Test corneal sensation to identify this lesser-known ocular condition
In this episode of The Spotlight Series Podcast, host Mario Nacinovich speaks with John Sheppard, MD, of Virginia Eye Consultants, about neurotrophic keratitis, a lesser-known condition that requires swift clinician awareness, diagnosis, and referral and/or treatment. They also discuss a “revolutionary” drug to treat NK.
Mario Nacinovich:
Welcome to The Spotlight Series Podcast, where we bring you the latest clinical insights. I’m Mario Nacinovich. Today, we turn our focus to shine some light on a challenging and often overlooked condition affecting the cornea in neurotrophic keratitis or NK. Joining me live here at the American Academy of Ophthalmology (AAO) in Orlando is Dr. John Sheppard, a board-certified ophthalmologist and fellowship-trained corneal specialist, serving as president of Virginia Eye Consultants and a senior leader with CVP Health.
He is a professor of ophthalmology at the Eastern Virginia Medical School and has led over 120 clinical trials in corneal disease and ocular pharmacology. Dr. Sheppard is recognized internationally for his research, clinical leadership, and contributions to advancing care in neurotrophic keratitis as well as complex ocular conditions. He’s also a longtime mentor and a personal friend. I’m happy to have him here live with us in Orlando. Dr. Sheppard, thank you for joining us.
John Sheppard, MD:
Thank you so much.
Mario Nacinovich:
To open our session, could you provide a high-level overview of neurotrophic keratitis? What makes this condition distinct, and why does it present such diagnostic challenges in the clinic?
John Sheppard, MD:
NK is ubiquitous. It comes in many forms. We have a distinct classification using Mackie’s designation from 1995. A Mackie 1 neurotrophic keratitis is punctate keratopathy, which is an almost ubiquitous condition. Mackie 2 is loss of epithelial integrity, a persistent epithelial defect. Mackie 3 is an erosive keratitis, an ulcer that thins the eye, degrading stroma, epithelium, and often, without management, leading to perforation, a very serious condition.
NK, we see all the time in our office, and many examiners, many clinicians, many technicians see NK and don’t realize it’s there. If you don’t know what it is, you won’t see it when you’re examining. How do we look for NK? Well, in its mild variety, the Mackie 1, you’ll see it in diabetics. Diabetics have notoriously deficient nerves throughout their body, including a degradation of the corneal nociceptors, those nerves in the eye that pick up stimuli from the environment, but also create neurotrophins and compounds that aid in the turnover and regular healing of the surface epithelium, and therefore visual clarity.
We see it all the time. It may be subclinical. It may not be a problem, but I would submit to you that all the more severe Mackie 2 and Mackie 3 patients were formerly Mackie 1 patients, and it’s up to us as astute clinicians to decide, decipher, predict which ones of those milder cases will progress to more serious disease. When a patient already has an ulcer, an epithelial defect, or erosive stromal keratitis, the cow is out of the barn. It’s too late to be proactive, but it’s very important to be aggressively reactive, because NK can be a blinding condition.
Ironically, the central epithelium through which we see is the furthest from the limbal arcade, from the nerves in the limbus, and from the progenitor limbal stem cells. It’s that central epithelium that is so frequently involved with neurotrophic keratitis, from mild to severe. Thus, it has the potential to have extremely adverse events on visual acuity, short-term and long-term. A healed serious NK ulcer may have residual stromal haze and epithelial or even stromal irregularities, permanently impacting vision.
It is important for us not only in a referral corneal practice where disasters arrive, but in all forms of eye care to pick up this disease, find the at-risk patient, diabetes, Parkinson’s, herpes simplex, herpes zoster, neurologic surgery, multiple eye, and corneal surgeries. Even repeated intravitreal injections are all patients at risk for neurotrophic keratitis.
Why don’t we recognize this disease? There are many hints. You can look at epithelial healing patterns. You can look at the demographics. You can look at the blink rate. People who don’t blink very often are often neurotrophic, older people, people with surface disease, Parkinsonism. People wearing contact lenses don’t blink as much. There’s not as much stimulus.
You and I, looking at our phones, blink about half as often as we do normally. These near tasks, these digital tasks slow the blink rate and send us on the road to ocular surface disease. With the increased inflammation of surface disease, other cofactors that aid and abet the development of NK, we must take care of these patients before the problems happen.
Mario Nacinovich:
You painted a very interesting picture of the prevalence of this disease and some specific patients, but what are some of the most common reasons or cases that initially present long before they get to you that are sometimes missed or misattributed to other ocular surface conditions?
John Sheppard, MD:
We look at limbal stem cell deficiency. The number 1 cause of limbal stem cell deficiency in the United States is chronic hypoxia from contact lens use. How many? Tens of thousands, tens of millions of patients have worn contact lenses for decades. This is a very common problem, often under-recognized, and we look at patients with dry eye. Maybe severe dry eye. Maybe Sjögren’s disease. We have the dissociation of signs and symptoms.
A horrible-looking cornea with a very, very bad blink rate and neurotrophic disease is at much greater risk than a more routine dry eye patient with extensive keratopathy. These are very common overlaps. It’s not necessarily you’re missing the diagnosis. You’re focusing on a comorbidity, and then, very commonly, hurting corneal nerves through planned ocular surgery.
Corneal transplant, 360 degrees. All the nerves are severed. LASIK flap, you sever all but one clock hour. You look at these patients, and you realize that perhaps they have neurotrophic surface changes, but they don’t complain as much, because the nerves have been cut, and unfortunately, they never grow back normally.
Mario Nacinovich:
Can’t complain about what you can’t feel.
John Sheppard, MD:
That’s exactly right. Those patients are the ones who present late, and I can’t tell you how many times on a Friday afternoon we see a 75-year-old diabetic person who, in the past, had RK, PRK, LASIK, cataract surgery, iatrogenic damage to the nerves, and they have a massive ulcer with a quarter-diameter hypopyon in their eye, and their complaint is blurred vision. A younger patient, someone without NK, would be in extreme pain rolling on the floor and present a week earlier. This is an extreme manifestation, but this is a very common presentation in a referral academic corneal practice.
Mario Nacinovich:
For the general eye care practitioner and for some specialists, and let’s even say for the general practitioner, why is it so important that we increase awareness and index of suspicion, NK, so much earlier? That diabetic patient is a good example.
John Sheppard, MD:
It’s important for the overall health of the population and avoiding disasters, and the awareness is important because you can get a feel from the history and the overall exam about the presence of NK, but you need to test corneal sensation to essentially establish the diagnosis. Not everybody does this. Nobody wants to do this. It takes time, and it’s very subjective. Plus, you must not anesthetize an eye prior to testing corneal sensation.
What do we do on all of our patients? We check their pressure. If you have a patient in your chair with a pressure on the chart, it’s too late to check corneal sensation. We have to make a separate visit. We have a no-touch visit to establish that diagnosis. You can touch the cornea. You can touch it with a cotton wisp, dental floss, little wisp of tissue paper, but those are contact methods.
There’s also a Cochet-Bonnet esthesiometer. It’s the industry standard. It’s been used for decades and is important in establishing at least a cursory level of NK that has decreased sensation in patients who are enrolling in very strictly regulated, very strictly included and excluded clinical trials. We need some type of parameter.
Unfortunately, the Cochet-Bonnet is a 6-cm nylon wire that you poke right into the middle of the cornea, right where the epithelium is most critical, and you keep decreasing the length. Therefore, the rigidity increases and causes more pressure on the cornea to determine that level of zero to 6 corneal sensation.
I find it somewhat barbaric in that we’re assessing that most critical part of the cornea. There is now a esthesiometer that is noncontact. It’s mounted on the slit lamp and gives us millibar readings from about 2 to 14, and this is the normal range of corneal sensation. It’s made by a company called Brill Engines out of Spain, and we’ve been using it in our office for 2 years. They are readily available now, and I think it should be part of everyone’s precorneal and cataract surgery regimen, because we are severing nerves, and we may be triggering that fall over the cliff of lingering NK to manifest NK.
That would be, of course, on the back of the surgeon, because they made the incision, and we have to prepare our patients for that. If they have preexisting markedly decreased corneal sensation and you’re incising the cornea, you’re just inviting complications. The number 1 thing that causes, in our practice, NK is herpetic disease. You can prophylax with a very safe medicine, oral valaciclovir, when you suspect in particular unilateral or asymmetric NK, which would implicate zoster or simplex above and beyond any other etiology.
We challenge our doctors to check for sensation and go to that extra effort to obtain some form of subjective reading, a replicative procedure and process where they use the same type of tactile stimulus. I use cotton wisp, and then we graduate in select offices, and I hope more and more offices, with the noncontact Brill esthesiometer.
Mario Nacinovich:
Well, it sounds like we’ve gone from almost pure barbarism, using your words, in diagnosing corneal sensitivity to something a little bit more humane, and I think the key to it is certainly getting that methodology beyond the clinical protocol and into the daily protocol within the offices to make that work. In your practice in specific, how do you integrate the use of corneal fluorescein staining or lissamine green to assess the extent and pattern of corneal epithelial damage? Are there established staging systems you rely on to guide that workup and management?
John Sheppard, MD:
You can certainly stage corneal fluorescein staining, and that has been done repeatedly. We have an Oxford scale, an NIH scale. We use those over and over and over again for ocular surface disease clinical trials, and they’re quite standardized. We find that it’s a little bit easier to have a more subjective scale in the office setting, and most people grade staining 0 to 4+. The key is, does it stain in the central cornea?
You can use a very rigid clinical research scale or a more practical scale in your examination. It’s important to be consistent. It’s essential to use fluorescein in virtually every patient you see if you’re assessing the cornea. I don’t know how you do it without that. A caveat is though that the fluorescein is most visible at about 30 seconds to 3 minutes.
Mario Nacinovich:
Time is critical.
John Sheppard, MD:
If you look too soon, you don’t see it. If you wait too long, it’s gone. What I do is I put in the fluorescein. I get out my 90D. I look at the optic nerves. It’s about 30 seconds, and then I’m ready to look at the fluorescein staining.
Mario Nacinovich:
That’s one potential pitfall, timing.
John Sheppard, MD:
Correct.
Mario Nacinovich:
Are there any other pitfalls in terms of diagnosis?
John Sheppard, MD:
Being in a rush. My scribe hands me the fluorescein. I put it in the eye. It makes me feel more efficient. It doesn’t take that long. It’s clearly a bare minimum part of our corneal external exam.
Mario Nacinovich:
We’ve talked about timely diagnosis, timely referral, if that’s appropriate and warranted, but let’s talk about early intervention, and guidelines certainly continue to emphasize timely treatment to minimize irreversible vision loss. You recommend that as well. But based on your experience, what are the most crucial factors that enable that early diagnosis and prompt initiation of therapy for NK?
John Sheppard, MD:
The critical factors are clinician recognition and patient awareness. The patient has to be educated. It’s not hard to tell a patient, “You have deficient nerves, and you can have healing problems or ulceration.” I mean, they get that. They don’t get that it does hurt, and that needs to be made clear. But having this early recognition allows the doctor to, one, aggressively manage any other preexisting ocular surface disease, and 2, be prepared for a surgical intervention where complications might occur.
Those are the 2 key points. Obviously, the clinician has to be aware of interventional strategies. Certainly, exquisitely precise ocular surface, dry eye management is essential to preventing complications, and the therapeutic regimens overlap for sure, especially for stage 1 Mackie. That being said, if you’ve given your best 1-2-3 punch with procedures and drops and plugs and serum tears and everything else you can think of, oral doxycycline or minocycline, lid scrubs at home, and a Demodex treatment, and they still have a bad cornea, something is going on.
That’s when you would start thinking about, “Well, maybe this NK is playing a major role. Maybe the limbal stem cell deficient is playing a major role and the 2 overlap.” It’s your next level of preventative care. You think about truly addressing the NK. If I get a reading of 14 on my Brill, or I have to go down to 2 on the Cochet-Bonnet, heaven forbid, or if I’m poking them with a Q-tip and nothing happens, there’s no blink, then we got a problem here, and I want to treat that before I’m putting my keratome or my trephine on that eye.
Mario Nacinovich:
Knowing that you’ve seen the absolute worst of the worst, can you talk a little bit about the potential long-term consequences of delayed recognition, that’s one step, but also the risk of corneal thinning, ulceration, perforation much later?
John Sheppard, MD:
Let’s look at the worst-case scenario, herpes zoster ophthalmicus. It’s easy to recognize a patient has scabs and scars all over their trigeminal 1 cutaneous distribution. Then what happens? Well, maybe they’re okay. Maybe they didn’t get uveitis. Maybe it’s their nondominant eye and they’re not complaining as much, but they’re just sitting ducks.
If that patient needs to have a cataract operation, one, I would really manage them aggressively, and I put them back on antivirals. Two, I would make them aware of the potential pitfalls and, “You really need this surgery.” Three, if I’m going to manage astigmatism, I’m not going to put in limbal relaxing incisions, because they cut even more nerves. You’re just waiting for a problem. Then four, the most difficult scenario is when there’s a permanent, legally blinding corneal opacity, because zoster patients do very poorly with transplants.
I mean, you can sew it in. The surgery goes great, and then they develop erosions. They develop thinning at the graft-host interface. They don’t epithelialize, and then you put in a tarsorrhaphy, and it’s just tragic. I tend to very conservatively approach a penetrating keratoplasty in a patient who truly needs one for visual rehabilitation, if they’ve had ophthalmic zoster. Those are challenges. They become lifelong patients, friends, frequent fliers, and they get to know everybody in the office. That’s fine, but I can think of better social endeavors.
Mario Nacinovich:
It’s certainly a better beginning to a relationship than that.
John Sheppard, MD:
Yes.
Mario Nacinovich:
I know we talked about this years ago when we were talking about fungal keratitis, but how do you educate referring providers? We’ve got a great listening audience here that can learn from this about the urgency of specialist referral once NK specifically is suspected.
John Sheppard, MD:
When a patient develops Mackie 2 or Mackie 3 with obvious risk factors or no identifiable risk factors, the cornea specialist is delighted to intervene. That’s what we do. We try to save eyes, and we have a system where someone can call and get the patient in that day or the next day. We have interventions that we are very aggressive with in advanced disease to save vision and to prevent transplants, to prevent that scarring that’s visually significant.
I think there’s a 1-2-3 punch available to actually regenerate corneal nerves. Formerly, we had a number of ways to produce an upgraded stimulus to the trigeminal afferents, and these are the corneal and conjunctival and nasal nociceptors that provide increased production of the tripartite tear film, and one of the things that we used was intranasal neurostimulation with a random voltage, and we felt that there was a stimulus there on regular use to increase regeneration of corneal nerves.
Whenever you regenerate corneal nerves, they don’t quite look the same as they used to. We’ve looked at other approaches, and we have much more clinical evidence of amniotic membranes producing a regenerative effect on corneal nerves by confocal microscopy, and I find that that’s a very effective way to begin the therapeutic journey in a patient with severe surface NK disease.
Finally, we have a therapeutic, a pharmacologic intervention that has changed the entire way we approach this disease, and that is cenegermin-bkbj. It’s recombinant human nerve growth factor, known as Oxervate, and we apply it to the patient’s eye, hopefully before the disaster, but certainly during the disaster if that’s the presentation, and it’s used, with motivation and counseling, 6 times a day for 8 weeks.
This is a revolutionary drug. It’s the product of 60 years of research, because nerve growth factor was identified shortly after World War II, and Rita Levi-Montalcini, the PhD scientist who identified this, won the Nobel Prize 30 years later; 30 years after that, a drug was approved. That’s a very slow bench to bedside, but we have this cenegermin-bkbj biologic, the first topical biologic for the eye, available to reverse corneal nerve changes, to regenerate nerves, to upregulate epithelialization and epithelial activity, and to promote direct effects on the lacrimal gland to upgrade tear production.
This is an endogenous factor, and it’s applied in 10,000-fold the normal concentration in tears. By producing this biologic effect, we hopefully restore normal homeostasis, normal function of that efferent unit in the trigeminal nerve, and prevent further scarring, prevent perforation, and improve the prognosis perhaps for an eventual ocular surgery.
In these patients, who I tend to identify, I really want to be careful not to be the cause. I don’t want my surgery to create a disaster, and we’ve seen that repeatedly. We get one or 2 referrals every quarter of a patient who apparently had nothing, had routine cataract surgery. We even have one with an SLT, one with a lamellar keratectomy, and then everything falls apart. At week 1, they have a massive ulcer. They’re seeing 20/200, and now we start the cenegermin.
It would have been far better to identify that, do a sensation, “Oh my goodness, this eye is great.” The operative eye has horrible sensation, and look really carefully. Maybe there’s a corneal opacity somewhere there. You put them on antivirals. You put them on cenegermin-bkbj, and then you wait. You can wait for your cataract surgery. You wait 6 weeks. You retest. You remeasure. The surface should look much better now, and you get more accurate biometry. It’s all about prevention, when you have the opportunity.
Mario Nacinovich:
How does that approach align, or how is that not aligned to current guidelines?
John Sheppard, MD:
I think it aligns to current guidelines. Cenegermin-bkbj is approved for stage 1, 2, and 3 Mackie NK, and guidelines indicate that we manage ocular surface disease and that we have particular concern for patients who’ve had previous infections. Herpetic infections are very common. Herpes simplex is a bad actor. Not quite as bad an actor as herpes zoster, but it’s a bad actor also, and that’s where we’ve seen all of these inordinately surprising referrals for severe, central, corneal, erosive NK disease after elective surgery, and many of these have been in premium IOLs too. That’s a major source of disappointment.
Mario Nacinovich:
You mentioned a plethora of medical management. You’ve talked about artificial tears, autologous serum. Amniotic membrane you mentioned as well. When do you proceed to the cenegermin-bkbj in accordance with current guidelines or accordance to John Sheppard guidelines?
John Sheppard, MD:
If I’m worried about a patient, if they are anticipating surgery, if they’re not healing normally with standard therapy, and if, heaven forbid, they’ve had previous complications, my threshold is very low.
Mario Nacinovich:
Your patients are usually very advanced as well.
John Sheppard, MD:
Exactly. It’s easy for me to say, because I have a high pathology corneal referral practice. In a comprehensive optometric or ophthalmologic practice, we still have a sense of the urgency and the severity potential of these cases, and you may see one once a month. Not every day, like I do, but they’re still out there, and you don’t want to ignore those patients.
Mario Nacinovich:
Can I get your thoughts on surgical outcomes following effective NK management? How does successful medical management of NK affect an individual patient’s candidacy and, ultimately, your surgical outcomes?
John Sheppard, MD:
That’s a great question, because you want to have the patient tuned up prior to their procedure. Tuning up a dry eye can take 3 months. You need to do that. Restoring the nerves in a clearly deficient patient is going to take eight weeks. There’s an important time level there. I find that if I do a permanent punctal occlusion with cautery in these really severely dry patients, it takes a couple of months for that to heal.
You really have to be prepared. If it’s a routine procedure, you’re probably okay. Glaucoma procedures are somewhat less deleterious to the cornea and the corneal nociceptors than corneal incisions with cataract and, heaven forbid, penetrating keratoplasty. In a patient with a decent prognosis who’s had maximal management, I’ll proceed with a cataract operation. I’m very hesitant though to recommend a multifocal lens in those patients.
I think with reliable, repeatable, verifiable biometric measurements and consistent astigmatic measurements in three different methodologies, that I am very comfortable giving them a toric lens, and maybe even an extended depth of focus, but I would essentially never give that patient a multifocal or a diffractive lens. That’s important, and it’s important for the patient’s informed consent process as well.
In terms of penetrating keratoplasty, I pray a lot when those are performed, and we often are presented with patients who have one eye, and it’s perforated. It’s scarred. They see nothing. Everything is on the line, and you have to make a really informed decision, and they will have a tarsorrhaphy at the time of surgery. I mean, you throw the books at these patients. You get them their full course of cenegermin-bkbj prior to the surgery. When their surface is as stable as it’s going to be, you judiciously consent and proceed.
Mario Nacinovich:
I think you made a number of interesting points. One that I just want to emphasize here for our listening audience, it’s don’t overlook this preoperatively, because when it’s postop, it perhaps may be too late.
John Sheppard, MD:
Or it’s certainly a bigger challenge postoperatively.
Mario Nacinovich:
I think we could agree on that.
John Sheppard, MD:
Yes. The prognosis is more guarded.
Mario Nacinovich:
I think another big challenge as we conclude our conversation here at the American Academy of Ophthalmology in Orlando, what are the most pressing unmet needs, or let’s call them priorities, in NK education and research right now?
John Sheppard, MD:
I think as far as research goes, there’s a lot of interest in the pharmaceutical world in producing this regenerative effect on corneal nerves, and it’s exciting to see that the success story of cenegermin has led to further potential innovations. In terms of education, it’s all about getting in front of the young audience.
The young audience is eager to absorb information, eager to pick up more unusual diagnoses, eager to bring the best possible therapeutic options to their patients. You educate someone at age 25. You got 3 decades of memory there. You hit somebody 65, and they’re more concerned about having a nice meal. I reflect on that bias personally.
Mario Nacinovich:
For the more mature practitioner or for that young practitioner, any resources, registries, referral networks that you would personally recommend they seek in terms of guidance, collaborative care for their patients, for themselves?
John Sheppard, MD:
There’s always a mixed blessing of patient self-education online, because many times, they pick an intervention that’s completely inappropriate for their condition. I had a lady last week who had Salzmann’s nodular degeneration, and she decided that she wanted a phototherapeutic keratectomy because she read about it on the internet.
Mario Nacinovich:
Dr. Google informed her.
John Sheppard, MD:
Yeah. Exactly. It sounds good. Patient education is important. It needs to be directed by the clinician, “This is what you need to look up, and this is the treatment you’re going to have.” The AAO has an amazing using source and patient explanation profile, and it’s plain English. At the same time, the AAO has a lot of practice management recommendations for even the most advanced clinicians. As a resource for information, I don’t think anybody would argue that the American Academy of Ophthalmology has the most comprehensive, accurate, peer-reviewed sources on the planet.
Mario Nacinovich:
Excellent. Well, thank you, Dr. Sheppard, for sharing your expertise on this vital topic of NK. For more resources and references on neurotrophic keratitis, please see our episode notes, and join us next time on The Spotlight Series for another deep dive into emerging clinical challenges. I’m Mario Nacinovich, and this is Dr. John Sheppard.
John Sheppard, MD:
Thank you.
Mario Nacinovich:
Pleasure.
