MOOREN’S ULCER IN IBADAN, SOUTHWEST NIGERIA

Abstract

Background:

Mooren's ulcer has been described as an idiopathic, painful, relentless, peripheral, chronic ulcerative keratitis that progresses circumferentially and centrally, with an unusual geographical presentation and distribution.

Aim:

To review the demographic and clinical characteristics of patients with Mooren’s ulcer at University College Hospital, Ibadan, Nigeria over seven years, and, the outcome of treatment.

Study design:

A retrospective case series

Methods:

Medical records of all patients with Mooren’s ulcer seen in the Eye Clinic, University College Hospital, Ibadan, South-west Nigeria between January 2007 and December 2013 were reviewed.

Results:

Thirty-one eyes of 23 patients, (18 males, overall mean age of all patients 30.9 ±16.6 years) were affected with Mooren’s ulcer. The right eye was involved in nine (39.1%) patients while eight (34.8%) patients had bilateral lesions. Most common presenting symptoms were redness and pain, with an average of 6.8 clock hour-limbal involvement. All patients received topical and systemic steroids, while eight patients with severe disease, or poor response to treatment with steroids, received systemic methotrexate or cyclophosphamide. Surgical treatment with conjunctivectomy and cryotherapy was undertaken in 14 (60.9%) patients (21 eyes) who had a minimum follow-up period of eight weeks, while those presenting with perforation also had scleral patch graft or direct repair. The ulcer healed with varying degrees of corneal opacity in 18 (85.7%) of the 21 eyes, while the remaining three eyes developed descemetocele, anterior staphyloma or corneal perforation. Large proportion (76.2%) of eyes maintained their presenting acuity.

Conclusion:

Mooren’s ulcer remains an uncommon disease in Ibadan, South-west Nigeria, and the clinical presentation is as recorded in other parts of the West African sub-region. Systemic immunosuppression with methotrexate and scleral patch graft appear to give good outcome in patients presenting with severe ulceration and/or perforation.

Introduction

Mooren’s ulcer was first described by Bowman in 18491 and later as a clinical entity in 18632. It is a chronic, painful peripheral ulceration of the cornea, largely of unknown etiology occurring in complete absence of any systemic disease that could be responsible for the progressive destruction of the cornea3. It is strictly a peripheral ulcerative keratitis (PUK), with no associated scleritis, occurring more commonly in Southern and Central Africa, China, and the Indian subcontinent, but rarely in the northern hemispher4. Mooren’s ulcer can occur in any age group, and has been described in a three-year-old child5 but, mostly, males between 40 and 70 years of age are affected6. Wood and Kaufmann7 described two clinical types of Mooren’s ulcer. The unilateral type, affecting older individuals, presents with mild to moderate symptoms and responds well to treatment; while the bilateral type is associated with greater pain, responds poorly to therapy and affects younger individuals. However, a more recent classification8 groups Mooren’s ulcer into three based on their clinical features, anterior segment fluorescein angiography and response to treatment. The unilateral ulceration is usually painful, progressive, with non-perfusion of superficial vascular plexus, and affects elderly individuals; while the bilateral aggressive lesion progresses circumferentially and centrally, with vascular leakage and new vessel formation at the base of the ulcer, and, affects younger individuals. The bilateral indolent ulceration, on the other hand, progresses circumferentially, with minimal inflammatory response, extension of new vessels to base of ulcer, and, affects middle-aged individuals. Treatment of Mooren’s ulcer is generally difficult due to its uncertain etiology, variable presentation and unusual geographical distribution8.

There are few studies9^-11 on Mooren’s ulcer in Nigeria, and the last report10 in South-west Nigeria was over two decades ago consisting of five case series. Hence, this study aims at reviewing the demographic and clinical characteristics of patients with Mooren’s ulcer at University College Hospital (UCH), Ibadan, South-west Nigeria and the outcome of treatment.

Reports

Methods

The medical records of all patients diagnosed with Mooren’s ulcer at the Eye Clinic of UCH, Ibadan, South-west Nigeria between January 2007 and December 2013 were reviewed. An average of 100 new patients are seen weekly at the Eye Clinic of UCH, Ibadan with various ocular diseases. Information retrieved from the medical records include patients’ demographics, presenting complaints, duration of symptoms, presenting visual acuity, laterality, degree and extent of ulcer, treatment received, outcome of treatment and duration of follow-up. The diagnosis of Mooren’s ulcer was made clinically based on typical features of painful progressive peripheral corneal ulceration with steep undermined leading and central edge, no sclera involvement and absence of other ocular infections or systemic diseases. Ethical clearance was received from the Hospital Ethical Review Committee.

Statistical analysis

Statistical analysis was performed on collected data with the aid of Statistical Package for Social Sciences version 15 software (SPSS Inc, Chicago IL., and USA) for exploratory and statistical analysis. Descriptive statistics such as means and standard deviations were used to summarize quantitative variables while categorical variables were summarized with proportions. The Pearson chi square test was used to investigate associations between any two categorical variables while the Mc-Nemar chi-square was used to investigate the change in proportion between pre-treatment and post-treatment visual acuity. All tests were declared significant at the 5% level of significance.

Results

Twenty-three patients (31 eyes) were seen during the study period. Eighteen were male (M: F = 3.6: 1), and age ranged from 16 to 79 years with an overall mean of 30.9 ±16.6 years. Males had a lower mean age (27.3 ± 14.4 years) compared with females (43.6 ± 19.2 years), but this was not statistically significant (p = 0.05); while patients with unilateral lesions had a significantly lower mean age (24.3 ± 6.8 years) than those with bilateral lesions (43.1 ± 22.6 years) (t= -3.033, df = 21, p = 0.006).

The right eye was involved in nine (39.1%) patients while eight (34.8%) patients had bilateral lesions. The presenting symptoms in the patients are shown in table 1. Nineteen (82.6%) patients presented with redness and 18 (78.3%) with pain. Figure 1 shows the clinical picture of a patient with the typical features of Mooren’s ulcer who presented with perforation. The average duration of symptoms at presentation in the patients was 12.5 ± 10.9 weeks (median 8 weeks, range 1-40 weeks)

Table 1: Presenting symptoms in patients with Mooren’s ulcer

Symptom	Frequency	Percent
Redness	19	82.6
Pain	18	58.1
Foreign body sensation	14	45.2
Tearing	14	45.2
Severe visual loss	6	19.4

Figure 1 .

Mooren’s ulcer presenting with perforation

Twelve (38.7%) eyes of 12 patients had perforated corneas at presentation, four of which were in patients with bilateral ulceration. Ulceration was limited to one quadrant of the limbus in six (19.4%) eyes, while all four quadrants were involved in 10 (32.3%) eyes Table 2. An average of 6.8 ± 3.4 clock hours of the limbus was involved in the ulceration in all 31 eyes with the temporal quadrant most affected in 28 (90.3%) eyes and the superior quadrant the least in five (16.1%) eyes (Figure 2). Visual acuity at presentation was >6/18 in nine (29.0%) eyes while four (12.9%) eyes had <3/60 acuity. All the four eyes with acuity of <3/60 had perforated corneas. Table 3

Table 2: Limbal quadrant involved

Specific quadrant	Frequency	Percent	No of quadrants	Frequency	Percent
Temporal	28	90.3	One	7	22.6
Inferior	26	83.9	Two	8	25.8
Nasal	19	61.3	Three	11	35.5
Superior	5	16.1	Four	5	16.1
			Total	31	100.0

Figure 2 .

Quadrant involvement in patients with Mooren’s ulcer

Table 3: Pre- and post-treatment visual acuity in patients with Mooren’s ulcer

Visual acuity	Pre-treatment		Post-treatment
	Frequency	Percent	Frequency	Percent
>6/18	9	29.0	11	52.4
6/18-3/60	18	58.1	6	28.6
<3/60	4	12.9	4	19.0
Total	31	100	21	100

All the patients were treated with topical dexamethasone 1% hourly and oral prednesolone 1mg/kg body weight and, these were tapered according to the response to treatment. However, over the last 18 months, patients presenting with corneal perforation (six patients), severe corneal ulceration involving the four quadrants (one patient), and, poor response to steroids after one week of treatment (one patient) were treated with systemic cyclophosphamide, 2mg/kg body weight/day (one patient) or methotrexate 7.5mg weekly (seven patients) in addition to tapering doses of steroids. Fourteen (60.9%) patients (21 eyes, 67.7%) had surgical treatment, the commonest procedure being conjunctivectomy with cryotherapy to the limbus in all 14 patients. Of the 12 patients with corneal perforation, seven patients had scleral patch graft, and two had direct closure alongside conjunctivectomy and cryotherapy, while the remaining three patients declined surgical intervention and subsequently defaulted from follow-up.

Fourteen (60.9%) patients (21 eyes) had a minimum follow-up of eight weeks (average follow-up duration 14.5 months, range 2 - 60 months), and were assessed for outcome of treatment. Considering the anatomical outcome, 18 (85.7%) of the 21 eyes healed with varying degree of corneal opacity, while one eye each developed anterior staphyloma, descemetocele or corneal perforation. The patient with corneal perforation subsequently had evisceration.

At last follow-up visit, acuity of >6/18 was recorded in 11 (52.4%) eyes while four (19.0%) eyes had acuity <3/60. Table 3. Comparing presenting acuity with post-treatment acuity in the 21 eyes, out of the 11 eyes presenting with visual impairment (visual acuity 6/18 – 3/60), vision improved to normal (visual acuity >6/18) in four eyes, worsened to blindness (visual acuity <3/60) in one eye, while the presenting acuity was maintained in the remaining patients. This change was however, not statistically significant (p = 0.23). Six (54.5%) of the 11 eyes with post-treatment normal visual acuity had perforation at presentation while three (75%) of the four eyes which became blind and three (50%) of the six with visual impairment presented with perforation.

Discussion

Mooren’s ulcer is reportedly more prevalent in West Africa, affecting younger individuals, and with a higher risk of perforation12. The mean age of the patients in this series (30.9 ± 16.6 years) is similar to earlier studies10^-12 in the country and the West African sub-region, but lower than studies6,13 among other races.

Also, similar to previous studies6,7,9,10,14^-16, males outnumbered females in this series (ratio 3.6:1). No reasons had been attributed for this, but, Lewallen and Courtright17 opined that it might be due to increased incidence of ocular trauma in men, and, cultural practices that discourage female hospital attendance in some countries rather than true biologic propensity for males. These factors however, do not cut across all ethnic groups and geographical regions, and may not fully explain this higher prevalence in males.

About a third of the patients in this series had bilateral lesions, and were in their older age (mean age 43.1 ± 22.6years). Previous studies7,9,10,12,18,19 had suggested that bilateral Mooren’s ulcer commonly affects younger individuals, however, Lewallen and Courtright17, after a review of all published cases of Mooren’s ulcer found that bilateral disease occurred in 43% of older patients and only one third of younger patients. Srinivasan et al6 also opined that the clinical characteristics of Mooren’s ulcer vary considerably from one region to another and that age, gender and race are not universal predictors of severity or natural history of the disease.

It is believed that Mooren’s ulcer is an autoimmune disease6,20^-22 but the exact etiology is unknown. The hypothesis10,23 linking Mooren’s ulcer in the West African sub-region with hookworm infection could not be verified by van der Gaag and associates24 in their study in Sierra Leone as they found circulating antibodies against corneal epithelium and hookworm in both patients with Mooren’s ulcer and controls.

A number of studies7,25,26 had linked ocular trauma and surgery with Mooren’s ulcer. Foster13 however, believed that peripheral ulcerative keratitis associated with trauma should be excluded as Mooren’s ulcer on the account that the etiology of the corneal ulceration resulting from trauma could be attributed to the accompanying vasculitis. In our series, only one patient gave a history of ocular trauma preceding onset of symptoms, and, an earlier report10 linking trauma and Mooren’s ulcer in the country might have been co-incidental.

The earliest typical clinical feature of Mooren’s ulcer is a grey swollen corneal patch 2-3 mm from the limbus which rapidly furrows8. This was however, not observed in any patient in this series, probably due to late presentation. The mean duration of symptoms in the affected eyes before presentation was 12.5 ± 10.0 weeks with the majority of the patients presenting with advanced disease.

About half of the patients had perforated ulcers at presentation. The incidence of perforation in Mooren’s ulcer varies in different studies. Keitzman9 reported about 36% perforation in an earlier study in the country, while Majekodunmi10 reported no perforations in a subsequent report of five cases. Perforation has been described in association with bilateral aggressive type of Mooren’s ulcer which is more common in the West African sub-region8,9,12,23. A third of the patients with perforations in this series had bilateral lesions, though, none presented with bilateral perforations, thus, suggesting an asymmetrical presentation of the disease in those with bilateral lesions.

The temporal limbus was the most commonly affected quadrant in the ulceration, similar to previous studies6,10. The inflammatory cells and circulating antibodies implicated in Mooren’s ulceration possibly reach the cornea through the limbal blood vessels.

Many treatment modalities had been tried over the years in patients with Mooren’s ulcer, and, due to the uncommon presentation of the disease, randomized and masked clinical trials to study the different therapeutic regimens had been difficult3. Treatment is however, aimed at arresting the destructive process, promoting healing and re-epithelialization of the corneal surface thereby ensuring good visual outcome22,27. Thus, a step-wise approach in management with topical steroids, conjunctival resection, systemic immunosuppression, additional surgical procedures and rehabilitation has been advocated3. In this series, all the patients were commenced on topical and systemic steroids at presentation, and these were subsequently tapered according to patients’ response. Administration of steroids is usually the first line of treatment for Mooren’s ulcer due to the autoimmune etiology3,8. Additionally, eight patients (six patients presenting with perforation, one patient each with 360⁰ severe ulceration or poor response to steroids after one week of treatment) were treated with systemic methotrexate (seven patients) or cyclophosphamide (one patient) with good response. They are presently on follow-up with close monitoring of their full blood count, electrolytes and urea, and urinalysis. Systemic immunosuppression with methotrexate or cyclophosphamide has been reported as good therapeutic choice for bilateral Mooren’s ulceration that is unresponsive to conventional medical and surgical treatment13,20,22.

Surgical modalities used in the management of Mooren’s ulcer with variable success include conjunctival resection22,27, lamellar keratoplasty28, keratoepithelioplasty29, delimiting keratotomy30, and patch grafts using periosteum31, fascia lata29, or Gore-Tex32 graft. Fourteen (66.7%) patients (21 eyes) in this series underwent conjunctivectomy and cryotherapy. This is aimed at reducing the antibody-producing inflammatory cells present in the conjunctiva adjacent to the ulcer3,33,34 and, a reasonably good response had been reported with this surgical treatment10,35,36. The major challenge in managing Mooren’s ulcer perforations with patch graft is recurrence of the ulcer in grafted material8,22. More than half of the patients with perforation in this series had scleral patch graft while about one fifth had direct closure alongside immunosuppression with good response

A small proportion of the patients who were followed up for a minimum of eight weeks had improvement in vision of between one and three lines of the Snellen’s chart while most patients maintained their pre-treatment visual acuity, similar to previous studies8,10,37. However, almost 90% of the 21 eyes achieved good anatomic outcome, the ulcers healing with varying degree of corneal opacity, while one eye each developed anterior staphyloma, descemetocele or corneal perforation. The major cause of blindness and visual impairment in the patients was leucomatous corneal opacity. The main challenge to corneal transplantation for visual restoration in these patients is the high rate of graft rejection and recurrence of the ulceration in the graft with some authors8,38,39 recommending conservative management of the corneal opacity.

The retrospective nature of the study and varied treatment modalities are major limitations to this study.

Conclusions

Mooren’s ulcer remains an uncommon disease in Ibadan, and the demographics and clinical presentation are similar to other parts of the West African sub-region. Systemic steroids and immunosuppression with appropriate surgical treatment appear to give good anatomic outcome and preservation of vision in our patients, and should be instituted early in those presenting with severe ulceration and/or perforation.