by Mark B. Abelson, M.D.; Michelle A. George; Christopher Garofalo & Dana Weintraub
(Dr. Abelson is a clinical senior scientist at Schepens Eye Research Institute, Harvard Medical School; Mrs. George is an assistant director of the ocular allergy laboratory at Schepens Eye Research Institute; Mr. Garofalo and Ms. Weintraub are lab assistants.)
Homeopathy, a form of medicine based on the theory, "like cures like," has found its way into the ophthalmic community. However, since it's difficult to determine the mechanism of action of most homeopathic remedies due to the minute concentration of active ingredients, many physicians feel they lack scientific merit. In the following study, we evaluated the efficacy of Similasan Eye Drops #2, a homeopathic treatment containing extracts of honey bee, eyebright, and cavadilla, in order to prove that it significantly reduces the signs and symptoms of allergic conjunctivitis as induced by allergen challenge.
Materials and Methods
Before beginning the study, we performed a baseline slit lamp exam to ensure that all subjects' ocular health was within normal limits. Subjects who exhibited greater than 1+ hyperemia, reported any ocular itching, or did not meet the entry criteria were not admitted into the study. To continue in this study, subjects had to agree to refrain from using any nonsteroidal anti-inflammatory (NSAIDS), antihistamines or steroids during the study. Contact lens wear was also prohibited for the study duration.
Antigen Challenge
We challenged 47 subjects with increasing doses of a solubilized antigen (short ragweed, cat dander, or timothy grass), to which they were sensitized, until each subject responded with at least 2+ itching and ciliary, conjunctival, and episcleral hyperemia bilaterally, as scored on a 0-4 scale (Table 1). We excused any subjects who failed to manifest 2+ itching and 2+ hyperemia.
| Sign/Symptom | 0 | 1 | 2 | 3 | 4 |
|---|---|---|---|---|---|
| Eyelid swelling, chemosis and hyperemia | none | mild | moderate | severe | unusually severe |
| Tearing | none | mild (eyes feel slightly watery) | moderate (blows nose occasionally) | severe (tears rolling down cheeks) | |
| Mucous discharge | none | present | |||
| Itching, as graded by the subject | none | an intermittent tickle sensation in the inner corner | a mild continuous itch not requiring rubbing | a definite itch; you would like to be able to rub | an incapacitating itch that requires eye rubbing |
At day seven, we confirmed the reproducibility of the initial response by instilling the final dose of antigen. Ten minutes after challenge, we evaluated conjunctival, episcleral and ciliary hyperemia, chemosis, lid swelling, tearing and itching.
Subjects who responded with 2+ itching and hyperemia (n=33) were given masked bottles marked "1" and "2" containing either drug or saline as control. We randomly assigned the masked bottles to each eye according to a predetermined randomization code (contralateral control within each individual). We instructed the subjects to administer 1-2 drops of the appropriate liquid into the assigned eye q.i.d. for 14 days.
At day 21, we performed a baseline ophthalmic examination and administered the final dose of the appropriate medication into each assigned eye. Within five minutes of installation, we examined the eye to assess comfort and safety. Ten minutes after installation, we challenged each subject with the final dose and evaluated conjunctival, episcleral, and ciliary hyperemia, chemosis, lid swelling, tearing and itching at 3, 10, and 30 minutes following the antigen challenge.
Results
We compared the means ±S.E.M. from the drug treated group at the 10-minute examination on visit 3 (after treatment installation) to the means ±S.E.M. from the drug treated group at the 10-minute exams in visits 1 and 2 (preceding treatment installation). We also made the same comparison within the placebo-treated group.
| Eye | Comparison of visits (A vs. B) | Mean ± S.E.M. A | Mean ± S.E.M. B | Mean Diff. (A-B) | pValue paired | pValue nonpaired |
|---|---|---|---|---|---|---|
| Drug | 1 vs. 2 | 7.25±0.28 | 7.38±0.31 | -0.13 | 0.99 | 0.77 |
| 2 vs. 3 | 7.38±0.31 | 6.11±0.43 | 1.27 | 0.0008 | 0.02 | |
| 1 vs. 3 | 7.25±0.28 | 6.11±0.43 | 1.14 | 0.004 | 0.03 | |
| Placebo | 1 vs. 2 | 7.25±0.27 | 7.27±0.29 | -0.02 | 0.96 | 0.97 |
| 2 vs. 3 | 7.27±0.29 | 6.13±0.42 | 1.15 | 0.003 | 0.03 | |
| 1 vs. 3 | 7.25±0.27 | 6.13±0.42 | 1.12 | 0.009 | 0.03 |
| Eye | Comparison of visits (A vs. B) | Mean ± S.E.M. A | Mean ± S.E.M. B | Mean Diff. (A-B) | pValue paired | pValue nonpaired |
|---|---|---|---|---|---|---|
| Placebo | 1 vs. 2 | 2.98±0.11 | 3.05±3.10 | -0.07 | - | - |
| 2 vs. 3 | 3.05±0.10 | 1.89±0.16 | 1.16 | 0.0001 | 0.0001 | |
| 1 vs. 3 | 2.98±0.11 | 1.89±0.16 | 1.09 | 0.0001 | 0.0001 | |
| Drug | 1 vs. 2 | 2.97±0.10 | 3.05±0.10 | -0.08 | 0.9999 | - |
| 2 vs. 3 | 3.05±0.10 | 2.00±0.20 | 1.05 | 0.0001 | 0.0001 | |
| 1 vs. 3 | 2.97±0.10 | 2.00±0.20 | 0.97 | 0.0002 | 0.0001 |
| Time (min) | Ciliary Hyperemia | Conjunctival Hyperemia | Episcleral Hyperemia | Itching | pValue paired | pValue nonpaired |
|---|---|---|---|---|---|---|
| 3 | Placebo: 1.50±0.16 | Placebo: 1.52±0.15 | Placebo: 1.42±0.17 | Placebo: 2.22±0.17 | 0.54 | 0.83 |
| Drug: 1.45±0.16 | Drug: 1.50±0.15 | Drug: 1.41±0.17 | Drug: 2.20±0.17 | 0.84 | 0.94 | |
| 10 | Placebo: 2.08±0.13 | Placebo: 2.06±0.14 | Placebo: 1.98±0.15 | Placebo: 1.89±0.16 | 0.99 | 0.99 |
| Drug: 2.08±0.13 | Drug: 2.05±0.14 | Drug: 1.98±0.15 | Drug: 2.00±0.20 | 0.87 | 0.94 | |
| 30 | Placebo: 1.98±0.15 | Placebo: 1.98±0.15 | Placebo: 1.97±0.16 | Placebo: 0.92±0.18 | 0.44 | 0.67 |
| Drug: 2.08±0.16 | Drug: 2.08±0.16 | Drug: 2.08±0.16 | Drug: 0.86±0.18 | 0.45 | 0.32 |
Discussion
These data suggest that Similasan Eye Drops #2, a homeopathic remedy, significantly reduces hyperemia and itching, the hallmark parameters of ocular hayfever. Ciliary and conjunctival hyperemia decreased by approximately a 0.5 unit in the scoring system, while itching was decreased by 1.0 unit. This 1.0 unit of itching is statistically significant. In addition, according to the FDA’s general dictate defining products of efficacy, 1.0 unit of difference indicates clinical relevance as well.
These significant differences were found in both the placebo-treated and the drug-treated groups, however, this may be indicative of a crossover effect. The lack of difference between visit 1 scores and visit 2 scores confirms the reproducibility of the model, and we would expect to record similar scores at visit 3 in the absence of a treatment effect. As the study design required dosing of 1-2 drops q.i.d. for two weeks, systemic absorption, resulting in an active effect in the placebo eye, is a possibility.
In our experience in placebo controlled trials of anti-allergic agents using this model, drug treated eyes have shown significant reductions in the signs and symptoms of acute allergic conjunctivitis, while placebo treated eyes have not. Placebo-treated eyes have generally shown negligible improvement relative to drug-treated eyes.
Thus, the values corresponding to the placebo-treated eyes posttreatment in this study are noteworthy. The highly significant p values, coupled with mean changes of 1.0 clinical unit, comparable to those in the active treated eye, are reminiscent of the crossover effects that have been observed with other compounds.
Consensual effects have been noted in previous animal studies following installation of a chemical into the eye, following experimental trauma, and following intracranial stimulation of the trigeminal nerve. One theory proposes that chemical mediators reach the fellow eye via the circulatory system. A second hypothesis suggests that centripetal neural impulses from the treated eye reach the fellow eye by either direct or antidromic neural transmission.
Conclusion
One to two drops of Similasan Eye Drops #2 administered q.i.d. for two weeks appears to significantly reduce the hyperemia and itching associated with allergic conjunctivitis. Our evaluation of safety showed this compound to be non-irritating. Recognizing the need for further investigation to elucidate these findings, we are conducting a double-masked, parallel group controlled crossover study in the same models.