INTRODUCTION

Top Abstract Introduction Materials & Methods Results Discussion References

ACCOMMODATION, the mechanism by which the normal eye changes its focus from distant to near objects, is lost with increasing age in humans and monkeys. This phenomenon, called presbyopia, is the most common human ocular affliction, and its pathophysiology is poorly understood. The classical theory of accommodation proposes that the ciliary muscle moves forward and axially in the eye during contraction, thereby releasing tension on the anterior zonular fibers and allowing the lens to round up and thicken axially (59). This increases the eye's refracting power and permits focus on near objects. During disaccommodation the ciliary muscle relaxes, allowing the elastic choroid and posterior zonular fibers to pull the muscle posteriorly, increasing the tension on the anterior zonules to flatten the lens. Theories to explain the pathophysiology of presbyopia fall into two main categories, involving dysfunction of either the lens or the ciliary muscle.

Changes in Lens Properties With Age

Ciliary Muscle Dysfunction

Technical Aspects

Subprimate animals accommodate either very little or by mechanisms anatomically and physiologically different from the human (11, 22, 31, 34, 39, 48). However, the rhesus monkey has an accommodative apparatus structurally and functionally identical to the human (2, 23, 33), a pharmacologically stimulated accommodative amplitude of 30-40 diopters in youth, and an age-related loss of accommodation with a similar relative age course as in humans (2, 23).

In rhesus monkeys, the iris may be surgically removed via a small limbal incision (24). This allows direct transcorneal visualization, imaging, and recording of the ciliary muscle-zonular-lenticular system in real-time by Scheimpflug- and goniovideography (40). Accommodation can be stimulated via the normal efferent neuronal pathway by means of an electrode permanently implanted in the midbrain Edinger-Westphal (E-W) nucleus (8). These previously described survival techniques permit repeated real-time study of accommodation dynamics and presbyopia in an animal model meaningfully comparable to the human. We describe here the quantitative image analysis methodology and initial findings obtained using this model.

	MATERIALS AND METHODS

Top Abstract Introduction Materials & Methods Results Discussion References

Data Collection

The anesthetized animal was placed prone in a head holder with the eyes in the primary position. Body temperature was maintained at 36-38°C by heating pads. Botulinum A toxin was injected into the medial and either the superior or inferior rectus muscle 2-7 days before the recording session to minimize eye movements by paralyzing the extraocular muscles. At the start of each experimental session, a 5-0 Dacron suture was passed beneath the lateral rectus muscle to allow tension to be applied to reduce any residual eye movements not eliminated by the toxin injections. Accommodation was stimulated centrally via the implanted electrode. Imaging and video recording of the anterior segment (Scheimpflug- and goniovideography, the latter employing a Swan-Jacobs gonioscopy lens) were accomplished using a modified Zeiss stereo photo slit-lamp microscope, equipped with a low-light, black and white video camera or a color video camera, a 3/4-in. VHS video recorder/player, and a time-date generator. Details of the equipment and procedures, as well as anesthesia, analgesia, and treatments for iridectomy, electrode implantation, and central stimulation have been described in detail previously (8, 24, 40) and in the interests of brevity are not repeated here.

The stimulus current amplitude (at constant frequency of 100 Hz)-accommodative response relationship was established for each animal. Scheimpflug video recordings of lenticular changes were then made during stimulation at four different current settings that yielded minimally suprathreshold, approximately one-third maximal, approximately two-thirds maximal, and maximal accommodative amplitude. Stimulus duration was 2.2 s, with videographic recording for at least 4 s to record the time course of onset and relaxation of accommodation. After the entire set of responses was recorded two or three times, the Scheimpflug camera was replaced with the gonioscopy lens to image movements of the ciliary body, zonule, and lens periphery, using the same stimulus parameters. Time to the nearest 0.01 s and onset and termination of electrical stimulation were electronically encoded on the videotape.

A millimeter rule was placed in the plane of focus for image calibration. The absolute magnification of the Scheimpflug and gonioscopic images within the eye could not be established, because in the former the cornea itself, and in the latter the gonio lens, alters the magnification of the image. However, uncertainty of magnification could have no effect on delay time or on observed time course of the various movements, the determination of which was the primary objective of the study. The values presented are therefore given as approximate millimeter change from baseline. Because in most cases interlace video mode was used, our time resolution was s. Using noninterlaced video mode to obtain -s resolution did not yield observable effects on the time course curves; hence this mode was abandoned because the loss of spatial resolution offset any gain from improved temporal resolution.

The lens and ciliary body movements were measured on frame-by-frame playback of video recordings from a Panasonic model AG-6300 video recorder and model WV5350 video monitor using an electronic caliper (Fowler; Max-Cal, Newton, MA) connected to an Apple IIe computer. The caliper was fitted with transparent plastic plates with hairlines to allow these markers to be placed as close to the images as allowed by the thickness of the glass face of the cathode ray tube (CRT). The measured values, together with the frame numbers and the times of onset and termination of the stimulus, were recorded in the computer.

In a later phase of the study, individual frames were entered into a Sun Microsystems model 150 computer via a Matrox (Montreal, PQ) frame grabber and graphic support boards. After image enhancement, similar measurements as described above were made using a mouse-controlled cursor. This technique eliminated possible parallax errors introduced by the thickness of the glass or the CRT when the electronic caliper was used to make this measurement. Because no systematic differences in measurement were observed between the use of the cursor versus the caliper, the data obtained with the caliper were accepted as accurate without corrections.

Data were collected only from image sequences where the measurement reference points could be viewed throughout the entire accommodation/disaccommodation sequence. For ciliary body movement, this measurement was the distance between a single, well-identified point on the ciliary body and a fixed point at the edge of the sclera. Lens thickness was measured as the distance between the slit-lamp beams reflected from the apexes of the lens anterior and posterior surfaces. Anterior chamber depth was measured as the distance between the slit-lamp beams reflected off the apex of the cornea and the apex of the anterior lens surface. Lens movements were measured relative to the static corneal slit-lamp reflection. Because lens thickness and ciliary body movements were measured using different imaging techniques, they were necessarily measured during different stimulations, although during the same session and at the same stimulus parameters. A running average of the last three data points was calculated and a best fit curve vs. time was plotted by the computer.

Statistical Analysis

	RESULTS

Top Abstract Introduction Materials & Methods Results Discussion References

Accommodation

View larger version (139K): [in this window] [in a new window]   Fig. 1.   Goniovideography of 4 (A-D)- and 24 (E and F)-yr-old rhesus monkey anterior segments following total iridectomy and midbrain electrode implantation. A and E: nonaccommodated ciliary process (CP), zonular fibers (Z), and lens (L) just before stimulation of Edinger-Westphal nucleus. B-D and F: after stimulation, accommodation progresses from B to maximum in D. In C, zonules fold and ciliary processes touch the lens. In D, ciliary ring continues to narrow even after tips of processes touch the lens. Line 1: distance between ciliary valley and lens. Lines 2-4, progressive narrowing of lens-ciliary body space during accommodation in 4-yr-old monkey. Lines 1 and 2 in 24-yr-old animal (E and F) indicate no change in lens-ciliary body space following stimulation.

View larger version (19K): [in this window] [in a new window]   Fig. 2.   Time course of lens and ciliary body movements during 14 diopters of centrally stimulated accommodation and disaccommodation in a 4-yr-old rhesus monkey. Stimulus parameters were supramaximal for accommodation in young animals (1,000 µA at 100 Hz). Reference points on ciliary muscle that did not disappear during accommodation or disaccommodation were selected to determine amplitude and velocity of movement from sclera.

Disaccommodation

Aging

Accommodation. Figure 1, E and F, shows the almost total lack of movement of the ciliary body during stimulation of the E-W nucleus in an old animal. Figure 3 shows the trends of ciliary body and lens movement during the entire stimulus train and the overall changes with age. Figures 4A and 5 show that the decline of both the lens and ciliary body response (amplitude and velocity) were significantly correlated with age, so that neither structure can be entirely excluded from involvement in the pathophysiology of presbyopia.

View larger version (25K): [in this window] [in a new window]   Fig. 3.   Age-dependent loss of lens thickening and ciliary body movement during centrally stimulated accommodation and disaccommodation in rhesus monkeys. Stimulus parameters were supramaximal for accommodation in young animals (1,000 µA at 100 Hz). Max Accomm, maximum accommodation; D, diopters.

View larger version (38K): [in this window] [in a new window]   Fig. 4.   A: age-dependent loss of ciliary body and lens response amplitude during centrally stimulated accommodation and disaccommodation vs. age in rhesus monkeys. Stimulus parameters were supramaximal for accommodation in young animals (1,000 µA at 100 Hz). , no lens response at age 24 yr. Solid curved line in A represents exponential decay model (see Statistical Analysis) to determine whether nonlinear decay evident in data was significant; the probability associated with the a1 variable (reported as coefficient ± SE) tests for presence of such a decline. Straight solid or dashed lines in A, B, and C represent least-squares regression of lens response amplitude (A), lens latent period (B), or ciliary body latent period (C) on age during accommodation (B and C, solid lines) and disaccommodation (B and C, dashed lines). Slopes and intercepts are coefficients ± SE; P, probability that slope or intercept = 0; r, correlation coefficient.

View larger version (33K): [in this window] [in a new window]   Fig. 5.   Age dependence of ciliary body (A and B) and lens (C) velocity during centrally stimulated (1,000 µA at 100 Hz) accommodation and disaccommodation. A: plot of first-phase velocity for young animals and monophasic velocity for older animals vs. age. B: plot of second-phase velocity for young animals and monophasic velocity for older animals vs. age. Straight solid or dashed lines in A, B, and C represent least-squares regression of ciliary muscle velocity (A and B) or lens thickening velocity (C) on age during accommodation (solid lines) and disaccommodation (dashed lines). Slopes and intercepts are coefficients ± SE.

View larger version (29K): [in this window] [in a new window]   Fig. 6.   Time-dependent loss of lens and ciliary body movements during repeated centrally stimulated maximum accommodation and disaccommodation in a 4-yr-old rhesus monkey. Stimulus parameters were 1,000 µA at 100 Hz, administered in a 2.2-s on, 2.0-s off sequence for 4 min. Dashed horizontal lines drawn to delineate subtle lens and ciliary body changes during stimulus train relative to the horizontal.

Disaccommodation. When the stimulus terminated, ciliary body relaxation was clearly slower in the three oldest compared with the three youngest animals (0.48 ± 0.06 vs. 4.75 ± 0.13 mm/s, respectively; P < 0.002) and was again essentially monophasic, but with slowing toward baseline state (Figs. 3, 5A; Table 3). Ciliary body disaccommodation velocity declined rapidly with age from ~10 yr to 15 yr (Fig. 5A). Composite ciliary body velocity in younger animals was greater during disaccommodation than during accommodation, both velocities declined with age, and the rate of decline was more rapid for disaccommodation, so that by the age of ~ 20 yr the velocities were about equal (Fig. 5, A and B; Table 3).

Fatiguing. When the 4-yr-old animal was maximally stimulated with repeated trains of 2.2-s duration separated by 2-s rest intervals, the magnitude of lens thickening declined progressively (Fig. 6, 7A; Table 4). After 4 min, maximum lens thickening was reduced by two-thirds and was reached within 0.5 s after stimulus onset. From 0.5 s after stimulus onset the lens tended to thin slightly whereas lens thickness increased continually throughout most of the first stimulus period. The latent period for lens thickening remained essentially constant with repeated stimulation while the latent period for thinning decreased progressively over time (Fig. 6, 7B; Table 4). With repetitive stimulation, velocity of lens thickening during accommodation did not clearly decline, but velocity of lens thinning during disaccommodation declined (Fig. 6, 8C; Table 4).

View larger version (41K): [in this window] [in a new window]   Fig. 7.   Amplitude (A) and latent period (B and C) of lens and ciliary body movement vs. time in a 4-yr-old rhesus monkey during repeated central stimulation. Conditions as in Fig. 6. Straight solid or dashed lines in B and C represent least-squares regression of lens latent period (B) or ciliary body latent period (C) on time during accommodation (dark, solid lines) and disaccommodation (dashed lines). Slopes and intercepts are coefficients ± SE.

View larger version (36K): [in this window] [in a new window]   Fig. 8.   Velocity of ciliary body (A: first phase; B: average of first and second phases) and lens movement (C) vs. time in a 4-yr-old rhesus monkey. Conditions as in Fig. 6. Solid lines in A, B, and C represent least-squares regression of ciliary body velocity (B) or lens thickening velocity (C) on time during accommodation. Slopes and intercepts are coefficients ± SE.

	DISCUSSION

Top Abstract Introduction Materials & Methods Results Discussion References

Technical Considerations

Total iridectomy is necessary for goniovideographic imaging of the lens equator and ciliary body. Complete removal of the iris reduces the maximal pharmacologically stimulated accommodative amplitude in rhesus monkeys, perhaps because the intense contraction of the iris enhances the normal accommodative response by either pinching the anterior lens or dragging the ciliary body still farther axially (9). Implanting a stimulating electrode in the living monkey brain and maintaining it for extended periods theoretically might injure the accommodative center, but no other methodology exists for repeatedly inducing rapid physiological accommodation and disaccommodation in subhuman primates. Minor week-to-week variability in the accommodative response occurs (8), both increases and decreases, perhaps due to variations in anesthetic depth, hydration, or the exact position of, or fibrosis surrounding, the electrode tip within the E-W nucleus. However, given all the theoretical sources of variability and systematic long-term effects on the ocular and central components of the mechanism, it is remarkable how consistent the responses of a given animal are over time (8). It is also remarkable how closely the cross-sectional age-dependent decline in electrically stimulated accommodation follows that for 1) pharmacologically induced accommodation in noniridectomized, non-electrode-implanted rhesus monkeys (2) and for 2) voluntary accommodation in humans, relative to lifespan (10, 29). Furthermore, the methodology was identical for all animals, and there is no evidence that young and old animals were affected differently by the procedures. Therefore, we conclude that our data reasonably reflect the true accommodative mechanism and its dynamics and aging. Ideally, one would like to observe the course of presbyopia in individual animals longitudinally. However, presbyopia is a continuous process and would require many years before discernible differences in accommodative amplitude became apparent in an individual animal. From the standpoints of both completing the study within a reasonable time period and maintaining a monkey with a functioning indwelling electrode in the midbrain, it was far more realistic to undertake a cross-sectional study.

Lens and Ciliary Body Dynamics and Their Relationship

The amplitude of the ciliary body movement declined exponentially while the lens thickening declined linearly (Fig. 4A). This indicates greater rate of age-dependent restriction of movement for the ciliary body than for the lens, especially in youth (Fig. 4A). The increasing inability of the lens to thicken might be caused by the increasing ciliary muscle restriction with age or by a hardening of the lens occurring separately but with a similar time course.

The latent period increased with age at virtually the same rate for the ciliary body and lens responses (Fig. 4, B and C). This may indicate increasing ciliary body restriction, perhaps due to an increasingly inelastic posterior attachment (33, 56). The increase in lens latent period could be due to the ciliary body restriction or, separately, to lens hardening. Figure 5 shows an age-dependent decline of ciliary body and lens velocities for both accommodation and disaccommodation, again consistent with restriction of the ciliary body; the reduced speed of lens deformation could again be due either to the ciliary body restriction or to lens hardening itself. The ciliary body velocity declines more rapidly with age than does lens velocity, suggesting a primary ciliary body-related component in presbyopia.

Figure 7A shows a nearly identical exponential decline in the amplitude of ciliary body and lens movement during repetitive stimulation in a young animal, reflecting the dependence of lens deformation on ciliary body movement and fatigue of the ciliary muscle with time. The ciliary body showed no time-dependent change in the latent period for disaccommodation, but the lens exhibited a decrease in latent period. Possibly the fatiguing ciliary body had progressively less antero-inward movement at the end of minutes 2, 3, and 4 during the stimulus train (Fig. 6), so that when the stimulus was turned off, return of zonular tension sufficient to deform the lens occurred sooner. Normally humans can accommodate to focus on near objects at a comfortable working distance for long periods of time, as during reading. The decline in amplitude of ciliary ring constriction and lens thickening within <4 min in our animal was most likely due to supramaximal stimulation, analogous not to normal near work but rather to maintaining sharp focus at or even closer than the near point of accommodation.

One might expect the velocities for both ciliary body and lens to change in parallel. However, they differed in the young animals; the ciliary body response was consistently biphasic (Figs. 2, 3, 6), whereas the lens responded monophasically, without a clear linear second phase in which to measure a constant velocity. Thus we did not analyze the lens response as having a first and second phase, as we did for the ciliary body in Table 1. The biphasic nature of the ciliary body response is likely real rather than artifact. In one animal a small transient baseline drop was occasionally observed near the stimulus onset. However, this occurred in only two traces (Fig. 2, immediately before stimulation; Fig. 7, 1-min stimulus train) and not in the 10 other traces examined in these young animals (Fig. 4; Fig. 7: 0-, 2-, 3-, and 4-min stimulus trains). The precise cause of this drop is uncertain, but in any event an artifactual change of this magnitude later during stimulation would be too small to account for the biphasic nature of the ciliary body response, which was biphasic both in the presence and absence of this apparent artifact.

The initial phase of ciliary muscle contraction appears to be lost with increasing age. We deduce this from three findings: 1) in the older animals, the ciliary body was slower to respond (longer latent period); 2) in the younger animals, the second-phase velocity was independent of age; and 3) the velocity of the monophasic movement in the older animals was similar to that of the second phase in young animals (Fig. 5B). Specifically, in the 18- and 24-yr-old animals the latent period (0.23 and 0.22 s, respectively) is as long or longer than the combined duration of the latent period and the first contraction phase in the 4, 9, and 10-yr-old animals (0.25, 0.14, and 0.16 s, respectively). However, it is also possible that the initial phase still exists in the older animals, but that the difference between the first and second phase is so small that we are not able to distinguish them.

Additionally, in the older animals, the combined duration of the latent period plus the remaining contraction period (0.40 ± 0.04 s; n = 3) is shorter than the entire biphasic contraction response in the younger eyes (1.06 ± 0.11 s; n = 3; P = 0.028), perhaps implying that at least some of the second contraction phase is also lost with age. The 15-yr-old animal appears to be intermediate in this regard (0.08-s latent period + 0.36-s remaining contraction period), perhaps because this animal has not completely lost its initial rapid phase. However, it also appears that ciliary movement latency changes may occur at a later age than the velocity and amplitude changes and therefore might be secondary to them. These various possibilities must remain tentative, because the findings are based on results from relatively few animals.

Our data also show that 1) maximum ciliary ring constriction in the young exceeds that needed for maximum lens rounding and thickening; 2) with fatigue during repetitive stimulation, the amplitude of ciliary body movement declines more rapidly than does lens thickening; and 3) despite the differences in lens and ciliary body dynamics, the maximum accommodative amplitude was virtually identical in all three young animals. The fatiguing experiment suggests that the rapid initial component is insensitive to fatigue, whereas the more slowly developing component is more sensitive.

On termination of the stimulus in the young animals, ciliary body relaxation begins almost immediately and continues in a rapid monophasic fashion but slows as the baseline state is approached (Fig. 3). The slowing (especially noticeable in the 10-yr-old animal) is most likely a consequence of the mechanical/elastic properties of the muscle-zonule-lens, rather than of gradual cessation of neuronal conduction and transmitter release and binding. Measuring this nonconstant deceleration might provide some insight into the elastic component of the accommodation apparatus and its change with age or fatiguing, but our limited data did not reveal any trends with age or duration of stimulus train here.

Ciliary body movement during accommodation in the young animals was biphasic, with an initial rapid and a subsequent slower velocity. It is likely that the latter is due to the elastic components of the system (i.e., choroid, lens, etc.). The decline with age in disaccommodation velocity and the increase with age in latent period of the ciliary body response are in all likelihood related to changes in properties of the ciliary muscle or lens, rather than to alterations in the neural signal. To verify this one could administer a cholinesterase inhibitor (i.e., eserine); if the latent period did not decrease in older animals it would point to a nonneural factor. The striking feature of the relationship of ciliary body disaccommodative velocity to age is its precipitous decline, rather than its linearity or sigmoidality.

Implications for Ciliary Body and Lens Involvement in Presbyopia

The basis for an extralenticular contribution to presbyopia is provided by the age-related loss of the movement of the rhesus ciliary muscle in response to pilocarpine stimulation, as measured in situ by histological criteria (33, 56). This loss of movement is believed to be due to a reduction in elasticity of the posterior tendons of the muscle and the elastic lamina of Bruch's membrane, the tissues that connect the ciliary muscle posteriorly to the sclera at a point near the optic nerve (56, 57). The diminished ciliary body response to pilocarpine is not due to any deficiency in the neuromuscular apparatus (47, 58). Furthermore, loss of muscle strength, especially as early in life as the loss of accommodation with presbyopia, is not characteristic of aging. The constancy of the pupillary light response throughout the rhesus lifespan (37) suggests that the parasympathetically dominated intraocular smooth muscles are no exception. Collectively, these data indicate that nonmuscular extralenticular factors are involved.

The ciliary muscle might be restricted due to the posterior tendons becoming fixed and rigid (56) or fixed and flaccid (3). If the posterior tendons became fixed and rigid, the normal forward movement of the contracting ciliary muscle would be restricted by the inextensible posterior tendons, and no accommodation would occur (56). If the posterior and intramuscular elastic tissue became flaccid, as proposed by Bito and Miranda (3), rather than rigid, the force of contraction of the ciliary muscle becomes irrelevant, because the system is "stuck" in the accommodated position assuming the lens/capsule remains elastic and continues to pull the ciliary muscle into an accommodated state. This scenario would also predict that the lens in the elderly eye would always have an accommodated (more spherical) configuration, with the refractive consequences (6, 27, 28) compensated for by a decreased lens refractive index, maintaining focus at distance rather than near (7, 44). However, Glasser and Campbell (19) argue against this hypothesis on the basis of finding that no amount of stretching or releasing zonular tension of older, presbyopic human lenses results in any optical change, whereas the same stretch applied to young lenses produced substantial optical changes. Furthermore, some of the neuromuscular and connective tissue degenerative changes and altered muscular and lenticular geometry sometimes cited in support of the "flaccid posterior attachment" hypothesis may represent postpresbyopic secondary or aging changes. Indeed, the specimens studied histologically (58) did not include individuals under the age of 35 yr, and very few under the age of 40 yr, by which time two-thirds of the accommodative amplitude has already been lost (10).

Another consideration is the change in the geometry of the lens suspension with age. In the human, the distance between the anterior zonular insertion onto the lens and the lens equator increases with age, whereas the distance between the insertion ring and the ciliary body remains relatively constant (12). It has been suggested that the zonular fibers become less able to relax with accommodation or that they become less able to exert an influence on the lens (27), but there are no other data to support this geometric theory (19), and complete relaxation of the zonular tension in vitro does not cause presbyopic human lenses to become accommodated (19). It remains unknown whether this phenomenon is present in the rhesus monkey.

In addition to a slow loss of accommodation before the age of 10, our data also suggest the possibility that something dramatic may be happening between the ages of 10 and 15 yr, when there is a loss of 8 diopters of accommodative ability. Furthermore, the decline of ciliary body response amplitude (Fig. 4A) and velocity (Fig. 5A) with age happens at a much faster rate from ages 10 to 15 yr than either before or after this age range. It could be that incremental changes seen before age 10 yr lead to a critical point where a new state is reached between ages 10 and 15 yr. However, a much more likely possibility is that of the few animals used, the 10 yr old had an unusually high accommodative amplitude and the 15 yr old an unusually low accommodative amplitude. The gradual decline in accommodation with the progression of presbyopia is such a characteristic feature of presbyopia [even in rhesus monkeys (2)] that it would be truly remarkable if something suddenly occurs between ages 10 and 15 yr in these animals.

Although our knowledge of the events that accompany the aging of the accommodative mechanism has increased vastly over the past decade, we still do not completely understand the age-related changes in any of the components, nor how they interact to produce presbyopia. Future studies utilizing the animal model and techniques described herein may enhance our insight into this most common of all ocular afflictions.

Perspectives