Current perspective on microfluctuations of accommodation Barry Winn* Department of Optometry and Vision Science Glasgow Polytechnic, Cowcaddens Road, Glasgow G4 OBA. UK Bernard Gilmartint Ophthalmic and Physiological Optics Research Group.. Department of Vision Sciences. Aston University, Aston Triangle, Birmingham B4 7ET. UK (Received 8 October 1991)
The collaboration of Fergus Campbell, Gerald Westheitner and John Robson in the 1950s produced insight into the nature of accommodation microfluctuations and instigated work which has led to the current view that the nominally steady-state accommodation response exhibits temporal variations which can be characterized by two dominant regions of activity: a low-frequency component (LFC < 0.6 Hz) and a high-frequency component (HFC ^ 1.0 ^ 2.3 Hz). A functional role has been attributed to these microfluctuations as they offer a means by which a directional cue could be derived from an even-error stimulus. However, there is no consensus regarding the respective contribution made by each of the dominant components in the accommodation control process. Using a newly-designed measurement and recording system we have conducted a series of experiments to investigate the nature and aetiology of the microfluctuations. The incidence and magnitude of microfiuctuations in LFCs and HFCs for central and peripheral lens zones were investigated while five young emmetropic subjects viewed a near target. The form of the power spectra of the fluctuations was found to be similar for central and peripheral zones although an overall reduction in magnitude was observed in the periphery. The HFCs are thus a consistent feature of microfluctuations in central zones and not. as previously suggested, merely a spurious feature of peripheral zones. A significant bet ween-subject variation in the location of HFCs was found and led us to consider the relationship between HFCs and other physiological systems which provide intraocular rhythmic variation. Simultaneous measurements of ocular accommodation and systemic arterial pulse made on 20 subjects demonstrate that the location of the HFC is significantly correlated with arterial pulse frequency. The intraocular mechanisms by which arterial pulse modulates the HFC are likely to involve rhythmic changes in both choroidal blood flow and intraocular pressure (IOP). Consequently, examination ofthe effect of 0.5% timolol maleate. a drug which exerts an ocular hypotensive effect and also induces a reduction in IOP pulse, on the accommodation microfluctuations of 10 emmetropic subjects when fixating a near target was undertaken. A double-blind protocol against saline showed timolol to reduce significantly the root-mean-square value ofthe fluctuations. The results thus indicate that factors relating to IOP and ocular vasculature will aflect the magnitude and frequency composition of accommodative fluctuations. The studies to date suggest that the complex waveform of the accommodative microfluctuations is a consequence of the combination of neurological control and physiological rhythmic variation: the former is probably attributable to the LFCs: the latter to arterial pulse.
The temporal instability in ocular accommodation has attracted the interest of numerous investigators over the last 50 years. The pioneering work of Collins using an ingenious electronic refractometer' was followed by the more systematic investigations of Arnulf and his colleagues using an elaborate double-pass ophthalmoscopic method to observe directly, or photographically. the temporal variations in retinal image quality of gratings"'^. The technique was later refined using an image intensifier* but the results of these early *MBCO. + FBCO.
investigations suggested that, rather than being a spurious characteristic of the accommodation response, microfluctuations could provide a mechanism for maintaining optimum mean response levels by monitoring small chatiges in contrast ofthe retinal image. Their potential as ati important component of the accommodative system attracted a number of further studies which used alternative methods of measurement, for example measurement of the small changes of curvature of the anterior surface of the lens and cine-photography of variations in the third Purkinje image^. It was, however, the collaboration of Fergus Campbell, Gerald Westheimer and John Robson in the late 1950s that produced special
252
© 1992 Butterworth-Heitiemann for British College of Optometrists 0275-5408/92/020252-05
Ophthal. Physiol. Opt., 1992. Vol. 12, April
Microfluctuations of accommodation: B. Winn and B. Gilmartin insight into the nature of accommodative microfluctuations^"'•\ Their objective continuously-recording infra-red optometer had sufficient sensitivity and time resolution to demonstrate not only the magnitude of microfluctuations but also their temporal characteristics and importantly their frequency spectra. A number of research laboratories have now consolidated their observation that the nominally steady-state accommodation response exhibits temporal variations which can be characterized by two dominant regions of activity: a low frequency component (LFC < 0-6 Hz) atid a high frequency component (HFC, 1.0-2.3 Hz I ' - ^ ' ^ ' M h e s e microfluctuations occur with a root-mean-square (r.m.s.) value of ^0.02 0.2 D at temporal frequencies of < 5 Hz and are positively correlated with increases in stimulus vergence''*''"''. A functional role has been attributed to these microfluctuations, as they offer a means by which a directional cue could be derived from an even-error stimulus'^ '". However, there is no consensus regarding the respective contribution made by each of the dominant components in the accommodation control process. With small pupils, HFCs are less pronounced and LFCs more obvious, the reverse being found with larger pupils'^, which suggests that the LFCs have a role in the accommodation feedback loop. This finding would support LFCs as a system for monitoring and maintaining retinal image contrast at an optimum level during steady-state accommodation. In contrast, the diminished fluctuations in HFCs evident with reduced pupil size has led to the proposal that they may not contribute directly to accommodative controi but may simply be a form of 'plant noise" derived from the mechanical and elastic properties of exposed parts ofthe peripheral lens and its support structure'*^. The source of HFCs is of particular interest as their characteristics do not appear to be related to changes in stimulus parameters. Using a newly-designed measurement and recording system a series of experimetits has been conducted to investigate the nature and aetiology ofthe microfluctuations.
Methods Accommodation microfluctuations were measured with a modified Canon Autoref R1 (IR) objective optometer^". The analogue output was fed into a digital storage oscilloscope which was connected to an on-line computer (IBM PC clone) through an IEEE-488 interface. A minimutn of five accommodation signals, each of 10 s duration, were collected at a sampling rate of 102.4 Hz in each of the studies described below. The data were smoothed with a high frequency cut at 10 Hz and a power spectrum was calculated for each individual trace with a frequency resolution of 0.1 Hz. Potential artefacts associated with changes in pupil size"' and eye movements were taken into account. Experiment 1 The incidence and magnitude of microfluctuations in LFCs and HFCs for central and peripheral lens zones were investigated while five young emmetropic subjects viewed a near target^^. To allow recording through separate lens regions it was necessary to dilate the pupil
of each subject with the mydriatic 10% phenylcphrine HCI. This procedure results in a slight reduction in accommodative amplitude (0.8 ± 0.3 D)^-* but allows suflicient reserve for young subjects to comfortably fixate a stimulus placed at a vergetice of - 4 . 0 D. The mean pupillary dilation of 8.5 mm induced was insuflicient to allow measurement through entirely discrete lens zones as the optometer utilizes a circular region of 3.52 mm diameter in continuously recording mode^'. An overlap of up to 10% in zone area was thus tolerated throughout the investigation. Power spectrum analysis was used to compare the frequency components oi^the accommodation microfluctuations for central and periperal lens zones. The form of the power spectra was found to be similar for central and peripheral zones although an overall reduction in magnitude was observed in the periphery (Figure I). The HFCs are thus a consistent feature of microfluctuations in central zones and not merely a spurious feature of peripheral zones. Although variation in the magnitude of dioptric change may be present across the lens surface the frequeticy characteristics are the same in all regions of the lens. This is perhaps not surprising as it is difficult to conceive of a mechatiism that would produce regions where the fluctuations occur at differing I'requencies. The results comply with mathematical models-*-^ which suggest that the lens capsule acts as a force distributor across the lens surface.
Experiment 2 The source of the HFCs is of particular interest as their characteristics do not appear to be reiated to changes in stimulus parameters^*". In this respect, the significant between-subject variation in the location of HFCs found in Experiment I led to consideration"^ ofthe relationship between HFCs and other physiological systems which provide intraocular rhythmic variation. The accommodative microfluctuations could be the result of a combination of both neurological control and localized plant noise. Simultaneous measurements of ocular accommodation and systemic arterial pulse made on 20 subjects demonstrated that the location of the HFC is significantly correlated (r = 0.99, P < 0.001) with arterial pulse frequency {Figure 2). The unusually high values of arterial pulse observed in many subjects could in part be attributable to the experimental conditions (e.g. dental bite and head restraint) which may have caused previous researchers to overlook this relationship. The correlation between arterial pulse frequency and the HFC was retained during the recovery phase of exercise-induced changes in pulse rate on three young subjects (F/(/wr£'.?). Additionally, the relationship between arterial pulse and the HFC was demonstrable in the sound eye of a 40-year-old unilateral aphakic but was absent from the aphakic eye^''. It seems clear that the source of HFCs is principally derived from the effect of arterial pulse on the accommodative plant as power changes associated with microdeformation ofthe eyebail in response to heartbeat which have been previously described"** are an order of magnitude less than those reported here.
Ophthal. Physiol. Opt., 1992, Vol. 12, April
253
Microfluctuations of accommodation: B. Winn and B. Gilmartin
5.40 il.2O Q •J
'o3.
00
X
.80
0.60
'. 0.30
0.90
1 .50 2.10
2.70
Frequency (Hz)
5.110-
5.40
Temporal lens
5.40-
Central lens
i
Nasal
ens
N
^4.20 3.00(-
^
3.00
0.30 -
!. '
"i
1 .80 •
0.60
Ar:.. A 0.30
0.90
1 .50
Frequency
2.10
0.60
0.60
. • 2.70
0.30
(Hz)
0.90
1.50
Frequency
5.40
2.10
2.70
(Hz)
^ -r P \-f 0.30
0.90
I.50
Frequency
t^.-=^ .... 2.10
2.70
(Hz)
Inferior lens
^4.20
7^3.00 X
aj 1 .80
o n. 0.60 0.30 0.90
1 .50
Frequency
2.10
2.70
(Hz)
Figure 1 Power spectra of the accommodalion microfluctuations are presented for central and peripheral zones of the crystalline lens, each with 10 degrees of freedom, for subject N.F, (age 21 years); , power spectrutn for the central region; — , for the peripheral lens zone indicated. Reproduced with permissioti from Wjnn e\ al.. ^^
Experiment 3 The intraocular mechanisms by which arterial pulse modulates the HFC are likely to involve rhythmic changes in both choroidal blood flow and intraocular pressure (IOP). In addition to reducing IOP the magnitude of IOP pulse is also reduced significantly by beta-receptor antagonists possibly due to their vasoconstrictive action or the choroidal vasculature^''. Consequently, the effect of the beta-antagonist 0.5Vo timolol maleate on the accommodation microfluctuations of 10 emmetropic subjects when fixating a near target was investigated-'". Following a double-blind protocol against saline, timolol was shown to reduce the r.m.s. value ofthe fluctuations (Figure 4). A reduction in the peak location of the HFC was observed in all subjects following treatment with timoiol and was correlated with the 4 6 beats min ~ ^ reduction in arterial pulse known to result from systemic absorption of timolol via ocular routes. To separate local from systemic effects, an experiment
254
Ophthai. Physiol. Opt., 1992, Vol. 12, April
was performed using four subjects whose treated and untreated eyes were examined to determine the effects of the consensual reduction in IOP and the position ofthe HFC. Results in the treated eyes were the same as in the previous experiment but for the untreated eyes the results were equivocal although the correlation between arterial pulse frequency and location ofthe HFC was maintained. Overall, the results indicate that factors relating to IOP and ocular vasculature will aflect the magnitude and frequency composition of accom;nodative fluctuations and suggest that measurement of microfluctuations may have potential as a non-invasive method of assessing the eflect of beta-adrcnergic antagonists on the eye.
Conclusions The studies to date suggest that the complex waveform ofthe accommodative microfluctuations is a consequence of the combination of neurological control and physiological rhythmic variation: the former is probably
Microfiuctuatiom of accommodation: B. Winn and B. Gilmariin attributable to the LFCs: the latter to arterial pulse. A functional role for the fluctuations as an error-deteetor in the accommodation system is, therefore, probably related to the maintenance of focus on a stationary target as the neurologically controlled component is too slow to provide the necessary information to optimize the response when rapid step changes in stimulus vergence occur. Although the HFCs may still be utilized by the contrast detection mechanism as part of the aggregate response, it would appear that only the LFCs could provide the necessary neurological control. If the accommodation controller operates to maintain a eonstant r.m.s. outptit to monitor retinal image contrast we would envisage that the magnitude of the neuroiogicaily-controlled LFC
eould be dependent upon the systemieally-modulated HFC. It is conceivable, therefore, that a substantial increase in the HFC could swamp the potential accommodative control offered by the LFC. This suggests that investigation of visual performance in relation to systemic variations of arterial pulse^' may be of importanee in identifying eausal agents of visual stress during sustained visual tasks. Acknowledgements This work was supported in part by the Visual Research Trust which we gratefully acknowledge. We also thank
2.4 2.2r2.2 2.02.0
1 .6
1 .4 -
1 .2
1 .2 }.2
1 .0 1 .0
1 .2
1.6
1 .4
2.0
Kigure 2 Corrclalioii between high-frequency component for P < 0.001 ). The regression line is with permission from Winn f( at.
arterial pulse frequency and the group data {n = 20: r = 0.99, y = 0.0604 + 0.9516 x. Reproduciid 1990'^
0.09
1.6
1.3
2.0
2.2
2.4
A c c o m m o d a t i o n high frequency component ( H z )
2.2
Accommodation high frequency component (Hz)
1.4
Figure 3 Correlation between arterial pulse frequency and the high-frequency component for the recovery phase (sequence 1-5) of exercise-induced changes in pulse. The subject was a 21-year-old woman whose base-line pulse was 1.5 Hz under the cxpcrimcntai conditions and was typical of the sample. Reproduced with permission from Winn er al. 1990"
0.09 Subject
^0.07
0.07
0.05
0.05
0.03
0.03
0.01
0.01
B.E,
- I '
0.30
0.90
1.50
2.10
Frequency (Hz)
2.70
0.30
0.90
1 .50
Frequency
2.10
2.70
(Hz)
Figure 4 Power spectra of the accommodation microfluctuations, each with 16 degrees of freedom, for two subjects whose eyes have been treated with timolol maleate 0.5%. Timolol induces iin overall reduction in power and a frequency shift in the high-frequency component which is consistent with a reduction in pulse rate of 6 beats min '; . timolol; , post-limolol. Reproduced with permission from Owens et al.. 1991'"
Ophthal. Physiol. Opt., 1992, Vol. 12, April
255
Microfluctuations of accommodation: B. Winn and B. Gilmartin Merck. Sharp and Dohme for the supply of timolol maleate and Drs H. Owens and J. R. Pugh for Iheir collaboration in the studies.
17. 18.
References 1. Collins. G. The electronic refractometer. Br. J. PhysinL Op!. I. 30 40 (1937). 2. Arnulf. A., Dupuy, O. and Klamnnl. F. LL-S micro liuclu at ions d'accommodation de Toeil el Facuite visuelle pour les diametres pupillaircsnalurelles.C .«.--1(W. S(/./'(//•/.S232.339-341 I 1951 ), 3. Arnulf. A., Dupuy. O. and Flamanl. F. Les microfluclualions de l'oeil el leur inlluence sur I'image retiniene. C. R. Hehd. Seancf.s Acad. SCI. Pans 232. 43S-450 (195! ). 4. Arnulf. A.. Santamaria, J. and Bescos. J. A cincmalogriiphic melhod for the dynamic sludy of the image formation by the human eye. Microiluctuations of accommodation. J. Optics (Paris) 12, 123-12K (1981). 5. Whiteside. T. C. D. Problems of Vi.sicn at Hiijh .4lii/iitli:s. Butterworths. London, p. 193 (1957). 6. Campbell, F. W. Correlation of accommodation between the two eyes. ./. Opi. Soc. Am. 50. 738 (1960). 7. Campbell, F. W. The accommodation response of the human eye. Br. J. Physiol. Opi. 16, 188 203 (i960). 8. Campbell. F. W. and Robson. J. G. High-speed infra-red optometer. J. Opi. .Sac. Am. 49. 268-272 (1959). 9. Ciimpbell, F. W. and Westheimer. G. Sensitivity of the eye to difTerenccs in focus. J. Phy.siol. 143. !8P |I958). 10. Campbell, F. W. and Westheimer. G. Factors inHueneing the accommodation responses of the human eye. J. Opi. Soc. .iDi. 49. 568-571 (1959). 11. Campbell. F. W. and Weslheimer. G, Dynamics of the accommodation responses of the human eye. J. Physio!. 151. 285-295 (1960). !2. Campbell. F. W.. Westheimer. G. and Robson, J. G. Sijiiiificance offlucttiations of accommodation. J. Opi. Soc. Am.4S.bb9 (1959). 13. Campbell, F. W.. Robson. J. G. and Westheimer, G. Fluctuations of accommodation under steady viewing conditions. J. Physial. 145. 579-594 (1959). 14. Denieul. P. Effects of stimulus vergence on near accommodation response, microlluetuations and accommodalion and optical quality of the human eye. \'isi(m Res. 23. 561-569 (1982). 15. Kotulak. J. C. and Schor. C. M. Temporal variations in accommodation during steady-slate conditions. J. Opi. Soc. Am. .4 3.223-227 (1986). 16. Alpern, M. Variability of accommodation during steady fixation
256
Ophthal. Physiol. Opt.. 1992, Vol. 12, April
19. 20 21,
22,
23, 24, 25 26. 27. 28. 29 30.
31.
at various levels of illuminance. J. Opi. Soc. Am. 48, 193-197 (1948). Crane, H. D. A theoretical analysis of the visual aecommodation system in humans. Report NASA CR-606. NASA. Washington (1966). Kotuiak, J. C. and Schor, C. M. A computational model of the error deteetor of human visual accommodation. Biol. Cyhvrii. 54, 189-194 (1986). Charman, W. N, The retinal image in the human eye. In Proijress in Retimil Re.seurch (eds N. Osborne. and G. Chader), Vol. 2, Pergamon. Oxford, pp. 1-50 (1983). Pugh. J. R. and Winn, B. Modification of the Ciinon Autoref RI for use as a eontinuously recording optometer. Ophlhal. Physiol. 0/J/. 9, 451-454 (1988). Winn, B,. Pugh, J- R-. Giimartin, B. and Owens, H. The effect of pupil size on static and dynamic measurements of accommodation using an infra-red optometer. Ophihat. Phv.m>l. Opt. 9. 277-283 (1989). Winn.B., Pugh, J. R., Gilmartin, B. and Ov^ens, H. The frequency characteristics of accommodative microfluctuations for central and peripheral zones of the human crystalline lens. Vi.sion Res. 30, 1093 1099 (1990). Rosenfield, M., Giimartin, B., Cunningham, E. and Oattani, N. The infiuence of alpha-;tdrenergic agents on tonic accommodation. Curr. Eye Res. 9. 267-272 (1990). Koretz, J. F. and Handeiman. G. H. Model of the accommodative meehanism in the human eye. I'ision Res. 22, 917-942 (1982). Koretz, J. F. and Handeiman. G. H. A model for aecommodation in the young human eye: the effects of lens anisotropy on the mechanism. Vi.sion Res. 23, 1679 1686 (1983). Denieul, P. and Corno. F. Accommodation et contraste. L'oplomelrie 32. 4-8 (1986), Winn, B., Pugh, J. R., Gilmartin, B. and Owens. H. Arterial pulse modulates steady-state ocular accommodation. Curr. Eve Res 9 971-975(1990). Fercher, A. F., Hu, H. Z., Steeger, P. F. and Briers, J. D. Eye deformation measurement by la.ser interferometry. Optica Acia. 29, 1401-1406 (1982). Colloton, E. and Perkins, E. S. The effect of timolol on the intraocular pulse pressure. Invest. Ophihalmol. Vi.sual Sci. 27(Suppl.), 102 (1986). Owens, H., Winn, B., Gilmartin, B. and Pugh, J. R. Effect of a topieal beta-adrenergic receptor antagonist on the dynamics of steady-state accommodation. Ophlhal. Phvsiol. Opt. 11, 99-104 Gros. B. L. and Cohn.T. E. The phase of the ocular arterial pulse affects deteclability of a brief threshold stimulus, /nrest. Ophthalmol. Visual Sci. 32(Suppl.), 699 (1991 ).
The collaboration of Fergus Campbell, Gerald Westheitner and John Robson in the 1950s produced insight into the nature of accommodation microfluctuations and instigated work which has led to the current view that the nominally steady-state accommodation response exhibits temporal variations which can be characterized by two dominant regions of activity: a low-frequency component (LFC < 0.6 Hz) and a high-frequency component (HFC ^ 1.0 ^ 2.3 Hz). A functional role has been attributed to these microfluctuations as they offer a means by which a directional cue could be derived from an even-error stimulus. However, there is no consensus regarding the respective contribution made by each of the dominant components in the accommodation control process. Using a newly-designed measurement and recording system we have conducted a series of experiments to investigate the nature and aetiology of the microfluctuations. The incidence and magnitude of microfiuctuations in LFCs and HFCs for central and peripheral lens zones were investigated while five young emmetropic subjects viewed a near target. The form of the power spectra of the fluctuations was found to be similar for central and peripheral zones although an overall reduction in magnitude was observed in the periphery. The HFCs are thus a consistent feature of microfluctuations in central zones and not. as previously suggested, merely a spurious feature of peripheral zones. A significant bet ween-subject variation in the location of HFCs was found and led us to consider the relationship between HFCs and other physiological systems which provide intraocular rhythmic variation. Simultaneous measurements of ocular accommodation and systemic arterial pulse made on 20 subjects demonstrate that the location of the HFC is significantly correlated with arterial pulse frequency. The intraocular mechanisms by which arterial pulse modulates the HFC are likely to involve rhythmic changes in both choroidal blood flow and intraocular pressure (IOP). Consequently, examination ofthe effect of 0.5% timolol maleate. a drug which exerts an ocular hypotensive effect and also induces a reduction in IOP pulse, on the accommodation microfluctuations of 10 emmetropic subjects when fixating a near target was undertaken. A double-blind protocol against saline showed timolol to reduce significantly the root-mean-square value ofthe fluctuations. The results thus indicate that factors relating to IOP and ocular vasculature will aflect the magnitude and frequency composition of accommodative fluctuations. The studies to date suggest that the complex waveform of the accommodative microfluctuations is a consequence of the combination of neurological control and physiological rhythmic variation: the former is probably attributable to the LFCs: the latter to arterial pulse.
The temporal instability in ocular accommodation has attracted the interest of numerous investigators over the last 50 years. The pioneering work of Collins using an ingenious electronic refractometer' was followed by the more systematic investigations of Arnulf and his colleagues using an elaborate double-pass ophthalmoscopic method to observe directly, or photographically. the temporal variations in retinal image quality of gratings"'^. The technique was later refined using an image intensifier* but the results of these early *MBCO. + FBCO.
investigations suggested that, rather than being a spurious characteristic of the accommodation response, microfluctuations could provide a mechanism for maintaining optimum mean response levels by monitoring small chatiges in contrast ofthe retinal image. Their potential as ati important component of the accommodative system attracted a number of further studies which used alternative methods of measurement, for example measurement of the small changes of curvature of the anterior surface of the lens and cine-photography of variations in the third Purkinje image^. It was, however, the collaboration of Fergus Campbell, Gerald Westheimer and John Robson in the late 1950s that produced special
252
© 1992 Butterworth-Heitiemann for British College of Optometrists 0275-5408/92/020252-05
Ophthal. Physiol. Opt., 1992. Vol. 12, April
Microfluctuations of accommodation: B. Winn and B. Gilmartin insight into the nature of accommodative microfluctuations^"'•\ Their objective continuously-recording infra-red optometer had sufficient sensitivity and time resolution to demonstrate not only the magnitude of microfluctuations but also their temporal characteristics and importantly their frequency spectra. A number of research laboratories have now consolidated their observation that the nominally steady-state accommodation response exhibits temporal variations which can be characterized by two dominant regions of activity: a low frequency component (LFC < 0-6 Hz) atid a high frequency component (HFC, 1.0-2.3 Hz I ' - ^ ' ^ ' M h e s e microfluctuations occur with a root-mean-square (r.m.s.) value of ^0.02 0.2 D at temporal frequencies of < 5 Hz and are positively correlated with increases in stimulus vergence''*''"''. A functional role has been attributed to these microfluctuations, as they offer a means by which a directional cue could be derived from an even-error stimulus'^ '". However, there is no consensus regarding the respective contribution made by each of the dominant components in the accommodation control process. With small pupils, HFCs are less pronounced and LFCs more obvious, the reverse being found with larger pupils'^, which suggests that the LFCs have a role in the accommodation feedback loop. This finding would support LFCs as a system for monitoring and maintaining retinal image contrast at an optimum level during steady-state accommodation. In contrast, the diminished fluctuations in HFCs evident with reduced pupil size has led to the proposal that they may not contribute directly to accommodative controi but may simply be a form of 'plant noise" derived from the mechanical and elastic properties of exposed parts ofthe peripheral lens and its support structure'*^. The source of HFCs is of particular interest as their characteristics do not appear to be related to changes in stimulus parameters. Using a newly-designed measurement and recording system a series of experimetits has been conducted to investigate the nature and aetiology ofthe microfluctuations.
Methods Accommodation microfluctuations were measured with a modified Canon Autoref R1 (IR) objective optometer^". The analogue output was fed into a digital storage oscilloscope which was connected to an on-line computer (IBM PC clone) through an IEEE-488 interface. A minimutn of five accommodation signals, each of 10 s duration, were collected at a sampling rate of 102.4 Hz in each of the studies described below. The data were smoothed with a high frequency cut at 10 Hz and a power spectrum was calculated for each individual trace with a frequency resolution of 0.1 Hz. Potential artefacts associated with changes in pupil size"' and eye movements were taken into account. Experiment 1 The incidence and magnitude of microfluctuations in LFCs and HFCs for central and peripheral lens zones were investigated while five young emmetropic subjects viewed a near target^^. To allow recording through separate lens regions it was necessary to dilate the pupil
of each subject with the mydriatic 10% phenylcphrine HCI. This procedure results in a slight reduction in accommodative amplitude (0.8 ± 0.3 D)^-* but allows suflicient reserve for young subjects to comfortably fixate a stimulus placed at a vergetice of - 4 . 0 D. The mean pupillary dilation of 8.5 mm induced was insuflicient to allow measurement through entirely discrete lens zones as the optometer utilizes a circular region of 3.52 mm diameter in continuously recording mode^'. An overlap of up to 10% in zone area was thus tolerated throughout the investigation. Power spectrum analysis was used to compare the frequency components oi^the accommodation microfluctuations for central and periperal lens zones. The form of the power spectra was found to be similar for central and peripheral zones although an overall reduction in magnitude was observed in the periphery (Figure I). The HFCs are thus a consistent feature of microfluctuations in central zones and not merely a spurious feature of peripheral zones. Although variation in the magnitude of dioptric change may be present across the lens surface the frequeticy characteristics are the same in all regions of the lens. This is perhaps not surprising as it is difficult to conceive of a mechatiism that would produce regions where the fluctuations occur at differing I'requencies. The results comply with mathematical models-*-^ which suggest that the lens capsule acts as a force distributor across the lens surface.
Experiment 2 The source of the HFCs is of particular interest as their characteristics do not appear to be reiated to changes in stimulus parameters^*". In this respect, the significant between-subject variation in the location of HFCs found in Experiment I led to consideration"^ ofthe relationship between HFCs and other physiological systems which provide intraocular rhythmic variation. The accommodative microfluctuations could be the result of a combination of both neurological control and localized plant noise. Simultaneous measurements of ocular accommodation and systemic arterial pulse made on 20 subjects demonstrated that the location of the HFC is significantly correlated (r = 0.99, P < 0.001) with arterial pulse frequency {Figure 2). The unusually high values of arterial pulse observed in many subjects could in part be attributable to the experimental conditions (e.g. dental bite and head restraint) which may have caused previous researchers to overlook this relationship. The correlation between arterial pulse frequency and the HFC was retained during the recovery phase of exercise-induced changes in pulse rate on three young subjects (F/(/wr£'.?). Additionally, the relationship between arterial pulse and the HFC was demonstrable in the sound eye of a 40-year-old unilateral aphakic but was absent from the aphakic eye^''. It seems clear that the source of HFCs is principally derived from the effect of arterial pulse on the accommodative plant as power changes associated with microdeformation ofthe eyebail in response to heartbeat which have been previously described"** are an order of magnitude less than those reported here.
Ophthal. Physiol. Opt., 1992, Vol. 12, April
253
Microfluctuations of accommodation: B. Winn and B. Gilmartin
5.40 il.2O Q •J
'o3.
00
X
.80
0.60
'. 0.30
0.90
1 .50 2.10
2.70
Frequency (Hz)
5.110-
5.40
Temporal lens
5.40-
Central lens
i
Nasal
ens
N
^4.20 3.00(-
^
3.00
0.30 -
!. '
"i
1 .80 •
0.60
Ar:.. A 0.30
0.90
1 .50
Frequency
2.10
0.60
0.60
. • 2.70
0.30
(Hz)
0.90
1.50
Frequency
5.40
2.10
2.70
(Hz)
^ -r P \-f 0.30
0.90
I.50
Frequency
t^.-=^ .... 2.10
2.70
(Hz)
Inferior lens
^4.20
7^3.00 X
aj 1 .80
o n. 0.60 0.30 0.90
1 .50
Frequency
2.10
2.70
(Hz)
Figure 1 Power spectra of the accommodalion microfluctuations are presented for central and peripheral zones of the crystalline lens, each with 10 degrees of freedom, for subject N.F, (age 21 years); , power spectrutn for the central region; — , for the peripheral lens zone indicated. Reproduced with permissioti from Wjnn e\ al.. ^^
Experiment 3 The intraocular mechanisms by which arterial pulse modulates the HFC are likely to involve rhythmic changes in both choroidal blood flow and intraocular pressure (IOP). In addition to reducing IOP the magnitude of IOP pulse is also reduced significantly by beta-receptor antagonists possibly due to their vasoconstrictive action or the choroidal vasculature^''. Consequently, the effect of the beta-antagonist 0.5Vo timolol maleate on the accommodation microfluctuations of 10 emmetropic subjects when fixating a near target was investigated-'". Following a double-blind protocol against saline, timolol was shown to reduce the r.m.s. value ofthe fluctuations (Figure 4). A reduction in the peak location of the HFC was observed in all subjects following treatment with timoiol and was correlated with the 4 6 beats min ~ ^ reduction in arterial pulse known to result from systemic absorption of timolol via ocular routes. To separate local from systemic effects, an experiment
254
Ophthai. Physiol. Opt., 1992, Vol. 12, April
was performed using four subjects whose treated and untreated eyes were examined to determine the effects of the consensual reduction in IOP and the position ofthe HFC. Results in the treated eyes were the same as in the previous experiment but for the untreated eyes the results were equivocal although the correlation between arterial pulse frequency and location ofthe HFC was maintained. Overall, the results indicate that factors relating to IOP and ocular vasculature will aflect the magnitude and frequency composition of accom;nodative fluctuations and suggest that measurement of microfluctuations may have potential as a non-invasive method of assessing the eflect of beta-adrcnergic antagonists on the eye.
Conclusions The studies to date suggest that the complex waveform ofthe accommodative microfluctuations is a consequence of the combination of neurological control and physiological rhythmic variation: the former is probably
Microfiuctuatiom of accommodation: B. Winn and B. Gilmariin attributable to the LFCs: the latter to arterial pulse. A functional role for the fluctuations as an error-deteetor in the accommodation system is, therefore, probably related to the maintenance of focus on a stationary target as the neurologically controlled component is too slow to provide the necessary information to optimize the response when rapid step changes in stimulus vergence occur. Although the HFCs may still be utilized by the contrast detection mechanism as part of the aggregate response, it would appear that only the LFCs could provide the necessary neurological control. If the accommodation controller operates to maintain a eonstant r.m.s. outptit to monitor retinal image contrast we would envisage that the magnitude of the neuroiogicaily-controlled LFC
eould be dependent upon the systemieally-modulated HFC. It is conceivable, therefore, that a substantial increase in the HFC could swamp the potential accommodative control offered by the LFC. This suggests that investigation of visual performance in relation to systemic variations of arterial pulse^' may be of importanee in identifying eausal agents of visual stress during sustained visual tasks. Acknowledgements This work was supported in part by the Visual Research Trust which we gratefully acknowledge. We also thank
2.4 2.2r2.2 2.02.0
1 .6
1 .4 -
1 .2
1 .2 }.2
1 .0 1 .0
1 .2
1.6
1 .4
2.0
Kigure 2 Corrclalioii between high-frequency component for P < 0.001 ). The regression line is with permission from Winn f( at.
arterial pulse frequency and the group data {n = 20: r = 0.99, y = 0.0604 + 0.9516 x. Reproduciid 1990'^
0.09
1.6
1.3
2.0
2.2
2.4
A c c o m m o d a t i o n high frequency component ( H z )
2.2
Accommodation high frequency component (Hz)
1.4
Figure 3 Correlation between arterial pulse frequency and the high-frequency component for the recovery phase (sequence 1-5) of exercise-induced changes in pulse. The subject was a 21-year-old woman whose base-line pulse was 1.5 Hz under the cxpcrimcntai conditions and was typical of the sample. Reproduced with permission from Winn er al. 1990"
0.09 Subject
^0.07
0.07
0.05
0.05
0.03
0.03
0.01
0.01
B.E,
- I '
0.30
0.90
1.50
2.10
Frequency (Hz)
2.70
0.30
0.90
1 .50
Frequency
2.10
2.70
(Hz)
Figure 4 Power spectra of the accommodation microfluctuations, each with 16 degrees of freedom, for two subjects whose eyes have been treated with timolol maleate 0.5%. Timolol induces iin overall reduction in power and a frequency shift in the high-frequency component which is consistent with a reduction in pulse rate of 6 beats min '; . timolol; , post-limolol. Reproduced with permission from Owens et al.. 1991'"
Ophthal. Physiol. Opt., 1992, Vol. 12, April
255
Microfluctuations of accommodation: B. Winn and B. Gilmartin Merck. Sharp and Dohme for the supply of timolol maleate and Drs H. Owens and J. R. Pugh for Iheir collaboration in the studies.
17. 18.
References 1. Collins. G. The electronic refractometer. Br. J. PhysinL Op!. I. 30 40 (1937). 2. Arnulf. A., Dupuy, O. and Klamnnl. F. LL-S micro liuclu at ions d'accommodation de Toeil el Facuite visuelle pour les diametres pupillaircsnalurelles.C .«.--1(W. S(/./'(//•/.S232.339-341 I 1951 ), 3. Arnulf. A., Dupuy. O. and Flamanl. F. Les microfluclualions de l'oeil el leur inlluence sur I'image retiniene. C. R. Hehd. Seancf.s Acad. SCI. Pans 232. 43S-450 (195! ). 4. Arnulf. A.. Santamaria, J. and Bescos. J. A cincmalogriiphic melhod for the dynamic sludy of the image formation by the human eye. Microiluctuations of accommodation. J. Optics (Paris) 12, 123-12K (1981). 5. Whiteside. T. C. D. Problems of Vi.sicn at Hiijh .4lii/iitli:s. Butterworths. London, p. 193 (1957). 6. Campbell, F. W. Correlation of accommodation between the two eyes. ./. Opi. Soc. Am. 50. 738 (1960). 7. Campbell, F. W. The accommodation response of the human eye. Br. J. Physiol. Opi. 16, 188 203 (i960). 8. Campbell. F. W. and Robson. J. G. High-speed infra-red optometer. J. Opi. .Sac. Am. 49. 268-272 (1959). 9. Ciimpbell, F. W. and Westheimer. G. Sensitivity of the eye to difTerenccs in focus. J. Phy.siol. 143. !8P |I958). 10. Campbell, F. W. and Westheimer. G. Factors inHueneing the accommodation responses of the human eye. J. Opi. Soc. .iDi. 49. 568-571 (1959). 11. Campbell. F. W. and Weslheimer. G, Dynamics of the accommodation responses of the human eye. J. Physio!. 151. 285-295 (1960). !2. Campbell. F. W.. Westheimer. G. and Robson, J. G. Sijiiiificance offlucttiations of accommodation. J. Opi. Soc. Am.4S.bb9 (1959). 13. Campbell, F. W.. Robson. J. G. and Westheimer, G. Fluctuations of accommodation under steady viewing conditions. J. Physial. 145. 579-594 (1959). 14. Denieul. P. Effects of stimulus vergence on near accommodation response, microlluetuations and accommodalion and optical quality of the human eye. \'isi(m Res. 23. 561-569 (1982). 15. Kotulak. J. C. and Schor. C. M. Temporal variations in accommodation during steady-slate conditions. J. Opi. Soc. Am. .4 3.223-227 (1986). 16. Alpern, M. Variability of accommodation during steady fixation
256
Ophthal. Physiol. Opt.. 1992, Vol. 12, April
19. 20 21,
22,
23, 24, 25 26. 27. 28. 29 30.
31.
at various levels of illuminance. J. Opi. Soc. Am. 48, 193-197 (1948). Crane, H. D. A theoretical analysis of the visual aecommodation system in humans. Report NASA CR-606. NASA. Washington (1966). Kotuiak, J. C. and Schor, C. M. A computational model of the error deteetor of human visual accommodation. Biol. Cyhvrii. 54, 189-194 (1986). Charman, W. N, The retinal image in the human eye. In Proijress in Retimil Re.seurch (eds N. Osborne. and G. Chader), Vol. 2, Pergamon. Oxford, pp. 1-50 (1983). Pugh. J. R. and Winn, B. Modification of the Ciinon Autoref RI for use as a eontinuously recording optometer. Ophlhal. Physiol. 0/J/. 9, 451-454 (1988). Winn, B,. Pugh, J- R-. Giimartin, B. and Owens, H. The effect of pupil size on static and dynamic measurements of accommodation using an infra-red optometer. Ophihat. Phv.m>l. Opt. 9. 277-283 (1989). Winn.B., Pugh, J. R., Gilmartin, B. and Ov^ens, H. The frequency characteristics of accommodative microfluctuations for central and peripheral zones of the human crystalline lens. Vi.sion Res. 30, 1093 1099 (1990). Rosenfield, M., Giimartin, B., Cunningham, E. and Oattani, N. The infiuence of alpha-;tdrenergic agents on tonic accommodation. Curr. Eye Res. 9. 267-272 (1990). Koretz, J. F. and Handeiman. G. H. Model of the accommodative meehanism in the human eye. I'ision Res. 22, 917-942 (1982). Koretz, J. F. and Handeiman. G. H. A model for aecommodation in the young human eye: the effects of lens anisotropy on the mechanism. Vi.sion Res. 23, 1679 1686 (1983). Denieul, P. and Corno. F. Accommodation et contraste. L'oplomelrie 32. 4-8 (1986), Winn, B., Pugh, J. R., Gilmartin, B. and Owens. H. Arterial pulse modulates steady-state ocular accommodation. Curr. Eve Res 9 971-975(1990). Fercher, A. F., Hu, H. Z., Steeger, P. F. and Briers, J. D. Eye deformation measurement by la.ser interferometry. Optica Acia. 29, 1401-1406 (1982). Colloton, E. and Perkins, E. S. The effect of timolol on the intraocular pulse pressure. Invest. Ophihalmol. Vi.sual Sci. 27(Suppl.), 102 (1986). Owens, H., Winn, B., Gilmartin, B. and Pugh, J. R. Effect of a topieal beta-adrenergic receptor antagonist on the dynamics of steady-state accommodation. Ophlhal. Phvsiol. Opt. 11, 99-104 Gros. B. L. and Cohn.T. E. The phase of the ocular arterial pulse affects deteclability of a brief threshold stimulus, /nrest. Ophthalmol. Visual Sci. 32(Suppl.), 699 (1991 ).
