Effectiveness of Three Different Mouthrinses in Dental Plaque Control and Early Wound Healing

Materials and Methods

Patients' enrollment and all clinical procedures were conducted at the Department of Periodontology, School of Dentistry, National and Kapodistrian University of Athens, Greece.

Clinical Procedures:

All surgical interventions were conducted by first and third year postgraduate residents. Supragingival debridement and polishing of the operated area was performed immediately before surgery. In cases of residual periodontal pocket elimination / reduction surgeries, intrasulcular incisions followed by papilla preservation incisions at interdental areas were applied. In crown lengthening cases, resective gingival incisions were applied in presence of sufficient width of keratinized tissues (≥ 3mm). Following flap elevation, root surfaces were thoroughly debrided with instruments and ultrasonic devices. Ostectomy / osteoplasty with carbide burs was applied, as deemed appropriate. Buccal and palatal / lingual flaps were fully adapted to each other and sutured together with monofilament slowly resorbable suture. Periodontal dressings were not used. No systemic antibiotics were prescribed. Ibuprofen 400mg, three times daily for 4 days, was suggested immediately after surgical intervention. Smokers were advised to cease smoking for the first postsurgical week. Sutures were removed at 7th postoperative day. During that period no accidental loss of sutures was recorded.

Randomization and allocation concealment:

Subjects were randomly assigned to one of three groups using a computerized assignment procedure (Random Sequence Generator, www.random.org) generated by the examiner A.G. Each subject was given an identical number between 1 and 42. Participants received bottles with one of the three solutions under investigation. Neither patients nor clinicians (surgeons and examiner A.G.) were informed about the identity of mouthrinse administered in each case (double - blind design). Masked bottles were given to participants by a third independent person who was not a dentist (member of Postgraduate Clinic's assistance staff).

Post - surgical maintenance protocol:

During the first 14 postoperative days, participants were instructed to rinse with 15ml of the administered solution for one minute, twice daily.

During the same period, subjects refrained from any mechanical plaque removal at the operated areas.

Intra-examiner reproducibility:

Prior to the beginning of the study, the examiner (A.G.) who performed the measurements was calibrated in order to establish intra - examiner reproducibility. Calibrations were performed with two subjects not included in the study. During post-operative maintenance using 0.12% chlorhexidine, at the 7th postsurgical day, early wound healing index (EHI) and plaque index (PI) were recorded on two separate occasions during the same day, however at least 8 hours apart. Intra-class correlation analysis was used to calculate intra-examiner agreement for repeated measurements. The calibration was accepted if both measurements were similar with a deviation of +/- 10% (intra-class correlation coefficient > 0.9).

Microbiological assessment:

Supragingival dental plaque pooled samples were collected at the 14th postoperative day from both interproximal tooth surfaces respectively to the interdental papilla with the highest EHI value, using curettes. Pooled samples were individually collected in 150μl TE buffer solution. Samples were stored at -80 until processing, which was performed at the Laboratory of Cell and Matrix Pathobiology at the Institute of Biosciences and Applications, National Center for Scientific Research (N.C.S.R.), Athens, Greece. Bacterial genomic DNA was isolated and purified using the PureLink® Genomic DNA Μini Kit, following the Gram (+) bacterial cell lysis manufacturer's protocol. Purified DNA concentrations were determined spectrophotometrically (NanoDrop 1000). Specimens were analyzed using quantitative real time PCR (qPCR) in order to evaluate the total bacterial count in each pooled sample. qPCR was conducted with application of a set of universal primers (forward primer: 5' TCCTACGGGAGGCAGCAGT 3' and reverse primer: 5' GGACTACCAGGGTATCTAATCCTGTT 3', with an amplicon size of 466 base pairs), reported to detect DNA sequences of gene encoding for 16S ribosomal subunit of at least 50 different bacterial species (aerobic and anaerobic Gram positive and Gram negative) and SYBR Green Ι fluorescent probe. Serial dilutions of DNA extract of pure bacterial cultures of Escherichia coli were performed in order to generate a standard curve for the qPCR. Reactions were performed for 40 cycles at 94 °C for 30 sec, 58 °C for 45 sec and 72 °C for 10 min in a Light Cycler 96 qPCR Device. Results were analyzed using the Light Cycler 96 1.1 Software. Total bacterial count of each sample is expressed as total bacterial DNA mass (ng) and bacterial copy numbers (x106).

Sample size calculation:

Sample size was estimated through one-way ANOVA based calculations, assuming that the minimum clinically significant difference is one unit in median EHI values at 14 days between at least one pair of groups. 14 patients for each group were needed for 80% power achievement and for median differences detection in under evaluation parameters among the groups.

Statistical analysis:

Demographic and clinical characteristics of the sample are presented by group in tables containing absolute and relative (%) frequencies for categorical variables and medians and interquartile ranges (Median - IQR) for quantitative variables. Between groups differences in these characteristics were assessed through Fisher's exact tests for categorical variables and Kruskal-Wallis tests for quantitative ones. The distributions of the indices of interest (Early wound healing index - EHI, Plaque index - PI, Plaque area index - PA%) were summarized through medians and IQRs.

Indices' data were analyzed through linear or generalized linear mixed models taking into account the potential correlations between multiple measurements taken on the same individual. In particular, for the analysis of EHI and PI, ordinal logistic regression models were used. For the PA% index, which was a proportion ranging from 0% to 100%, a linear model was used after a logit transformation of the dependent variable. Model selection, including checks for potential interaction effects, for the final multivariable models was based on likelihood ratio tests.

Total microbial count differences among groups or in relation to other sample characteristics were analyzed using linear regression model after log10 transformation. Results are presented as relative differences %.

P-values less than 0.05 were considered statistically significant. All analyses were performed using the statistical package Stata 13.1 (Stata Corp., USA).