By: Shanu Subbiah*

Corneal melting syndromes are an uncommon group of conditions characterised by keratolysis often in response to iatrogenic and infective agents but most commonly the condition is autoimmune in origin. They can lead to rapid and permanent loss of vision in the affected eye.

Autoimmune corneal melt

Peripheral ulcerative keratitis (PUK) is a form of ocular inflammation that involves the outer portions of the cornea and may be associated with autoimmune conditions. By far the commonest association of PUK is rheumatoid arthritis (Figure 1). It is also seen in a variety of other conditions including Wegener’s granulomatosis, systemic lupus erythromatous, relapsing polychondritis, polyarteritis nodosa and inflammatory diseases such as roscea (Figure 2).  PUK has also been reported following surgical procedures such as cataract surgery and in association with infective keratitis. Mooren’s ulcer is a rare form of PUK.

Figure 1 – Rheumatoid corneal melt (photograph courtesy of Professor Charles McGhee)

Figure 2 – Rosacea associated corneal melt (photograph courtesy of Professor Charles McGhee)

Severe forms of PUK can lead to the progressive marginal corneal thinning and perforation of the cornea. 

The incidence of PUK is approximately 3 per million per year. Despite the fact that it may be expected that there would be a female preponderance it appears that both sexes are affected equally.1 

The clinical manifestations of PUK include ocular irritation, variable pain and redness, photophobia and corneal opacity to the more serious features of decreased visual acuity, corneal perforation and subsequent blindness. Clinically PUK typically presents as a crescent-shaped area of ulceration that occurs within 2-mm of the limbus. The corneal stroma is thinned with absence of the overlying epithelium. There are varying degrees of associated stromal cellular infiltration often depending on the cause or systemic association of the ulceration. Rheumatoid PUK can be preceded by sclerosing keratitis characterised by peripheral corneal thickening and opacification (which can look remarkably similar to marginal keratitis). There is often a history of connective tissue disease and exacerbations in systemic disease frequently correlate with PUK.  There is usually inflammation of the adjacent conjunctiva and sclera. In a retrospective review of scleritis patients Tauber et al noted 14% had associated PUK, conversely 36% of patient with PUK had scleritis present.2 PUK is most often seen in patients with necrotising scleritis and is bilateral in up to 40% of cases.2, 3 

Mooren’s ulcer was typically characterised as existing in two forms: the limited form is typically seen as unilateral disease in middle aged/ elderly Caucasians and is fairly responsive to treatment; more aggressive bilateral disease is seen in younger pigmented individuals and is less responsive to treatment4. Lewallen and Courtwright in 1990 however found in their series that 43% of older patients had bilateral disease whilst bilateral disease was only present in a third of patients younger than 35 years-of-age and that more whites than black patients were affected with bilateral disease.5

Unlike PUK, on examination there is no perilimbal clear zone and no associated scleritis. The leading edge of the ulceration is grey and undermines the epithelium and advances centrally and circumferentially with underlying stromal thinning (Figure 3). It is almost universally painful. By definition the cause of Mooren’s ulcer are unknown (PUK is due to known disease e.g. rheumatoid and rosacea). Precipitating factors are thought to include trauma and parasitic infection. An association with hepatitis C has been postulated as many of these patients have responded to interferon therapy.6 The exact pathogenesis of Mooren’s ulcer remains unclear. Studies correlating the human leukcocyte antigen system (specifically HLA-DR17) and antibodies to auto-antigens in the corneal stroma (known as cornea-associated antigen) support an autoimmune basis for the disease.7, 8 

Figure 3 – Mooren’s Ulcer (photograph courtesy of Professor Charles McGhee)

PUK is often the primary manifestation of an underlying vasculitis.1, 2, 8 Tauber et al reported in a retrospective study from a tertiary referral center an underlying systemic disease was found 25 of 47 patients.2 Although most systemic conditions were known prior to the time of diagnosis, approximately 25% of the predisposing conditions were undiagnosed. Thus the importance of a careful medical history, comprehensive systemic review, and appropriate laboratory testing must be emphasized in newly diagnosed PUK. Rheumatoid arthritis was found in 34% of patients as the most commonly associated disease.2 PUK associated with rheumatoid is well reported to herald systemic vasculitis (up to 50%) and carries a high mortality rate if not aggressively treated.8

The aetiology of PUK is still uncertain but both T cell and antibody mediated pathways have been implicated in the disease process. Biopsy of the adjacent conjunctiva (especially in the rheumatoid model of the disease) suggests immune complex deposition at the limbus causes an obliterative vasculitis. This results in corneal inflammation with activation of the complement system, subsequent increase in cytokine production with recruitment of neutrophils and macrophages that secrete proteases and collagenases and thus corneal melt. Studies have shown that increased cytokines may stimulate an increase in matrix metalloproteinase 1 (MMP-1) by keratocytes as well as a reduction in tissue inhibitor of matrix metalloproteinase 1 (TIMP‐1), resulting in rapid corneal keratolysis.9

PUK is associated with significant eye and systemic morbidity. In one retrospective review of 24 patients with scleritis associated PUK, all 24 patients had impending corneal perforation, 67% had an associated anterior uveitis, and 83% had decreased vision (defined as a decrease in visual acuity of 2 or more Snellen lines at the end of follow up or visual acuity of 20/80 or worse at presentation3). In another retrospective review of 47 patients with PUK, 34% had impending or frank corneal perforations (defined as peripheral corneal thinning of 75-100%), 47% required a tectonic graft procedure, 9% had an associated anterior uveitis, and 43% had visual acuity of 20/400 or worse at presentation.2 

Iatrogenic corneal melt

By far the commonest groups of drugs implicated in corneal melting syndromes are the non-steroidal anti-inflammatories (NSAIDS) and topical corticosteroids. NSIADS are commonly used for pain and inflammatory control following cataract and refractive surgery. With regards to topical NSAIDS particularly there are multiple reports in published literature of corneal toxicity, ulceration and perforation.11, 12 

Cyclo-oxygenase inhibitors such as diclofenac block the conversion of arachidonic acid to prostaglandins that are important mediators of inflammation. Excess arachidonic acid is also metabolized via a second pathway catalysed by lipoxygenase, resulting in the production of leukotrienes (pro-inflammatory mediators). The resulting accumulation of leukotrienes and their intermediate products are potent chemo-attractants for neutrophils as well as a factor for degranulation of neutrophils. The granules released by neutrophils during the inflammatory response contain collagenase as well as other hydrolytic enzymes. When released in the cornea, the accumulation of collagenase may potentiate corneal melting13. Laboratory data suggests that NSAIDs may also inhibit keratocyte proliferation in-vitro; thereby decreasing wound healing and increasing the risk of corneal ulceration. In a model comparing diclofenac to topical corticosteroids, NSAIDs were found to have a more potent negative effect on wound healing.14

NSAIDs may also have an effect on MMPs. These have been found to be elevated in ulcerated and melting corneas. There is evidence that MMPs are involved in stromal degradation as well as other processes that occur during corneal ulceration and repair, such as re-epithelialisation and tissue remodeling.15

The evidence that NSAIDs play a role in corneal melting are often seen in the use of the agents in a post operative setting where corneal trauma has been induced as well as in concomitant use with topical steroid or when there is underlying collagen vascular disease. It does appear that the presence of one of the above factors is often a requisite for corneal melt.13 

Other rare causes of iatrogenic melt have been described following cataract surgery, keratoprosthesis implantation, keratoplasty, pterygium surgery (Figure 4) and use of topical anti-VEGFs. It is once again worth noting in many of the above cases there were underlying collagen vascular disease (implying an autoimmune pre-disposition to melting), concurrent steroid use or underlying corneal pathology.

Figure 4 – Mitomycin associated corneal melt following pterygium treated with lamellar keratoplasty, amniotic membrane and conjunctival flap (photograph courtesy of Professor Charles McGhee)

Ocular and systemic infections may also cause or be associated with corneal melt. Microbial pathogens implicated in the aetiology of melt include bacteria, spirochetes, Mycobacteria, viruses (herpes simplex virus, varicella zoster virus), acanthameoba, and fungi.16

Extensive stromal inflammation from the immune response to microbial pathogens leads to proteolytic stromal degradation and liquifactive necrosis of the cornea. 

Treatment of corneal melts

The treatment of corneal melts is determined by the severity of findings within the cornea and the extent of any associated systemic disease. 

Goals of treatment are to reduce inflammation (ocular and potential systemic), promote epithelial healing, minimise stromal loss and treat or remove any underlying inciting pathology. 

Systemic corticosteroids are the mainstay of therapy for autoimmune corneal melts, although only after any potential microbial aetiology is excluded. Pulsed treatment is often used to good effect (e.g. 1g methylprednisolone a day for three days) often followed by oral steroid.17 Because of the potential for further proteolysis topical corticosteroids are not an appropriate therapy for corneal melt especially in the presence of a disrupted epithelium. 

Preservative free lubricants and hyaluronate are useful in managing the ocular surface and oral tetracyclines may have a beneficial effect in reducing further stromal loss by inhibiting protease activity. A number of topical collagenase inhibitors such as acetylcysteine and sodium citrate may also be of value.

Alkylating agents such as cyclophosphamide and other steroid-sparing immunomodulators (e.g. Methotrexate, azathioprine, mycophenolate mofetil, and the biologic agents) may be used in conjunction with corticosteroids in cases of PUK with imminent danger of corneal perforation and in cases associated with a systemic vasculitis. The vast majority of immunomodulators however take up to 6-8 weeks to achieve therapeutic levels.  As with treatment of systemic autoimmune diseases, steroid-sparing agents are also added when an initial course of high-dose corticosteroid fails to control the disease, when control cannot be maintained with less than 10 mg/day of prednisone, and when there are significant concerns related to the well known adverse effects of corticosteroids.17 

The treatment of autoimmune corneal melts with immunosuppressive medications has the potential for treatment-related morbidity and mortality (including steroid induced bone loss and organ toxicity). Patients require careful monitoring, with frequent clinic visits and blood monitoring. Often treatment will be in conjunction with a rheumatologist.

PUK that has failed to respond to medical treatment may respond to conjunctival resection presumably by eliminating the source of inflammatory cells and collagenases.2 

In cases where perforation has occurred or is imminent techtonic procedures are necessary to maintain globe integrity. These include the use of cyanoacrylate glue, bandage contact lens, amniotic membrane and corneo-scleral grafting.

Conclusion

Corneal melt continues to prove be a complex management problem. Whilst the condition is thankfully rare prompt recognition of the clinical signs with rapid medical therapy can reduce the progression of the condition to globe perforation with consequent severe visual loss. Early diagnosis decreases the need for demanding surgery as well as identifying significant associated systemic co-morbidity.

References

1. McKibbin M, Isaacs JD, Morrell AJ. Incidence of corneal melting in association with systemic disease in the Yorkshire Region, 1995-7. Br J Ophthalmol. 1999;83(8):941–3.
2. Tauber J, Sainz de la Maza M, Hoang-Xuan T, Foster CS. An analysis of therapeutic decision making regarding immunosuppressive chemotherapy for peripheral ulcerative keratitis. Cornea. 1990;9(1):66–73
3. Sainz de la Maza M, Foster CS, Jabbur NS, Baltatzis S. Ocular characteristics and disease associations in scleritis-associated peripheral keratopathy. Arch Ophthalmol. 2002;120(1):15–9
4. Wood T, Kaufman H. Moorens ulcer. Am J Ophthalmol. 1971;71:417-422
5. Lewallen S, Courtright P. Problems with current concepts of the epidemiology of Mooren’s corneal ulcer. Ann Ophthalmology. 1990;22:52-55
6. Erdem U, Auran JD, Florakis GJ, Wilson SE, Srinivasan DB. Interferon treatment of Mooren’s ulcers associated with hepatitis C. Am j Ophthalmol. 1995;119:365-366
7. Zelefsky JR, Taylor CJ, Srinivasan M et al. HLS-DR17 and Mooren’s ulcer in South India. Br J Ophthalmol. 2008;92:179-181
8. Gottsch J, Liu S, Minkovitz JB et al. Autoimmunity to a cornea-asscoaited stomal antigen in patients with Mooren’s ulcer. Invest Ophthalmol Vis Sci. 1995;36:1541-1547
9. Foster CS, Forstot SL, Wilson LA. Mortality rate in rheumatoid arthritis patients developing necrotizing scleritis or peripheral ulcerative keratitis; effects of systemic immunosuppression. Ophthalmology 1984;91:1253–63
10. O’Day D, Horn JF. The eye and rheumatic disease. In: Ruddy S, Harris ED and Sledge CM, eds. Kelley’s textbook of rheumatology. 6th ed. Philadelphia: Saunders; 2001
11. Di Pascuale MA, Whitson JT, Mootha VV. Corneal melting after use of nepafenac in a patient with chronic cystoid macular edema after cataract surgery. Eye Contact Lens. 2008;34(2):129-130.
12. Wolf EJ, Kleinman LZ, Schrier A. Nepafenac-associated corneal melt. J Cataract Refract Surg. 2007;33(11):1974-5. 
13. Guidera AC, Luchs JI, Udell IJ. Keratitis, ulceration, and perforation associated with topical nonsteroidal anti-inflammatory drugs. Ophthalmology. 2001;108(5):936-44.
14. K.L Lu, W.R Wee, T Sakamoto, P.J McDonnell. Comparison of in vitro anti-proliferative effects of steroids and non-steroidal anti-inflammatory drugs on human keratocytes. Cornea. 1996;15;185–190
15. Apte RS, Hargrave SL, Fini EM, et al. NSAID and Matrix Metalloproteinases in Postoperative Corneal Melts. American Academy of Ophthalmology Meeting Poster, October, 2000.
16. Dana M, Qian Y, Hamrah P. Twenty-five-year panorama of corneal immunology: emerging concepts in the immunopathogenesis of microbial keratitis, peripheral ulcerative keratitis, and corneal transplant rejection. Cornea. 2000;19(5):625–43
17. Jabs DA, Nussenblatt RB, Rosenbaum JT. Standardization of uveitis nomenclature for reporting clinical data. Results of the First International Workshop. Am J Ophthalmol. 2005;140(3):509–16.

Time Bomb

Back to Top

By: Geraint Phillips & Sarah Welch*

Case History 

A 38yr old female presented with a two day history of occasional but increasing flashes in the peripheral vision of her left eye. She didn’t think her visual acuity had changed and she hadn’t noticed any floaters or other visual symptoms. About three years previously, she had noticed similar symptoms in her right eye and was found to have a small retinal tear for which she received some barrier laser photocoagulation. 

She had a history of progressive myopia during her teens which had stabilised by her early twenties at about -7.00DS in both eyes. Her ocular history was otherwise unremarkable and there was no family history of myopia or other eye problems.

Her best spectacle corrected visual acuity was 6/6 in both eyes, the IOPs were 14mmHg and the anterior segments showed no abnormalities. Her pupils were dilated with a combination of tropicamide and phenylephrine. 

The most significant posterior segment findings were in the superio-temporal periphery of both eyes. In the right eye there was pigmentary changes and scarring from the previous laser therapy surrounding the small retinal tear. This in turn was adjacent to an area of lattice degeneration, within which were several atrophic holes. These findings were visualised using a three mirror Goldmann-type contact lens. (Figure 1). In the superio-temporal periphery of the left eye there was also an area of lattice degeneration but without evidence of any atrophic holes, retinal tears or localised vitreo-retinal attachments. 

The vitreous of both eyes was clear and there were no signs of posterior vitreous detachment.

The patient was referred to a vitreo-retinal specialist for an opinion as to whether laser would be considered appropriate in the left eye given the increasing symptoms of flashes in that eye and the history of retinal tear in the other eye. 

Figure 1

Optometry View: Geraint Phillips

Lattice degeneration is a vitreoretinal condition that has been found to be present in about 6 - 10% of the population. It is more commonly seen in myopic eyes and is typically bilateral. It is usually located in the temporal region of the peripheral retina, between the equatorial region and the vitreous base, more often superiorly than inferiorly. Areas of lattice degeneration tend to run circumferentially and appear as round, oval or linear patches within which can be seen glistening white speckles and variable amounts of pigment clumping. A crisscrossing meshwork of white vessels may also be seen traversing the lesion. Just under half the time there will be small atrophic retinal holes present within lattice and these can remain stable for long periods, with no accompanying retinal detachment. 

Histologically, there is retinal thinning, adhesion of the vitreous to the margins of the lattice and the overlying vitreous is liquefied. Atrophic holes are full thickness breaks with no flap or overlying operculum. 

There are two types of retinal detachment (RD) that have been linked with lattice degeneration – RDs associated with tears at the margin of the lattice and RDs associated with atrophic holes. Tears at the margin can occur during a posterior vitreous detachment and subretinal fluid present at the margin of an atrophic hole can represent a subclinical RD that can progress. 

Lattice degeneration is present in 30- 40% of eyes that develop a RD. It has been reported that with no history of previous RD in either eye, the chances of a RD developing in an eye with lattice degeneration is less than 1% over an average of 11years. Where there is lattice degeneration in both eyes and a history of RD in one eye, the risk of RD developing over 7 years in the fellow eye is 2-5%. Where there are atrophic holes present, the fellow eye risk of RD has been reported as 12-39%. 

In the case presented here, there is bilateral lattice degeneration with atrophic holes in one eye. There is a history of retinal tear and laser repair in the eye with the holes and a history of increasing flashing lights in the fellow eye, although there were no retinal breaks present in this eye. A specialist vitreo-retinal opinion was sought which was certainly appropriate considering the findings and history. The decision whether or not to apply laser in these types of cases can be a source of debate. There is no clear evidence that prophylactic treatment is useful in many cases of lattice degeneration but the risk of RD is a spectrum, increasing with the additional presence of atrophic holes, significant myopia, the development of a posterior vitreous detachment and a history of RD in the fellow eye. 

Examination of the peripheral retina in this case was made easier with a widely dilated pupil using antimuscarinic and adrenergic dilating drugs. There are also a number of contact and non-contact indirect fundus lenses that provide excellent views of this region of the retina. It would be wise for practitioners to look carefully at cases of lattice degeneration to assess for the presence and the risk of retinal breaks. 

Ophthalmology View: Sarah Welch

Patients presenting with flashes and floaters are a common problem facing most ophthalmologists and especially vitreoretinal specialists.  The aim of the examination is: firstly, to exclude any rhegmatogenous retinal detachment (RRD) and then to determine if there are any changes in the retina that may lead to a RRD. Such changes require treatment.

In this case our patient has a worrying history including:

High myopia (greater than -5). About half of RRDs that occur without trauma, occur in myopes. A low myope (<-3 dioptres) has an increased risk of retinal detachment of 4x and a high myope has an increased risk of 10x compared to that of non-myopes.

A history of a retinal tear in the fellow eye. Many people have symmetrical vitreoretinal changes in their fellow eyes and so if I find an area that is suspicious in one eye I ensure that I examine the corresponding area in the other eye very carefully. Of those patients with a RRD in one eye then the risk is about 10% for the fellow eye.

Recent onset of flashes. Asymptomatic retinal tears and breaks are much less likely to proceed to RRD and so the presence of symptoms, especially flashes suggests to me there is an area of vitreous traction that is tugging on the retina causing symptoms.

Other key points that must be covered in the history include:

A family history of retinal detachment. This was negative in this case, but can point to syndromes like Sticklers.

The presence of previous ocular surgery. Cataract surgery increases your risk of retinal detachment to about 1%.

A history of trauma or of retinopathy of prematurity can also predispose to RRD.

Having taken a complete history, the most important step is a thorough examination.  In such cases, I find the examination of the vitreous helpful in guiding me to the likelihood of retinal pathology. In this case, our patient had a clear vitreous with no evidence of blood or tobacco dust (Schafer’s sign).  That these are absent makes a retinal tear or a RRD much less likely. Similarly, that there was no posterior vitreous detachment also makes a break less likely. 

Our patient demonstrated the common ‘mirror image’ location of pathology with lattice in both retinas in the superior-temporal periphery.  The location of a superior temporal break is the most common site for a tear to progress to a RRD. Our patient had appropriate laser retinopexy in her right eye to treat her symptomatic retinal break three years ago.  

In the left eye there is lattice with no evidence of a horseshoe retinal tear (also know as a flap tear).  These are the principal retinal breaks that lead to RRD. Because the flap still has vitreous traction on it there is a far greater risk of progression to RRD.  Lattice alone, as in this case, is often the site of where breaks may occur but there is no evidence that treating lattice without breaks has any benefit in stopping future breaks or RRDs and I would not recommend laser in this case. Lattice alone is rarely treated.  There is debate as to whether it should be treated if the fellow eye had had an RRD but there is insufficient evidence to show there is any benefit in treating or not. Even lattice with atrophic holes rarely progresses to RRD and mostly does not require treatment.  In some young people there is progression of subretinal fluid under atrophic holes and in these cases if progression is documented then they should be treated.

The most common treatment is laser retinopexy but in some cases cryo retinopexy is used. Treatment is generally reserved for horse-retinal tears as these are the most likely to progress to RRD.  The aim of any treatment is to form a scar around the break thereby stopping subretinal fluid accumulating and a detachment developing. A few patients will still progress to retinal detachment despite treatment.

Our patient needs to be followed annually to ensure there is no progression of asymptomatic retinal pathology. I would expect an optometrist to mange these regular follow-ups. Also, most importantly our patient must be warned to return urgently should there be any increase of symptoms – flashes, floaters or a field defect.

About the Authors

* Geraint Phillips is a senior lecturer and clinic director at the Department of Optometry and Vision Science, The University of Auckland. Sarah Welch is a consultant ophthalmologist specialising in vitreo-retinal surgery at Greenlane Hospital, Auckland, and in private practice at Auckland Eye.

References

1. Manjunath V, Taha, M, Fujimoto JG, Duker J. Posterior Lattice Degeneration Characteristics by Spectral Domain Optical Coherence Tomography. Retina 2011;31:492-496.
2. Wilkinson CP. Interventions for asymptomatic retinal breaks and lattice degeneration for preventing retinal detachment. Cochrane Database. Syst Rev 2005:CD003170. 
3. Gonzales CR, Gupta A, Schwartz SD. The fellow eye of patients with phakic rhegmatogenous retinal detachment from atrophic holes of lattice degeneration without posterior vitreous detachment. Br J Ophthalmol 2004;88:1400-1402.