Clinical features and outcome of corneal opacity associated with congenital glaucoma

Congenital glaucoma (CG) is a rare disease with the incidence largely varying upon the ethnicity [1‐3]. Studies reported that the annual incidence of CG was 2.85 to 5.41 in 100,000 live births in Caucasian populations [3, 4], whereas it was higher in South Asian children [3]. Despite its rarity, CG is often associated with poor visual and functional outcome [5, 6], and it is estimated that glaucoma is responsible for 4–18% of childhood blindness [3, 6‐8]. Hence, early detection of the disease and proper treatment are necessary for the vision in pediatric patients with CG.

The classical triad of symptoms in congenital glaucoma includes epiphora, photophobia, and blepharospasm [1]. However, the most common signs first recognized by parents or doctors are corneal abnormalities such as corneal enlargement (buphthalmos) due to increased intraocular pressure (IOP) or corneal clouding as a result of Descemet’s membrane tears (Haab striae) or stromal edema [9‐13]. In addition, as it is one form of developmental anomaly of anterior segment, CG is often combined with corneal opacity as a sequel to anterior segment dysgenesis. In these cases, corneal opacity can lead to sensory deprivation amblyopia, and the visual outcome can be poor despite optimal control of IOP. One study reported that corneal opacity along with anisometropia was responsible for vision loss in 50% of childhood glaucoma patients [14]. Therefore, evaluation for corneal opacity and its management are critical for early diagnosis of the disease and the favorable outcome in patients with CG.

We performed this study to evaluate the incidence and clinical characteristics of corneal opacity combined with CG and to investigate the surgical outcome of penetrating keratoplasty (PK) and clinical factors affecting the outcome in eyes with CG.

This retrospective study was approved by the Institutional Review Board (IRB No. 1706–094-860). Medical records were reviewed for 320 eyes of 193 Korean patients who were diagnosed with CG between January 1981 and January 2016.

The diagnosis of CG was taken as recorded in the charts. The diagnosis was usually made on the basis of two or more of the following ocular findings: elevated IOP (> 21 mmHg), buphthalmos (enlarged corneal diameter, Fig. 1a), Haab striae, corneal stromal edema, and glaucomatous optic disc change which had been present at birth or shortly after birth. Glaucoma of childhood or juvenile onset was excluded.

Data collected were the patient demographics, laterality of disease, IOP, corneal diameter, ocular comorbidities (aniridia, Peters anomaly, cataract, posterior segment anomalies), systemic abnormalities (Sturge-Weber syndrome, neurofibromatosis, congenital heart disease, TORCH positivity, cerebral palsy, Wilms tumor, chromosomal anomaly), ocular surgery (PK, glaucoma surgery, lens extraction), and last-recorded visual acuity.

In addition, anterior segment photographs at first presentation of a patient were reviewed for the presence of corneal opacity, and the corneal findings were classified as follows: 1) completely clear cornea, 2) Haab striae only (Fig. 1b, c), and 3) corneal opacity. The severity of corneal opacity was further graded based on both four-stage system and two-scale system. The four-stage system followed the corneal opacity scoring system suggested by Gupta et al. [15, 16] with modification: stage 1 = minimal opacity (Fig. 1d), stage 2 = moderate stromal opacity (anterior chamber and iris both well visualized, Fig. 1e), stage 3 = significant stromal opacity (pupil visible with haze, Fig. 1f), and stage 4 = intense stromal opacity (pupil invisible, Fig. 1g). The two-scale system was composed of nebulomacular corneal opacity (iris details visible through opacity) and leukomatous corneal opacity (unable to visualize iris details through opacity). Overall, nebulomacular corneal opacity included stage 1 and 2 opacities, and leukomatous corneal opacity comprised stage 3 and 4 opacities.

The data were additionally analyzed in 12 eyes of 10 CG patients with stage 4 corneal opacity who underwent PK. The graft failure was determined when corneal clarity was irreversibly lost under slit-lamp examination, and the graft survival time was defined as the period from PK to the day when the graft failure was first noted. When multiple PKs were performed in a patient, the first surgery was taken for analysis. The association of age, IOP at the time of surgery, corneal diameter, donor/recipient trephine sizes, difference of sizes between donor and recipient trephines, presence of ocular comorbidities, and concurrent lens or glaucoma surgeries with corneal graft outcome was evaluated.

Data were presented as mean ± SD. The GraphPad Software (GraphPad Prism, La Jolla, CA) was used for statistical analysis. Comparison of quantitative variables between two groups was made by using Student t test. The correlation between IOP and the severity of corneal opacity was tested by using Pearson r coefficient and two-tailed P value. Survival analysis was performed using the Kaplan–Meier method to estimate the median time to graft failure. The association with the surgical outcome and each clinical factor was analyzed using Fisher’s exact test for qualitative variables and two-tailed Student t test for quantitative variables. A P value < 0.05 was considered statistically significant.

Corneal findings are important for suspicion and diagnosis of CG. Studies reported that cloudy cornea and buphthalmos are the most common presenting signs found in over 40% of patients with CG [11, 12]. In our study, 248 of 320 eyes (77.5%) with CG had corneal opacification at disease presentation, whereas 22.5% had completely clear cornea. However, among 77.5% of CG eyes with corneal opacity, 16.6% presented with Haab striae alone and 21.2% had grade 1 minimal opacity. Therefore, visually significant and recognizable corneal opacity (grade 2, 3 or 4 opacity) was observed in 39.7% of the eyes with CG, which is similar to previous reports [11] [12].

Other notable findings of our study are that male was more prevalent than female in CG patients at a ratio of 60.1 to 39.9, and bilateral involvement was more common compared to unilateral involvement at a ratio of 65.8 to 34.2. Overall, 18.7% of CG patients had combined systemic diseases, and the most common systemic comorbidity was Sturge-Weber syndrome which was found in 12.9% of patients with CG. However, Sturge-Weber syndrome was not associated with severe corneal opacity since no patient with Sturge-Weber syndrome underwent PK because of corneal opacity. Neurofibromatosis was combined in two patients (1.0%) in our study. Quaranta et al. [18] reported that neurofibromatosis patients showed gonioscopic findings characteristic of underdevelopment of the iridocorneal angle, suggesting the vulnerability to glaucoma. Of ocular morbidities combined with CG, aniridia was the most frequent (found in 7.3% of CG patients), followed by Peters anomaly (found in 3.6% of CG patients). Combined aniridia and Peters anomaly predisposed patients with CG to severe non-resolving corneal opacity which required PK. Seven out of 14 patients with CG + aniridia and 5 of 7 patients with CG + Peters anomaly underwent PK.

It is generally accepted that the presence of glaucoma and concurrent glaucoma operation during PK are risk factors for poor graft survival in a pediatric population [19, 20]. However, there have just a few case series evaluating the surgical outcome of PK in CG patients, and the results greatly vary upon studies. In our study, the overall graft survival rate was 50% (6 of 12) in eyes with CG during a mean follow-up of 80.6 months, and 33.3% of patients with PK achieved ambulatory vision at the last follow-up. Of note, in the graft failure group, all grafts failed within 10 months (at the mean 7.3 months) after PK. Ariyasu et al. [21] reported that the graft success occurred in 67% of grafts (6 of 9) in eyes with CG during 24 months of follow-up, and ambulatory vision was achieved in 75% of eyes. In a report by Al-Torbak [22], 43 and 17% of corneal grafts survived at 24 and 48 months, respectively after combined PK and Ahmed valve implantation. In our patients, PK was combined with glaucoma valve implantation in 4 eyes, 3 of which had graft failure. By contrast, 3 of 8 eyes without glaucoma surgery had graft failure. This suggests that concurrent glaucoma valve implant surgery might be associated with poor graft outcome, although it did not reach statistical significance due to the small sample size of the current study. Further study with larger sample size would be necessary to confirm the effect of simultaneous glaucoma surgery on the corneal transplant survival.

The only variable affecting the graft outcome was the corneal diameter in our study. The cornea was significantly smaller in the graft failure group, compared to the graft success group. This result is consistent with our previous findings in patients with Peters anomaly or sclerocornea [23]. The proximity of the donor graft to the recipient limbus and more exposure to the host immune system might be related to the high rate of graft failure in eyes with smaller corneas.

Our study is limited by its retrospective nature. Although we here presented the data on corneal opacity determined from the corneal photography in 320 eyes of 193 patients, it was not possible to evaluate other corneal abnormalities such as topographic changes, endothelial cell counts, or hysteresis. Since it was reported that corneal topographic abnormalities were commonly present in CG [13], further prospective study evaluating the cornea from multiple aspects would be helpful to better understand corneal anomalies associated with CG.

Corneal opacity was a common feature of CG and found in 77.5% of CG patients. The survival rate of corneal allografts in eyes with CG and severe stromal opacity was 50%, and the graft outcome was poorer in eyes with small corneal diameters.