Program # 2800

Saccadic Lens Instability Increases with Accommodative Stimulus in PresbyopesLin He, Scott B. Stevenson, Adrian Glasser College of Optometry, University of Houston, Houston, TX, USA;

William J. Donnelly III Breault Research Organization, Inc., Tucson, AZ, USA.

INTRODUCTIONThe etiology of presbyopia remains unclear. Presbyopia could occur consequent to loss of ciliary muscle function or loss of lens function. Recent literature suggests loss of lens compli-ance as the primary pathology. However, the extent to which ciliary muscle contraction during accommodation is preserved in the presbyopic eye has been debated1,2. Lens instability during and after saccadic eye movement may reflect ciliary muscle function and zonular tension. The dual Purkinje image (dPi) eye tracker can be used to evaluate this instability through the “lens wobble artifact”3,4. The amplitude of the lens wobble may indirectly reflect the extent of ciliary muscle contraction. In this study, saccadic lens wobble was quantified in presbyopes with accommodative stimuli of 9 different amplitudes.

METHODSTen presbyopic subjects participated. Subjects executed 32 4-degree saccades at one second intervals between targets arranged in a cross on illuminated cards at each of 9 viewing distances ranging from 0.5 to 8 D accommodative demands. Viewing was binocular; targets were aligned with the dilated (2.5% phenylephrine) left eye. Testing was also performed in the dilated eye of a 49 year old subject without and on another occasion with two drops of tropicamide administered to paralyze accommodation. Lens wobble could be directly recorded in one of the presbyopic subjects by video-based tracking of a cuneiform cataract using an infrared sensitive video camera with a 60 Hz frame rate. The dPi left eye Purkinje (P) image channels that sampled P1 position signals (H1/V1) and signals from the relative difference in position between P1 and P4 signal (θH/θV) were recorded at 360 Hz. To exclude the possibility that the post-saccadic lens wobble artifacts originate from saccadic eye-movement “dynamic overshoots”5,6, the artifacts were extracted by subtraction of H1/V1 from θH/θV signals (Fig. 1).

RESULTS

REFERENCES

1. Strenk, S.A., Strenk, L.M., & Semmlow, J.L. (2000). High resolution MRI study of circumlental space in the aging eye. Journal of Refractive Surgery, 16, S659-60. 2. Stachs, O., Martin, H., Kirchhoff, A., Stave, J., Terwee, T., & Guthoff, R. (2002). Monitoring accommodative ciliary muscle function using three-dimensional ultrasound. Graefe's Archive for Clinical and Experimental Ophthalmology, 240, 906-12.3. Crane, H.D., & Steele, C.M. (1978). Accurate 3-Dimensional Eye tracker. Applied Optics, 17, 691-705. 4. Deubel, H., & Bridgeman, B. (1995). Fourth Purkinje image signals reveal eye-lens deviations and retinal image distortions during saccades. Vision Res, 35, 529-38.5. Bahill, A.T., Clark, M.R., & Stark, L. (1975). Dynamic overshoot in saccadic eye movements is caused by neurological control signed reversals. Experimental neurology, 48, 107-22.6. Kapoula, Z.A., Robinson, D.A. (1986). Motion of the eye immediately after a saccade. Experimental Brain Research, 61, 386-94.

DISCUSSIONThe lens wobble artifact provides the opportunity to understand the consequences of ciliary muscle function in presbyopic eyes. Subtraction of H1/V1 and θH/θV traces provides a useful, quanititative measure of this artifact. This study shows larger lens wobble artifacts occur with increasing accommodative demands in presbyopes, indicating increased lens instability. Lens wobble following saccades likely results from ciliary muscle contraction and reduced zonular tension. Eye modeling suggests that the lens wobble could occur from either lens translation or lens rotation. Therefore, the results suggest, as shown in prior studies1,2, that ciliary muscle contraction continues to occur in the presbyopic eye with accommodative effort.

2 2.05 2.1 2.15 2.2 2.25

−5

−4

−3

−2

−1

0

1

2

Time (seconds)

Am

plitu

de (d

eg)

H1 channelθH channel

θH - H1

Fig. 1: Recordings from a four degree, leftward saccade to a 0.5 D accommodative stimulus. The output signals from the H1 channel (P1: dotted line) and the θH channel (P4 – P1: solid line) are shown. Subtracting H1 from θH, shows the lens wobble artifact (dash-dot line) that is assumed to represent the deviation of the lens within the eye.

0 100 200

0.5

1.0

2.0

3.0

4.0

5.0

6.0

7.0

8.0

LeftwardArtifacts

Time (ms)0 100 200

RightwardArtifacts

Time (ms)

Diopter4º

Fig. 2: The lens wobble artifacts as a function of time for nine accommodative stimuli during horizontal saccades from one subject (HEB). The 4° scale applies to all recordings. Lens wobble amplitude increases with accommodative stimulus demand.

Fig. 3: Artifact amplitude (in degrees) plotted as a function of the saccade amplitude for different accommodative stimuli (in diopters) from all the subjects. All the liner regression lines shown are statistically significant at the p<0.001 level as shown in Table 1.

Accommodative Stimulus (D)

0 2 4 6 8

Rela

tive

Art

ifact

/Sac

cad

e Ra

tio

-0.2

0.0

0.2

0.4

0.6

0.8

Fig. 5: To better present the ratios of leftward artifact/saccade amplitude from all the subjects on the same scale, each subject’s average ratio for the 0.5 D accommodative stimulus was subtracted from the ratios for all stimuli for that subject. There is a significant increase in the ratio as a function of the accommodative stimulus amplitude. Slope = 0.034 and r2 = 0.169 [F(1, 649) = 146.49, p < 0.001].

Accommodative Stimulus (D)0 2 4 6 8

oitaR edaccaS/tcafitr

A

0.0

0.2

0.4

0.6

0.8

1.0

LeftwardRightward

819.01958.005215.0

2 =

+=

rxy

745.02955.005470.0

2 =

+=

rxy

Fig. 6: The artifact/saccade ratio as a function of accommodative stimulus from one subject. Two-way ANOVA shows leftward (abducting) ratios are statistically different from rightward (adducting) ratios [F(1, 131) = 20.21, p < 0.001]. Specifically, the regression slopes are similar [t(128) = 0.783, p=0.435], but the leftward saccades have a larger intercept [t(128) = 7.11, p < 0.001].

Lens Rotation (degrees)

Lens Translation (mm)

Le

ns

Wo

bb

le A

rtif

ac

t (d

eg

ree

s)

Lens Rotation

Lens Translation

Table 1: Slopes (b1) and r2 values for the linear regression lines fit to the lens wobble artifacts as a function of saccade amplitude from the different accommodative stimuli calculated from the results in Fig. 3.

ACKNOWLEDGEMENTS

This project was funded in part by a sVRSG from UHCO to LH. Thanks to Dorothy Win-Hall for the clinical assistance and to Dr. Harold Bedell for help in developing the project.

DISCLOSURE

LH, None; SBS, None; AG, None; WJD, Breault Research Organization, Inc., E.

Author email addresses: lhe5@uh.edusbstevenson@uh.edu aglasser@uh.eduwdonnelly@breault.com

A. Without Cycloplegia

Accommodative Stimulus (D)0 2 4 6 8

o itaR edacc aS /t caf itr

A

00..00

0.2

0.4

0.6

0.8

001.07562.0

2721.00495.02

<=

+=

pr

xy

Fig. 7: The artifact/saccade ratio as a function of accommodative stimulus amplitude from a pre-presbyopic subject with 1.34 D of accommodation without (A) or with (B) 1% tropicamide cyclople-gia. The ratio is not dependent on stimulus amplitude after cyclople-gia, suggesting that accommoda-tive contraction of the ciliary muscle releases zonular tension to allow the lens wobble to occur.

Fig. 8: (A) A ray tracing eye model (Advanced Human Eye Model, AHEM, Breault Research Organiztion, Inc.) was used to model the fourth Purkinje image shifts for a certain range of lens translations and rotations. The relative positions of the Purkinge images (Red: P1; Blue: P4) were identified so their movements can be measured and quantified. (B) Ray tracing analysis of movements of P1 and P4 with either lens rotation or lens translation shows the extent of lens wobble artifact in degrees. For instance, to achieve a lens wobble artifact of ±1°, the lens needs to either translate ±0.125 mm or rotate ±2.5°.

B.A.

Fig. 4: Since lens wobble artifact amplitude is dependent on saccade amplitude, the ratio of artifact amplitudes to saccade amplitude was calculated and plotted as a function of 9 different accommodative stimuli. Data shown are from all leftward saccades from ten subjects. The scale bar shows a ratio of 0.5, which means the artifact amplitude is half the size of saccade amplitude. Linear regression fits show that in 9/10 subjects (all except CN), the ratio increases significantly (p<0.01) with accommodative stimulus amplitude.

The ratio of the subtracted profile amplitude to saccade amplitude was analyzed as a function of increasing accommo-dative stimuli. A ray tracing eye model (Advanced Human Eye Model, AHEM, Breault Research) was also employed to model P1 and P4 shifts for a range of lens translations and rotations.

Saccade Amplitude

Art

ifact

Am

plit

ud

e

0.5

1.0

2.0

3.0

4.0

5.0

6.0

7.0

8.0

1º1º

Diopter

Accommodative Stimulus (D)

0 2 4 6 8

CN

RNB

LP

DTG

CK

DH

SH

LM

HEB

RSH

Subject

0.4

0.5

Arti

fact

/Sac

cade

Rat

io

B. With Cycloplegia

Accommodative Stimulus (D)0 2 4 6 8

oitaR edaccaS/t cafit r

A

0.0

0.2

0.4

0.6

0.8

035.00597.0

2726.00051.02

==

+=

pr

xy

Accommodative Stimulus (D) b1 r2 P-value

0.5 0.2632 0.2160 <0.001 1.0 0.2776 0.2316 <0.001 2.0 0.3534 0.3775 <0.001 3.0 0.4457 0.4262 <0.001 4.0 0.4834 0.4850 <0.001 5.0 0.6204 0.5287 <0.001 6.0 0.5902 0.5886 <0.001 7.0 0.5330 0.4458 <0.001 8.0 0.3962 0.3872 <0.001