- Email: [email protected]
Low frequency temporal modulation of light promotes a myopic shift in refractive compensation to all spectacle lenses
Experimental Eye Research 83 (2006) 322e328 www.elsevier.com/locate/yexer
Low frequency temporal modulation of light promotes a myopic shift in refractive compensation to all spectacle lenses Sheila G. Crewther a,*, Ayla Barutchu a, Melanie J. Murphy a, David P. Crewther b a
School of Psychological Science, La Trobe University, Plenty Rd, Bundoora, Melbourne, Vic. 3086, Australia b Brain Sciences Institute, Swinburne University of Technology, Hawthorn 3122, Australia Received 26 June 2005; accepted in revised form 15 December 2005 Available online 31 March 2006
Abstract Emmetropization, the process by which ocular growth of young animals adapts to ensure focussed retinal images, can be disrupted by high frequency flicker, causing a hypermetropic shift. Emmetropization can also be disrupted differentially, in a sign dependent manner, by pharmacological alteration of the balance of activation of the ON and OFF retinal sub-systems in normal light or by rearing in an environment with a moving spatiotemporally varied diamond pattern (yielding local sawtooth illumination on the retina). Thus the aim of this experiment was to determine whether low frequency temporal modulation alone was sufficient to cause defocus sign-dependent interference with compensation. Chicks were reared for 6 or 7 days with monocular 10 D, 0 D, or No Lenses in a 12 h light/dark cycle. Luminance of the environment was temporally modulated during the light cycle with a non-square wave profile pulse of 250 msec duration, with the illumination fluctuating between 1.5 and 180 lux at 1 Hz, 2 Hz, 4 Hz or with no flicker (0 Hze180 lux). Final refractive state and ocular dimensions, measured using retinoscopy and A-scan ultrasonography, demonstrated that in the absence of temporal luminance modulation (0 Hz), chicks compensated to induced defocus in the expected sign-dependent manner. However, under 1, 2 and 4 Hz flickering light conditions, there was an overall myopic offset of approximately 6D across lens groups with refractive compensation to positive lenses more strongly inhibited. This myopic offset was reflected by increases in the depth of both vitreous and anterior chambers. However, luminance modulation had no effect on refraction or ocular parameters in the No Lens conditions. This is a hitherto unreported strong interaction between lens wear and low frequency temporally modulated light, with the refractive compensation mechanism being overridden by a generalized myopic shift. Ó 2006 Elsevier Ltd. All rights reserved. Keywords: myopia; flicker; low temporal frequency luminance modulation; on and off response; hyperopia; lens defocus
1. Introduction The process by which postnatal growth of the axial components of the eyes of most young animals including humans, primates and chickens, interacts with the visual environment, to actively match the optical power of the cornea and lens is termed emmetropization. However, despite years of research elucidating how varied light conditions affect * Corresponding author. Tel.: þ61 3 9479 2290; fax: þ61 3 9479 1956. E-mail addresses: [email protected] (S.G. Crewther), a.barutchu@ latrobe.edu.au (A. Barutchu), [email protected] (M.J. Murphy), [email protected] (D.P. Crewther). 0014-4835/$ - see front matter Ó 2006 Elsevier Ltd. All rights reserved. doi:10.1016/j.exer.2005.12.016
emmetropization the biological mechanisms driving this process are unclear (reviewed Crewther, 2000; Fredrick, 2002; Wallman and Winawer, 2004) and in most cases remain relatively unexplored. One week of optical defocus (OD) induces almost complete refractive and axial compensation (Irving et al., 1991, 1992; Schaeffel et al., 1988; Schmid and Wildsoet, 1997; Wallman et al., 1978; Wildsoet and Wallman, 1995) and morphological changes in retina, choroid and sclera (Beresford et al., 2001). In most cases the strong relationship between abnormal axial growth and resultant refractive error is assumed to derive primarily from changes in vitreous chamber depth though anterior chamber depth changes have also been implicated on a number of occasions (Barutchu et al., 2002;
S.G. Crewther et al. / Experimental Eye Research 83 (2006) 322e328
Gottlieb et al., 1987; Lauber and Oishi, 1987; Napper et al., 1995). Emmetropization and ocular growth have long been known to be perturbed by rearing chicks in unusual visual conditions such as dim light or constant bright light, with dim light inducing axial elongation and a myopic shift in refraction while constant light causes increased ocular growth, but a flatter cornea resulting in a net hyperopic shift (Lauber and Oishi, 1987). Many other variations in distribution of visual experience and the consequent biometric affect on emmetropization have been examined (see (Winawer and Wallman, 2002) for an extensive review). Interpretation has usually been in terms of how much experience is needed for disruption of refractive compensation with interpretation in terms of the construct of ‘‘STOP/GO’’ signals rather than in more physiological terms such as which neural elements are involved and how retinal signal transduction in the ON and OFF systems could be perturbed by the different patterns of light stimulation. Stroboscopic flicker of frequencies above 6 Hz was originally shown to suppress the development of form deprivation myopia in chicks (FDM) (Gottlieb et al., 1987; Vingrys et al., 1991) with the effects being mimicked by pharmacological manipulations of the retinal ON and OFF systems (Crewther and Crewther, 1990). Inhibition of the OFF retinal pathway in form deprived chicks with D-a aminoadipic acid (DAAA) or with cis 2,3 piperidine dicarboxylic acid (PDA) (Crewther et al., 1996) resulted in less myopia than with occlusion alone. Similarly, intravitreal kainic acid reduces form deprivation induced eye growth in chickens (Ehrlich et al., 1990) with damage to photoreceptors, amacrine, bipolar cells and a small proportion of ganglion cells, presumably with more damage to OFF type bipolars and amacrines than ON type neurons (Dvorak and Morgan, 1983; Golcich et al., 1990). More recently form deprivation and defocus induced myopia (DIM) have been reported to be inhibited by square wave flicker of 6 Hz or above, while defocus induced hypermetropia (DIH) could only be inhibited by flicker above 12 Hz (Schwahn and Schaeffel, 1997). Schwahn and Schaeffel interpreted this as the extent of inhibition of FDM or DIM (but not DIH) being related to the duty cycle of the flicker. They concluded ‘‘that two different retinal circuits with different temporal characteristics are involved in the processing of hyperopic defocus/deprivation and of myopic defocus, [and only] the first one is dependent on flicker ERG amplitude’’. Perturbation of emmetropia has also been recently reexamined by Winawer and Wallman, in terms of the minimum visual exposure the eye needs to exhibit what they consider the principle components of ocular compensation: a scleral mechanism of altered rate of elongation and altered choroidal thickness. They found that, in chicks, such emmetropization processes can be activated by very brief periods of light in animals otherwise reared in darkness. Chicks required 14 min (made up of 7 2 min periods) of total patterned light experience per day rather than the same duration in 1 or 2 sessions, to initiate sign appropriate compensation to optical defocus (Winawer and Wallman, 2002). Their results also suggest that the refractive response to plus lenses is dominant over
323
that to negative lenses. They also suggested that very brief episodes of experience, even if very frequent, are relatively ineffective, however, they did not reconcile these results with the earlier reports of higher frequency flicker on refractive compensation. Although emmetropization can be activated by extremely short durations of normal light it can also be disrupted in a sign dependent manner by light of normal intensity and duration but with a dynamically changing pattern of spatiotemporal luminance modulation or with differential pharmaceutical activation of retinal ON and OFF pathways (Crewther and Crewther, 2002, 2003). Spatiotemporal manipulation of the environment with a local sawtooth temporal luminance profile, asymmetric with Fast OFFset and Slow ONset at a frequency of 4 Hz, suppressed compensation to þ10 D but not to 10 D defocus, whereas Fast ONset/Slow OFFset inhibited compensation to 10 D defocus but not to þ10 D. By comparison, rearing chicks in an environment of a stationary sawtooth pattern gave almost complete refractive compensation to all lens powers used, indicating that asymmetrical spatial luminance patterns per se are unlikely to affect emmetropization or refractive compensation and that the temporal luminance modulation (TLM) profile is necessary to interfere with refractive compensation (Crewther and Crewther, 2002). Thus it was the aim of the present investigation to further explore the effect on ocular growth and refractive compensation to optical defocus of whole environment low frequency temporal (1, 2 and 4 Hz) non-square-wave luminance modulation. 2. Methods 2.1. Animals In total, 229 day-old male chicks (white-Leghorn blackAustralorp cockerels) obtained from a local commercial hatchery were raised from days 4e10 or days 7e14 post-hatching, with a monocular (right eye) goggle containing one of 10 D, 0 D lens or No Lens (see Table 1 for ages and group sizes). Changes in hatchery production schedules were responsible for differences in rearing days. Within each lens condition, chicks were randomly allocated to one of four light profile conditions for 12 h: constant light (0 Hz), 1 Hz, 2 Hz or 4 Hz. Chicks were housed in batches of 20 with ad libitum access to food and water in a custom, light-sealed, ventilated box. Each batch contained 5 chicks from each of the 4 lens groups. Temperature was maintained at an average of 29 C (2 C) during the experimental period, and illumination was provided by a 12 V, 20 W halogen lamp on a programmed 12 h light/dark day cycle. 2.2. Experimental protocol and biometric measures At the beginning of the experimental period (day 4 or 7) lenses were affixed over the right eye, while the fellow eye was used as a control. Optical defocus was induced using
S.G. Crewther et al. / Experimental Eye Research 83 (2006) 322e328
324
Table 1 Experimental conditions and number per group and age of chicks at time of analysis Light profile
Lens
Experimental duration in days
Age of animal at analysis
Number of chicks
0 Hz
þ10 0 No Lens 10 D
6 6 6 6
10 10 10 10
15 14 14 15
1 Hz
þ10 0 No Lens 10 D
6 6 6 6
10 10 10 10
15 13 14 15
2 Hz
þ10 0 No Lens 10 D
7 7 7 7
14 14 14 14
7 6 20 10
4 Hz
þ10 0 No Lens 10 D
6 6 6 6
10 10 10 10
18 17 19 17
positively or negatively powered (10 D) modified human PMMA contact lenses (8.1 mm in diameter), while 0 D lenses and the No Lens condition were used to assess the effects of lens wear in general. Detailed lens preparation and attachment procedures are found in Crewther and Crewther (2002). During the 12 h light component of the diurnal cycle, environmental luminance was either maintained constantly at 180 lux (0 Hz) or manipulated via TLM of 1, 2 or 4 Hz produced by a function generator linked to a Field Effect Transistor (FET) and a halogen globe. Environmental luminance was measured with a light meter probe fixed at the level of the chick’s eyes attached to an oscilloscope while animals were moving around in the box. Continuous measurements were recorded over 15 min with little variation in the upper and lower limits, which were then read off the oscilloscope. The TLM profile indicated that the flicker consisted of two phases: a light pulse consisting of a gradual rise and slightly sharper decline in luminance between 1.5 and 180 lux (pulse length 250 msec, with full width half maximum of each flash 100 msec) and a second dark (1.5 lux) phase. Thus for each frequency the duration of darkness (duty cycle) varied (see Fig. 1 for flicker profiles). A natural consequence of maintaining the same peak luminance is that the mean luminance varied with frequency. Throughout the experimental period the health of chicks and cleanliness of lenses were monitored twice daily. At the end of each experimental period, chicks were anaesthetized with an intramuscular injection 4.5 mg/kg xylazine: 45 mg/ kg ketamine and both eyes were refracted by retinoscopy (Keeler, Vista Diagnostic Instruments) and axial dimensions were obtained with the average of at least three A-Scan ultrasonography traces (A-Scan Ophthal A-Scan III, TSL: Teknar, Inc. St Louis). Anterior chamber depth was determined by subtracting A-Scan measures of the distance from anterior of the lens to retina from measures of total axial length. Chicks were then euthanized via a barbiturate overdose (Lethobarb;
Fig. 1. Temporal Luminance Modulation (TLM) profiles for 1 Hz (with 750 ms of dark phase per second), 2 Hz (with 500 ms of dark phase per second) and 4 Hz (with no extended dark phase). The pulse profile for all frequencies was maintained at a width of 250 ms. During each flicker cycle, luminance varied between maximum/minimum light levels of 180 lux and 1.5 lux.
Loveridge, Southampton, Hants, UK). All procedures were conducted in strict accordance with La Trobe University Animal Ethics Committee guidelines and adhere to the ARVO Statement for the use of Animals in Ophthalmic and Vision Research. 2.3. Data analysis To minimise variability related to extraneous variables, such as variations in eye size related to overall body weight, refractive and biometric measurements were converted to raw score difference between experimental and fellow eyes. (All mean raw data is presented in Table 2). To assess whether the two different experimental periods affected final refractive status and or axial dimensions, a 2(experimental periods) 4(lens conditions) factorial ANOVA was used to compare the differences between experimental and fellow eyes of chicks reared from days 4e10 and days 7e14 under different lens conditions (10 D, 0 D and No Lens) with constant (dayperiod) illumination at 180 lux (0 Hz condition). No significant differences in either refractive or biometric measures (P > 0.25 for all comparisons) were found between the two experimental periods. Thus, the data from the two rearing time groups were collapsed prior to the statistical analysis of the effect of lens wear under different luminance conditions on refractive sate and ocular growth using a series of ANOVAs. 3. Results As expected, exposure to the 12 h of continuous daylight cycle (0 Hz condition) induced good refractive compensation to positive and negative lenses, and little refractive change in 0 D and No Lens groups (see Fig. 2). As is apparent from Table 2, a series of paired t-tests between the various temporal frequencies of rearing did not show any significant differences in refraction or ocular dimensions for the No Lens group. However, TLM of the environment systematically altered growth compensation in the positive, plano and negative lens groups. TLM at frequencies from 1e4 Hz resulted in
S.G. Crewther et al. / Experimental Eye Research 83 (2006) 322e328
325
Table 2 Means (SEM) of experimental and control eyes for axial length, vitreous chamber depth and anterior chamber length Refractive state
Axial length
Vitreous chamber
Anterior chamber
Experimental
Control
Experimental
Control
Experimental
Control
Experimental
Control
8.57 0.53 13.0 0.94 13.88 1.12 13.12 0.60
0.70 0.31 0.34 0.58 0.15 0.29 0.32 0.32
9.26 0.04 9.67 0.05 9.91 0.11 9.62 0.05
8.80 0.04 9.07 0.04 9.24 0.08 9.00 0.04
5.55 0.05 5.88 0.04 6.00 0.08 5.78 0.05
5.19 0.04 5.41 0.03 5.42 0.05 5.36 0.03
1.49 0.03 1.62 0.03 1.59 0.03 1.57 0.03
1.40 0.02 1.47 0.02 1.49 0.03 1.43 .02
0D 0 Hz 1 Hz 2 Hz 4 Hz
0.16 0.32 6.47 1.59 4.58 0.89 6.21 1.58
0.75 0.18 0.07 0.29 0.90 0.72 0.49 0.31
8.89 0.04 9.29 0.08 9.62 0.10 9.25 0.08
8.84 0.03 8.89 0.09 9.43 0.15 8.96 0.05
5.21 0.04 5.53 0.05 5.61 0.09 5.45 0.07
5.24 0.03 5.31 0.05 5.54 0.09 5.30 0.04
1.45 0.03 1.56 0.05 1.72 0.06 1.54 0.04
1.43 0.02 1.38 0.07 1.56 0.08 1.39 0.02
No Lens 0 Hz 1 Hz 2 Hz 4 Hz
0.70 0.22 0.52 0.40 0.15 0.26 0.07 0.31
0.41 0.24 0.28 0.38 0.44 0.14 0.81 0.27
9.07 0.04 9.08 0.05 9.77 0.05 8.82 0.05
9.03 0.03 9.03 0.06 9.70 0.05 8.79 0.05
5.39 0.03 5.47 0.04 5.80 0.40 5.22 0.04
5.37 0.03 5.42 0.05 5.77 0.04 5.19 0.04
1.39 0.02 1.43 0.02 1.52 0.02 1.37 0.02
1.41 0.02 1.46 0.05 1.54 0.02 1.39 0.02
þ10 D 0 Hz 1 Hz 2 Hz 4 Hz
9.30 0.32 3.45 1.74 0.43 1.90 1.74 1.17
1.15 0.19 0.30 0.03 0.68 0.35 0.39 0.60
8.51 0.06 9.29 0.13 9.15 0.25 8.88 0.08
8.82 0.06 8.99 0.08 9.14 0.19 8.91 0.05
4.80 0.06 5.46 0.08 5.32 0.17 5.17 0.05
5.21 0.05 5.36 0.07 5.37 0.14 5.30 0.05
1.51 0.04 1.61 0.08 1.55 0.08 1.50 0.05
1.45 0.03 1.42 0.03 1.44 0.06 1.39 0.09
10 D 0 Hz 1 Hz 2 Hz 4 Hz
a generalized myopic shift in refractive state and ocular elongation in all lens-treated eyes with no apparent relation between temporal frequency and duty cycle for the 10 or 0 D lens groups. A non-significant trend towards a less myopic shift was seen in refractive compensation by the þ10 D lens group as temporal frequency increased from 1 to 4 Hz. Refractive development was significantly dependent on the interaction between lens induced defocus and the frequency of TLM, F9,212 ¼ 4.39, P < 0.001. A large myopic shift was observed in 10 D treated eyes which developed myopia of 3 D
Difference in Refractive State (D)
10
5 0 Hz 1 Hz 2 Hz 4 Hz
0
-5
-10
-15 - 10 D
0D
No Lens
+ 10 D
Lens Type Fig. 2. Mean difference in refractive state (SEM) for experimental minus fellow refractions for chicks reared using temporal luminance modulation of 0, 1, 2 and 4 Hz and across the different lens groups No Lens, 0 D, þ10 D, 10 D.
exceeding the power of the imposed defocus and 4 D with respect to the no flicker 10 D rearing group. The effect of TLM was even greater on 0 D (nearly 5 D of myopic shift) and þ10 D groups (10 D of myopic shift) with respect to the corresponding 0 Hz lensed groups. Similarly TLM led, not only to the suppression of compensation to positive lenses at frequencies of 2 and 4 Hz, but to an excessive myopic shift in response to both the þ10 D and plano lenses in the 1 Hz condition. Interestingly, TLM did not significantly perturb the normal refractive development of eyes in the No Lens condition, nor the untreated fellow eyes in any condition (see Table 2). Note also the much lower variability in refractions seen in all groups raised in constant light during the day period. Table 2 indicates that the axial lengths of the eyes of the No Lens group are usually equal to or greater than those of the non-lens-wearing eye of the lensed groups. It is also apparent that the anterior chamber depth of non-lens-wearing eyes is smaller than the AC depth of all lens-wearing groups. Changes in refractive states were also accompanied by changes in ocular dimensions including axial and vitreous chamber elongation (Fig. 3). Significant lens-by-light profile interactions were found for both axial length (F9,212 ¼ 4.68, P < 0.001) and vitreous chamber depth (F9,212 ¼ 6.28, P < 0.001). Negative and plano lens-wear led to increased ocular elongation in all flicker conditions; the degree of elongation observed in the negative lens condition was significantly enhanced under TLM, corresponding to the approximately 5 diopter difference in refractive compensation observed in the plano lens group under TLM in comparison to 0 Hz (constant daytime illumination) (see Table 2). Chicks wearing positive lenses showed a significant but varied, increase in axial length and vitreous chamber depth when exposed to TLM of 1 Hz, and 2 and 4 Hz compared to those raised in 0 Hz modulation.
S.G. Crewther et al. / Experimental Eye Research 83 (2006) 322e328
326
Ocular Dimensions (mm)
0.7
0.5
0.3
0.1
-0.1
-0.3
-0.5 4
2
1
Hz
Hz
Hz
Hz
Hz
Hz
Hz
No Lens
0
4
2
Hz
Hz
Hz
Hz
Hz
Hz
Hz
Hz
Hz
0D
1
0
4
2
1
0
4
2
1
0
- 10 D
+ 10 D
Lens Type Fig. 3. Mean differences (SEM) in axial length (black), vitreous chamber depth (grey) and anterior chamber depth (white) between experimental and fellow eyes of chicks across all conditions of temporal luminance modulation and lens wear.
However as can be readily observed in Fig. 3, the vitreous chamber depths cannot account for the total change in axial length. In particular note the plano lens group. A factorial ANOVA examining anterior chamber depth measures revealed a significant main effect for both lens type (F9,212 ¼ 11.05, P < 0.001) and TLM frequency (F9,212 ¼ 2.66, P ¼ 0.049). Anterior chambers were significantly deeper in eyes reared with lenses compared to No Lens groups, and in eyes exposed to 1 Hz flicker compared to other TLM conditions. 4. Discussion This study demonstrates that full-field luminance modulation at low temporal frequencies with variation in mean luminance and duty cycle perturbs axial growth and refractive compensation to optical defocus, resulting in a generalized myopic shift in all lens wearing groups, but not in any of the No Lens groups. While significant in all lens groups, the suppression of compensation was greatest in chicks reared with positive lens defocus. Such a myopic shift by all (i.e. negative, positive and plano) lens wear groups is in the opposite direction to all previous reports of the effects of stroboscopic flicker on refractive compensation to optical defocus, however the statistical strength of the data is undeniable. Currently it is accepted that square wave flicker induces suppression to form deprivation myopia and negative lens myopia at 6 and 12 Hz (Schwahn and Schaeffel, 1997) or at 10, 15 and 20 Hz (Schmid and Wildsoet, 1997) and suppression of positive lens wear at around 12 Hz. By contrast, in our experiment, all groups wearing negative lenses in the presence of low frequency TLM conditions showed similar degrees of enhanced development of myopia beyond the dioptric power of the 10 D lenses, despite variation in duty cycle and mean luminance. Our results also demonstrate suppression of refractive compensation to þ10 D lens wear using the same TLM with the greatest effect observed at 1 Hz modulation. One Hz TLM
with þ10 D defocus not only eliminated the development of DIH, but led to w4 D of myopia while positive lens wear combined with 2 Hz and 4 Hz TLM resulted in minimal generation of hyperopia. Furthermore the pattern of our results for negative, positive, and plano lens groups did not demonstrate a duty cycle relationship with compensation as reported under higher flicker rates (Schwahn and Schaeffel, 1997). It is noteworthy that the myopic shift cannot be merely attributed to differences in the duty cycle of the TLM, as Winawer and Wallman have recently shown that refractive compensation to optical defocus can take place with only 14 min total visual exposure per day (2 min per 2 h 7 times per 14 h day) (Winawer and Wallman, 2002). This is less total light exposure in a day than our 1 Hz TLM animals received in an hour i.e. 15 min (250 msecs per sec per hour). It is also likely that our much greater total duration of visual exposure may account for the fact that although all of our 1 Hz TLM animals received only 250 msec per sec per exposure, this was adequate to give greater than normal refractive compensation to negative lenses and a further sign dependent suppression of positive lens refractive compensation on top of the generalized myopic shift. This also makes it difficult to explain why similar durations of visual exposure under the conditions of TLM for the positive lens group resulted in total suppression of compensation, if as posited (Schwahn and Schaeffel, 1997; Winawer and Wallman, 2002) there are 2 different mechanisms for positive and negative lens refractive compensation and if refractive compensation to plus lens wear is dominant over refractive compensation to negative lens wear (Winawer and Wallman, 2002). As few previous studies have concurrently used both plano lens and No Lens controls and certainly none have reported a difference in relative refraction the question of why it occurs needs to be addressed. In our studies all animals were raised in the same enclosures under the same conditions with lens conditions mixed equally in each rearing batch, so the results are
S.G. Crewther et al. / Experimental Eye Research 83 (2006) 322e328
327
results reflect a strong difference between lens-reared and No Lens groups. Irrespective of the power of lens employed, anterior chambers were deeper after exposure to low frequency TLM in conjunction with lens wearing than TLM in No Lens wearing groups. Simply stated, there is more fluid within the vitreous and anterior chambers of the lensed eyes receiving flicker stimulation, whether or not they are also in receipt of optical defocus. Whether this fluid build up is due to increased secretion or to impaired outflow in the front or back of the eye will require further research. Certainly 1 Hz flicker can be observed to increase pupillary constriction and expansion and this may act to slow aqueous removal.
unlikely to be due to unusual visual degradation of the plano group compared to the others. While an overall myopic shift in refraction could be attributed to improper maintenance of the lens-cleaning regimen resulting in form deprivation myopia, this argument can be rejected on the grounds that the 0 Hz flicker þ10 D group showed excellent compensation, achieving a mean refraction of þ9.3 D of hyperopia. Furthermore, since the lensed and non-lensed groups without flicker showed little variability in refraction or ocular dimensions, the mechanism resulting in excessive myopia appears to be related to the effect of low frequency TLM on lensed eyes. Also, while flicker duty cycle leading to possible imbalance of ON and OFF retinal pathway activation has been shown to have some effects on refractive and growth response in negative lens wear groups (Schwahn and Schaeffel, 1997), the uniform myopic shift across all lensed eyes of all powers is much greater than any variation in duty cycle with temporal frequency (i.e. with duty cycle ranging from 0e750 msec). Three possible mechanisms to explain the results are suggested (all requiring further experimentation): i) peripheral restriction of field through the lens supports leading to myopia; ii) low frequency flicker leading to a resultant increase in the fluid in the anterior chamber, either being due to increased ciliary fluid secretion or restricted outflow possibly caused by the attachment of the lens to the periocular feathers; iii) The nature of the temporal profile of the light pulse, being biased towards a Slow-ON, Fast-OFF profile, may engender the myopia shift and enhance suppression of refractive compensation to positive lenses.
The results presented here demonstrating a much stronger suppression of compensation to positive lens defocus than to negative defocus are reminiscent of the greater suppression of refractive compensation to þ10 D compared to 10 D induced-defocus with low temporal frequency Fast OFF/Slow ON sawtooth luminance profiles (Crewther and Crewther, 2002). Indeed the measured pulse profile used here is biased towards a Fast OFF/Slow ON shape. As all flickered lenses groups showed a myopic shift, it is likely that our results reflect the effects of the temporal contrast modulation profile used. However the absence of an effect on the No Lens group to this flicker profile would require that there is an extra element, such as peripheral occlusion which combines with the effects of temporal profile.
4.1. Peripheral restriction
5. Conclusion
The two layers of matched Velcro we use to hold our spectacle lens reduces the field of view from 180 to 90 and may play a role in myopic shifts in refraction. This is similar to a much earlier suggestion (Nathan et al., 1985) that peripheral dysfunction through visual disease may play a part in inducing myopic shifts in refractions in children with low vision. Nathan et al. observed an association between more myopic refractions and disease with a peripheral or peripheral plus central impairment of vision whereas those conditions in which foveal vision was primarily impaired showed a mild hypermetropic trend. Also, it has been recently reported that peripheral vision restriction perturbs emmetropization in monkeys and induces a slight myopic shift (Smith et al., 2005). Thus, a similar aetiolgy for the myopic shift seen in all our lens groups is possible but this explanation still cannot account for the fact that the plano lens group (0 Hz) did not show the myopic shift.
Positive, plano and negative lens wear in the presence of a non-square wave profile low frequency (1, 2 and 4 Hz) temporal luminance modulation results in a general myopic shift in refractive compensation and enhanced ocular growth. The results also highlight the role of abnormal temporal modulation in accurate refractive compensation and emphasize the interaction between flicker induced changes in the activation of the ON and OFF retinal pathways and lens rearing.
4.2. Flicker induced fluid and aqueous outflow restriction Consistent with previous findings, refractive changes were accompanied by changes in ocular growth patterns, both in vitreous and anterior chamber depth. The anterior chamber depth
4.3. Temporal modulation profile
References Barutchu, A., Crewther, S.G., Crewther, D.P., 2002. Effects of optical defocus and spatial contrast on anterior chamber depth in chicks. Clin. Experiment. Ophthalmol. 30, 217e220. Beresford, J.A., Crewther, S.G., Kiely, P.M., Crewther, D.P., 2001. Comparison of refractive state and circumferential morphology of retina, choroid, and sclera in chick models of experimentally induced ametropia. Optom. Vis. Sci. 78, 40e49. Crewther, D.P., 2000. The role of photoreceptors in the control of refractive state. Prog. Retin. Eye Res. 19, 421e457. Crewther, D.P., Crewther, S.G., 1990. Pharmacological modification of eye growth in normally reared and visually deprived chicks. Curr. Eye Res. 9, 733e740. Crewther, D.P., Crewther, S.G., 2002. Refractive compensation to optical defocus depends on the temporal profile of luminance modulation of the environment. Neuroreport 13, 1029e1032.
328
S.G. Crewther et al. / Experimental Eye Research 83 (2006) 322e328
Crewther, D.P., Crewther, S.G., Xie, R.Z., 1996. Changes in eye growth produced by drugs which affect retinal ON or OFF responses to light. J. Ocul. Pharmacol. Ther. 12, 193e208. Crewther, S.G., Crewther, D.P., 2003. Inhibition of retinal ON/OFF systems differentially affects refractive compensation to defocus. Neuroreport 14, 1233e1237. Dvorak, D.R., Morgan, I.G., 1983. Intravitreal kainic acid permanently eliminates off-pathways from chicken retina. Neurosci. Lett. 36, 249e253. Ehrlich, D., Sattayasai, J., Zappia, J., Barrington, M., 1990. Effects of selective neurotoxins on eye growth in the young chick. Ciba Found. Symp. 155, 63e84. discussion 84e68. Fredrick, D.R., 2002. Myopia. BMJ 324, 1195e1199. Golcich, M.A., Morgan, I.G., Dvorak, D.R., 1990. Selective abolition of OFF responses in kainic acid-lesioned chicken retina. Brain Res. 535, 288e300. Gottlieb, M.D., Marran, A., Xu, A., Nickla, D.L., Wallman, 1987. The emmetropization process in chicks is compromised by dim light. Invest. Ophthalmol. Vis. Sci. 32 (Suppl.), 2616. Irving, E.L., Callender, M.G., Sivak, J.G., 1991. Inducing myopia, hyperopia, and astigmatism in chicks. Optom. Vis. Sci. 68, 364e368. Irving, E.L., Sivak, J.G., Callender, M.G., 1992. Refractive plasticity of the developing chick eye. Ophthalmic. Physiol. Opt. 12, 448e456. Lauber, J.K., Oishi, T., 1987. Lid suture myopia in chicks. Invest. Ophthalmol. Vis. Sci. 28, 1851e1858. Napper, G.A., Brennan, N.A., Barrington, M., Squires, M.A., Vessey, G.A., Vingrys, A.J., 1995. The duration of normal visual exposure necessary
to prevent form deprivation myopia in chicks. Vision Res. 35, 1337e 1344. Nathan, J., Kiely, P.M., Crewther, S.G., Crewther, D.P., 1985. Disease-associated visual image degradation and spherical refractive errors in children. Am. J. Optom. & Physiol. Opt. 62, 680e688. Schaeffel, F., Glasser, A., Howland, H.C., 1988. Accommodation, refractive error and eye growth in chickens. Vision Res. 28, 639e657. Schmid, K.L., Wildsoet, C.F., 1997. Contrast and spatial-frequency requirements for emmetropization in chicks. Vision Res. 37, 2011e2021. Schwahn, H.N., Schaeffel, F., 1997. Flicker parameters are different for suppression of myopia and hyperopia. Vision Res. 37, 2661e2673. Smith, E.L.d., Kee, C.S., Ramamirtham, R., Qiao-Grider, Y., Hung, L.-F., 2005. Peripheral vision can influence eye growth and refractive development in infant monkeys. Invest. Ophthalmol. Vis. Sci. 46, 3965e3972. Vingrys, A.J., Squires, M.A., Napper, G.A., Barrington, M., Vessey, G.A., Brennan, N.A., 1991. Prevention of form deprivation myopia in post-hatch chickens. Invest. Ophthal. Vis. Sci. 32 (Suppl.), 1203. Wallman, J., Turkel, J., Trachtman, J., 1978. Extreme myopia produced by modest change in early visual experience. Science 201, 1249e1251. Wallman, J., Winawer, J., 2004. Homeostasis of eye growth and the question of myopia. Neuron 43, 447e468. Wildsoet, C., Wallman, J., 1995. Choroidal and scleral mechanisms of compensation for spectacle lenses in chicks. Vision Res. 35, 1175e1194. Winawer, J., Wallman, J., 2002. Temporal constraints on lens compensation in chicks. Vision Res. 42, 2651e2668.
Low frequency temporal modulation of light promotes a myopic shift in refractive compensation to all spectacle lenses Sheila G. Crewther a,*, Ayla Barutchu a, Melanie J. Murphy a, David P. Crewther b a
School of Psychological Science, La Trobe University, Plenty Rd, Bundoora, Melbourne, Vic. 3086, Australia b Brain Sciences Institute, Swinburne University of Technology, Hawthorn 3122, Australia Received 26 June 2005; accepted in revised form 15 December 2005 Available online 31 March 2006
Abstract Emmetropization, the process by which ocular growth of young animals adapts to ensure focussed retinal images, can be disrupted by high frequency flicker, causing a hypermetropic shift. Emmetropization can also be disrupted differentially, in a sign dependent manner, by pharmacological alteration of the balance of activation of the ON and OFF retinal sub-systems in normal light or by rearing in an environment with a moving spatiotemporally varied diamond pattern (yielding local sawtooth illumination on the retina). Thus the aim of this experiment was to determine whether low frequency temporal modulation alone was sufficient to cause defocus sign-dependent interference with compensation. Chicks were reared for 6 or 7 days with monocular 10 D, 0 D, or No Lenses in a 12 h light/dark cycle. Luminance of the environment was temporally modulated during the light cycle with a non-square wave profile pulse of 250 msec duration, with the illumination fluctuating between 1.5 and 180 lux at 1 Hz, 2 Hz, 4 Hz or with no flicker (0 Hze180 lux). Final refractive state and ocular dimensions, measured using retinoscopy and A-scan ultrasonography, demonstrated that in the absence of temporal luminance modulation (0 Hz), chicks compensated to induced defocus in the expected sign-dependent manner. However, under 1, 2 and 4 Hz flickering light conditions, there was an overall myopic offset of approximately 6D across lens groups with refractive compensation to positive lenses more strongly inhibited. This myopic offset was reflected by increases in the depth of both vitreous and anterior chambers. However, luminance modulation had no effect on refraction or ocular parameters in the No Lens conditions. This is a hitherto unreported strong interaction between lens wear and low frequency temporally modulated light, with the refractive compensation mechanism being overridden by a generalized myopic shift. Ó 2006 Elsevier Ltd. All rights reserved. Keywords: myopia; flicker; low temporal frequency luminance modulation; on and off response; hyperopia; lens defocus
1. Introduction The process by which postnatal growth of the axial components of the eyes of most young animals including humans, primates and chickens, interacts with the visual environment, to actively match the optical power of the cornea and lens is termed emmetropization. However, despite years of research elucidating how varied light conditions affect * Corresponding author. Tel.: þ61 3 9479 2290; fax: þ61 3 9479 1956. E-mail addresses: [email protected] (S.G. Crewther), a.barutchu@ latrobe.edu.au (A. Barutchu), [email protected] (M.J. Murphy), [email protected] (D.P. Crewther). 0014-4835/$ - see front matter Ó 2006 Elsevier Ltd. All rights reserved. doi:10.1016/j.exer.2005.12.016
emmetropization the biological mechanisms driving this process are unclear (reviewed Crewther, 2000; Fredrick, 2002; Wallman and Winawer, 2004) and in most cases remain relatively unexplored. One week of optical defocus (OD) induces almost complete refractive and axial compensation (Irving et al., 1991, 1992; Schaeffel et al., 1988; Schmid and Wildsoet, 1997; Wallman et al., 1978; Wildsoet and Wallman, 1995) and morphological changes in retina, choroid and sclera (Beresford et al., 2001). In most cases the strong relationship between abnormal axial growth and resultant refractive error is assumed to derive primarily from changes in vitreous chamber depth though anterior chamber depth changes have also been implicated on a number of occasions (Barutchu et al., 2002;
S.G. Crewther et al. / Experimental Eye Research 83 (2006) 322e328
Gottlieb et al., 1987; Lauber and Oishi, 1987; Napper et al., 1995). Emmetropization and ocular growth have long been known to be perturbed by rearing chicks in unusual visual conditions such as dim light or constant bright light, with dim light inducing axial elongation and a myopic shift in refraction while constant light causes increased ocular growth, but a flatter cornea resulting in a net hyperopic shift (Lauber and Oishi, 1987). Many other variations in distribution of visual experience and the consequent biometric affect on emmetropization have been examined (see (Winawer and Wallman, 2002) for an extensive review). Interpretation has usually been in terms of how much experience is needed for disruption of refractive compensation with interpretation in terms of the construct of ‘‘STOP/GO’’ signals rather than in more physiological terms such as which neural elements are involved and how retinal signal transduction in the ON and OFF systems could be perturbed by the different patterns of light stimulation. Stroboscopic flicker of frequencies above 6 Hz was originally shown to suppress the development of form deprivation myopia in chicks (FDM) (Gottlieb et al., 1987; Vingrys et al., 1991) with the effects being mimicked by pharmacological manipulations of the retinal ON and OFF systems (Crewther and Crewther, 1990). Inhibition of the OFF retinal pathway in form deprived chicks with D-a aminoadipic acid (DAAA) or with cis 2,3 piperidine dicarboxylic acid (PDA) (Crewther et al., 1996) resulted in less myopia than with occlusion alone. Similarly, intravitreal kainic acid reduces form deprivation induced eye growth in chickens (Ehrlich et al., 1990) with damage to photoreceptors, amacrine, bipolar cells and a small proportion of ganglion cells, presumably with more damage to OFF type bipolars and amacrines than ON type neurons (Dvorak and Morgan, 1983; Golcich et al., 1990). More recently form deprivation and defocus induced myopia (DIM) have been reported to be inhibited by square wave flicker of 6 Hz or above, while defocus induced hypermetropia (DIH) could only be inhibited by flicker above 12 Hz (Schwahn and Schaeffel, 1997). Schwahn and Schaeffel interpreted this as the extent of inhibition of FDM or DIM (but not DIH) being related to the duty cycle of the flicker. They concluded ‘‘that two different retinal circuits with different temporal characteristics are involved in the processing of hyperopic defocus/deprivation and of myopic defocus, [and only] the first one is dependent on flicker ERG amplitude’’. Perturbation of emmetropia has also been recently reexamined by Winawer and Wallman, in terms of the minimum visual exposure the eye needs to exhibit what they consider the principle components of ocular compensation: a scleral mechanism of altered rate of elongation and altered choroidal thickness. They found that, in chicks, such emmetropization processes can be activated by very brief periods of light in animals otherwise reared in darkness. Chicks required 14 min (made up of 7 2 min periods) of total patterned light experience per day rather than the same duration in 1 or 2 sessions, to initiate sign appropriate compensation to optical defocus (Winawer and Wallman, 2002). Their results also suggest that the refractive response to plus lenses is dominant over
323
that to negative lenses. They also suggested that very brief episodes of experience, even if very frequent, are relatively ineffective, however, they did not reconcile these results with the earlier reports of higher frequency flicker on refractive compensation. Although emmetropization can be activated by extremely short durations of normal light it can also be disrupted in a sign dependent manner by light of normal intensity and duration but with a dynamically changing pattern of spatiotemporal luminance modulation or with differential pharmaceutical activation of retinal ON and OFF pathways (Crewther and Crewther, 2002, 2003). Spatiotemporal manipulation of the environment with a local sawtooth temporal luminance profile, asymmetric with Fast OFFset and Slow ONset at a frequency of 4 Hz, suppressed compensation to þ10 D but not to 10 D defocus, whereas Fast ONset/Slow OFFset inhibited compensation to 10 D defocus but not to þ10 D. By comparison, rearing chicks in an environment of a stationary sawtooth pattern gave almost complete refractive compensation to all lens powers used, indicating that asymmetrical spatial luminance patterns per se are unlikely to affect emmetropization or refractive compensation and that the temporal luminance modulation (TLM) profile is necessary to interfere with refractive compensation (Crewther and Crewther, 2002). Thus it was the aim of the present investigation to further explore the effect on ocular growth and refractive compensation to optical defocus of whole environment low frequency temporal (1, 2 and 4 Hz) non-square-wave luminance modulation. 2. Methods 2.1. Animals In total, 229 day-old male chicks (white-Leghorn blackAustralorp cockerels) obtained from a local commercial hatchery were raised from days 4e10 or days 7e14 post-hatching, with a monocular (right eye) goggle containing one of 10 D, 0 D lens or No Lens (see Table 1 for ages and group sizes). Changes in hatchery production schedules were responsible for differences in rearing days. Within each lens condition, chicks were randomly allocated to one of four light profile conditions for 12 h: constant light (0 Hz), 1 Hz, 2 Hz or 4 Hz. Chicks were housed in batches of 20 with ad libitum access to food and water in a custom, light-sealed, ventilated box. Each batch contained 5 chicks from each of the 4 lens groups. Temperature was maintained at an average of 29 C (2 C) during the experimental period, and illumination was provided by a 12 V, 20 W halogen lamp on a programmed 12 h light/dark day cycle. 2.2. Experimental protocol and biometric measures At the beginning of the experimental period (day 4 or 7) lenses were affixed over the right eye, while the fellow eye was used as a control. Optical defocus was induced using
S.G. Crewther et al. / Experimental Eye Research 83 (2006) 322e328
324
Table 1 Experimental conditions and number per group and age of chicks at time of analysis Light profile
Lens
Experimental duration in days
Age of animal at analysis
Number of chicks
0 Hz
þ10 0 No Lens 10 D
6 6 6 6
10 10 10 10
15 14 14 15
1 Hz
þ10 0 No Lens 10 D
6 6 6 6
10 10 10 10
15 13 14 15
2 Hz
þ10 0 No Lens 10 D
7 7 7 7
14 14 14 14
7 6 20 10
4 Hz
þ10 0 No Lens 10 D
6 6 6 6
10 10 10 10
18 17 19 17
positively or negatively powered (10 D) modified human PMMA contact lenses (8.1 mm in diameter), while 0 D lenses and the No Lens condition were used to assess the effects of lens wear in general. Detailed lens preparation and attachment procedures are found in Crewther and Crewther (2002). During the 12 h light component of the diurnal cycle, environmental luminance was either maintained constantly at 180 lux (0 Hz) or manipulated via TLM of 1, 2 or 4 Hz produced by a function generator linked to a Field Effect Transistor (FET) and a halogen globe. Environmental luminance was measured with a light meter probe fixed at the level of the chick’s eyes attached to an oscilloscope while animals were moving around in the box. Continuous measurements were recorded over 15 min with little variation in the upper and lower limits, which were then read off the oscilloscope. The TLM profile indicated that the flicker consisted of two phases: a light pulse consisting of a gradual rise and slightly sharper decline in luminance between 1.5 and 180 lux (pulse length 250 msec, with full width half maximum of each flash 100 msec) and a second dark (1.5 lux) phase. Thus for each frequency the duration of darkness (duty cycle) varied (see Fig. 1 for flicker profiles). A natural consequence of maintaining the same peak luminance is that the mean luminance varied with frequency. Throughout the experimental period the health of chicks and cleanliness of lenses were monitored twice daily. At the end of each experimental period, chicks were anaesthetized with an intramuscular injection 4.5 mg/kg xylazine: 45 mg/ kg ketamine and both eyes were refracted by retinoscopy (Keeler, Vista Diagnostic Instruments) and axial dimensions were obtained with the average of at least three A-Scan ultrasonography traces (A-Scan Ophthal A-Scan III, TSL: Teknar, Inc. St Louis). Anterior chamber depth was determined by subtracting A-Scan measures of the distance from anterior of the lens to retina from measures of total axial length. Chicks were then euthanized via a barbiturate overdose (Lethobarb;
Fig. 1. Temporal Luminance Modulation (TLM) profiles for 1 Hz (with 750 ms of dark phase per second), 2 Hz (with 500 ms of dark phase per second) and 4 Hz (with no extended dark phase). The pulse profile for all frequencies was maintained at a width of 250 ms. During each flicker cycle, luminance varied between maximum/minimum light levels of 180 lux and 1.5 lux.
Loveridge, Southampton, Hants, UK). All procedures were conducted in strict accordance with La Trobe University Animal Ethics Committee guidelines and adhere to the ARVO Statement for the use of Animals in Ophthalmic and Vision Research. 2.3. Data analysis To minimise variability related to extraneous variables, such as variations in eye size related to overall body weight, refractive and biometric measurements were converted to raw score difference between experimental and fellow eyes. (All mean raw data is presented in Table 2). To assess whether the two different experimental periods affected final refractive status and or axial dimensions, a 2(experimental periods) 4(lens conditions) factorial ANOVA was used to compare the differences between experimental and fellow eyes of chicks reared from days 4e10 and days 7e14 under different lens conditions (10 D, 0 D and No Lens) with constant (dayperiod) illumination at 180 lux (0 Hz condition). No significant differences in either refractive or biometric measures (P > 0.25 for all comparisons) were found between the two experimental periods. Thus, the data from the two rearing time groups were collapsed prior to the statistical analysis of the effect of lens wear under different luminance conditions on refractive sate and ocular growth using a series of ANOVAs. 3. Results As expected, exposure to the 12 h of continuous daylight cycle (0 Hz condition) induced good refractive compensation to positive and negative lenses, and little refractive change in 0 D and No Lens groups (see Fig. 2). As is apparent from Table 2, a series of paired t-tests between the various temporal frequencies of rearing did not show any significant differences in refraction or ocular dimensions for the No Lens group. However, TLM of the environment systematically altered growth compensation in the positive, plano and negative lens groups. TLM at frequencies from 1e4 Hz resulted in
S.G. Crewther et al. / Experimental Eye Research 83 (2006) 322e328
325
Table 2 Means (SEM) of experimental and control eyes for axial length, vitreous chamber depth and anterior chamber length Refractive state
Axial length
Vitreous chamber
Anterior chamber
Experimental
Control
Experimental
Control
Experimental
Control
Experimental
Control
8.57 0.53 13.0 0.94 13.88 1.12 13.12 0.60
0.70 0.31 0.34 0.58 0.15 0.29 0.32 0.32
9.26 0.04 9.67 0.05 9.91 0.11 9.62 0.05
8.80 0.04 9.07 0.04 9.24 0.08 9.00 0.04
5.55 0.05 5.88 0.04 6.00 0.08 5.78 0.05
5.19 0.04 5.41 0.03 5.42 0.05 5.36 0.03
1.49 0.03 1.62 0.03 1.59 0.03 1.57 0.03
1.40 0.02 1.47 0.02 1.49 0.03 1.43 .02
0D 0 Hz 1 Hz 2 Hz 4 Hz
0.16 0.32 6.47 1.59 4.58 0.89 6.21 1.58
0.75 0.18 0.07 0.29 0.90 0.72 0.49 0.31
8.89 0.04 9.29 0.08 9.62 0.10 9.25 0.08
8.84 0.03 8.89 0.09 9.43 0.15 8.96 0.05
5.21 0.04 5.53 0.05 5.61 0.09 5.45 0.07
5.24 0.03 5.31 0.05 5.54 0.09 5.30 0.04
1.45 0.03 1.56 0.05 1.72 0.06 1.54 0.04
1.43 0.02 1.38 0.07 1.56 0.08 1.39 0.02
No Lens 0 Hz 1 Hz 2 Hz 4 Hz
0.70 0.22 0.52 0.40 0.15 0.26 0.07 0.31
0.41 0.24 0.28 0.38 0.44 0.14 0.81 0.27
9.07 0.04 9.08 0.05 9.77 0.05 8.82 0.05
9.03 0.03 9.03 0.06 9.70 0.05 8.79 0.05
5.39 0.03 5.47 0.04 5.80 0.40 5.22 0.04
5.37 0.03 5.42 0.05 5.77 0.04 5.19 0.04
1.39 0.02 1.43 0.02 1.52 0.02 1.37 0.02
1.41 0.02 1.46 0.05 1.54 0.02 1.39 0.02
þ10 D 0 Hz 1 Hz 2 Hz 4 Hz
9.30 0.32 3.45 1.74 0.43 1.90 1.74 1.17
1.15 0.19 0.30 0.03 0.68 0.35 0.39 0.60
8.51 0.06 9.29 0.13 9.15 0.25 8.88 0.08
8.82 0.06 8.99 0.08 9.14 0.19 8.91 0.05
4.80 0.06 5.46 0.08 5.32 0.17 5.17 0.05
5.21 0.05 5.36 0.07 5.37 0.14 5.30 0.05
1.51 0.04 1.61 0.08 1.55 0.08 1.50 0.05
1.45 0.03 1.42 0.03 1.44 0.06 1.39 0.09
10 D 0 Hz 1 Hz 2 Hz 4 Hz
a generalized myopic shift in refractive state and ocular elongation in all lens-treated eyes with no apparent relation between temporal frequency and duty cycle for the 10 or 0 D lens groups. A non-significant trend towards a less myopic shift was seen in refractive compensation by the þ10 D lens group as temporal frequency increased from 1 to 4 Hz. Refractive development was significantly dependent on the interaction between lens induced defocus and the frequency of TLM, F9,212 ¼ 4.39, P < 0.001. A large myopic shift was observed in 10 D treated eyes which developed myopia of 3 D
Difference in Refractive State (D)
10
5 0 Hz 1 Hz 2 Hz 4 Hz
0
-5
-10
-15 - 10 D
0D
No Lens
+ 10 D
Lens Type Fig. 2. Mean difference in refractive state (SEM) for experimental minus fellow refractions for chicks reared using temporal luminance modulation of 0, 1, 2 and 4 Hz and across the different lens groups No Lens, 0 D, þ10 D, 10 D.
exceeding the power of the imposed defocus and 4 D with respect to the no flicker 10 D rearing group. The effect of TLM was even greater on 0 D (nearly 5 D of myopic shift) and þ10 D groups (10 D of myopic shift) with respect to the corresponding 0 Hz lensed groups. Similarly TLM led, not only to the suppression of compensation to positive lenses at frequencies of 2 and 4 Hz, but to an excessive myopic shift in response to both the þ10 D and plano lenses in the 1 Hz condition. Interestingly, TLM did not significantly perturb the normal refractive development of eyes in the No Lens condition, nor the untreated fellow eyes in any condition (see Table 2). Note also the much lower variability in refractions seen in all groups raised in constant light during the day period. Table 2 indicates that the axial lengths of the eyes of the No Lens group are usually equal to or greater than those of the non-lens-wearing eye of the lensed groups. It is also apparent that the anterior chamber depth of non-lens-wearing eyes is smaller than the AC depth of all lens-wearing groups. Changes in refractive states were also accompanied by changes in ocular dimensions including axial and vitreous chamber elongation (Fig. 3). Significant lens-by-light profile interactions were found for both axial length (F9,212 ¼ 4.68, P < 0.001) and vitreous chamber depth (F9,212 ¼ 6.28, P < 0.001). Negative and plano lens-wear led to increased ocular elongation in all flicker conditions; the degree of elongation observed in the negative lens condition was significantly enhanced under TLM, corresponding to the approximately 5 diopter difference in refractive compensation observed in the plano lens group under TLM in comparison to 0 Hz (constant daytime illumination) (see Table 2). Chicks wearing positive lenses showed a significant but varied, increase in axial length and vitreous chamber depth when exposed to TLM of 1 Hz, and 2 and 4 Hz compared to those raised in 0 Hz modulation.
S.G. Crewther et al. / Experimental Eye Research 83 (2006) 322e328
326
Ocular Dimensions (mm)
0.7
0.5
0.3
0.1
-0.1
-0.3
-0.5 4
2
1
Hz
Hz
Hz
Hz
Hz
Hz
Hz
No Lens
0
4
2
Hz
Hz
Hz
Hz
Hz
Hz
Hz
Hz
Hz
0D
1
0
4
2
1
0
4
2
1
0
- 10 D
+ 10 D
Lens Type Fig. 3. Mean differences (SEM) in axial length (black), vitreous chamber depth (grey) and anterior chamber depth (white) between experimental and fellow eyes of chicks across all conditions of temporal luminance modulation and lens wear.
However as can be readily observed in Fig. 3, the vitreous chamber depths cannot account for the total change in axial length. In particular note the plano lens group. A factorial ANOVA examining anterior chamber depth measures revealed a significant main effect for both lens type (F9,212 ¼ 11.05, P < 0.001) and TLM frequency (F9,212 ¼ 2.66, P ¼ 0.049). Anterior chambers were significantly deeper in eyes reared with lenses compared to No Lens groups, and in eyes exposed to 1 Hz flicker compared to other TLM conditions. 4. Discussion This study demonstrates that full-field luminance modulation at low temporal frequencies with variation in mean luminance and duty cycle perturbs axial growth and refractive compensation to optical defocus, resulting in a generalized myopic shift in all lens wearing groups, but not in any of the No Lens groups. While significant in all lens groups, the suppression of compensation was greatest in chicks reared with positive lens defocus. Such a myopic shift by all (i.e. negative, positive and plano) lens wear groups is in the opposite direction to all previous reports of the effects of stroboscopic flicker on refractive compensation to optical defocus, however the statistical strength of the data is undeniable. Currently it is accepted that square wave flicker induces suppression to form deprivation myopia and negative lens myopia at 6 and 12 Hz (Schwahn and Schaeffel, 1997) or at 10, 15 and 20 Hz (Schmid and Wildsoet, 1997) and suppression of positive lens wear at around 12 Hz. By contrast, in our experiment, all groups wearing negative lenses in the presence of low frequency TLM conditions showed similar degrees of enhanced development of myopia beyond the dioptric power of the 10 D lenses, despite variation in duty cycle and mean luminance. Our results also demonstrate suppression of refractive compensation to þ10 D lens wear using the same TLM with the greatest effect observed at 1 Hz modulation. One Hz TLM
with þ10 D defocus not only eliminated the development of DIH, but led to w4 D of myopia while positive lens wear combined with 2 Hz and 4 Hz TLM resulted in minimal generation of hyperopia. Furthermore the pattern of our results for negative, positive, and plano lens groups did not demonstrate a duty cycle relationship with compensation as reported under higher flicker rates (Schwahn and Schaeffel, 1997). It is noteworthy that the myopic shift cannot be merely attributed to differences in the duty cycle of the TLM, as Winawer and Wallman have recently shown that refractive compensation to optical defocus can take place with only 14 min total visual exposure per day (2 min per 2 h 7 times per 14 h day) (Winawer and Wallman, 2002). This is less total light exposure in a day than our 1 Hz TLM animals received in an hour i.e. 15 min (250 msecs per sec per hour). It is also likely that our much greater total duration of visual exposure may account for the fact that although all of our 1 Hz TLM animals received only 250 msec per sec per exposure, this was adequate to give greater than normal refractive compensation to negative lenses and a further sign dependent suppression of positive lens refractive compensation on top of the generalized myopic shift. This also makes it difficult to explain why similar durations of visual exposure under the conditions of TLM for the positive lens group resulted in total suppression of compensation, if as posited (Schwahn and Schaeffel, 1997; Winawer and Wallman, 2002) there are 2 different mechanisms for positive and negative lens refractive compensation and if refractive compensation to plus lens wear is dominant over refractive compensation to negative lens wear (Winawer and Wallman, 2002). As few previous studies have concurrently used both plano lens and No Lens controls and certainly none have reported a difference in relative refraction the question of why it occurs needs to be addressed. In our studies all animals were raised in the same enclosures under the same conditions with lens conditions mixed equally in each rearing batch, so the results are
S.G. Crewther et al. / Experimental Eye Research 83 (2006) 322e328
327
results reflect a strong difference between lens-reared and No Lens groups. Irrespective of the power of lens employed, anterior chambers were deeper after exposure to low frequency TLM in conjunction with lens wearing than TLM in No Lens wearing groups. Simply stated, there is more fluid within the vitreous and anterior chambers of the lensed eyes receiving flicker stimulation, whether or not they are also in receipt of optical defocus. Whether this fluid build up is due to increased secretion or to impaired outflow in the front or back of the eye will require further research. Certainly 1 Hz flicker can be observed to increase pupillary constriction and expansion and this may act to slow aqueous removal.
unlikely to be due to unusual visual degradation of the plano group compared to the others. While an overall myopic shift in refraction could be attributed to improper maintenance of the lens-cleaning regimen resulting in form deprivation myopia, this argument can be rejected on the grounds that the 0 Hz flicker þ10 D group showed excellent compensation, achieving a mean refraction of þ9.3 D of hyperopia. Furthermore, since the lensed and non-lensed groups without flicker showed little variability in refraction or ocular dimensions, the mechanism resulting in excessive myopia appears to be related to the effect of low frequency TLM on lensed eyes. Also, while flicker duty cycle leading to possible imbalance of ON and OFF retinal pathway activation has been shown to have some effects on refractive and growth response in negative lens wear groups (Schwahn and Schaeffel, 1997), the uniform myopic shift across all lensed eyes of all powers is much greater than any variation in duty cycle with temporal frequency (i.e. with duty cycle ranging from 0e750 msec). Three possible mechanisms to explain the results are suggested (all requiring further experimentation): i) peripheral restriction of field through the lens supports leading to myopia; ii) low frequency flicker leading to a resultant increase in the fluid in the anterior chamber, either being due to increased ciliary fluid secretion or restricted outflow possibly caused by the attachment of the lens to the periocular feathers; iii) The nature of the temporal profile of the light pulse, being biased towards a Slow-ON, Fast-OFF profile, may engender the myopia shift and enhance suppression of refractive compensation to positive lenses.
The results presented here demonstrating a much stronger suppression of compensation to positive lens defocus than to negative defocus are reminiscent of the greater suppression of refractive compensation to þ10 D compared to 10 D induced-defocus with low temporal frequency Fast OFF/Slow ON sawtooth luminance profiles (Crewther and Crewther, 2002). Indeed the measured pulse profile used here is biased towards a Fast OFF/Slow ON shape. As all flickered lenses groups showed a myopic shift, it is likely that our results reflect the effects of the temporal contrast modulation profile used. However the absence of an effect on the No Lens group to this flicker profile would require that there is an extra element, such as peripheral occlusion which combines with the effects of temporal profile.
4.1. Peripheral restriction
5. Conclusion
The two layers of matched Velcro we use to hold our spectacle lens reduces the field of view from 180 to 90 and may play a role in myopic shifts in refraction. This is similar to a much earlier suggestion (Nathan et al., 1985) that peripheral dysfunction through visual disease may play a part in inducing myopic shifts in refractions in children with low vision. Nathan et al. observed an association between more myopic refractions and disease with a peripheral or peripheral plus central impairment of vision whereas those conditions in which foveal vision was primarily impaired showed a mild hypermetropic trend. Also, it has been recently reported that peripheral vision restriction perturbs emmetropization in monkeys and induces a slight myopic shift (Smith et al., 2005). Thus, a similar aetiolgy for the myopic shift seen in all our lens groups is possible but this explanation still cannot account for the fact that the plano lens group (0 Hz) did not show the myopic shift.
Positive, plano and negative lens wear in the presence of a non-square wave profile low frequency (1, 2 and 4 Hz) temporal luminance modulation results in a general myopic shift in refractive compensation and enhanced ocular growth. The results also highlight the role of abnormal temporal modulation in accurate refractive compensation and emphasize the interaction between flicker induced changes in the activation of the ON and OFF retinal pathways and lens rearing.
4.2. Flicker induced fluid and aqueous outflow restriction Consistent with previous findings, refractive changes were accompanied by changes in ocular growth patterns, both in vitreous and anterior chamber depth. The anterior chamber depth
4.3. Temporal modulation profile
References Barutchu, A., Crewther, S.G., Crewther, D.P., 2002. Effects of optical defocus and spatial contrast on anterior chamber depth in chicks. Clin. Experiment. Ophthalmol. 30, 217e220. Beresford, J.A., Crewther, S.G., Kiely, P.M., Crewther, D.P., 2001. Comparison of refractive state and circumferential morphology of retina, choroid, and sclera in chick models of experimentally induced ametropia. Optom. Vis. Sci. 78, 40e49. Crewther, D.P., 2000. The role of photoreceptors in the control of refractive state. Prog. Retin. Eye Res. 19, 421e457. Crewther, D.P., Crewther, S.G., 1990. Pharmacological modification of eye growth in normally reared and visually deprived chicks. Curr. Eye Res. 9, 733e740. Crewther, D.P., Crewther, S.G., 2002. Refractive compensation to optical defocus depends on the temporal profile of luminance modulation of the environment. Neuroreport 13, 1029e1032.
328
S.G. Crewther et al. / Experimental Eye Research 83 (2006) 322e328
Crewther, D.P., Crewther, S.G., Xie, R.Z., 1996. Changes in eye growth produced by drugs which affect retinal ON or OFF responses to light. J. Ocul. Pharmacol. Ther. 12, 193e208. Crewther, S.G., Crewther, D.P., 2003. Inhibition of retinal ON/OFF systems differentially affects refractive compensation to defocus. Neuroreport 14, 1233e1237. Dvorak, D.R., Morgan, I.G., 1983. Intravitreal kainic acid permanently eliminates off-pathways from chicken retina. Neurosci. Lett. 36, 249e253. Ehrlich, D., Sattayasai, J., Zappia, J., Barrington, M., 1990. Effects of selective neurotoxins on eye growth in the young chick. Ciba Found. Symp. 155, 63e84. discussion 84e68. Fredrick, D.R., 2002. Myopia. BMJ 324, 1195e1199. Golcich, M.A., Morgan, I.G., Dvorak, D.R., 1990. Selective abolition of OFF responses in kainic acid-lesioned chicken retina. Brain Res. 535, 288e300. Gottlieb, M.D., Marran, A., Xu, A., Nickla, D.L., Wallman, 1987. The emmetropization process in chicks is compromised by dim light. Invest. Ophthalmol. Vis. Sci. 32 (Suppl.), 2616. Irving, E.L., Callender, M.G., Sivak, J.G., 1991. Inducing myopia, hyperopia, and astigmatism in chicks. Optom. Vis. Sci. 68, 364e368. Irving, E.L., Sivak, J.G., Callender, M.G., 1992. Refractive plasticity of the developing chick eye. Ophthalmic. Physiol. Opt. 12, 448e456. Lauber, J.K., Oishi, T., 1987. Lid suture myopia in chicks. Invest. Ophthalmol. Vis. Sci. 28, 1851e1858. Napper, G.A., Brennan, N.A., Barrington, M., Squires, M.A., Vessey, G.A., Vingrys, A.J., 1995. The duration of normal visual exposure necessary
to prevent form deprivation myopia in chicks. Vision Res. 35, 1337e 1344. Nathan, J., Kiely, P.M., Crewther, S.G., Crewther, D.P., 1985. Disease-associated visual image degradation and spherical refractive errors in children. Am. J. Optom. & Physiol. Opt. 62, 680e688. Schaeffel, F., Glasser, A., Howland, H.C., 1988. Accommodation, refractive error and eye growth in chickens. Vision Res. 28, 639e657. Schmid, K.L., Wildsoet, C.F., 1997. Contrast and spatial-frequency requirements for emmetropization in chicks. Vision Res. 37, 2011e2021. Schwahn, H.N., Schaeffel, F., 1997. Flicker parameters are different for suppression of myopia and hyperopia. Vision Res. 37, 2661e2673. Smith, E.L.d., Kee, C.S., Ramamirtham, R., Qiao-Grider, Y., Hung, L.-F., 2005. Peripheral vision can influence eye growth and refractive development in infant monkeys. Invest. Ophthalmol. Vis. Sci. 46, 3965e3972. Vingrys, A.J., Squires, M.A., Napper, G.A., Barrington, M., Vessey, G.A., Brennan, N.A., 1991. Prevention of form deprivation myopia in post-hatch chickens. Invest. Ophthal. Vis. Sci. 32 (Suppl.), 1203. Wallman, J., Turkel, J., Trachtman, J., 1978. Extreme myopia produced by modest change in early visual experience. Science 201, 1249e1251. Wallman, J., Winawer, J., 2004. Homeostasis of eye growth and the question of myopia. Neuron 43, 447e468. Wildsoet, C., Wallman, J., 1995. Choroidal and scleral mechanisms of compensation for spectacle lenses in chicks. Vision Res. 35, 1175e1194. Winawer, J., Wallman, J., 2002. Temporal constraints on lens compensation in chicks. Vision Res. 42, 2651e2668.