Early Human Development, 23 (1990) 85-91 Elsevier Scientific Publishers Ireland Ltd.
85
EHD 01074
Visual evoked potentials in term light-forgestational-age infants and infants of diabetic mothers Sten Petersena, Ole Pryds” and Werner Trojaborgb Departments of ONeonatologyand bNeurophysiology, State University Hospital (Rigshospitalet), DK-2100 Copenhagen #(Denmark) (Received 20 January 1990; revision received 15 April 1990; accepted 30 April 1990)
Summary The latency and amplitude of the first negative peak of visual evoked potentials (VEP) were evaluated in 52 term infants, investigated within 48 h after birth. Sixteen were light-for-gestational-age (LGA), 16 were appropriate-for-gestational-age (AGA) and 20 were infants of diabetic mothers (IDM). The VEP latency was shorter in LGA infants compared to AGA infants, and it was closely related to the birth weight deviation. The VEP latency was inversely related to gestational age and positively related to head circumference. When corrected for gestational age and head circumference, the VEP latency was not significantly different between the subgroups, nor related to the birth weight deviation, ponderal index or skinfold thickness. Thus, it could be argued that the high conduction velocity in LGA infants is due to stress maturation or alternatively due to the smaller head circumference. The VEP amplitude was higher in LGA infants when compared with AGA infants, and inversely related to the birth weight deviation. No differences were found in VEP latency or VEP amplitude between IDM and AGA infants. visual evoked potentials; light-for-gestational-age;
infants of diabetic mothers.
Introduction Fetal malnutrition severely affects fetal body composition, maturation and development. Birth weight (BW) below the 10th percentile (light-for-gestational age, LGA) is associated with poor postnatal growth and development and an increased risk of neurological sequelae [1,2]. In term LGA infants total body fat is approximately 15% of that measured in infants with birth weights between 10th and 90th percentile (appropriate-for-gestational age, AGA) [3]. Severe malnutrition may influence myelination within the central nervous system and thereby affect impulse transmission. 0378-3782/90/$03.50 0 1990 Elsevier Scientific Publishers Ireland Ltd. Published and Printed in Ireland
Changes of the visual evoked potentials (VEP) reflect the ongoing maturation of the brain. With increasing gestational age the VEP latency decreases due to increased myelination and increased diameter of the conducting fibers. The VEP configuration also changes with maturity, the appearance of an initial positive peak (Pl) indicating the development of synaptic formations in the visual cortex 14-61. Delayed early fetal growth and neurological development has been observed in diabetic pregnancy [7,8], and coincided with subsequent impaired neurological function of the infants [9]. The aim of this study was to evaluate the VEP in newborn term infants and to investigate whether the VEP parameters in LGA infants, AGA infants and infants of diabetic mothers (IDM), were of value as indicators of brain maturation. Material and Methods Fifty-two infants with gestational age ranging from 37 to 41 weeks were examined within 48 h after birth. The infants had not experienced perinatal asphyxia or postnatal hypoglycaemia, and all were healthy at the time of the study. According to our reference material of normal fetal growth [lo], 16 were LGA and 16 were AGA. A further 20 infants were IDM, of whom five belonged to group A, according to the White classification, four to group B, four to group C, six to group D and one to group F. Clinical data from LGA, AGA and IDM infants are presented in Table I. Gestational age (GA) was verified from an accurate menstrual history and confirmed in 31 cases by ultrasonographic assessment of the biparietal diameter before week 20 of gestation. In seven cases GA was based on an early ultrasound examination alone because of unreliable menstrual history. The newborn infant’s crown-heel length was measured by means of a Harpenden infantometer, and head circumference (greatest fronto-occipital circumference) with a calibrated measuring tape. Skinfolds were measured using a Harpenden skinfold caliper at five sites (quadriceps, pectoralis, biceps, subscapular and triceps) [ 111, and skin-fold thickness (ST) expressed as the sum of these five measurements. Ponderal index (PI) was calculated as birth weight (gram) x lOO/crown-heel length (cm)3. VEP was recorded by means of a Neuromatic 2000 (Dantec, Denmark). Three disc electrodes were placed on the head at C3 and Oz with a frontal ground according to the lo-20 international EEG system. The light source was placed about 10 cm in front of the infants eyes. Normal background illumination was reduced to 20 lux. The duration of the flash was 1.6 ps and the peak energy per flash was 1.4 klux. Ten VEPs were recorded with about 30-s intervals, and the averaged signal was used for the analysis. Another series of 10 VEPs served for comparison. All VEPs were recorded while the infant was sleeping quietly with closed eyes; care was taken not to disturb the infant during the investigation. The VEP consisted of an initial negative peak (Nl), followed by a positive deflection and then a long negative wave (Fig. 1). In most infants an initial positive peak (Pl) was also present. The VEP configuration was graded according to whether Pl was present or not. The latency and amplitude of Pl and Nl were measured by means of a ruler.
TABLE 1 Clinical data in 52 term infants.
Gestational age (days) Mean SD. Birth weight (g) Mean SD. Birth weight deviation (SD.-scores) Mean S.D. Crown-heel length (mm) Mean S.D. Head circumference (mm) Mean S.D. Abdominal circumference (mm) Mean S.D. Skinfold thickness (mm) Mean S.D. Ponderal index Mean S.D. VEP latency (ms) Mean S.D. VEP amplitude (nV) Mean S.D.
LGA n = 16
AGA n = 16
IDM n = 20
272 7.1
215 8.9
272 9.0
2029 225 -3.3 0.5
3398 414 0.1 0.8
3907 577 3.2 1.8
P ANOVA
n.s.
< 0.05
< 0.05
4U.b 20
5w
515h 26
< 0.05
16
3lP.h 12
346” 13
352b 14
< 0.05
262 17
327 13
339 20
< 0.05
13.0 3.1 2.30 0.17
21.2 3.6 2.64 0.13
26.8 5.2 2.85 0.24
< 0.05
< 0.05
235”,b 17
252” 19
26oh 17
< 0.05
38”.b
22’ 17
22b 13
< 0.05
27
a or b indicates that significant differences are restricted to pairs of subgroups identified by the StudentNeuman-Keuls test.
The study was approved by the Ethics Committee of Greater Copenhagen, and parental informed consent was obtained for each infant before enrollment in the study. Statistical methods Based on previous observations [ 121it was necessary to study at least 15 infants in each group to detect a significant difference in VEP latency of 15 ms (alpha 0.05, beta 0.10). Because the VEP latency decreases with gestational age a difference of 15 ms may be comparable with a difference in gestational age of about one month.
88
JJV -2oo-
0
N1
0
I 0.5
I 1 Sec.
Fig. 1. Visual evoked potential of a LGA infant with gestational age 38 weeks and birth weight 2150 g.
Fulfillment of the null hypotheses requires that the VEP configuration and the VEP latency are identical in LGA, IDM and AGA infants. Because the VEP latency depends on GA and the transmission distance from the retina to the visual cortex, GA and head circumference were included in the analysis. Differences between groups were tested using ANOVA including the StudentNewman-Keuls test for comparison between groups (SPSS/PC + ). Dichotomous variables were tested with the Fisher’s exact test. Relations between variables were tested with univariate and bivariate linear regression (SPSWPC + ). Results The VEP configuration vs. LGA, IDM and AGA infants Pl was present in 15 of 16 LGA infants, in 17 of 20 IDM infants and in 12 of 16 AGA infants. These differences were not significant (P > 0.05). Nl was present in all cases. Because the latency of Pl was very closely related to the latency of Nl (P < O.OOOl),only the latter was used for further analysis. The VEP latency vs. LGA, IDM and AGA infants The latency of Nl was significantly shorter in LGA infants compared to AGA infants and IDM, whereas there was no difference between the two latter groups (Table I). Thus, the latency was closely related to the birth weight deviation (P = 0.0005; Fig. 2). The VEP latency was, however, inversely related to the gestational age within the actual range (37-41 weeks) (P = 0.015) and positively related to the head circumference (P = 0.038). These two variables achieved an even higher statistical relation to the latency when included in a bivariate analysis (P = 0.003 and P = 0.0014, respectively). The residuals from this bivariate regression analysis of VEP latency by GA and head circumference were not significantly different between the three groups nor related to the birth weight deviation, ponderal index or skinfold thickness.
290 0 280 270
* *
$ 280 E - 250 E
f3 0 *
f 240 Cr a 230
**lb* w * -4
t
tm
0
**
210
t
4
* Y
220
200 t -8
i
0
3
9
tt t1ttitt
89
L2
$ 0
0
Birth weight deviation
t
it
to
2
4
I 8
(SD-scores)
Fig. 2. VEP latency versus birth weight deviation in 16 LGA infants (*), 16 AGA infants (0) and 20 IDM (+).
The VEP amplitude
vs. LGA, IDA4 and AGA infants
The amplitude of Nl was significantly higher in LGA infants when compared to AGA’infants and IDM (P = 0.03). There was no difference between the amplitude in IDM and AGA infants (Table I). Thus, the VEP amplitude was inversely related to the birth weight deviation (P = 0.027; Fig. 3). Higher VEP amplitudes were recorded in infants with small heads (P = 0.011) without any relation to the gestational age (P = 0.094). Discussion Neurophysiological studies of growth retarded newborn infants demonstrated that peripheral nerve conduction velocities were not significantly different from those recorded in normal term infants of the same gestational age [13]. Soares et al. [14] compared brainstem auditory evoked potentials from LGA infants with those from AGA controls matched for gestational age (30-41 weeks) and concluded that the responses found in the LGA infants could be explained by immaturity of the basal end of the cochlea. In the present study, shorter VEP latencies were recorded in LGA infants compared to AGA infants and IDM. The higher conduction velocity in LGA infants may be explained by a stress-induced maturation. Alternatively, it could be argued that the shorter VEP latency was due to the smaller head circumference in LGA infants. The identical VEP configuration in all three groups support the assumption
90
90 80
* * *
70 9 560
*
t* #* * 10
*
*** *
01
-6
-4
-2
I
0
2
4
I
6
Bkth wdght devlatlon (SD-scores) Fig. 3. VEP amplitude versus birth weight deviation in 16 LGA infants (*), 16 AGA infants (0) and 20 IDM(+).
of an almost identical degree of maturation. Recently, a magnetic resonance spectroscopy study substantiated the uniformity of cerebral maturation in LGA and AGA infants [ 151. The higher VEP amplitude in LGA infants may be due to a minor amount of subcutaneous tissue, implying less attenuation of the signal. It should be noted that our LGA infants were asymmetrically growth retarded and may therefore be considered to have developed normally until a few weeks before birth. Accordingly, a short period of substrate lack does not seem to affect the myelination markedly. Alternatively, minor alterations in the constitution of myelin may not alter the axonal conductivity. Symmetrically growth retarded infants, on the other hand, may have a high risk of delayed maturation since central nervous myelination commences at approximately the 28th week of gestation. The VEP parameters did not differ between IDM and AGA infants, probably as a result of the improved monitoring and treatment of the diabetic mother, even though several of these diabetic mothers had diabetic complications. References Ounsted, M.K., Moar, V.A. and Scott, A. (1984): Children of deviant birthweight at the age of seven years: health, handicap, size and developmental status. Early Hum. Dev., 9,323-340. Rantakallio, P. (1985): A 14-year follow-up of children with normal and abnormal birth weight for their gestational age. Acta. Paediatr. Stand., 74.62-69. Petersen, S.. Gotfredsen, A. and Knudsen, F.U. (1988): Lean body mass in smah for gestational age and appropriate for gestational age infants. J. Pediatr.. 113,886-889.
91 Ellingson, R. (1960): Cortical electrical responses to visual stimulation in the human infant. Electroencephalogr. Clin. Neurophysiol., 12,663-677. Hrbek, A., Karlberg, P. and Olsson, T. (1973): Development of visual and somatosensory evoked responses in preterm newborn infants. Electroencephalogr. Clin. Neurophysiol., 34,225-232. Kurtzberg, D., Vaughan, H.G., Courchesne, E. et al. (1984): Developmental aspects of event related potentials. Ann. N.Y. Acad. Sci., 425,300-398. Pedersen, J.F. and Melsted-Pedersen, L. (1979): Early growth retardation in diabetic pregnancy. Brit. Med. J., i, 18-19. Visser, G.H.A., Bekedam, D.J., Mulder, E.J.H. and van Ballegooie, E. (1985): Delayed emergence of fetal behaviour in type-l diabetic women. Early Hum. Dev., 12, 167-172. Bloch Petersen, M., Pedersen, S.A., Greisen, G., Pedersen, J.F. and Mdlsted-Pedersen, L. (1988): Early growth delay in diabetic pregnancy: relation to psychomotor development at age 4. Brit. Med. J., 2%,598-600. 10 Larsen, T., Petersen, S., Greisen, G. and Falck Larsen, J. (19%): Normal fetal growth evaluated by longitudinal ultrasound measurements. Early Hum. Dev., in press. 11 Weile, B., Bach-Mortensen, N. and Peitersen, B. (1986): Caliper skinfold measurements in newborns: Analysis of a method. Biol. Neonate, 50, 192-199. 12 Pryds, O., Trojaborg, W., Carlsen, J. and Jensen, J. (1989): Determinants of visual evoked potentials in preterm infants. Early Hum. Dev., 19, 117-125. 13 Schulte, F.J., Michaelis, R., Linke, I. and Nolte, R. (1968): Motor nerve conduction velocity in term, preterm, and small-for-dates newborn infants. Pediatrics, 42, 17-26. 14 Soares, I., Collet, L., Morgon, A. and Salle, B. (1988): Effect of brainstem auditory evoked potential stimulus intensity variations in neonates of small for gestational age. Brain Dev., 10, 174-177. 15 Azzopardi, A., Wyatt, J.S., Hamilton, P.A., Cady, E.B., Delpy, D.T., Hope, P.L. and Reynolds, E.O.R. (1989): Phophorus metabolites and intracellular pH of normal and small for gestational age infants investigated by magnetic resonance spectroscopy. Pediatr. Res., 25,440~444.
85
EHD 01074
Visual evoked potentials in term light-forgestational-age infants and infants of diabetic mothers Sten Petersena, Ole Pryds” and Werner Trojaborgb Departments of ONeonatologyand bNeurophysiology, State University Hospital (Rigshospitalet), DK-2100 Copenhagen #(Denmark) (Received 20 January 1990; revision received 15 April 1990; accepted 30 April 1990)
Summary The latency and amplitude of the first negative peak of visual evoked potentials (VEP) were evaluated in 52 term infants, investigated within 48 h after birth. Sixteen were light-for-gestational-age (LGA), 16 were appropriate-for-gestational-age (AGA) and 20 were infants of diabetic mothers (IDM). The VEP latency was shorter in LGA infants compared to AGA infants, and it was closely related to the birth weight deviation. The VEP latency was inversely related to gestational age and positively related to head circumference. When corrected for gestational age and head circumference, the VEP latency was not significantly different between the subgroups, nor related to the birth weight deviation, ponderal index or skinfold thickness. Thus, it could be argued that the high conduction velocity in LGA infants is due to stress maturation or alternatively due to the smaller head circumference. The VEP amplitude was higher in LGA infants when compared with AGA infants, and inversely related to the birth weight deviation. No differences were found in VEP latency or VEP amplitude between IDM and AGA infants. visual evoked potentials; light-for-gestational-age;
infants of diabetic mothers.
Introduction Fetal malnutrition severely affects fetal body composition, maturation and development. Birth weight (BW) below the 10th percentile (light-for-gestational age, LGA) is associated with poor postnatal growth and development and an increased risk of neurological sequelae [1,2]. In term LGA infants total body fat is approximately 15% of that measured in infants with birth weights between 10th and 90th percentile (appropriate-for-gestational age, AGA) [3]. Severe malnutrition may influence myelination within the central nervous system and thereby affect impulse transmission. 0378-3782/90/$03.50 0 1990 Elsevier Scientific Publishers Ireland Ltd. Published and Printed in Ireland
Changes of the visual evoked potentials (VEP) reflect the ongoing maturation of the brain. With increasing gestational age the VEP latency decreases due to increased myelination and increased diameter of the conducting fibers. The VEP configuration also changes with maturity, the appearance of an initial positive peak (Pl) indicating the development of synaptic formations in the visual cortex 14-61. Delayed early fetal growth and neurological development has been observed in diabetic pregnancy [7,8], and coincided with subsequent impaired neurological function of the infants [9]. The aim of this study was to evaluate the VEP in newborn term infants and to investigate whether the VEP parameters in LGA infants, AGA infants and infants of diabetic mothers (IDM), were of value as indicators of brain maturation. Material and Methods Fifty-two infants with gestational age ranging from 37 to 41 weeks were examined within 48 h after birth. The infants had not experienced perinatal asphyxia or postnatal hypoglycaemia, and all were healthy at the time of the study. According to our reference material of normal fetal growth [lo], 16 were LGA and 16 were AGA. A further 20 infants were IDM, of whom five belonged to group A, according to the White classification, four to group B, four to group C, six to group D and one to group F. Clinical data from LGA, AGA and IDM infants are presented in Table I. Gestational age (GA) was verified from an accurate menstrual history and confirmed in 31 cases by ultrasonographic assessment of the biparietal diameter before week 20 of gestation. In seven cases GA was based on an early ultrasound examination alone because of unreliable menstrual history. The newborn infant’s crown-heel length was measured by means of a Harpenden infantometer, and head circumference (greatest fronto-occipital circumference) with a calibrated measuring tape. Skinfolds were measured using a Harpenden skinfold caliper at five sites (quadriceps, pectoralis, biceps, subscapular and triceps) [ 111, and skin-fold thickness (ST) expressed as the sum of these five measurements. Ponderal index (PI) was calculated as birth weight (gram) x lOO/crown-heel length (cm)3. VEP was recorded by means of a Neuromatic 2000 (Dantec, Denmark). Three disc electrodes were placed on the head at C3 and Oz with a frontal ground according to the lo-20 international EEG system. The light source was placed about 10 cm in front of the infants eyes. Normal background illumination was reduced to 20 lux. The duration of the flash was 1.6 ps and the peak energy per flash was 1.4 klux. Ten VEPs were recorded with about 30-s intervals, and the averaged signal was used for the analysis. Another series of 10 VEPs served for comparison. All VEPs were recorded while the infant was sleeping quietly with closed eyes; care was taken not to disturb the infant during the investigation. The VEP consisted of an initial negative peak (Nl), followed by a positive deflection and then a long negative wave (Fig. 1). In most infants an initial positive peak (Pl) was also present. The VEP configuration was graded according to whether Pl was present or not. The latency and amplitude of Pl and Nl were measured by means of a ruler.
TABLE 1 Clinical data in 52 term infants.
Gestational age (days) Mean SD. Birth weight (g) Mean SD. Birth weight deviation (SD.-scores) Mean S.D. Crown-heel length (mm) Mean S.D. Head circumference (mm) Mean S.D. Abdominal circumference (mm) Mean S.D. Skinfold thickness (mm) Mean S.D. Ponderal index Mean S.D. VEP latency (ms) Mean S.D. VEP amplitude (nV) Mean S.D.
LGA n = 16
AGA n = 16
IDM n = 20
272 7.1
215 8.9
272 9.0
2029 225 -3.3 0.5
3398 414 0.1 0.8
3907 577 3.2 1.8
P ANOVA
n.s.
< 0.05
< 0.05
4U.b 20
5w
515h 26
< 0.05
16
3lP.h 12
346” 13
352b 14
< 0.05
262 17
327 13
339 20
< 0.05
13.0 3.1 2.30 0.17
21.2 3.6 2.64 0.13
26.8 5.2 2.85 0.24
< 0.05
< 0.05
235”,b 17
252” 19
26oh 17
< 0.05
38”.b
22’ 17
22b 13
< 0.05
27
a or b indicates that significant differences are restricted to pairs of subgroups identified by the StudentNeuman-Keuls test.
The study was approved by the Ethics Committee of Greater Copenhagen, and parental informed consent was obtained for each infant before enrollment in the study. Statistical methods Based on previous observations [ 121it was necessary to study at least 15 infants in each group to detect a significant difference in VEP latency of 15 ms (alpha 0.05, beta 0.10). Because the VEP latency decreases with gestational age a difference of 15 ms may be comparable with a difference in gestational age of about one month.
88
JJV -2oo-
0
N1
0
I 0.5
I 1 Sec.
Fig. 1. Visual evoked potential of a LGA infant with gestational age 38 weeks and birth weight 2150 g.
Fulfillment of the null hypotheses requires that the VEP configuration and the VEP latency are identical in LGA, IDM and AGA infants. Because the VEP latency depends on GA and the transmission distance from the retina to the visual cortex, GA and head circumference were included in the analysis. Differences between groups were tested using ANOVA including the StudentNewman-Keuls test for comparison between groups (SPSS/PC + ). Dichotomous variables were tested with the Fisher’s exact test. Relations between variables were tested with univariate and bivariate linear regression (SPSWPC + ). Results The VEP configuration vs. LGA, IDM and AGA infants Pl was present in 15 of 16 LGA infants, in 17 of 20 IDM infants and in 12 of 16 AGA infants. These differences were not significant (P > 0.05). Nl was present in all cases. Because the latency of Pl was very closely related to the latency of Nl (P < O.OOOl),only the latter was used for further analysis. The VEP latency vs. LGA, IDM and AGA infants The latency of Nl was significantly shorter in LGA infants compared to AGA infants and IDM, whereas there was no difference between the two latter groups (Table I). Thus, the latency was closely related to the birth weight deviation (P = 0.0005; Fig. 2). The VEP latency was, however, inversely related to the gestational age within the actual range (37-41 weeks) (P = 0.015) and positively related to the head circumference (P = 0.038). These two variables achieved an even higher statistical relation to the latency when included in a bivariate analysis (P = 0.003 and P = 0.0014, respectively). The residuals from this bivariate regression analysis of VEP latency by GA and head circumference were not significantly different between the three groups nor related to the birth weight deviation, ponderal index or skinfold thickness.
290 0 280 270
* *
$ 280 E - 250 E
f3 0 *
f 240 Cr a 230
**lb* w * -4
t
tm
0
**
210
t
4
* Y
220
200 t -8
i
0
3
9
tt t1ttitt
89
L2
$ 0
0
Birth weight deviation
t
it
to
2
4
I 8
(SD-scores)
Fig. 2. VEP latency versus birth weight deviation in 16 LGA infants (*), 16 AGA infants (0) and 20 IDM (+).
The VEP amplitude
vs. LGA, IDA4 and AGA infants
The amplitude of Nl was significantly higher in LGA infants when compared to AGA’infants and IDM (P = 0.03). There was no difference between the amplitude in IDM and AGA infants (Table I). Thus, the VEP amplitude was inversely related to the birth weight deviation (P = 0.027; Fig. 3). Higher VEP amplitudes were recorded in infants with small heads (P = 0.011) without any relation to the gestational age (P = 0.094). Discussion Neurophysiological studies of growth retarded newborn infants demonstrated that peripheral nerve conduction velocities were not significantly different from those recorded in normal term infants of the same gestational age [13]. Soares et al. [14] compared brainstem auditory evoked potentials from LGA infants with those from AGA controls matched for gestational age (30-41 weeks) and concluded that the responses found in the LGA infants could be explained by immaturity of the basal end of the cochlea. In the present study, shorter VEP latencies were recorded in LGA infants compared to AGA infants and IDM. The higher conduction velocity in LGA infants may be explained by a stress-induced maturation. Alternatively, it could be argued that the shorter VEP latency was due to the smaller head circumference in LGA infants. The identical VEP configuration in all three groups support the assumption
90
90 80
* * *
70 9 560
*
t* #* * 10
*
*** *
01
-6
-4
-2
I
0
2
4
I
6
Bkth wdght devlatlon (SD-scores) Fig. 3. VEP amplitude versus birth weight deviation in 16 LGA infants (*), 16 AGA infants (0) and 20 IDM(+).
of an almost identical degree of maturation. Recently, a magnetic resonance spectroscopy study substantiated the uniformity of cerebral maturation in LGA and AGA infants [ 151. The higher VEP amplitude in LGA infants may be due to a minor amount of subcutaneous tissue, implying less attenuation of the signal. It should be noted that our LGA infants were asymmetrically growth retarded and may therefore be considered to have developed normally until a few weeks before birth. Accordingly, a short period of substrate lack does not seem to affect the myelination markedly. Alternatively, minor alterations in the constitution of myelin may not alter the axonal conductivity. Symmetrically growth retarded infants, on the other hand, may have a high risk of delayed maturation since central nervous myelination commences at approximately the 28th week of gestation. The VEP parameters did not differ between IDM and AGA infants, probably as a result of the improved monitoring and treatment of the diabetic mother, even though several of these diabetic mothers had diabetic complications. References Ounsted, M.K., Moar, V.A. and Scott, A. (1984): Children of deviant birthweight at the age of seven years: health, handicap, size and developmental status. Early Hum. Dev., 9,323-340. Rantakallio, P. (1985): A 14-year follow-up of children with normal and abnormal birth weight for their gestational age. Acta. Paediatr. Stand., 74.62-69. Petersen, S.. Gotfredsen, A. and Knudsen, F.U. (1988): Lean body mass in smah for gestational age and appropriate for gestational age infants. J. Pediatr.. 113,886-889.
91 Ellingson, R. (1960): Cortical electrical responses to visual stimulation in the human infant. Electroencephalogr. Clin. Neurophysiol., 12,663-677. Hrbek, A., Karlberg, P. and Olsson, T. (1973): Development of visual and somatosensory evoked responses in preterm newborn infants. Electroencephalogr. Clin. Neurophysiol., 34,225-232. Kurtzberg, D., Vaughan, H.G., Courchesne, E. et al. (1984): Developmental aspects of event related potentials. Ann. N.Y. Acad. Sci., 425,300-398. Pedersen, J.F. and Melsted-Pedersen, L. (1979): Early growth retardation in diabetic pregnancy. Brit. Med. J., i, 18-19. Visser, G.H.A., Bekedam, D.J., Mulder, E.J.H. and van Ballegooie, E. (1985): Delayed emergence of fetal behaviour in type-l diabetic women. Early Hum. Dev., 12, 167-172. Bloch Petersen, M., Pedersen, S.A., Greisen, G., Pedersen, J.F. and Mdlsted-Pedersen, L. (1988): Early growth delay in diabetic pregnancy: relation to psychomotor development at age 4. Brit. Med. J., 2%,598-600. 10 Larsen, T., Petersen, S., Greisen, G. and Falck Larsen, J. (19%): Normal fetal growth evaluated by longitudinal ultrasound measurements. Early Hum. Dev., in press. 11 Weile, B., Bach-Mortensen, N. and Peitersen, B. (1986): Caliper skinfold measurements in newborns: Analysis of a method. Biol. Neonate, 50, 192-199. 12 Pryds, O., Trojaborg, W., Carlsen, J. and Jensen, J. (1989): Determinants of visual evoked potentials in preterm infants. Early Hum. Dev., 19, 117-125. 13 Schulte, F.J., Michaelis, R., Linke, I. and Nolte, R. (1968): Motor nerve conduction velocity in term, preterm, and small-for-dates newborn infants. Pediatrics, 42, 17-26. 14 Soares, I., Collet, L., Morgon, A. and Salle, B. (1988): Effect of brainstem auditory evoked potential stimulus intensity variations in neonates of small for gestational age. Brain Dev., 10, 174-177. 15 Azzopardi, A., Wyatt, J.S., Hamilton, P.A., Cady, E.B., Delpy, D.T., Hope, P.L. and Reynolds, E.O.R. (1989): Phophorus metabolites and intracellular pH of normal and small for gestational age infants investigated by magnetic resonance spectroscopy. Pediatr. Res., 25,440~444.
