Late-onset bilateral epithelial ingrowth following rapid corneal decompensation owing to amantadine

Abstract

Epithelial ingrowth (EI) can occur at the interface between the flap and the stromal bed of the patient's cornea following laser assisted in situ keratomileusis (LASIK) surgery. The introduction of corneal epithelial cells into the interface can happen during surgery or after the procedure owing to loss of contact inhibition and the presence of a pathway to the interface. Epithelial ingrowth is a relatively uncommon complication, with a reported incidence of 3.9% following primary LASIK and up to 12.8% following enhancement with a flap lift.1Mohamed T.A. Hoffman R.S. Fine I.H. et al.Post-laser assisted in situ keratomileusis epithelial ingrowth and its relation to pretreatment refractive error.Cornea. 2011; 30: 550-552Crossref PubMed Scopus (22) Google Scholar While usually asymptomatic, EI may lead to decreased vision owing to direct intrusion of cells into the visual axis, irregular corneal astigmatism, or melt of the overlying flap. Risk factors for EI have been described; however, systemic drug intake is not reported. We report a case of late-onset bilateral EI induced by acute corneal edema 6 years following primary LASIK in a patient taking amantadine, botulinum toxin (BT) injections, and natalizumab for multiple sclerosis. In April 2019, a 49-year-old white female patient was referred for acute bilateral corneal edema unresponsive to topical corticosteroids. Her past medical ocular history was notable for uncomplicated bilateral LASIK in 2013, with uncorrected visual acuity (UCVA) of 20/15 in her right eye (OD) and monovision of J1 in her left eye (OS). (Best spectacle corrected visual acuity 20/15 OS with a −1.75 diopter [D] correction). No EI OU. Her past medical history was relevant for multiple sclerosis, diagnosed in 2012 and treated with 200 mg of oral amantadine daily since 2012, BT injections every 3 months for leg spasticity since 2013, and monthly intravenous natalizumab since 2017. On examination, UCVA was 20/200 OD and 20/150 OS (J8 and best spectacle corrected visual acuity 20/80 OS). Anterior segment was unremarkable OU with only rare cells OS. Slit lamp examination showed moderate corneal edema, no keratic precipitates, normal LASIK flaps, no diffuse lamellar keratitis, no infiltrates, no EI, and no interface haze (Fig. 1A, B). Ultrasound pachymetry revealed a central corneal thickness of 666 microns (μm) OD and 763 μm OS. Digital and tonometer intraocular pressure was normal and fundus examination was unremarkable OU. Hourly topical prednisolone acetate 1% was continued while sodium chloride 5% drops and ointment were added without improvement. The patient underwent sequential Descemet stripping endothelial keratoplasty (DMEK) grafts OU. Due to a mild head tremor, the patient required rebubbling OU for partially nonadherent grafts and a second DMEK following failed rebubbling OD. Epithelial ingrowth appeared 20 weeks and 25 weeks after the initial decrease in vision for OD and OS respectively (Fig. 1C, D). The EI occurred before DMEK in OD and after DMEK in OS. Pathology reported rarified endothelial cells with no inflammation. The diagnosis of rapid and irreversible endothelial decompensation owing to amantadine was emitted. Amantadine was stopped to prevent decompensation of the corneal grafts. As corneal edema resolved, the size of EI decreased OU but remained central and visually significant in OD with an UCVA of 20/60. Surgical excision of EI was performed OD in January 2020, with resulting UCVA of 20/30 with mild interface haze and peripheral non–visually significant EI recurrence (Fig. 1E). The EI in OS was non–visually significant and remained stable with a UCVA of 20/20 (Fig. 1F). As of May 2020, both eyes remained phakic, with no development of cataract, and did not need further surgery. To our knowledge, this is the first documented case of late-onset bilateral EI owing to acute bilateral corneal decompensation following BT injection in a patient taking amantadine. Higher daily dosage of amantadine (400 mg daily) was found to induce greater changes in endothelial parameters than a long-term intake.2Tandon R. Daggumilli S. et al.Corneal evaluation in patients with Parkinsonism on long-term amantadine therapy.Cornea. 2019; 38: 1131-1136Crossref PubMed Scopus (7) Google Scholar Subsequent corneal edema is a rare but serious side effect that can occur any time during treatment with a dose-dependent increase in risk and can be reversible if amantadine is stopped early.3Lee P.Y. Tu H.P. Lin C.P. Chang C.H. Cheng K.C. Lin C.C. Hsu S.L. Amantadine use as a risk factor for corneal edema: a nationwide cohort study in Taiwan.Am. J. Ophthalmol. 2016; 171: 122-129Abstract Full Text Full Text PDF PubMed Scopus (8) Google Scholar While a case report of a patient acquiring a necrotising retinopathy secondary to varicella zoster while on natalizumab was described in the literature,4Haseltine Van Tassel S. Gupta M.P. Orlin A. et al.Progressive outer retinal necrosis in a multiple sclerosis patient on natalizumab.Neurology. 2015; 84: 2198-2199Crossref PubMed Scopus (8) Google Scholar no cases of corneal endothelial damage were described. One case report described acute anterior uveitis activated subsequent to BT injection in a patient with Behçet's disease, which was attenuated by infliximab.5Sasajima H. Syunsuke Y. Hiromu O. Masahiro Z. Botulinum toxin-induced acute anterior uveitis in a patient with Behçet's disease under infliximab treatment: a case report.Journal of Medical Case Reports. 2017; 11: 124Crossref PubMed Scopus (1) Google Scholar However, no ocular side effects have been described from the combination of BT injections and intravenous natalizumab. It is possible that the BT injection inducing the myasthenic syndrome also induced a bilateral anterior uveitis, left undiagnosed owing to the long delay between the decrease in vision and the first eye exam (1 month). The anterior uveitis could have led to a rapid and permanent corneal edema in a patient with an already weakened endothelium from the amantadine. In conclusion, patients taking amantadine on a long-term basis should be referred in ophthalmology for monitoring and referred quickly if vision decreases after BT injection, as it can lead to visual complications. Endothelial cell count should be monitored, and amantadine stopped if endothelial cell change occurs to avoid irreversible damage. In the rare event of a rapid and irreversible corneal decompensation occurring in a patient with previous LASIK, prompt endothelial graft should be considered before epithelial ingrowth further complicates restoration of vision. The authors have no proprietary or commercial interest in any materials discussed in this article.