Control of Peri-Implant Mucous Inflammation by Using Chlorhexidine or Ultraviolet C Radiation for Cleaning Healing Abutments. Double-Blind Randomized Clinical Trial

Two-phase implants must be exposed to the external environment after the period of osteointegration has elapsed. For this purpose, a healing abutment is placed passing through the mucosa while forming the emergence profile. The continuous connection and disconnection can lead to an alteration in the tissue maturation, both because of the contact of bacterial plaque and because of the mechanical trauma that involves its manipulation, manifesting with different degrees of erythema or bleeding. To assess whether this epithelium disruption can be counteracted, a blinded study design was developed on 150 unitary implant patients divided into three groups (n = 50), applying chlorhexidine (group 1), ultraviolet C (UV-C) at a wavelength of 254 nm (group 2)and no treatment as a control group (group 3), during each of the disconnections and connections during the prosthodontic treatment (1 time per week for four weeks). All groups showed a better epithelium aspect at the end of the evaluation. Although there were no statistically significant differences in the degree of inflammation, the UV-C treated group had the lowest plaque accumulation, and the highest was for the chlorhexidine-treated group.


Introduction
Dental implants pass through the oral mucosa, establishing a connection between the inner and outer environments [1,2]. Most implants performed today belong to the "two-stage" type. This type of implant involves the appearance of an interface (between the implant and the abutment) called the GAP. When the GAP is positioned at or below the bone level, an increase in initial peri-implant bone loss occurs [3][4][5][6]. This is characteristic for external and internal butt-joint connections but not for Morse tapered (conical) connections, where implants can be placed subcrestally and have long-term crestal bone stability [7,8].
In butt-joint connections, histologically, the process usually manifests itself as an inflammatory state of the implant-abutment complex [9,10], and can trigger a possible loss of crestal bone [11]. One of the possible causes that favor the inflammatory state may be chemical or biological contamination [12,13]. The healing abutments were composed of titanium grade 5 (TiAl6V4). Once the abutments had been connected during the second surgery, they were left to heal for one week. After this week, the patients were randomly distributed into three groups: group 1 (G1), application of chlorhexidine gel; group 2 (G2), control using abutment as received from the manufacturer; group 3 (G3), exposure for 10 min in a UV-C sterilizer at a wavelength of 254 nm and 80 W.
After the first week of healing, the abutments were unscrewed following the sequence: take impressions after a week, testing of the metal structure of the prosthesis, testing of the ceramic prosthesis, and final delivery. The test groups (G1 and G3) received the assigned treatment upon each disconnection, while those of the control group (G2) were kept in a sterile container.
At all times, the data related to the degree of clinical inflammation were recorded. Plaque index was recorded at the end of the evaluation period. A total of four inflammation assessments were performed over four consecutive weeks and a single plaque evaluation at the end of the study period.  Random distribution was made before the assignment, following the Internet program https: //www.random.org The healing abutments were composed of titanium grade 5 (TiAl6V4). Once the abutments had been connected during the second surgery, they were left to heal for one week. After this week, the patients were randomly distributed into three groups: group 1 (G1), application of chlorhexidine gel; group 2 (G2), control using abutment as received from the manufacturer; group 3 (G3), exposure for 10 min in a UV-C sterilizer at a wavelength of 254 nm and 80 W.
After the first week of healing, the abutments were unscrewed following the sequence: take impressions after a week, testing of the metal structure of the prosthesis, testing of the ceramic prosthesis, and final delivery. The test groups (G1 and G3) received the assigned treatment upon each disconnection, while those of the control group (G2) were kept in a sterile container.
At all times, the data related to the degree of clinical inflammation were recorded. Plaque index was recorded at the end of the evaluation period. A total of four inflammation assessments were performed over four consecutive weeks and a single plaque evaluation at the end of the study period.
Each of the evaluations was interpreted by a blinded evaluator, according to the following evaluation protocol ( Figure 2 The degree of plaque accumulation on the abutment was recorded following the modified Mombelli plaque index [35].
All patients followed CONSORT guidelines to ensure the quality of randomized clinical trials [36].

Randomization
The distribution of patients was carried out using the Internet App (https://www.random.org/), and the result was kept in a closed envelope with the corresponding number from 1 to 150. As patients attended the appointment for their second surgery, the envelope corresponding to their numerical sequence was opened, and the assigned treatment was applied to the abutment.

Blinding
Patients were informed of the purpose of the study but were unaware of the type of treatment they would receive, and the abutment was stored in a different room when removed for processing, so the group was blinded for the patient. The samples were also blinded to the statistician.

Statistical analysis
The SPSS version 23.0 statistical package (SPSS Inc., Chicago, IL, USA) was used for statistical comparisons, and an independent statistician reviewed the results. Statistical significance was considered at p < 0.05.
Data were presented as descriptive statistics (mean and I.C.) using Tukey's exploratory analysis. Adjustment to normality was determined using the Kolmogorov-Smirnov test, and between-group comparisons were performed using the Mann-Whitney U test and ANOVA test. The results were reviewed by an independent statistician (http://estadisticamurcia.com/web/#2). The degree of plaque accumulation on the abutment was recorded following the modified Mombelli plaque index [35].
All patients followed CONSORT guidelines to ensure the quality of randomized clinical trials [36].

Randomization
The distribution of patients was carried out using the Internet App (https://www.random.org/), and the result was kept in a closed envelope with the corresponding number from 1 to 150. As patients attended the appointment for their second surgery, the envelope corresponding to their numerical sequence was opened, and the assigned treatment was applied to the abutment.

Blinding
Patients were informed of the purpose of the study but were unaware of the type of treatment they would receive, and the abutment was stored in a different room when removed for processing, so the group was blinded for the patient. The samples were also blinded to the statistician.

Statistical Analysis
The SPSS version 23.0 statistical package (SPSS Inc., Chicago, IL, USA) was used for statistical comparisons, and an independent statistician reviewed the results. Statistical significance was considered at p < 0.05.
Data were presented as descriptive statistics (mean and I.C.) using Tukey's exploratory analysis. Adjustment to normality was determined using the Kolmogorov-Smirnov test, and between-group comparisons were performed using the Mann-Whitney U test and ANOVA test. The results were reviewed by an independent statistician (http://estadisticamurcia.com/web/#2).

Results
The results are presented as mean and confidence interval (C.I.) for each variable. A total of 150 patients were included in the study: 62 men and 88 women with an average age of 60.7 years (C.I. 58.9-62.4). The female average age was 60.4 years (C.I. 58.3-62.6), while the male average age was 61.0 years (C.I. 58.0 / 64.0). There were no statistically significant differences between male and female age (t-test).
Twelve implants were placed in anterior position (incisors or canines), 58 in premolars, and 80 in molars.
Seventeen patients were excluded from the study, with four implant failures, five dropouts, and eight"non-compliance" with appointments.
Inflammation of the overall sample was 1. Results for each group were as follows: No statistically significant differences in the degree of inflammation among the three groups (ANOVA) were found.
The plaque level was statistically significant in favor of group 3 treated with UV-C, with no statistical difference between groups 1 and 2. Clinically, group 1 (chlorhexidine) had the greatest accumulation of plaque ( Figure 3). Materials 2020, 13, x FOR PEER REVIEW 6 of 11

Discussion
When connecting an abutment with the implant, there is very often an inflammatory phase, but this is usually limited in severity. Some authors claim that the source of this inflammation is the presence of bacteria inside of the GAP [14,[37][38][39]. Such GAP colonization appears to come from the patient's own oral cavity [14,[39][40][41][42], and structured flora has been found in the peri-implant furrow two weeks after connection [43,44]. To reduce contamination of the healing abutment, different compositions have been investigated, such as the use of anatase crystallographic form, with favorable results [45]. In our study, all pillars were grade V (Ti-6Al-4V). So we could not observe this difference. On the other hand, new antiseptics, such as Bactercline® (NMTECH ITALIA SRL), have been developed for decreasing the bacterial charge on the surface of these screws, with hopeful results [46]. In our study, we only use chlorhexidine with no appreciable effect.
In the case of bacterial colonization of the GAP, migration to the inside of the implant seems to depend on the time elapsed since the connection of the abutment. Early initial contamination has been reported after 5 h [47], and 100% of the inside of the implants seems to be contaminated after five years of functioning [13]. This bacterial reservoir can cause bone loss and depends on the socalled "extended arm" [48], "effective radius of action" [49] or "inflammatory front" [50].
Other authors have mentioned that trauma from the continuous connection and disconnection of the abutment could also develop this inflammatory state [51][52][53][54][55]. Finally, for other authors, the mobility of the abutment is a major factor, producing a micro-pumping effect [6,56].
Whatever the case, this inflammatory state can cause bone reabsorption, which will depend on the position of the GAP in the bone [5,57,58].To prevent this adverse effect, the continuity solution between the internal and external environment must be sealed by epithelial and connective adhesion [59,60]. To improve this adhesion, various modifications have been made to the materials of the healing abutments [61,62] and the surface's topography [63]. However, other authors find no clear improvement from taking such actions [64,65]. To avoid these possible confounding variables, all implants in this study had an internal conical connection, due to their increased sealing characteristics [66]. Likewise, all the healing abutments were used as received from the manufacturer, with a length of 2 mm, and screwed at a standardized torque of 20 N/cm. The only modification they received was that associated with their group treatment. The abutments were screwed in and unscrewed on five time intervals (first connection, impression appointment, metal try-in, ceramic and prosthesis delivery), although they were only evaluated on four occasions to determine inflammatory changes (impression taking, metal test, ceramic test, and delivery).
This implies that the continuous connection and disconnection of the abutments would involve the re-establishment of a new biological width to a more apical level of the implant [51]. Both causes

Discussion
When connecting an abutment with the implant, there is very often an inflammatory phase, but this is usually limited in severity. Some authors claim that the source of this inflammation is the presence of bacteria inside of the GAP [14,[37][38][39]. Such GAP colonization appears to come from the patient's own oral cavity [14,[39][40][41][42], and structured flora has been found in the peri-implant furrow two weeks after connection [43,44]. To reduce contamination of the healing abutment, different compositions have been investigated, such as the use of anatase crystallographic form, with favorable results [45]. In our study, all pillars were grade V (Ti-6Al-4V). So we could not observe this difference. On the other hand, new antiseptics, such as Bactercline®(NMTECH ITALIA SRL), have been developed for decreasing the bacterial charge on the surface of these screws, with hopeful results [46]. In our study, we only use chlorhexidine with no appreciable effect.
In the case of bacterial colonization of the GAP, migration to the inside of the implant seems to depend on the time elapsed since the connection of the abutment. Early initial contamination has been reported after 5 h [47], and 100% of the inside of the implants seems to be contaminated after five years of functioning [13]. This bacterial reservoir can cause bone loss and depends on the so-called "extended arm" [48], "effective radius of action" [49] or "inflammatory front" [50].
Other authors have mentioned that trauma from the continuous connection and disconnection of the abutment could also develop this inflammatory state [51][52][53][54][55]. Finally, for other authors, the mobility of the abutment is a major factor, producing a micro-pumping effect [6,56].
Whatever the case, this inflammatory state can cause bone reabsorption, which will depend on the position of the GAP in the bone [5,57,58].To prevent this adverse effect, the continuity solution between the internal and external environment must be sealed by epithelial and connective adhesion [59,60]. To improve this adhesion, various modifications have been made to the materials of the healing abutments [61,62] and the surface's topography [63]. However, other authors find no clear improvement from taking such actions [64,65]. To avoid these possible confounding variables, all implants in this study had an internal conical connection, due to their increased sealing characteristics [66]. Likewise, all the healing abutments were used as received from the manufacturer, with a length of 2 mm, and screwed at a standardized torque of 20 N/cm. The only modification they received was that associated with their group treatment. The abutments were screwed in and unscrewed on five time intervals (first connection, impression appointment, metal try-in, ceramic and prosthesis delivery), although they were only evaluated on four occasions to determine inflammatory changes (impression taking, metal test, ceramic test, and delivery).
This implies that the continuous connection and disconnection of the abutments would involve the re-establishment of a new biological width to a more apical level of the implant [51]. Both causes (the connection of the abutment and its colonization) favor the presence of a reservoir between the implant and the abutment [9,62,63,67,68].
To the best of our knowledge, there have been no studies comparing the use of chlorhexidine and UV-C in comparison to untreated controls. In this study, we wanted to determine whether the use of chlorhexidine digluconate gel, the use of UV-C at 254 nm for 15 min or an untreated abutment (as received) lead to differences in the degree of gingival inflammation and plaque accumulation.
As for the degree of clinical inflammation, in our study, there were no significant differences in any of the three groups studied. This induces us to think that the most important factor during this period of what might be termed "prosthetic preparation" is the mechanical trauma involved in the repeated connection and disconnection of the abutments.
During the four weeks of follow-up, all the groups showed gradual improvement due to the maturation of the tissues and reconditioning of the supracrestal insertion in a natural healing process.
These results are consistent with those in the literature that point to the abutment trauma and its continuous changes as a fundamental cause of tissue inflammation [51,69].
As for abutments' plaque accumulation, many clinicians use chlorhexidine to disinfect prosthetic components and prevent bacterial colonization. However, this procedure remains controversial, and our study indicates clinical differences between the three established groups. The chlorhexidine group showed the worst results, although no statistically significant differences were found with the control group. However, this finding is consistent with the adverse effects described for chlorhexidine, which depend on its substantivity. Indeed, both staining and calcium phosphate crystal deposits increase with the use of chlorhexidine [70]. By contrast, some authors found an improvement in bacterial load with the use of chlorhexidine [71,72]. Our results showed that this decrease in the bacterial load did not result in any clinical improvement during the four weeks of the study. These results agree with those obtained by other authors, who also observed no improvement with the use of chlorhexidine [73,74]. In addition, chlorhexidine may alter the surface properties of titanium, the clinical impact of which has yet to be determined [75].
In the present study, the best result was obtained by the UV-C treated group, which showed statistically significant differences from the other groups regarding plaque accumulation.
The effect of UV-C is complex and is due, on the one hand, to the rejuvenating effect on titanium, eliminating the carbonic compounds deposited on it [76] and, on the other hand, to the alteration caused on the surface load of titanium and its ability to become moist [77]. Finally, UV-C radiation itself is capable of producing ozone, which could also be involved in the disinfection of the abutments [78].
From the clinical standpoint, the stabilized blood clot at the second stage surgery seems to be a physiologic barrier contributing as a fundamental biologic seal to control bacterial penetration to the peri-implant tissues, The clinical relevance of our study is that for the time taken for the tissue to mature, as mentioned above, all groups showed a clear improvement in the degree of inflammation over time, an improvement that was clearly evident two weeks after connection. Therefore, we recommend that, when taking the final impressions, a minimum period of two weeks between the healing abutment connection and the final impression. This aspect is of great significance when the prostheses are located in an aesthetic zone.

Conclusions
Within the limitations of this study, we can conclude that: • No treatment showed a statistically significant decrease in the degree of clinically recorded inflammation.

•
The worst result in terms of plaque accumulation was observed in the chlorhexidine treated group.

•
The best result regarding plaque accumulation was presented by the abutments that were UV-C treated at 254 nm for 15 min, with statistically significant differences. • Peri-implant tissue maturation improves with time.
• Since the most important changes take place during the first two weeks, we recommend taking impressions at this time, although for the best aesthetic results, we recommend waiting four weeks for soft tissue maturation. Funding: This study was funded by a contract between the Office of Transfer of Research Results and the company Mozo Grau.