This randomized, controlled, parallel-group clinical trial was performed from June 2017 to July 2018 in the Department of Periodontology of the Second Affiliated Hospital of Zhejiang University School of Medicine, Hangzhou (SAHZU), China. This randomized trial followed the CONSORT 2010 Checklist protocol[21]. The study protocol was approved by the Human Research Ethics Committee of SAHZU (NO. 2017037) and registered at the Chinese Clinical Trials Registry (No. ChiCTR-IPR-17011678, http://www.chictr.org.cn/) before patient enrolment and conducted in full accordance with the Declaration of Helsinki.
Inclusion and exclusion criteria
Each participant was fully informed about the research and signed the written consent prior to enrollment. The patient inclusion criteria were: (1) aged 18 to 60 years; (2) good general health without systemic disease; (3) no smoking history or quitting smoking for at least 2 years; (4) with a diagnosis of moderate or severe periodontitis [22, 23], who needed to be treated by non-surgical periodontal treatment (scaling and root planing) in one session. Briefly, moderate periodontitis was diagnosed as clinical attachment level (CAL) = 3 or 4 mm and probing depth (PD) ≥ 4 in ≥ 2 non-adjacent teeth, and severe periodontitis was diagnosed as CAL ≥ 5 and PD ≥ 5 in ≥ 2 non-adjacent teeth; (5) with at least three teeth in the buccal face existing dentine hypersensitivity [visual analogue scale (VAS) ≥ 2.0] after non-surgical periodontal therapy, which was evaluated by air-blast test. The patient exclusion criteria were: (1) patients with gross oral mucosal disease (oral lichen planus, oral ulcer, et al), reflux or bulimia, extremely advanced periodontitis who cannot endure one session non-surgical periodontal therapy; (2) patients receiving surgical or non-surgical periodontal therapy within 12 months; (3) patients using desensitizing agents in the past 6 months; (4) hypersensitive teeth with mobility greater than 1°; (5) hypersensitive teeth which were the second or third molars; (6) hypersensitive teeth with extensive and/or defective restorations, suspected caries, pulpitis or cracked enamel; (7) patients with chronic use of antihistamines, anticonvulsants, sedatives, antidepressants, tranquilizers or daily analgesics within 1 month; (8) pregnant or lactating females; (9) patients presenting allergies to the test product, or ever allergic to oral care consumer products; (9) patients who have participated in another desensitizing dentifrice study.
Sample Size Estimation
The main outcome was the VAS difference across groups between the mean changes in air-blast test from baseline evaluation to the end of the follow-up. According to previous data [24, 25], the expected baseline mean VAS score was 5.5±2.0. The hypothesized mean VAS score at the end of the follow-up was 2.0±1.9 for the test dentifrice and 3.5±1.8 for the control product, and the mean changes were 3.5±1.8 and 2.0±1.5, respectively. Using an unpaired t-test and assuming an α-error = 0.05, power = 80% (two-tailed comparison), a minimum of 21 participants per group were requested using a single allocation ratio (1:1) (G*Power version 3.1 for Mac, Franz Haul, University of Kiel, Germany). To compensate for the possibility of 15% dropouts, 48 patients per group were aimed to be recruited.
Non-surgical periodontal therapy
The enrolled patients received full-mouth scaling and root planing in one session. The non-surgical periodontal therapy was performed using ultrasonic scaler (P5 Newtron, Acteon Satelec, France) combined with hand instruments (Gracey curettes, Hu-Friedy, Chicago, USA) after local anesthesia by 4% atticacaine and 1/100,000 adrenaline (Primacaine, Produits Dentaires Pierre Rolland, France). After the therapy, chairside irrigation with 0.12% Chlorhexidine (Koutai, South China Pharmaceutical, China) for 1 min. All patients were taught to brush teeth by modified bass technique using the same kind of soft toothbrush (Systema, Lion, Japan) provided, twice a day for 3 minutes during the 6-week trial.
All periodontal treatments were performed by the same experienced periodontist PHD, who was masked from the patient allocation during the whole study.
Clinical evaluation
Baseline periodontal parameters were recorded before non-surgical periodontal therapy. Periodontal examination was performed for each tooth including gingival recession (GR), probing depth (PD), bleeding on probing (BOP) and clinical attachment level (CAL).
Baseline air-blast evaluation (post-scaling evaluation) for each tooth was taken 12-24 hours after scaling and root planing. First, the tooth was isolated by the cotton roll from the adjacent teeth. Next, a blast of air from a standard dental unit syringe at 60 ± 5 psi at 18-22 °C was directed onto the exposed middle 1/3 buccal surface for 1 second at a distance of approximately10mm. Then, each patient reported the sensitivity he/she sensed using the Visual Analogue Scale (VAS) [26] and Schiff Cold Air Sensitivity scale [27]. VAS scale was scoring from 0 (no pain) to 10 (intense pain). Schiff scale was scored from 0 to 3: score “0” means no response, score “1” means response without request of discontinuation of stimulus, score “2” means response with request of discontinuation of stimulus, and score “3” means pain with request of discontinuation of stimuli.
According to the baseline evaluation, patients who presented at least three hypersensitive teeth with baseline VAS ≥ 2.0 were suitable to be included. 48 patients were sequentially enrolled by the dentist AD and randomly assigned to either test or control group.
Once enrolled, each patient would immediately receive one consecutive 5-second polishing of the assigned dentifrices by rubber cup at a moderate speed (about 1000 rpm) to all teeth by the same dentist JPH. The dentifrices were as follows: (1) test group: n-CAP dentifrice containing 20% n-CAP (Dentiguard Sensitive, Daewoong Co, Korea); (2) control group: calcium carbonate-based dentifrice free of n-CAP or other desensitizing ingredients (Honghua, Saky, China). Both of the dentifrices were without any form of fluoride. The two dentifrices had identical appearance, which were over-wrapped to hide their original packages and labeled with different numbers from 1 to 48. Neither investigators nor patients knew the codes. After application, each patient was subjected to post-polishing evaluation (0-week evaluation) of DH. Thereafter, patients would be called back to evaluate DH after 2, 4 and 6 weeks. At the 6-week follow-up, the same periodontal examination as baseline was repeated. Participants were instructed to use only the assigned products to brush teeth twice a day throughout the 6-week trial. They were also instructed not to eat acidic food before toothbrushing and one hour before the DH evaluation. Participant adherence to dentine hypersensitivity treatment was evaluated by inquiry and examination of the remaining volume of dentifrice at each follow-up.
All periodontal and hypersensitivity outcomes were measured by the same experienced examiner HJH, who had been well-trained before patient enrolment. The same procedure as employed at baseline was used as the standardized method throughout the clinical trial. At each visit, the occurrence of potential adverse effects was assessed by investigators by both intraoral examination and patient inquiry.
Randomization and allocation concealment
Simple randomization was adopted in this study. Random allocation list was generated using a computer program (Rand function, Excel 2016 for Mac, Microsoft, Redmond, VA, USA) by JML before patient recruitment. The assigned dentifrices were saved in opaque envelopes in advance. Other investigators only knew the number over the envelope but not the allocation sequence, and thus they were blind to the group allocation during the whole research period.
Statistical analyses
The data of patients who adhered to the assigned intervention and completed the predefined process were included in the statistical analysis. Teeth with baseline VAS ≥ 2 and baseline Schiff score ≥ 1 were included. The normality of data was assessed by Shapiro-Wilk test. Data with skewness was converted through logarithmic conversion before analysis. Intragroup and intergroup comparisons were analyzed by paired t test and independent t test, respectively. The results of VAS and Schiff score were analyzed using the mixed linear model adjusted to age, gender, tooth type as well as the number of teeth per participant contributed to the air-blast test. The corresponding baseline results of VAS or Schiff were covariates. All statistical analyses were performed using SPSS (version 24.0 for Mac, Chicago, IL, USA). A two-tailed p value < 0.05 was considered significant. Primary outcomes were change of VAS and change of Schiff at all evaluation stages.