Introduction

One of the most common chronic illnesses in children remains dental decay, with early childhood caries (ECC) posing a significant public health challenge. The prevalence of ECC is alarmingly high, affecting over 50% of children under the age of five in many regions, particularly in underserved populations [1]. Unlike permanent enamel, deciduous enamel is thinner, more porous, and contains a higher proportion of organic material, making it more susceptible to acid dissolution and rapid demineralization [2]. Due to these structural differences, primary teeth experience faster caries progression, often leading to early tooth loss and long-term oral health complications, including mal­occlusion, speech difficulties, and increased caries risk in permanent dentition [3].

The mineralized structure of enamel plays a crucial role in resisting acid dissolution; however, exposure to organic acids produced by cariogenic bacteria initia­tes demineralization. During this process, phosphate (PO₄^3–) and carbonate (CO₃²^–) ions are lost, weakening the enamel’s crystalline structure [4]. The loss of phosphate contributes to mineral degradation. Carbonate substitution within the hydroxyapatite lattice creates crystal defects and increases solubility, therefore, its pre­sence renders enamel more susceptible to acid dissolution and structural porosity [5].

Conversely, remineralization is a natural repair process, where calcium, phosphate, and fluoride ions are redeposited into enamel, forming a more acid-resistant mineral phase. Fluoride is particularly effective in promoting remineralization by forming fluorapatite, which has lower solubility than hydroxyapatite [6]. However, remineralized enamel may not fully restore the original structure, leading to variations in mineral density and composition at different depths [7].

Despite advances in pediatric dentistry, early caries detection in young children remains challenging. Traditional diagnostic methods, such as visual inspection and bitewing radiographs, have limitations in identifying subsurface demineralization before cavitation occurs [8]. Alternative detection tools, such as fluorescence-based caries detection and transillumination techniques, have improved sensitivity but may still lack depth penetration and specificity, leading to false positives from surface stains or enamel hypoplasia [9].

Given the limitations, there is a growing need for noninvasive, high-resolution diagnostic techniques that can accurately quantify early enamel mineral loss and monitor remineralization treatments in primary teeth. Raman spectroscopy has emerged as a promising tool for noninvasive enamel analysis [10]. This optical technique utilizes laser light scattering to detect molecular vibrations, providing a spectral fingerprint of enamel composition. The most relevant Raman spectral bands for enamel analysis include phosphate and carbonate, which serve as key indicators of mine­ral content [11].

Two advanced Raman-based methods, i.e., spontaneous Raman spectroscopy (SpRS) and stimulated Raman scattering spectroscopy (SRS), offer significant advantages for caries detection. Notably, SRS enhances Raman signals by several orders of magnitude, enabling rapid imaging of enamel mineralization changes without background interference [12]. Moreover, SRS enables depth-resolved analysis, extending from 0 to 100 µm, through its intrinsic optical sectioning capability, a key feature of its nonlinear optical process. This allows for the acquisition of high-resolution, three-dimensional (3D) images by precisely controlling the laser focus, without the need for sample’s physical slicing. This is a significant advantage over traditional SpRS, which is limited to surface-level measurements (0 µm depth) [13]. According to Ando et al. [14], SRS might be able to determine the depth of lesions from the enamel surface without sectioning the tooth/specimen that would not be possible under clinical conditions.

A potential method for detecting dental caries at an early stage is Raman spectroscopy, which can identify changes in the quantity of minerals in enamel brought on by initial caries [15]. Moreover, Raman spectrometers are small enough to fit in a pocket and can be controlled remotely via optical fibers. Therefore, the Raman method is a quantitative instrument for dental diagnostics, meeting the preventive management of oral care necessary in modern dentistry due to the availability of strong correlations with recognized spectroscopic techniques [16].

Phosphate/carbonate ratio (P/C ratio) is a critical metric for evaluating enamel integrity. It has been shown that a higher P/C ratio corresponds to sound enamel and a lower P/C ratio indicates mineral loss due to demineralization, while an increase in P/C ratio suggests partial recovery of mineral content during reminerali­zation [14]. This change occurs because acidic demine­ralization causes a change in enamel structure due to the loss of phosphate and carbonate ions. The loss is not always proportional, with phosphate removed at 18.5% and carbonate at 1.0%. Despite rapid decreases in carbonate, nearly 99% can remain in the lesion body, which leads to a decrease in phosphate relative to carbonate, causing a decrease in P/C ratio. Conversely, remine­ralization results in partial mineral content restoration, increasing P/C ratio [14]. By quantifying the P/C ratio at different depths, researchers can better understand the dynamics of enamel demineralization and remine­ralization, paving the way for improved preventive and restorative strategies [17]. Understanding mineral changes during both demineralization and reminerali­zation is critical for developing effective caries prevention and treatment strategies [18].

Objectives

The aim of this study was to measure the P/C ratio in sound, demineralized, and remineralized deciduous enamel samples at the enamel surface using SpRS, and at various depths using stimulated SRS. The null hypo­thesis was that there are no significant differences in the P/C ratio across sound, demineralized, and reminera­lized deciduous enamel.

Material and methods

Ethical considerations

All procedures involving human deciduous enamel specimens were conducted in accordance with ethical guidelines and regulations. The Research Ethics Committee of the Faculty of Dental Medicine, Al-Azhar University, Cairo, Egypt, approved the study procedure
(authorization code: 1213/1178).

Specimen preparation

Thirty deciduous enamel specimens (5 × 5 mm) were prepared from 30 freshly extracted, non-carious primary incisors (single sample from every incisor), which were stored in 0.1% thymol solution at 4°C before use to prevent microbial contamination. Specimens were attached to polishing discs, and the dentin portion was ground with 500-grit silicon carbide sandpaper (Struers Inc., Cleveland, PA, USA) to a consistent thickness of 2.4 mm. Enamel surfaces were sequentially polished using 1200-, 2400-, and 4000-grit silicon carbide sandpapers, followed by 1-µm diamond polishing suspension to ensure a smooth and uniform surface [19]. This polishing was performed to standardize the enamel surface, minimizing topographical variations and signal scattering artifacts. While this polishing could alter the very outermost natural enamel layer, it provided a consistent starting point (0 µm depth) for comparative analy­sis across all samples.

Each specimen was ultrasonically cleaned in deionized water for 3 minutes to eliminate any debris or contamination. Only specimens with enamel thickness of minimum 0.5 mm and no visible cracks were chosen; they were stored in a humid environment at 4°C to maintain hydration until further use [20].

Demineralization protocol

To create artificial enamel lesions, three distinct experimental windows were created on the surface of each specimen (n = 30): (1) sound enamel (protected, no treatment); (2) 24-hour demineralization (24 h-demin); (3) 48-hour demineralization (48 h-demin). Deminera­lization was performed using a Carbopol-lactic acid solution designed to mimic early enamel caries lesions. In the solution, pH 5.0 was achieved by adjusting 0.1 mol/l of lactic acid and 0.2% Carbopol 907TM (B.F. Goodrich Co., Chemical Group, Ohio, USA) to a saturation state by hydroxyapatite [21].

Acid-resistant tape was applied to selectively expose the windows. First, the 48 h-demin window was exposed, and the specimens were submerged in demi­neralization solution at 37°C. After 24 hours, the samples were removed, and the tape protecting the 24 h-demin window was removed for exposure, while the sound window remained protected. The specimens were then resubmerged for an additional 24 hours, ensuring that the 48 h-demin window received a total of 48 hours of acid exposure (the 24 h-demin window received 24 hours). After each demineralization period, the samples were rinsed thoroughly with deionized water and stored in a humidified environment before further processing.

Group allocation

Following the demineralization process, the 30 specimens were randomly divided into two independent groups: group 1 (demineralized, n = 15): specimens were rinsed and stored in a humidified environment for immediate Raman analysis, and group 2 (remineralized, n = 15): samples were subjected to remineralization protocol.

Remineralization protocol

Specimens in group 2 undergone a remineralization regimen using a 15-day pH cycling model to simulate natural remineralization process. This model incorporated daily fluoride exposure, saliva immersion, and periodic acid challenges to mimic oral conditions. Each daily cycle consisted of the following phases [22]:

• fluoride treatment: specimens were exposed to a 1100 ppm fluoride dentifrice slurry (Crest Cavity Protection, USA) for 1 minute, followed by a deioniz­ed water rinse;
• saliva immersion: samples were then immersed in a freshly prepared synthetic saliva for 60 minutes to promote mineral uptake; synthetic saliva had the following composition: 1.5 mM/l calcium chloride (CaCl₂), 50 mM/l potassium chloride (KCl), 0.9 mM/l potassium phosphate monobasic (KH₂PO₄), and adjusted to pH = 7.4 using potassium hydroxide (KOH);
• acid challenge: specimens were subjected to a 4-hour acid challenge (pH 5.0 demineralization solution) to simulate cariogenic conditions;
• final remineralization phase: after acid challenge, samples were immersed in synthetic saliva overnight to simulate natural oral conditions.

The pH cycling process was carried out for 15 consecutive days. Over the weekends, samples were stored in a humidified chamber to prevent dehydration.

Raman spectroscopy analysis

Spontaneous Raman spectroscopy (SpRS)

SpRS was conducted using an inVia Renishaw Raman microscope equipped with a 785 nm laser. SpRS measurements were exclusively performed at the enamel surface (0 µm depth), and spectra were collected at these locations per specimen:

• demineralized specimen: sound (sound-demin), 24 h-demin, and 48 h-demin;
• remineralized specimen: sound (sound-remin), 24 h-demin, 15-day remineralization (24 h d-remin), 48 h-demin, and 15-day remineralization (48 h d-remin).

For each specimen, five spectra were obtained within an 1 × 1 mm² acquisition area to ensure representativeness, and these were averaged before analysis. For the P/C ratio analysis presented in this study, the 785 nm laser was employed because its near-infrared wavelength produces significantly lower fluorescence interference than shorter-wavelength lasers. This reduced background noise allows for clearer detection and quantification of spectral bands corresponding to phosphate and carbonate, which are key components of dental enamel [23]. Each spectrum was recorded in a 1 × 1 mm² acquisition area, with an integration time of 10 seconds per scan. Spectral analysis focused on phosphate (959 cm^–1) and carbonate (1,070 cm^–1) peaks, quantified with WiRE 3.4 software (Renishaw, UK) [24].

Stimulated Raman scattering spectroscopy (SRS)

SRS imaging was performed using a hyperspectral SRS system based on a dual-output ultrafast laser (InSight DeepSee, Spectra-Physics). The pump beam (942 nm, 120 fs pulse duration) and Stokes beam (1040 nm, 220 fs pulse duration) were modulated at 2.3 MHz. A 50× water immersion objective (NA = 1.0, Olympus) was employed to focus the laser. At the sample, the laser power for the pump beam was set at 200 mW and for the Stokes beam at 150 mW. Image acquisition speed was 0.5 frames per second for a 256 × 256 pixel image. For each specimen, SRS images were acquired from multiple regions of interest within each treatment area. From these images, spectra for the P/C ratio calculation were extracted at 10-µm intervals from 0 to 100 µm depth (from surface to dentin-enamel junction). For each depth point, data from a 10 × 10 µm region of interest (comprising 100 pixels) within a SRS image were averaged to generate a representative spectrum, from which the P/C ratio was calculated.

P/C ratio calculation

The P/C ratio was similarly determined for both SpRS and SRS by calculating the ratio of the integrated peak intensities of the phosphate band (959 cm^–1) to the carbonate band (1,070 cm^–1) at each depth [14]. Prior to calculating integrated peak intensities, a polynomial baseline correction was applied to all spectra using WiRE 3.4 software. For a 1000 to 1100 cm^–1 region, where phosphate and carbonate signals can overlap, a curve-fitting procedure utilizing a mixed Gaussian-Lorentzian function was performed to deconvolve the composite band. This deconvolution ensured accurate determination of individually integrated intensities for both the phosphate and carbonate peaks.

Statistical analysis

For both SRS and SpRS at the surface (0 μm), the P/C ratio was calculated. For SRS, this ratio was similarly determined at 10-μm intervals up to 100 μm from the surface, reaching dentin-enamel junction. Due to nonnormal data distribution, nonparametric tests were employed, and descriptive statistics were presented as median and interquartile range (IQR: Q1-Q3). Wilcoxon signed-rank tests were used to compare different experimental windows (sound vs. 24 h-demin vs. 48 h-demin) within the same specimens (paired data). Mann-Whitney U tests were employed to compare the P/C ratio between independent demineralized (group 1) and remineralized (group 2) specimens. A significant level of 0.05 was used for all statistical tests.

Results

Spontaneous Raman spectroscopy (SpRS) at the enamel surface (0 µm)

The P/C ratio at the enamel surface was significantly affected by the demineralization and remineralization processes. The median P/C ratio for sound enamel was 13.00 (12.50-14.00). Following demineralization, this value decreased to 12.30 (11.80-13.00) for the 24 h-demin group and 11.90 (11.70-12.80) for the 48 h-demin group. After the remineralization period, the P/C ratio recovered to values nearly similar to sound enamel, with 13.60 (13.30-14.60) in the 24 h d-remin group and 13.40 (13.10-14.20) in the 48 h d-remin group. The complete descriptive data for the SpRS analysis are presented in Table 1.

Statistical analysis confirmed that decreases observed after both 24 h and 48 h demineralization were statistically significant (p < 0.05) compared to sound enamel. Furthermore, the recovery following reminera­lization was significant for both demineralized groups (p < 0.001), with the final remineralized values showing no significant difference from sound enamel (p > 0.05). The summary of these statistical comparisons is provided in Table 2.

Stimulated Raman scattering (SRS) depth-resolved analysis (0-100 µm)

The depth-resolved analysis using SRS provided detailed insights into mineral changes occurring beneath the enamel surface. The median P/C ratios for all treatment groups at depths from 0 to 100 µm are shown in Table 3. Overall, demineralization caused a reduction in the P/C ratio at all depths compared to sound enamel. This effect was most pronounced in superficial depths (0-20 µm); for instance, at a depth of 20 µm, the P/C ratio of sound enamel of 1.59 (1.32-1.70) decreased to 1.32 (0.98-1.56) in the 48 h-demin group.

Following the remineralization protocol, a partial recovery of the P/C ratio was observed, which was again most evident in superficial depths (0-20 µm). At deeper depths (30-100 µm), where no significant initial mineral loss was detected, the P/C ratio values remained stable and comparable to baseline sound enamel.

The statistical significance of these changes is presented in Table 4. When comparing sound versus demineralized enamel, a significant reduction in the P/C ratio was found at superficial depths (0-20 µm, p < 0.05); however, at deeper depths (30-100 µm), no significant differences were observed (p > 0.05). When comparing demineralized versus remineralized enamel using
the Mann-Whitney U test, the 24 h-demin group showed a statistically significant partial recovery of the P/C ratio in superficial layers (0-20 µm, p < 0.05). For the 48 h-demin group, remineralization did not result in significant increase in the P/C ratio in superficial layers (p > 0.05) and as expected, no significant changes were observed in deeper, unaffected layers.

Discussion

Primary and permanent teeth have different enamel surface morphologies. Also, primary teeth’s enamel is more prone to caries than permanent teeth’s due to primary teeth having a lower mineral concentration and a higher proportion of organic content than permanent teeth [25]. The P/C ratio is a critical indicator of enamel mineralization and caries progression [4].

In this study, Raman spectroscopy (SpRS and SRS) was used to quantify P/C ratio changes in sound, demi­neralized, and remineralized primary enamel at various depths. While both techniques provided valuable insights into mineral changes, it is important to reiterate that absolute P/C ratio values obtained from SpRS and SRS are not directly comparable due to inherent differences in their signal acquisition and processing [26]. However, the observed trends in demineralization and reminera­lization were consistently detected by both methods.

The results confirmed that demineralization significantly decreased phosphate content, whereas remine­ralization led to partial recovery. The initial phosphate concentration in sound enamel was significantly higher compared to demineralized enamel in superficial layers (0-20 µm), which aligns with a previous study conducted by Alkattan et al. [27], who stated that phosphate loss is an early indicator of enamel demineralization. This loss is attributed to the acidic dissolution of hydroxy­apatite (HA) crystals, leading to a relatively higher proportion of carbonate compared to phosphate in the lesion area [28].

After remineralization, the P/C ratio showed partial recovery, particularly in superficial layers. Phosphate ions contributing to this remineralization primarily originated from the synthetic saliva solution used in the 15-day pH cycling model. The presence of calcium and phosphate ions in saliva is crucial for enamel remi­neralization [29], and the contribution of these ions to mineral recovery was quantitatively verified by the observed increase in the P/C ratio.

At depths beyond 30 µm, the phosphate content remained stable throughout the experiment, showing no significant reduction during the demineralization phase. This confirms that the acid challenge was primarily surface-limited, leaving deeper subsurface layers intact [30]. Consequently, the lack of significant ‘recovery’ in these deep layers should be interpreted as a preservation of sound enamel structure rather than a failure of the remineralization agent. Significant remineralization effects were therefore naturally confined to superficial layers (0-20 µm), where the mineral loss actually occurred. This aligns with Godoi et al. [31], who found that fluoride treatments predominantly affect the outer­most enamel layers, where the demand for ion deposition is highest. A higher susceptibility of the surface enamel to acid dissolution compared to deeper layers is supported by previous research attributing this phenomenon to a higher porosity and carbonate content of the outer enamel [32].

However, for superficial layers that did undergo demineralization (especially in the 48 h-demin group), the restoration of P/C ratio was incomplete, which suggests that severe surface demineralization may result in irreversible structural changes [4]. Additionally, this finding demonstrates that remineralization is time-dependent; the 15-day period may have been insufficient for full restoration. Newly formed mineral deposits often differ structurally from natural hydroxyapatite, containing higher carbonate levels and lower crystallinity [33]. This difference in mineral quality explains why, even with fluoride treatment, the P/C ratio in the affected superficial layers did not fully return to baseline levels.

This supports previous research demonstrating that surface enamel is more vulnerable to acid attack due to its higher porosity and carbonate content [32]. Furthermore, the observed negative relationship across the lesion depth and SpRS P/C ratio (r = –0.39, p < 0.05) aligns with Hookham et al. [34], who revealed that the mineral loss increases with lesion progression.

Sound vs. demineralized enamel

The P/C ratio was significantly lower in deminera­lized enamel in superficial layers (0-20 µm), which may be because acid exposure leads to progressive mineral loss [35]. The most pronounced phosphate reduction occurred in the outermost 20 µm of enamel, which supports prior studies showing that surface layers are more susceptible to acid dissolution due to their higher poro­sity than sound enamel [36]. A study by Eisenburger [37] claimed that partial demineralization of enamel results from erosive impacts, and reported that erosive acid and interprismatic porosities lead to a 16.0 µm loss of material. The rate of demineralization of enamel was 62% for calcium and 64% for inorganic phosphorus when compared to initial mineral concentrations.

Demineralized vs. remineralized enamel

Remineralization led to a significant recovery of phosphate levels at superficial layers (0-20 µm) for the 24 h-demin group. However, for this samples at deeper depths (30-100 µm), no significant mineral loss was observed following demineralization. Consequently, the lack of significant increase in the P/C ratio in these layers during the remineralization phase reflects the preservation of sound enamel structure rather than a failure of remineralization treatment. For the 48 h-demin group, remineralization did not lead to a signi­ficant recovery of phosphate levels in superficial layers (0-20 µm) and as expected, deeper layers remained unchanged. This limited effectiveness in superficial lesion, particularly after longer demineralization, suggests that the 15-day remineralization period was insufficient to restore the mineral content in enamel subjected to more severe acid challenge, or that structural damage at the surface was irreversible [31].

Sound vs. remineralized enamel

While remineralized enamel demonstrated phosphate recovery, it remained significantly different from sound enamel, particularly in superficial layers. This observation, combined with findings from other study, implies that newly formed mineral deposits differ structurally from native hydroxyapatite [38]. Moreover, stu­dies using techniques, such as X-ray diffraction, have directly shown differences in crystallographic texture and crystallinity between native and remineralized enamel [39, 40], which is in line with our compositional data indicating incomplete restoration of original P/C ratio in the affected depths. According to our findings, the null hypothesis was rejected as there were significant diffe­rences in the P/C ratio across sound, demineralized, and remineralized deciduous enamel samples, with sound enamel exhibiting the highest ratio and demineralized enamel the lowest.

The outcomes of the present study have several important clinical consequences for dentistry. Raman spectroscopy offers a promising, noninvasive, and chairside method for early caries detection and monitoring remi­neralization, particularly beneficial for young patients, who may find traditional methods uncomfortable. By accurately measuring the P/C ratio, subclinical demine­ralization can be detected before visible lesions appear and track mineral recovery from therapies, such as fluo­ride, enabling proactive assessment of enamel health and improved preventive care in pediatric dentistry.

Limitations of this study include a short remine­ralization period, as remineralization after only 15 days was assessed. Also, using a single remineralizing agent is a limitation. The effectiveness of fluoride-based, calcium-based, and biomimetic remineralization strategies should be compared to determine the most efficient approach for phosphate recovery. Future studies should investigate the effects of extended remineralization protocols and different mineralizing agents.

Conclusions

Within the limitations of this in vitro study, it can be concluded that demineralization significantly reduces the P/C ratio in deciduous enamel, with the greatest mineral loss occurring in superficial layers. Reminera­lization led to partial P/C ratio recovery, primarily in superficial layers, though full restoration of the original mineral composition was not achieved. Raman spectroscopy with SpRS providing surface-level insights and SRS enabling depth-resolved analysis, demonstrated potential as a quantitative method for monitoring these mineral changes.

Disclosures

Author contributions: Conceptualization: A.S.T.A., N.A.A.; Methodology: A.S.T.A., N.A.A.; Questionnaire adaptation: A.S.T.A.; Investigation and data collection: A.S.T.A., N.A.H.A.; Formal analysis: N.A.A.; Data curation: A.S.T.A., N.A.A.; Writing of original draft: N.A.H.A., N.H.A.; Writing – review and editing: A.S.T.A., N.A.A., N.A.H.A., N.H.A.; Supervision: A.S.T.A.; Project admi­nistration: A.S.T.A. All authors have read and agreed to the published version of the manuscript.

Funding: This research received no external funding.

Institutional Review Board statement: Not applicable.

Informed consent statement: Not applicable.

Data availability statement: All the data generated or analyzed in this study are included in this manuscript. The data generated in this study may be requested from the corresponding author.

Acknowledgments: None.

Conflicts of interest: The authors declare no conflicts of interest.

AI use statement: Not applicable.

References