DRY EYE DISEASE SPECIAL SERIES: PAIN

A NOD to a painful, puzzling diagnosis

Nonobvious dry eye disease may have a neuropathic basis, so look for symptoms; signs might prove elusive.

By Richard A. Adler, MD

As ophthalmologists, we see its effects nearly every day. We know about it, but rarely discuss it with fellow physicians. We diagnose it, but rarely treat it. The Schirmer’s score is normal, tear break-up appears normal, there’s no corneal staining, no significant lid disease, and the osmolarity test is within normal limits. However, the patient is miserable, complaining of persistent ocular surface disease symptoms. How can this be? Can we still make a diagnosis of dry eye disease (DED) with normal tear parameters?

For years, the traditional view of DED required a neat formula wherein DED had to equal the ocular surface minus tears. We have gradually developed a more sophisticated perspective on this equation and have correctly added meibomian gland dysfunction to this formula. Yet, there is a large contingent of unhappy patients who experience neither tear deficiency nor meibomian gland dysfunction and fit neither category of aqueous-deficient nor evaporative DED. Our well-intended descriptors such as “Dysfunctional Tear Syndrome”, “Chronic Ocular Surface Disease”, and “Keratoconjunctivitis Sicca” seem to give way to the proverbial “Normal Eye Exam” leaving the patient dissatisfied and leaving the doctor eager to attend the next patient on the schedule.

As this article will address, a mounting body of evidence is starting to provide us with an answer and, hopefully, a solution. The traditional DED equations need new factors and subsequent recalculation, because not all patients present with abnormal tear parameters. An underrecognized, but potentially widely prevalent form of dry eye, dry eye disease (NOD), might explain the dissatisfaction of both physicians and patients concerning the treatment of ocular surface symptoms. While many patients with DED do have deficient tear parameters and benefit from traditional therapies, NOD patients may instead have neuropathic eye pain. These patients often do not respond to traditional treatments no matter how many tears, plugs, or procedures we prescribe their way.

Only symptoms?

The concern that logically follows from this discussion is whether one can truly have DED with only symptoms and no signs. We are generally comfortable diagnosing DED in patients with signs but no symptoms, but can the converse be true? Does the well-established truism that signs and symptoms do not correlate well in DED allow for this possibility? Perhaps we can look toward the FDA process of evaluating potential DED drugs for some guidance on this. For a new drug to receive FDA approval for a dry eye indication, it must meet co-primary endpoints of improvement in signs and symptoms. However, the FDA does not require that both endpoints be met in a single trial; meeting only one of the two endpoints in separate trials can sufficiently prove its efficacy.

As further evidence for the possibility of a diagnosis based exclusively on symptoms, we can also look toward our experience with other medical conditions. For example, the presence of symptoms without signs in fibromyalgia neatly parallels the similar disconnect found in DED. The current diagnostic criteria for fibromyalgia however, do not require any objective testing.¹ Similarly, we are comfortable making the diagnosis of migraine headache based solely upon symptoms and even go so far as to categorize migraine by symptomatology alone; migraineurs either have migraine with or without aura. However, our current categorization of DED relies largely upon the binary choice of aqueous-deficient or evaporative signs and, in doing so, potentially misses a more meaningful categorization based on symptoms.

DED diagnosis challenges

The notoriously diverse and protean symptoms of DED partly explain the challenges of DED diagnosis and management. From visual symptoms such as blur and glare to the widely varied experiences of sandiness to shooting pain, the specter of scattered symptoms perhaps affords us the opportunity to better organize and manage our different dry eye patient types. In addition to the traditional dichotomy of evaporative and aqueous-deficient DED, imagine if we divided our dry eye patients by symptoms as well. Could we possibly better tailor our treatment plans?

Does this clear cornea indicate a healthy, DED-free eye? Don’t yet rule out nonobvious dry eye disease.

Ocular pain, like other forms of pain, can be thought of as occurring in two broad categories: nociceptive pain and neuropathic pain, both present in DED. Nociceptive pain results from tissue damage on the ocular surface (e.g., punctate keratitis, mechanical abrasions, thermal or chemical injury) and can be characterized by transient pain. In contrast, neuropathic pain is chronic and results not from direct damage to surrounding tissue, but rather from damage to the nerves themselves. There is often a persistent, phenotypic change in the nerve that explains the sustained hypersensitivity, but the tissue itself appears undamaged; the cornea is clear. The slit lamp exam is normal. The co-occurrence of nerve damage and the absence of corneal signs form the basis of NOD.

Neuropathic pain

Neuropathic eye pain may present as dysesthesia (spontaneous shooting pain), hyperalgesia (amplification of normal pain), or as allodynia (pain in response to typically nonpainful stimuli, such as wind or cold). Our current screening surveys, such as the Ocular Surface Disease Index score, do not segregate neuropathic from non-neuropathic symptoms; in fact, they even lump together visual function symptoms and pain symptoms into a single composite score.

However, treatment of neuropathic dry eye pain may require different strategies from non-neuropathic pain, thus obligating us to better segregate these symptom types.

The neuropathic origins of NOD may further be classified into peripheral and central causes. Peripheral neuropathic pain results from the hyperexcitability of peripheral nerve endings due to increased activity of nerve terminal sodium channels.² While numerous types of peripheral nerve fibers exist, those designed to monitor temperature changes on the cornea appear to be in a class involved in neuropathic pain.³ Normally, these thermo-sensitive nerves provide a remarkably precise real-time mechanism to maintain corneal tear thickness. Corneal tear thickness is essential to the refractive power of the eye’s optical system and, as such, a real-time monitoring system would be evolutionary advantageous.

However, what happens when this evolutionary advantage misfires? What happens in a hyper-sensitized state? In nonsensitized states, these nerves fire when the interblink ocular surface drops 4-5⁰C, alerting the brain to the potentially deleterious effects of ocular surface desiccation. In the sensitized state, these nerves inappropriately fire at only a 2⁰C drop, creating the experience of a desiccating threat in the absence of an actual one. To the patient, this may be experienced as a hypersensitivity to wind or cold or, could possibly result in spontaneous shooting pain. What signs should we expect on slit lamp exam corresponding to these symptoms? In many cases, there will be none at all.

However, even without obvious pathology on slit lamp exam, pathology can still exist. Recent investigations point to the role of inflammation in the induction and maintenance of neuropathic pain.4 Neuro-inflammation may be induced and maintained by a diverse group of inflammatory mediators such as IL-1, IL-6, TNF-α, and MMP-9; neuro-inflammation does not require slit-lamp signs to be present. The peripheral sensitization induced by inflammation occurs due to resultant changes in the distribution and function of voltage-gated channels in nerve terminals. For example, inflammation may result in an increased insertion of sodium channels into nerve terminal plasma membranes along with an increased propensity of these sodium-gated channels to open. A corresponding decrease in the number of potassium-gated channels occurs as well. The combined effect of increased depolarizing sodium channels and decreased hyperpolarizing potassium channels can account for the peripheral sensitization seen in NOD.⁴

Central mechanisms

Descending inhibitory central pathways, originating from multiple areas of the somatosensory cortex and midbrain, synapse in peripheral ganglia. At this location, the descending central pathways typically dampen the pain response originating from peripheral nerves. In certain pathological states, however, these descending inhibitory pathways are themselves inhibited, leading to increased subjective pain awareness.^5,6

One of these pathological states may be DED. This possibility emerged in a remarkable study from 2013, in which investigators confirmed a link between DED and pain sensitivity.⁷ What made these results remarkable was that the pain sensitivity investigated was not on the eye, but on the forearm. Specifically, subjects with and without DED were exposed to noxious thermal stimulation on their forearms. The DED group had a statistically significant decreased threshold to pain as compared to the control group without DED. This conclusion points decisively toward a unifying central mechanism for the enhanced sensitivity to pain in DED patients.⁷ Furthermore, it provides yet another explanation for NOD — this time, from the central nervous system.

Conclusion

An improved awareness of NOD and of its pathophysiological mechanisms can lead to improved management of these difficult patients. Ophthalmologists need no longer succumb to the appearance of a normal eye exam. No longer must the “pain without stain” patient have no basis for her symptoms nor should her lack of signs leave her devoid of diagnosis. Ophthalmologists once confined to the binary solutions of aqueous-deficient or evaporative DED have new diagnostic options. They can now sort DED by differing symptoms, not only by differing signs, while patients can be categorized by their presenting signs and symptoms as well. The neuropathic dry eye patient can be identified separately from the non-neuropathic one; this is the first step in improving our management of NOD.

Early research with GABA-analogues, such as gabapentin, has led to the intriguing possibility of these drugs’ use in neuropathic eye pain, particularly following photorefractive surgery.⁸ Additionally, research involving diclofenac suggests it has a unique anti-allodynic and antihyperalgesic effect due to its specific impact on hyperpolarizing postassium-gated channels.⁹ However, we need not wait for progress in the lab in order to make progress in the office. Perhaps the greatest benefit we can offer our patients at this point may simply be an explanation for their pain. We must never underestimate the enormous impact that an answer can have to a patient with a question. And while we as ophthalmologists may marvel at the prospect of unhitching DED from simple tear dysfunction, our patients will no doubt marvel at the prospect of not being told that their eye exam is simply normal. Managing the normal eye exam may not be as easy as it seems. It is time to admit that the absence of signs is not necessarily a sign of absence. OM

REFERENCES

About the Author
	Dr. Adler is the Director of Ophthalmology at the Belcara Health Premier Multi-Specialty Center in Baltimore, Md. He is an assistant professor of ophthalmology at the Wilmer Eye Institute. Dr. Adler received his medical degree at the Johns Hopkins School of Medicine and completed both his ophthalmology residency and cornea fellowship at the Wilmer Eye Institute.