Scleral Fine Needle Diathermy Using A 30-Gauge Needle And Intravascular Lidocaine To Treat Lipid Keratopathy

Fine needle diathermy (FND) is an established method of occluding corneal vessels and is conventionally performed at the limbus or intrastromally. Limitations are risk of intrastromal bleeding and failure. There are also technical challenges in depth assessment, risk of inducing astigmatism or corneal scarring with repeated intrastromal diathermy. By use of a 30G needle into a scleral vessel and adjunctive use of lidocaine it is possible to safely isolate and treat single isolated vessels. Lidocaine at low doses is vasoconstrictor which can help confirm causative vessels. Here, we report a modification to conventional FND which allows safe, precise isolation and treatment of causative corneal vessels in lipid keratopathy.

We report a single case at the Norfolk and Norwich University Hospital, a tertiary centre corneal service based in the United Kingdom.

A 69-year-old woman was referred for symptomatic reduced left eye (OS) visual acuity. She presented with logMAR best corrected visual acuity (BCVA) of 0.00 in her right eye (OD) and 0.20 OS. She had history of long-term soft contact lens wear but with no history of microbial keratitis. She had no significant medical history or surgical history, and no family history of hyperlipidaemia.

Clinically, examination OS demonstrated discrete 2 x 2.5 mm grey-white corneal opacity with a single feeder vessel. Her examination OD was unremarkable. She had laboratory workup which demonstrated hypercholesterolaemia of 8.6 mmol/L but was otherwise unremarkable and viral polymerase chain reaction (PCR) of conjunctival swabs were negative for herpes simplex or varicella zoster. Findings were most in keeping with lipid keratopathy.

Under local anaesthetic, a localised conjunctival peritomy was performed to obtain access to the causative vessel proximally at the scleral level. Once the needle was introduced to the correct depth, lignocaine 2% was injected into the suspected causative vessel through a 30G needle until the vessel blanched. Diathermy was then applied via the same 30G needle and used to close the exposed conjunctiva. Post-operative dexamethasone and moxifloxacin were given.

At one month follow up, her left visual acuity remained 0.20 but the patient reported they were symptomatically better and had visible regression of the feeder vessel and associated area of lipid keratopathy.

We report a safe modification to conventional FND through scleral vessel access with 30G needle and intravascular lignocaine which reduces risk of iatrogenic corneal damage, failure, and intrastromal bleeding. In this case, we have described the technique in treating lipid keratopathy, but the same technique is applicable for neovascularisation from herpetic disease, in high-risk patients with stromal vascularisation prior to keratoplasty or patients post-keratoplasty.