girls) of the 78 investigated neonates (27%) had abnormal retinal vessel morphology, such as increased vascular tortuosity (n = 3), decreased number of vascular branching points (n = 12), increased number of vascular branching points (n = 1) and narrow arterioles and/or venules (n = 13). None of them showed signs of ROP. The 78 neonates were examined at a mean postnatal age of 7 days (ranging from 1 to 28 days). Infants showing abnormal retinal vascularisation underwent an eye examination at the same postnatal age (mean 7 days; ranging from 3 to 19 days) as those with normal retinal vascularisation (mean 7 days; ranging from 1 to 28 days). Infants with abnormal retinal vascularisation (e.g. increased vascular tortuosity, decreased/increased number of vascular branching points and/or narrow arterioles and/or venules) had significantly lower birth weight, birth length and head circumference at birth, as well as lower GA at birth (Table 2). They had also a higher percentage of SGA
Table 1 Auxological data at birth, infant data and maternal data of 153 infants born moderately to late preterm (MLP) divided into those who had an ophthalmological examination performed (study group, n = 78) and those who did not (n = 75) Auxological data at birth
All newborns n = 153
Study group n = 78
Not examined n = 75
p-Value
Gestational age (week) Birth weight (g) Birth length (cm) Head circumference (cm) Infant data Gender Girl Boy Twin No Yes SGA/AGA AGA SGA Maternal data Pre-eclampsia/hypertension No Yes Pre-eclampsia Hypertension Diabetes mellitus No Yes
35 + 4 2500 47.0 32.5
34 + 4 2250 45.0 31.8
36 + 2 2660 47.0 33.0
0.0001 0.0001 0.0001 0.0001
(32 + 2; 36 + 6) (1190; 3575) (38.0; 51.0)* (27.5; 36.5)
(32 + 2; 36 + 6) (1190; 3575) (38.0; 50.0)** (27.5; 36.5)
(32 + 2; 36 + 6) (1585; 3420) (42.0; 51.0) (29.5; 35.0)
67 (43.8%) 86 (56.2%)
34 (43.6%) 44 (56.4%)
33 (44%) 42 (56%)
1.000
120 (78.4%) 33 (21.6%)
59 (75.6%) 19 (24.4%)
61 (81.3%) 14 (18.7%)
0.51
124 (81%) 29 (19%)
57 (73.1%) 21 (26.9%)
67 (89.3%) 8 (10.7%)
0.018
114 39 34 5
57 21 19 2
57 18 15 3
(74.5%) (25.5%) (22.2%) (3.3%)
146 (95.4%) 7 (4.6%)
(73.1%) (26.9%) (24.3%) (2.6%)
73 (93.6%) 5 (6.4%)
(76%) (24%) (20%) (4%)
73 (97.3%) 2 (2.7%)
0.820
0.476
For categorical variables, n (%) is presented. For continuous variables, median (min; max) are presented. *n = 152; **n = 77; AGA, Appropriate for gestational age; SGA, Small for gestational age.
596
©2014 Foundation Acta Pædiatrica. Published by John Wiley & Sons Ltd 2014 103, pp. 594–600
Allvin et al.
Birth weight and retinal vascularisation
Table 2 Auxological data at birth of 78 infants born moderately to late preterm (MLP) divided into those with abnormal (n = 21) and normal (n = 57) retinal vascularisation Auxological variables Gestational age (week) Birth weight (g) Birth length (cm) Head circumference (cm)
Abnormal retinal vascularisation (n = 21)
Normal retinal vascularisation (n = 57)
p-Value
34 + 1 (32 + 2; 34 + 6 (32 + 35 + 3) 2; 36 + 6) 1850 (1190; 3260) 2320 (1330; 3580)
0.0039
