The Ethics Committee at Pharos University approved this comparative in vitro study under the reference number (PUA-UREAC-27-10-2024). Using G*Power version 3.1.9.4, an ANOVA test was conducted with an effect size of 0.76 [17], an alpha level of 0.05, and a power of 0.95. The calculation determined that a minimum sample size of 30 specimens, divided equally into three groups (n = 10 per group), was necessary to detect a significant effect. The current study follows CRIS Guidelines (Checklist for Reporting In-vitro Studies) [18].

Human freshly extracted non-carious permanent premolars were obtained from dental clinics at Pharos University, as well as from public and private clinics. The teeth were extracted for orthodontic reasons. Patients were aged 18–25 years. Prior to extraction, written informed consent was obtained from each patient, allowing their teeth to be used for research purposes. The inclusion criteria required teeth to have intact enamel surfaces free of caries. Teeth with enamel defects, fractures, visible cracks, previous restorations, or signs of hypomineralization were excluded from the study.

The extracted teeth were thoroughly cleaned and disinfected through scaling and polishing to remove any remaining tissue and debris. They were then stored in a 0.9% sodium chloride (normal saline) solution at room temperature until further use [19, 20], which was changed daily to ensure a sterile environment and prevent bacterial growth until further use.

The teeth were sectioned longitudinally, with the roots removed, leaving specimens that included the buccal and lingual enamel surfaces. A 2 × 2 mm section of self-adherent tape was placed to be centered in the middle third of each specimen. The surfaces were then coated with acid-resistant nail polish (Maybelline [New York, NY, USA]). Two layers were applied using a micro-applicator brush to ensure complete sealing. After allowing the teeth to dry, the tapes were removed, uncovering a 2 × 2 mm area of each enamel specimen.

The specimens’ smooth surface was examined using DIAGNOdent. The laser probe moved in a pendulum movement without applying pressure, scanning the enamel. The maximum value was recorded (value from 0:12), which denotes a healthy tooth; a value greater than 12 indicates the presence of caries [21].

For 3 days, the specimens were submerged in a demineralizing solution that was pH 4.4 adjusted and contained 2.2 mM calcium chloride, 2.2 mM potassium dihydrogen phosphate, 0.05 M acetic acid, and 1 M potassium hydroxide (KOH). A pH meter was used to check the solution’s pH every day, and tiny amounts of NaOH were added to bring it down to 4.2. Refreshing amounts of 30% HCl solution were added as necessary to keep the pH between 4.2 and 4.25 [22]. Every day, the solution was updated.

The specimens were then reassessed with DIAGNOdent to verify the extent of demineralization. The readings were expected to range from 13 to 21, ensuring that all specimens experienced a consistent level of demineralization.

One gram of sodium carboxymethyl cellulose powder was weighed and mixed with 100 ml of deionized water to form a homogeneous solution [23]. Then, 100 ml of bioactive glass nanoparticle powder (BG 45S5- Nanotech Egypt) with an average size of 50 ± 20 nm was added to 2 ml of the sodium carboxymethyl cellulose and ethanol mixture, in a 1:1 ratio. The resulting mixture was sonicated for 1 h to ensure complete homogenization and then refrigerated at 5 °C in a dark container till use [24].

The specimens were labeled with codes ranging from 1 to 30 and a randomization list was generated to assign the specimens to one of three groups based on the treatment applied:

Group 1: Bioactive glass varnish (10 spacemen).

Group 2: 5% Na fluoride varnish (10 spacemen).

Group 3: Control group (10 spacemen).

The varnishes were applied to the enamel surfaces of the specimens in Groups 1 and 2 using a microbrush with a 2 mm head size, allowing for the even application of a thin layer of varnish on the enamel.

Immediately after treatment, the specimens were subjected to a pH cycling process for 4 weeks. This involved exposure to a demineralizing solution for 3 h, followed by a remineralizing solution for 21 h each day; the cycle was repeated daily throughout the 1 month duration of the study.

The remineralizing solution contained Na3PO4 (3.90 mM), NaCl (4.29 mM), KCl (17.98 mM), CaCl2 (1.10 mM), MgCl2 (0.08 mM), H2SO4 (0.50 mM), NaHCO3 (3.27 mM), and distilled water, with the pH adjusted to 7.2 [20].

The specimens were reassessed using DIAGNOdent after 1 month to determine the degree of remineralization.

Scanning electron microscopy with energy-dispersive X-ray spectroscopy (SEM-EDAX) (JEOL JSM-IT200 Scanning Electron Microscope, USA) was used to assess the calcium (Ca) and phosphate (P) mass percentages of enamel. The analysis was conducted at Alexandria University’s Faculty of Science, with the EDAX detector optimized at a working distance of 10 mm.

All DIAGNOdent and EDAX for calcium and phosphate measurements were performed at 3 points for each specimen. The mean of the 3 readings was calculated and used for all statistical analyses to minimize random error. After completion of EDAX analysis, the specimens were sputter-coated with a thin layer of gold solely to enhance image quality during scanning electron microscopy (SEM). SEM examination was then carried out at the Faculty of Science, Alexandria University, to evaluate enamel surface morphology at high magnification.

Data analysis was performed using SPSS software version 25.0 (IBM Corp., Armonk, New York, USA). The normality of the data distribution was assessed using the Shapiro–Wilk test. For DIAGNOdent readings, the analysis indicated a non-normal distribution in several groups (p < 0.05). Consequently, non-parametric tests were employed. Differences between the three groups were analyzed using the Kruskal–Wallis test, followed by Dunn’s post hoc test for pairwise comparisons. Variations across follow-up periods were assessed using the Friedman test. Data for DIAGNOdent are presented as medians and interquartile ranges (IQR). In contrast, EDAX measurements exhibited a normal distribution (p > 0.05). Therefore, parametric tests were applied. Differences between groups were compared using one-way ANOVA, followed by Tukey’s post hoc test. Variations across different follow-up periods were analyzed using repeated measures ANOVA. Data for EDAX are presented as mean and standard deviation (SD). A p-value of ≤ 0.05 was considered statistically significant for all analyses.