Specific Objectives and Hypotheses

The aim of this study was to compare the effectiveness of application of 0.8% metronidazole gel or mobile telephone SMS oral hygiene reminders with placebo gel on gingivitis in 66 adult patients for a period of 4 weeks. The null hypothesis was that there would be no difference between the interventions (0.8% metronidazole gel and mobile telephone SMS reminders) on reducing the gingival inflammation in orthodontic patients.

Trial Design

This study was a three-arm parallel, randomized, double-blind controlled trial with an allocation ratio of 1:1:1. Ethical approval was obtained from the Ethics Review Committee of the Aga Khan University Hospital (5320-Sur-ERC-18). The trial was registered in the clinical trials registration database (ClinicalTrials.gov; NCT03508999). There were no deviations to the protocol after trial initiation. The study was conducted according to the World Medical Association's Declaration of Helsinki and the principles of Good Clinical Practice.

Participants, Eligibility Criteria, and Settings

The sample was collected from the outpatient orthodontic clinic from May 2018 to November 2018. Eligibility criteria included patients aged between 18–40 years with moderate to severe crowding undergoing orthodontic treatment for at least 6 months with all teeth mesial to the first molars bonded; systemically healthy patients with no comorbidities such as rheumatic fever, blood dyscrasias, congenital heart disease, or diabetes mellitus; and patients with gingivitis as assessed by GI (score ≥ 2), BI (score = 2), and OPI (score ≥ 2) on any three standard teeth (teeth numbers 13,15, 21, 33, 35, 41). Only those patients who gave written consent were included in the study.

The exclusion criteria were women who were pregnant or lactating and patients with histories of surgical or nonsurgical periodontal therapy in the past 6 months, use of antibiotic or anti-inflammatory drugs in the past 30 days, smoking, or allergy to metronidazole; clinical attachment loss of greater than 2 mm at two sites; or removable or fixed dental prosthesis.

Interventions

All patients were treated with the same fixed orthodontic appliances using metal brackets (Di-MIM Bracket System, Ortho Organizers, Carlsbad, Calif) on the incisors, canines, and premolars and bands on the first molars. Bonding of the brackets was completed by orthodontic residents (H.T.M., F.F., S.I., D.M., E.F.) with a direct bonding approach and use of chemical cure adhesives (3M Transbond, Saint Paul, Minn). The type of ligation for each patient was conventional (elastomeric ligatures). General oral hygiene instructions were given verbally as well as provided in a brochure to each participant of the study by the treating clinician.

Group A participants were given 0.8% metronidazole gel. Group B participants were given placebo gel and biweekly SMS reminders for reinforcing oral hygiene (“Please keep brushing twice daily with a soft bristle toothbrush and fluoridated toothpaste.”). Group C participants were given placebo gel (control) only. All patients were instructed to apply a pea-sized amount of gel on the gingiva twice daily for 4 weeks. Patient compliance for gel use was assured by providing a log sheet and further checked by having them return the tubes of gels. Another log sheet given to each participant was used to confirm whether they received the reminder SMS at biweekly intervals.

The 0.8% metronidazole gel contained metronidazole tablets and 2% carboxy-methyl cellulose as a base solution. The placebo gel contained similar ingredients except the metronidazole tablet. The preparation, packing, labeling, storage, and dispatch of the gel were performed by the clinical trial unit (CTU).

Outcomes

The primary outcome was measurement of the indexes during the study period from baseline (T₀) to 4 weeks (T₁) of metronidazole gel/reminder SMS therapy (Figure 1). GI¹³ and BI¹⁴ were evaluated by assessing and probing the gingiva. OPI was evaluated by using plaque disclosing tablets (Produits Dentaires SA, Vevey, Switzerland).¹⁵

View Figure

• Download as Powerpoint
• Download Figure

Six standard sites on the incisors, canines, and premolars were used in this study as described by Gettinger et al.¹⁶ The banded first molars were not included because banding itself was expected to lead to a compromise in periodontal conditions. The study sites were assessed using a Williams probe and included six proximal line angles on the following teeth: right maxillary second premolar, mesiobuccal line angle; right maxillary canine, distobuccal line angle; left maxillary central incisor, distopalatal line angle; right mandibular central incisor, distolingual line angle; left mandibular canine, distobuccal line angle; and left mandibular second premolar, mesiobuccal line angle. If a study tooth was missing, the corresponding tooth on the contralateral side was examined. There were no changes in the outcome assessment after trial initiation.

The patients were included in the trial if GI, BI, and OPI scores were found to be ≥2 on any three or more of the six teeth that were evaluated in the trial. The cumulative GI and BI indexes were obtained by calculating the average of the measured values at T₀ and T₁. For OPI, the score at each time interval was obtained by counting the number of teeth with plaque accumulation.

Sample Size

Sample size was calculated using an OpenEpi (version 3.01 developed by Emory University, Atlanta, GA.) sample size calculator. Martin et al.¹⁷ reported that a mean GI at 4–6 weeks in the gel group was 1.56 ± 0.14, whereas in the control group it was 1.68 ± 0.12. With that difference at a level of significance (α) of 5% and power of (1-β) 95%, at least 19 observations were needed for each arm. The sample size was inflated by 10% to get 21 participants per arm for any loss to follow-up or noncompliance. Because there were three experimental groups, a total of 63 participants were needed.

Interim Analyses and Stopping Guidelines

None.

Randomization (Sequence Generation, Allocation Concealment, Implementation)

Participants were assigned to one of the three study groups using a computer-generated randomization list. The randomization was performed by CTU using a random permuted block sampling of 6 and 9.

Patients recruitment was performed by one pair of investigators (H.T.M. and F.F.) who explained the aims and objective and the design of the study to the patients. Allocation was concealed from the patients and investigators by labeling the gels with the same prescription and using sequentially numbered codes. All of the measurements at T₀ and T₁ were recorded by the second set of investigators (S.I., A.Y., and E.F.) on separate T₀ and T₁ sheets. The reminder SMS was sent by another clinician (D.M.) who was not part of the recruitment stage.

Blinding

The trial was double blinded (patient and investigator), and only the CTU pharmacist was unblinded in the study. The randomization codes were not revealed until data analysis was performed.

Statistical Methods

Data were analyzed using SPSS version 22.0 (UNICOM Systems, Inc., Los Angeles, Calif.). Descriptive statistics for all baseline clinical parameters such as GI, BI, and OPI scores were calculated. To test the normality of data, the Shapiro-Wilk test was applied, and it showed normal distribution; hence, parametric tests were applied. Paired-sample t-tests were applied to compare the mean differences (T₀–T₁) of GI, BI, and OPI in groups A, B, and C. To evaluate the effect of interventions with controls, independent-sample t-tests were applied among the GI, BI, and OPI variables in groups A vs C and group B vs C. The level of significance was kept at P ≤ .05.

Participant Flow

A total of 70 patients were assessed for eligibility in the trial, of which 4 declined to participate. Thus, 66 patients were randomly assigned with 22 patients per group. There was one drop out in both groups A and B (Figure 2).

View Figure

• Download as Powerpoint
• Download Figure

Baseline Data

The number of female participants was higher in each group because of the fact that a greater number of women sought orthodontic care. The mean ages of the participants in each group were comparable (Table 1).

Table 1. Baseline Demographic Characteristics of the Participants^a

View Fullscreen

Numbers Analyzed, Outcomes, and Estimation

The means and standard deviations (SDs) at T₀ and T₁ of groups A, B, and C are given in Table 2. There were statistically significant differences (P < .05) in the values at T₀ and T₁ of GI, BI, and OPI in all groups. The values at T₁ showed improvements in all indexes for each intervention. All interventions showed resolution of gingival inflammation. Greater reductions in OPI scores further improve the GI and BI indexes.

Table 2. Comparison of Means and SDs of GI, BI, and OPI Scores in Metronidazole Gel, Reminder SMS, and Control Groups at T₀ and T₁^a

View Fullscreen

Although there were statistically significant differences within each intervention between T₀ and T₁, the group-wise (group A vs group C and group B vs group C) comparisons of GI, BI, and OPI values showed no significant differences (Tables 3 and 4). Hence, all interventions were similarly effective in the resolution of gingivitis.

Table 3. Comparison of Means and SDs of GI, BI, and OPI Scores in the Metronidazole Gel and Control Groups^a

View Fullscreen

Table 4. Comparison of Means and SDs of GI, BI, and OPI Scores in Reminder SMS and Control Groups^a

View Fullscreen

Harms

None.

Main Findings and Interpretations

In this trial, no significant differences were found in the effectiveness of either metronidazole gel or reminder SMS therapies. The comparison of metronidazole gel and reminder SMS therapy with the control group did not lead to superior results for any intervention. Effective reductions in all index scores were also found in the placebo group. As standard oral hygiene instructions were given to each participant of the trial, the similar positive effects with all interventions could also have been attributed to patients' proficient compliance and management of oral hygiene.

The factors related to the clinically acceptable effects in the placebo group could have been attributed to patient expectations toward outcome, patient–doctor relationships, or the Hawthorne effect.^18–20 This effect was frequently reported in various studies as the sole criterion for oral health improvements of control groups that received placebo treatments.^21,22 Feil et al.²¹ also reported a reduction in plaque scores with intentional application of the Hawthorne effect and changes in patients' behavior and compliance. The effective reduction in all indexes in each group could have been attributed to the Hawthorne effect.

Numerous researchers have reported significant improvement in the GI, BI, and plaque index with the topical application of 2% chlorhexidine,²³ 0.4% stannous fluoride,²⁴ and antioxidant-essential oil gels.²⁵ Pradeep et al.²⁶ reported a reduction in the GI and plaque index and microbiological count after 24 weeks of application of 10 mg metronidazole gel. Perinetti et al.²⁷ reported a decrease in bleeding on probing, probing depth, and clinical attachment levels after four weekly applications of 1% metronidazole gel. In the current study, improvements in the GI, BI, and OPI scores was also found after 4 weeks of application of 0.8% metronidazole gel. Nevertheless, the previous studies were conducted in nonorthodontic patients, and no other study had previously evaluated the effects of this gel in orthodontic patients.

In orthodontics, active reminders are persuasive in keeping up the good rapport between doctors and patients, decreasing the level of self-reported pain by the patient and improving oral hygiene compliance.^12,28 Bowen et al.²⁹ sent biweekly SMS reminders for 4 weeks about oral hygiene compliance and found significantly lower plaque coverage around teeth. Iqbal et al.³⁰ sent weekly SMS reminders for 3 months and found significantly lower BI and plaque and modified gingival index scores in their studied population. In the current trial, biweekly SMS reminders were sent for 4 weeks and similarly resulted in significantly lower GI, BI, and OPI scores.

Limitations

The main limitation of this study was the relatively short follow-up period. There were more female participants in each arm, and their presence may have confounded the results as women have previously been shown to be generally more compliant.³¹ Metronidazole gel may result in a metallic taste, which may cause nausea. Participants were informed of the taste in the written consent form, and compliance with using the gel was ensured by having them fill the log sheets. The concentration of metronidazole gel that was used was also unique and not readily available in every health care setting.

Generalizability

The generalizability of the results could be limited because the study was conducted in one center only. The strength of the study was that there was limited attrition of the sample. The double blinding of the trial reduced observational and detection biases. In addition, a pair of clinicians was involved in the recruitment stage and another set of clinicians were involved in obtaining the measurements of the indexes_, leading to further reduction in selection and measurement biases. Furthermore, the clinician who sent the reminder SMS was not part of the trial participant recruitment.

Recommendation

Additional trials for testing the efficacy of metronidazole gel and reminder SMS at biweekly intervals with a greater follow-up period are needed to confirm the reported findings of this study and to determine the maximum effectiveness of such interventions.