Effects of preterm extrauterine visual experience on the development of the human visual system: a flash VEP study Syuichi Tsuneishi MD PhD, Department of Paediatrics, Kobe University School of Medicine, Kobe, Japan; Paul Casaer* MD PhD FRCP (Edin), Division of Pediatric Neurology and Developmental Neurology Research Unit, Department of Pediatrics, University Hospital Gasthuisberg, Leuven, Belgium.
An increasing number of preterm infants survive without neurological sequelae. This fact gives us new opportunities to study the underlying mechanism of neurological development. Preterm birth exposes infants earlier to extrauterine life and to a variety of stimuli that are not present in utero. The question posed is whether these unexpected environmental conditions influence neurological maturation (Touwen 1980). Flash visual evoked potentials (VEPs) are easy to record even in preterm infants. There are, however, difficulties with the clinical application of VEPs due to a lack of agreed standard methodology and normative data. Peak latencies of VEPs have been considered to be less significant than VEP wave-form changes. Recently, Tsuneishi et al. (1995) reported normative data for preterm infants with good reproducibility by taking the two components (N1a, early peak; N1b, late peak) of the N1 wave into account. Using VEPs intra- and extrauterine maturational courses of the human visual system can be compared using cross-sectional and longitudinal approaches. Methods PARTICIPANTS
*Correspondence to second author at Division of Pediatric Neurology and Developmental Neurology Research Unit, Department of Pediatrics, University Hospital Gasthuisberg, Herestraat 49, B-3000 Leuven, Belgium E-mail:
[email protected]
Preterm infants (gestational age, GA: 24 to 36 weeks) admitted to the University Hospital Gasthuisberg, Belgium, were included in this study and parental informed consent was obtained before testing. CROSS - SECTIONAL STUDY
To compare the functional maturation of the human visual system between intra- and extrauterine course flash visual evoked potentials (VEPs) in preterm infants (gestational age 24 to 36 weeks). Previously established normal values, with special reference to the two components of the N1 wave, were employed (Tsuneishi 1995). A cross-sectional analysis of 124 infants at 36 weeks postmenstrual age (PMA), showed that there are no differences in the absolute values of VEP peak latencies depending on the postnatal age (PNA). Conversely, the N1 wave form changes with increasing PNA from a wave in which the early peak (N1a) has a higher amplitude than the late peak (N1b) into the reverse situation with a higher amplitude of the N1b as compared to N1a. This observation may correlate with the maturation of the neuronal networks in the visual cortex. In a longitudinal analysis of 50 infants followed for more than 5 sessions of weekly recordings, we found that the individual rapid decrease in the N1a latency, which may reflect the initiation of myelination in the optic radiation, most frequently occurs at around 37 weeks PMA, regardless of PNA. Preterm extrauterine visual experience has little effect on the myelination process in the visual pathway, but has a marked effect on the developmental changes in VEP wave form which reflect the developmental changes of the neuronal networks in the visual cortex.
One-hundred and twenty-four healthy preterm infants of appropriate birthweights for GA were studied at their postmenstrual age (PMA) of 36 weeks ± 5 days. The GA was assessed by mothers’ menstrual dates, ultrasound examination, and postnatal clinical assessment. Infants with neurological problems or anomalies, perinatal infectious diseases, and other severe systemic conditions were excluded. All infants included were scanned with a 7.5 MHz transducer at least within the first 3 days after birth and afterwards weekly. Infants who had abnormal echo density, suggesting periventricular/intraventricular hemorrhage or periventricular leukomalacia were excluded. The infants in the study were neither on respiratory support nor under phototherapy. The infants did not receive medication that might influence the nervous system. The 124 infants were divided into four subgroups according to their postnatal ages (PNAs) at recording: within 7 days, 8 to 21 days 22 to 35 days, and more than 36 days. The absolute values of peak latencies and amplitudes, and VEP wave configuration were compared among subgroups. LONGITUDINAL STUDY
Fifty healthy preterm infants of appropriate birthweights for GA, who were followed weekly for more than 5 sessions (weeks), were included in the longitudinal part of this study. All 50 infants fulfilled the clinical criteria described above. Participants were divided into two subgroups according to their GAs at birth: less than 31 weeks or 31 weeks or more. The weekly decreasing rates of N1a latency were analysed intraindividually and the occurrence and timing of the weeks in which there was a large decrease in latencies was studied and compared between the two subgroups.
Developmental Medicine & Child Neurology 2000, 42: 663–668
663
VEP TECHNIQUE
The technique of VEP recording and the principle of interpretation used in this study have been described in detail by Tsuneishi et al. (1995). In brief, we used a Neuropack 2 (Nihon Kohden, Tokyo, Japan) with three dish electrodes: the active electrode at inion, the reference at Fz, and the ground at the ear lobe. The interelectrode impedance was kept at less than 5 KΩ. Flash lights were emitted from a light-emitting diode plate (10 ×10 cm) placed about 15 cm in front of the infant’s eyes. It gave red light stimulation with a duration of 10 ms and a mean level of energy of more than 100 cd/M2. Stimulating frequency was set at 0.2 Hz, the sweep time at 1 s and 50 responses were averaged with a bandpass of 1 to 100 Hz. Each recording session was performed in natural sleep with more than two trials to check
reproducibility. The most prominent negative wave arising at around 300 ms is defined as N1, in which we found two components: N1a (early peak) and N1b (late peak). The positive peak emerging before N1a with a mean latency of 200 ms is defined as P1. The next positive peak after the N1 complex is P2 and then N2 (Fig. 1). Peak latencies and amplitudes of N1a, N1b, P2, and N2 have been previously reported by Tsuneishi et al. (1995) Changes in wave configuration were also noted. The detectability of peak P1, which had more than 2 µV in amplitude, and the maturational changes of the N1 wave form, which were assessed from the amplitudes of the two components of the N1 wave (N1a and N1b), were also considered. In the longitudinal study a week, between the two exami-
Table I: Clinical characteristics of infants in cross-sectional study
N2 N1a
1–7 Mean (SD)
N1b
5 days PNA (GA: 35W6D)
Total nr of data 31 Sex m/f 15/16 GA (wk) 35.3 (0.7) At recording: PNA (d) 3.9 (2.3) PMA (wk) 35.9 (0.5) BW (g) 2136 (235) HC (cm) 32.2 (1.1)
P2 P1 N1a
21 days PNA (GA: 33W0D)
Postnatal age (d) 8–21 22–35 ≥ 36 Mean (SD) Mean (SD) Mean (SD) 33 17/16 33.8 (0.6) 15.0 (3.9) 36.0 (0.4) 2108 (208) 32.3 (0.8)
32 14/18 31.9 (0.7)
28 12/16 29.9 (1.4)
28.3 (3.4) 43.9 (9.9) 36.0 (0.5) 36.1 (0.4) 2120 (288) 2149 (189) 32.7 (1.4) 32.8 (1.0)
PNA, postnatal age; PMA, postmenstrual age; GA, gestational age; BW, bodyweight; HC, head circumference. There are no statistically significant differences among clinical characteristics except for GA and PNA at the time of the recording.
N1b
33 days PNA (GA: 31W4D)
Table II: Peak latencies of VEP waves Latency 1–7
N1b N1a N1b P2 N2
N1a
60 days PNA (GA: 27W1D)
263 (11) 354 (12) 409 (21) 501 (40)
Postnatal age (d) 8–21 22–35 261 (13)a 340 (17)a 399 (26)a 491 (31)a
262 (11)a 341 (18)a 403 (21)a 493 (26)a
≥ 36aa 259 (10)a 338 (18)a 406 (19)a 495 (31)a
PNA, postnatal age, a p
