Removing benzalkonium chloride may not reduce irritation

Switching to another prostaglandin analogue that does not contain benzalkonium chloride (BAK) may not be the answer for managing patients complaining of dryness and irritation while using latanoprost 0.005% (Xalatan, Pfizer), according to the results of a prospective, double-masked, randomized clinical study conducted by eye-care professionals at Lackland Air Force Base (AFB), San Antonio, TX.

The study included 33 patients with glaucoma already being treated with latanoprost for at least 6 months who reported ocular dryness and irritation. Patients were randomized to Q.D. instillation of latanoprost in one eye and travoprost 0.004% preserved with SofZia (Travatan Z, Alcon Laboratories) in the fellow eye. Latanoprost contains 0.005% BAK and SofZia is a proprietary ionic buffered preservative system considered milder and less toxic than BAK.

Differences in scores

At each follow-up visit, mean corneal staining scores were worse in the travoprost-treated eyes compared with the latanoprost-treated eyes. The difference achieved statistical significance at 2 months. By month 3, staining had increased in the latanoprost eyes so that there was no longer any significant difference between groups.

However, OSDI results showed a significant difference favoring the latanoprost-treated eye for less dryness/irritation. This data corresponded to patients' ratings using the simplified grading scale. Additionally, in comparing dryness and irritation between their two eyes, the proportion of patients noticing no difference decreased throughout the study whereas the proportion who felt the travoprost-treated eye was worse increased, reported J. Richard Townley, MD, clinical director of ophthalmology services, Lackland AFB.

Both treatments effectively controlled IOP and no differences between groups were identified in Schirmer testing or tear break-up time.

"Ocular irritation and dryness are common complaints among patients using topical IOP-lowering medication. BAK exposure is considered a cause for this problem," Dr. Townley noted. "Animal and human studies have shown less ocular irritation using travoprost with SofZia compared with the original product containing BAK. Anecdotally, patients outside of this study [who] switched from latanoprost to travoprost without BAC report improvement. Therefore, our study results were surprising."

Patient population

"Although we can postulate several explanations to account for the findings in this investigation, the main conclusion is that further investigation in a larger patient population seems warranted," he continued.

For study recruitment, the investigators identified all patients from Wilford Hall Medical Center at Lackland AFB or Brooke Army Medical Center, Fort Sam Houston, San Antonio, TX, who had a prescription for latanoprost monotherapy. They were able to contact 180 patients, but only 89 (41%) were experiencing dryness/irritation, and only 33 were interested in participating in the study; 27 patients completed the 3-month treatment period.

"We did not know what magnitude of difference we might see between eyes and arbitrarily used a 10% difference in our power calculation," Dr. Townley said. "Based on that number, it was determined that the study would need to include about 100 patients for adequate statistical power. Therefore, we were surprised to find a statistically significant difference between treatments with our much smaller population."

Trying to account for the discrepancy between the study findings and previous research, he noted that results of a rabbit study found eyes treated with travoprost with SofZia exhibited fewer corneal changes and less conjunctival inflammation than eyes treated with latanoprost with BAK, it is possible that human eyes may respond differently to these products.

In addition, the prostaglandin analogue itself and not just the preservative may be a cause for ocular surface changes and symptomatic complaints. Repackaging of the two commercial products into similar containers in order to maintain study blinding may introduce another variable because each formulation may react with the polymer of the container to generate a byproduct that might cause irritation.

"Taking the latter possibility into account, future studies comparing these products should also use containers that have the same composition as the bottles containing the commercially available medications," Dr. Townley said.