Concurrent Ocular Pain in Patients with Neurotrophic Keratopathy

Abstract

PURPOSE:

To illustrate that ocular pain may occur in patients with neurotrophic keratopathy (NK) that typically are thought to lack symptoms of discomfort. A subset of these patients may also present with neuropathic corneal pain (NCP).

METHOD:

Retrospective case series of 7 stage 1 NK patients who presented with concurrent ocular pain, as confirmed by clinical examination, proparacaine challenge test and in vivo corneal confocal microscopy (IVCM). Records were assessed for results of ocular surface disease index (OSDI), pain on visual analog scale (VAS), ocular pain assessment survey (OPAS), best-corrected visual acuity (BCVA), corneal fluorescein staining (CFS) score, and IVCM findings. IVCM findings were compared to that of 20 healthy reference controls.

RESULTS:

Mean age of patients was 63.7±11.6 (range 44–76) years and 56.9±8.6 (range 42–74) years in reference controls (p=0.11). At presentation, ocular discomfort was 8.0±1.3 (range 7–10) on VAS and mean OSDI scores were 72.26±6.81 (range 62.50–79.54). Mean BCVA was 20/40, and mean CFS scores were 3.43±0.79 (range 2–4) on the Oxford scale. IVCM analysis showed significant decrease in mean total, main and branch nerve densities in ranges consistent with NK as compared to normal controls (p<0.001 for all), increased dendritiform cell density in three patients (p<0.001), and the presence of micro-neuromas in six of the patients.

CONCLUSION:

Patients with neurotrophic keratopathy are thought to present with hypoesthesia. However, nerve damage and inflammation, which play a role in the development of NK may result in the development of chronic ocular pain, such as NCP, resulting in potential under diagnosis of either disease.

INTRODUCTION

The cornea is highly innervated by the ciliary nerves, which originate from the ophthalmic branch of the trigeminal nerve.¹ Corneal nerves enter the cornea radially and migrate anteriorly to perforate the Bowman’s layer and comprise the subbasal plexus.¹ Corneal nerves and their function can be affected by a variety of ocular and systemic conditions that may result in neurotrophic keratopathy (NK).² Neurotrophic keratopathy is a rare neurodegenerative disease of the cornea caused by trigeminal nerve damage, subsequent breakdown of the corneal epithelium and poor healing, which presents with corneal sensory loss and decreased tear production and blink rate due to reduction of trigeminal reflexes.^3,4 Common causes of NK are corneal herpetic keratitis, chronic use of preserved eye drops, thermal and chemical burns, chronic or improper use of contact lenses, diabetes, and cranial neurosurgery, among others.² The diagnosis, and treatment of NK are based on disease severity, which has been broadly classified into 3 stages.⁵

Neuropathic corneal pain (NCP) is an ill-defined entity, which is currently diagnosed by clinical history, sign/symptoms disparity, ophthalmological evaluation, and the presence of nerve injury shown by in vivo confocal microscopy (IVCM).^6–8 Diagnosing NCP is difficult due to the lack of understanding of this disease, as well as minimal or absent clinical signs on examination. These difficulties are further confounded when NCP concurrently exists with other conditions, as NCP may be masked in such situations.^6,9–13 Regardless of whether NCP occurs in isolation or concurrently with other ocular diseases, evidence-based clinical recommendations are needed to manage these complicated patients.^6,12,14 The underlying causes of NCP may be systemic diseases (diabetes or small fiber neuropathy) or ocular diseases such as herpes simplex keratitis, herpes zoster ophthalmicus, ocular surgeries, chronic contact lens wear, radiation keratopathy, or chronic conditions, including dry eye disease (DED).^7,15–18 In both conditions, confirmation of diagnosis requires assessment of corneal nerve morphology and/or function, that can be performed by IVCM that is a non-invasive and high-resolution imaging modality allowing real-time visualization of corneal nerves, and by corneal esthesiometry, such as with a Cochet-Bonnet esthesiometer.^1,19,20 Microneuromas are may be seen in IVCM in NCP patients which were defined as abnormalities of the nerves and may present as irregularly shaped and the enlargements of terminal subbasal nerve ending(s). Microneruomas may have variable hyperreflectivity and to distinguish them from terminal nerve beading, all layers of the cornea are generally screened with around 300 images. Recent publications showed that presence of the microneuromas is an objective biomarker for diagnosis of NCP.²¹

NCP can be further divided into peripheral (nociceptor sensitization), central (alterations in central pain processing pathways), and mixed types. Determining the origin of pain is a crucial step for selecting the appropriate treatment, as the peripheral type may be treated with topical or non-central nervous system (CNS) penetrant drugs, whereas the central type requires the use of CNS penetrant drugs. The different types may be distinguished by instillation of anesthetic drops, which abolish peripheral pain without having any effect on non-ocular pain, such as from central sensitization. Thus, following instillation of anesthetic drops, patients experiencing complete relief suffer from peripheral NCP. Those with no relief suffer from centralized NCP, and those with residual pain in proparacaine challenge test have a mixed phenotype with both peripheral and centralized NCP.⁷

The current case series demonstrates that, ocular pain, such as NCP, may also occur concurrently in patients with stage I NK in addition to other causes of chronic ocular painful conditions, and that these conditions are not mutually exclusive.

METHODS

This is a retrospective case series of 7 stage 1 NK patients, seen at the Cornea Service of the New England Eye Center, Tufts Medical Center, Boston, MA, between December 2015 and March 2020. Classification of NK is based on ocular surface findings and corneal sensation. MacKie et al. classified stage I NK with reduced tear production, decreased epithelial turnover and decreased tear film thickness, presenting with punctate epithelial erosions, resulting in superficial punctate keratopathy and reduced corneal sensation. More recently Dua et al. proposed an adaptation for this classification with additional clinically relevant information, with the presence of superficial punctate keratopathy, tear film instability, and reduced or absent corneal sensation in one or more quadrants of cornea.²² All our cases are considered stage 1 NK based on these classifications. We also evaluated the patients to identify the underlying cause of the pain and NCP was diagnosed in a subset of our patients. For NCP diagnosis, we use applied our previously published approach.⁶ Patients were diagnosed with NCP when symptoms were out of proportion to signs, patients’ pain did not respond to extensive prior therapies, symptoms did no resolve with anesthetic drops after 90 seconds, and patients demonstrated the presence of microneuromas on IVCM. In cases with complete resolution of pain to anesthetic drops, lack of pain response to extensive prior therapies, symptoms out of proportion to signs, and presence if microneuromas were used for the diagnosis of NCP.²¹

Ocular surface and in vivo confocal microscopy (IVCM) findings of the patients were compared to that of 20 healthy reference controls. The study was approved by the Institutional Board Review of Tufts Medical Center/Tufts University Health Sciences. The protocol conformed to the Declaration of Helsinki and adhered to the Health Insurance Portability and Accountability Act (HIPAA).

Data extracted from the medical records included patient demographics, medical history, ocular history, previous treatment, best-corrected visual acuity (BCVA), clinical findings, including vital dye staining by Oxford Scale, tear metrics, and symptom questionnaires, including ocular surface disease Index (OSDI),²³ as well as the ocular pain assessment survey (OPAS).²⁴ Moreover, proparacaine challenge test (PCT) results from the same visit were recorded. For the PCT, patients were asked to report their pain relief based on visual analogue scale after 90 seconds of installation of 0.5% proparacaine hydrochloride eye drops (Alcaine; Novartis Ophthalmics, East Hanover, NJ).⁶

Laser IVCM (Heidelberg Retina Tomograph 3 with the Rostock Cornea Module, Heidelberg Engineering GmbH, Heidelberg, Germany) was conducted routinely on all patients presenting with pain and discomfort for confirmation of corneal subbasal nerve alterations, as previously described.²⁵ IVCM was obtained from all patients and compared to age- and sex-matched reference controls from a prospectively enrolled normative database. Images were obtained with the help of a 63× objective immersion lens with a numerical aperture of 0.9 (Olympus, Tokyo, Japan). This microscope uses a 670-nm red wavelength diode laser source producing images representing a coronal section of the cornea of 400 × 400 μm (horizontal × vertical). Digital images were recorded at of 30 frames/s. Adjacent images are separated by 1 μm, with a lateral resolution of 1 μm/pixel. To perform this procedure, both eyes were topically anesthetized using 0.5% proparacaine hydrochloride (Alcaine; Novartis Ophthalmics). This was followed by administration of a drop of hydroxypropyl methylcellulose 2.5% (GenTeal gel, Alcon, Fort Worth, TX) to improve the optical coupling with the cornea module of the microscope. The cornea module was mounted with a disposable, sterile polymethylmethacrylate cap (Tomo-Cap; Heidelberg Engineering GmbH), filled with a layer of hydroxypropyl methylcellulose 2.5% (GenTeal gel; Alcon), gel was also applied to the surface of the cap. The equipment is manually advanced until the gel on the cap comes in contact with the surface of the central cornea.²⁵ Fifty to hundred images of the corneal subbasal layer were obtained via six to eight sequence, and a masked observer (L.Y.S) selected the three most representative images determined as best focused, single layer, minimum folds and good contrast of the subbasal nerve plexus. Two masked observers (L.Y.S, B.N.B.) analyzed IVCM images for morphology and density of subbasal nerve plexus (SNP) as well as dendritiform cells (DCs).

A priori power analysis was conducted using data from a previous study in NK patients²⁶ to estimate a sufficient sample size to achieve a power of 80% and alpha error of 0.05. Post hoc power analysis was conducted after the study was completed to determine the power of the study, confirming the appropriate sample size for the study with a power of more than 98% for statistically significant variables. Statistical analyses were performed with SPSS software version 22.0 (SPSS Inc., IBM, Chicago, IL, USA). Distribution of data was analyzed by Kolmogorov-Smirnov test. Independent sample t test and Chi-squared were used to assess the differences in demographic and clinical parameters as well as IVCM parameters of comparison of the patients and healthy reference controls. P values lower than 0.05 were considered statistically significant.

CASE PRESENTATIONS

Demographic features of the patients and healthy age-sex matched reference controls are listed in Table1. The ocular and past medical history, medications, and clinical features of the patients are summarized in Table 2. Pertinent summary results of the OPAS are summarized in Table 3. IVCM parameters of the patients and healthy reference age- and sex-matched controls are listed in Table 4.

Table 1:

The summary table for demographic features and clinical findings of the cases and age- and sex-matched reference controls

Parameter	Cases (n=7) (mean ± SD)	Controls (n=20) (mean ± SD)	p value
Age (years)	63.7±11.6 (44–76)	56.9±8.6 (42–74)	p=0.11
Gender (female/male)	4/3	11/9	p=0.92
Schirmer’s score II (mm)	7.00±4.08	>15.00	p<0.001
Corneal fluorescein staining (0–5 Oxford scale)	3.43±0.79	0.00	p<0.001
OSDI	72.26±6.81	<5.00	p<0.001

SD: Standard deviation, Sec: Second, mm: millimeter, OSDI: Ocular surface disease index

Table 2:

Clinical features of the patients with neurotrophic keratopathy

	Age	Sex	Ocular History	Non-Ocular History	BCVA	Central CFS	Sensation	Proparacaine Challenge (%relief)	Ocular Medication	Systemic Medication
CASE 1	76	F	DED, Cataract Surgery Chronic Blepharitis	Scleroderma, Reynaud’s Disease, Colon CA, Lung Transplantation, HTN	20/70	4	Central 2 cm with CBE	100%	AT	Amlodipine, Ondansetron, Prilocaine, Tacrolimus. Silver Sulfadiazine, Furosemide, Ropinirole
CASE 2	75	F	Herpes Simplex Keratitis, LASIK Surgery, DED, Primary Open angle Glaucoma	Migraine, Depression, Hypothyroidism, Colon Cancer	20/25	3	Central 1.5 cm with CBE	100%	AT, FSO, timolol maleate ophthalmic solution 0.5%, Latanoprost Ophthalmic Solution 0.05%,	Levothyroxine, Omeprazole, Ipratropium bromide, Rosuvastatin, Triazolam, Triamcinolone, Bupropion
CASE 3	70	M	DED	Sjoegren’s Disease, Small Fiber Neuropathy Rheumatoid Arthritis, Depressive Disorders Cardiac Arrythmia, Chronic Kidney Disease, Gout Disease, HTN	20/30	4	N/A	14%	AT, AST, Fish Oil	Aspirin, Allopurinol, Escitalopram, Vitamin D, amlodipine besylate, Gabapentin
CASE 4	66	M	Lower Eye Lid SCC, cataract surgery, lower eye, DED	MOHS surgery, DM, Idiopathic Pulmonary Fibrosis, Lung Transplantation, HTN, HTN, Chronic Renal Insufficiency, CHF	20/25	3	Central 0 cm with CBE	57.1%	AT, Lifitegrast Ophthalmic Solution 5%, Cyclosporine Ophthalmic Solution 0.05%	Alfuzosin, Amlodipine, Carvedilol, Sirolimus/Rapamune, Azithromycin, Warfarin/coumadin, B12 Biotin
CASE 5	60	F	DED, Cataract Surgery, YAG Capsulotomy	Sjoegren’s Disease, Colon CA, Cardiac Stent and Valve Replacement, Hysterectomy	20/70	4	N/A	71.4%	AT, AST, Cyclosporine Ophthalmic Solution 0.05%, Loteprednol Etabonate Ophthalmic Solution 0.5%	Trazadone, Gabapentin, Lamotrigine, Lorazepam
CASE 6	55	M	Herpetic keratitis, Fuchs’ Dystrophy, DED	Crohn Disease, Total Thyroidectomy, tension headache	20/50	3	N/A	80%	AT	Vitamin D, Iron, Mesalamine
CASE 7	44	F	DED, Penetrating Eye Trauma, Aphakia, Multiple Seconder IOL Implantation	Anxiety	HM	4	Central less than 1mm with CBE	62.50%	AT, Doxycycline Monohydrate 50mg, Atropine sulfate,	Alprazolam, Topiramate

DED: Dry Eye Disease, SCC: Squamous Cell Carcinoma, HTN: Hypertension, AT: Artificial Tears, AST: Autologous Serum Tears, FSO: Flaxseed Oil, CBE: Cochet-Bonnet Esthesiometer IOL=Intra ocular Lens HM: Hand Motion

Table 3:

Ocular Discomfort and symptom questionnaire results of the patients

	OSDI score	Pain Intensity for the last 24 hours	Pain Intensity for the last 2 weeks	QoL Score (mean ± SD(Range))	How much pain increased when exposed to wind, dry air, heat, air conditioni ng (%)	How often eye pain accompanied by redness (%)	How often eye pain accompanied by burning (%)	How often eye pain accompanied by sensitivity to light (%)
Case1	75.0	0	0	3±3.7	50	100	100	20
Case2	N/A	4	4	8.8±1.6	80	20	50	90
Case3	79.54	2	2	N/A	100	80	80	10
Case4	79.0	5	5	6.8±2.1	80	90	50	90
Case5	75.0	8	8	7±1.8	90	10	30	10
Case6	65.0	2	3	6.5±3.9	N/A	100	100	100
Case7	62.5	4	4	2.8±1.1	100	0	50	50

OSDI: Ocular surface disease index, QoL: Quality of life, SD: Standard deviation

Table 4.

In vivo confocal microscopy parameters of the affected eyes of patients with neurotrophic keratopathy and healthy reference age and sex matched controls

IVCM Parameters	Affected Eyes with NK (n=7) (mean ± SEM)	Controls (n=20) (mean ± SEM)	p value
Total Nerve Density (μm/mm2)	5,331.18±1,349.72	24,200.78±735.14	p<0.001
Main Trunk Nerve Density (μm/mm2)	2,752.84±813.22	10,753.59±338.20	p<0.001
Branch Nerve Density (μm/mm2)	2,577.15±569.50	13,447.19 ±602.31	p<0.001
Dendritiform cell density (cells/mm2)	65.47±29.96	20.5±3.6	p<0.05

NK: Neurotrophic Keratopathy, SEM: Standard error of mean

Case 1

A 76-year-old Caucasian female presented with a four-month history of burning, redness, light sensitivity and ocular discomfort in her left eye with the severity of 7/10 on the VAS. Her past ocular history was significant for long-standing DED, unresponsive to conventional therapy, bilateral cataract surgery three years ago, and 1 mm lagophthalmos with poor Bell’s reflex bilaterally. Her past medial history was significant for scleroderma, Reynaud’s syndrome, colon resection for colon cancer, and hypertension. Ocular treatments at presentation were warm compresses, preservative free artificial tears as needed, and cyclosporine 0.05% twice daily bilaterally, which did not help patient’s symptom relief (Table 2). The patient had an OSDI score of 75.0 (Table 3). On examination, her BCVA was 20/25 in the right eye 20/70 in the left eye, slit lamp examination showed diffuse +3 corneal fluorescein staining in OD, diffuse +4 corneal fluorescein staining in OS, including the central cornea (Fig. 1A–B). Her corneal sensation as measured by Cochet-Bonnet esthesiometer was centrally 3.0 cm OD, and 2.0 cm OS. Schirmer’s test with anesthesia was 1.0 mm OD, and 4.0 mm OS. The proparacaine challenge test resulted in decreased ocular discomfort from 7/10 intensity to 0/10 intensity after 90 seconds bilaterally. Further, IVCM demonstrated 77.0% loss for corneal nerve density as compared to healthy reference controls (Fig. 2A–B), with the mean total, main trunk, and branch nerve densities of 5,212.36±2,219.88 μm/mm², 1,931.7±1,446.30, and 3,272.31±801.04 respectively. In addition, IVCM showed presence of microneuromas (Fig. 3A). Moreover, IVCM showed a DC density of 6.8±3.0 cells/ mm². The loss of corneal sensation, long-standing history of DED, which was unresponsive to conventional therapy, severe central corneal fluorescein staining without epithelial defect, and severe nerve loss by IVCM supported the diagnosis of stage 1 NK. Additionally, the history of biopsy-proven small fiber neuropathy, ocular pain assessment scores, ocular discomfort unresponsive to conventional therapy, and presence of morphological nerve abnormalities by IVCM, supported chronic ocular pain diagnosis in this patient. This patient presented with stage I NK but demonstrated severe concurrent symptoms of burning and discomfort due to small fiber neuropathy.

Figure 1. Anterior Segment Photography with and without corneal fluorescein staining demonstrate the diagnosis of Stage 1 NK.

Anterior segment photography of Case 1 shows the absence of the corneal epithelial defect (A). Slit-lamp examination shows diffuse corneal fluorescein staining in OS in Case 1 (B). Conjunctival injection without epithelial defect of Case 6 (C). The slit lamp examination shows diffuse confluent corneal fluorescein staining of Case 6 (D) Anterior segment photography of Case 7 demonstrates the corneal vascularization and corneal scarring (E). The slit-lamp examination shows diffuse corneal fluorescein staining centrally and temporally in Case 7 (F).

Figure 2. Laser in vivo confocal microscopy images of corneal nerves in patients.

In vivo confocal microscopy image of healthy reference control (A). Severe decrease in nerve density in Case 1, as a result of chronic inflammation and ocular surgery (B). The decrease of total corneal nerve density and branching in Case 2 with herpes simplex keratitis (C). Severe decrease in nerve density in Case 3 with concurrent chronic systemic inflammatory disease (D). Decreased subbasal cornea nerve density and branching of Case 4 (E). Severe decrease in total nerve in Case 5 with history of ocular trauma and multiple intraocular surgeries, long standing dry eye disease and history of ocular surgery (F). Central corneal subbasal nerve plexus alterations in Case 6 with a presumed history of herpetic keratitis (G). Severe decrease in nerve density in Case 7, as a result of ocular trauma and multiple intraocular surgeries (H).

Figure 3. Laser in vivo confocal microscopy images of corneal nerves, demonstrating morphological nerve alterations in patients with symptomatic neurotrophic keratopathy.

Central corneal images of the Cases 1–5 demonstrating the presence of microneuromas (A–F).

Case 2

A 75-year-old Caucasian female presented with six-year history of left eye pain and light sensitivity with an intensity of 7–8/10 on VAS. Her past ocular history was significant for epithelial dendritic herpes simplex keratitis, laser in situ keratomileuses (LASIK) surgery OU, longstanding dry eye disease (DED), and primary open angle glaucoma. Her past medical history was significant for migraines, hypothyroidism, depression, and colon cancer (Table 2). On presentation, her ocular medications included artificial tears, latanoprost 0.05% 1x/day, timolol maleate 0.5% 2x/day and oral flaxseed oil 2000 mg/day which did not help symptom relief (Table 2). She reported a range of pain being 1–7/10 on OPAS (Table 3). On examination, her BCVA was 20/25 in the right eye and 20/25 in the left eye. Slit-lamp examination showed central +2 SPK in her right eye with mild haze between 3 and 9 o’clock, +3 focal central SPK with superficial inferior corneal vascularization, and a superior pannus between 9 and 2 o’clock in her left eye. Tear break-up time was 8.0 seconds OD, and 12.0 seconds OS. Schirmer’s test with anesthesia was 6.0 mm OD, and 9.0 mm OS. Her corneal sensation as measured by Cochet-Bonnet esthesiometry was 2.0 cm centrally in her right eye, 1.5 cm centrally in her left eye. The proparacaine challenge test resulted in a decreased pain intensity from 7/10 to 0/10 intensity within 90 seconds. IVCM of her left eye demonstrated 89.7 % decrease in corneal nerve density compared to healthy reference controls (Fig. 2C), with a mean total, main trunk, and branch nerve densities of 2,325.90±474.84 μm/mm², 1,177.51±598.49^, and 1,148.38±137.46 respectively. In addition, IVCM showed the presence of microneuromas (Fig. 3B). Further, DC density of DC’s was 2.4±0.7 cells/mm². The history of herpetic keratitis and LASIK surgery, the dense central corneal staining without epithelial defect, significantly decreased corneal sensation, and the profound decrease in corneal nerve density by IVCM supported the diagnosis of stage 1 NK. In addition, the protracted and severe pain, as well as the presence of morphological nerve abnormalities by IVCM support the diagnosis of chronic ocular pain, likely due to post-herpetic neuralgia. The patient presented with severe stage I NK and concurrent ocular pain and light sensitivity, likely due to either post-herpetic or post-LASIK neuralgia, or a combination of both.

Case 3

A 70-year-old Caucasian male was referred from the neurology clinic with a three-month history of ocular discomfort, burning sensation and light sensitivity. His past ocular history was significant for long standing DED secondary to Sjögren’s syndrome. His past medical history was significant for Sjögren’s syndrome, rheumatoid arthritis, small fiber neuropathy, gout, depressive disorder, and cardiac arrhythmia (Table 2). On presentation, his ocular medication included preservative free artificial tears 6x/day, autologous serum tears (AST) 20% 6x/day, and fish oil 2000 mg 1x/day (Table 2). In addition, he was on gabapentin 100 mg three times a day. The patient had previously failed multiple previous therapies, including lid hygiene, thermal pulsation device, cyclosporine drops, amniotic fluid drops, and lubrication which did not help patient’s symptoms relief either. His OSDI score was 77.8, and the patient rated his ocular discomfort as 7/10 on the VAS (Table 3). On examination, BCVA was 20/30 in the right eye and 20/20 in the left eye. Slit-lamp examination showed +4 diffuse central corneal fluorescein staining bilaterally. Schirmer’s test with anesthesia was 15 mm OD, and 14 mm OS. The proparacaine challenge test resulted in decreased discomfort from 7/10 to 6/10 intensity after 90 seconds. IVCM demonstrated 91.8% loss of corneal nerve density compared to healthy reference controls (Fig. 2D) with the mean total, main trunk, and branch nerve densities were 1,980.53±506.00 μm/mm², 815.85±815.85, and 1,164.67±735.26 respectively. In addition, IVCM demonstrated the presence of microneuromas (Fig. 3C). IVCM further showed a 5.5-fold increase in DC density compared to controls, with 131.0±0.93 cells/mm² (Fig. 4A, B), which is significantly higher than in control eyes. The long-standing Sjögren’s DED, lack of improvement with conventional ocular surface therapies, the dense central corneal fluorescein staining without epithelial defect, and profound nerve loss by IVCM support the diagnosis of stage 1 NK. Additionally, the significant ocular discomfort, not resolving with anesthetic drops, and the presence of corneal nerve abnormalities by IVCM support the concurrent diagnosis of NCP with a component of centralization. Residual pain after the proparacaine challenge test supported central NCP diagnosis in this patient. This patient presented with stage I NK due to longstanding Sjögren’s syndrome and concurrent small fiber neuropathy resulting in severe ocular discomfort.

Figure 4. Laser in vivo confocal microscopy images of healthy cornea and patients with increased dentritiform cell density.

In vivo confocal microscopy image of healthy reference control (A). Central corneal images of the Cases 3, 5, 7 demonstrating the increased dentritiform cell density (B, C, D respectively).

Case 4

A 66-year-old Caucasian male presented with an eight-months history of stabbing pain and light sensitivity in his right eye with the severity of 7/10 on VAS, which started progressively within one month after radiation therapy for squamous cell carcinoma of his right forehead and temple. His past ocular history was significant for cataract surgery OU, and right upper lid entropion repair, and stage II NK with a prior persistent epithelial defect that had healed with patching. His past medical history was significant for multiple skin cancers, involving multiple excisions and MOHS surgery for squamous cell carcinoma five years ago, type II diabetes, idiopathic pulmonary fibrosis requiring lung transplantation, chronic renal insufficiency, hypertension, and congestive heart failure (Table 2). At the time of initial presentation, his ocular medication included lifitegrast 5% twice a day, cyclosporine 0.05% twice per day, as well as preservative-free lubricating eye drops six times a day which did not provide complete symptom relief (Table 2). His symptom severity on OSDI was 79.54, and the range of pain severity on OPAS fluctuating between 1–9/10 during the day (Table 3). On examination, his best-corrected visual acuity (BCVA) was 20/25 in the right eye, and 20/20 in his left eye. Slit-lamp examination showed bilateral madarosis and polyosis, grade 2 meibomian gland dysfunction (MGD), +3 diffuse superficial punctate keratopathy (SPK) OD, involving the center, and +1 diffuse SPK OS. Schirmer’s test with anesthesia was 9.0 mm OD, and 20.0 mm OS. The proparacaine challenge test resulted in decreased discomfort from 7/10 to 3/10 intensity after 90 seconds. His corneal sensation as measured by Cochet-Bonnet esthesiometry was 0 cm centrally in the right eye and 4.5 cm in the left eye. IVCM of the right affected eye with NK demonstrated 74.9 % decrease in nerve density as compared to healthy reference controls, and the presence of microneuromas (Fig. 2E, Fig. 3D). The mean total, main trunk, and branch nerve densities were 6,063.58±547.28 μm/mm², 3,119.26±173.10, and 2,944.32±626.62 in his right eye respectively (Table 4). IVCM further showed DC density of 4.5±1.9 cells/mm². Thus, the patient history, the decreased central corneal sensation, the dense and diffuse CFS involving the central cornea, as well as the severe loss of cornea nerves by IVCM support the diagnosis of stage 1 NK. Further, the severe pain and light sensitivity unresponsive to conventional ocular surface therapy, and the abnormal nerve morphology by IVCM support the concurrent diagnosis of NCP.^6,19,27 Residual pain after the proparacaine challenge test supported central NCP diagnosis in this patient. This patient with stage I NK, significant corneal nerve density and complete loss of central corneal sensation after periocular radiation, presented with both concurrent recalcitrant pain and light sensitivity.

Case 5

A 60-year-old Caucasian female was referred with a five-month history of stabbing pain and severe light sensitivity in her left eye with the intensity 7–8/10 on VAS. Her ocular history was significant for severe DED secondary to uncontrolled Sjoegren’s disease, recurrent corneal ulcers due to long-standing NK, relapsing stage II NK that was persistent to medication and only recently healed with patching, concurrent DED unresponsive to conventional treatment, bilateral cataract surgery 1.5 years ago, and YAG laser capsulotomy one year ago. Her past medical history was significant for uncontrolled Sjoegren’s disease, headaches, anxiety, depressive disorder, colon cancer, hearing loss, cardiac valve replacement and hysterectomy (Table 2). On presentation, her ocular medications included AST 20% 4x/day, cyclosporine 0.05% 2x/day, loteprednol etabonate 0.5% eye drop 4x/day and preservative-free lubricating eye drops 6x/day in her left eye but patient did not get complete symptom relief (Table 2). Her OSDI score was 75.0, while on OPAS she reported a range of pain of 2–8/10 (Table 3). On examination, her BCVA was 20/20 in the right eye and 20/70 in the left eye. Slit-lamp examination showed +1 corneal fluorescein staining OD, corneal fluorescein +4 diffuse staining, including the central cornea OS. Her corneal sensation measured by Cochet-Bonnet esthesiometer was centrally 3.5 cm OD, and 1.5 cm OS. Schirmer’s test with anesthesia was 1.0 mm OD, and 4.0 mm OS. The proparacaine challenge test resulted in decreased discomfort from 7/10 to 2/10 intensity after 90 seconds. IVCM demonstrated of her left affected eye demonstrated a decrease in corneal nerve density by 62.3% as compared to healthy reference controls (Fig. 2F) with the mean total, main trunk, and branch nerve densities of 9,111.83±2,607.72 μm/mm², 5,510.71±938.18, and 3,601.13±1,823.67 respectively. In addition, her IVCM showed the presence of microneuromas in her left eye (Fig. 3E). IVCM also showed a 8.1-fold increase of DC density compared to controls, with 182.5±4.8 cells/mm² (Fig. 4C). Based on the history of severe long-standing DED secondary to uncontrolled Sjoegren’s disease, which resulted NK with recurrent epithelial breakdown and neurotrophic ulcers in her left eye, the decrease in corneal sensation, and IVCM findings with the severe loss of nerve density supported the stage 1/2 NK. Further, the severe left eye pain and the light sensitivity, unresponsive to conventional therapy and presence of microneuromas in her left eye by IVCM supported concurrent NCP. Residual pain after the proparacaine challenge test supported central NCP diagnosis in this patient. The findings in this patient demonstrate that stage I/II NK is not always pain free and can present with severe stabbing pain, despite loss of corneal sensation and loss of corneal nerves.

Case 6

A 55-year-old Caucasian male presented with a one-year history of blurry vision, ocular discomfort, burning sensation and severe light sensitivity with a severity of 10/10 on VAS being mostly in his left eye. His past ocular history was significant for herpetic keratitis in his left eye, neurotrophic keratopathy, Fuchs’ endothelial corneal dystrophy, long-standing DED, and chronic blepharitis. His past medical history was significant for dry mouth, tension headache, Crohn’s disease, and total thyroidectomy (Table 2). The ocular medications at presentation were preservative-free lubricating eye drops 6x/day bilaterally. On presentation, the patient reported his symptom severity with on OSDI with a score of 65.0. In addition to severe burning, he had a pain severity of 2/10 (range 0–4) on OPAS (Table 3). On examination, his BCVA was 20/20 in the right eye and 20/50 in the left eye. Slit-lamp examination with corneal fluorescein staining showed +2 SPK inferiorly and temporally OD, and confluent diffuse +3 SPK, including the corneal center OS, as well as corneal guttata OU (Fig. 1C–D). His corneal sensation measured by Cochet-Bonnet esthesiometer was 5.5 cm centrally OD, and 1.0 cm OS. Schirmer’s test with anesthesia was 4.0 mm OD, and 4.0 mm OS. The proparacaine challenge test demonstrated decreased ocular discomfort from 10/10 intensity to 2/10 intensity after 90 seconds. IVCM demonstrated of his left affected eye showed a 55.5% decrease in corneal nerve density compared to healthy reference controls (Fig. 2G) mean total, main and branch nerve densities were 10,755.36±2,459.45μm/mm², 5,847.05±832.46^, and 4,908.31±2,204.50 respectively. In addition, left eye IVCM demonstrated the presence of microneuromas (Fig. 3F) and DC density of 0.6±0.4 cells/mm². The long-standing history of NK due to herpetic keratitis OS, unresponsive to conventional therapy, confluent diffuse corneal fluorescein staining including the corneal center, loss of corneal sensation and severe nerve loss by IVCM supported the diagnosis of stage 1 NK. Additionally, severe light sensitivity/ocular discomfort, not completely resolving with anesthetic drops, and the presence morphological nerve abnormalities by IVCM supported the diagnosis of NCP. Residual pain after the proparacaine challenge test supported central NCP diagnosis in this patient. This case demonstrates that NK can present with severe neuropathic symptoms of burning and light sensitivity and does not have to present without symptoms of ocular discomfort.

Case 7

A 44-year-old Caucasian female was referred for long-standing severe left eye pain (intensity of 8/10 on VAS), which started progressively after recurrent ocular surgeries (x7) related to penetrating trauma and intraocular lens dislocation 25 years ago. Her past ocular history was significant for several ocular surgeries in her left eye, including lens extraction, secondary intraocular lens implantation with iris fixation, and intraocular lens exchange due to a dislocated intraocular lens, and pars plana vitrectomy following the initial trauma. Her past medical history was otherwise not significant. (Table 2). Her ocular medication included preservative free artificial tears 6x/day, atropine sulfate 1% 1x/day OS for her left eye, and doxycycline monohydrate 50 mg 2x/day but patient did not get complete symptom relief (Table 2). On presentation, the patient reported the severity of symptoms with an OSDI score of 62.5, and a mean pain severity on OPAS for last 24 hours 4/10 (range 3–8) and 4/10 (range 4–8) for past two weeks. On examination, her BCVA was 20/30 in right eye, and hand motions in left eye. Slit-lamp examination showed +1 inferior SPK OD, and a superior corneal scar from 12–4 o’clock, as well as diffuse 4+ corneal staining centrally and temporally in OS (Fig. 1E–F). Corneal sensation measured with Cochet-Bonnet esthesiometry was 6.0 cm centrally and was OD less than 1.0 cm centrally OS. Schirmer’s test with anesthesia was 3 mm OD and 7 mm OS. The proparacaine challenge test resulted in decreased pain from 8/10 intensity to 3/10 intensity after 90 seconds. IVCM of her left affected eye demonstrated 92.2% decrease in corneal nerve density compared to healthy reference controls, severe decrease in nerve density with the mean total, main trunk, and branch nerve densities of 1,868.68±1,102.03 μm/mm², 867.75±867.75, 1,000.94±527.63 respectively (Fig. 2F). IVCM further showed a 5.5-fold increase in DC density compared to controls, with 130.4±0.6 cells/mm² (Fig. 4D). The history of multiple ocular surgeries after penetrating trauma, the severe decrease in central corneal sensation, diffuse central corneal staining, and severe nerve loss by IVCM supported the diagnosis of stage 1 NK. In addition, the presence of severe, long-standing and refractory pain that did not completely resolve with anesthetic drops, demonstrates concurrent presence of NCP. Residual pain after the proparacaine challenge test supported central NCP diagnosis in this patient. This case demonstrates that a history of trauma and multiple ocular surgeries can result in the concurrent presence of NCP in the setting of stage I NK.

DISCUSSION

Herein we present a case series of seven patients with stage I NK who concurrently suffer from concomitant ocular pain and discomfort. While NK is typically thought of as a disease with decreased corneal sensation and lack of symptoms of discomfort,^22,28 the current case series highlights that the presence of stage I NK and chronic ocular pain association. Additionally, NCP was also demonstrated in subset of these patients,^{6,8,10,14,29,30} which supports that they are not mutually exclusive. The purpose of the current paper is thus to demonstrate that even patient with features of NK (decreased corneal sensation, severe epithelial damage, lack of response to conventional therapies) may have concurrent NCP (from neuronal sensitization of remaining nerves from inflammation) especially in patients with concurrent chronic ocular pain. The current series highlights that patients may have abnormal mechanical corneal nerve function (low sensation by esthesiometry) and decreased corneal nerves (as shown on IVCM), but that if the remaining nerves are sensitized at one point (typically from inflammation during the course of the disease), these remaining nerves may be overly sensitive, resulting in hyperalgesia and allodynia (features of NCP), and thus present with both features of NK and NCP. Considering these two diseases as being mutually exclusive, may thus result in underdiagnosis of either condition, and subsequently may result in increased morbidity.

Corneal nerves are essential for corneal homeostasis and function, such as for corneal sensation, epithelial integrity and functioning, epithelial proliferation and wound healing.^31–33 Corneal nerve injury may result due to both ocular and systemic diseases, including infectious keratitis,^25,34,35 herpetic keratitis,^36,37 DED,³⁸ ocular surgeries,³⁹ diabetes,^40,41 small fiber polyneuropathy,⁴² as well as cranial ischemia or stroke and intracranial lesions, including trigeminal nerves,^22,43,44 leading to NK. In our case series, herpetic keratitis, long-standing dry eye disease and ocular surgeries were the likely underlying causes of NK.

NK generally presents with absence of symptoms of discomfort, as the disease results in corneal hypoalgesia or anesthesia.²² The presentation of NK may, however, differ based on the underlying condition resulting in NK, co-morbidities, or co-existing inflammation.^2,4 Inflammation may result in additional nerve loss and sensitization of peripheral corneal nerves through pro-inflammatory cytokines that can alter peripheral pain signaling pathways.^6,8 Persistent presence of peripheral sensitization may then result in altered threshold changes in the central nervous system neurons that result in their over-responsiveness and increased pain perception (central sensitization).⁴⁵ Central sensitization was likely present in five of our cases without complete relief to anesthetic drops. The overlap of the underlying causes of NK and NCP could thus result in concomitant presence of both conditions in patients.

In contrast to nociceptive pain, the International Association for the Study of Pain defined neuropathic pain as being caused by a lesion or disease of the somatosensory pathways in the peripheral and/or central nervous system.⁴⁶ NCP can present with corneal pain (allodynia), light sensitivity (photoallodynia), or hyperalgesia.^14,47 Similar to NK, the underlying causes can include diabetes,⁴⁸ long-standing dry eye disease,²⁹ degenerative or ischemic diseases,^12,14 systemic autonomic diseases,^12,14 intracranial tumors or infarcts,⁴⁹ post-herpetic neuralgia,^12,14,16 and chronic inflammatory conditions, among others.⁵⁰ In our cases, autoimmune diseases such as dysimmune small fiber neuropathy, Sjögren’s syndrome, and Crohn’s disease can result in chronic inflammation and nerve damage, which may play a role in the progression of both NK and result in the development of concurrent NCP.³⁹ In our study, there are two patients with chronic ocular pain who had complete relief after proparacaine challenge test. The possible underlying cause of the ocular pain may be inflammatory, nociceptive or peripheric neuropathic corneal.^51–53

Despite the presence of severe decrease in mechanical sensation and loss of corneal nerves by IVCM, patients presented with variable symptoms of discomfort, including severe burning, severe pain, hyperalgesia, allodynia, and light sensitivity, or a combination thereof. The OPAS questionnaire also assesses light sensitivity, burning and tearing sensation for the evaluation of the quality of life of the pain patients.²⁴ The severity of pain levels generally affects quality of life negatively.²⁴ Quality of life scores are calculated in the OPAS, by assessing a group of questions, including daily activities, such as reading and/or computer use, driving and/or watching TV, general activities like walking, doing house chores, mood, sleep, enjoying life/relations with other people. Herein, the mean quality of life score was 5.83±2.40, and 71.4% of cases had severe enough symptom to have moderate impact on daily activities such as reading, computer use, driving, watching TV, walking, doing house chores. Additionally, the patients had accompanying symptoms, with 57.1% redness, 65.7% of burning, of 52.8% light sensitivity.

Similarly, neuropathic pain in diabetes or small fiber neuropathy has various presentations; presence of both negative sensory symptoms (numbness or reduced sensation) or positive sensory symptoms (pinprick sensation, pressure sensation, pain sensation after completion of stimulus).⁵⁴ In addition, neuropathic pain may also induce various degrees of physical disorders, such as depression, anxiety, sleep disorders, and a more reduced quality of life than patients with painless-DPN, particularly with moderate to severe neuropathic pain.⁵⁵ Similarly, four of our cases reported depression, anxiety or tension headache in their presentations. The clinical presentation of small fiber neuropathy is also characterized by the by autonomic dysfunction.^54,56

IVCM allows for quasi-histopathological examination of the subbasal corneal nerves in health and disease.²² The average decrease of subbasal corneal nerves measured by IVCM as compared to healthy references controls for our patients was 77.7%. Three of the patients had severe nerve loss with over 89% decrease in subbasal nerves and one case demonstrated near absent nerves by IVCM, consistent with NK. Further, IVCM analysis for DCs demonstrate that DC density was increased in 3 patients compared to healthy reference controls, suggesting a possible role for inflammation in these patients.

Additionally, following nerve injury and Wallerian degeneration of the distal segment “microneuromas” can be seen in patients with NCP due to aberrant nerve regeneration.^6,7 Six of the patients in our case series showed concurrent presence of severe decreased of corneal nerves and the presence of microneuromas, corresponding with severe symptoms of discomfort, suggestive of concomitant presence of NK and NCP. The last case showed near absent nerves. Given the variable underlying etiologies, with some of these patients having bilateral NK due to systemic conditions, and others having unilateral NK, and the fact that we have previously published extensively on the contralateral effect of corneal nerves,^26,57,58 contralateral eyes were not assessed for this study.

Given the complex nature of neurotrophic keratopathy and neuropathic corneal pain, there is likely no single mechanism by which NK and NCP may concomitantly be present. It has been demonstrated in animal models of nerve injury, such as axotomy, that not all nerve fibers may be capable of regeneration. Thus, within these models there exist three types of nerve fibers: axotomized-regenerating, axotomized-non-regenerating, and axotomy-spared.⁵⁹ While it may seem counterintuitive that such significant nerve loss can lead to apparent hyperalgesia, this is also seen in the setting of phantom limb pain (PLP).⁶⁰ Such patients often report dysesthesias and pain in the region of the lost limb. Such patients are thought to experience pain as a result of underlying sensitization mechanisms, both central and peripheral.⁶⁰

In our study, Case 3 was on Gabapentin therapy, which is occasionally used as a single agent for the treatment of painful diabetic neuropathy, post-herpetic neuralgia, and other peripheral neuropathies. The patient’s NCP were neither due the post-herpetic or diabetic neuropathy. In our experience, patients with other non-diabetic and non-herpetic NCP etiologies have demonstrated a limited respond to gabapentin.⁶ Nevertheless, response to gabapentin or lack thereof should not be used as a diagnostic criterion, as most patients with neuropathic pain in general require multi-model therapies.^6,30,61

While our case series has a small sample size and lacks long-term follow-up, we clearly demonstrate that early stages of NK can occur with concomitant symptoms of discomfort, such as with NCP. Further, we cannot completely exclude that at least part of the pain could be due to inflammatory and/or nociceptive pain in patients with resolution of pain after the proparacaine challenge test. Nevertheless, our series demonstrates that the presence of symptoms alone should not be used to exclude the diagnosis of NK. Conversely, when assessing patients with NK, vision care providers should consider and treat symptoms of discomfort that could result due to underlying NCP.