|Home | About | Journals | Submit | Contact Us | Français|
To develop a measurement protocol for changes in the shape and size of the ciliary muscle with accommodation using the Zeiss Visante™ Anterior Segment Optical Coherence Tomographer (AS-OCT) and to determine the test-retest repeatability of these measurements.
Subjects were 25 adults ages 23–28 years. The ciliary muscle was imaged at two visits with the Visante™ while accommodative response was monitored during imaging using the PowerRefractor. Ciliary muscle thickness was measured at 1 mm (CMT1), 2 mm (CMT2), and 3 mm (CMT3) posterior to the scleral spur and at the point of maximal thickness (CMTMAX). Thickness was measured at these locations while subjects viewed a target at distance and at a 4.00-D accommodative stimulus. Outcome measures were the change in thickness between distance and the 4.00-D stimulus and the change in thickness per diopter of accommodative response (PowerRefractor). Finally, the repeatability measurements between visit 1 and visit 2 were determined with a Bland-Altman analysis.
The statistically significant modeled changes in ciliary muscle thickness were as follows: CMTMAX = 69.2 μm (4.00-D stimulus) and 18.1 μm (per diopter of accommodation); CMT1 = 45.2 μm (4.00-D stimulus) and 12.3 μm (per diopter of accommodation); and CMT3 = −45.9 μm (4.00-D stimulus) and −12.0 μm (per diopter of accommodation); p < 0.0001 for all.
The combination of the Visante™ and the PowerRefractor is a feasible tool for measuring thickening of ciliary muscle at more anterior locations and thinning at more posterior locations during accommodation. We noted a wide range of accommodative responses during the time of image capture in this study indicating that the most accurate estimates of the change in ciliary muscle dimensions with accommodation may be obtained by using accommodative response rather than stimulus values and by using measurements taken simultaneously with image capture.
Over the past several decades, researchers have produced a body of literature that has given us an in-depth understanding of ciliary muscle anatomy. In vivo function remains an elusive aspect to study because of visualization difficulties. Recent advances in imaging technology have enabled us to continue to build on our current understanding of normal ciliary muscle function. These advances will elevate the level of comprehensive eye care we can provide for the entire visual system.
Patients present with symptoms of accommodative dysfunction in optometric practices daily. While reading additions successfully alleviate these symptoms in the young and old alike, continued research is needed to gain a greater understanding of the in vivo muscle function in young patients with accommodative dysfunction and also after presbyopia onset. A recent review article has shown that multiple companies are working to develop accommodative intraocular lenses (IOLs) that rely on the action of the ciliary muscle; although, there remains a limited understanding of presbyopic muscle function.1
In addition to presbyopia, other types of accommodative dysfunction are also important to consider because our daily activities increasingly place demands on our ability to see clearly at close distances, i.e., computer use, video games, electronic book readers, smart phones, etc. Few studies have explored the prevalence of accommodative dysfunction in the general population, but 60% to 80% of patients diagnosed with binocular vision disorders have been found to suffer from accommodative dysfunction.2–3 The symptoms that are commonly associated with accommodative anomalies include blurred vision, headache, ocular discomfort, ocular or systemic fatigue, and loss of concentration during a task performance.
Studies have shown vision therapy can improve both accommodative amplitude and facility, as well as increase the magnitude and velocity of accommodative responses.4–5 Nonetheless, specific muscular and/or neurological changes that might take place during vision therapy remain unmeasured and the exact effects of therapy remain unknown. Similarly, an increased accommodative lag has been observed after the onset of myopia, but we have been unable to identify the source of this anomaly in the accommodative system.6
Further studies are needed to provide a complete understanding of the childhood development and then the gradual decline of the human accommodative system and how it may relate to myopia, accommodative dysfunction, and the feasibility of IOLs that restore accommodation. Until recently, however, we have not had the imaging capability to measure in vivo changes in the ciliary muscle with accommodation. Recent studies have successfully used magnetic resonance imaging (MRI), ultrasound biomicroscopy (UBM), and anterior segment optical coherence tomography (AS-OCT) to visualize the ciliary muscle and/or to make measurements of the muscular changes that take place with accommodation.7–10 The present study combined the imaging capabilities of the Visante™ AS-OCT (Carl Zeiss Meditec™, Dublin, CA), which is capable of imaging the ciliary muscle and allows for further extraction of thickness measurements,11 with simultaneous monitoring of accommodative status (vertical meridian) by the PowerRefractor II (MultiChannelSystems, Reutlingen, Germany). Combining data from these two devices with knowledge of ciliary muscle anatomy allows one to make measurements of changes in ciliary muscle thickness while accounting for the subject’s accommodative response. In addition, the Visante™ is a non-contact, non-invasive, imaging system that is also easy to operate. It acquires images in a rapid fashion that makes it possible to image the ciliary muscle in adults and children.12–13
Subjects were recruited within the Ohio State University College of Optometry. Adults, under the age of 30 years were eligible. Twenty-five subjects ages 23 to 28 years (mean ± SD = 24.2 ± 1.1 years) participated in the study. At the time of planning this study, we were uncertain of what amount of change in the ciliary muscle thickness should be expected. A sample of 25 was thought to be sufficient for this exploratory investigation. Seventeen of the subjects (68%) were female. Subjects could have any refractive error, i.e., myopia, hyperopia, or astigmatism, as long as it was correctable to 20/20 with contact lenses. Exclusion criteria were any history of ocular diseases or disorders other than refractive error; a history of strabismus; a history of any previous eye surgery; and the use of any systemic or topical medications that are known to affect the ciliary muscle. After a presentation and discussion of the study procedures, all subjects provided written informed consent. The study was approved by the Institutional Review Board of The Ohio State University.
All subjects underwent the same testing protocol, as described below, at two, separate, study visits that were scheduled approximately two weeks apart. All subjects who required refractive error correction wore contact lenses while all measurements were taken. All measurements were made on the right eye while the left eye was occluded.
Images of the temporal ciliary muscle of the right eye were taken without the use of any pharmaceutical agents. Images were obtained with the Zeiss Visante™ Anterior Segment OCT while the subjects viewed targets at 1.00 m (1.00-D stimulus) and at 25 cm (4.00-D stimulus). The target placed at 1.00 m will be referred to as “distance” throughout the manuscript because the PowerRefractor monitoring showed that mean ± SD accommodative response to the 1.00-D target was −0.20 ± 0.54 D, the median was −0.08 D, and the range was −1.12 D to +0.99 D, indicating that most subjects’ accommodative systems were at rest when viewing the 1.00-D stimulus. Six ciliary muscle images were obtained from each subject for both targets in the Enhanced High Resolution Corneal Mode (axial resolution = 18 μm and transverse resolution = 60 μm, according to the manufacturer). The scleral spur was selected in each image by one examiner (MDB) who was masked to both the subject identification number as well as the accommodative state for the image. Thickness measurements were obtained from these ciliary muscle images using a semi-automatic algorithm to obtain a maximum thickness measurement (CMTMAX) as well as thickness measurements at 1 mm (CMT1), 2 mm (CMT2) and 3 mm (CMT3) posterior to the scleral spur (Figure 1).11
To determine the accommodative state of the eye during ciliary muscle imaging with the Visante™, simultaneous measurements of monocular accommodative response were taken using the PowerRefractor II. The PowerRefractor is an autorefractor that is capable of determining the power of the eye when positioned approximately 1.0 m from the subject’s eye. It samples at a rate of 25 Hz, has a range of −8.75 to +4.00 D,14 and requires a pupil size larger than 3.7 mm.15 While subjects were positioned in the Visante™ head and chin rest, they viewed either the distance or the 4.00-D targets. At the same time, the PowerRefractor was positioned to read the power of the eye continuously through a mirror positioned on the side of the Visante while six images of the ciliary muscle were obtained using the Visante™. The PowerRefractor data were filtered to eliminate changes in accommodation greater than 10 D/s because an accommodative change of that magnitude is faster than what is known to be physiologically possible.16 The analysis used the mean PowerRefractor reading for the horizontal meridian over the entire time subjects were viewing the distance and the 4.00-D targets. Thus, at each visit, each subject had one mean PowerRefractor reading for the distance target and one mean PowerRefractor reading for the 4.00-D target. A diagram of the setup for the ciliary muscle image acquisition with simultaneous accommodative monitoring is shown in Figure 2.
All 25 subjects completed the required two visits. For six subjects, PowerRefractor data were not available for one or both visits due to an equipment malfunction, so we were unable to include the data from these subjects’ visits in analyses that accounted for accommodative response. In addition, one visit’s data was removed from all analyses for two subjects because he or she did not exhibit an appropriate accommodative response during testing, i.e., the subject did not have an accommodative response of at least 2.00 D when he or she was presented with the 4.00-D stimulus. This was likely due to a lack of attention or understanding of the task on the part of the two subjects, rather than an inability to accommodate, because the problem only occurred at one of the two study visits.
While the ciliary muscle imaging was completed according to our laboratory’s previous publication,11 some minor adjustments to the protocol were needed for some of the 4.00-D images. In our original publication on measuring the ciliary muscle,11 all of the images were obtained under cycloplegic conditions. In the present study, some images that were taken while subjects responded to the 4.00-D stimulus were darker than cycloplegic or non-accommodative images in the anterior region of the muscle. Figure 3 is an example of one subject who had darker images under accommodative conditions. For some subjects, an adjustment was made to the contrast settings in our algorithm in order to obtain the outline depicted as a solid white line as opposed to the outline depicted as a white dashed line in Figure 3. This algorithm adjustment allowed the outline of the ciliary muscle to include the entire anterior region of the muscle, such that the CMTMAX measurement was located at the apex of the muscle.
The action of the ciliary muscle during accommodation was described by our data in two ways. First, we determined the mean change in the thickness of the ciliary muscle for all subjects as they changed their focus from a distance target to a 4-D stimulus. This calculation did not account for the actual accommodative response of each subject, and it is referred to as the “change for a 4-D stimulus” in ciliary muscle thickness in this report. Second, we divided the change for a 4-D stimulus by the accommodative response recorded by the PowerRefractor (the difference between the distance and the 4-D measurements taken during Visante™ imaging). This ratio (change for a 4-D stimulus/accommodative response) represented the amount of thickening or thinning that occurred in the ciliary muscle per diopter of accommodative response, or the “change per diopter of accommodation.” For each measurement of change in ciliary muscle thickness (CMT1, CMT2, CMT3 and CMTMAX), we fitted the following multilevel model:
In the model μ is the population mean for change in thickness. Estimating μ was the goal of the analysis. The deviation from the mean had two components, a deviation due to biological variability between subjects (δi) and within-subject measurement error (εij).
Analyses suggested by Bland and Altman (1986) were used to assess the inter-visit repeatability of the change for a 4-D stimulus and change per diopter of accommodation for each ciliary muscle thickness measurement.17 Plots of the difference versus the mean of the measurements were visually inspected to determine if the difference between measurements was related to the mean. The mean difference between measurements was calculated and compared to zero using a one sample t-test to determine if any bias was present. The mean of the differences and its standard deviation were used to calculate the degree of repeatability also commonly referred to as the limits of agreement. The limits of agreement (mean ± [1.96 x standard deviation]) characterize the expected difference between repeated measures.
Table 1 shows the mean ciliary muscle thickness values for the distance and 4.00-D measurements at both visits. Note that even though this sample was relatively small, i.e., 25 subjects, the mean and median are similar, suggesting that the data were distributed symmetrically around the mean.
Descriptions of the observed change for a 4-D stimulus and the change per diopter of accommodation in ciliary muscle thickness with accommodation are shown in Table 2. The results for the modeled change for a 4-D stimulus and the change per diopter of accommodation, which provided 95% confidence intervals, are shown in Table 3. The modeled results showed a statistically significant thickening of the ciliary muscle with accommodation at both CMT1 and CMTMAX. No thickening or thinning was found for CMT2 with the change for a 4-D stimulus or the change per diopter of accommodation. Statistically significant thinning of the ciliary muscle was found with accommodation at CMT3 for both the change for a 4-D stimulus and the change per diopter of accommodation.
The results of the Bland-Altman analysis are shown in Table 4. No statistically significant bias was noted between visits for any ciliary muscle thickness measurement, which was not surprising because no learning effects were anticipated to occur for the accommodative task. The coefficients of repeatability for all ciliary muscle thickness measurements were less than 130 μm for the change for a 4-D stimulus and less than 40 μm for the change per diopter of accommodation.
There is a growing body of literature exploring changes in the ciliary muscle structure with accommodation. Recent advances in imaging technology have made it possible to more easily visualize in vivo changes in the ciliary muscle with accommodation.11 The primary objective of this study was to determine if the Visante™ AS-OCT was capable of providing images with enough resolution to observe changes in ciliary muscle shape with accommodation. During the course of our efforts to develop a protocol using the Visante™, we did determine that it is important to monitor accommodation during the imaging as it did differ slightly from measurements obtained prior to imaging on an autorefractor (data not shown).
In a published letter,18 we have attempted to encourage those who measure the ciliary muscle in accommodative research to discuss and consider how the ciliary muscle should be measured. Two previous studies have demonstrated that the Visante™ is capable of imaging the ciliary muscle in vivo,13, 19 and reported changes in the ciliary muscle morphology that are similar to our results even though their data collection and image analysis techniques were different from the ones used in this report.18 While Sheppard and Davies (2010) did not use exactly the same locations for thickness measurements as we did, i.e., their thickness measurements were referenced to the overall length of the muscle rather than the scleral spur, their CM25 measurement is probably in the general area of our CMT1, and their CM75 is probably similar to our CMT3. Both studies included an evaluation of ciliary muscle thickness differences, comparing distance and 4.00-D accommodative stimuli. When the measurements from our study and Sheppard and Davies (2010) are compared, we found a 45.2 μm mean thickening at CMT1, and Sheppard and Davies (2010) found a mean increase in thickness from 550 μm to 571 μm (or 21 μm) at CM25. Similarly, we reported a mean thinning of −45.9 μm at CMT3 and Sheppard and Davies (2010) reported a mean thinning from 174 μm to 166 μm (or 8 μm) at CM75. Sheppard and Davies (2010) also reported a measurement at 2 mm posterior to the scleral spur (CM2), and they found a very small, but statistically significant thinning of 21 μm at this location where we observed no significant change (−7.3 μm, p = 0.5). It is possible that we might have observed a statistically significant change if we had included more than 25 subjects in our study.
Our results and previous studies confirm that it is possible to measure the action of the ciliary muscle during accommodation using images from the Visante™, and that there appears to be a thickening of the anterior portion and a thinning of the posterior portion of the ciliary muscle during accommodation. Still, there are improvements that could be made to the protocol for making these measurements. It is possible that the best protocol for obtaining these measurements in future studies would include having a measure of the accommodative response at the exact time of image capture. This suggestion is based on observations from a couple of subjects with a wide range of accommodative responses during Visante™ imaging. For example, one subject had similar mean ± SD PowerRefractor readings while viewing the distance target at the two study visits, but the range of responses was more variable than some other subjects [Visit 1: −0.92 ± 0.17 D (range = −0.28 to −1.44) and Visit 2: −0.98 ± 0.19 D (range = −0.33 to −1.62)]. This subject also had more variable ciliary muscle measurements than other subjects. For example, CMTMAX at visit 1 was 780 μm and was 700 μm at Visit 2. We noted at least two subjects with discrepancies as extreme as these. Of course when we are measuring human subjects and relying on their responses and attention in a study such as this, some variability is inevitable and likely unavoidable. Nonetheless, in future studies, we will use an imaging protocol that includes measurements of accommodative response at the exact time each image is obtained to try to reduce the variability of the measurements and improve estimates of how much change in ciliary muscle thickness is required per diopter of accommodative response.
While accommodative monitoring during image acquisition is certainly one major limitation of the estimates of the change in ciliary muscle thickness during accommodation for the present study, we should also acknowledge one additional limitation. In our original publication of the semi-automatic algorithm we use to analyze the dimensions of the ciliary muscle, we included only cycloplegic images. When we began analyzing images that had been obtained during accommodation, it was clear that the anterior portion of the ciliary muscle was, at times, much darker than the cycloplegic images. It would not be surprising that the spacing of fibers within the muscle might vary between the accommodative and cycloplegic states of the muscle, resulting in different levels of contrast in the two types of images. The process of adjusting the contrast for some of the 4.00-D images, as depicted in Figure 3, did require some subjectivity on our part. For future studies, we are exploring how image capture and/or analysis might be adjusted so that all images could be analyzed with the same contrast settings in the algorithm.
While we are certain that refinements of the methods used to monitor in vivo accommodation in humans is important, we would also like to remind the reader that we have already discussed the fact that our results were similar to those of Sheppard and Davies (2010), even though our measurement techniques were not identical.18 In addition, we have collected data similar to these for additional studies, i.e., a study of children20 and also in a study of pre-presbyopic and presbyopic adults (unpublished data), where we made protocol adjustments but still obtained remarkably similar results. Based on the fact that we have repeated these results in other samples, we are confident that the results presented in this study are a good estimate of the change in ciliary muscle thickness per diopter of accommodation, despite the limitations of the study that are discussed above.
In summary, the combination of the Visante™ AS-OCT and the PowerRefractor provide a feasible method for future studies to continue to explore the exact nature of ciliary muscle contraction in adult subjects. In future studies, we intend to further refine the estimates of how the shape of the ciliary muscle changes with accommodation, and refine our procedures for determining the subject’s accommodative response. The information about ciliary muscle contraction and also the evolution of our methodology and protocols provided here and in our companion paper in this issue of the journal describing these measurements in children,20 should allow for more definitive and insightful studies of the ciliary muscle during accommodation in the future.
This research was supported by Beta Sigma Kappa (Lossing), the National Science Foundation NSF DMS 0811003 and Sloan Fellowship (Kao), and National Institutes of Health, National Eye Institute Grants R24-EY014792 (Sinnott), T32-EY013359 and K23EY019097 (Richdale).
The project described was supported by Award Number KL2 RR025754 (Bailey) from the National Center for Research Resources, funded by the Office of the Director, National Institutes of Health (OD). The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Center for Research Resources or the National Institutes of Health.
The Ohio State University has filed a provisional patent application on the behalf of the authors: U.S. Provisional Patent Application No. 61/594,027, filed February 2, 2012, entitled: “Semiautomatic Extraction of Algorithm for Images of the Ciliary Muscle”.