Vision Res. Vol. 31, No. IO, pp. 1837-1840, 1991 Printed in Great Britain. All rights reserved

0042-6989/91 $3.00 + 0.00 Copyright 0 1991 Pergamon Press plc

SHORT NOTE

TRANSMISSION OF LIGHT ACROSS THE ADULT AND

NEONATAL EYELID IN VIVO

JUDITH ROBINSON,' SUSAN C. BAYLISS' and ALISTAIR R. FIELDER’

’ Department of Ophthalmology, University of Birmingham, Birmingham and Midland Eye Hospital, Church Street, Birmingham B3 2NS and

2 Department of Physics, Leicester University, L&ester LEl 7RH, U.K.

(Received IS August 1990; in revised form 4 February 1991)

Abstract-Light transmission characteristics of the human adult and neonatal eyelid were measured in vim. Light was delivered via a grating monochromator through a fibre-optic mounted onto a contact lens placed under the eyelid, and detected using a photodiode on its external skin surface. Data from 5 adult and 9 preterm neonatal subjects indicate that the eyelid acts as a predominantly red-pass filter, with mean transmissions at 700nm of 14.5% in the adult and 21.4% in the neonate, declining to 63% in both groups below 580 nm. The relevance of this data to clinical electrophysiology and to estimates of retinal irradiance is discussed.

Light Eyelid Adult Transmission Infant-premature Photometry Human

INTRODUCTION

The infant born preterm is exposed to light at an earlier stage of development than his/her full term counterpart. Whether this early exposure may affect the immature visual system, includ- ing the maturation of acuity, or the develop- ment of retinopathy of prematurity (ROP) is of interest, but as yet unknown. A number of factors contribute to the light dose incident on the surface of the neonatal eye. These include first, the environmental illumination (Robinson, Moseley & Fielder, 1990) and second, factors which control light reaching each eyelid such as head position, shielding by clothing, eyepatches and equipment (Robinson, 1990). Third, aspects relating to the eyelid, including the frequency and duration of opening (Robinson, Moseley, Thompson & Fielder, 1989), and the amount of light reaching the surface of the eye when the eyelids are closed. The last is the topic of this article.

Light transmission through skin has been studied, mainly by transillumination (Sisson & Wickler, 1973). Only two articles refer to the eyelid; Crawford and Marc (1976) studied the cat and macaque monkey, and Moseley, Bayliss and Fielder (1988) the adult human. According to these studies eyelids act as red-pass filters, with transmission vaues at 700 nm ranging from 10% for the adult human through 20% for

macaques (M. mulatta) to 40% for the eyelid of a white cat. There is no corresponding data for the neonatal eyelid.

Here we report on in uiuo measurements of eyelid transmission in both adults and preterm neonates.

MATERIALS AND METHODS

This investigation was divided into two parts. First, a pilot study of five adults, aged 21-29 yr. Second, a neonatal study of nine preterm infants born at gestational ages (GA) ranging from 25 to 32 weeks (Table 1). Measurements were undertaken when the infant’s general condition permitted, and informed parental consent had been obtained for the neonatal study.

Table 1. Details of neonates. Infants of Indo-Pakistani origin are referred to as Asian

Gestational Postnatal Birth- age age weight

Baby (weeks) (weeks) (g) Ethnic origin

I 25

I’!!1 29 30 IV 30 V 31 VI 31 VII 32 VIII 34 IX 35

5 630 Caucasian 3 1640 Caucasian 3 1515 Afro-Caribbean 4 1270 Afro-Caribbean 3 1310 Asian 3 1870 Asian 3 1085 Caucasian 1 1460 Caucasian 1 1850 Asian

1837

1838 Short Note

The apparatus consisted of a dedicated light source, the “ophthalmic spectrophotometer” and a contact lens assembly. The “opthahnic spectrophotometer” contained a halogen light source, monochromator, chopper, power supply unit, lens and titer assembly. The halogen bulb and chopper were powered by a 12 V/5 A power supply unit. The chopper blade was a rotating disc divided into eight sectors-alternatively clear and optically opaque. An alternating cur- rent source and synchronized electrical reference signal were fed into a “lock-in” amplifier. Input signals outside the reference frequency were thus rejected, enabling measurement of low intensity light signals under normal ambient room illumination. Except for that incident on the lens-filter assembly, white light emitted from the halogen bulb was absorbed by graphite paint. The grating monochromator transmitted light of known and variable wave- length via the fibre-optic (500 pm dia.) embed- ded onto the anterior surface of an opaque PMMA contact lens. The fibre-optic tip was bevelled to direct light perpendicularly away from the contact lens and cornea and through the eyelid. Plastic fibre-optic was used in prefer-

ence to glass for safety reasons recognizing, however, the poorer transmission properties at wavelengths < 400 nm.

The subject’s anterior ocular surface was anaesthetized with benoxinate (oxybuprocaine) hydrochloride 0.4% eyedrops and the contact lens positioned under the tarsus of the upper eyelid. The detector, a silicon photodiode (RS Components-dia. 1 .O cm) operating in photo- conductive mode, attached to a foam rubber surround was held gently against the external surface of the dosed eyelid. Measurements were made from one, usually the right, eye of each subject. The study period did not exceed 5 min.

The size difference between source (dia. 500 pm) and detector (dia. 1 cm), enabled col- lection of both on and off axis transmitted light rays. Measurements were taken, after appropri- ate amplification (x lo’), at 10 nm intervals in adults (2Onm for neonates) across the visible spectrum (400-700 nm).

Signals were maximized at 700 run (I-) by moving the detector over the surface of the eyelid. Baseline measurements were obtained with the contact lens held against the cheek and detector 2 mm from the light source. Finally, a

Wavelength (nm)

Fig. 1. Transmission of light through the adult and neonatal human ayeMid. Each point rrprer#nts the me-an of data obtained from 5 adult subjects (0) or 9 neon&to8 (e). The bars indk%te the sm.

Short Note

background level was measured by removing the light source.

Each transmission value was corrected for system effects by dividing by the baseline. Spectra were normalized at 700 nm, the maxi- mum intensity of transmitted light (I,,) being obtained by scanning of the eyelid with the detector. No second order spectra were detected since the fibre and detector windows cut off wavelengths below 350 nm. Analysis of variance was used to compare the eyelid transmission of the adult and neonatal subjects.

Ethical permission for these studies were granted by the Leicestershire and West Birming- ham Health Authorities.

RESULTS

The human eyelid acts as a predominantly red-pass filter in both the adult and neonate (Fig. 1). At 700 nm, mean transmission in adults (mean 14.5%, range 13-19%) was significantly less than for neonates (mean 21.4%, range 20-34%) (P < 0.01). Transmission declined to < 3% at wavelengths Q 580 nm for both groups.

DISCUSSION

The eyelid of both adults and preterm neonates acts as a predominantly red-pas filter (Fig. 1). Data obtained from adults was broadly similar to that of Moseley et al. (1988). Compared with the macaque monkey eyelid (Crawford & Marc, 1976) our results, for adults, at 700 nm are m 5-10% less. This could be due to eyelid stretching caused by the positioning of the sensing probe in the former study. Alter- natively, the variation may represent species differences of eyelid composition.

The greater absolute transmission for the adult human eyelid at each wavelength com- pared with Moseley et al. (1988) may be ex- plained by differences in the measurement techniques. The geometry of the present exper- iment was as follows-the detector (dia. 10 mm) rested directly on the eyelid (thickness < 2 mm) under which the fibre-optic source (dia. 0.5 mm) also directly contacted. Assuming that the eyelid is a perfect infinitely thin diffuser sitting at the source then using Lambert’s Law it is possible to estimate that 0.86 of the light was collected. This approximation for the system gives the least bound for the fraction of light collected. The adult eyelid, however, is not a perfect diffuser and less “small-angle-scattered” inten-

sity is expected (by calculation of the geometry of the system) than is found experimentally. The apparatus used by Moseley et al. (1988) required precise alignment of source (dia. I cm) and detector (dia. 500 pm) to obtain a maximum signal. In addition, in the current study the sample and reference signals were approx. 200 times greater, enabling measurements to be made with good signal to noise ratio in illuminated rooms.

Greater “absolute” intensity of transmitted light across the neonatal compared with the adult eyelid may be due to the thinner eyelids of preterm neonates, but this parameter is difficult to measure and was not attempted in this study. But, reduction in percentage light transmission through tissue, with increasing source-to-detector distance, has been reported (e.g. Sisson & Wickler, 1973; Weaver & Reppert, 1989). We did not detect any signifi- cant racial differences in adults (5 Caucasian, 1 Asian) or neonates (4 Caucasian, 3 Asian and 2 Afro-Caribbean), in agreement with Moseley et al. (1988) who studied three subjects (1 Asian, 1 Caucasian and 1 Oriental).

Our measurements were made using mono- chromatic light over the visible region, which is however a relatively narrow wavelength range. To estimate the total amount of white light transmitted through the eyelid integration, between 400 and 700 nm, was performed. Thus the neonatal eyelid transmits 38% of white light incident on its outer surface. A greater pen- etrance of broad spectrum white (flourescent) light through the abdominal wall of the Wistar rat was noted by Sisson and Wickler (1973) when compared with the penetrance of narrow spectrum blue light.

This investigation has examined the light transmission characteristics of the human eye- lid. The relevance of these findings can be considered under two headings: ocular electro- physiology and the neonatal ocular light dose. For visual pathway electrophysiology the white light stimulus may be required to pass across the closed eyelid, as in certain disease states (Jaffe, Caruso, de Monasterio & Nussenblatt, 1987), and in neonates. Clearly, both light inten- sity and spectral composition are consequen- tially altered inducing selective photoreceptor stimulation (Jaffe et al., 1987) and affecting response amplitude (Grose, Harding, Wilton & Bissenden, 1989).

Less than 3% of light below 580 nm and between 14.5 and 21.4% at 700 nm, depending

1840 Short Note

on age, reaches the anterior surface of the eye through the closed eyelid. These ikiings are a necessary prerequisite for calculating the ocular light dose of the preterm neonate, and further investigation of possible effects of light on the immature visual system.

~C~o~~e~e~nts-J~~th Robinson was funded by the Royal National Institute for the BBnd. The Birmingbam Hospitals Saturday Fund provided a grant towards the cost of the apparatus. Thanks are also due to Dr J. Hennessy, Department of Physics, L&ester University who constructed the ophthalmic spectropbotometer.

REFERENCES

Crawford, M. L. J. & Marc, R. E. (1976). Light trans- mission of cat and monkey eyelids. Vi&n Besea*ch, Id, 323-324.

Grose, J., Harding, G. F. A., Wilton, A. Y. & Bissenden, J. G. (1989). The matumtion of the pattern reversal and

Rash ERG in preterm infants. CIinicol Vi&m Sciences, 4.

239-246. Jaffe, M. J., Caruso, R. C., de Monasterio, F. M. &

Nussenblatt, R. B. (1987). Chronic eyelid closure and its effect upon the human cone and rod &ctroretinogram. American Journal of Optometry Physiological Optics, 64. 373-376.

Moseley, M. J., Bayliss, S. C. & Fielder, A. R. (1988). Light transmission through the human eyeIi& in f&o measure- ment. ~~~t~ ad Physi~ogica~ Optics, 8, 229-230.

Robinson, J. (1990). Light and the eye of the preterm neonate, Ph.D. thesis, University of Birmingham.

Robinson, J., MoseIey, M. J., Thompson, J. R. & Fielder, A. R. (I 989). Eyelid opening in preterm neonates. Archives of Diseases in ChMood, 64, 943-948.

Robinson, J., hioseley, M. J. & Fielder, A. R. (1990). Ihuminance of neonatal units. ArcMves of Diseases in Childhood, 65, 679-682.

Simon, T. R. C. & Wickler, M. (1973). Transmission of light through living tissue. Pediatric Research, 316, 88.

Weaver, D. R. & Reppert, S. M. (1989). Direct in utero perception of light h$ the mammalian fetus. Deuelop- mental Brain Research, 47, 151-155.