3.1. Study selection
In the initial search, a total of 487 studies were identified; 171 PubMed, 166 Embase and 139 from the Cochrane Library database. Three papers were found through the Open Grey search and eight papers were selected following a manual search. After removal of duplicates (n=221), 266 papers were included in the selection phase of titles and abstracts. A total of 240 articles were excluded, and 27 papers were selected for full-text reading. In this phase, 10 studies were further excluded (Table S1) and 17 papers were finally included in the qualitative analysis2,10-12, 17-19, 26-35. The kappa value for inter-reviewer agreement was 0.92 indicating high degree of inter-rater reliability. Figure 1 shows the study identification flowchart based PRISMA19 with the reasons for exclusion.
3.2. Description of the included studies
Seventeen articles met the criteria and were included for qualitative analysis. Thirteen studies reported subgingival application of CHX gel adjunct to SRP at selected sites with a moderate to deep probing depth (at least 4mm in all studies) 8-10, 25-34. Nine studies were split-mouth RCTs10,12,27-29,32-35, four were parallel RCTs11,26,30,31 and three studies used a placebo in the control12,31,33. From the 13 papers, 10 showed the clinical outcomes of adjunctive subgingival delivered Xanthan-based CHX gel (XAN-CHX gel) in SRP and SRP alone26-35; CHX concentration in the XAN-CHX gel was 1.5% in nine studies and 2.5% in one study34, and another three studies reported the use of gels containing 0.5%, 1% and 2% CHX without Xanthan gum10-12. Patient samples ranged from five to 98. One included study compared the clinical outcomes between SRP plus XAN-CHX gel and SRP alone for patients with diabetes mellitus type 227. The timing and frequency of CHX Gel application varied between the trials. In all 13 studies but four10, 12, 27, 32, the CHX gel was applied once at baseline after SRP. In the other four studies, the application of CHX gel was described as three times at baseline, 10 day and 20 day follow-ups27, once at 1 month after treatment31 and three times within 10 min at baseline10,12. The follow-ups ranged from 1 month to 6 months after SRP.
An additional arm of the four studies evaluated the results between FMD and FMSRP2,17-19. All studies were RCTs, and one used a placebo gel and solution in the FMSRP group. The number of participants ranged from 18 to 38. Follow-ups varied from 1 month to 12 months. One study included patients with diabetes mellitus type 218. A 1% CHX gel was used in all of the trials. The timing and frequency variations for the CHX gel ranged from once at baseline17,19 and three times in 10 min at baseline18 to three times within 10 min at first session, second session of FMSRP and at 1 week of follow-up, respectively2. Table 1 shows the summary of the characteristics of the included studies.
3.3. Risk of bias assessment
All studies were RCTs. Seven studies did not report on their randomization and allocation methods in detail11,12,27,28,30,32,34,35, and from these, six studies also did not describe the blinding methods of participants and personnel as well as their assessment11,27,28,30,32,35, continued CHX rinsing stains the tooth and tongue surfaces, examiners could deduce which subjects were receiving CHX though these changes, and all the four studies in the analysis for comparison between FMD and FMSRP were considered at most to be single-blinded2,1719. Given examiner blinding was performed strictly in three studies, the detection bias for the three articles was qualified as ‘unclear’ 2,17,19. Overall, for all 17studies, six were assessed to have a low risk of bias10,17,18,26,29,31, two were judged as an unclear risk of bias12,33, nine were considered to have a high risk of bias2,11,19,27,28,30,32,34,35,and six were excluded from the quantitative analysis.11,27,28,30,32,35 . The summary of quality assessment is showed in Table 2.
3.4. Synthesis of results
All 17 studies reported on clinical outcomes with the use of adjunctive CHX Gel. The clinical results of these studies are summarized in supplemental Table 2. There was no consensus on the clinical efficacy of adjunctive CHX gel to SRP at selected sites. A significant improvement in PPD and/or CAL was reported in a number of studies using XAN-CHX gel27,28,30,32-35. Whereas, several studies showed no additional benefit in clinical outcomes with the adjunctive use of CHX gel10-12,26,29,31. In addition, all three studies using CHX gels that did not contain Xanthan gum reported no clinical benefits in the test group10-12. For comparing FMD and FMSRP, one study showed a significant improvement of PPD at 6 months17. In the other three studies, no sufficient evidence supported that FMD provided any significant improved clinical outcomes in terms of PPD and CAL2,18,19.
Quantitative analysis was performed when data on at least three studies at 3 and/or 6 months follow-up (± 2 months) was obtained. Six trials were excluded because of an unreached methodological quality for the requirement of this meta-analysis. Four trials were not included in the quantitative synthesis due to a lack of clinical outcomes in terms of PPD and CAL at follow-up2,12,33,34. Finally, four studies were included for the quantitative analysis of subgingival application of CHX gel at selected sites in terms of PPD reduction and CAL gain10,,26,29,31, three studies were included for analysis of full-mouth subgingival application of CHX gel in terms of the mean PPD and mean CAL at 3–4 and 6–8 months17-19. Four trials reported the adverse events after treatment18,19,29,32. Changes in PPD and CAL at selected sites 6 months after CHX gel administration and the mean bleeding of probing (BOP) value at follow-ups after treatment were not conducted due to a lack of data available in the meta-analysis.
3.5.Pooled outcomes
For the adjunctive application of CHX gel to SRP compared to SRP alone at selected sites, the meta-analysis showed a significant improvement in PPD reduction, with a mean MD of 0.15mm (MD: 0.15 [95% CI: 0.04–0.25]; p=0.005), no heterogeneity was observed among the studies (I2=0%) (Fig. 2a); No significant differences were found on the CAL gain between the groups (MD: 0.03 [95% CI: -0.09–0.15]; p=0.09) and moderate heterogeneity was indicated (I2=54%) (Fig. 2b). For subgroup analysis, adjunctive XAN-CHX gel provided a significant PPD reduction, with a MD of 0.15mm with no heterogeneity (MD: 0.15 [95% CI: 0.04–0.25]; p=0.005, I2=11%) (Fig. 3a). Whereas, no additional benefit for CAL gain was showed in the XAN-CHX group with a low heterogeneity among the studies (MD: 0.05 [95% CI: -0.05–0.15]; p=0.33, I2=50%) (Fig. 3b).
For full-mouth use of CHX gel, both the mean PPD and CAL showed no significant differences at 3-4 and 6-8 months. The overall effect size for PPD was -0.18mm at 3-4 months and -0.12mm at 6-8 months, and a high heterogeneity was observed among the studies [3-4 months (MD: -0.43 [95% CI: -0.63–0.27]; p=0.43, I2=76%) (Fig. 4a), 6-8 months (-0.12 [95% CI: -0.58–0.35]; p=0.62, I2=78%) (Fig. 4b)]. CAL was 0.09mm at 3-4 months and 0.05mm at 6-8 months with no heterogeneity [3-4 months (MD: 0.09 [95% CI: -0.27–0.46]; p=0.61, I2=0%) (Fig. 5a), 6-8 months (MD: 0.05 [95% CI: -0.29–0.39]; p=0.78, I2=0%) (Fig. 5b)].
3.6. Adverse events
Four studies reported adverse effects after treatment 17,18,28,31. Only one study comparing FMD and FMSRP reported that 17 subjects in the FMD and 12 in the FMSRP groups had one or two adverse events following mouth rinses, including changes in taste perception, dry mouth and staining17.