Effects of demineralization around orthodontic brackets bonded with different adhesive agents on fatigue-based shear bond strength: an in vitro comparative study Authors: Deniz Arslan Aydoğan, Orhan Cicek, Mehmet Yetmez Published: 2026 ABSTRACT Background and aimThere is limited evidence regarding the effect of demineralization around orthodontic brackets on shear bond strength (SBS). The aim of this in vitro fatigue study was to evaluate the effect of demineralization developing around orthodontic brackets bonded with different adhesive agents on fatigue-based SBS under laboratory conditions.MethodsA total of 60 extracted human maxillary premolars bonded with Gemini metal brackets were randomly allocated into three groups (n = 20). Bracket bonding was performed using Transbond™ XT (acid etching + primer + adhesive) in Group 1, GC Ortho Connect® (acid etching + one-step adhesive) in Group 2, and Transbond™ Plus (self-etch primer + adhesive) in Group 3. The mean demineralization values of the specimens were measured using a DIAGNOdent pen before demineralization (T0) and after 28 days of exposure to an artificial cariogenic environment (T1). Fatigue-based SBS was evaluated under cyclic loading (10 N, 0.5 Hz) at a crosshead speed of 300 mm/min using a low-cycle fatigue testing machine and expressed as the number of shear strokes to failure. The level of statistical significance was set at p < 0.05.ResultsThere were no significant differences in demineralization values among the groups at either T0 or T1 (p > 0.05). However, all groups showed a significant increase in demineralization at T1 compared with T0 (p < 0.05). Group 2 (GC Ortho Connect®) had a significantly higher number of shear strokes to failure than Groups 1 (Transbond™ XT) and 3 (Transbond™ Plus) (p < 0.05). No significant difference was observed between Groups 1 and 3 (p > 0.05). A significant negative correlation was observed between demineralization and shear strokes to failure across all groups (p < 0.05). Hard tissue (enamel and dentin) damage occurred in nine specimens in Groups 1 and 3, and in seven specimens in Group 2.ConclusionsIncreased demineralization may negatively affect fatigue-based SBS and increase the risk of hard tissue damage. Under demineralized conditions, GC Ortho Connect® may demonstrate more favorable fatigue-based bond performance compared with the other adhesive agents evaluated; however, further studies under clinically representative conditions are needed. Introduction The historical evolution of fixed orthodontic treatments clearly reflects a transition from conventional banded appliances to contemporary, minimally invasive bonding systems [1]. The introduction of acid-etching techniques accelerated this progression by enabling the direct bonding of orthodontic brackets to enamel surfaces [2, 3]. In this context, adhesive agents, which are responsible for bracket retention, have been recognized as critical determinants of bonding success. Numerous studies have highlighted the decisive influence of adhesive type and bonding protocol on bond strength [4–6]. Conventional orthodontic bonding procedures are based on the fundamental etch-and-rinse technique consisting of sequential steps of acid etching, rinsing, and drying. Typically performed using phosphoric acid, this method remains the gold standard due to the high bond strength it provides to enamel surfaces [7]. However, this multi-step protocol has several limitations, including technique sensitivity, prolonged application time, and increased susceptibility to contamination [8, 9]. Self-etch adhesive systems, which etch and prime in a single acidic solution, were introduced to address these drawbacks. These systems can be applied either in two steps involving primer and adhesive application or as single-step one-bottle formulations, substantially simplifying bonding procedures [8]. More recently, primer-integrated composite resins and one-step orthodontic adhesive systems, such as GC Ortho Connect®, have been introduced to further simplify bonding procedures and enhance clinical efficiency. Nevertheless, balancing reducing enamel damage during debonding with the potential reduction in bond strength resulting from simplified bonding protocols remains a subject of ongoing investigation in orthodontic adhesive research. The success of orthodontic bonding largely depends on effectively bonding brackets to the enamel surface using the appropriate techniques and meticulous application. Securely bonding brackets to enamel ensures mechanical stability throughout treatment and directly influences the overall success of the treatment. Ideal bond strength should be high enough to maintain bracket stability during treatment yet low enough to prevent enamel damage during debonding and allow for easy removal of residual adhesive [10]. This balance is particularly critical in the presence of demineralization. Previous studies have reported that demineralization can negatively impact the micromechanical and chemical bonding mechanisms of adhesive agents, resulting in reduced shear bond strength (SBS) [11, 12]. Fixed orthodontic appliances create new retention sites for dental plaque, facilitating the colonization of cariogenic bacteria such as Streptococcus mutans and Lactobacillus species, and triggering enamel demineralization through localized pH reduction [13, 14]. White spot lesions, which represent the earliest clinical manifestation of this process, may result in esthetic and structural problems following orthodontic treatment [15, 16]. Residual adhesive material around brackets and the biological properties of the adhesive agents used directly influence both the risk of demineralization and the success of orthodontic bonding [17]. Adhesive agents capable of fluoride release and exhibiting antibacterial properties have been reported to offer potential advantages in reducing enamel demineralization [18–20]. However, the effects of these agents on bond strength and their performance on demineralized enamel surfaces are still controversial. The SBS of orthodontic brackets, which reflects their ability to withstand functional masticatory forces, can be objectively assessed using SBS tests. Cyclic loading applied by fatigue testing machines simulates clinical mastication, enabling a realistic evaluation of bonding performance [21, 22]. Unlike conventional SBS measurements expressed in MPa, which are based on a single load-to-failure event, fatigue-based evaluation reflects the ability of the bracket–adhesive–enamel interface to withstand repeated subcritical forces over time. This approach is considered to better simulate intraoral conditions, where orthodontic brackets are subjected to cyclic masticatory forces rather than a single static load. Therefore, the use of the number of shear strokes to failure provides a more clinically relevant assessment of bond strength. Despite extensive research on the factors affecting conventional SBS [23–25], studies specifically investigating the effect of demineralization around brackets bonded with different adhesive agents on fatigue-based SBS remain limited. Therefore, evaluating the performance of different adhesive agents under demineralized conditions is essential for predicting orthodontic bonding success and supporting evidence-based clinical material selection. The aim of this in vitro study was to compare the effect of demineralization around orthodontic brackets bonded to extracted human maxillary first premolars using different adhesive agents on fatigue-based SBS under cyclic loading conditions. Brackets bonded using different adhesive agents and surface preparation protocols were evaluated in terms of their bond performance, expressed as the number of shear strokes to failure. The first null hypothesis was that there would be no significant differences among the groups in terms of the number of shear strokes to failure. The second null hypothesis was that there would be no significant association between demineralization levels and the number of shear strokes to failure. Materials and methods Ethical approval and sample size calculation This study was conducted using 60 extracted human maxillary first premolar teeth obtained from patients who applied to the Department of Orthodontics, Faculty of Dentistry, Zonguldak Bülent Ecevit University, for orthodontic treatment. Ethical approval was granted by the Zonguldak Bülent Ecevit University Non-Interventional Clinical Research Ethics Committee (approval date: 31 May 2023; decision no: 2023/11–14). A priori power analysis was performed using G*Power software (version 3.1.9.4; Heinrich Heine University Düsseldorf, Düsseldorf, Germany). Sample size estimation was based on data from a previous study by Joseph et al. [26] comparing bond strength among different adhesive agents, using one-way ANOVA. With an effect size of 0.4566, a statistical power (1–β) of 0.80, and an alpha level of 0.05, a minimum of 51 specimens (17 per group) was required (actual power: 0.8147). To enhance statistical robustness, a total of 60 specimens (n = 20 per group) were included in the present study. Sample preparation and bracket bonding procedures The specimens used in the present study were derived from a specialty thesis conducted in 2022 under the supervision of Assoc. Prof. Orhan Cicek [27, 28] and were reused for a secondary analysis focusing on fatigue-based SBS under cyclic loading conditions. In the original study, extracted human maxillary premolars were bonded with orthodontic brackets, subjected to the demineralization protocol, and subsequently evaluated for shear bond performance under standardized experimental conditions. The present manuscript transparently reports the secondary use of these previously prepared specimens, while the fatigue-based bond performance was re-evaluated and comparatively analyzed within the scope of the current study under the same experimental testing conditions. It should be noted that the specimens used in the present study had already undergone the demineralization process in the original study and were reused in their demineralized state. No additional demineralization procedures were performed, and the present study was limited to fatigue-based SBS testing. Inclusion criteria required teeth to be free of restorations, caries, or fillings, with no enamel cracks, fractures, or forceps marks, and without fluorosis; donor age, sex, and extraction quadrant were not considered. Prior to bracket bonding, the teeth were stored in 0.1% thymol solution at room temperature in a dark environment, with the solution renewed monthly [28]. A total of 60 extracted teeth were randomly assigned to three experimental groups (n = 20 per group), and all specimens were bonded with Gemini metal brackets (3 M Unitek, Monrovia, CA, USA). To standardize the specimens and restrict the analysis to the coronal enamel, the teeth were sectioned at the cementoenamel junction prior to bonding, and the exposed pulp chambers were sealed with composite resin. In Group 1, enamel surfaces were conditioned with 37% phosphoric acid, followed by the application of Transbond™ XT Primer, and brackets were bonded using Transbond™ XT Light Cure Adhesive. In Group 2, enamel conditioning was performed with 37% phosphoric acid, after which brackets were bonded using the self-priming, one-step adhesive GC Ortho Connect® Light Cure Adhesive without separate primer application. In Group 3, Transbond™ Plus Self-Etching Primer was applied to the cleaned enamel surfaces without prior acid etching, and bracket bonding was completed using Transbond™ XT Light Cure Adhesive. Light polymerization was performed in all groups using an ESPE Elipar S10 light-curing unit (3 M ESPE, St Paul, MN, USA). The adhesive materials and their compositions according to the experimental groups are presented in Table 1 [27, 28]. Table 1Adhesive agents and surface conditioning protocols used in the groups and their chemical compositionsGroupsEtching AgentAdhesive AgentAdhesiveCompositionCompositeManufacturerGroup 137% phosphoric acidTransbond™ XT PrimerBisphenol A diglycidyl ether dimethacrylate (BISGMA)Triethylene glycol dimethacrylate (TEGDMA)4-(Dimethylamino)-benzene ethanolTransbond™ XT3 M Unitek, Monrovia, CA, USAGroup 237% phosphoric acidGC Ortho Connect→4,4′-Isopropylidenediphenol ethoxylated 2-methylprop-2-enoic acid esterification productsUrethane dimethacrylate (UDMA)Methacryloxydecyl dihydrogen phosphate6-tert-butyl-2,4-xylenolDiphenyl (2,4,6-trimethylbenzoyl) phosphine oxide-GC Corporation, Tokyo, JapanGroup 3-Transbond™ Plus Self-Etching Primer2-propenoic acid, 2-methyl-, 2-hydroxyethyl ester reaction products with phosphorus oxide (P₂O₅)DL-camphorquinoneN, N-dimethylbenzocaineTransbond™ XT3 M Unitek, Monrovia, CA, USA Table 1Adhesive agents and surface conditioning protocols used in the groups and their chemical compositionsGroupsEtching AgentAdhesive AgentAdhesiveCompositionCompositeManufacturerGroup 137% phosphoric acidTransbond™ XT PrimerBisphenol A diglycidyl ether dimethacrylate (BISGMA)Triethylene glycol dimethacrylate (TEGDMA)4-(Dimethylamino)-benzene ethanolTransbond™ XT3 M Unitek, Monrovia, CA, USAGroup 237% phosphoric acidGC Ortho Connect→4,4′-Isopropylidenediphenol ethoxylated 2-methylprop-2-enoic acid esterification productsUrethane dimethacrylate (UDMA)Methacryloxydecyl dihydrogen phosphate6-tert-butyl-2,4-xylenolDiphenyl (2,4,6-trimethylbenzoyl) phosphine oxide-GC Corporation, Tokyo, JapanGroup 3-Transbond™ Plus Self-Etching Primer2-propenoic acid, 2-methyl-, 2-hydroxyethyl ester reaction products with phosphorus oxide (P₂O₅)DL-camphorquinoneN, N-dimethylbenzocaineTransbond™ XT3 M Unitek, Monrovia, CA, USA Adhesive agents and surface conditioning protocols used in the groups and their chemical compositions Bisphenol A diglycidyl ether dimethacrylate (BISGMA) Triethylene glycol dimethacrylate (TEGDMA) 4-(Dimethylamino)-benzene ethanol 4,4′-Isopropylidenediphenol ethoxylated 2-methylprop-2-enoic acid esterification products Urethane dimethacrylate (UDMA) Methacryloxydecyl dihydrogen phosphate 6-tert-butyl-2,4-xylenol Diphenyl (2,4,6-trimethylbenzoyl) phosphine oxide 2-propenoic acid, 2-methyl-, 2-hydroxyethyl ester reaction products with phosphorus oxide (P₂O₅) DL-camphorquinone N, N-dimethylbenzocaine Following completion of the demineralization protocol, the specimens were stored in distilled water for 24 h prior to testing, in accordance with the storage protocol used during the original specimen preparation. The fatigue-based testing protocol applied in the present study was specifically implemented to evaluate bond performance under cyclic loading conditions, while maintaining consistency with the original specimen preparation procedures. Demineralization protocol and DIAGNOdent measurements Demineralization was induced, as previously described in detail [27], by exposing the specimens to an artificial cariogenic environment designed to simulate clinical conditions, consisting of artificial saliva supplemented with Streptococcus mutans and sucrose. Each tooth specimen was individually incubated in 2 mL of the cariogenic suspension under continuous agitation using a rotary shaker to ensure homogeneous contact. The experimental setup was maintained at 37 °C under a 10% CO₂ atmosphere for 28 days, during which the cariogenic suspension was renewed every 48 h [28–30]. Following completion of the bonding procedures, demineralization values were measured using a DIAGNOdent pen (KaVo, Biberach, Germany) and expressed in arbitrary units (0–99) [31]. Measurements were obtained from the mesial, distal, occlusal, and gingival margins of each bracket, and the mean of these values was recorded as baseline (T0). After all specimens had been stored in artificial saliva containing Streptococcus mutans for 28 days, measurements were repeated at the same sites, and the mean values were recorded as T1 [27, 28]. Following T1 assessment, all specimens were immediately stored in distilled water supplemented with 0.1% thymol in a dark environment to prevent bacterial proliferation until SBS testing was performed [30]. Specimen preparation and fatigue-based SBS testing Prior to testing, all teeth were removed from the storage solution, rinsed, and air-dried. Specimens were embedded in acrylic blocks using a four-compartment stainless steel mold (40 × 25 mm height-to-width ratio, 15 mm depth). To standardize specimen positioning and minimize angular deviations, a 0.019 × 0.025-inch stainless steel archwire was inserted into the bracket slots and secured with elastic ligatures. The ends of the archwire were stabilized with wax, and the molds were filled with autopolymerizing cold-curing acrylic resin (Vertex, Vertex Dental B.V., Zeist, The Netherlands), leaving the buccal surfaces exposed (see Fig. 1). Care was taken to prevent contact between the resin and areas adjacent to the bracket margins. Fig. 1Fixation of the specimens in the molds and embedding in acrylic resin with the buccal surfaces exposed: A Occlusal view of the mold assembly during specimen fixation; B Lateral view showing the positioning of the acrylic blocks and exposed buccal surfaces Fixation of the specimens in the molds and embedding in acrylic resin with the buccal surfaces exposed: A Occlusal view of the mold assembly during specimen fixation; B Lateral view showing the positioning of the acrylic blocks and exposed buccal surfaces After polymerization, specimens were removed from the molds, numbered, and stored in distilled water for 24 h prior to testing. No thermal cycling procedure was applied prior to fatigue-based SBS testing. Fatigue-based SBS was evaluated using a cyclic loading protocol and expressed as the number of shear strokes to failure. Testing was performed using a low-cycle, closed-loop controlled fatigue testing machine operating at a crosshead speed of 300 mm/min and subjected to a constant load of 10 N. This loading magnitude was selected based on previous in vitro studies simulating masticatory forces acting on orthodontic brackets and was considered appropriate for providing standardized and clinically relevant cyclic loading conditions for comparative fatigue-based evaluation [22, 32]. During testing, the angle between the metal loading nozzle and the bracket slot was standardized at 90°. Acrylic blocks were rigidly secured in a custom-made fixture, and cyclic loading was applied perpendicular to the bracket–enamel interface until failure occurred. Loading was performed at a frequency of 0.5 Hz under load-controlled conditions using a triangular waveform. The load ratio (R) was set to 0 (minimum load: 0 N; maximum load: 10 N), and one shear stroke was defined as a complete loading–unloading cycle (0–10–0 N). The number of shear strokes to failure was recorded for each specimen (see Fig. 2). In cases where failure occurred before completion of a full loading–unloading cycle, the number of shear strokes to failure was recorded as a fractional value (e.g., 0.5). This indicating that failure occurred during the loading phase of a cycle. Fig. 2Preparation of acrylic block specimens and fatigue-based SBS testing setup: A Frontal view of an acrylic block specimen showing the bonded bracket on the buccal surface; B Lateral view showing specimen orientation within the acrylic block; C Placement of the acrylic block specimens in the fatigue testing machine; D Custom-made fixture and loading configuration used to stabilize the specimens during fatigue-based SBS testing Preparation of acrylic block specimens and fatigue-based SBS testing setup: A Frontal view of an acrylic block specimen showing the bonded bracket on the buccal surface; B Lateral view showing specimen orientation within the acrylic block; C Placement of the acrylic block specimens in the fatigue testing machine; D Custom-made fixture and loading configuration used to stabilize the specimens during fatigue-based SBS testing Hard tissue damage assessment After fatigue-based SBS testing was completed, all specimens were inspected for signs of hard tissue damage. A single examiner (D.A.A) conducted the evaluation under consistent lighting conditions to ensure standardization. Enamel damage was identified based on surface irregularities, such as cracks, structural loss, or disruptions at the bracket–enamel interface. Dentin involvement was noted when the underlying dentin was exposed following bracket debonding. Since no magnification or microscopic techniques were used, the assessment was limited to a macroscopic visual examination. Hard tissue damage was recorded qualitatively as either present or absent for each specimen. Statistical analysis Statistical analyses were performed using SPSS software (Statistical Package for the Social Sciences, version 26.0; IBM Corp., New York, NY, USA). Data normality was assessed using the Shapiro–Wilk test. As demineralization data followed a normal distribution, intergroup comparisons were conducted using one-way analysis of variance (ANOVA), with post hoc multiple comparisons performed using the Tukey test. Intragroup comparisons between T0 and T1 were evaluated using the paired-sample t-test. As the number of shear strokes to failure did not show a normal distribution, intergroup comparisons were performed using the Kruskal–Wallis test, followed by pairwise comparisons with the Mann–Whitney U test. The relationship between demineralization values and the number of shear strokes to failure was assessed using Spearman’s rho correlation coefficients. The level of statistical significance was set at p < 0.05. Results No statistically significant differences were observed among the groups in baseline demineralization values at T0 (p = 0.996). Similarly, demineralization values measured at T1, following 28 days of exposure to the artificial cariogenic environment, did not differ significantly between groups (p = 0.356). ANOVA revealed no significant intergroup differences in the magnitude of demineralization change between T0 and T1 (p = 0.486) (Table 2). Table 2Intergroup comparison of demineralization values at baseline (T0), at 28 days (T1), and changes between T0 and T1Sum of squaresdfMean squareF p T0Between groups0.00120.0010.0040.996Within groups7.193570.126Total7.19459T1Between groups0.89020.4451.0510.356Within groups24.137570.423Total25.02759T0/T1Between groups0.81220.4060.7300.486Within groups31.715570.556Total32.52759df: degrees of freedom, T0: baseline demineralization measurement, T1: demineralization measurement at 28 days, T0/T1: change in demineralization values between baseline and 28 days, F: F-statistic, p: significance value Table 2Intergroup comparison of demineralization values at baseline (T0), at 28 days (T1), and changes between T0 and T1Sum of squaresdfMean squareF p T0Between groups0.00120.0010.0040.996Within groups7.193570.126Total7.19459T1Between groups0.89020.4451.0510.356Within groups24.137570.423Total25.02759T0/T1Between groups0.81220.4060.7300.486Within groups31.715570.556Total32.52759df: degrees of freedom, T0: baseline demineralization measurement, T1: demineralization measurement at 28 days, T0/T1: change in demineralization values between baseline and 28 days, F: F-statistic, p: significance value Intergroup comparison of demineralization values at baseline (T0), at 28 days (T1), and changes between T0 and T1 df: degrees of freedom, T0: baseline demineralization measurement, T1: demineralization measurement at 28 days, T0/T1: change in demineralization values between baseline and 28 days, F: F-statistic, p: significance value In contrast, paired-sample t-test analysis demonstrated a significant increase in demineralization values at T1 compared with T0 in all groups (p = 0.001 for all groups). The mean ± standard deviation values for demineralization at T0, at T1, and the T0–T1 changes for each group are presented in Table 3. Table 3Intergroup and intragroup comparisons of demineralization values at baseline (T0), at 28 days (T1), and changes between T0 and T1Group 1Group 2Group 3 p T0(Mean ± SD)2.66 ± 0.352.65 ± 0.252.66 ± 0.420.996 AT1(Mean ± SD)6.34 ± 0.756.08 ± 0.596.08 ± 0.590.356 A Intragroup p 0.001* P0.001* P0.001* PT0/T1(Mean ± SD)-3.67 ± 0.80-3.65 ± 0.71-3.41 ± 0.700.486 AAone-way analysis of variance (ANOVA) for intergroup comparisons, Ppaired-sample t-test for intragroup comparisons, T0: baseline demineralization measurement; T1: demineralization measurement at 28 days, T0/T1: change in demineralization values between baseline and 28 days, SD Standard deviationp: significance value, *p < 0.05 Table 3Intergroup and intragroup comparisons of demineralization values at baseline (T0), at 28 days (T1), and changes between T0 and T1Group 1Group 2Group 3 p T0(Mean ± SD)2.66 ± 0.352.65 ± 0.252.66 ± 0.420.996 AT1(Mean ± SD)6.34 ± 0.756.08 ± 0.596.08 ± 0.590.356 A Intragroup p 0.001* P0.001* P0.001* PT0/T1(Mean ± SD)-3.67 ± 0.80-3.65 ± 0.71-3.41 ± 0.700.486 AAone-way analysis of variance (ANOVA) for intergroup comparisons, Ppaired-sample t-test for intragroup comparisons, T0: baseline demineralization measurement; T1: demineralization measurement at 28 days, T0/T1: change in demineralization values between baseline and 28 days, SD Standard deviationp: significance value, *p < 0.05 Intergroup and intragroup comparisons of demineralization values at baseline (T0), at 28 days (T1), and changes between T0 and T1 Aone-way analysis of variance (ANOVA) for intergroup comparisons, Ppaired-sample t-test for intragroup comparisons, T0: baseline demineralization measurement; T1: demineralization measurement at 28 days, T0/T1: change in demineralization values between baseline and 28 days, SD Standard deviation p: significance value, *p < 0.05 Statistically significant differences were observed among the groups in the number of shear strokes to failure (p = 0.016). The lowest median number of shear strokes was observed in Group 1 bonded with Transbond™ XT Primer [1 (IQR: 1–2.75; mean ± SD: 4.25 ± 9.06)], whereas the highest median value was recorded in Group 2 bonded with GC Ortho Connect® [2 (IQR: 2–17; mean ± SD: 15.15 ± 26.75)]. Group 3 bonded with Transbond™ Plus demonstrated a median of 1.5 (IQR: 1–4.25; mean ± SD: 6.00 ± 13.05) (Table 4). Pairwise comparisons revealed that the number of shear strokes in Group 1 was significantly lower than those in Group 2 (p = 0.009), while no significant difference was detected between Groups 1 and 3 (p = 0.542). In contrast, Group 3 exhibited a significantly lower number of shear strokes than Group 2 (p = 0.029). Overall, Group 2 demonstrated a significantly higher number of shear strokes compared with both Groups 1 and 3 (Table 4). Table 4Comparison of the number of shear strokes to failure between groupsVariablesGroup 1aGroup 2bGroup 3c p Shear strokesMin–Max1–401–931–58Mean ± SD4.25 ± 9.0615.15 ± 26.75ac6.00 ± 13.050.016 * KMedian (Q1–Q3)1 (1–2.75)2 (2–17)1.5 (1–4.25)SD Standard deviation, Min Minimum, Max MaximumKKruskal–Wallis test (pairwise comparisons were performed using the Mann–Whitney U test)p: significance value, *p < 0.05a Significant difference compared with the Transbond™ XT groupb Significant difference compared with the GC Ortho Connect® groupc Significant difference compared with the Transbond™ Plus group Table 4Comparison of the number of shear strokes to failure between groupsVariablesGroup 1aGroup 2bGroup 3c p Shear strokesMin–Max1–401–931–58Mean ± SD4.25 ± 9.0615.15 ± 26.75ac6.00 ± 13.050.016 * KMedian (Q1–Q3)1 (1–2.75)2 (2–17)1.5 (1–4.25)SD Standard deviation, Min Minimum, Max MaximumKKruskal–Wallis test (pairwise comparisons were performed using the Mann–Whitney U test)p: significance value, *p < 0.05a Significant difference compared with the Transbond™ XT groupb Significant difference compared with the GC Ortho Connect® groupc Significant difference compared with the Transbond™ Plus group Comparison of the number of shear strokes to failure between groups SD Standard deviation, Min Minimum, Max Maximum KKruskal–Wallis test (pairwise comparisons were performed using the Mann–Whitney U test) p: significance value, *p < 0.05 a Significant difference compared with the Transbond™ XT group b Significant difference compared with the GC Ortho Connect® group c Significant difference compared with the Transbond™ Plus group Correlation analysis demonstrated a statistically significant negative association between demineralization values and the number of shear strokes to failure in all groups (p < 0.05). Spearman correlation coefficients were − 0.585 for Group 1 (p = 0.007), − 0.518 for Group 2 (p = 0.019), and − 0.556 for Group 3 (p = 0.011) (Table 5). Table 5Spearman correlation analysis between demineralization and the number of shear strokes to failure by groupNumber of shear strokes to failureGroup 1Group 2Group 3DemineralizationGroup 1Spearman rho-0.585 p 0.007*Group 2Spearman rho-0.518 p 0.019*Group 3Spearman rho-0.556 p 0.011*p: significance value; *p < 0.05 Table 5Spearman correlation analysis between demineralization and the number of shear strokes to failure by groupNumber of shear strokes to failureGroup 1Group 2Group 3DemineralizationGroup 1Spearman rho-0.585 p 0.007*Group 2Spearman rho-0.518 p 0.019*Group 3Spearman rho-0.556 p 0.011*p: significance value; *p < 0.05 Spearman correlation analysis between demineralization and the number of shear strokes to failure by group p: significance value; *p < 0.05 Following fatigue-based SBS testing, hard tissue damage involving both enamel and dentin was observed in all groups. Enamel–dentin damage was identified in 9 specimens in Group 1, 7 specimens in Group 2, and 9 specimens in Group 3 (see Fig. 3). Fig. 3Representative examples of hard tissue (enamel and dentin) damage involving both enamel and dentin after fatigue-based SBS testing: A Group 1, B Group 2, and C Group 3 Representative examples of hard tissue (enamel and dentin) damage involving both enamel and dentin after fatigue-based SBS testing: A Group 1, B Group 2, and C Group 3 Discussion Reliable bonding of orthodontic brackets depends on the long-term stability of the bracket–adhesive–enamel interface, and disruption of this interface has been associated with increased bracket failure and reduced clinical efficiency [28, 29]. In addition to patient-related factors, the selection of adhesive systems and bonding protocols plays a decisive role in clinical bonding outcomes [33, 34]. In this regard, Dudas et al. [35] demonstrated that variations in adhesive agents and application procedures significantly influence bracket failure rates. Furthermore, clinical studies have reported a direct association between enamel demineralization and bracket failure during fixed orthodontic treatment [36, 37]. Despite these findings, most previous investigations have examined either bond strength or demineralization independently, while studies addressing both parameters simultaneously remain limited [28, 38]. Therefore, the present study evaluated the fatigue-based SBS of orthodontic brackets bonded with three different adhesive agents following demineralization, using cyclic fatigue loading, with bond performance expressed as the number of shear strokes to failure. Although in vivo bond strength studies more closely reflect clinical reality, the inability to control intraoral variables constitutes a major limitation [39]. In contrast, in vitro designs enable strict standardization of experimental conditions, allowing objective evaluation of the tested materials under controlled settings [39]. Accordingly, bond performance in the present study was evaluated under in vitro conditions, with all specimens tested in a uniform and standardized manner. Selveraj et al. [40] reported that orthodontic brackets and archwires are exposed to multidirectional forces in the oral environment, and that such complex loading patterns play a critical role in material fatigue behavior and clinical longevity. Consistent with this concept, Cicek et al. [21] and Eser et al. [22] employed fatigue testing machines to simulate cyclic mechanical forces associated with masticatory function under laboratory conditions. The present study adopted a fatigue testing approach using a custom-designed loading configuration under controlled cyclic loading conditions (10 N, 0.5 Hz, 300 mm/min) [21, 22, 32]. In contrast to conventional SBS testing performed using universal testing machines (e.g., Instron), which rely on a single load-to-failure event, the present approach evaluates bond performance under cyclic loading conditions by recording the number of shear strokes to failure. For this reason, fatigue-based testing was preferred over conventional SBS testing, as it allows a more clinically relevant evaluation of bond strength under conditions that better simulate intraoral functional loading. However, the relatively low median number of shear strokes to failure observed in the present study may indicate that some specimens failed at early loading cycles rather than undergoing prolonged fatigue. This suggests that, under the selected loading conditions (10 N, 0.5 Hz), failure may have occurred in a quasi-static manner in certain cases, potentially limiting the ability of the test protocol to fully capture long-term fatigue behavior. Nevertheless, the use of a constant cyclic loading protocol provides a standardized and reproducible framework for comparing bond strength, which cannot be adequately captured by conventional single-load SBS measurements expressed in MPa. Numerous studies investigating bracket failure in fixed orthodontic treatment have emphasized the influence of adhesive selection and bonding protocols on clinical outcomes [7, 34, 41]. In their 2025 systematic review and meta-analysis, Sánchez et al. [7] reported that the bond strength of metal brackets varies significantly depending on both the adhesive type and the application procedure. Consistent with these findings, the present study demonstrated that fatigue-based bond performance differed among adhesive agents, despite comparable levels of demineralization. Specifically, brackets bonded with the self-priming one-step adhesive (GC Ortho Connect®) exhibited higher fatigue-based SBS compared with both the conventional three-step system (Transbond™ XT) and the two-step self-etch adhesive system (Transbond™ Plus). These findings suggest that differences in adhesive composition and application protocols may influence bond durability under cyclic loading conditions. Orthodontic brackets bonded to tooth surfaces have been shown to create areas that retain microbial dental plaque and predispose adjacent enamel to demineralized, caries-related lesions [42]. In their 2024 systematic review, Duś-Ilnicka et al. [13] reported a direct association between fixed appliance therapy, particularly in adult patients, and increased caries prevalence. The present study assessed the effects of demineralization on bonded teeth by exposing bracketed specimens bonded with different adhesive systems to an artificial cariogenic environment at 37 °C for 28 days. Analyses revealed a significant increase in mean demineralization values from T0 to T1 across all groups. These results align with prior studies indicating that orthodontic brackets encourage plaque accumulation and subsequent demineralization in the surrounding enamel [42, 43]. Previous studies evaluating the association between adhesive type and demineralization around orthodontic brackets have reported variable results. Horan et al. [44] observed increased white spot lesion incidence across brackets bonded with Transbond™ XT, Transbond™ Plus, and GC Ortho Connect®, with significantly greater lesion number and severity in the GC Ortho Connect® group. In contrast, the present study demonstrated significant demineralization increases in all adhesive groups, with comparable magnitudes among them, a discrepancy that may be explained by methodological differences between clinical and in vitro designs. Similarly, Narendran and Raghunath [45] reported greater demineralization around brackets bonded with conventional etch-and-rinse systems, particularly Transbond™ XT, but attributed this finding primarily to plaque-related carious activity rather than the etching protocol itself. Consistent with this interpretation, no significant differences in demineralization were observed between conventionally etched groups (Transbond™ XT and GC Ortho Connect®) and the self-etch adhesive group (Transbond™ Plus) in the present study. Comparative evaluations of bond strength among orthodontic attachments bonded with different adhesive agents have produced inconsistent results. Several studies have reported higher SBS for Transbond™ XT than for alternative adhesive systems [46–48]. Rosidin et al. [48] and Kamboj et al. [49] demonstrated superior bond strength for Transbond™ XT compared with Transbond™ Plus, using metal brackets and orthodontic molar tubes, respectively. However, Singh et al. [50] reported higher bond strength for brackets bonded with Transbond™ Plus, and Yadala et al. [51] found no significant difference in bond strength between Transbond™ XT and Transbond™ Plus. This variability may be attributed to differences in bracket type, sample size, storage conditions, and testing protocols. In the present study, no significant difference was observed between Transbond™ XT and Transbond™ Plus with respect to the number of shear strokes to failure. Recently developed adhesive systems have been introduced to simplify orthodontic bonding by reducing the complexity of conventional multi-step etch-and-rinse protocols. Among these, the self-priming one-step adhesive GC Ortho Connect® has been reported to reduce application time and technique sensitivity while maintaining adequate bond performance (GC Orthodontics Europe, Product Information Sheet, 2016). Previous studies have demonstrated that GC Ortho Connect® achieves bond strength values comparable to those of conventionally etched TransbondTM XT, with clinically acceptable performance [4, 52–55]. Uslu et al. [56] found no significant difference in SBS between GC Ortho Connect® and Transbond™ XT when bonding metal brackets, with both systems exhibiting higher values than Transbond™ Plus. Similarly, Perkovic et al. [52] reported comparable bond strength between Transbond™ XT and GC Ortho Connect®, highlighting the latter as a favorable alternative due to reduced clinical steps. In contrast, Bahrami et al. [6] observed significantly higher SBS for GC Ortho Connect® compared with Transbond™ XT, while Helou et al. [54] reported superior performance of GC Ortho Connect® relative to both Transbond™ XT and Transbond™ Plus. In agreement with this variability, the present study demonstrated that fatigue-based bond performance varied among adhesive agents, with GC Ortho Connect® exhibiting a significantly higher number of shear strokes to failure compared with both Transbond™ XT and Transbond™ Plus, whereas no significant difference was observed between the latter two. These findings suggest that simplified adhesive agents may provide enhanced bond durability under cyclic loading conditions, potentially due to differences in adhesive composition and interfacial interaction with enamel. Accordingly, the first null hypothesis was partially rejected. Previous studies have demonstrated that the SBS of orthodontic metal brackets may vary depending on adhesive type and bonding protocol [4–6], whereas evidence regarding adhesive performance under demineralized conditions remains limited. Turğut [57] reported comparable SBS and similar increases in white spot lesions for GC Ortho Connect® and Transbond™ XT after 12 months of clinical use. In contrast, the present study demonstrated that, despite similar levels of demineralization across all groups, the number of shear strokes to failure differed significantly among the adhesive agents. Importantly, median values demonstrated a consistent trend, supporting the robustness of the observed differences despite the non-normal distribution of the data. GC Ortho Connect® exhibited higher fatigue-based SBS compared with Transbond™ XT under identical etching and demineralization conditions. These differences may be attributed to variations in adhesive composition and their micromechanical interaction with the enamel surface. Furthermore, the observed decrease in the number of shear strokes to failure with increasing demineralization supports the rejection of the second null hypothesis. From a clinical perspective, while acid etching of the enamel surface enhances adhesive bonding strength, it has also been associated with an increased risk of hard tissue damage during bracket debonding [58]. Rosidin et al. [48] reported longer enamel cracks in brackets bonded with Transbond™ XT compared with Transbond™ Plus following bond strength testing, while Kamboj et al. [49] observed reduced structural damage and surface irregularities in teeth bonded with Transbond™ Plus. However, the relationship between SBS and enamel damage remains controversial [59]. Rodríguez-Chávez et al. [59] demonstrated that etch-and-rinse systems may achieve high bond strength without inducing enamel damage, whereas limited damage was observed even in some lower-strength self-etch systems. In the present study, hard tissue damage involving both enamel and dentin was observed in 25 of 60 specimens. Although similar demineralization conditions were applied, the incidence of damage varied among adhesive agents, with GC Ortho Connect® showing the lowest frequency of enamel–dentin damage. Notably, this finding was observed despite its higher fatigue-based SBS, suggesting that bond strength under cyclic loading may not be directly associated with an increased risk of hard tissue damage. This study has several limitations that should be considered when interpreting the findings, including the absence of a non-demineralized control group, which limits the ability to directly compare the effects of demineralization on bond performance. As the present study was conducted under in vitro conditions, the results may not fully reflect the complex biological and mechanical environment of the oral cavity. In addition, the absence of thermal cycling may be considered a limitation, as thermal fluctuations in the oral environment could influence bond performance over time. Furthermore, as the present study involved the secondary use of previously prepared specimens, the findings should be interpreted within the context of the original specimen preparation and storage protocol. Additionally, donor age was not considered during sample selection. Although enamel properties such as mineral content and bonding characteristics may vary with age, previous in vitro studies have reported that methodological variability—particularly differences in enamel preparation and bonding protocols—can significantly influence bond strength outcomes, highlighting the importance of standardized procedures in minimizing such variability [28, 60]. The evaluation was further limited to metal brackets and three adhesive systems, restricting the generalizability of the findings to other bracket types and bonding materials. Moreover, the use of fatigue-based SBS as an outcome measure may not fully capture all aspects of clinical bond behavior. The relatively low number of loading cycles to failure observed in some specimens further suggests that the applied loading conditions may not fully represent long-term fatigue behavior. Future studies incorporating a broader range of materials, more advanced diagnostic methods, and appropriate control groups are warranted to further elucidate the effects of demineralization on bond performance. Despite these limitations, the study was conducted under highly standardized experimental conditions, which enhances the reliability of the findings. Conclusions The findings of this in vitro study suggest that increased demineralization around orthodontic brackets may negatively affect fatigue-based SBS and increase the risk of hard tissue damage. Although demineralization levels were comparable among adhesive agents, differences were observed in fatigue-based bond performance. Among the materials evaluated, GC Ortho Connect® showed a tendency toward higher fatigue-based SBS with a lower incidence of hard tissue damage compared with the other adhesive agents. These findings highlight the potential influence of adhesive selection under demineralized conditions; however, further clinical and experimental studies are needed to confirm these findings.