Corneal Nerve Reconstruction (Treatment for Neurotrophic Keratopathy)

Key Points at a Glance

Corneal neurotization (CN) is the only surgical procedure that fundamentally restores corneal sensation for neurotrophic keratopathy.
Transfer healthy donor sensory nerves (such as the supratrochlear and supraorbital nerves) to the corneal limbus to promote axonal regeneration.
There are direct and indirect methods (e.g., sural nerve graft), with no significant difference in sensory recovery at 12 months ¹⁾.
A scoping review of 164 eyes showed an average success rate of 90%, with preoperative CB value improving from 2.7 mm to 36.0 mm postoperatively ²⁾.
Sensory recovery begins at 4–6 months and peaks at 12 months ¹⁾²⁾.
Better outcomes are expected in younger patients and with early intervention. Older age and long-term denervation are poor prognostic factors ⁷⁾.

1. What is Corneal Neurotization?

Corneal neurotization (CN) is a surgical procedure that transfers healthy sensory nerves to the cornea to achieve reinnervation for neurotrophic keratopathy (NK) ¹⁾. NK is a refractory disease caused by trigeminal nerve damage, leading to reduced or absent corneal sensation and disruption of epithelial homeostasis, with a prevalence of approximately 5 per 10,000 people.

Conventional NK treatments such as artificial tears and tarsorrhaphy are symptomatic and do not address the underlying nerve damage. CN is the only surgical method that can stabilize the ocular surface and improve long-term prognosis by restoring corneal sensation ¹⁾.

Historical Background

1972: Samii proposed the first concept. He bridged the greater occipital nerve and proximal ophthalmic nerve with a sural nerve graft, but it required a large craniotomy and was not practical ³⁾
2009: Terzis reported the first case of direct corneal neurotization. The contralateral supratrochlear and supraorbital nerves were directly transferred to the corneoscleral limbus ¹⁾⁴⁾
2014: Elbaz et al. developed the indirect minimally invasive technique (MICN) using a sural nerve graft, which could be performed through small incisions ⁴⁾

Since then, surgical techniques have rapidly diversified, including endoscopic approaches and the introduction of acellular allografts ¹⁾.

2. Main Symptoms and Clinical Findings

Corneal Neurotization for Neurotrophic Keratitis image

Diana Carmen Dragnea, Iva Krolo, Carina Koppen, Callum Faris, et al. Corneal Neurotization—Indications, Surgical Techniques and Outcomes 2023 Mar 13 J Clin Med. 2023 Mar 13; 12(6):2214 Figure 2. PMCID: PMC10059768. License: CC BY.

A shows punctate keratopathy, B shows persistent epithelial defect, and C shows more advanced lesions with stromal melting. Fluorescein staining reveals the different stages of neurotrophic keratitis.

The symptoms and findings of neurotrophic keratopathy that are the target of CN are shown below.

Subjective symptoms

Patients with NK are less aware of symptoms due to decreased corneal sensation.

Lack of foreign body sensation: Difficulty feeling pain even when a foreign body enters the cornea
Decreased visual acuity: Slowly worsens with progression of epithelial damage and corneal opacity
Dry eye symptoms: Dryness due to decreased reflex tear secretion
Redness: Conjunctival injection due to secondary inflammation

Clinical findings (findings confirmed by physician examination)

The severity of NK is assessed using the Mackie classification.

Stage 1 (Mild)

Punctate superficial keratopathy: Irregularity and hyperplasia of the corneal epithelium

Shortened tear breakup time: Decreased tear film stability

Superficial neovascularization: Vascular invasion from the limbus

Stage 2 (Moderate)

Persistent epithelial defect: Epithelial defect with rolled edges

Stromal edema: Accompanied by Descemet’s membrane folds

Anterior chamber inflammation: Mild flare

Stage 3 (Severe)

Corneal melting: Stromal thinning and ulceration

Corneal perforation: Most severe, with progressive melting

Secondary infection: Increased risk of secondary infection

Corneal sensitivity is quantitatively assessed using the Cochet-Bonnet esthesiometer (CB). On a scale from 0 mm (no sensation) to 60 mm (normal), values below 20 mm indicate hypoesthesia¹⁾.

3. Causes and Risk Factors

NK results from damage at any point along the trigeminal nerve pathway¹⁾³⁾.

Viral infection: Herpetic keratitis (HSV, VZV) is the most common, accounting for 27–32% of all cases¹⁾
Post-neurosurgery: After acoustic neuroma or meningioma resection, the most common surgical cause. Due to proximal trigeminal nerve injury³⁾
Trauma: Includes chemical injury (alkali, acid), iatrogenic causes after corneal surgery (LASIK, corneal transplantation, etc.)
Congenital: Congenital trigeminal nerve hypoplasia/aplasia, Goldenhar-Gorlin syndrome
Systemic diseases: Diabetes (diabetic keratopathy), multiple sclerosis, leprosy
Drug-induced: Chronic use of topical anesthetics, beta-blocker eye drops, benzalkonium chloride-containing preparations

4. Diagnosis and Examination Methods

Corneal Sensitivity Measurement

Cochet-Bonnet esthesiometer (CBA): Uses a 0.12 mm diameter nylon filament. Start at 60 mm and shorten in 10 mm increments until a response is obtained. Measure at the central cornea and four quadrants¹⁾
Qualitative assessment: Lightly touch the tip of a cotton thread to the cornea and check for the presence of a blink reflex¹⁾

In Vivo Confocal Microscopy (IVCM)

Non-invasively observe the density and morphology of the subbasal nerve plexus²⁾. It is useful for preoperative baseline and postoperative nerve regeneration monitoring. Improvement in nerve parameters is confirmed from 3 months postoperatively²⁾⁴⁾.

Preoperative Evaluation of Donor Nerves

Evaluate sensation of the supratrochlear, supraorbital, and infraorbital nerves using a cotton thread or Semmes-Weinstein monofilament¹⁾
If ipsilateral nerve function is reduced, also test the greater auricular nerve (sensation of the earlobe)¹⁾
When planning autograft using the indirect method, also check sensation at the sural nerve harvest site (posterior lower leg, lateral foot)¹⁾

Other Tests

Anterior segment OCT: Evaluate corneal epithelial thickness and limbal stem cell status (Vogt’s palisades)²⁾
Schirmer test: Measures basal tear secretion. Used to evaluate dry eye associated with NK.

Q How is corneal sensitivity accurately measured?

The Cochet-Bonnet esthesiometer (CBA) is the standard quantitative test. A 0.12 mm diameter nylon filament is applied perpendicularly to the cornea starting at 60 mm (maximum length, minimum pressure) and shortened in 10 mm increments until sensation is perceived. Shorter filament length indicates higher pressure and more severe hypoesthesia. Less than 20 mm indicates low sensitivity, and 50 mm or more indicates high sensitivity.

5. Standard Treatment

Conservative Treatment (Conventional Therapy)

Treatment of NK is based on a stepwise approach according to the Mackie stage.

Artificial tears and eye ointments: Frequent use of preservative-free formulations. Combination of oily eye ointment instillation and forced eyelid closure.
Punctal plugs: Inserted for severe tear deficiency.
Bandage contact lens (BCL): Protection of persistent epithelial defects.
Tarsorrhaphy: Useful for promoting epithelial defect healing in stages 2–3.
Amniotic membrane transplantation: Protective cover for refractory epithelial defects.

Pharmacotherapy (New Treatments)

Cenegermin (Oxervate™): Recombinant human nerve growth factor (rhNGF) eye drops. Indicated for NK stages 2–3. The REPARO trial showed a 72% rate of epithelial defect closure⁴⁾. However, significant improvement in corneal sensitivity has not been demonstrated.
Substance P + IGF-1 (FGLM-NH₂+SSSR): A minimal essential peptide combination that promotes corneal epithelial wound healing. Promising in early NK.
Autologous serum eye drops: Contain growth factors that promote epithelial healing. Mild improvement in corneal sensitivity has also been reported.

Corneal Neurotization

Patients who do not respond to conservative treatment or have persistent corneal anesthesia for more than one year are candidates for CN ¹⁾.

Direct Corneal Neurotization (DCN)

A method of transferring a donor nerve directly to the corneal limbus.

The contralateral or ipsilateral supratrochlear/supraorbital nerve is mobilized and its distal end is fixed to the corneoscleral junction ¹⁾
No graft is needed, so there are no donor site complications
Because the nerve is continuous, sensory recovery tends to be faster ¹⁾

Indirect Corneal Neurotization (ICN)

A method of bridging the donor nerve and the cornea using an interposition nerve graft. It is used in 63% of all cases ²⁾.

The sural nerve is the most common graft, with up to 30 cm available ¹⁾
Acellular allograft nerve grafts (e.g., Avance Nerve Graft) have comparable outcomes to autografts and avoid donor site complications ¹⁾²⁾
Can be performed through small incisions and is applicable to bilateral NK ⁴⁾

Item	Direct Method	Indirect Method
Speed of sensory recovery	Fast	Somewhat slow
Outcomes at 12 months	Equivalent	Equivalent
Secondary surgical site	None	Present

At 12 months, there is no significant difference in sensory recovery between the direct and indirect methods¹⁾²⁾.

Donor nerve selection

Donor nerve selection is based on nerve viability, axon count, and proximity to the cornea⁵⁾.

Nerve	Number of myelinated axons	Characteristics
Supraorbital nerve	Approximately 6,000	Close to cornea, first choice
Supratrochlear nerve	Approximately 2,500	Anatomical variation possible
Great auricular nerve	Approximately 6,530	Useful for bilateral NK

Terzis et al. proposed a cutoff of 900 myelinated axons required for effective nerve reconstruction ⁴⁾. The distal portions of the supraorbital and supratrochlear nerves each contain more than 2,000 myelinated axons, fully meeting this criterion ⁴⁾⁵⁾.

Nerve anastomosis and corneal fixation

End-to-end anastomosis: The donor nerve is completely transected, supplying all axons to the graft. This maximizes axonal load but causes numbness in the donor area ¹⁾
End-to-side anastomosis: A window is opened in the donor nerve sheath to partially divert axons. Donor sensation can be preserved ¹⁾
The nerve is divided into fascicles (usually 3–4) and placed in each quadrant of the cornea ¹⁾
Fixation into a corneoscleral tunnel is recommended. Compared to subconjunctival placement under the limbus, it shows better sensory recovery at 3 months postoperatively ¹⁾²⁾

Postoperative Management

Protect the cornea with a bandage contact lens and temporary tarsorrhaphy¹⁾
Provide aggressive lubrication with preservative-free artificial tears and lubricating eye ointment¹⁾
Use antibiotic eye drops for the first week postoperatively. Extend if epithelial defect persists¹⁾
There are attempts to start cenegermin adjunctively from one month postoperatively¹⁾

Q Which is better, direct or indirect corneal neurotization?

Comparative studies show no significant difference in corneal sensation recovery at 12 months. Direct method does not require a graft and sensation recovery is slightly faster (CB values improve statistically significantly at 3–6 months), but indirect method can be performed through small incisions and is applicable to bilateral NK. The choice of technique depends on donor nerve availability, distance to the cornea, and surgeon experience.

6. Pathophysiology and Detailed Mechanisms

Corneal Innervation

The cornea is one of the most densely innervated tissues in the body, with approximately 7,000 nociceptors per mm². Its sensitivity is 40 times that of dental pulp and 400–600 times that of skin.

Innervation reaches the cornea via the first branch of the trigeminal nerve (ophthalmic nerve) → nasociliary nerve → long ciliary nerves. The nerves penetrate Bowman’s layer and form the subepithelial nerve plexus. 70% of receptors are polymodal (responding to chemical, thermal, and mechanical stimuli), 20% are mechanoreceptors, and 10% are cold receptors.

Role of Neurotrophic Factors

Corneal nerves contain and release neurotransmitters such as substance P, CGRP, neuropeptide Y, and VIP. These regulate epithelial cell proliferation, migration, and regeneration, and are essential for maintaining corneal homeostasis.

Trophic factors such as NGF, BDNF, GDNF, NT-3, and NT-4/5 are also supplied by corneal nerves. Loss of these factors is the root cause of delayed epithelial healing and homeostatic disruption in NK.

Mechanism of Action of Corneal Neurotization

In CN, a healthy donor nerve is repositioned at the corneoscleral junction ¹⁾.

In the direct method, the distal end of the donor nerve is directly implanted into the cornea. In the indirect method, an interposition graft bridges the donor nerve and the cornea. In both cases, axonal fibers extend from the donor nerve into the corneal stroma and subepithelial layer, resulting in corneal reinnervation ¹⁾.

Axonal regeneration follows the typical pattern of Wallerian degeneration ¹⁾. After initial axonal breakdown, macrophages remove degenerated myelin, and Schwann cells differentiate and migrate from the healthy proximal portion of the donor nerve to promote axonal regeneration. Upregulation of NGF and IL-1 plays an important role in promoting axonal sprouting ¹⁾.

In a rat model, CN increased corneal nerve density, but only a small proportion of regenerated axons through the graft reached the cornea. It has been suggested that the cornea may selectively allow the growth of only unmyelinated nerve fibers with a specific phenotype ¹⁾.

Time Course of Reinnervation

2–4 weeks postoperatively: Axonal regeneration begins toward the graft ⁴⁾
3–6 months: Reinnervation is observed near the limbus. Onset of subjective sensory recovery ¹⁾²⁾
6–12 months: Reinnervation extends to the basal layer of the central cornea. Sensation peaks ¹⁾²⁾
12–18 months: In some cases, IVCM findings become comparable to the contralateral healthy eye ¹⁾

The trophic function of corneal nerves tends to recover earlier than sensory function. Therefore, even if CB value improvement is poor at 12 months, there are cases where corneal nerves are confirmed by IVCM and epithelial defects have healed ¹⁾.

7. Recent Research and Future Perspectives

Evidence from Large-Scale Reviews

A 2025 scoping review (12 studies, 164 eyes) reported a success rate of CN of 60.7–100% (mean 90%). The mean preoperative CCS (CBA) of 2.7 mm significantly improved to 36.0 mm postoperatively (range: 21.1–49.7 mm). Corneal sensation appeared at a mean of 4.1 months and peaked at 12 months ²⁾.

The indirect method (sural nerve graft) was the most common, accounting for 63% of all cases. Acellular allografts showed comparable outcomes to autografts, with advantages of shorter surgical time and avoidance of donor site complications ²⁾.

Favorable Outcomes in Pediatric Cases

In a report by Catapano et al. on 19 eyes (mean age 12.5 years), 87% of patients achieved CB 40 mm, and 64% recovered full sensation. The group with fascicular insertion into the corneoscleral tunnel showed earlier sensory recovery ²⁾.

In an 11-year-old boy with Mackie stage 1, objective sensory improvement appeared 6 months after MICN, and central corneal sensation reached 20 mm at 21 months. Schirmer test improved from 15 mm to 30 mm, and visual acuity recovered from 6/18 to 6/5 ⁶⁾.

Elucidation of Poor Prognostic Factors

In two cases of advanced age (62 and 70 years) and long-term denervation (23 years), sustained sensory improvement was not achieved after MICN. Proximal trigeminal nerve injury (e.g., after acoustic neuroma surgery) may have low “functional reserve” and lead to poorer prognosis compared to peripheral causes (e.g., herpetic keratitis) ⁷⁾.

These findings suggest that advanced age, long-term denervation, and proximal trigeminal nerve injury are poor prognostic factors for CN and should be considered in patient selection and counseling ⁷⁾.

Future Challenges

Optimization of direct and indirect methods through large-scale randomized controlled trials (RCTs) ²⁾
Standardization of patient-reported outcomes (PROs) ²⁾
Further accumulation of evidence for acellular allografts ¹⁾
Validation of the efficacy of cenegermin as postoperative adjuvant therapy ¹⁾
Noninvasive monitoring of corneal nerve fibers using anterior segment OCT ²⁾

Q What is the success rate of corneal nerve reconstruction surgery?

A 2025 scoping review (164 eyes) reported success rates ranging from 60.7% to 100% (mean 90%). Preoperative corneal sensation (CBA) improved significantly from an average of 2.7 mm to 36.0 mm postoperatively, and at 12 months, over 60% of children achieved normal sensation. However, outcomes may be poorer in elderly patients and those with long-term denervation.

8. References

Hubschman S, Rosenblatt MI, Cortina MS. Corneal neurotization for the treatment of neurotrophic keratopathy. Curr Opin Ophthalmol 2025;36:294-301.

Samoilă O, Samoilă L, Petrescu L. Corneal Neurotization, Recent Progress, and Future Perspectives. Biomedicines 2025;13:961.

Dragnea DC, Krolo I, Koppen C, et al. Corneal Neurotization-Indications, Surgical Techniques and Outcomes. J Clin Med 2023;12:2214.

Saini M, Jain A, Vanathi M, et al. Current perspectives and concerns in corneal neurotization. Indian J Ophthalmol 2024;72:1404-11.

Serra PL, Giannaccare G, Cuccu A, et al. Insights on the Choice and Preparation of the Donor Nerve in Corneal Neurotization for Neurotrophic Keratopathy: A Narrative Review. J Clin Med 2024;13:2268.

Lee BWH, Khan MA, Ngo QD, et al. Minimally invasive, indirect corneal neurotization using an ipsilateral sural nerve graft for early neurotrophic keratopathy. Am J Ophthalmol Case Rep 2022;27:101585.

Ting DSJ, Pradhan SP, Barnes E, et al. Minimally invasive corneal neurotization for neurotrophic keratopathy: The potential effect of age, denervation chronicity and lesion location. Am J Ophthalmol Case Rep 2023;29:101804.

Related keywords