Results

Bacterial recovery and descriptive findings

All 48 implants were included in the analysis, with 12 implants per group. The untreated control group exhibited the highest bacterial load, confirming successful biofilm formation and recovery. Mean log10(CFU/implant) values decreased progressively from control (5.09 +/- 0.33) to CHX (4.03 +/- 0.12), GalvoSurge (3.77 +/- 0.25), and laser treatment (2.52 +/- 0.39), indicating a clear reduction gradient across the tested protocols. Descriptive statistics and corresponding log₁₀ reductions relative to the control group are presented in Table 2.

Group | n | Mean log10(CFU) | SD | Approx. mean CFU/implant | Approx. log10 reduction vs control
Control | 12 | 5.09 | 0.33 | 155.333 | Reference
CHX | 12 | 4.03 | 0.12 | 11.125 | 1.06
GalvoSurge | 12 | 3.77 | 0.25 | 6.883 | 1.32
Laser | 12 | 2.52 | 0.39 | 515 | 2.57

Among the tested protocols, laser treatment resulted in the lowest residual bacterial burden, followed by GalvoSurge and CHX. To further assess the difference between treatment outcomes, inferential analysis focused on the comparison between CHX and laser groups, representing the least and most effective active protocols. Detailed descriptive statistics for this comparison are shown in Table 3.

Compared group | n | Mean | SD | SE | 95% CI lower | 95% CI upper | Minimum | Maximum
CHX | 12 | 4.0297 | 0.1220 | 0.0352 | 3.9522 | 4.1072 | 3.8633 | 4.2967
Laser | 12 | 2.5175 | 0.3874 | 0.1118 | 2.2714 | 2.7637 | 2.0000 | 3.4150
Total | 24 | 3.2736 | 0.8218 | 0.1678 | 2.9266 | 3.6206 | 2.0000 | 4.2967

The laser group demonstrated markedly lower mean log₁₀(CFU) values compared with CHX. The assumption of homogeneity of variance was evaluated using Levene’s test. Results indicated a violation of this assumption (p < 0.05). One-way ANOVA results for the CHX versus laser comparison showed a highly significant difference between groups (p < 0.001). The magnitude of the treatment effect was very large (η² = 0.883), indicating a substantial difference between groups.

Overall, statistical testing showed a highly significant difference in log10(CFU/implant) values (p < 0.001), with a very large effect size (eta squared = 0.883). Although variance homogeneity was not fully satisfied, the conclusion was supported by Welch robust testing, strengthening confidence in the observed treatment-related difference.