In Vivo Confocal Microscopy of the Cornea in Nephropathic Cystinosis

We report a case of nephropathic cystinosis, with corneal crystals, imaged and analyzed by in vivo confocal microscopy. The subject had a visual acuity of 6/6 OU. Slitlamp biomicroscopy revealed dense, hyperreflective cysteine crystals corresponding to a "Gahl score" of 2.75 to 3.00 in both eyes. In vivo confocal microscopy highlighted dense, polyhedral crystals throughout the posterior stroma and crisscrossing crystals of similar density throughout the anterior stroma. Quantitative assessment suggested that crystal density was lowest in the middle stroma. The longitudinal and transverse diameters of a total of 100 crystals for anterior and posterior stromal layers were measured and analyzed. The advantages of in vivo confocal microscopy for microstructural analysis of a living human cornea might provide useful information about the natural history of crystal deposition and growth throughout the stromal layers in subjects with cystinosis.

Infantile nephropathic cystinosis may affect vision due to corneal disease, retinopathy, or glaucoma.¹ Of these 3 complications, glaucoma is the one most commonly associated with significant visual loss in adult subjects.² Corneal crystals may be observed as early as age 1 year, but they generally have a benign course and rarely cause severe visual loss, although photophobia is a common symptom.³ Recently, corneal crystals have been successfully treated with cysteamine drops.^3,4

A 24-year-old man received a diagnosis of cystinosis at the age of 14 months and underwent renal transplantation at the age of 22 years. From the age of 1 year he was treated with oral cysteamine hydrochloride that was changed to oral cysteamine bitartrate because of a manufacturer's change in 1999. He has always been aware of having photophobia but reported a worsening of his symptoms following the change in his systemic medication; however, because of serious life-threatening problems, his eye symptoms have been considered less important. Unfortunately, the subject cannot be offered treatment with topical cysteamine drops because the medication is currently not registered in New Zealand.

The subject's examination in our unit included photobiomicroscopy, in vivo confocal microscopy, tonometry, and fundus dilatation. The subject had a visual acuity of 6/6 OU. Slitlamp biomicroscopy revealed dense, hyperreflective cysteine crystals corresponding to a score of 2.75 to 3.00 in both eyes based on the library published by Gahl et al.³ Careful examination of the optical section revealed a random distribution of the crystals throughout the corneal thickness; however, there was apparently greater density toward the limbus (Figure 1).

Crystals were also identified over the anterior surface of the iris. Intraocular pressures were 12 and 13 mm Hg, OD and OS, respectively, and there were no signs of posterior segment involvement.

The methodology for in vivo confocal examination has been published previously.⁵ In the case reported herein, 4 passes at a 900-µm working distance were used. To minimize image glare and reflections from the hyperreflective crystals, the intensity of the light was decreased to approximately half the usual intensity for human subjects. The bright reflections prevented visualization of the cellular elements of the stroma, but the endothelial mosaic and superficial epithelial layers were clearly recognizable. Only the central cornea was examined because the reflections from the peripheral cornea were very bright. Two acquisitions were performed for each eye and a total of 900 images were saved onto a hard disk drive. Despite being very photophobic the subject had no difficulties undergoing the examination.

In vivo confocal microscopy visualized the crystals in detail and their appearance in both eyes was very similar. Immediately in front of Descemet membrane the crystals were dense and polyhedral, the longitudinal and transverse diameters measuring 85 ± 37 µm to 43 ± 29 µm (mean ± SD) (n = 50 measurements), respectively. Anteriorly, a few larger crystals were highlighted; however, most of the smaller crystals were needle shaped. Crystals in the anterior fifth of the stroma appeared to be crisscrossing and of similar density, with longitudinal and transverse diameters measuring 57 ± 41 µm to 21 ± 17 µm (mean ± SD) (n = 50 measurements), respectively. However, quantitative assessment suggested that crystal density was lowest in the middle stroma (Figure 2). The corneal endothelium and epithelium appeared to be normal in respect to structure and cell density; however, stromal keratocytes were shadowed by the crystals thus making evaluation impossible.

The cornea has been called "the window to cystinosis"; however, the cause for the severity and progression of the corneal presentation is still not clear.³ It has been proven that cysteamine drops may have a beneficial effect in up to 62% of symptomatic patients and patients with recurrent erosion syndrome.³ However, the effect of the medication has been evaluated "semiquantitatively" on the basis of clinical photography.³

We report the application of in vivo confocal microscopy for imaging both corneas of a subject who has nephropathic cystinosis. By careful modulation of the confocal settings, the crystals can be visualized in detail, providing an excellent opportunity for qualitative and quantitative analysis. We found the central cornea to be best for analysis in our subject, but we believe that this site may be beneficial for several reasons: the central cornea is thinner with fewer crystals, providing easier structural discrimination and measurement, the examination can be repeated using the pupillary margin as a landmark, and, finally, the dynamics of the central cornea may be more closely associated with the patient's symptoms such as photophobia and glare. In addition, the examination by in vivo confocal microscopy is relatively quick (2-5 minutes) and in this report our subject coped well with the procedure with no unusual symptoms.

An alternative method of examination in these subjects is ultrasound biomicroscopy (UBM).² The advantage of this technique is visualization of a larger area of the cornea and adjacent structures including the conjunctiva, iris, anterior chamber angle, and lens.² However, owing to low magnification, comparable to the conventional slitlamp, UBM provides only a semiquantitative analysis of the corneal crystals.² Our report demonstrates the advantage of in vivo confocal microscopy, not only in visualizing the crystals but also in accurately measuring their dimensions. In vivo confocal microscopy will equally cover the same age groups as previously examined by UBM (>16 years); however, it might provide better follow-up of the corneal changes especially following application of cysteamine eyedrops.²

Our in vivo confocal observations of the distribution of the corneal crystals throughout the corneal thickness (z-dimension) did not concur with the slitlamp observations. In the latter the crystals appeared to be evenly distributed throughout the optical slice, whereas, confocal microscopy highlighted greater density anteriorly and posteriorly, with least crystal density in the middle stroma. By isolating the area of interest, confocal microscopy offers the advantage of minimizing the effect of adjacent regions and in the future might provide useful information about the natural history of crystal deposition and growth throughout the stromal layers in subjects with cystinosis.

The study was supported in part by an unrestricted grant from the Maurice and Phyllis Paykel Trust.

The authors have no financial interest to declare.

Corresponding author: Charles McGhee, PhD, FRCOphth, Discipline of Ophthalmology, University of Auckland, Private Bag 92019, Auckland, New Zealand (e-mail: c.mcghee@auckland.ac.nz).