Materials and methods

Thirty patients who had a previous clinical history of VKC and who were affected by active mild to moderate VKC (mean age 12 (SD 1.1) years; 21 males, nine females) were enrolled in the study. Only patients with a sum clinical score of not less than 3 and not more than 14 for allergic signs and symptoms were included. All patients were free from medication for at least 5 days. After clinical evaluation of ocular signs and symptoms, tear samples were collected using a capillary micropipette for tear cytology and ECP determination. A 5 μl aliquot was used for tear cytology, with the remainder stored at −20°C for subsequent determination of ECP.

Patients were divided in a double masked, randomised fashion into two groups (n=15 per group) to be treated bilaterally four times daily with either DSCG 4% (group 1) or lodoxamide 0.1% (group 2). In both groups, two patients with monolateral VKC were included. Clinical evaluation and tear collection (for tear cytology and ECP measurement) were repeated 10 days after commencement of therapy.

A clinical score (0–4: 0 = absent; 4 = severe) was determined at each visit and for each eye for the four major symptoms: itching, tearing, photophobia, and foreign body sensation, and for the six major signs: conjunctival erythema, conjunctival chemosis, discharge, papillae, (mild hyperaemic scattered papillae =1+; moderate diffuse hyperaemic swollen papillae = 2+; as before but more severe = 3+; hyperaemic swollen giant papillae covering the superior tarsal plate = 4+), limbal infiltrates (mild limbus hyperaemia and swelling =1+; moderate limbus hyperaemia and swelling =2+; as before but more severe = 3+; 360° limbus hyperaemia and swelling=4+), and corneal epithelial disease (fine superficial epithelial defects involving less than half of the cornea = 1+; diffuse fine superficial epithelial defects involving more than half of the cornea = 2+; confluent epithelial defects or mucous plaque formation = 3+; oval corneal ulcers = 4+).

Tear cytology was evaluated using precoloured slides (Testsimplet, Boehringer, Mannheim, Germany), counting the eosinophils present in five microscopic fields of 0.15 mm² each.

ECP was measured using a commercially available radioimmunoassay technique (Kabi Pharmacia, Uppsala, Sweden). The results were expressed as micrograms per litre (μg/l).

Statistical analyses of the normally distributed clinical data were performed using the paired Student’s t test for intratreatment comparisons, and unpaired Student’s t test for between treatment comparisons. The Wilcoxon sign rank test was used for the laboratory ECP values and for eosinophil number in tears, both not normally distributed data. Results are listed as the mean plus or minus standard deviation of the mean (SD). Correlations were obtained using the Spearman correlation analysis. For statistical significance, the assigned p value was ≥0.05.

Results

At baseline, the mean score of individual and total symptoms and signs and the mean tear ECP levels were similar in both groups of patients.

In DSCG treated patients, the mean score of clinical signs and symptoms did not change significantly from baseline. In 12 of the 28 DSCG treated eyes (42%), the mean score of symptoms and signs was unchanged. The mean tear ECP increased from 343 (SD 363) μg/l to 571 (777) μg/l after treatment (Table 1) due to a marked elevation in six of the treated eyes. However, in 17 of the samples (65%), tear ECP levels did show a decrease after treatment. The number of eosinophils in tears was reduced from 18.7 (14.2) to 12.5 (12.7).

In patients treated with lodoxamide, the mean total clinical score of both signs and symptoms decreased significantly (p<0.001) compared with baseline. In only five of 28 treated eyes (15%), the mean score was unchanged. Of the individual signs and symptoms, itching, photophobia, foreign body sensation, tearing, hyperaemia, chemosis, limbal infiltrates, and discharge were all significantly improved (p<0.001). Lodoxamide treatment also significantly reduced tear ECP compared with baseline levels (560 (756) μg/l v 241 (376) μg/l; p<0.01) (Table 1), with 83% of samples showing decreased tear ECP after treatment. The number of eosinophils in tears was reduced, but not significantly, from 16.2 (13.7) to 9.6 (10.2).

When comparing the two treatments, lodoxamide was significantly more effective in reducing clinical signs and symptoms (p<0.005) than was DSCG (Figs 1 and 2). Lodoxamide was superior to DSCG in the reduction of itching, photophobia, foreign body sensation, tearing, hyperaemia, chemosis, discharge, and corneal epitheliopathy, but not limbal infiltrates and papillae. Tear ECP was lower in patients treated with lodoxamide (241 (376) μg/l) than in those treated with DSCG (571 (777) μg/l). However, a cross group comparison of the values of these two groups did not show a statistically significant difference (p<0.06).

• Download figure
• Open in new tab
• Download powerpoint

Figure 1

Mean (SD) sum score of the six major ocular clinical signs of vernal keratoconjunctivitis (VKC) (conjunctival erythema, chemosis, discharge, papillae, limbus infiltrates, and epithelial disease) before and after topical, ocular treatment of 30 patients with VKC with either sodium cromoglycate 4% (DSCG) or lodoxamide 0.1%. *Statistically significantly lower than baseline ocular signs, p <0.005

• Download figure
• Open in new tab
• Download powerpoint

Figure 2

Mean (SD) sum score of the four major ocular clinical symptoms of vernal keratoconjunctivitis (VKC) (itching, tearing, photophobia, and foreign body sensation) before and after topical ocular treatment of 30 patients with VKC with either sodium cromoglycate 4% (DSCG) or lodoxamide 0.1%. *Statistically significantly lower than baseline ocular symptoms, p <0.005.

Significant correlations (p = 0.0001) were observed between ECP levels in all samples (independent of therapy) and signs, symptoms, corneal involvement, and number of eosinophils in tears.

Discussion

Vernal keratoconjunctivitis is a chronic inflammatory disorder in which both IgE and cellular mediated mechanisms are recognised as pathogenic factors.9 The typical histological features of VKC are the conjunctival infiltration of eosinophils and basophils, a constant, increased number of mast cells, and a connective tissue hyperplasia with an increased deposition of collagen.1011An accumulation of Th2-like helper T cells, producing interleukins 3, 4, and 5, has also been shown to occur in VKC, accounting for the frequent occurrence of high levels of tear immunoglobulin E, mast cell proliferation, and the abundant presence of eosinophils.12Activated eosinophils are known to release toxic factors, such as the currently studied ECP, in addition to major basic protein (MBP), eosinophil peroxidase, eosinophil derived neurotoxin/eosinophil protein X (EDN/EPX). Eosinophil MBP has already been shown to play a role in the pathogenesis of epithelial erosions observed in VKC, indicating the importance of eosinophil activation in severe forms of this disease.13 Conjunctival eosinophil activation and degranulation, accompanied by high levels of tear ECP, was also previously reported in seven patients with VKC¹⁴ and then confirmed by us in a larger group of patients.7 Higher levels of serum ECP and EPX were also shown, indicating that VKC should be considered a ‘systemic’ disorder with circulating activated eosinophils.15 The toxic effect of ECP on corneal epithelial cells has been shown in vitro,45 suggesting a relation between this mediator and VKC related corneal damage and that reduction of eosinophil activation may be an important objective in the treatment of VKC.

In this study, ECP was easily detectable and measurable in tears with a high degree of sensitivity. The wide variation of ECP levels found in baseline tears of VKC patients may have reflected a varying range of activation of eosinophils. Lodoxamide significantly decreased mean tear ECP levels, which reflected reductions in 83% of individual eyes. Mean ECP was increased after DSCG because six of the treated eyes showed a remarkable increase of ECP associated with a worsening in clinical signs and symptoms. However, DSCG did reduce tear ECP in 66% of treated eyes.

Clinical signs and symptoms were significantly decreased only by lodoxamide treatment. The finding that lodoxamide significantly reduced ECP levels but not eosinophil number suggests that this drug is more effective on eosinophil activation and less on chemotaxis. ECP levels were found to be highly correlated with both signs and symptoms, in addition to corneal involvement and, less strongly, with the number of eosinophils in tears. The correlation between ECP and corneal epitheliopathy substantiates in vitro studies on the epithelial toxicity of ECP.

DSCG has been shown to be effective in the long term management of VKC,16 but often moderate to severe cases need additional therapy with steroids or the alternative, topical treatment with cyclosporin A.17

Lodoxamide is a newly available mast cell stabiliser more active than DSCG in inhibiting the passive cutaneous anaphylaxis and conjunctival allergic reactions in animal models.1819 Recently, using the conjunctival provocation test in humans, we reported the efficacy of lodoxamide in reducing histamine release and the development of induced clinical and cytological allergic early and late phase reactions.20 In other studies, lodoxamide has been shown to be more effective than DSCG in alleviating signs and symptoms of VKC²¹ and to be more effective at decreasing steroid use.22 Additionally, in a long term study, lodoxamide was shown to decrease corneal epitheliopathy in VKC.8

With the present data, lodoxamide was shown to reduce eosinophil activation and the release of toxic proteins. The finding of a parallel decrease in signs and symptoms, which were shown to be correlated with ECP levels, suggests that inhibition of eosinophil activation should be an important part of VKC management. Thus, the measurement of ECP in tears is a new tool for use not only in the monitoring of ocular allergic diseases, but also in the evaluation of topical therapies, for which an objective factor is needed to complement clinical assessments.