Comparison of hydroxyapatite and fluoride oral care gels for remineralization of initial caries: a pH-cycling study

Dental caries remains highly prevalent worldwide. While fluoride therapies are well established for caries prevention and remineralization, fluoride alone—especially in high-risk patients—may be insufficient and carries a fluorosis risk in young children if ingested. Biomimetic calcium phosphates, notably synthetic hydroxyapatite (HAP), resemble tooth mineral, are highly biocompatible, and have shown efficacy in remineralizing initial lesions and reducing bacterial adhesion without bactericidal effects. Given prior in situ and clinical evidence supporting HAP toothpastes, this study aimed to test whether daily application of a 15% HAP oral care gel is as effective as weekly application of a high-concentration fluoride gel (12,500 ppm F) in remineralizing early enamel caries lesions under a pH-cycling model, compared with artificial saliva.

Design: In vitro, randomized, three-arm study using a net-remineralization pH-cycling model over 28 days. Specimens: From 20 bovine teeth, three enamel blocks were prepared per tooth (total 60 blocks). Artificial caries-like subsurface lesions were created by 4 days of demineralization in acidified gel. Randomization: Blocks were assigned to three groups (n=20/group): A) Hydroxyapatite (HAP) gel (Karex gelée; 15% HAP; fluoride-free), B) Fluoride gel (Elmex gelée; 12,500 ppm F), C) Artificial saliva (AS) control. pH-cycling protocol: All groups underwent one daily 2 h demineralization challenge. Treatments were: - HAP gel: applied once daily for 3 min. - Fluoride gel: applied once weekly for 3 min. - AS control: remained in AS only. Between challenges and applications, specimens were stored in fresh AS; AS was magnetically stirred at 350 rpm; demineralization solution (DS) was static. Gel and DS phases were conducted at 37°C; other phases at room temperature. Fresh AS and DS were prepared daily (changed after acid challenge). Duration: 28 days. Outcome measures: Transverse microradiography (TMR) quantified mineral loss (vol%.µm) pre-test (baseline, Az1) and post-test (Az2). Tooth slices (~150 µm) were taken pre- and post-treatment; pre- and post-slices were microradiographed together for consistency. Percentage remineralization (% mineral gain) was calculated as (Az1 − Az2)/Az1 × 100. Microradiographs were examined qualitatively for remineralization patterns. Sample size and power: Based on prior studies (mean % remin ≈28.76, SD 7.2), n=20 blocks/group provided >0.85 power at one-sided α=0.05 to detect ≥10% mean remineralization. Statistics: Normality (Shapiro–Wilk) and variance homogeneity (Levene) were assessed; variances were heterogeneous. Intra-group changes were tested with paired t-tests. Between-group differences in % remineralization were assessed with Welch’s ANOVA and Games–Howell post hoc tests. Non-inferiority margin for HAP vs fluoride was set at ≤20% difference; superiority over AS was assessed using 95% and 80% confidence intervals (CIs). Analyses used STATA 10.0 with α=0.05.

- All groups showed significant remineralization from baseline to post-test (paired t-tests, p < 0.001). Pre-/post-/change (vol%.µm): HAP: 1713 ± 130.23 to 1016.50 ± 153.39; change 696.50 (95% CI 640.19–752.81), Cohen’s d 5.789. Fluoride: 1726 ± 104.95 to 1030 ± 232.18; change 696.00 (95% CI 622.19–769.65), d 4.423. Artificial saliva (AS): 1759 ± 130.80 to 1652.25 ± 122.89; change 106.25 (95% CI 83.56–128.94), d 2.192. - Percentage remineralization (mean ± SD): HAP 39.27% ± 6.98 (95% CI 36.00–42.53), Fluoride 40.76% ± 10.75 (95% CI 35.74–45.79), AS 6.01% ± 2.62 (95% CI 4.78–7.24). - Welch’s ANOVA: F ≈ 270.048, p < 0.001; Games–Howell: both gels > AS (p < 0.001). HAP vs Fluoride: no significant difference (mean difference ≈ −1.50%; p = 0.861). - Superiority over AS established: difference vs AS 95% CIs: HAP 29.10–37.42%; Fluoride 28.53–40.99%. - Non-inferiority: HAP was non-inferior to fluoride within a pre-specified ≤20% margin; statistical testing supported non-inferiority (p = 0.861). - Qualitative microradiography: Fluoride-induced remineralization localized to outer surface zone without lesion depth reduction; HAP induced more homogeneous remineralization throughout the lesion with reduced lesion depth. - Total experimental duration 28 days; daily 2 h acid challenge; HAP applied daily 3 min; fluoride weekly 3 min; storage in AS otherwise.

The study addressed whether daily use of a 15% hydroxyapatite gel can match the remineralization efficacy of a weekly 12,500 ppm fluoride gel on initial enamel caries lesions. Both gels significantly increased mineral content compared with baseline and were markedly superior to artificial saliva alone. The absence of a significant difference between HAP and fluoride, along with non-inferiority within a clinically pre-defined 20% margin, indicates comparable efficacy under pH-cycling conditions. Mechanistically, HAP likely supports remineralization by adsorbing to enamel, filling microporosities, acting as crystallization nuclei, and supplying/calcium-phosphate ions, leading to more homogeneous lesion repair. Fluoride’s effect concentrated in the surface zone, consistent with established lamination that may limit deeper remineralization. Clinically, daily HAP gel could be an alternative to prescription-strength high-fluoride gels—particularly advantageous for populations where fluoride ingestion risk is a concern—while maintaining efficacy. Artificial saliva elicited modest yet significant remineralization, reflecting saliva’s natural, limited capacity to promote repair, more pronounced at neutral pH.

A fluoride-free oral care gel containing 15% hydroxyapatite was as effective as a high-fluoride gel (12,500 ppm F) in remineralizing initial enamel caries lesions under a 28-day pH-cycling model. Both gels were significantly more effective than artificial saliva. The more homogeneous remineralization pattern with HAP suggests potential benefits for deeper lesion repair. Future work should include clinical trials to confirm effectiveness and preventive benefits of HAP gels in vivo across different risk groups and age populations.

- In vitro pH-cycling models, while mirroring alternating de-/remineralization, lack biological processes of the oral environment (e.g., dental plaque biofilm dynamics). - The demineralization challenges used may be more aggressive per exposure than typical intraoral acid attacks. These factors may affect generalizability to clinical conditions; thus, clinical trials are warranted.