This split-mouth, randomized controlled clinical trial (Chinese Clinical Trial Registry, ChiCTR2000029831) of 6-month duration was conducted in the Department of Periodontology, Hospital and School of Stomatology, Peking University, China. All procedure followed the Consolidated Standards of Reporting Trials (CONSORT) guideline.
Subjects selection
This study protocol was documentarily approved by the Ethics Committee of Peking University School of Stomatology (No. PKUSSIRB-201944047) and was conducted in accordance with the Helsinki Declaration of 1975, as revised in 2013. 20 patients were recruited into the study, depending on the following inclusion criteria: (1) patients diagnosed as generalized chronic periodontitis with CAL loss in > 30% of sites [22] and (2) patients systemically healthy, without any disease compromising wound healing. Exclusion criteria were as follows: (1) smoking, (2) pregnancy or lactation, (3) systemic antibiotics taken within previous 6 months, (4) subgingival scaling received within previous 6 months and (5) less than 10 maxillary teeth except for wisdom teeth. All periodontal sites associated with a PD ≥ 4 mm and bleeding on probing were selected as study sites in maxillary dentition of each subject, except wisdom teeth and teeth considered hopeless or to be extracted. Patients received a description of the study design and provided written informed consent before enrolled in the study.
Ultrasonic equipment
A piezoelectric ultrasonic scaler (SKL A7) fabricated two coolant containers was sponsored by SKL Medical Instrument Co. Ltd. (Guangdong, China), which could generate ultrasonic vibration at a frequency of 24–33 kHz. Two different subgingival scaler tips used in treatment represented a conventional design (P3, SKL, Guangdong, China) and a new-type terminal-outlet design (Patent No. ZL201820316091.3) which allows coolants to be delivered through the tip and ejected from the apex outlet in addition to ultrasonic vibration (Fig. 1). The flow rate of coolants (either water or EGCG solution) was turned up to maximum during ultrasonic scaling.
EGCG testing and solution preparation
The EGCG used in this study was well-purified Sunphenon® EGCG purchased from Taiyo Green Power Co. Ltd. (Wuxi, Jiangsu, China). Determined by high performance liquid chromatography, the purity of this EGCG product was higher than 92%. Within 30 min before each treatment, 5 mg/mL EGCG aqueous solution was prepared with distilled water and stored in one labelled 500 mL coolants container, which was masked and with the cap tightened to avoid light and oxygen interference. The other labelled coolants container was filled with distilled water as control.
Clinical examination
Primary outcome: PD and CAL were measured at 6 sites per tooth (disto-buccal, mid-buccal, mesio-buccal, mesio-lingual, mid-lingual and disto-lingual), using a Williams periodontal probe (Hu-Friedy®, Chicago, United States) with 1 mm increment. The discrepancy from the gingival margin and cemento-enamel junction to the base of the periodontal pockets was recorded to the nearest millimeter as PD and CAL, respectively.
Secondary outcome: bleeding index (BI, Mazza, 1981) and plaque index (PLI, Silness & Löe, 1964) were measured at 2 sites per tooth (buccal and lingual) and the scores from 0 to 5 and 0 to 3 were assessed respectively.
The intra-examiner reproducibility had been calibrated within 2 patients, 336 sites, prior to commencement of the measurement. The intraclass correlation coefficient (ICC) of PD and CAL was calculated as 0.969 and 0.959 respectively in duplicate measurements (n = 336), 30 min apart.
Clinical protocol
Before enrollment, the randomized allocation of maxillary contra-lateral quadrants was determined with a coin-toss method in each subject by the investigator (Q.Y.). All patients received proper oral hygiene instruction and full-mouth supragingival scaling at the first visit. After 2 weeks, baseline recording of clinical parameters and microbiological sampling were performed by the examiner (Y.W.) who was blinded to the allocation. Then, the subgingival treatments were performed by the operator (J.Z.) and accomplished in 4 sessions arranged 1 week apart in each patient. As the split-mouth study design, the left or right maxillary quadrant sites with PD ≥ 4 mm was scaled and root planed with either EGCG solution or distilled water as coolants of the ultrasonic scaler. The sequence of each quadrant treatment was as follows: removing massive calculus and subgingival plaque with ultrasonic scaling (P3 Scaler Tip, SKL), thorough scaling and root planing to the bottom of periodontal pockets with hand instrumentation (Gracey Standard Curettes, Hu-Friedy®, Chicago, United States) until tooth surfaces were smooth and hard, then ultrasonic scaling at least 1 min per tooth (New-type Scaler Tip). Average 30–40 min was demanded for each quadrant where local infiltration anesthesia was conducted if necessary. The volume of either EGCG solution or distilled water consumed in each treatment was approximately 300–500 mL. The patients were informed to report if any discomfort related to the treatment was felt within 1 week.
The patients were recalled 3 and 6 months after treatment. All clinical parameters were recorded as baseline and microbiological samples were taken by the examiner (Y.W.), and the oral hygiene was reinforced when necessary. If there still existed study sites associated with PD ≥ 4 mm, the patients would receive ultrasonic scaling using the new-type scaler tip at a frequency of 24–33 kHz in accordance with the former allocation.
Microbiological procedures
Based on the baseline clinical examination, a pair of study sites with PD ≥ 4 mm within contra-lateral premolars or the first molars were selected for subgingival plaque sampling. These selected sites were consistent among baseline, 3-month and 6-month re-evaluation. The subgingival plaque samples were collected by paper points method and stored at − 80 °C for microbiological testing. The PCR products of bacterial 16S rRNA were generated and sequenced after DNA isolation, as described in Liu et al. [23]. After annotated against the Human Oral Microbiome Database [24], the mean relative abundances of red complex pathogens were calculated in subgingival microbiome.
Safety of treatments
The patients were instructed to report if any discomfort was felt during each treatment and 1 week after all treatment procedures. The adverse effects associated were as follows: (1) postoperative pain and dentinal hypersensitivity; (2) staining of tooth surfaces associated with medication; (3) subgingival emphysema related to the new-type scaler tip, which might manifest as noticeable swelling and crepitus of soft tissues; (4) fracture of the new-type scaler tip, which was monitored by checking post-treatment length and shape changes with the customized alginate impression of each labeled tip.
Sample size calculation
A sample size of 20 patients was estimated according to the previous study which applied a sustained-release catechins mixture as an adjunct to SRP [18], thus 393 and 369 sites were obtained from 18 patients (2 drop-outs) in two groups respectively. Then a post hoc analysis at a two-side type I error of 0.05 (α) and 80% power of detection (1 − β) was conducted based on the primary outcome PD at month 6, determining that 287 sites per group would be necessary.
Statistical analysis
The effect of different treatment groups on these clinical parameters at each interval was compared by Student’s t test. Within each group, the Bonferroni method (Paired-sample t test) was performed to compare the clinical parameters at different intervals. Pearson chi-square test was performed to compare the percentage of treatment sites between groups. Comparison of relative abundance in subgingival microbiome was performed with Wilcoxon signed rank test. Data processing and analysis was performed on SPSS 23.0 (IBM, Chicago, United States) and R (Stats package; R Foundation for Statistical Computing, Vienna, Austria). p-values < 0.05 were accepted for statistical significance.