A prospective study was carried out in order to evaluate the response to treatment of 100 implants (from the brand Biomet 3i) with peri-implantitis, placed in the Prosthodontics and Occlusion Teaching Unit of the University of Valencia. The implants included in the study were selected during the control visits carried out in the year 2021. It was imperative that they had been loaded with their definitive prosthesis for at least one year, presented clinical signs of peri-implantitis following the criteria indicated in the 2017 Workshop  and had an adequate general periodontal health status.
Initial records related to the patient were collected (age, sex, tobacco consumption habits and history of periodontitis), having placed all implants in patients already treated and periodontally stabilized ; as well as other features related to the implants: Millimeters of keratinized mucosa measured in the mid zone of the buccal site in implants using a periodontal probe CP 15 by Hu-friedy® (Chicago,IL, USA); Recession: millimeters of gingival margin retraction with respect to the prosthetic crown (where the gingival margin was located at the time of prosthesis placement) measured in the mid-buccal site; Silness and Löe plaque index; Bleeding and suppuration: its presence or absence was measured. Probing depth using the CP15 periodontal probe (at six points: mesiobuccal, buccal, distobuccal, mesiopalatal, palatal and distopalatal); Gingival inflammation (Löe and Silness Gingival Index); and percentage of bone loss: measured using Rhinoceros software (Robert Mcneel & Associates. Seattle, USA) on radiographs taken with Rinn system positioners (Denstsply, Illinois, USA). The years since prosthesis placement were also recorded. All data were collected by a single operator.
Once the initial records were obtained, a non-surgical decontamination treatment was performed under local anesthesia (articaine 4% and adrenaline 1:100,000) which consisted in the removal of plaque and calculus from the implant surface using ultrasound (SP Newtron, Satelec Acteon, Olliergues, France) and H3 tip (Satelec Acteon, Olliergues, France), implant-specific titanium curettes by Hu-friedy® (Chicago,IL, USA), curettage of the internal epithelium of the pocket with Hu-friedy® 4R/4L curette (Chicago, IL, USA), subgingival irrigation with an antiseptic (chlorhexidine 0.2%), administration of antibiotic (metronidazole 500 mg every 8 hours/7 days) and antiseptic (0.2% chlorhexidine gel 3 times a day for 15 days). Supragingival scaling treatment was also performed in the rest of the oral cavity and oral hygiene instructions were provided. There were no patients allergic to the drugs administered, therefore, all of them received the same treatment.
After completing the treatment, a re-evaluation was carried out at 1 month, 3 months and 6 months to assess the outcome of the treatment, in which the same data as at baseline were collected to analyze both the initial results and the medium-term evolution (Table 1). In the evaluation at one month it was also determined whether the patient should go on to maintenance or whether a previous surgical phase was required.
A statistical analysis was performed with the data obtained.
Pre-study (T0) Day 0
1 month (T1)
3 months (T2)
6 months (T3)
Baseline records and treatment
Ethics and consent to participate
The study was approved by the Human Research Ethics Committee of the University of Valencia with procedure number 183748 in accordance with the Declaration of Helsinki. All participants signed a written informed consent.
Simple binary logistic regression models were estimated using generalized estimating equations (GEE) to explain the probability of implant recession, bleeding and suppuration on probing or inflammation of the implant depending on the presence or absence of KM over time.
Likewise, for clinical variables of ordinal or continuous nature (probing depth, bone loss, plaque) generalized linear models were also applied under the GEE approach. This type of approach is justified by the multiplicity of implants per patient, thus controlling the dependence of the observations.
The models were replicated adjusting the time since implant placement and the the amount of KM (no presence, <=2, >2mm). In order to study the association between the presence of KM and the pre-treatment clinical situation, the above methodology was also employed.
The significance level used in the analyses was 5% (α = 0.05). Due to the multilevel design of the data (several implants per patient), assuming a high intra-subject correlation (ρ = 0.75), a power of 89% was obtained.