Use of Topical NSAID to Reduce Pain in Oral Lichen Planus and Oral Lichenoid Lesions.

A four-week randomized, double-blind, cross-over, placebo-controlled trial was planned. Ethical approval was obtained from the Institutional Review Board (IRB) at the University of Washington before the start of the study, participants agreed to participate in this study had to sign a written informed consent. The present trial was also registered (#NCT03509675). The research was conducted in accordance with the ethical principles stated in the World Medical Association's Helsinki Declaration.

A coin was flipped to decide which medication would be used first to start the randomization. Based on the coin toss, drug A was chosen for the first block of two participants. Then, block randomization was further assigned by blocks of two, alternating the drugs. The medicines were distributed in identical plastic containers, packed by an external pharmacist who was unaware of the protocol and successfully labeled as A and B containers.

During the intervention, neither investigators nor the participants knew which of the treatments they were using.

Population and Study Sample

Potential participants were screened through the Oral Medicine patient registry at the University of Washington. Prospective patients aged 18-80 were examined by the attending at the Oral Medicine Clinical Services (OMCS) faculty. All potential study participants with a clinical diagnosis of symptomatic oral lichenoid mucositis with or without a biopsy were identified. Individuals were eligible for inclusion if they met the following criteria:

• Speak English
• Have a symptomatic form of the disease, pain level of 3 out 10 on VAS
• At least 18 years of age

The exclusion criteria were as follows:

• The occurrence of dysplasia in the histopathological specimen
• Known or suspected sensitivity to NSAID medication
• History of asthma
• History of gastrointestinal ulceration
• History of bleeding disorders
• Pregnancy

The patients were asked to participate in the research at their first or follow-up visits at the OMCS.

Sample Size and Sample Selection

The most similar study to the present one was done by Saxen et al., in which a 45% change in VAS scores between the treatment groups was found. With the assumption of a 45% difference in VAS scores, a standard deviation of 0.69, a two-sided significance level of 0.05, and a statistical power of 80%, a sample size of 36 was calculated. When accounting for the effect of a cross-over trial and analyses, a sample size of 36 should be able to show significance with a change in VAS scores between 25% and 30%, significantly less than that found in the work of Saxen et al.

Collection of Data

The data collected included the following:

• VAS at baseline, end of day 4 and following the end-of-the-week application of the active intervention and the placebo
• Gender
• Race and ethnicity
• Duration of the primary chief complaint
• Medication trials for the condition
• Current medical conditions
• Current prescription drugs
• Current OTC drugs
• Known allergies to drugs or food substances
• Clinical diagnosis of the disease
• Classification of the OLP (if noted): reticular, erythematous, erosive, bullous Topical Formulation

The topical suspension of the topical NSAID was 100 mg per 5 ml concentration of ibuprofen, with similar ingredients as OTC children's ibuprofen and was compounded by an external drug service. Another placebo suspension was also compounded with the same taste but without the active ingredient.

Assessment

We used tracking forms for the participants to record both their daily usage and VAS scores at baseline, day 4, and day 7. (see Appendix A). Participants were asked to mark their pain level before starting the first bottle in the morning of day one. 100mm VAS scales were used with the anchors of (non-intense) on the right side and (extremely intense) on the left. The patients were seen for a free assessment to evaluate their condition during the intervention and report any adverse reactions.

Intervention

After we obtained written consent from the patients regarding their participation in the research, they received a research packet with the relevant data forms, the consent form, and two bottles; each bottle had an individual label marked wither "A" or "B". The coding which identified the contents of A and B was sealed by the pharmacy and was kept with one of the committee members as well as with the pharmacy

The participants were asked to record their baseline score of spontaneous pain on a horizontal 10 cm VAS before commencing the use of the provided rinses. If the participants were already on active treatment at the time of enrollment, they were asked to discontinue for 7 days prior for a wash-out period before starting the research protocol.

All participants were instructed to use the first suspension four times a day for 7 days. They were instructed to rinse before the meals; breakfast, lunch, dinner and also before bedtime. After every application of the rinse, they were asked to check a box in order to record their use of the rinse. The instructions included rinsing with 5 ml of the suspension for 1 minute without swallowing it, and then expectorating. The patients were instructed not to eat or drink for the following 20 min after the application of the drug. At the end of days 4 and 7, the participants were asked to record their spontaneous pain level on the VAS. After the first week, they discontinued any treatment for 7 days (wash-out) before starting the second suspension.

All participants were instructed to use the second suspension on the same schedule as the first, and with the same instructions. The participants were contacted initially after the first day of the intervention to discuss any concerns or questions they may have. Every week, reminder phone calls were made for them to fill out the forms from the investigator and to check for any side effects from the intervention.

Data We protected the patients' information by assigning each individual with a study number that was correlated with the medical record number on a master list. The master list was secured in a password-protected file on a password-protected computer stored in a locked office. The collected data were stored separately. The list was not placed on a portable device. Only the research committee members and the personnel required for the statistical analysis had access to the files.

The study data will be kept indefinitely but will only be linked until December 31, 2019. Only the investigators will have access to the identifiable data unless otherwise required by law.

Data Analysis Strategies

The VAS results were measured in millimeters and entered as whole numbers. T-tests for means were used to compare VAS scores at the following time points: baseline, days 4 and 7 of treatment condition # 1, at baseline #2 (following the wash-out period), and at days 4 and 7 of treatment condition #2.

Multiple linear regression was used to determine the difference in effect between the active medication and placebo groups by including the difference between reports at baseline, day 4 and at day 7. Both univariate models and models including age and duration of chief complaint were used to assess the effect of these variables. There was insufficient variance in both gender and race to assess for these effects. Although randomization of the treatment order should eliminate order effect, tests for order effect were performed using regression analyses.

For drug effect assessment, in addition to using VAS scores themselves, percent change in VAS scores was used to normalize distributions.

Ethics and Human Participant Issues

All experimental protocols were reviewed and approved by the institutional review board IRB and Committee on human participant research at the University of Washington.