ORIGINAL RESEARCH |
https://doi.org/10.5005/jp-journals-10015-2281 |
Evaluation of the Effect of Bioactive Varnish Application on the Marginal Micro Gap of Class II Composite Restoration: SEM Analysis
1–6Department of Conservative Dentistry & Endodontics, Vishnu Dental College, Bhimavaram, Andhra Pradesh, India
Corresponding Author: Bagu V Sindhuja, Department of Conservative Dentistry & Endodontics, Vishnu Dental College, Bhimavaram, Andhra Pradesh, India, Phone: +91 8885581369, e-mail: sindhujabagu@gmail.com
Received on: 02 August 2023; Accepted on: 04 September 2023; Published on: 13 October 2023
ABSTRACT
Aim: This research aimed to evaluate the marginal micro gap of class II composite resin restorations sealed with bioactive calcium (Ca) phosphate (P)—based fluoride (F) varnish using the scanning electron microscope (SEM) at 200× magnification.
Materials andmethods: Box-only cavities were prepared in 24 samples and restored with prewarmed bulk-fill composite. Samples in the control group (n = 12) were left intact as a conventional restorative protocol. Cavosurface margins of restorations in the test group (n = 12) were sealed with casein phosphopeptide-amorphous calcium phosphate fluoride (CPP-ACPF) varnish (MI varnish, GC America). After the thermocycling procedure for 500 cycles, samples were stored in artificial saliva for 30 days. Then washed samples were dried and sectioned mesiodistally and observed under SEM at 200× for marginal micro gap and mineral deposition along the cavosurface margin with energy-dispersive X-ray analysis (EDAX). Samples were scored according to Blunck and Zaslansky’s criteria. Results were statistically analyzed with a Fisher’s exact test and Mann–Whitney U test.
Results: Fisher’s exact test showed a significant difference between the control group and test group between the buccal wall (p = 0.02), base (gingival wall) (p = 0.05), and palatal wall (p = 0.02) for the marginal micro gap. The mean Ca/P ratios of the CPP-ACPF varnish group, that is, the test group were higher than the control group, without statistically significant differences regarding the mineral analysis.
Conclusion: The test group with CPP-ACPF application showed a lesser marginal micro gap with statistically significant results than the control group.
Clinical significance: The application of bioactive varnish along the cavosurface margins reduces the marginal micro gap. This might prevent postoperative sensitivity and secondary caries leading to long-term clinical success.
How to cite this article: Sajjan GS, Sindhuja BV, K MV, et al. Evaluation of the Effect of Bioactive Varnish Application on the Marginal Micro Gap of Class II Composite Restoration: SEM Analysis. World J Dent 2023;14(9):772–776.
Source of support: Nil
Conflict of interest: None
Keywords: Bio-active varnish, Bulk fill composite, Class II restorations, Composite surface sealer, In vitro study, Marginal micro gap, Scanning electron microscope
INTRODUCTION
Dental composite resins have proven to be materials of choice for many dentists due to the increased demand for high-quality esthetic results in everyday dentistry.1
However, problems such as polymerization shrinkage and marginal microleakage still occur despite the continuous evolution of these resins. Due to dimensional change, micro gaps can develop, resulting in adhesive interface degradation, microleakage of saliva and bacteria, secondary caries, pulpal changes, and clinical failure of the restoration.2 It is, therefore, imperative to maintain the marginal integrity of composite resin restoration to achieve long-term clinical success.3
An average gap of 11.47 µm from the prepared tooth surface was exhibited with room-temperature bulk-fill composite restorations.4 Marginal micro gaps of 0.7–0.8 µm in enamel and 5.2–5.5 µm in dentin with prewarmed bulk fill restorations were reported.5 A recent clinical trial of class II composite restorations reported failure of restorations due to sensitivity and fractures at the end of 2 years.6 Many randomized clinical trials have reported better success with composite restorations but with the persisting issue of marginal integrity.7 So, long-term sealing of composite resin-tooth interface has gained significant importance. Many surface sealants were developed to address the marginal micro gap associated with composite restorations. Many low-viscosity sealants containing resin monomers were developed, which could flow into the micro gaps at the resin-tooth interface for restrengthening the junction. However, the real perfect seal was not achieved.8 A 12-month clinical trial evaluated the influence of surface sealants on class I composite restorations and reported no increase in clinical success.9
So, the quest for newer, better sealing materials is on. The ideal requirements of a surface sealant are better flow and good wettability to occupy the interface. If the sealing happens with the natural mineralization process, it can be sustained for a long time.
Recently bioactive materials have gained popularity in the management of incipient caries lesions. A combination of fluoride (F) and casein phosphopeptide-amorphous calcium phosphate fluoride (CPP-ACPF) showed improved release of calcium (Ca), phosphate (P), and F ions with the possibility of enhanced remineralization.10
Hence, the authors conceived the novel use of bioactive varnish for sealing the marginal interface. No studies have used bioactive Ca P-based varnish for marginal sealing. Hence a study was designed to evaluate the effect of the application of a bioactive varnish on the marginal micro gap of class-II composite restorations using scanning electron microscope (SEM) and energy-dispersive X-ray analysis (EDAX).
The null hypothesis tested was that there would be no change in the marginal micro gaps of composite resin restorations sealed with CPP-ACPF varnish compared with a conventional restorative protocol.
MATERIALS AND METHODS
The study was carried out from March 2022 to November 2023 in the Department of Conservative Dentistry and Endodontics at Vishnu Dental College, Bhimavaram, Andhra Pradesh, India. Ethical Committee clearance was obtained from the Institutional Review Board for using natural teeth in the study. (No: IECVDC/2021/PG01/CE/IVT/21). Reporting of this research was prepared according to Clinical Research Information System guidelines.11
Selection Criteria
A total of 50 human maxillary premolars extracted for orthodontic purposes with intact crown structures, without any caries, were selected.
The effect of bioactive varnish application on the marginal micro gap of class II composite resin restorations was assumed to be 80% in the test and 20% in the control groups. Therefore, considering the 95% level of significance and 80% power of the study, the estimated sample size was 24.
On the distal side of each tooth, box-only class II cavities were prepared with similar dimensions.5 Selective etching was done on enamel and dentin, followed by a standard bonding procedure. The cavities were filled with prewarmed composite resin (Tetric N-Ceram–Ivoclar Vivadent products, Delhi) as per the instructions and photoactivated for 20 seconds using the light emitting diode curing source with an intensity of 1000/cm2 (Woodpecker, Guilin, China). After finishing and polishing, the prepared cavities were distributed in two groups. 12 samples were allocated to a control group, and the other 12 samples were allocated to the test group using computer-generated randomization (www.randomizer.org).
In the control group, no sealing agent was applied along the cavo surface margins of samples simulating the conventional restorative protocol.
In the test group, CPP-ACPF varnish (MI varnish, GC America) was coated across the resin tooth interface using a micro brush in a uniform motion as per the manufacturer’s instructions.12
Specimens were subjected to 500 thermocycles5 and were stored in synthetic saliva for 30 days at a simulated oral temperature. Saliva was replenished for every alternate day. All the samples were subjected to SEM (SEM, JEOL; MODEL NO: JSM-IT500LV; JAPAN) evaluation at 200× magnification13 and EDAX. The SEM photographs were decoded by a colleague who was not part of the research for blinding the principal investigator. The marginal gap at the tooth restoration interface at different walls was scored according to Blunck’s criteria14 (Figs 1 and 2). The elemental investigation was done with EDAX to evaluate the minerals deposited.15 The presence of Ca and P was expressed in the EDAX graph (Figs 3 and 4), and the Ca/P ratio was calculated.
Fig. 1: Scanning electron microscope (SEM) image (200×) of the sample in the control group
Fig. 2: Scanning electron microscope (SEM) image (200×) of the sample in the test group
Fig. 3: Image showing Ca and P peaks of a sample in the control group
Fig. 4: Image showing Ca and P peaks of a sample in the test group
Statistical Analysis
Statistical analysis was done using IBM Statistical Package for the Social Sciences (SPSS) Statistics for Windows, Version 20 (IBM Corp., Armonk, N.Y., USA). Intra and intergroup comparison of scores obtained by assessing the marginal micro gap at the tooth-restoration interface using Fisher’s exact test. A comparison of the Ca/P ratio between two independent groups was done using the Mann–Whitney U test. All the samples underwent both analyses.
RESULTS
All 12 samples in the control group showed a gap of ≥2 µm on any one of the three surfaces. 8/12 buccal surfaces, 5/12 base (gingival wall), and 6/12 palatal surfaces showed a gap of ≥2 µm (Fig. 1). The samples in the test group performed well, with the least marginal micro gap score in most of the samples. Four out of 12 samples showed marginal micro gap formation. These samples showed a gap of ≥2 µm on any one of the three surfaces. 3/12 buccal surfaces, 1/12 base (gingival wall), and 0/12 palatal surfaces showed a gap of ≥2 µm (Fig. 2).
Statistical analysis suggested a statistically significant difference between the control and test groups in marginal micro gap formation. The results mentioned above have shown that the buccal (p = 0. 020 > 0.05), gingival wall (p = 0.050), and palatal walls (p = 0. 020 > 0.05) of the Test group showed fewer marginal micro gaps than the control group samples (Table 1).
| Surface | Group | Score 1 | Score 2 | Score 3 | Score 4 | p-value |
|---|---|---|---|---|---|---|
| Buccal | Control | 3 (12.5%) | 1 (4.2%) | 0 | 8 (33.3%) | 0.020* |
| Test | 9 (37.5%) | 0 | 1 (8.3%) | 2 (8.3%) | ||
| Gingival wall | Control | 3 (12.5%) | 4 (16.7%) | 0 | 5 (20.8%) | 0.050* |
| Test | 9 (37.5%) | 2 (8.3%) | 0 | 1 (4.2%) | ||
| Palatal | Control | 5 (20.8%) | 1 (4.2%) | 0 | 6 (25%) | 0.020* |
| Test | 10 (41.7%) | 2 (8.3%) | 0 | 0 |
*, statistical significance
Intergroup comparisons of Ca/P ratios between control and test groups were made using the Mann–Whitney U test. On the palatal wall, the mean Ca/P ratio of the control group was 1.47, the mean Ca/P ratio of the test group was 1.66 group, and the p-value was 0.514. On the gingival wall, the mean Ca/P ratio of the control group was 1.40, the mean Ca/P ratio of the test group was 1.56 group, and the p-value was 0.089. On the buccal wall, the mean Ca/P ratio of the control group was 1.40, the mean Ca/P ratio of the test group was 1.54 group, and the p-value was 0.061. Although the mean Ca/P ratios of the test group were relatively more than the control group, there was no significant difference observed in terms of mineral deposition (Table 2).
| Surface | Group | Mean | Standard deviation | Mann–Whitney U value | p-value |
|---|---|---|---|---|---|
| Palatal | Control | 1.4775 | 0.45324 | 60.000 | 0.514 |
| Test | 1.6642 | 0.32020 | |||
| Gingival wall | Control | 1.4075 | 0.23978 | 42.000 | 0.089 |
| Test | 1.5642 | 0.22909 | |||
| Buccal | Control | 1.4067 | 0.19865 | 36.500 | 0.061 |
| Test | 1.5450 | 0.08723 |
DISCUSSION
The peripheral seal is vital for the success of direct composite resin restorations. Despite unceasing evolution in material science and clinical techniques, the presence of marginal micro gap still persists.5 Many randomized clinical trials have reported better success with composite restorations but with the persisting issue of marginal integrity.7 So, long-term sealing of composite resin-tooth interface has gained significant importance.
Many surface sealants like Fortify, Fortify Plus, Biscover, Optiguard, Protect It, PermaSeal, and Adper single bond were tested to address the marginal micro gap associated with composite restorations, but the real perfect seal was not achieved.8 Hence a novel method of application of bioactive varnish was evaluated for its effect on the marginal micro gap of class II composite restorations. 100% of samples in the control group without any surface sealant demonstrated marginal micro gaps of scores 3 and 4. The reasons attributed could be inherent polymerization shrinkage of the composite material, and polymerization stresses compromising the adhesive interface.16 Prewarming of composite resins is developed as one of the methods to manage polymerization shrinkage. An in vitro study reported that room-temperature bulk-fill composite restorations exhibited an average gap of 11.47 µm from the prepared tooth surface.4 In the present study, all samples were restored with prewarmed bulk-fill composite resin (Tetric N-Ceram–Ivoclar Vivadent products, Delhi). Blalock et al. reported a 23–27% reduction of film thickness with nanohybrid composites by preheating.17 So, the adaptation of resin to the cavity walls might have improved.
An average gap of 5.25 µm was observed in the present research. Darabi et al. reported 5.2–5.5 µm marginal micro gaps in dentin in the investigation of marginal adaptation of prewarmed composites.5 This result is in accordance with the present study.
In the present study, test group cavosurface margins were sealed with CPP-ACPF varnish. The results obtained are promising as fewer samples demonstrated a very minimal marginal micro gap. Around 66.6% of samples demonstrated a score of 1, but 33.3% showed a score of 4 on only one wall.
In general, for prevention of loss and regain of minerals, dental varnish containing F was recommended for the management of incipient carious lesions. However, the limited bioavailability of Ca and P ions in compromised situations like xerostomia resulted in less efficacy of F varnish in preventing demineralization. This led to the rationale for incorporating F in Ca and P formulations like CPP-ACPF. CPP-ACPF is designed with the patented Recaldent (CPP-ACP) developed from milk protein casein and F. Basically, CPP-ACPF is indicated for the prevention of hypersensitivity after scaling, occlusal wear, hypo mineralized areas, and cervical areas. CPP-ACPF sets in contact with water or saliva and should remain undisturbed for 4 hours. Hence the patient should be instructed to avoid rinsing, eating, and any oral hygiene practices for 4 hours. The application of CPP-ACPF forms a tenacious coat on the tooth surface with good wear resistance for a few days.
The composition of CPP-ACPF includes sodium fluoride and CPP-ACP as active ingredients, along with polyvinyl acetate, hydrogenated resin, ethanol, and silicon dioxide as inactive ingredients. Polyvinyl acetate is a water-soluble polymer; hence, it dissolves quickly, releasing the highest F ions in a shorter time. Polyvinyl acetate has high polarity and affinity to carbon dioxide making it highly soluble. Ethanol is another ingredient of CPP-ACPF. This easily evaporates, creating pores for water transport; hence the varnish takes up the water to swell, dissolve, and release ions. As the hydrodynamic radius of Ca, P, and F nano complexes is 2.12 ± 0.26 nm, they can easily penetrate the pores created to reach the deeper surfaces. This might be the reason for fewer marginal micro gaps in the test group.10
Stabilization of F, Ca, and P ions is potentially aided by the presence of peptide σS1-CN (59–79 sequence) in the CPP-ACPF structure. With initial high F release and bioavailability of Ca, P, and F, hydroxyapatite crystals are formed into the dentinal tubules in the affected area.18
Similarly, in the present research, mineral deposition across the polymerization gap mimicking the natural mineralization might have resulted in a minimal gap (average of 0.71 µm). Literature supports that the presence of the casein phosphopeptides stabilizes the amorphous Ca P ions to deliver bioavailable Ca, P, and F ions to enamel and dentin to promote remineralization.19 Cai et al. reported that CPP-ACP complexes adhere to the tooth surface and form a supersaturation state. Thus, they display good wettability and better contact with the tooth surface.20
Cochrane et al., in their study, concluded that CPP-ACPF showed an almost constant F release for 6 hours while the remaining F varnishes, including PreviDent varnish, exhibited very low F release overall.21 Subsequently, CPP-ACPF demonstrated higher F uptake. Tuloglu et al. showed that CPP-ACP produces more acid-resistant primary enamel in the pH cycling model.22 In the test group samples, the marginal gap would have been formed as in the control group. However, the application of CPP-ACPF might have resulted in the mineralization of the gap. This finding is supported by research by Shen et al. on the mineral deposition of laboratory-induced white spot lesions by CPP-ACPF promoted remineralization by 41%. This was attributed to the higher level of F, Ca, and P ion release helping in the sub-surface lesion remineralization.18 Reynolds et al. reported that CPP was found to bind 21 calcium and 14 phosphorus ions per molecule.23
In the current research, the control group showed a marginal micro gap of 5.25 µm, whereas the CPP-ACPF group showed a minimal marginal micro gap of 0.7 µm. The diffusion of CPP-ACP nanocomplexes and natural remineralization could be attributed to the reduced marginal micro gap.
Wierichs et al. demonstrated that CPP-ACP protected dentin from mineral loss. His in vitro study with the pH cycling model proved CPP-ACPF to be superior to other agents tested in preventing dentin demineralization. The bio-availability of Ca, F, and P ions from CPP-ACPF varnish helps in F uptake with an improved pattern of remineralization along with the remineralization of the entire body of the lesion.24 Application of only F did not result in optimum remineralization on enamel lesions compared to a mixture of F and CPP-ACP.25
Energy-dispersive X-ray analysis (EDAX) evaluated the mineral content in the current research. The mean Ca/P ratios of the test group were relatively higher than the control group at the resin-tooth interphase (Table 2). This is supported by the research wherein application of CPP-ACPF varnish resulted in 80 vol% mineral gain and also resulted in high F release and Ca and P ions.18 Cai et al. reported 8.52% mineral gain upon application of CPP-ACPF on root caries lesions using micro-computed tomography analysis.20 In the current search, the mean Ca/P ratios of the CPP-ACPF varnish group were higher than the control group, without statistically significant differences in the mineral deposition.
The null hypothesis was rejected as there was a statistically significant difference between the control and test groups, as the samples in the test group showed better performance. 8/12, that is, 70% of the samples showed no marginal micro gap on SEM imaging after the cavosurface sealing.
LIMITATIONS
In the future, X-ray diffraction analysis can be recommended to understand the nature of the deposition at the tooth and resin restoration interface. As it was an in vitro study, all the oral environmental factors like pH and occlusal loading were not stimulated in the study. Clinical success is only discovered after prolonged clinical use. So clinical studies are needed to assess the success and longevity of restoration with cavosurface sealing with CPP-ACPF varnish.
CONCLUSION
The test group with CPP-ACPF application demonstrated a statistically significant decrease in marginal micro gap compared to the control group. Only 30% of the samples in the test group with MI Varnish cavosurface sealing showed a marginal gap on SEM imaging with an average of 0.71 µm. 100% of the samples in the control group with no cavosurface sealing showed a marginal gap on SEM imaging with an average of 5.25 µm. EDAX results of the test and control groups did not demonstrate a statistically significant difference.
REFERENCES
1. Ferracane JL. Resin composite–state of the art. Dent Mater 2011;27(1):29–38. DOI: 10.1016/j.dental.2010.10.020
2. Guo J, Holmes B, Yang B, et al. Determining the temporal development of dentin-composite bond strength during curing. Dent Mater 2016;32(8):1007–1018. DOI: 10.1016/j.dental.2016.05.009
3. Malhotra N, Kundabala M, Shashirashmi A. Strategies to overcome polymerization shrinkage–materials and techniques. A review. Dent Update 2010;37(2):115–115. DOI: 10.12968/denu.2010.37.2.115
4. Correia A, Andrade MR, Tribst J, et al. Influence of bulk-fill restoration on polymerization shrinkage stress and marginal gap formation in class V restorations. Oper Dent 2020;45(4):E207–E216. DOI:10.2341/19-062-L
5. Darabi F, Tayefeh-Davalloo R, Tavangar SM, et al. The effect of composite resin preheating on marginal adaptation of class II restorations. J Clin Exp Dent 2020;12(7):e682–e687. DOI: 10.4317/jced.56625
6. Torres CRG, Mailart MC, Crastechini É, et al. A randomized clinical trial of class II composite restorations using direct and semidirect techniques. Clin Oral Investig 2020;24(2):1053–1063. DOI: 10.1007/s00784-019-02999-6
7. Sabbagh J, El Masri L, Fahd JC, et al. A three-year randomized clinical trial evaluating direct posterior composite restorations placed with three self-etch adhesives. Biomater Investig Dent 2021;8(1):92–103. DOI: 10.1080/26415275.2021.1939034
8. D’Alpino PH, Pereira JC, Rueggeberg FA, et al. Efficacy of composite surface sealers in sealing cavosurface marginal gaps. J Dent 2006;34(3):252. DOI: 10.1016/j.jdent.2005.06.010
9. Nahsan F, Wang L, Modena K, et al. A 12-month clinical trial examining the effects of a surface sealant on Class I composite resin restorations. Gen Dent 2016;64(2):18–20.
10. Sleibi A, Tappuni AR, Karpukhina NG, et al. A comparative evaluation of ion release characteristics of three different dental varnishes containing fluoride either with CPP-ACP or bioactive glass. Dent Mater 2019;35(12):1695–1705. DOI: 10.1016/j.dental.2019.08.113
11. Krithikadatta J, Gopikrishna V, Datta M. CRIS Guidelines (checklist for reporting in-vitro studies): a concept note on the need for standardized guidelines for improving quality and transparency in reporting in-vitro studies in experimental dental research. J Conserv Dent 2014;17(4):301–304. DOI: 10.4103/0972-0707.136338
12. https://r.search.yahoo.com/_ylt=AwrgxRDGM1JkcjsLG9ZXNyoA;_ylu=Y29sbwNncTEEcG9zAzMEdnRpZAMEc2VjA3Ny/RV=2/RE=1683137607/RO=10/RU=https%3a%2f%2fwww.gc.dental%2famerica%2fproducts%2foperatory%2fpreventive%2fmi-varnish/RK=2/RS=E3UNjJHoyzTeJ7XclRabrieFKbE.
13. Mazumdar P, Das UK, Kundu R. SEM evaluation of gap at the resin dentin interface in Class II composite resin restoration: an in vitro study. Health Renaissance. 2012 Jul 28;10(2):98–104. DOI: 10.3126/HREN.V10I2.6571
14. Sabatini C, Blunck U, Denehy G, et al. Effect of pre-heated composites and flowable liners on Class II gingival margin gap formation. Oper Dent 2010;35(6):663–671. DOI: 10.2341/10-094-L
15. Hegde MN, Moany A. Remineralization of enamel subsurface lesions with casein phosphopeptide-amorphous calcium phosphate: a quantitative energy dispersive X-ray analysis using scanning electron microscopy: An in vitro study. J Conserv Dent 2012;15(1):61–67. DOI:10.4103/0972-0707.92609
16. Didron PP, Chrzanowski W, Ellakwa A. Effect of temperatures on polymerization stress and microleakage of class V composite restorations. Open J Compos Mater 2013;3(4):107. DOI: 10.4236/OJCM.2013.34011
17. Blalock JS, Holmes RG, Rueggeberg FA. Effect of temperature on unpolymerized composite resin film thickness. J Prosthet Dent 2006;96(6):424–432. DOI: 10.1016/j.prosdent.2006.09.022
18. Shen P, McKeever A, Walker GD, et al. Remineralization and fluoride uptake of white spot lesions under dental varnishes. Aust Dent J 2020;65(4):278–285. DOI: 10.1111/adj.12787
19. Shen P, Manton DJ, Cochrane NJ, et al. Effect of added calcium phosphate on enamel remineralization by fluoride in a randomized controlled in situ trial. J Dent 2011;39(7):518–525. DOI: 10.1016/j.jdent.2011.05.002
20. Cai J, Burrow MF, Manton DJ, et al. Remineralising effects of fluoride varnishes containing calcium phosphate on artificial root caries lesions with adjunctive application of proanthocyanidin. Dent Mater 2021;37(1):143–157. DOI: 10.1016/j.dental.2020.10.021
21. Lippert F, Hara AT, Martinez-Mier EA, et al. Laboratory investigations into the potential anticaries efficacy of fluoride varnishes. Pediatr Dent. 2014;36(4):291–295. PMID: 25197993.
22. Tuloglu N, Bayrak S, Tunc ES, et al. Effect of fluoride varnish with added casein phosphopeptide-amorphous calcium phosphate on the acid resistance of the primary enamel. BMC Oral Health 2016;16(1):103. DOI: 10.1186/s12903-016-0299-4
23. Cochrane NJ, Cai F, Huq NL, et al. New approaches to enhanced remineralization of tooth enamel. J Dent Res 2010;89(11):1187–1197. DOI: 10.1177/0022034510376046
24. Wierichs RJ, Stausberg S, Lausch J, et al. Caries-preventive effect of NaF, NaF plus TCP, NaF plus CPP-ACP, and SDF varnishes on sound dentin and artificial dentin caries in vitro. Caries Res 2018;52(3):199–211. DOI: 10.1159/000484483
25. Tao S, Zhu Y, Yuan H, et al. Efficacy of fluorides and CPP-ACP vs fluorides monotherapy on early caries lesions: a systematic review and meta-analysis. PLoS One 2018;13(4):e0196660. DOI: 10.1371/journal.pone.0196660
________________________
© The Author(s). 2023 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (https://creativecommons.org/licenses/by-nc/4.0/), which permits unrestricted use, distribution, and non-commercial reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.