Purpose: To evaluate the clinical outcome of patients with refractory Neurotrophic Keratopathy (NK) in stages 2 and 3 treated with topical insulin.

Methods: A retrospective, observational analysis of eyes with NK in stages 2 and 3 refractory to standard medical and/or surgical treatment which were treated with topical insulin. Topical insulin (1 unit per mL) was applied 4 times a day; treatment was continued until persistent epithelial defect (PED) or ulcer resolved and then tapered accordingly. The primary outcome of the study was the complete resolution of the PED or ulcer. Best-corrected visual acuity (BCVA), days until complete resolution as well as anterior segment photographs were obtained.

Study type

Retrospective, observational, single-center study at the Department of Ophthalmology of Centro Hospitalar Vila Nova de Gaia e Espinho, EPE.

Data was obtained from patients' medical records and anterior segment photographs from October 1, 2018 to November 1, 2020.

Each participant and/or relative was given a detailed explanation and a written informed consent regarding possible alternatives, risks and benefits of off-label use of topical insulin drops. This study was approved by the board of Centro Hospitalar de Vila Nova de Gaia e Espinho (UIEC-2020-981914339595fa024c567ec5). This study complied with the tenets of the Declaration of Helsinki.

Patient selection

Patients included in this study were provided from the Cornea department of the Ophthalmology center in Centro Hospitalar de Vila Nova de Gaia e Espinho. Every patient included had an established diagnosis of NK (ICD10: H16.2) in stages 2 or 3 that was refractory to standard medical and/or surgical treatment and underwent treatment with topical insulin. All patients underwent complete ophthalmological exam including best-corrected visual acuity, slit-lamp examination, corneal sensitivity in the center and four quadrants of the cornea and fundoscopic evaluation. NK was graded based on slit-lamp and fluorescein stain findings - stage 2: epithelial defect (with or without a rim of loose epithelium) without stromal ulceration; stage 3: corneal ulceration and/or stromal lysis.

Topical insulin preparation and administration

Topical insulin drops were prepared by diluting 1 unit of fast-acting insulin per 1 mL of an artificial tear with a propylene glycol base. Drops were preserved at low temperature (2ºC) and were applied four times a day. A therapeutic corneal CL was placed in every patient and fluoroquinolone drops were applied to prevent possible CL side-effects.

Treatment was continued until NK PED or Ulcer resolved and tapered accordingly. Patients would discontinue topical insulin if the condition did not improve within 30 days or worsen. Follow-up was ensured on the 3rd,5th, and 7th day and then individualized during the full-extent of treatment. Anterior segment photos were taken on each visit.

Inclusion criteria included:

• Eyes with established diagnosis of NK in stages 2 or 3 refractory to standard treatment* that underwent topical insulin treatment.

The diagnosis of NK was based on the following criteria:

1. Decreased or absence of corneal sensation; AND

2. Corneal injury [PED and/or corneal ulcer] refractory to standard treatment*; AND

3. Clinical history of conditions related to trigeminal innervation impairment;

   • Standard treatment - lubrication with artificial tears, topical and/or oral antibiotics, topical and/or oral steroids, antiviral drugs, punctal plugs, therapeutic CL, amniotic membrane surgery and temporary tarsorrhaphy.´

We excluded every patient with corneal ulcer and signs of infection.

Established diagnosis of NK in stages 2 or 3 refractory to standard treatment that underwent topical insulin treatment.