Neuropathic ocular surface pain: Emerging drug targets and therapeutic implications

Abstract

Introduction –

Dysfunction at various levels of the somatosensory system can lead to ocular surface pain with a neuropathic component. Compared to nociceptive pain (pain due to a noxious stimulus at the ocular surface), neuropathic pain tends to be more chronic and refractory to therapies, making it an important source of morbidity in the population. An understanding of the options available for neuropathic ocular surface pain, including new and emerging therapies, is thus an important topic.

Areas covered –

This review will examine studies focusing on ocular surface pain, with an emphasis on those examining patients with a neuropathic component. Attention will be placed towards recent (e.g. after 2017) studies that have examined new and emerging therapies for neuropathic ocular surface pain.

Expert opinion –

Several therapies have been studied thus far, and continued research is needed to identify which individuals would benefit from specific therapies. Gaps in our understanding exist, especially with availability of in-clinic diagnostics for neuropathic pain in particular. A focus on improving diagnostic capabilities and researching gene-modulating therapies could help us to provide more specific mechanism-based therapies for patients. In the meantime, continuing to uncover new modalities for chronic pain and examining which are likely to work depending on pain phenotype remains an important short-term goal

1). Introduction

The cornea and conjunctivae derive their innervation from the trigeminal nerve, and these corneal nerves serve important functions in ocular surface protection and healing. As such, structural changes and/or dysfunction at any point along the somatosensory pathway (stemming from the ocular surface to the central nervous system (CNS) and back to the ocular surface) can negatively impact ocular surface health and/or vision. Furthermore, the ocular surface sensory pathway can be divided into different components, including peripheral corneal and conjunctival nerves, various structures within the CNS, and the autonomic nervous system (ANS); dysfunction within any of these locations may also give rise to chronic ocular surface pain with a neuropathic component. This review will summarize the current paradigm for the workup and treatment of neuropathic ocular surface pain, as well as discuss several emerging therapies for refractory pain.

2). Where does ocular surface pain come from?

The sub-basal nerve system within the cornea originates from the ophthalmic branch of the trigeminal nerve, and is involved in several important functions including processing of sensation, epithelial healing, and ocular homeostasis.(1) This system is among the most densely innervated in the body, with a total surface area estimated at 90 mm² and consisting of 5400–7200 nerve bundles. Each bundle separates into side branches containing 3–7 axons each, and overall, this accounts for an estimated 19,000–44,000 total axons. Each nerve fiber houses 10–20 nerve terminals, and thus it is estimated that there are 315,000–630,000 terminals in the cornea, or approximately 7,000 per mm.(2, 3)

The corneal sensory nerves are primarily involved in processes that mediate the sensation of pain, either physiologically as a result of sensing exogenous noxious stimuli (nociceptive), or aberrantly, due to internal nervous dysfunction after injury or excessive sensitization (neuropathic). The pathways that convey signals of pain start on the ocular surface and arrives at the CNS through a chain of neurons.(4, 5) The cell bodies of the primary order neurons are in the trigeminal ganglion (TG). These afferent neurons sense noxious stimuli on the ocular surface and convey the pertinent pain signals to projection second order neurons in the spinal trigeminal nucleus (SpV) in the medulla. After synapsing, the axons of the second order neurons decussate and reach the CNS, specifically the thalamus, to relay the pain signal via contralateral trigeminothalamic tracts. At the thalamus, they synapse to third order neurons, which project to cortical and subcortical centers of the brain, where the sensation of pain is perceived or wherein behavior- affective and neuro-endocrine responses to pain are mediated, respectively.(4, 5)

Along with these corneal and conjunctival nerves branching off the trigeminal nerve, abnormalities in the periocular sensory fibers can also contribute to chronic ocular surface pain.(6) Specifically, trigeminal nerve afferents in the periocular skin follow a similar path as primary afferents to the ocular surface. Cell bodies exist within the TG - they first synapse within the SpV, the second synapse in the thalamus, and then these fibers continue alongside their ocular surface fiber counterparts and terminate in specific brain structures involved in the sensation and emotional/behavioral response to pain.(7) Convergence of incoming pathways may occur at various levels that may mediate cross-sensitization and amplification of pain.

Finally, autonomic nerve fibers involved in the descending pathways of pain must also be understood when examining somatosensory fibers involved in chronic ocular surface pain. Sympathetic nervous system (SNS) efferent fibers project to the cornea from the superior cervical ganglion, while the parasympathetic nervous system (PSNS) sends fibers from the ciliary ganglion.(8) Both sympathetic and parasympathetic efferent mechanisms may contribute to chronicity of pain, as they may also contribute to painful conditions of other parts of the body (e.g. fibromyalgia(9), cluster headaches(10, 11), and complex regional pain syndrome(12)).

3). Nociceptive and neuropathic pain

The International Association for the Study of Pain (IASP) defines pain as a “an unpleasant sensory and emotional experience associated with, or resembling that associated with, actual or potential tissue damage.”(13) When applied to the eye, many individuals report having chronic sensations at the level of the ocular surface that are often described as dryness, burning, tenderness, aching, and/or foreign body sensation, to name a few. These sensations can occur spontaneously and/or be elicited by normally non-noxious stimuli, such as wind or light. In summary, the eye is an organ that should not be felt, so when individuals report persistent unpleasant eye sensations at the level of the ocular surface, they are termed as having ocular surface pain.

Ocular surface pain is a leading cause of clinic visits, increased ophthalmic healthcare costs, and social isolation and absenteeism from work. Symptoms are thus a significant burden to individuals and societies worldwide. Ocular surface pain can occur as a manifestation of an ocular condition, such as dry eye disease (DED), as a result of ocular surgery or trauma, and/or be co-morbid with pain elsewhere in the body (e.g., migraine, fibromyalgia, pelvic pain). Beyond its prevalence, chronic ocular surface pain is a major cause of disability and morbidity as it negatively impacts quality-of-life via impaired social, physical, and mental functioning, leading to decreased productivity.(14)

Ocular surface pain can be driven by nociceptive, neuropathic, or mixed mechanisms.(15–17) Nociceptive pain occurs from noxious stimuli at the level of the ocular surface, while neuropathic mechanisms occur with dysfunction within the somatosensory nerves supplying the cornea and/or CNS as a pain contributor.(4, 5) Neuropathic pain can further be distinguished by location of the lesion within the somatosensory system, specifically categorized as peripheral (e.g. dysfunction at the level of corneal and/or periocular nerve fibers), central (e.g. dysfunction within ascending and descending CNS fibers), and autonomic.(4, 5) Importantly, ocular surface pain can have a mixed background, involving multiple sources of pain.

In current literature, several terms have been utilized to describe the concept that ocular surface pain can arise from neuropathic mechanisms (peripheral and central), including ‘neuropathic ocular surface pain’, ‘neuropathic ocular pain’,(18) ‘ocular neuropathic pain’,(19) and ‘neuropathic corneal pain’(16), all of which convey the same concept. Furthermore, there are no agreed upon diagnostic criteria to define the entity, although certain examination findings suggest the diagnosis. Ultimately, the diagnosis of neuropathic ocular surface pain remains a clinical one.

4). Workup of neuropathic ocular surface pain

When approaching an individual with chronic ocular surface pain, it is important to obtain a thorough history and complete ocular and neurologic examination to identify potential contributors to pain. The workup begins with an evaluation of the ocular structures as well as of the periocular skin and eyelid anatomy, looking for skin diseases (rosacea, seborrheic dermatitis, atopy) and abnormal eyelid anatomy (ectropion, entropion, laxity, decreased blink rate). Next, point of care tests can be used, if available, to detect ocular surface inflammation (InflammaDry, Quidel, San Diego, CA) and high or unstable tear osmolarity (TearLab, San Diego, CA). Ocular sensation is then qualitatively tested with a cotton tip applicator or dental floss (0=none; 1=reduced; 2=normal; 3=increased). Next, a slit lamp is used to examine the ocular surface with the aid of fluorescein and lissamine green stains with assessment of tear lake height, tear stability, corneal and/or conjunctival staining, and anatomic abnormalities of the eyelids and eyelashes (anterior blepharitis), conjunctiva (conjunctivochalasis, pterygium), and cornea (map dot fingerprint). Subsequently, the patient is asked to rate their current ocular pain intensity in the right and left eye (0-10 numerical rating scale, NRS) followed by placement of an anesthetic drops, and rerating of pain 30 seconds to 2 minutes later. The Schirmer’s test can then be used to quantify tear production, and pressure on the meibomian glands can highlight the quality of the meibum.(20) If available, ancillary testing can be performed such as infrared imaging of the Meibomian glands to highlight areas of dropout, and confocal microscopy to examine corneal nerve morphology and density, and dendritic cells in the central corneal. Overall, this examination is performed to examine for potential nociceptive sources of pain that need to be addressed.

It is important to note that not all clinical findings are found in every patient with neuropathic ocular surface pain, and that there is not one ‘gold standard’ diagnostic test that confirms the diagnosis. Instead, the clinician must examine the patient in total, looking for the features noted above. Some key findings that point to a clinical diagnosis of neuropathic ocular surface pain include presence of pain co-morbidities (migraine, fibromyalgia, traumatic brain injury) or ocular surface pain that started after trauma or surgical injury (refractive surgery).(20) Other clues include the presence of specific symptoms such as sensitivity to wind and light (ocular hyperalgesia and allodynia, respectively) and overall, symptoms that are out of proportion to signs of disease (“pain without stain”).(21, 22) On examination, it is important to recognize any abnormalities in corneal and/or adjacent cutaneous sensation (decreased or increased sensation)(23) and abnormalities in the morphology and density of nerves on in vivo confocal microscopy (IVCM). Neurologic abnormalities that suggest central neuropathic pain include persistent ocular pain after placement of a topical anesthetic (proparacaine challenge, suggestive of centralized neuropathic mechanisms) and the presence of cutaneous allodynia (pain to light touch around the eye).(24) Finally, neuropathic ocular surface pain (or pain coming from another area of the eye or a non-ocular source of pain) should be considered if pain persists despite treatment of ocular surface abnormalities with a variety of modalities.(25)

5). Assessments of chronic ocular surface pain

Several questionnaires can be used to quantify ocular surface pain. Some were initially developed as DED questionnaires but incorporate questions about pain. Commonly utilized questionnaires include the Dry Eye Questionnaire 5 (DEQ5) which asks about the frequency and intensity of discomfort (“During a typical day in the past month, how often did your eyes feel discomfort? When your eyes feel discomfort, how intense was this feeling of discomfort at the end of the day, within two hours of going to bed?” (26)), and the Ocular Surface Disease Index (OSDI) which asks about the frequency of various pain complaints (“Eyes that are sensitive to light? Eyes that feel gritty? Painful or sore eyes?”, along with questions about vision, triggers, and impact on life(27)). Subsequently, questionnaires have been developed specifically to assess the intensity and quality of ocular pain. These include the Numerical Rating Scale (NRS; pain only, rated on Likert scale 0-10), the Neuropathic Pain Symptom Inventory modified for the Eye (NPSI-E(28); specific for neuropathic pain descriptors, such as burning, squeezing, pressure, sensitivity to wind, light] and the Ocular Pain Assessment Survey (OPAS(29); pain, descriptors, and related quality of life).

6). Current strategies for the treatment of chronic ocular surface pain

The first strategy in addressing chronic ocular surface pain is to treat nociceptive sources of pain. If this strategy is not successful, or if minimal nociceptive sources are identified one examination, treatment of neuropathic pain should be considered.

6.1 –. Current strategies for treating peripheral neuropathic pain

Generally, topical therapies are used as first line agents in individuals with nociceptive or peripheral neuropathic pain. Most topical therapies can improve both ocular surface parameters and peripheral afferent function, as chronic ocular surface abnormalities are one cause of peripheral nerve dysfunction. Such treatments include artificial tears and anti-inflammatories, to name a few. Numerous studies have evaluated the efficacy of such treatment modalities. For example, a randomized controlled trial (RCT) evaluated 0.1% cyclosporine in 292 subjects with chronic ocular surface pain in the setting of dry eye disease (OSDI, Schirmer ≤5 mm, staining). A significant improvement in pain was noted every month for up to 6 months follow-up in the cyclosporine 0.1% group (p≤0.01; data not provided), while improvement was only noted in the vehicle alone group starting at month 6 (p≤0.01; data not provided).(30) Similar findings were noted in an open label Turkish study of individuals with Sjögren’s who received 0.05% cyclosporine (n=36) vs. saline (n=20). A higher frequency of individuals in the cyclosporine group reported total resolution of discomfort present at baseline at 1-month follow-up [resolution in 14 of 24 individuals with baseline discomfort (58.3%)] compared to resolution in 1 of 17 (5.88%) in the control group (p=0.01). Similar trends were noted in regards to photosensitivity (14 of 23 (60.9%) vs. 1 of 14 (5.88%); p=0.001) and for burning symptoms (8 of 23 (34.8%) vs. 0 of 18 (0.0%); p=0.0001).(31) Beyond improving tear abnormalities, topical anti-inflammatories have been shown to affect peripheral nerve metrics – in an open label study 30 individuals with Sjögren’s were treated with 0.05% cyclosporine BID for 6 months. Overall, nerve density (via what confocal microscope) increased after treatment (12.4±4.5 to 16.5±5.9, p<0.001).(32)

Autologous blood products (serum tears, platelet enriched plasma) are another group of topical therapies used in individuals with nociceptive and peripheral neuropathic). These therapies contain several blood-bound neurotrophic and epithelial growth factors [e.g. nerve growth factor (NGF), insulin-like growth factor-1, brain-derived neurotrophic factor, neurotrophin-3, and glial cell line-derived neurotrophic factor] which have been shown to promote the growth, regeneration, and restoration of corneal nerves.(33) In one study, 51 subjects with systemic autoimmune diseases and DED (defined using OSDI, corneal staining, and failure of previous DED therapy) were treated with 20-50% autologous serum tears (AST) for a mean of 14.3 ± 11.7 months. Overall, OSDI symptom scores improved (baseline 62 ± 24.2 to final visit 47 ± 22.3, p = 0.03) in 85.1% of individuals, with a third of the subjects (34.6%) noting complete resolution of ocular discomfort and 14.9% reporting no benefit.(34) Beyond improving tear abnormalities, AST have been shown to improve peripheral nerve metrics. In a study of 16 individuals with presumed neuropathic ocular surface pain (defining characteristics not included), ocular pain severity decreased (9.10 ± 0.20 (range 8–10) to 3.10 ± 0.30 (range 0–4), p < 0.0001) and nerve count increased (10.50 ± 1.40 to 15.10 ± 1.60 nerves/frame, p < 0.0001) with AST treatment (mean 3.8 ± 0.5 month, range 1–8).(35) These data suggest that in individuals with nociceptive and/or peripheral neuropathic pain, topical agents that improve tear parameters and peripheral nerve function can improve chronic ocular surface pain.

It is important to consider that AST may modulate both nerve and immune function. In this regard, AST may inhibit the release of pro-inflammatory molecules such as the neuropeptides substance P (sP) and calcitonin gene-related peptide (CGRP), which are released after nerve injury.(36) The release of these molecules causes neurogenic inflammation by triggering downstream local release of pro-inflammatory cytokines like IL-1, TNF, and IL-6, via NF-κB pathway activation.(37) In summary, there is an interplay between the immune and nervous systems and molecules that impact nerve function often have an impact on inflammatory pathways as well.

6.2 –. Current strategies for treating central neuropathic pain

Many individuals with chronic ocular surface pain have a central component and, in these individuals, oral neuromodulators are first-line therapies. These medications, overall, aim to suppress neuronal excitability and signal propagation to the CNS. Commonly used agents include α₂δ ligands (gabapentin or pregabalin), tricyclic antidepressants (TCAs, nortriptyline or amitriptyline), serotonin-norepinephrine reuptake inhibitors (SNRIs, duloxetine), and anti-convulsants/sodium channel blockers (carbamazepine, oxcarbazepine or topiramate).(38, 39) These agents can be administered alone, but are often combined in cases where monotherapies fail to control symptoms. For example, α₂δ ligands may be combined with TCAs (nortriptyline), SNRIs (duloxetine), and/or opioid-targeting agents.(40) Various mechanisms of action have been implicated. Specifically, gabapentin and pregabalin inhibit presynaptic voltage-gated calcium channels (thus resulting in reduced release of excitatory neurotransmitters at presynaptic terminals) and suppress synaptic propagation of pain signaling and neuronal excitability,(41) antidepressants activate endogenous inhibitory analgesic pathways, while some antidepressants (amitriptyline) and anticonvulsants inhibit sodium channels and NMDA glutamate receptors.(42, 43)

It is important to remember that these agents are neuromodulators rather than direct acting analgesics, and that they may also have unwanted effects. Therefore, patients need to be counseled on expectations, anticipated benefits, risks, and the anticipated time course of analgesic/anti-hyperalgesic effects. This understanding improves compliance with long term therapy that is necessary in most of these patients. Selection of the most appropriate agent for initiation of therapy should be tailored to an individual’s specific needs and underlying medical profile. After proper selection, each medication should be generally started at a low dose and slowly titrated to a therapeutic dose (“start low and go slow”). If not tolerated, it should be discontinued and another drug, from a different category should be tried. Pain improvements typically occurs after 2-3 months and the effect may be variable and gradual, with continued improvements up to one year or more. In case of a partial response, addition of a second agent (from a different category of agents may be necessary, such as, for example, addition of a nortriptyline to gabapentin or to pregabalin). Our general strategy is to find a medication or a combination that is well tolerated and reduces pain, and to maintain each patient on the therapy for 2-3 years prior to weaning off.

In an case series of 8 subjects living in South Florida with a clinically diagnosed neuropathic ocular surface pain (pain out of proportion to clinical findings, poor response to topical therapies), gabapentin (300mg daily, escalation to 600-900 TID) or pregabalin (75mg daily, escalation to 150 mg BID) led to complete pain relief in 2 subjects (NRS=0 on a 0–10 scale), significant improvements in 3 subjects (NRS≤2), and slight noticeable improvement in 1 subject (NRS=10->7), while 2 subjects found no improvements.(44) Interestingly, the two subjects who noted complete relief were also on duloxetine (20 mg daily, escalation to 60 mg daily). While no studies have examined duloxetine as a treatment for chronic ocular surface pain, studies have demonstrated its efficacy in diabetic neuropathy – in a multicenter RCT, 348 type 1 or 2 diabetics received duloxetine 60 mg once daily, duloxetine 60 mg twice daily, or placebo for 12 weeks (n=116 each). Individuals randomized to duloxetine more frequently reported a 30% reduction in pain at 12 weeks (0-10 scale, 24 hour recall: daily 68%, twice daily 64%, placebo 43%).(45) Finally, tricyclic antidepressants such as nortriptyline have also been evaluated in chronic ocular surface pain. In an open label British study of 25 individuals with neuropathic ocular surface pain treated with nortriptyline (10 to 25mg starting dose escalated to 100mg daily based on effect and tolerance), pain scores were lower at 4 weeks post- vs. pre-treatment (via NRS; 3.8 ± 2.4 vs. 6.4 ± 2.2; p < 0.0001), with improvement reported in 84% of subjects (n = 21), including 28% with >50% improvement (n = 7), 40% with 25–50% improvement (n = 10), and 32% with < 25% improvement (n = 8).(46) It is important to remember that while several oral therapies have been used to treat neuropathic ocular surface pain, it is not possible to predict which patient will respond to which therapy and as such, a trial and error approach is needed to optimize therapy, often in combination with topical agents.

7). Alternative & emerging therapies for neuropathic pain

7.1 –. Current understanding of neuropathic pain & targets being evaluated in the pre-clinical setting

Despite the high prevalence of neuropathic ocular surface pain and its morbidity, there is a disproportionate paucity of relevant preclinical studies. The current understanding of the mechanisms that drive neuropathic ocular pain (peripheral and central) is based on these few models,(47–50) which are summarized by Puja et al(51) and by Bista & Imlach(52), as well as from extrapolating information from non-specific preclinical models of pain.(53) For example, in a mouse model of ocular pain due to corneal injury (10 second corneal surface exposure to local application of an alkali solution (0.75 N NaOH), treatment with an opioid antagonist (naltrexone or naloxone-methiodide, subcutaneously (3 mg/kg) or topically (eye drop, 100 lM)) was found to reinstate pain-induced eye-wiping behaviors after normal healing, suggesting that endogenous opioid receptors are involved in latent pain sensitization (attenuation of nerve hypersensitivity which may normally lead to chronic ocular pain after corneal injury).(49) In another model of guinea pigs previously subjected to a corneal surgical lesion (healed 4mm corneal flap created via microkeratome, similar to the wound during LASIK) action potential discharges of polymodal nociceptors were increased in response to chemical stimulation (98.5% CO2 jet) but not to mechanical stimulation, suggesting that chemical responsiveness becomes potentiated (i.e. sensitized) after corneal injury. Interestingly, sodium channel blockers (lidocaine, carbamazepine, and amitriptyline) applied 24-48 hours after the procedure significantly decreased this response to acidic CO2 stimulation, suggesting their utility in treating post-surgical spontaneous pain, hyperalgesia, and allodynia.(50)

Furthermore, identification and design of mechanism-targeting treatments is similarly based on these models, as well as on extrapolating results from non-ocular trigeminal,(52, 54) spinal or peripheral nerves,(55) or “generic”(53, 56) pain models. The relative lack of such studies may reflect the inherent difficulties in assessing chronic pain behaviors in animals subjected to such states,(57–59) which may be reflexive(60) or not adequately specific(61) for the purpose of identifying clinically pertinent therapeutic targets or mechanism-based therapies. While major limitations include methodological difficulties in techniques and in assessing and measuring the behavioral expressions of “pain” in such states, there is no doubt that such models will be necessary identifying mechanisms and therapeutic targets for translational studies.(62–64) We hope that more focused studies will emerge in the future, as these will help determine more specific mechanisms pertinent to the pathogenesis and pathophysiology of chronic ocular neuropathic pain mechanisms, as well as help to guide mechanistically pertinent treatments. Until then, a variety of basic studies focused in trigeminal, non-ocular, or spinal neuropathic states that intriguingly may apply to ocular neuropathic pain.(65)

The role of corneal receptors and transduction channels involved in corneal nociception has been most extensively investigated in basic studies.(37, 51, 66, 67) Sodium channel(68) and calcium channel(69–71) alterations has been investigated the most as having a role in peripheral neuropathic pain,(72) including in trigeminal nerve-related pain states.(73–75) As such, targeting sodium and calcium channels with appropriate agents(76–78) has been established as a mainstream therapeutic strategy in such states (with potential for neuropathic ocular pain, too), while in comparison little is known about other potential targets. For example, drugs acting as voltage-gated calcium channel blockers (the levels of which upregulate after nerve injury)(79, 80) are commonly used to treat these conditions, while the use of sodium blockers has been so-far limited by incomplete responses.(81) Yet, recent preclinical studies(82, 83) have provided material for future translational studies.(84) These include the development of PF-0508977 (a novel small orally bioavailable molecule and peripherally acting arylsulfonamide prototype which has been shown to act as a NaV1.7 selective inhibitor),(85) as well as novel blockers that inhibit sodium currents not by blocking channel activation, but by enhancing inactivation, thus resulting in significant hyperpolarization shifts.(86)

Neuronal potassium channels may play a significant role in generating and maintaining neuropathic pain,(87) including trigeminal pain.(88, 89) These channels include the ATP-sensitive potassium (K_ATP) channels,(90),(91) and their altered upstream regulation by cytosolic messengers, such as cytosolic calcium and/or cytosolic kinases (e.g. calmodulin (CaM) and calmodulin-dependent kinase II (CaMKII)), has been shown to drive neuropathic pain, at least in spinal nerves.(92) Animals with deficient mobilization of calcium from cytosolic sources develop neuropathic pain after injury, in contrast to animals with intact calcium release mechanisms that have been subjected to the same type of injury. In animals that have hyperalgesia after nerve injury, loss of calcium-dependent signaling results in suppressed CaMKII activity and a decrease in K_ATP channel opening, with subsequently enhanced membrane excitability and excitatory transmitter release leading to hyperalgesic responses and pain behavior.(92) Furthermore, these same K_ATP channels are available for activation by appropriate pharmacologic agents,(93) and channel opening agents may thus have a role in managing neuropathic pain in the eye. These preclinical studies have highlighted the need for examining such drugs (e.g. diazoxide, minoxidil, diclofenac) in this clinical context. Additional benefits of drugs that act as K_ATP channel openers are neuroprotection, including protecting retinal ganglion cells from ischemic stress and from glutamate-induced toxicity,(94) as well as regulation of the intraocular pressure,(95, 96) suggesting the need for further studies in patients with ocular pain, underlying ischemia, and/or glaucoma.

Other drugs, such as desipramine or fluoxetine,(97, 98) may activate the deficient CaMKII in these settings for therapeutic benefit.(99) What drives the generation and development of neuropathic pain and hyperalgesia nerve injury is the specific deficiency of the calcium/CaM/CaMKII signaling that regulates K_ATP channel opening, that may originate from decreased calcium mobilization in the animals that develop pain only. This may indicate that unique genetic traits or mechanisms are responsible, as suggested for other types of trigeminal pain,(100) and this may require further investigation. Identification of individuals, genetically susceptible to development of neuropathic pain after ocular surgery, may then lead to preventative strategies that would minimize the transition to pain, or exclude those patients from elective surgical procedures with high risks for transition to intractable pain.

The contribution of basic studies on endogenous inhibitory effects, either GABA-ergic, glycinergic, or descending serotoninergic, onto the oculofacial pain processes has been summarized by Puja et al.(51) The use of agents with neurotrophic properties has been discussed in this article, as well as therapies targeting inflammation on the ocular surface. Yet, these therapies should be combined with those aiming at reducing the afferent signaling that leads to enhanced neuroexcitatory transmission, sensory perception, and behavioral manifestations of pain. This is particularly the case in neuropathic pain, wherein both peripheral and central neuronal sensitizing mechanisms have been implicated in photophobia and emotional/behavioral manifestations of pain. Many preclinical studies have shown that the input from various converging pathways sensitizes central neuronal centers to the point that they generate photophobia, and these have been comprehensively summarized by Digre and Brennan.(101) Regarding the emotional and behavioral manifestations of neuropathic ocular pain, the relevant mechanisms are very complex, and probably based in both neurobiological processes as well as in psycho-social reinforcing mechanisms (in humans). While the elucidation of the latter requires complex psycho-social studies, some of the former have been based on animal investigations that have shown the close connectivity and interactions between the trigeminal nociceptive afferent system and centers of the brain that control emotions and behaviors, such as the parabrachial systems.(102, 103)

Yet, considering the chronic and refractory nature of neuropathic pain in cases wherein central sensitization has been established other preclinical models should be applicable. These models investigate the role of long-term potentiation mechanisms(104, 105) including processes at brainstem centers, such as the interpolaris/caudalis (Vi/Vc) and Vc/upper cervical cord (Vc/C1) regions,(106) but equally or most importantly the role of glial activation.(107, 108) The later has been recognized as a major contributor to pain, to the point that chronic pain may be considered as a “gliopathy.”(108, 109) Various preclinical models have been investigated, but so far specific focus in neuropathic ocular pain is lacking. Other models have highlighted the role of cytokines and “neuroinflammation” maintaining the glia in an active status dynamically contributing to pain,(110) while little is known regarding reversal of the sensitized state. Adenosine A3 receptor agonists (MRS5698, MRS5980)(108), as well as low dose naltrexone (LDN),(111, 112) have been proposed as emerging therapies with analgesic potentials, based on relevant basic research.(113) LDN reduces glial inflammatory response by modulating Toll-like receptor 4 signaling,(114–116) and may have an analgesic benefit in chronic pain states. It has been incorporated into clinic practice, as monotherapy or in combination with other drugs, with good results, although long term therapy is necessary.(117)

It is important to note these multiple interacting mechanisms that may contribute to the manifestations of neuropathic pain;(104, 105) therefore, multimodal therapies targeting as many of those mechanisms as possible is often necessary for therapeutic improvement.(16, 17, 39) This may be particularly helpful in refractory cases wherein monotherapies have failed to produce relief. Multiple converging peripheral neuronal pathways are involved in the mechanisms of neuropathic pain, and these include both those directly exposed to injury or disease, as well as intact nerves, adjacent to injured ones. Animal models have highlighted the role of nerves not directly injured but adjacent to those subjected to injury(118–120), and targeting those nerves clinically with nerve blocking agents(39, 121) has been successful in alleviating pain, as we have shown.

7.2 –. Alternative therapies for peripheral neuropathic pain

In instances wherein first-line therapies for neuropathic ocular surface pain are non-tolerable, or if pain is refractory, alternative and adjuvant therapies can be considered, depending on the location of the lesion.(122)

Besides AST, amniotic membrane transplant (AMT) has been used as an adjuvant modality in peripheral neuropathic ocular surface pain. In a study of 9 patients living in Boston with a clinical diagnosis of neuropathic ocular surface pain (ocular pain and IVCM abnormalities), placement of an AMT (PROKERA, Bio-Tissue, Miami, FL; mean retention 6.4 ± 1.1 days) reduced pain (6.3 ± 0.8 to 1.9 ± 0.6, p = 0.0003) and increased nerve density on laser-scanning IVCM (17,700.9 ± 1315.7 to 21,891.3 ± 2040.5 μm/mm2, p = 0.05) over an average follow-up period of 9.3 ± 0.8 months (range 7.6–13.8 months).(123) While there is interest, no data are available on the use of recombinant NGF (other than as part of an AST formulation) for the treatment of peripheral neuropathic pain.(124)

More recently, TRP (transient receptor potential vanilloid 1 (TRPV1), and transient receptor potential melastatin 8 (TRPM8)) ion channel binding agents (agonists and antagonists) have transitioned into human testing. TRPV1 is involved in the transduction of signals of pain from the ocular surface to the CNS. Agonist molecules induce an analgesic effect by maintaining these ion channels in an open state, allowing sustained influx of calcium resulting in calcium-induced cytotoxicity and nociceptive nerve neurodegeneration. Antagonistic compounds block the receptor and induce analgesia by preventing nociceptive signaling directly.(125) In one study, the effect of a TRPV1 antagonist vs placebo was tested in 40 subjects who were undergoing photorefractive keratectomy (PRK). Individuals who received the TRPV1 antagonist had lower mean visual analog scale (VAS) pain scores 1 hour postop vs subjects who received placebo (−53%) and the difference in pain scores persisted up to 30 hours postop (p≤0.10).(126) Based on these promising results, a larger randomized trial is underway which aims to examine effects of a TRPV1 antagonist in individuals with chronic post-surgical ocular surface pain that is partially modulated by a topical anesthetic (NCT04630158).

TRPM8 is instead involved in tear secretion and is activated by evaporative cooling and by hyperosmolar solutions. Agonists such as menthol and icilin induce cooling effects on the ocular surface directly to relieve pre-existing symptoms, however, they are rarely applied directly to the ocular surface because they subsequently cause discomfort and stinging. Antagonists, on the other hand, prevent evaporative cooling and hyperosmotic stimuli as the triggering cause of pain in the first place.(127) In an open label pilot study of 15 individuals living in Korea with DED-associated neuropathic ocular surface pain, a TRPM8 agonist, cryosim-3 (C3), was applied to the eyelid 4 times/day for 1 month. On Ocular Pain Assessment Survey (OPAS), ocular pain (scale 0-60) was decreased at 1 week (26.47 ± 11.45; p=0.01) and 1 month (21.53 ± 10.84; p=0.02) as compared to baseline (30.60 ± 12.84), and quality of life (scale 0-60) improved in a similar fashion (1 week: 27.60 ± 15.49 and 1 month: 27.17 ± 16.06) when compared to baseline (33.53 ± 14.24; p=0.003 and 0.02 respectively).(128)

Similar to AST, it is important to consider that beyond neuromodulation, altering TRPV1 and TRPM8 function can have immunomodulatory effects as well. For example, TRPV1 antagonists have been shown to inhibit the release of sP after neuronal injury.(37)

7.3 –. Emerging therapies for peripheral neuropathic pain

Several other emerging therapies are still being evaluated in pre-clinical studies, including other molecules that bind TRP ion channels, omega 3 fatty acid derivatives, and cannabinoid agonists. For example, one animal study examined use of the TRPV1 agonist resiniferatoxin (RTX) in controlling ocular surface pain after capsaicin challenge (50 μL 0.02% CAP) in a rat model. This study found that pain in the form of eye wipe count was reduced in a dose-dependent manner in RTX-treated mice. Specifically, the difference in mean eye wipe counts between RTX-treated eyes and contralateral untreated eyes became significant at doses ≥1μg.(129) In a similar manner, studies utilizing rat models have demonstrated efficacy of TRPV1 blockage(130–132) and TRPM8 agonism(133) in reducing ocular surface pain.

Omega 3 fatty acid molecules are also being examined in the preclinical stage. For example, docosahexaenoic (DHA) and similar precursor omega 3 fatty acids promote the release of docosanoids, a family of endogenous anti-inflammatory lipid mediators (e.g. neuroprotectin-1 or resolvin-1) which have been found to have anti-inflammatory activity. These molecules work in a cascade along with other growth factors to increase synthesis of NGF for increased neurogenesis, and have also been found to have immunomodulatory effects in inhibiting production of pro-inflammatory cytokines (IL-1, TNF-a).(134) In a rabbit model of nerve injury (lamellar keratectomy), eyes were treated with a collagen shield soaked in a combined DHA-pigment epithelium derived growth factor (PEDF) wash or vehicle. Six weeks after treatment, the corneal nerve area stained with βIII tubulin (as a surrogate calculation for nerve density) was found to be higher in the PEDF-DHA group (PEDF-DHA: 26.7±2.6% vs. vehicle: 11.7±1.7%; p=0.01). Corresponding signals in corneal sensation were noted on Cochet Bonnet, with a complete absence of sensation in all groups (0 cm) after the injury, and a greater return of sensation in the PEDF-DHA group at 7 weeks compared to vehicle (PEDF-DHA: 75% return (2.8 cm) vs. vehicle: ~47% return (~1.7cm); p<0.05).(135) Similar findings of increased nerve regeneration and decreased pain have been noted in pre-clinical studies focusing on resolvin, a similar molecule.(136)

Cannabinoid receptor 1 (CB1) agonists also been investigated in animal models of ocular surface pain.(137) In a mouse model, chemical cauterization of the cornea (causing corneal hyperalgesia) followed by treatment with a combination CB1 agonist compound (CB1 ligands GAT211, GAT228, or GAT229, either alone or in combination with another CB1 orthosteric agonist Δ8-tetrahydrocannabinol). Topical administration of GAT211 or GAT229 in combination with 0.4% Δ8-THC, or treatment with GAT228 alone, were all found to reduce pain (assessed by number of eye wipes) induced by a 1 μM capsaicin challenge 6 hours post-cauterization to a greater degree than vehicle (p<0.05; specific data not provided).(138) Other compounds currently being examined for their therapeutic potential in pre-clinical stages for neuropathic pain include melatonin analogs,(139) cibinetide,(140) and vitamin B12.(141) In summary, several therapies currently in the pre-clinical stage of study may become viable as they continue to progress through the various stages of investigation.

7.4 –. Alternative therapies for central neuropathic pain

As above, several therapies have been used to treat individuals with a suspected central component to their neuropathic ocular surface pain (α₂δ ligands, TCA antidepressants). Several alternative and adjunctive agents may be added to the more traditionally utilized pain-control agents, such as opioid-based agents (low dose naltrexone or tramadol), anticonvulsants, and adjuvant therapies targeting the periocular afferents.

Opioid receptor-modulating agents have been combined with α₂δ ligands and TCA antidepressants in individuals with chronic ocular surface pain. For example, low dose naltrexone has been studied in the treatment of chronic ocular surface pain, most often in conjunction with a TCA.(117) The effects of low dose naltrexone are attributed to antihyperalgesia(142) (transient blockade of μ- and δ opioid receptors) as well as reduced neuroinflammation (antagonistic binding to the Toll-like receptor-4).(143) In an open label study of 59 patients living in Boston (n=14 postsurgical) with centralized neuropathic ocular surface pain (presence of neuropathic symptoms, IVCM findings, and/or persistent pain after topical anesthetic), naltrexone 4.5mg nightly was given along with other oral medications (e.g. nortriptyline (26.6%), selective serotonin reuptake inhibitors (SSRIs – 26.6%), and gabapentin (23.3%)) for a mean of 14.87±11.25 months. Overall, a 49.2% improvement in pain intensity was noted from baseline (3.23±2.60 from 6.13±1.93, p<0.001, range 0-10).(117) Tramadol is another agent used in the treatment of neuropathic pain, but it has not been formally studied in relationship to chronic ocular surface pain. Tramadol is thought to promote analgesia by acting as a weak μ-opioid receptor agonist.(144)

Anticonvulsants (carbamazepine, phenytoin, lamotrigine, valproate) have been used in individuals with presumed central neuropathic ocular surface pain who failed to respond or could not tolerate first line agents.(145) Rationale exists as mechanisms that underlie neuropathic pain (neuronal hyperexcitability, abnormal expression and/or activation of sodium channels, increased glutamate activity, altered GABA inhibition, altered calcium influx into cells) overlaps with abnormalities noted in epilepsy.(146) While no studies have examined the effects of these anticonvulsants in a systematic manner, studies have shown efficacy in improving neuralgia in other systemic pain conditions including trigeminal neuralgia(147, 148) and diabetic neuropathy.(149, 150) We have anecdotally used topiramate in a number of patients with neuropathic ocular surface pain, with benefit noted in some individuals who have failed therapy with first-line agents.

Several adjuvant therapies can be used to target the periocular trigeminal afferents, such as nerve blocks, botulinum toxin injections (BoNT-A), and non-invasive neurostimulation devices, all of which can be used in conjunction with topical and oral therapies. In individuals with cutaneous allodynia, nerve blocks may be utilized in addition to or instead of oral medications. This modality entails a reversible blockade of sodium channels, which prevents generation of action potentials involved in propagating the pain signal and synaptic transmission. Sodium channel blockade may thus minimize sensitization, particularly combined with a long-acting corticosteroid that potentiates the effect by reducing ectopic generation of nociceptive firing and by various genomic and non-genomic modulating effects on primary afferents.(151) A case series reported on outcomes of 11 subjects with presumed neuropathic ocular surface pain after periocular (supraorbital, supratrochlear, infratrochlear, and infraorbital) nerve blocks (4 mL of 0.5% bupivacaine with 1 mL of 80 mg/mL methylprednisolone acetate). In total, 7 subjects experienced pain relief, lasting from hours to several months.(44)

BoNT-A exerts analgesia by inhibiting release of neuroexcitatory transmitters (e.g. CGRP) and attenuating nociception via unmyelinated C-fibers involved in relaying signals contributing to pain perception, photophobia, and dryness.(152) In a study of 76 individuals living in South Florida with migraine who received BoNT-A (100–150 units every 3 months), improvements in photophobia were noted following BoNT-A (via Visual Light Sensitivity Questionnaire-8 (VLSQ-8); 4.89±2.97 to 3.37±2.54, p<0.001). Furthermore, dry eye symptoms significantly improved but only in the subset of patients with severe DED symptoms at baseline (DEQ5 score ≥12; n=38) (via DEQ5; 15.40±2.47 to 13.80±4.02, p=0.03).(153) In a subsequent study, improvements in pain were found not to be related to tear parameters. In a study of 62 individuals with migraine treated with BoNT-A (100–150 units every 3 months), the positive effect of BoNT-A on migraine pain and photophobia (4.9 ± 3.0 to 3.4 ± 2.5 vs.; p<0.001) 6 weeks post- vs. pre-treatment was found to be independent of tear film volume.(153) Fortunately, similar effects of BoNT-A were noted in individuals with ocular surface pain and photophobia but without migraine. In one case series of 4 individuals, 35U of BoNT-A were injected in 7 areas of the forehead. At one month, all patients noted decreased photosensitivity, with notable improvements in frequency of light sensitivity to outdoor daylight. DED symptoms also improved with regards to in frequency and severity of pain-related symptoms.(154)

Finally, non-invasive neurostimulation is a treatment that can be applied to peripheral (e.g. transcutaneous electrical nerve stimulation [TENS]) or central (e.g. transcranial magnetic stimulation [TMS]) nerves.(152, 155, 156) One study evaluating the efficacy of TENS found that an in-office 30 minute session (RS Medical RS-4i Plus Sequential Stimulator) improved ocular pain in an open-label fashion in 14 individuals with chronic ocular pain. Overall, mean pain was reduced 5 minutes post-vs. pre-treatment (0–10 NRS: right eye 4.54±3.20 to 1.92±2.50, p=0.01; left eye 4.46±3.36 to 2.00±2.38, p=0.01).(157) This therapy can be incorporated into the long term treatment of chronic pain, as demonstrated by persistent use (mean 6.6 ± 3.6 months) in 10 individuals with chronic ocular surface pain from a variety of etiologies - pain decreased post- vs. pre-treatment by 27% (5.6 vs. 7.6; p=0.02).(158) A positive effect on pain has been found with various TENS units. One device (Cefaly, Cefaly Technology, Herstal, Belgium) has been cleared for used in migraines. In an open label study of 18 individuals living in South Florida with clinically diagnosed neuropathic ocular surface pain (persistent pain after topical anesthesia, discordance between symptom severity and ocular signs, burning, sensitivity to light and wind comorbid chronic pain conditions), all individuals received the trigeminal neurostimulator device and reported monthly on their pain intensity. Over the 6 months follow up period (mean frequency of use 3.7 ± 1.9 sessions for 25.8 min at month 1 to 2.7 ± 2.3 sessions of 28.0 min at month 6), ocular pain intensity continued to decrease, with significance noted at 3 months compared to baseline. At 6 months, a 31.4% decreased in ocular pain intensity scores were noted (NRS; 3.8±3.5 to 2.7 ± 3.0, p=0.02), and individuals reported reduced light sensitivity (5.7 ± 3.2 to 4.7 ± 2.6, p=0.11) and burning (3.3 ± 3.4 to 2.3 ± 2.8, p=0.03).(159) Overall, several therapies that modulate periocular peripheral afferents surrounding the eye have been found to improve ocular in individuals with neuropathic contributors, and these strategies can be utilized in conjunction with oral and topical medications.

Finally, given the close association of ocular surface pain and psychological dysfunction,(160, 161) ancillary therapies such as cognitive–behavioral therapy (CBT) must be considered. While no studies have examined CBT for treating ocular surface pain specifically, studies have demonstrated that CBT is a viable treatment adjuvant for other pain conditions, such as low back pain. Exemplifying this point, a British RCT separated 598 adults with low-back pain into CBT (n = 399) and control groups (n = 199). Individuals in the CBT had greater improvements in pain levels compared to controls at 12 months, measured via a Von Korff pain scale (range 0–100%; mean decrease in pain scale of 13.8% vs. 5.4%; p<0.0001).(162) Other therapies that have also been examined for use in varying chronic pain conditions include acupuncture, exercise, and massage therapies.(122, 163)

7.5 –. Alternative therapies for individuals with ans-related nerve dysfunction

While less investigated as a cause of chronic ocular surface pain compared to peripheral and central mechanisms of pain, some individuals have been observed to have parasympathetic or sympathetic contributors to pain. In such individuals, use of sphenopalatine ganglion (SPG) or superior cervical ganglion blocks may be considered to block parasympathetic or sympathetic responses, respectively. Although biologic plausibility exists, further studies are needed to evaluate how to determine whether autonomic mechanisms impact ocular surface pain and which individuals would benefit from autonomic ganglion blocks. These approaches have been examined in migraine, however, which has a related pathophysiology to neuropathic ocular surface pain.(164) In particular, one study of 88 individuals in Pennsylvania with migraine underwent a suprazygomatic SPG block and reported a mean 67.2% reduction (on a 10 point Likert scale) in headache severity beginning 30 minutes after the procedure (p<0.001) with no decrease in efficacy on repeated injections, however a major limitation of this study is the lack of follow-up data to examine duration of pain relief days and weeks post procedure.(165)

8). Conclusion

Neuropathic mechanisms contribute to chronic ocular surface pain in some individuals, the manifestations of which can be variable, and include sensory dysfunction (allodynia, photophobia) and behavioral and emotional dysfunction in addition to pain. Therefore, patient-centered, comprehensive, and multi-modal therapies applied over a continuum of time are necessary to ensure optimal outcomes (Table 1). Treatments include topical and oral neuromodulating medications, and, in addition, adjuvant therapies in individuals intolerant to first-line therapies or in those with refractory pain. Cognitive-behavioral therapy, ongoing emotional support, and counseling is essential in this context as well. While several therapies have been outlined in this review, continued research is needed to identify more options, test current therapies more robustly, and identify which individuals would benefit most from particular therapies.

Table 1.

Summary of Key Targets for the Treatment of Peripheral and Central Neuropathic Pain

MEDICATION	DOSAGE	TARGET	MOA
Oral:
α2δ ligands (gabapentin and pregabalin)(44)	Gabapentin: start 100-300 mg nightly, escalate to 600-900 TID; Pregabalin: start 25-75 mg nightly, escalate to 150-200 mg BID	Ca2+ channels at presynaptic terminals	Inhibition of presynaptic Ca2+ channels, reducing release of excitatory neurotransmitters and therefore suppressing propagation of pain
TCA (amitriptyline, nortriptyline)(166)	Start at 10-25 mg nightly, escalate gradually up to 100 mg per day	Serotonin and norepinephrine reuptake at nerve terminals	Reinforcement of the descending inhibitory pathways of pain by prevention of monoaminergic neurotransmitter reuptake (which are inhibitory for pain transmission)
SNRI (duloxetine)(17)	Start at 20mg daily, escalate up to 60 mg per day	Same as TCA	Same as TCA
Anticonvulsant (carbamazepine)(167)	Start at 100 mg BID, escalate up to 200-400 mg BID	Neuronal Na+ channels; Central nervous system (CNS) inflammation	Believed to alter pro-inflammatory signaling mechanisms (e.g. GABA, glutamate), and exert a blockade on excitatory sodium channel activity and calcium channel influx
Anticonvulsant (topiramate)(166)	Start at 25 mg BID and escalate weekly up to 200 mg BID within 6 weeks	Same as carbamazepine	Same as carbamazepine
Opioid antagonist (Naltrexone)(166)	1.5-2 mg nightly, escalate to 4-4.5mg	Glial cells; Neuro-inflammation	Inhibition of microglial cell activation in the central nervous system (CNS) via TLR4 modulation, thereby reducing the production of inflammatory and excitatory molecules
Opioid agonist (Tramadol)(17)	50 mg once or twice daily as needed	μ-opioid receptor; serotonin and norepinephrine reuptake at nerve terminals	Weak agonist at the μ-opioid receptor (direct analgesia) and downstream inhibition of noradrenaline and serotonin re-uptake (indirect analgesia)
Topical:
Autologous Serum Tears(168)	Drops applied 4-8 times per day (concentration of 20% most commonly used, 20%-100% available)	Corneal nerves	Contain several growth factors (e.g. nerve growth factor, insulin-like growth factor-1, brain-derived neurotrophic factor, neurotrophin-3) which are believed to promote nerve regeneration as a cause of pain and ocular disease
Topical corticosteroid(17)	No standardized dosing regimen but in general, start at 4 times daily and taper off over a 4- to 12-week period	Ocular inflammation	Prevent inflammation primarily by binding to glucocorticoid receptors and suppressing pro-inflammatory transcription factor NF-kB, thereby regulating expression and proliferation of pro-inflammatory mediators and lymphocytes
Topical anti-inflammatories (cyclosporine, lifitegrast, tacrolimus)(17)	Most often utilized 2 times per day	Ocular inflammation	Prevent activity of T-lymphocytes by modulating calcineurin, which is involved in a cascade that normally leads to pro-inflammatory cytokine release in response to stress
Adjuvant:
Botulinum toxin(154)	In individuals without migraine, 35 units over 7 sites in forehead, adjusted as needed	CNS and meningeal inflammation	Prevents release of pro-inflammatory substrates e.g. cGRP and sP, blocks nociceptive signaling of trigeminal C-fibers, and inhibits ACh release at presynaptic nerve terminals, reducing muscle activity
Transcutaneous electrical stimulation(169)	Optimal usage not yet established; generally used as a 20-minute treatment nightly at least 3 sessions per week	Trigeminal afferent pathway; sympathetic inhibition	Prevention of pain signaling along this the pathway via a combination of peripheral and central mechanisms (169)
Peri-ocular nerve blocks(170)	Dosage should be individualized. Usually prepare a mixture of 3-4 mL of 0.5% bupivacaine mixed with 1 mL of 80 mg/mL methylprednisolone acetate and inject 0.5-1 mL around supra-orbital, supra-trochlear, infraorbital nerves	Na+ channels	Reversible block of sodium channels, involved in pain-related action potential trafficking into the central nervous system; suppression of ectopic firing on sensitized nerves
Parasympathetic ganglion block (sphenopalatine) and/or Sympathetic ganglion block (superior cervical)(171)	Various approaches (transnasal, transoral, supra/infrazygomatic) for sphenopalatine; one study found pain control with delivery of 0.3 mL of 0.5% bupivacaine in patients with migraine; 1-2 mL of 2-4% lidocaine via transnasal route in each nostril for pain.Direct image-guided blockade of superior cervical ganglion with local anesthetic (bupivacaine 0.25%).	Parasympathetic or Sympathetic ganglion	Reversal of parasympathetically mediated ocular vasodilation, increased permeability, and release of pro-inflammatory mediators, causing sensitization of nociceptors; Superior cervical blockades suppress the sensitizing sympathetic efferent drive to nociceptive nerves (in case of sympathetically maintained pain)

MOA=mechanism of action; Ca2+=calcium; α₂δ=alpha 2 delta; TLR4= Toll-like receptor 4; cGRP= calcitonin gene-related peptide; sP=substance P; ACh=acetylcholine; Na+=sodium; TID=three times per day; BID=two times per day; TCA=tri-cyclic antidepressant; SNRI=serotonin and norepinephrine reuptake inhibitor

9). Expert opinion

In summary, our understanding of ocular surface pain is still in evolution. Studies thus far have provided us with countless insights into the pathophysiology and clinical manifestation of pain, especially regarding nociceptive vs. neuropathic origins and mapping of pain signaling from the periphery to the CNS. Studies have also highlighted the pertinent molecular and clinical pharmacology, and based on these findings, we have been able to find a handful of therapies that are effective in patients depending on type of pain, many of which are outlined in this review. Yet, several exciting therapies in the pre-clinical and clinical stages may add to this list in the near future.

It is important to highlight the recent strides we have made in our understanding of ocular surface pain. First, of utmost importance has been the discovery of several novel modalities that are being studied clinically with promising results, which we were able to identify in a long research process that stemmed from mechanistic considerations and extrapolation from other neuropathic conditions across the body. For example, currently utilized therapies like anticonvulsants, antidepressants, and adjuvant botulinum toxin injection were all eventually studied in ocular surface pain based on their observed effects in other pain conditions. Similarly, a finding that has been continually evolving in recent years that we hope will be fully understood within the next decade is our ability to precisely categorize patients with signs of neuropathic pain based on potential area and mechanisms of underlying dysfunction. These kinds of discoveries will greatly aid our ability to start patients on the appropriate type-specific and mechanism-based therapeutic ladder as early as possible.

However, crucial gaps in our current scientific understanding still exist. A major challenge that exists in the field currently is our reliance on clinical diagnosis for patients with signs of neuropathic ocular surface pain. While an array of high specificity/sensitivity in-clinic tools exist for nociceptive pain, this is in stark comparison to the paucity of available in-clinic testing which can help to precisely determine whether there is a dysfunction within the somatosensory system, and if so, where it exists in the body, as well as what might be the underlying mechanisms. This lack of diagnostic capability binds together researchers in the field as it forces reliance on clinical judgment, and thereby brings forth an important target in future research: improved diagnostic testing, which can in turn help us to identify which individuals may be more likely to benefit from specific mechanism-based therapies.

In the meantime, there are several avenues of focus that can help us to further develop our understanding of ocular surface pain and our ability to diagnose and treat it. First, developing our understanding of pain pathways and, as discussed, testing the efficacy of new and emerging diagnostic tools is of utmost importance (Figure 1). Next, studies should continue to examine therapeutic modalities across a spectrum of patients with varying symptom intensities and pain descriptors, in order to hone our understanding of which modalities are likely to be most effective depending on a particular clinical phenotype. Finally, a more distant focus is examining genetic determinants of pain susceptibility and development, as well as gene-modulating therapies and response to these therapies, which may provide significant breakthroughs and help us to better guide preventative techniques prior to surgical procedures (such as refractive surgery). On the clinical side, earlier recognition and integration of counseling is important given the chronicity of pain, particularly in explaining the expected time course of response (with years of therapy often needed) and to discuss approaches to cope with the emotional component of chronic ocular surface pain.

Figure 1. Alternative & Emerging Therapies for Neuropathic Pain.

Schematic detailing the trigeminal pathway of pain and various therapeutic targets along this system. Ascending ocular pain pathways originate from ocular and/or dural blood vessels (black), or peripheral afferents of the corneal surface and periocular skin (green). These signals are all transmitted via the ophthalmic branch (V1) of the trigeminal nerve and converge at the trigeminal ganglia (TG) before piercing the trigeminal nucleus caudalis (TNC) and terminating in the posterior thalamus and primary somatosensory cortex. An overlapping pathway of photophobia which includes the sphenopalatine ganglion shares some of these locations; these parasympathetic fibers enter the same ocular and dural vessels and function to promote vasodilation.

AST=autologous serum tears; AMT=amniotic membrane transplant; TRPV1=transient receptor potential vanilla 1; TENS=Transcutaneous electrical nerve stimulation; TMS=transcranial magnetic stimulation

Article highlights.

1) Dysfunction at several levels of the somatosensory system (peripheral ocular, peripheral periocular, central, or autonomic nerves) can lead to chronic ocular surface pain with a neuropathic component.

2) While a number of first-line agents exist for each type of nervous injury, in patients with neuropathic ocular surface pain who cannot tolerate or are refractory to first-line pain modalities, alternative and adjuvant therapies may be considered.

3) The use of alternative or adjunctive therapies heavily dependent on the likely location of the nervous lesion along the somatosensory pathway (e.g. peripheral, central, autonomic, etc.)

4) While several alternative therapies have been identified, several emerging modalities are currently being studied in the pre-clinical stage and may be readily available for treatment of refractory pain in the near future.

5) A patient-centered, comprehensive approach for multi-modal therapies must be applied over a continuum of time to ensure optimal outcomes in patients with chronic refractory ocular surface pain with a neuropathic component.