In the present study, significant reductions were observed in the clinical parameters of peri-implant inflammation at 2, 4, 12 and 24 weeks relative to baseline after debridement with the chitosan brush seated in an oscillating dental drill piece. No progression in radiographic bone loss was reported at any of the implants at the final evaluation, and the method was thus judged safe to use in cases with mild peri-implantitis.

In comparison, a randomized clinical trial by Sahm et al. [35] compared amino acid glycine powder versus mechanical debridement using carbon curettes and antiseptic therapy with chlorhexidine digluconate. At the 6-month final evaluation, PPD reductions of 0.6 and 0.5 mm, respectively, were reported. Similarly, Renvert et al. [36] performed a randomized clinical trial comparing an air-abrasive device and an Er:yttrium aluminium garnet (YAG) laser in the non-surgical treatment of peri-implantitis and reported mean PPD reductions of 0.9 and 0.8 mm, respectively, in the two groups. In the present study, a mean PPD reduction of 1.1 mm was determined at the final evaluation at 6 months which is comparable to findings reported in other studies.

A study by Riben-Grundstrom and co-workers [37] compared the use of glycine powder air-polishing and the ultrasonic treatment of peri-implant mucositis and utilized inclusion criteria comparable to those used in the present study for mild peri-implantitis. The inclusion criteria were (1) the presence of one or more sites diagnosed with peri-implant mucositis. The diagnostic criteria used were a probing depth ≥4 mm combined with bleeding with or without suppuration and (2) bone loss ≤2 mm assessed from the implant shoulder subsequent to the bone healing and remodelling process. They used dichotomous values for BoP and observed reductions of 27% in the air-polishing group and 31% in the ultrasonic group after 6 months.

When pooling the index scores to dichotomous values in the present study, the mBoP score was found to have decreased from 2 or 3 to 0 or 1 in 55% of the samples. Although the complete absence of inflammation was difficult to achieve in most implants, significant and stable reductions in the parameters of inflammation were demonstrated in most sites up to 6 months after treatment with the chitosan brush. Nicotine interferes with the bleeding response in soft tissues and may, consequently, lead to false negatives for the BoP; therefore, to avoid positively skewing the results due to smoking, we decided to exclude current smokers from this study. According to the literature, bleeding on probing has low sensitivity as a predictor for active peri-implant disease because of the high frequency of false-positive responses, but it has a high level of specificity as no bleeding on probing indicates peri-implant health [38]. Due to the absence of perpendicular periodontal fibres in dental implants, a lighter probing force should be used than when probing the gingival crevice in teeth. Similarly, the standardization of the examiners’ technique is critical [39]. A pressure-sensitive periodontal probe was used to record PPD and mBoP. We also used a 3-graded bleeding index to further distinguish between true disease and bleeding from the base of the pocket as the result of excessive pressure and rupture of the junctional epithelium. Scores of 1 and 0 and scores of 2 and 3 were pooled to create a more rigid, dichotomous score. This strengthens the positive results because significant differences were obtained when both the graded and dichotomous BoP scores were analysed. However, previous smokers were included but the outcomes in this patient group did not differ from finding in never smokers. Similarly, patients taking anticoagulants were excluded to avoid false-positive bleeding scores because of the increased bleeding response caused by the medication. Salvi and co-workers [40] studied the reduction in experimental peri-implant mucositis and revealed that 3 weeks of reinstituted plaque control did not yield pre-experimental levels of peri-implant health. While infection control was carefully installed prior to baseline, and the included implants were plaque free, we found significant reductions in the parameters of inflammation as early as 2 weeks after treatment with the chitosan brush. These results were stable up to 6 months after treatment, indicating a fast and stable response. More of the crown margins were exposed 2 weeks after debridement with the brush and the levels were thereafter stable. The more apical position the crown margins is most probably related to reduction in the soft tissue oedema from the inflammation.