A Case Report Illustrating the Postoperative Course of Descemetorhexis without Endothelial Keratoplasty with Topical Netarsudil Therapy

case reports in ophthalmology
18 Nov, 2019 ,

Pimpiroon Ploysangam and Sangita P. Patel report a 51-year-old patient with visually significant confluent central guttae from FECD who underwent a successful DSO with immediate post-operative use of the Rho-associated protein kinase (ROCK) inhibitor (netarsudil). They report the preoperative and post-operative slit lamp images, specular microscopy data, and corneal topography, thickness, and densitometry data. These represent a unique data set for this new surgical treatment option for FECD. Despite a small descemetorhexis, they show the improvement in corneal thickness and opacity extends beyond the boundaries of the descemetorhexis. Early initiation of a ROCK inhibitor was a successful treatment for this patient.

Full content

The patient is a 51-year-old Caucasian female with history of FECD who presented with gradually worsening vision over the past year. She had noted fluctuating vision, photophobia, difficulty reading, and glare with oncoming headlights while driving. Her ocular history was remarkable for ocular rosacea with inspissated meibum and eyelid margin vascularization and glaucoma suspect due to cup to disc asymmetry. Ocular medications included intermittent use of artificial tears and sodium chloride 5% ointment.

Her eye exam revealed a corrected Snellen visual acuity of 20/40 (right; −0.75 +3.00 × 100) and 20/30 (left; −0.75 +2.75 × 085). Her corneas had central confluent endothelial guttae in both eyes measuring 4.0 × 4.0 mm (right) and 3.8 × 4.0 mm (left) with associated stromal haze in both eyes. The peripheral corneas in both eyes were relatively spared of guttae. There was no clinically apparent corneal edema on slit lamp exam. Ultrasound pachymetry (DGH 555 Pachette 3 Ultrasonic Pachymeter, DGH Technology, Inc., Exton, PA) showed a central corneal thickness of 608 µm (right) and 595 µm (left). She also had 1+ nuclear sclerotic cataract in both eyes. Dilated fundus exam was unremarkable.

Specular microscopy imaging (Konan Specular Microscope Noncon Robo Series NSP-9900, Konan Medical Inc., Japan) was completed and showed central confluent guttae (unable to count endothelial cell density). The peripheral cornea showed scattered, nonconfluent guttae with cell densities of 1508–1550 cells/mm2 in the left eye. The location of “peripheral” corneal endothelial imaging by the Konan Specular Microscope is not specified, but we suspect a midperipheral location based upon the distribution of guttae in the corresponding red-reflex image. Scheimpflug tomography (Pentacam, Oculus, Inc., Arlington, WA) showed 2 diopters of regular with-the-rule astigmatism. The minimum central thickness was 593 µm in the left eye and the decreased back surface elevation suggested the presence of subclinical corneal edema.

Sagittal corneal curvature, back surface elevation, and pachymetry maps pre- and post-DSO for the left eye.

The patient's visual symptoms were felt to be secondary to corneal changes from FECD and mild cataract. Surgical options were reviewed with her including cataract extraction with intraocular lens placement with or without DSEK or DSO. Patient decided to proceed with cataract extraction with intraocular lens placement and DSO for her left eye, with the understanding that she might need corneal transplant should her cornea fail to clear following DSO.

The patient underwent routine cataract surgery with phacoemulsification and intraocular lens placement under topical anesthesia. Following lens implantation, the central corneal area of densest guttae was marked at 4 × 4 mm with calipers on the corneal surface. While the eye was still filled with cohesive viscoelastic, a reverse sinskey hook was used to score and lift the Descemet's membrane/endothelium for approximately 1–2 clock hours in the marked region. Using the lifted edge of Descemet's membrane, a continuous descemetorhexis was pulled approximating the marked central region using utrata forceps. Viscoelastic was then removed from the anterior chamber, and the remainder of the case was completed per standard cataract surgery protocol.

On her postoperative day 1 visit, uncorrected vision in the left eye was 20/125 (pinhole to 20/60). Her cornea showed removal of greatest area of guttae in her left eye measuring 4 mm (vertical) × 3.4 mm (horizontal), with slight inferior displacement of area of stripping relative to the pupil, corresponding to the area with densest guttae preoperatively. Ultrasound pachymetry was 936 µm. The anterior chamber was quiet, and posterior chamber intraocular lens was in proper position. The patient was started on the following eye drop regimen for the left eye: prednisolone acetate 1% 4x/day, polymyxin b sulfate-trimethoprim 4x/day, netarsudil at bedtime, and polymyxin B/neomycin/dexamethasone ointment at bedtime.

At postoperative week 1, her visual acuity in the left eye was 20/250 (pinhole to 20/100). Her cornea showed stromal and focal reticular epithelial edema in the left eye measuring 3 mm (vertical) × 2.8 mm (horizontal). Reticular edema has previously been noted with netarsudil use. The area of stripping was clearly delineated and edges of this area were noted to be clearing. The anterior chamber remained deep and quiet. Scheimpflug tomography imaging (Pentacam) showed inferocentral thickening and steepening in the area of Descemet's stripping. At this visit, the patient noted pain which was thought to be secondary to the corneal epithelial edema. Frequent application of artificial tears did not alleviate the pain. A bandage soft contact lens was placed for 1 week. All eye medications were continued except for the polymyxin B / neomycin/dexamethasone ointment.

At the postoperative one-month visit, her uncorrected vision measured 20/150 (pinhole 20/50). The corneal edema had resolved, but she continued to have a central anterior stromal haze. Scheimpflug tomography (Pentacam) showed a minimum central thickness at 565 µm. The endothelium could be imaged by specular microscopy (Konan) and showed central endothelial cell counts of approximately 473 cells/mm2. The polymyxin B sulfate-trimethoprim and netarsudil eye drops were stopped, and the prednisolone acetate 1% eye drop was tapered off over the next 3 weeks.

Her vision continued to improve at postoperative month two, with her best corrected vision at 20/40 (right; −0.75 +3.00 × 100 OD) and 20/25 (left; −1.50 +2.25 × 080). Due to persistent pain, a bandage soft contact lens was again placed at the postoperative month two visit for one more week, and the patient was started on fluorometholone eye drops 2x/day in the left eye. Pain has been reported as a side effect of instillation of netarsudil eye drops; however, she had stopped using netarsudil eye drops 1 month prior.

At 3 months, vision remained stable from the prior month. Specular microscopy (Konan) showed a central cell density of 754 cells/mm2. Scheimpflug tomography (Pentacam) demonstrated decreased corneal thickness compared to all prior measurements for the left eye. In addition, evaluation of corneal opacity by Scheimpflug corneal densitometry showed improved (reduced) densitometry values from both the anterior and posterior corneal layers, and from the central and peripheral corneal zones compared to preoperatively.

She continued to have pain in her left eye which did not respond to bandage contact lens placement but was alleviated with periodic use of fluorometholone 0.1% eye drops once daily as needed. The etiology of the pain was not evident on exam, thus we suspect a neuropathic component. Fluoromethalone eye drops were continued for an additional 2 months. At 9 months postoperatively, her left eye pain had resolved. Her central corneal endothelial cell density was 700 cells/mm2 and corneal pachymetry was 555 µm. The cell density of the peripheral corneal endothelium was noted to be decreased from pre-operatively.