The evolution of ophthalmic imaging

From the earliest days of the first ophthalmoscope, ophthalmologists have had an advantage over most other specialties in their ability to see inside the body to evaluate the fundus and other ocular structures of their patients. Indeed, the ophthalmoscope, described by Hermann Helmholz in 1851, is still part of regular practice, and the concepts of magnification and illumination continue to be the basis of most imaging technologies now in use. Ophthalmologists currently have at their fingertips an impressive array of imaging tools for diagnosing disease, guiding interventions, and managing outcomes. These tools are a key part of everyday practice in anterior segment surgery, particularly cataract and refractive surgery, corneal transplantation, keratoconus, glaucoma and ocular surface disease. Imaging of the anterior segment has evolved significantly over the past decade. The technology has improved in many areas including resolution, acquisition time, speed of acquisition, better output maps etc, notes Soosan Jacob MD, FRCS, Director and Chief, Dr. Agarwal's Refractive and Cornea Foundation, Chennai, India. “In addition, the ability to image more accurately the anterior and posterior surface as well as layer imaging in the form of epithelial mapping have improved the ability to understand the cornea better. Integration of better software, calculations, nomograms and predictors also help us to analyze corneal maps better,” she told EuroTimes. Keratoconus screening and diagnosis Screening for corneal ectasia in general and keratoconus in particular is one of the biggest challenges facing clinical ophthalmology. The nature of the cornea leads to unexpected outcomes in refractive and keratoplasty surgeries even when the procedures seem to have gone well. This is attributable to variations in the biomechanical properties of individual corneas and inadequacies in the ability to diagnose ectasia in the early stages. Screening at the primary level includes slit-lamp evaluations and corneal topographic imaging. Corneal specialists can then call on many imaging options including optical coherence tomography (OCT) such as the Visante (Carl Zeiss Meditec), Scheimpflug imaging with Oculus Pentacam HR, and Placido Disk keratography (Oculus Keratograph) to evaluate the anterior and posterior surfaces of the eye as well as the pachymetry. More recently, the ORA ocular response analyzer (ORA Reichert) and the Corneal Visualization Scheimpflug Technology (Corvis ST) have entered the arena. Both of these use a puff of air to evaluate the biomechanical properties of the cornea. Researchers are still working to translate data provided by these systems into useful diagnostic values. “I think epithelial mapping with either OCT or very high-frequency (VHF) digital ultrasound (Artemis, ArcScan) would be of help in screening for keratoconus. We also have tools that have been added to imaging technologies such as indices and screening tests such as the BAD D value, Corvis Biomechanical Index (CBI), Tomographic Biomechanical Index (TBI). Devices based on biomechanical properties still have a far way to go before being able to be translated into perfect accuracy for each patient every time, but if perfected, can become an invaluable tool,” commented Dr Jacob. While these instruments are invaluable for cornea specialists, none of them can provide a reliable diagnosis of keratoconus for the simple reason that there is currently no consensus on how best to classify and predict ectasia. The best known classification systems are the Amsler-Krumeich and Collaborative Longitudinal Evaluation of Keratoconus (CLEK), but neither are considered to be the perfect solution. In the search for a more accurate screening technique, Seok Hyun Yun PhD, and colleagues at Harvard University Medical School, Cambridge, USA, are using a non-invasive technique known as Brillouin microscopy imaging. This technique utilizes a low-power near infrared laser to determine the mechanical compressibility of tissue, otherwise known as the longitudinal modulus, by analyzing the return signal spectrum. Brillouin imaging maps small frequency changes, known as Brillouin frequency shifts, in order to created a 3D scan of the cornea. Early research findings are promising. Clinical studies have revealed significant differences in the elastic properties of normal corneas compared to those diagnosed with mild and severe keratoconus. Brillouin imaging studies have also demonstrated biomechanical changes after corneal collagen cross-linking treatment in keratoconus patients. The technique has also been used to image the crystalline lens and the sclera. Research with Brillouin microscopic imaging is furthest along in the area of keratoconus diagnostics. Dr Yun believes that the technique should become clinically useful in the not- too -distant future. “Our data warrant further clinical validation. A promising next step is to derive metrics that combines the bio-mechanical data with morphological features and establish its clinical benefits—sensitivity and specificity—over current morphology-only diagnostic criteria,” Dr Yun told EuroTimes. Beyond applications in keratoconus diagnostics, Dr Yun believes Brillouin microscopy has potential applications in many areas of ophthalmology, including cataract and refractive surgery, noting that this imaging technique has the potential to predict the refractive changes in the cornea after corneal incisions and post-LASIK more accurately. Dr Yun and colleagues built the instruments and the related software used in his research in his lab at Harvard Medical School. He is also the medical research director of start-up company, Intelon Optics, that hopes to commercialize a Brillouin eye scanner product. INTRAOPERATIVE OCT? Optical coherence tomography imaging is now used in virtually every aspect of ophthalmology, from detailed studies of the optic nerve and retina to guiding femtosecond assisted cataract surgery, corneal surgery, and even sub-micron analysis of ocular surface disease. Intraoperative imaging has become a useful tool for many cataract and refractive procedures.  However, the best use of this approach, particularly in the area of corneal transplantation, is still being debated. Nino Hirnschall MD, PhD, FEBO, Vienna, Austria and Massimo Busin, MD, PhD, University of Ferrara, Italy, discussed the pros and cons of intraoperative OCT for corneal transplantation in a EuroTimes Eye Contact interview. https://player.escrs.org/eurotimes-eye-contact/intraoperative-oct-for-the-anterior-segment-massimo-busin Dr Hirnschall noted that the visualisation provided by intraoperative OCT could allow Descemet's membrane endothelial keratoplasty (DMEK) to be performed in an eye with a cloudy cornea, where otherwise Descemet's stripping automated endothelial keratoplasty (DSAEK) would be needed by default Intraoperative OCT can also identify the remnants of Descemet's membrane, allowing complete Descemet's removal in both DMEK and DSAEK. This is important because the detachment rate is is higher if the graft is placed on top of remnant's of Descemet's membrane. There is also published clinical research indicating that the improved visualisation provided by intraoperative OCT could mean less graft manipulation time,  and so less endothelial cell loss. OCT also assures proper graft orientation. Professor Busin, a high volume corneal surgeon, is not convinced. “In my field of corneal transplantation I feel that intraoperative OCT is really not that useful, and doesn’t justify the expense. It has two main limitations- it only give small slices, and you have to scan the whole surface yourself. It is difficult to see what you are looking for. Plus it moves while you are using it and is very awkward to use. It is difficult to obtain useful information that you can use to adjust your surgery,” he said. Dr Hirnschall noted that during both DMEK and DSAEK, intraoperative OCT enables identification of fluid in the donor-graft interface and of graft adherence. For deep anterior lamellar keratoplasty (DALK) or penetrating keratoplasty procedures, intraoperative OCT gives surgeons exact information about trephination depth and, during DALK, it can guide cannula depth when creating the big bubble. Dr Busin commented that there are techniques besides OCT that surgeons use to determine DMEK graft orientation, including by looking through the microscope. If still uncertain, other practical and less expensive approaches include the handheld slit beam, an S or F stamp, the Moutsouris sign or endoillumination. Dr Jacob agrees that although intra-operative OCT is proving useful for certain corneal surgeries such as DMEK or pre-Descemet's endothelial keratoplasty (PDEK), where it can help identify graft orientation, there are other ways to accomplish this accurately with other techniques such as the endoilluminator-assisted DMEK or PDEK described by her. “Intraoperative OCT may also help identify fluid in the interface," she said. However, with the air pump assisted technique, again described by her, interface fluid can be decreased in these surgeries so that confirmation with i-OCT would not be required.  In DALK too, though it has been similarly claimed to be of benefit, an experienced surgeon can generally make do well without the aid of intraoperative OCT. "Current disadvantages of introperative OCT such as additional time taken, the need to constantly keep focusing the scan exactly on the area of interest in the x,y and z axes, rapid shifting of focus on eye movements, inversion of the image, as well as, of course, the additional cost involved, make it less practical," she commented. Discovering the basis of dysphotopsia Even as IOL designs continue to improve and biometry and improved surgical techniques evolve, dysphotopsia remains the leading cause of patient dissatisfaction after uncomplicated cataract surgery. While the glare and halo symptoms of positive dysphotopsia are believed to be attributable to square edged IOL design, the causes of negative dysphotopsia are less well understood. Negative dysphotopsia presents as persistent dark shadows in the patient’s peripheral visual field even after an uneventful cataract surgery With support from an ESCRS clinical research award, Jan-Willem Beenakker MD and colleagues at the Leiden University Medical Centre in the Netherlands are conducting a study known as vRESPOND, (Virtual REfractive Surgery for the Prevention Of Negative Dysphotopsia), the goal of which is to uncover the pathological origin of negative dysphotopsia. The researchers are using advanced magnetic resonance imaging (MRI) techniques and other imaging modalities to make models to study the exact path of light through the eye. They are using the imaging data to build personalized three-dimensional models that link the optical characteristics of negative dysphotopsia with  patients' subjective complaints. Dr Beenakker notes that conventional optical techniques used to measure distances in the eye have significant systematic errors for off-axis measurements due to refraction. Since MRI is not affected by refraction it can be an important tool to study the retinal shape. However, the sensitivity of MRI to eye-motion has limited these evaluations to two dimensional or low-resolution three dimensional studies. The Leiden team have a developed a method that takes advantage of advances in high field MRI to quantify full three dimensional images. “The acquisition of the data has been finished and a large part of the data have been analysed. We have found significant differences in the anterior chamber configuration of patients with and without negative dysphotopsia. These findings provide the first clinical confirmation of earlier ray-tracing studies on the relation between the angle kappa and negative dysphotopsia,” Dr Beenakker told EuroTimes. The novel, patient-specific, 3D eye models provided by this study should prove useful for many different future applications in refractive surgery, enabling innovative clinical solutions for diseases which affect the quality of vision, he predicted. Objective basis of subjective complaints Related research is being conducted in Portugal by Joaquim Murta MD, PhD, Miguel Castelo Brano, MD, PhD and Andreia Rosa MD, phD  in the department of ophthalmology, University of Coimbra. With support from another ESCRS clinical research award. The Neuroadaptation after Cataract and Refractive Surgery Study, or NECSUS, is carried out jointly by the Universities of Coimbra and Maastricht. It is set to study the occurence of patient-reported subjective difficulties and neuroadaptation after cataract surgery with different IOLs (multifocal, monofocal and extended depth of focus lenses) andf levels of adaptation, using functional magnetic resonance imaging (fMRI). A group of unhappy multifocal IOL patients expecting explantation surgery will also be included. Prof Murta notes that positive dysphotopsia, in the form of glare, halos and starbursts remain an important cause of dissatisfaction after cataract surgery. Moreover, these unwanted effects are considered to be limiting the more widespread use of multifocal intraocular lenses. In a pilot study funded by the ESCRS in 2016, it was shown, for the first time, an association between patients's reported subjective difficulties and fMRI outcomes, independently of optical parameters and psycophysical performance. The increased activity of cortical areas dedicated to attention, learning, and cognitive control and to task goals probably represent the beginning of the neuroadaptation process of multifocal intraocular lenses. Thus was demonstrated a positive change in brain activity months after surgery, accompanying clinical improvement. The NECSUS study aims to extend this knowledge by comparing different IOL designs and evaluating functional connectivity among visual and attention-related areas, especially in patients in which neuroadaptation failed to occur. Information produced by this research should provide valuable knowledge leading to the identification of therapeutic targets as well as intraocular lens characteristics that are more likely to trigger neuroadaptation circuits effectively. Hopefully, this could lead to practical clinical applications for better medical and surgical outcomes, notes Prof Murta. Imaging the future When asked what she would you like to see in imaging technology that would help in day- to -day clinical work in the future, Dr Jacob commented, “Among things I would anticipate greater use of artificial intelligence and greater accuracy in layer-by layer analysis. We need better devices to measure corneal biomechanics reliably. We need a greater understanding of the eye's refractive mechanisms for extremely accurate IOL power measurements, especially in aberrated and post-surgical corneas. I also hope we will see microscopic level imaging of the trabecular meshwork and outflow pathways in the glaucoma area.   Massimo Busin: mbusin@yahoo.com Jan Willem Beenakker: j.w.m.beenakker@lumc.nl. Soosan Jacob: dr_soosanj@hotmail.com Seok Hyun Yun: syun@mgh.harvard.edu Joaquim Neto Murta: jmurta@netcabo.pt