Evaluation of the Inactivation of Microorganisms by a Blue Laser (445 nm)—An In Vitro Study

Abstract

Background: Blue laser light has been the subject of research regarding the inactivation of microorganisms as a possible alternative to chemical treatment methods for a number of years. In dentistry, blue light could be used, for example, in the treatment of periodontitis/peri-implantitis, as well as in endodontics and against caries. It could serve as an alternative or supplement to traditional chemical and/or invasive methods. The antimicrobial effectiveness of a blue laser in relation to the speed of treatment is investigated using three different microbial test organisms in order to identify possible species differences. Methods: The test organisms Enterococcus faecalis, Streptococcus mutans, and Candida albicans were applied to smooth zirconium discs and treated twice with a diode laser at 445 nm wavelength with a traversing speed of 1, 2, and 4 mm/s. The antimicrobial effect was analysed based on the resulting colony-forming units on agar plates. The temperature was measured during the treatment. Preliminary tests were carried out using the MTT dye test to determine relevant setting parameters and the required energy dose. Results: Statistically significant differences were found between the negative control and the treated samples for all three tested organisms, with a maximum viability reduction of 1.8 log₁₀ CFU/mL for Enterococcus faecalis, 2.5 log₁₀ CFU/mL for Streptococcus mutans, and 1.0 log₁₀ CFU/mL for Candida albicans at 1 mm/s traversing speed, regarding estimated marginal means (p < 0.001). The temperature on the substrate surface reached 30 to 42 °C for all samples evaluated. Conclusions: Blue laser light (445 nm) demonstrates antimicrobial activity, which increases with prolonged exposure. Further research is needed to assess all key influencing parameters and define possible clinical applications.

1. Introduction

Investigations into the use of lasers in dentistry began shortly after the development of the ruby laser in the 1960s. The first applications of a CO₂ laser on patients were reported in 1968 [1]. Commercially available lasers for dental applications typically operate at 445 nm, 810 nm, 970 nm, or 2940 nm and have proven themselves in conservative dentistry, endodontology, periodontology, implantology, and oral surgery [2,3]. However, few studies have investigated microbial reduction using blue lasers in endodontic, periodontic, and caries treatments [4].

Blue diode lasers have their primary indication in invasive surgery, as they are well able to penetrate chemical structures such as porphyrin, e.g., in haemoglobin and enzymes with prosthetic groups such as the flavin group [5]. Promising research is also being conducted in other areas, such as supporting teeth whitening or removing ceramic braces [6,7]. Additionally, blue light is known for its antimicrobial effect [8]. It is thought to interfere with microbial enzymes that contain, for example, the flavin group that absorbs light in the 380–500 nm wavelength range [9], which can lead to structural changes, such as the release of reactive oxygen species such as singlet oxygen from the structures, which in turn can damage other cell structures such as DNA, cell compartments, cell membranes, or cell walls [10,11]. This means that no extracellular photosensitisers are required, in contrast to photodynamic therapy [12], and the effect is conclusively directed at the microorganisms. Another advantage of blue light is its minimal absorption by water [13], resulting in little to no heating of water-containing compartments [14]. This implies that it is theoretically possible to selectively “heat” microbial intracellular components through light absorption by specific chemical compounds without concomitantly heating the surrounding aqueous environment through which the light must pass. This selectivity is not observed with lasers operating at longer wavelengths, such as those above 900 nm, which are strongly absorbed by water. However, various implant and dental materials are capable of absorbing blue light, potentially leading to the heating of the material’s surface [15,16]. Temperature control is therefore advisable in order to avoid damaging the surrounding tissue during clinical use [16]. Based on the reasons mentioned above, namely that blue light can act directly on the target without the addition of chemicals by generating reactive oxidative species and heat and that it is not absorbed by surrounding water, intensive research is currently being conducted on the effects of blue light on microorganisms in general and on multidrug-resistant organisms in particular [8,17].

In dental clinical practice, the antimicrobial efficacy of blue light may be used as an adjunctive treatment for endodontic infections [18], periodontitis [19], peri-implantitis (currently only in combination with photoactivatable compounds) [20], and dental caries [21].

Consequently, the aim of this study is to evaluate the antimicrobial effects of 445 nm blue laser light and to determine species-specific treatment parameters for Enterococcus faecalis, Streptococcus mutans, and Candida albicans.

2. Results

2.1. Test Group Comparisons

The descriptive results of the antimicrobial tests on the test organisms, Enterococcus faecalis, Streptococcus mutans, and Candida albicans, are listed in Table 1 and presented graphically as a box plot in Figure 1.

Table 1.

Descriptive statistics of colony-forming units in log₁₀ CFU/mL of the microbial cell concentration of the initial suspension, the traversing speed test groups, and control groups for the test organisms Enterococcus faecalis, Streptococcus mutans, and Candida albicans; mean (M), standard deviation (SD), median (Med), the quantile (Q1, 1st quantile; Q3, 3rd Quantile), and the reduction factor (RF) regarding negative control are given. Data source [22].

Test Organisms	Enterococcus faecalis							Streptococcus mutans							Candida albicans
	n	M	SD	Med	Q1	Q3	RF	n	M	SD	Med	Q1	Q3	RF	n	M	SD	Med	Q1	Q3	RF
Initial suspension	14	7.9	0.3	7.9	7.6	8.2		14	8.2	0.2	8.3	8.0	8.4		14	7.3	0.2	7.2	7.0	7.5
Test groups
4 mm/s (250 J/cm2)	4	5.7	0.1	5.7	5.6	5.8	− 4.9	14	4.5	0.8	4.3	4.1	4.8	−4.1	13	5.3	0.7	5.1	4.9	5.3	−4.1
2 mm/s (500 J/cm2)	13	5.8	0.4	5.7	5.5	6.1	−5.3	14	4.5	0.5	4.3	4.2	4.5	−3.7	14	5.4	0.5	5.2	5.0	5.8	−4.6
1 mm/s (1000 J/cm2)	14	5.5	0.5	5.3	5.1	5.8	−5.0	14	4.0	0.6	3.9	3.5	4.4	−3.7	4	4.4	0.1	4.4	4.3	4.4	−3.2
Control groups
Negative control	14	7.3	0.3	7.3	7.2	7.5		14	6.5	0.5	6.4	6.0	6.9		14	5.8	0.6	5.4	5.3	6.2
Positive control	14	3.7	0.9	3.8	3.1	4.3	−3.4	14	0.9	0.8	0.9	0.0	1.7	−1.0	14	1.5	1.2	2.0	0.0	2.4	−1.2

Figure 1.

A box plot of the log₁₀-transformed colony-forming unit (CFU) values of Enterococcus faecalis, Streptococcus mutans, and Candida albicans in untreated samples (negative controls), laser-treated samples (test groups 4 mm/s, 2 mm/s, and 1 mm/s), and positive controls treated with chlorhexidine digluconate (CHX). Data source [22].

The statistical model described in the Section 4 (Material and Methods) was used for the group comparisons in order to account for the different group sizes. The estimated marginal means determined for the other group comparisons are shown in Table 2.

Table 2.

Estimated marginal mean (EMM) of the test organisms Enterococcus faecalis, Streptococcus mutans, and Candida albicans in log₁₀ CFU/mL; standard error (SE) and the 95% confidence interval (CI) are given. Data source [22].

Test Organisms	Enterococcus faecalis			Streptococcus mutans			Candida albicans
	EMM	SE	95% CI	EMM	SE	95% CI	EMM	SE	95% CI
Test groups
4 mm/s (250 J/cm2)	6.02	0.18	5.56–6.49	4.49	0.21	3.95–5.04	5.35	0.23	4.75–5.96
2 mm/s (500 J/cm2)	5.84	0.15	5.45–6.24	4.47	0.21	3.92–5.01	5.40	0.23	4.80–6.00
1 mm/s (1000 J/cm2)	5.48	0.15	5.09–5.87	3.95	0.21	3.41–4.50	4.75	0.25	4.10–5.40
Negative control	7.26	0.15	6.87–7.65	6.49	0.21	5.95–7.04	5.77	0.23	5.17–6.37

The results of the CFU, the estimated values, and the superiority test compared to CHX were carried out individually for each test organism.

For Enterococcus faecalis, the untreated control samples (negative controls) showed mean values of 7.3 log₁₀ CFU/mL. Mean values of 5.7 log₁₀ CFU/mL for 4 mm/s, 5.8 log₁₀ CFU/mL for 2 mm/s, and 5.5 log₁₀ CFU/mL for 1 mm/s were observed for the traversing speed groups (Table 1). One sample from the Enterococcus faecalis test series, which was assigned to the 2 mm/s group, could not be analysed for laboratory reasons and was, therefore, excluded from the overall evaluation, resulting in a treatment number of n = 13 instead of n = 14 [22].

The estimated marginal mean (Table 2) changed compared to the mean values (Table 1) in the groups 4 mm/s (n = 4) and 2 mm/s (n = 13), as these were less filled than the remaining groups, and the information from all groups was used to determine the estimated marginal means and their standard errors. Only in the 4 mm/s test group did the estimated marginal mean value increase by 0.3 to 6.02 log₁₀ CFU/mL, in relation to the observed mean value of 5.7 log₁₀ CFU/mL; the other values remained comparable (Table 1 and Table 2) [22].

The positive controls, which were treated with CHX (2%), had a mean value of 3.7 log₁₀ CFU/mL (Table 1) and the largest standard deviation among the analysed Enterococcus faecalis samples with a value of 0.9 log₁₀ CFU/mL. Compared to the group to be tested (1 mm/s), there was a statistically significant difference (p < 0.001). The non-inferiority of the 1 mm/s test group compared to CHX could therefore not be demonstrated [22].

For Streptococcus mutans, a mean value of 6.5 log₁₀ CFU/mL was shown for the negative controls. In the treated samples, the 4 mm/s and 2 mm/s groups had the highest mean value of 4.5 log₁₀ CFU/mL. This was followed by mean values of 4.0 log₁₀ CFU/mL in the 1 mm/s group (Table 1) [22].

The estimated marginal means of Streptococcus mutans corresponded to the observed means and were, therefore, not adjusted for the estimates, as the sample size is the same for all groups (Table 2).

The main effect of the positive controls was 0.9 log₁₀ CFU/mL and is significantly smaller (p < 0.001) than the group to be tested (1 mm/s) at 4.0 log₁₀ CFU/mL, which is why non-inferiority could be ruled out (Table 1) [22].

For Candida albicans, the samples treated at three different speeds had mean values of 5.3 log₁₀ CFU/mL at 4 mm/s, 5.4 log₁₀ CFU/mL at 2 mm/s, and 4.4 log₁₀ CFU/mL at 1 mm/s (Table 1). There were statistically significant differences (p < 0.001 in each case) between the untreated (negative controls) and treated samples of group 1 mm/s and the positive control (Table 3).

Table 3.

Group comparisons of the traversing velocities and negative control (NC) for the test organisms Candida albicans, Enterococcus faecalis, and Streptococcus mutans; contrast values (CV) are stated in log₁₀ (CFU/mL), 95% confidence interval (CI), and resulting p-values (Bonferroni corrected, α = 0.01) are given. Data source [22].

Test Organisms	Enterococcus faecalis			Streptococcus mutans			Candida albicans
Comparisons	CV	95% CI	p Value	CV	95% CI	p Value	CV	95% CI	p Value
NC vs. 4 mm/s	1.24	0.86–1.62	<0.001	2.00	1.50–2.50	<0.001	0.42	0.18–0.65	<0.001
NC vs. 2 mm/s	1.41	1.18–1.65	<0.001	2.02	1.53–2.52	<0.001	0.37	0.15–0.60	<0.001
NC vs. 1 mm/s	1.78	1.55–2.01	<0.001	2.54	2.04–3.04	<0.001	1.02	0.65–1.39	<0.001
4 mm/s vs. 2 mm/s	0.18	−0.20–0.56	0.861	0.020	−0.47–0.52	1.000	−0.04	−0.28–0.19	1.000
1 mm/s vs. 4 mm/s	−0.54	−0.92–−0.16	<0.001	−0.54	−1.04–−0.04	0.004	−0.6	−0.97–−0.23	<0.001
1 mm/s vs. 2 mm/s	−0.36	−0.60–−0.13	<0.001	−0.52	−1.01–−0.02	0.007	−0.65	−1.02–−0.28	<0.001

The estimated marginal means (Table 2) were close to the observed mean values (Table 1). Only the estimated marginal mean value at 1 mm/s increased slightly by approximately 0.4 to 4.75 log₁₀ CFU/mL in relation to the observed mean value (Table 1 and Table 2) [22].

Since the main effect of the positive controls (CHX), the active control group, was 1.5 log₁₀ CFU/mL (Table 1), the tolerance limit of 10%, at which the test group to be compared (1 mm/s) can be assigned to non-inferiority, is therefore 1.65 log₁₀ CFU/mL. The comparison of the test group with the active control group showed a clear and statistically significant difference (p = 0.0006) for the mixed model restricted to the test group with 1 mm/s and the active control group [22]. Accordingly, there was no non-inferiority.

Additionally, a comparison of the test groups for all three test organisms by the mixed model are summarised in Table 3. The significance value (p) was <0.001 for the global test for differences between the groups. For all test organisms, all three laser treatment groups were able to achieve a statistically significant vital bacterial reduction compared to the untreated control samples, and the highest antimicrobial effect of the laser treatment was observed at a laser speed of 1 mm/s. The highest contrast value of the treated samples was of 2.54 log₁₀ CFU/mL for Streptococcus mutans. Within the velocity groups, the results of all three test organisms correlate: there is a statistically significant difference between the parameters 1 mm/s and 2 mm/s, as well as 1 mm/s and 4 mm/s, but not between 2 mm/s and 4 mm/s (Table 3) [22].

2.2. Temperature Measurements

The temperatures measured at all settings were between 30.5 °C and 34.2 °C for Enterococcus faecalis, 31.8 °C and 34.9 °C for Streptococcus mutans, and 35.8 °C and 39.8 °C for Candida albicans (Table 4). One sample of Candida albicans showed a temperature of 80 °C during laser treatment at a speed of 4 mm/s and was excluded from the overall evaluation in accordance with the exclusion criteria described in the Section 4 (Material and Methods) [22].

Table 4.

Temperature during laser treatment of Enterococcus faecalis, Streptococcus mutans, and Candida albicans in degrees Celsius. Value 1 corresponds to the measured area directly next to the laser beam and value 2 to the area below the laser tip. Data source [22].

Test Organisms	Traverse	Temperature Value 1 [°C]				Temperature Value 2 [°C]
	Speed	Mean	SD	Min.	Max.	Mean	SD	Min.	Max.
Enterococcus faecalis	4 mm/s	30.5	0.5	29.7	31.7	33.3	1.0	32.0	34.8
	2 mm/s	30.7	0.6	29.3	32.0	34.2	2.8	29.5	39.9
	1 mm/s	30.6	1.0	28.3	32.4	33.6	1.5	29.7	37.1
Streptococcus mutans	4 mm/s	31.8	0.7	30.3	33.4	33.7	1.9	30.1	38.2
	2 mm/s	32.1	0.9	30.5	33.7	33.9	2.0	31.0	37.9
	1 mm/s	32.5	1.3	30.4	35.1	34.9	2.6	31.0	39.8
Candida albicans	4 mm/s	37.5	1.9	34.9	41.7	38.6	2.2	34.8	41.9
	2 mm/s	37.4	1.6	35.0	41.4	39.8	1.7	36.7	41.8
	1 mm/s	35.8	0.7	34.2	36.6	38.5	1.9	34.7	41.9

3. Discussion

In this study, the effect of blue laser light on Enterococcus faecalis, Streptococcus mutans, and Candida albicans on ceramic discs at different traversing speeds was investigated.

Ceramic was chosen as the test substrate, as it is a material frequently used in dentistry [23,24] and in antimicrobial testing for dental applications [25]. Moreover, it has been shown to be less susceptible to problems associated with excessive surface heating during blue laser treatment [15]. The low absorption of blue light and the smooth surface of the ceramic appear to be the appropriate choice for initial studies aimed at determining the antimicrobial effect of blue light alone. Therefore, the influence of the substrate should be minimised.

The test organisms used in this study play an important role in dentistry. Enterococcus faecalis is often involved in persistent root canal infections and can contribute to peri-implantitis [26]. Streptococcus mutans may indirectly contribute to periodontal disease by promoting supragingival biofilm accumulation through glucan-rich extracellular polysaccharide synthesis, thereby sustaining gingival inflammation and increased gingival crevicular fluid flow [27,28]. This inflammation-driven nutrient shift favours periodontal dysbiosis. Additionally, Streptococcus mutans can bind via the SpaP–RadD adhesin–receptor interaction to Fusobacterium nucleatum, a key bridging organism, potentially facilitating the incorporation of periodontitis-associated taxa such as Porphyromonas gingivalis, Tannerella forsythia, Treponema denticola, and Aggregatibacter actinomycetemcomitans [29]. Candida albicans is a commensal organism of the oral cavity with pathogenic potential. According to the literature, it can cause opportunistic infections under conditions of oral dysbiosis, especially when the commensal microbiota are reduced [30,31]. Further, it is often associated with mucositis and can interact with different pathogenic bacteria such as Streptococcus mutans through physicochemical and biochemical interactions [32].

The distance between the laser tip and the sample surface of 1 mm and the traversing speed of 2 mm/s selected on the basis of the preliminary investigations (Appendix A, Table A1) were also used in other laser studies [18,33]. The time component is interesting, because the treatment times with a common dental laser are longer and more time-consuming than with conventional methods [34]. Therefore, we investigated traversing speeds of 1 to 4 mm/s in order to determine the influence of different treatment speeds. Based on the results of the preliminary tests and in order to minimise the use of resources, only two travel speeds were initially estimated for Enterococcus faecalis and Candida albicans (with n = 14), and the third travel speed was added in the course of the study (here, n = 4). The small sample size of n = 4 is justified by the theory and practice of sequential trials, which are now widely accepted [35]. Moreover, mixed models were used, which allow information to be borrowed across groups [36]. This reduces the necessary resources in terms of materials, time, and costs without compromising the significance of the results.

Additionally, the mixed models are advantageous with regard to potential bias [36]. To demonstrate this advantage in the bias–variance trade-off, a comparison was made with conventional linear regression for Candida albicans (Appendix B, Table A3). All three microorganisms were analysed in a single mixed model to better estimate the possible effect of bias due to low case numbers and to better incorporate the available information (Appendix B, Table A4).

A number of challenges were identified during the pre-studies. One of the challenges was to determine the relevant parameters to achieve a reliable antimicrobial effect, fixed to an energy dose value. This proved to be difficult, as changing one parameter changes the calculated energy dose per area, which had to be compensated for by changing another parameter in order to work with comparable energy dose values. Finally, individual application parameters were determined for all three test organisms (Appendix A, Table A1). Another difficulty was the impossibility of using a uniform carrier solution for the suspensions and the treatment procedure of the three selected test organisms. Either crystals were formed, the cells were washed off the samples, or the viability was reduced (Appendix A, Figure A1). In order to avoid errors by the preparation of samples, different carrier solutions were used for the test organisms as described in the Section 4 (Material and Methods).

The antiseptic CHX shows antimicrobial effects in concentrations between 0.1 and 2%, against Gram-positive germs and fungi (e.g., streptococci, enterococci, Candida albicans) [37,38,39]. Preliminary studies showed that the frequently used concentration of 0.2% CHX for 10 min resulted in only a lower reduction in viable cells than 1 min with 2% CHX. The differences were 1.4 log₁₀ CFU/mL (Enterococcus faecalis), 0.9 log₁₀ (Streptococcus mutans), and 1.1 log₁₀ (Candida albicans) (Appendix A, Table A2). A “non-inferiority” of the test group 1 mm/s would then not have been detectable in all cases. Therefore, the concentration of 2% CHX for 60 s treatment time was chosen in this study, as a significant antimicrobial effect was always achieved for all test groups and thus a stable positive control.

In all test organisms, a statistically significant reduction in viable cells was always achieved with the laser compared to the untreated control samples. Differences between the test groups were only significant for the most antimicrobial effective group with a traversing speed of 1 mm/s, but not between the two groups of 2 and 4 mm/s (Table 3), which is also reflected in the results of the preliminary test (Appendix A, Table A1). The contrast values and the reduction factors of the CFU values show that Candida albicans is the least sensitive and Streptococcus mutans the most sensitive to the blue laser light used (Table 3). An influence of the temperature development during the treatments can be ruled out, as the temperatures were checked as described. Here, the temperature of the ceramic surface rose to an average of 35 °C for Enterococcus faecalis and Streptococcus mutans and to 40 °C for Candida albicans. This observation suggests that cellular structures or surrounding extracellular components may influence the temperature development. As reported in the results, the temperature in one out of fourteen samples of Candida albicans rose to 80 °C. The cause of this observation remains unclear. Contamination of the affected sample is considered the most likely explanation, as it showed markedly increased absorption at 445 nm. In contrast, intra-species variability in the cell structure of Candida albicans appears unlikely, since the phenomenon was observed only in a single instance. Therefore, this outlier was excluded from the analysis.

In various studies, there are two wavelength ranges of visible blue light that have been investigated solely as antimicrobial active components, 400–420 nm and 450–470 nm. It is generally assumed that the mechanism of the antimicrobial effect of blue light is the stimulation of endogenous intracellular metal-free porphyrins (especially at 405 nm) and flavins (especially at 445 nm) behaving as a photosensitising dye, generating singlet oxygen, which leads to further reactive oxygen species that react with cell components containing lipids, nucleic acids, and proteins that lead to oxidative stress and cell death [9,11,17,40,41,42].

Based on the literature, this effect appears mostly to be higher at around 410 nm than at around 450 nm [40,41,43]. Candida albicans has porphyrins and flavins. However, the antimicrobial effect of blue light is increasingly discussed in Candida albicans, due to the presence of porphyrins [44]. This may be one reason why Candida albicans showed the lowest reduction in this study with a 445 nm laser light (porphyrins are less affected). Enterococcus faecalis and Streptococcus mutans do not have original porphyrin structures (however, Enterococcus faecalis is known to be able to absorb and utilise them [45]), but they are sensitive to blue light at the wavelengths of 405 and 445 nm, indicating more complex mechanisms than a sole reaction by flavine group structures involved in the antimicrobial mechanisms of Enterococcus faecalis and Streptococcus mutans [42,46].

This study is one of the few to have analysed the antimicrobial efficacy of the 445 nm wavelength without an additional photosensitiser on microorganisms on solid surfaces. A comparison with other studies is, therefore, only possible to a very limited extent. Only examinations in dental canals are known, where, for anatomical reasons, no fixed treatment distance is possible. Studies with investigations on suspensions and agar plates with antimicrobial blue light at wavelength at or close to 445 nm show that the reduction values are all in a similar range to those in this study, even though a significantly lower energy dose was used in most cases.

It should be noted here that in some studies an LED was used instead of a laser light spot, as used in our study. The LED light can illuminate the samples broadly, and no “sample traversing” is necessary. Therefore, the information on the energy input to the surface must be interpreted differently. For (the limited) comparability with other studies, the results are expressed as the energy doses required for a 90% reduction in viable cells (1 log₁₀ CFU/mL) based on the contrast values for Enterococcus faecalis at 250 J/cm², for Candida albicans at 1000 J/cm², and for Streptococcus mutans for a 99% reduction (2 log₁₀ CFU/mL) at 250 J/cm².

In the study of Plavskii et al. (2018) using Staphylococcus aureus, Escherichia coli, and Candida albicans as test organisms in suspension, the application of a 445 nm laser light, which was expanded using a 10-fold telescope to irradiate the entire 3 mL volume in a glass tube (p = 50 mW), resulted in varying degrees of microbial reduction [9]. After 15 min of exposure, a reduction in viable cells of 0.1 to 0.3 log₁₀ CFU/mL was observed. A 90% reduction was achieved for Candida albicans after 120 min (366 J/cm²), whereas for Escherichia coli and Staphylococcus aureus, the same reduction required 200 min of irradiation (~606 J/cm²). This indicates that Candida albicans required less energy for inactivation compared to the Gram-positive S Staphylococcus aureus, which needed approximately three times the energy required for Streptococcus mutans and Enterococcus faecalis in this study. Sousa et al. (2015) found no dose-dependent correlation in their study [47]. However, they were able to show that an effective growth inhibition of the test organisms Staphylococcus aureus, Pseudomonas aeruginosa, and Escherichia coli in suspension (300 µL in well plate) was achieved at 6 J/cm² after 86 s of irradiation (laser, 450 nm wavelength, continuous wave at 70 mW power, irradiation area 1 cm²). No stronger effect was observed up to 24 J/cm².

What about studies on microbial biofilms? Microbial biofilms are generally more resistant to environmental influences [48]. Due to the higher cell density and the biofilm matrix, Martegani et al. (2020) were unable to demonstrate a reduction in the viable cells of Pseudomonas in biofilm after treatment with blue LED light at a wavelength of 455 nm, but the biofilm development was inhibited [41]. In another study on 14-day-old Candida albicans biofilms with blue LED light (entire surface illuminated at 455 nm, 75 mW/cm² on surface, distance 2 cm), there were reductions approximately between 0 log₁₀ (after 2 min [9 J/cm²]), 0.35 log₁₀ (after 5 min [22.5 J/cm²]), and a more than 99% reduction of 2.4 log₁₀ CFU/mL after 10 min 45 J/cm² [49]. Compared to our study with 90% reduction at 1000 J/cm², the required energy doses by Rosa et al. (2016) for 99% reduction at 45 J/cm² were very low, indicating that the overall energy dose may depend on more factors than the species tested and its sample preparation [49]. For example, a supplementary study by Bumah et al. (2015) with Staphylococcus aureus strains on agar plates showed that the density of the microorganisms on the surface has an influence on the required energy doses [50]. Here, repeated irradiation (LED, 470 nm) with 55 J/cm² for “low dense” cultures (3 × 10⁶ CFU/mL) or 220 J/cm² for “dense cultures” (7 × 10⁶ CFU/mL) resulted in 100% bacterial suppression [50].

One advantage of blue light is its ability to penetrate translucent materials such as dentine. A clinical study showed that the antimicrobial blue light (445 nm) can penetrate 500 µm thick dentine to reduce viable Streptococcus salivarius up from 443 J/cm², which indicates its application in caries treatment [21].

The ability to penetrate dentine is also interesting for endodontic therapies. An in-vitro study showed that the 445 nm light (diode laser application at a 5° angle, 0.6 W) can penetrate dentine up to 1000 µm to reduce viable Enterococcus faecalis in root canals by 0.8 log₁₀ CFU/mL after 40 s cumulative treatment time (4 intervals) [18] or by 1 log₁₀ CFU/mL (0.6 W, power input 3.7 kW/cm²) [51]. In another clinical study, the use of a 445 nm laser diode (0.8 W, 45° angle application, 60–80 s complete treatment session) also showed a significant benefit for the treatment of chronic periodontitis within an observation period of 6 months [19].

The possibility of resistance to antimicrobial blue light is controversial in the literature. Amin et al. (2016) found no evidence of resistance development in Pseudomonas aeruginosa, and Zhang et al. (2014) reported a similar finding for Acinetobacter baumannii strains after 10 sublethal doses [52,53]. However, Rapacka-Zdonczyk et al. (2021) observed species-dependent tolerance development for Escherichia coli, Klebsiella pneumoniae, and Pseudomonas after 5, 10, and 15 days of exposure to 415 nm blue light, which may be an adaptation to reactive oxygen stress [54].

The following limitations must be noted. No oral pathogenic Gram-negative bacterial strains were tested in this study. This would require anaerobic experimental conditions, which could not be achieved here. For a species-specific effect, further investigations would have to be carried out to break down intracellular mechanisms. It is possible that other wavelengths or broadband light sources in the blue range could overcome species-specific differences. The effectiveness of blue light can vary greatly between pathogenic microorganisms and is, therefore, not necessarily replicable in other test organisms under the same experimental conditions. This study tested the efficacy on microorganisms applied to a surface and not on microorganisms embedded in an adherent biofilm. Typically, microorganisms in the oral environment occur in multispecies biofilms, which are expected to be less sensitive, due to the matrix and layering.

In addition to the aforementioned applications in caries and root canal treatment, blue light could also be used in the treatment of peri-implantitis. However, the measured antimicrobial effects are currently low, meaning that blue light therapy can only be used as a supplement to other treatment methods at present. The temperature development should always be monitored for each specific substrate to prevent overheating. For example, when bleaching teeth with a 445 nm laser with higher laser power, the deeper tooth pulp can also be heated [16]. The data on efficacy cannot yet be directly transferred to practice, as too many environmental factors influence efficacy. Further research is necessary. Additionally, antimicrobial blue light can have toxic effects on eukaryotic cells, which decrease with increasing wavelengths [55]. The relevance of the issue of blue light cytotoxicity depends on the treatment goal. Intentional or unintentional tissue treatment or even scattered light can reach human cells. Studies on the treatment of periodontitis with a light wavelength of 445 nm showed no dangerous changes in human tissue as a result of the treatment [19,56]. Nevertheless, the question of cytotoxicity should be considered in further studies on clinical application.

4. Material and Methods

4.1. Test Specimens

For the experiments, polished and sintered yttria stabilised zirconium dioxide discs (VITA Zahnfabrik, Bad Säckingen, Germany) with a diameter of 5 mm, thickness of 1 mm, and a roughness average of R_a = 251 ± 21 nm were used. The roughness was measured by a Dektak 3 St Surface Profilometer (Irvine, CA, USA).

4.2. Test Organism and Specimen Preparations

The test organisms Enterococcus faecalis (ATCC^® 29212™), Streptococcus mutans (DSM 20523/ATCC^® 25175™), and Candida albicans (SC5314/ATCC MYA-2876D-5™) were used for the main experiments: two Gram positive bacteria and one fungus as representatives of important dental pathogens.

The microorganisms were cultured on blood agar plates (Columbia Agar +5% sheep blood, bioMérieux, Nürtingen, Germany) were harvested with an inoculation loop and resuspended in a tube contained 4 mL of Triptic soy Bouillon, Carl Roth, Karlsruhe, Germany) for Enterococcus faecalis and Streptococcus mutans or Yeast Peptone Dextrose Broth (Sigma-Aldrich, Y1375, St. Louis, MO, USA) for Candida albicans. The tubes were incubated for 24 h at 37 °C in a shaker incubator (Heidolph Titramax 1000, Heidolph Inkubator 1000, Fa. Heidolph Instruments, Schwabach, Germany) under aerobic conditions. The suspensions were washed two times by centrifugation for 5 min at 3000 rpm (rounds per minute) (Megafuge R8, Thermo Scientific SL8, Heraeus, Hanau, Germany) and resuspension in deionised water for Candida albicans and Enterococcus faecalis or potassium phosphate buffer solution (50 mM di-Potassium hydrogen phosphate trihydrate, pH 7.5) for Streptococcus mutans. The respective resuspension medium was evaluated for the test organisms to prevent disruptive organic residues or crystal formation (Appendix A Figure A1) on the surface without compromising the viability of the test organisms. To ensure a sufficient number of microorganisms of around 10⁷ to 10⁸ CFU/mL for Candida albicans and 10⁸ to 10⁹ CFU/mL for Enterococcus faecalis and Streptococcus mutans for the tests, the suspension concentrations were adjusted to the defined values based on the optical density by diluting with the washing medium. The defined OD_{620 nm} values were 2.13–2.20 for Candida albicans, 1.0–1.21 for Enterococcus faecalis, and 1.99–2.06 for Streptococcus mutans, which were evaluated in accordance with DIN EN 13697:2019-10 [57], and measured by a spectral photometer (Synergy™ HTX Multi-Mode Microplate Reader, BioTek^® Instruments, Winooski, VT, USA).

The tube was shaken by a vortex, and 10 µL of the initial test suspension was dropped and spread on each specimen surface. After drying under laminar flow, these inoculating steps were repeated two times, resulting in a bacterial cell density of 8.3 CFU/cm², before treatment within 60 min. One test run with two samples per test group was carried out on each of seven different days (n = 14 for each test group). Deviations are mentioned separately.

4.3. Laser Application and Temperature Measurement

Microbially covered specimens were treated by a diode laser (SIROLaser Blue with EasyTip 320 µm, numerical aperture 0.22, Dentsply Sirona, Bensheim, Germany) at a wavelength of 445 nm in continuous wave mode with the output power set to 1 W. The specimen was fixed in a sample holder that was positioned on a PC controlled x-y-z-table (SMC corvus eco, PI miCos, Shrewsbury, MA, USA). The laser handpiece was fixed perpendicularly above the test specimen with a distance of 1 mm between the specimen and the distal end of the application fibre (Figure 2). This setup results in a circular illumination area on the specimen with a diameter of approximately 0.8 mm corresponding to a mean power density of 200 W/cm². The sample was moved twice in a meandering pattern under the fibre tip using the x-y-z table. The line spacing was 0.1 mm, and the scan area was 5 × 5 mm. The movement and overlap of the illuminated area cause the laser beam profile to be spatially averaged. The mean energy density on the specimen is calculated as the product of the mean power density multiplied by the illumination time. The illumination time corresponds to the scan time multiplied by the ratio of the illumination area to the scan area. This results in a mean energy density of about 1000, 500, and 250 J/cm² at scan speeds of 1 mm/s, 2 mm/s, and 4 mm/s, calculated according to the standard procedure [49]. The application parameters were based on preliminary tests.

Figure 2.

(a). Schematic test setup (not true to dimensions); (b), measurement of the temperature at this location shortly after the laser has irradiated the sample; (c), measurement of the temperature at this location on the test specimen during laser treatment. Figure source [22].

During the laser treatments, the temperature was recorded using an infrared camera (Optris PI, Optris GmbH, Berlin, Germany). The camera was aligned at an angle of 45° to the laser in the direction of the specimen surface to be treated. Two areas were defined in the camera software (Optris PIX Connect, Rel. 2.9.2147.0) for the temperature measurement for each specimen. The first area was defined directly next to the laser beam, behind the direction of movement, in order to determine the temperature on the sample after the laser had irradiated it (Figure 2b). The second area was defined below the laser tip to determine the temperature between the laser beam and the sample during the movement of the laser over the sample (Figure 2c). The mean value of the two measured values per area was used to determine the temperature per treated specimen. Treatment temperatures above 42 °C were excluded from the evaluation to ensure that no thermal effect alone could account for the antimicrobial effect.

After treatment, the specimens were transferred into wells of microplate (48-well-micro titre plate, Grainer) for colony-forming units assay.

4.4. Treatment with Chlorhexidine

Chlorhexidine digluconate (CHX, pharmacy university medicine Greifswald, Germany) was used as a positive control. For this purpose, the samples stored in microtitre plates were covered with 300 µL CHX (2%) for 60 s. The liquid was then pipetted off and discarded. To inactivate the effect of the CHX, 300 µL Lipofundin (MCT/LCT 20%, B. Braun, Melsungen, Germany) was added [58,59] and removed again after 10 min incubation time at room temperature.

4.5. Antimicrobial Effect by Colony-Forming Units (CFU) Assay

To detach the microorganisms from the surface of the test specimens (recovery), ultrasound was used in 48-well plates with 0.5 mL sodium chloride solution (0.9%) and five glass beads per well for a period of 10 min, similar to other experiments [60,61]. The temperature of the ultrasonic bath (Elmasonic S 30 H, Elma Schmidbauer GmbH, Singen, Germany) was controlled to ensure it remained below 30 °C. The microtitre plate was then shaken on a shaker for 5 min at a frequency of 600/min. The same procedure was repeated twice with a 5 min ultrasonic bath and 3 min shaking each time, mixed, and made up to 1 mL total volume with a further 0.5 mL sodium chloride solution (0.9%). Starting from these suspensions, dilution series were then prepared (from 100 to 10⁻⁷), and 100 µL were applied to two agar plates per dilution step and spread evenly and then incubated in an incubator for 48 h at 37 ± 1 °C. After incubation of the agar plates, the colony-forming units on the agar plates were counted. In accordance with DIN EN 1040:2006-03 [58], only agar plates with a CFU count between 15 and 300 were considered for the evaluation. The determined CFU was calculated back to CFU/mL. Differences between the treated samples and the controls and the statistical analysis were carried out using the log₁₀-transformed CFU/mL values.

4.6. Statistical Analyses

4.6.1. Group Comparisons

For the group comparisons in relation to the test group traversing speed, the three test groups 4 mm/s, 2 mm/s, 1 mm/s, and the negative control group (untreated samples) were considered. These four groups were compared as fixed factors in mixed models, which are advantageous when the test groups are not equally sized, with the seven days modelled as a random factor [62]. The mixed models were analysed using Stata (v. 14) [63], using the Kenward–Roger correction for small samples [64]. The specified “contrast value” represents the difference between the estimated marginal means of the respective groups to be compared. The number of samples was primarily 14 for each test group and test organism. For the Enterococcus faecalis test group 4 mm/s and the Candida albicans test group 1 mm/s, the number of samples was 4 in order to conserve resources for conducting the study. Therefore, the significance level α was set at 0.01 due to the sequential testing after 4, 10, and 14 observations, for which an alpha-spending approach was used [65]. Moreover, Bonferroni correction was applied to the six pairwise comparisons between the four groups.

For all three microorganisms, the test group that showed the highest effect (test group 1 mm/s) was tested against the positive control. The non-inferiority was analysed. Based on the positive control, a tolerance limit was defined that corresponds to around 10% of the main effect. The main effect corresponded to the mean value of the measured colony-forming units of the positive controls. The main effect of the group to be analysed should then lie within this range in order to discuss the non-inferiority.

4.6.2. Sensitivity Analyses

For the sensitivity analyses as bias analyses were additionally performed within the sample of Candida albicans (linear regression) and for the three microorganisms in one model, which are listed as additional information in Appendix B. Sensitivity analyses.

5. Conclusions

The treatment of the three representative oral pathogenic microorganisms with the blue laser resulted in a statistically significant reduction in microbial organisms within the specified treatment parameters. The strongest effects were observed at the slowest traversing speed at 1 mm/s. Streptococcus mutans demonstrated higher sensitivity to blue light with a wavelength of 445 nm than Enterococcus faecalis and Candida albicans. This shows that the speed of the treatment has a significant influence on the effect. The reduction in microbial organisms in this study and in the cited literature are still too low for the light at 445 nm to be considered a clinically relevant antiseptic. However, its promising properties suggest that the blue light could be used as an adjunct to various dental therapies to improve the outcomes. Further studies are needed to gain a deeper understanding of the parameters that enhance the antimicrobial efficacy and to extend the practical applications of blue laser technology in clinical settings.

Abbreviations

CFU	colony-forming units’ assay
CHX	chlorhexidine digluconate
DNA	deoxyribonucleic acid
LED	light-emitting diode
MTT	tetrazolium dye (3-(4,5-di methyl thiazol-2-yl)-2,5-diphenyltetrazolium bromide)

Appendix A. Supplementary Preliminary Investigations to Evaluate Suitable Parameter Settings

Appendix A.1. Tables

Table A1.

Summary of the findings from smaller preliminary tests for the expected effective antimicrobial effects on ceramics depending on the microorganism. The number of samples was four for the distance and eight for each test group and test organism of treatment repetition (2- or 5-fold) and traversing speed (2 or 4 mm/s). The treatment was a line across the sample surface. The antimicrobial effect was evaluated based on the lack of blue crystal formation of the MTT assay along the treatment line in its visible intensity and the resulting line width. Data source [22].

Parameter		Test Organisms
		Enterococcus faecalis	Streptococcus mutans	Candida albicans
Energy	Power input	1–2.5 W	0.4–1.5 W	0.8–2.0 W
Treatment option	Distance	1–2 mm	1–2 mm	0.5–2 mm
	Treatment repetition	No *	No *	No *
	Traversing speed	Low *	Low *	Low *

* Information on the influence of the parameters on the antimicrobial effect: “No”: no effect recognisable/present; “Low”: influence on antimicrobial effect is recognisable, but not dominant.

Table A2.

Summary of results of the mean values for 0.2% CHX treatment for 10 min and 2% CHX treatment for 1 min (n = 3) for each test organism.

Test Organisms	0.2% CHX	2% CHX	Control
	10 min	1 min
Enterococcus faecalis	4.93	3.52	7.54
Streptococcus mutans	1.90	1.02	6.82
Candida albicans	2.36	1.31	6.42

Appendix A.2. Figure

Figure A1.

Testing of the carrier solutions: phosphate buffered saline (PBS), deionised H₂O, potassium phosphate buffer (KPO₄), sodium potassium phosphate buffer (NaKPO₄) and Tris-buffered saline with hydrochloric acid (Tris-HCl) by applying the microorganisms to the ceramic test specimens, vital staining using MTT, and subsequent microscopic evaluation of the microbial converted blue formazan. Tris-HCl reduced the viability slightly for Enterococcus faecalis and strongly for Streptococcus mutans. Sodium potassium phosphate buffer was not suitable for any of the test organisms. PBS was also not optimal for this test setup, with clearly recognisable salt crystals forming. Stable viability results were shown with the H₂O (deionised) solution for Candida albicans and Enterococcus faecalis and with the potassium phosphate buffer solution for Streptococcus mutans. Figure source [22].

Appendix B. Sensitivity Analyses

Appendix B.1. Bias Analysis Within the Sample of Candida Albicans (Linear Regression)

A comparison with conventional linear regression was performed for Candida albicans. The seven days were modelled as clusters, so that robust variances were estimated [66].

In the linear regression applied here, the point estimate of 4.37 log₁₀ CFU/mL for group 1 mm/s with n = 4 (Table A3) corresponded to the observed mean value (Table 1), while in the mixed model information from the other groups, the point estimate of 4.75 log₁₀ CFU/mL was closer to the other groups (Table 2) [22].

Mixed models are less susceptible to bias. This is also emphasised by the differences in the estimated standard error, which was much too small at 0.01 in the conventional linear regression (compared to 0.25 in the mixed model). It can, therefore, be said that the uncertainties in the estimation of small groups are better considered in the mixed model.

Table A3.

Conventional linear regression, estimated values, Candida albicans log₁₀ (CFU/mL). Data source [22].

Method	Estimated Marginal Mean [log10 CFU/mL]	Standard Error [log10 CFU/mL]
Negative controls	5.77	0.25
4 mm/s	5.32	0.29
2 mm/s	5.40	0.21
1 mm/s	4.37	0.01

Appendix B.2. Bias Analysis: The Three Microorganisms in One Model

In addition, all three microorganisms were analysed in a single mixed model in order to better estimate a possible bias effect due to small numbers and to better incorporate the available information. Here, 146 observations were modelled together, with the method, the microorganism, and the interaction between method and microorganism as fixed factors. The day was included in the model as a random factor.

For the 1 mm/s group with n = 4 for Candida albicans, the estimated marginal mean here was 4.45 log₁₀ CFU/mL (Table A4) and thus closer to the observed mean value of 4.37 log₁₀ CFU/mL (Table 1) than to the value of 4.75 log₁₀ CFU/mL estimated in the Candida albicans model (Table 2) [22].

Table A4.

Bias analysis: all three microorganisms Enterococcus faecalis, Streptococcus mutans, and Candida albicans in one model log₁₀ (CFU/mL). Fixed factors are the method (4 mm/s, 2 mm/s, 1 mm/s), the microorganism, and the interaction between method and microorganism. In total, 146 observations were modelled together, the day as a random factor. Data source [22].

Interaction Microorganism/ Method	Estimated Marginal Mean [log10 CFU/mL]	Standard Error [log10 CFU/mL]
Enterococcus faecalis/
Negative control	7.26	0.18
4 mm/s	5.78	0.26
2 mm/s	5.85	0.18
1 mm/s	5.48	0.18
Streptococcus mutans/
Negative control	6.49	0.18
4 mm/s	4.49	0.18
2 mm/s	4.47	0.18
1 mm/s	3.95	0.18
Candida albicans /
Negative control	5.77	0.18
4 mm/s	5.33	0.18
2 mm/s	5.4	0.18
1 mm/s	4.45	0.26

Author Contributions

All authors (R.M., L.D., C.S., L.J., T.K. and A.W.) contributed to the study conception and design. Material preparation, data collection, and analysis were performed by L.D. and R.M. The statistical analysis was performed by C.S. The draft of the manuscript was written by L.D., R.M. and A.W. All authors have read and agreed to the published version of the manuscript.

Institutional Review Board Statement

Not applicable.

Informed Consent Statement

Not applicable.

Data Availability Statement

The datasets used and/or analysed during the current study are available from the corresponding author on reasonable request.

Conflicts of Interest

The authors declare no conflicts of interest.

Funding Statement

This work was realised within the research project ‘PeriPlas’, supported by the Federal Ministry of Education and Research (BMBF, grant number 13N14480). The spectral photometer was supported by the European Regional Development Fund (ERDF/EFRE, registration number GHS-19-0013). Support for the publication fee was provided by University of Greifswald’s publication fund and Dentsply Sirona.