Remember me
ORIGINAL ARTICLE
Year : 2023 | Volume
: 13
| Issue : 1 | Page : 21-27
In vitro antimicrobial efficacy of riboflavin, ultraviolet-A radiation, and combined riboflavin/ultraviolet-A radiation on ocular pathogens
Ashok Sharma1, Rajan Sharma1, Jagdish Chander2, Verinder S Nirankari3
1 Cornea Centre, Chandigarh, India
2 Department of Microbiology, Government Medical College, Chandigarh, India
3 Department of Ophthalmology, University of Maryland, College Park, MD, USA
Correspondence Address:
Source of Support: None, Conflict of Interest: None
DOI: 10.4103/tjo.tjo_28_21
PURPOSE: There is a lack of studies evaluating the antimicrobial effect of riboflavin and ultraviolet-A (UVA) radiation using an epithelium-off cross-linking technique (Dresden protocol) in the literature. This study aimed to demonstrate the antimicrobial properties of riboflavin and UVA radiation and combined riboflavin/UVA radiation on various ocular pathogens Dresden protocol.
MATERIALS AND METHODS: This was a prospective and comparative study. This study included four groups: Staphylococcus aureus, Pseudomonas aeruginosa, Candida albicans, and Acanthamoeba. All the groups were tested 12 different times with riboflavin alone, UVA light alone, and combined riboflavin with UVA. The growth inhibition zones were measured. Descriptive statistics were expressed as mean and standard deviation. The main outcome was measurement and analysis of growth inhibition zones.
RESULTS: The mean growth inhibition zones following riboflavin and UVA exposure were 9.70 ± 1.63 mm for P. aeruginosa and 7.70 ± 1.08 mm for S. aureus. The mean growth inhibition zones for P. aeruginosa were greater than for S. aureus (t = 2.395, P = 0.038). The mean growth inhibition zones for C. albicans and Acanthamoeba were 0. The mean growth inhibition zones following UVA exposure alone and riboflavin alone for all organisms were also 0.
CONCLUSION: Combined riboflavin and UVA in the clinical dose was effective against P. aeruginosa and S. aureus but ineffective against C. albicans and Acanthamoeba. Our study shows the potential of combined riboflavin and UVA in the treatment of bacterial keratitis.
Keywords: Acanthamoeba, Candida albicans, Pseudomonas aeruginosa, riboflavin, Staphylococcus aureus, ultraviolet-A radiation
Microbial keratitis is a sight-threatening corneal disease and resulting corneal opacities constitute a major cause of corneal blindness worldwide.[1] Infective keratitis in developing countries (Tamil Nadu, South India, 113.0/100,000 people) is at least 10 times more prevalent compared to the United States (Minnesota, 11.0/100,000 people).[2],[3] Corneal epithelial trauma is the most common risk factor; other risk factors include wearing contact lenses, steroid exposure, and ocular surface disease. Although clinical examination helps to categorize infective keratitis into bacterial, fungal, viral, and parasitic origins, identification of pathogens is vital for initiating appropriate treatment.[4] Patients suffering from microbial keratitis need urgent and aggressive treatment. Ophthalmologists face several challenges in the treatment of microbial keratitis.
The prior steroid exposure, association of systemic diseases (autoimmune or diabetes mellitus), and delay in establishing diagnosis may be responsible for less favorable response to medical therapy. Increasing anti-microbial resistance against third and fourth-generation fluoroquinolones makes medical treatment ineffective in certain cases.[5],[6] Patients receiving fluoroquinolone monotherapy need close monitoring and if a patient shows deterioration, one can consider switching to fortified broad-spectrum antibiotics.[7] Treatment of fungal corneal ulcers is even more challenging, as the available antifungal agents are limited. The most commonly used anti-fungal agent, natamycin 5% suspension, a polyene derivative, has limited corneal penetration.[8] Fluconazole has better ocular bioavailability, but due to its narrow spectrum, it is not considered the first choice against keratitis due to filamentous fungi.[9] Cases of Acanthamoeba keratitis are increasing worldwide.[10]Acanthamoeba keratitis occurs in people who wear contact lenses, but the disease is also seen frequently in people who do not wear contact lenses in developing nations.[11] Diagnosis is often considerably delayed as the disease misdiagnosed as bacterial, fungal, or viral keratitis. Early cases respond to medical therapy (polyhexamethylene biguanide 0.02%, chlorhexidine 0.02%, and propamidine isethionate 0.1%), but advanced cases show a less favorable response to drug treatment often resulting in the failure of medical treatment.[12] Patients not responding to medical therapy may need emergency penetrating keratoplasty.[13] Due to inadequate eye banking facilities and nonavailability of quality donor eyes in developing countries, it is difficult to perform therapeutic grafts and save these eyes.[14] Therefore, it is of paramount importance to innovate and evaluate the newer treatment options. Corneal collagen cross-linking (CXL) is a new promising treatment modality for infectious keratitis.
Corneal collagen cross-linking (CXL) with riboflavin (CXL) was introduced in 2003 by Wollensak et al.[15] Since then, CXL has become a widely performed procedure to arrest the progression of ectasia and stabilize corneal topography in progressive keratoconus.[16] CXL has also been used to treat pellucid marginal degeneration and postlaser-assisted in situ keratomileusis (LASIK) corneal ectasia with encouraging results.[17] The Dresden protocol is the standard and most commonly performed CXL technique.[15] The technique involves epithelial debridement (epi-off), corneal stroma saturation with riboflavin, and exposure to UV-A radiation (370 nm) at 3 mW/cm2 for 30 min to achieve a surface dose of 5.4 J/cm2.[15] CXL without the removal of the corneal epithelium (epi-on) has been advocated to reduce the risk of infective keratitis and postoperative pain, experienced with epi-off CXL.[18],[19]
Another variant accelerated CXL has also been found effective in stabilizing progressive keratoconus by reducing the time of UV-A exposure and increasing the UV-A radiation power.[20],[21] Despite these developments, due to its superior efficacy, the Dresden protocol still remains the gold standard CXL procedure for keratoconus.[22] UV-A irradiation in the presence of riboflavin as a photosensitizer causes the release of singleton oxygen and superoxide radicals.[23] These radical species increase the strength of stromal collagen bonds, thus enhancing the biomechanical strength of the cornea.[24] These reactive oxygen species also cause unspecific oxidative damage to the DNA of various microbes.[25] Due to this property, CXL is postulated as an adjunct treatment for infectious keratitis.
Microbial inactivation using riboflavin exposed to ultraviolet-A (UVA) radiation was first reported in 1960.[26] Several studies have attributed this antimicrobial effect to the nonspecific oxidative damage mediated by the production of singlet oxygen, superoxide ions, and hydroxyl radicals.[27] These reactive oxygen species cause cytotoxic effect and death of bacteria by causing damage to either DNA or the lysis of cell wall.[25] In a study on inactivation of Staphylococcus aureus using hematoporphyrin as a photosensitizer, cell membrane damage has been shown as the main result.[28] The DNA damage has been evidenced by the breaking of single-stranded DNA, double-stranded DNA, and the disappearance of plasmid superhelix fragments.[29] Photoactivated chromophore for keratitis–corneal CXL (PACK) has been reported to inactivate a wide spectrum of microorganisms including viruses, fungi, bacteria, and parasites.[30] It has been postulated that UV radiation exposure in the presence of riboflavin may also be used to treat microbial keratitis. Several clinical studies and case reports have documented favorable results of this procedure infective keratitis.[31],[32] Schrier et al.[33] reported that CXL is an effective modality to eradicate the bacteria S. aureus, methicillin-resistant S. aureus, and Pseudomonas aeruginosa. Makdoumi et al.[34] reported that riboflavin photoactivation using UVA (365 nm) significantly reduced P. aeruginosa, S. aureus, Staphylococcus epidermidis, and that the combination of riboflavin and UVA is more effective in reducing bacterial number than UVA alone. Martins et al.[35] reported that riboflavin/UVA was effective against S. aureus, S. epidermidis, P. aeruginosa, methicillin-resistant S. aureus, multidrug-resistant P. aeruginosa, and drug-resistant Streptococcus pneumoniae but was ineffective on Candida albicans. Kashiwabuchi et al.[36] found that combined riboflavin/UVA did not eradicate Acanthamoeba trophozoites.
In previous in vitro, ex vivo, and clinical studies, the parameters used were mostly different from the standard Dresden protocol [Supplementary Table 1] and [Supplementary Table 2]. Both the concentration of riboflavin solution and the total UV-A radiation used were too high to produce clinical benefit for patients [Supplementary Table 1] and [Supplementary Table 2]. We used a different in vitro experiment design to previous studies that was more straightforward and could be reapplied for in future investigations. The present study evaluates antimicrobial effect of riboflavin, UVA, and combined riboflavin/UVA on methicillin sensitive S. aureus, P. aeruginosa, C. albicans, and Acanthamoeba using Dresden protocol parameters.
MethodsIn a prospective comparative study, the antimicrobial properties of combined UVA irradiation and riboflavin, UVA irradiation alone, and riboflavin alone were studied on methicillin-sensitive S. aureus, P. aeruginosa, C. albicans, and Acanthamoeba. Approval was obtained from the Cornea Centre Institutional Review Board (IRB No. 0009/21). The study was conducted with adherence to the tenets of the Declaration of Helsinki.
We used three exposure groups for each microorganism tested: group A – combined UVA irradiation and riboflavin, group B – UVA irradiation alone, and Group C – riboflavin alone. Standard inoculations of methicillin-sensitive S. aureus, P. aeruginosa, C. albicans, and Acanthamoeba were prepared in the laboratory under direct supervision of an experienced microbiologist (JC). P. aeruginosa and methicillin-sensitive S. aureus were selected from the human clinical isolates. S. aureus culture of freeze-dried microorganisms was used for quality control. Freeze-dried S. aureus was subjected to two subcultures before testing. A McFarland turbidity greater than or equal to 0.5 was considered as the endpoint.
A sterile cotton swab was dipped into the inoculum and excess was removed by pressing the swab against the wall of the vial. The agar plates were swabbed three times ensuring even distribution. C. albicans strains prepared by microbiology department and acquired from human keratitis patients were inoculated on Sabouraud dextrose agar medium. Sabouraud dextrose agar plates were inoculated by streaking, as with standard bacteriological media. These inoculations were then exposed to riboflavin with UVA radiation (Group A), UVA alone (Group B), and riboflavin alone (Group C). All the four groups of microbes were tested at 12 different times. Separate Petri dishes were used for different specimens.
A 7.5 mm diameter paper disc was punched out using a disposable corneal trephine. This paper was adhered to the center of the outer surface of transparent Petri dish containing culture media. A circle was drawn with a black marker, using the paper disc as a guide and its center was also marked. For Group A, a drop of a riboflavin solution (0.1% riboflavin 5-phosphate and 20% dextran T-500) was placed on the inoculated media at the center of the marked circle [Figure 1]a. The riboflavin drop was allowed to diffuse into the inoculated media for 20 min. The area was then exposed to UVA radiation.
Figure 1: (a) Instillation of riboflavin 0.1% drop in the center of 7.5 mm circle marked on the back of inoculated agar plate. (b) Exposure of the riboflavin saturated area to ultraviolet-A irradiance of 3.0 mW/cm2 (c) Growth inhibition zone following photoactivated chromophore for keratitis–corneal cross-linking on Pseudomonas aeruginosa inoculate (d) Growth inhibition zone following photoactivated chromophore for keratitis–corneal cross-linking on Staphylococcus aureus inoculate. (e) No growth inhibition zone following photoactivated chromophore for keratitis–corneal cross-linking on Candida albicans inoculate. (f) 1 – Nonnutrient agar plate with Escherichia coli overlay, 5 mm disc of cultured Acanthamoeba, with cultured surface facing Escherichia coli. 2 – Magnified view of inverted 5 mm disc of cultured Acanthamoeba. 3 – Acanthamoeba growing from the margin of the disc (×10)For Acanthamoeba, nonnutrient agar plates with Escherichia coli overlay were used. From Acanthamoeba culture, a 5 mm diameter circular disc was removed with the help of skin biopsy punch. A 5 mm disc of Acanthamoeba culture was then removed and placed in the punched-out area. A 7.0 mm paper disc was attached to the backside of the culture plate. Cultures were subdivided into the same three exposure groups as for the other pathogens, and a fourth group that was not exposed to any treatment served as a control. All the culture plates were sealed and incubated at 30°C.
For all cases using UVA, the exposure was performed using the device CL-UVR corneal CXL system (Appasamy Associates, First Street, Arumbakkam, 20, SBI Officers Colony, Chennai, 600106, Tamil Nadu, India). A calibrated UV meter was used to ensure an irradiance of 3.0 mW/cm2 before each treatment session. The parameters of emitted UV light included wavelength 370 ± nm, irradiance 3 mW/cm2, and diameter 7.5 mm. Using these parameters, exposure of the culture plate surface was done for 30 min [Figure 1]b.
In the second group (UVA alone), UVA exposure was performed on the inoculated media in the same manner as Group 1. In the third group (riboflavin alone), isotonic riboflavin solution (0.1% riboflavin 5-phosphate and 20% dextran T-500) was placed on the inoculated media and allowed to diffuse for 20 min.
The growth inhibition zones were measured by a microbiologist in the microbiology department who had no prior information about the type of exposure. The area showing no obvious growth visible to the naked eyes was considered the zone of inhibition [Figure 1]c, [Figure 1]d, [Figure 1]e. The growth inhibition zones were measured with a transparent ruler from the back of the plate. The plate was held against the light and the diameter of the growth inhibition zone was measured in millimeters (mm). For Acanthamoeba, the distance (mm) from the margin of the inoculation disc to the point of observation of trophozoites and cysts in the agar plate was measured in the treatment and control group. For Acanthamoeba, the growth inhibition zone was the distance for control minus the treatment group [Figure 1]f. Culture plates were incubated at 34°C–35°C in an ambient air incubator for 48 h. The growth inhibition zones in all groups were measured at 24 and 48 h after treatment.
To test for normality, Shapiro–Wilk test was performed and was found to be nonsignificant (P > 0.05). Student's t-test was used to compare the two groups.
ResultsThe growth inhibition zones were larger at 24 h than those at 48 h for P. aeruginosa and S. Aureus. The values for C. albicans and Acanthamoeba were 0 at 24 h. The measurements obtained after 24 h of incubation were considered for the analysis.
The mean growth inhibition zones following combined riboflavin and UVA (Group 1) exposure were 9.70 ± 1.63 for P. aeruginosa and 7.70 ± 1.08 mm for S. aureus. The mean growth inhibition zones for C. albicans and Acanthamoeba were 0. The mean growth inhibition zones following UVA exposure alone (Group 2) and riboflavin alone (Group 3) for S. aureus, P. aeruginosa, C. albicans, and Acanthamoeba were also 0.
Combined riboflavin and UVA (Group 1) treatment was effective against P. aeruginosa and S. aureus, whereas UVA alone (Group 2) or riboflavin alone (Group 3) treatments showed no effect against any of the microorganisms. Within the treatment Group 1, mean growth inhibition zone was significantly more for P. aeruginosa than for S. aureus (t = 2.395, P = 0.038). The results of the treatment groups on various ocular pathogens are shown in [Table 1].
Table 1: Growth inhibition zones on cultures of Pseudomonas aeruginosa, Staphylococcus aureus, Candida albican, and Acanthamoeba on exposure to combined riboflavin with ultraviolet-A, ultraviolet-A alone, and riboflavin alone Discussion In this study, growth inhibition was observed in P. aeruginosa and S. aureus with the use of combined riboflavin and UVA. When the two treatment options were used separately, there was no bactericidal effect. The results show that riboflavin with UVA exposure has the potential to play a key adjunct role in the treatment of bacterial keratitis. Our findings did not show a treatment effect against C. albicans or Acanthamoeba. As the effect is dependent upon the distance of the treatment area from the UVA source, the bactericidal effect of UVA may not be uniform. Studies have shown the antibacterial effect of riboflavin with UVA radiation.[33],[34] The effect has been found to be dose dependent, and in some of the studies, a higher dose has been used. We evaluated the effect using riboflavin 0.1% and UVA wavelength 370 nm, irradiance 3 mW/cm2, and diameter of the exposed area 7.5 mm. These parameters mirror what is currently used clinically to perform collagen CXL in keratoconus or post-LASIK ectasia patients. We chose these parameters because we will be able to use these parameters clinically to treat infective keratitis.
In our study, riboflavin and UVA exposure combined achieved P. aeruginosa and S. aureus growth inhibition but not in the case of C. albicans and Acanthamoeba. Our results were similar to those reported by Schrier et al.[34] These authors reported that riboflavin in combination with UVA was effective in eradicating the bacteria S. aureus, methicillin-resistant S. aureus, and P. aeruginosa.[34] In another study by Kashiwabuchi et al.,[37] the combination of riboflavin 0.1% and UVA light at 365 nm did not exhibit antimicrobial activity against oxacillin-susceptible Staphylococcus aureus. Makdoumi et al.[34] also reported that riboflavin photoactivation using UVA (365 nm) significantly reduces P. aeruginosa, S. aureus, and S. epidermidis. These authors found that the combination of riboflavin and UVA is more effective in reducing bacterial number than UV alone. We did not observe any growth inhibition with either UVA or riboflavin alone. Riboflavin and UVA alone may not be useful in the treatment of infective keratitis.
Riboflavin and UVA induce alteration in the biochemical and chemical properties of the collagen. The photochemical reaction produces covalent bonds and cross-links between collagen fibrils.[54] These biochemical changes make the corneal stroma stiffer and more resistant to enzymatic bacterial degradation. Due to this effect, CXL may reduce the progression of corneal melt in infective keratitis. Several studies have shown that riboflavin and UVA exposure may cause damage to the DNA and RNA of the bacteria, viruses, and fungi, and thus exert an antimicrobial effect.[55],[56] One study has shown that a photosensitizer (riboflavin) attaches to the nucleic acid of bacteria and causes damage to DNA by the photochemical reaction.
Riboflavin and UVA (280–370 nm) may damage nucleic acid by direct electron transfer, production of singlet oxygen, and generation of hydrogen peroxide with the formation of hydroxyl radicals.[57] Some studies have shown that due to the complex cell-wall structure of Gram-negative species compared to Gram-positive ones, Gram-positives could be eliminated with a lesser photosensitizer and light penetrating the cell-wall structure.[58] Another study showed that, compared to S. epidermis, S. aureus is more likely to form biofilm, which prevents penetration of a photosensitizer or light inside the complex. Therefore, elimination of bacteria with biofilm formation may need higher light energy or longer irradiation time.[59]
Martins et al.[35] also reported that riboflavin/UVA was effective against S. aureus, S. epidermidis, P. aeruginosa, methicillin-resistant S. aureus, multidrug-resistant P. aeruginosa, and drug-resistant S. pneumonia, but the same was ineffective against C. albicans. In this study, the authors used UVA exposure time of 1 h. According to the Dresden protocol, clinically UVA exposure time of 30 min is recommended. In a recent experimental study, the antifungal efficacy of corneal collagen CXL with photoactivated riboflavin (PACK-CXL) and voriconazole was compared in Fusarium solani and C. albicans keratitis models.[42] The authors reported lower keratitis score for F. solani and C. albicans with combined riboflavin and UVA treatment.[42] Since the earlier study has shown a lower fungicidal effect with 0.1% riboflavin, the authors therefore used 0.25% riboflavin.[42] In clinical practice, riboflavin drops are used to saturate corneal stroma with riboflavin and the concentration of riboflavin eye drops is 0.1% in 20% dextran.
We did not observe any growth inhibition of Acanthamoeba with any of the treatment combinations. In a study by del Buey et al., the combined riboflavin/UVA did not eradicate two strains of Acanthamoeba.[40]C. albicans is known to form biofilm and may require higher dose of UV-A irradiation. The reason for observing no growth inhibition against C. albicans and Acanthamoeba may be the lower photobiological dose (energy per unit area) in our study.
CXL is relatively a simple and safe surgical procedure.[60] Transient corneal haze and epithelial defects are common. Serious complications including sterile corneal ulceration, peripheral ulcerative keratitis, infective keratitis, and recurrence of Herpes simplex keratitis due to UVA exposure have also been reported.[61],[62],[63],[64] Penetration of UVA is known to occur up to 400 μ corneal depth.[23] The maximal effect of the riboflavin-UV is in the anterior corneal stroma (up to 300 μ) and in patients with epi-off corneal thickness more than 400 μ, the endothelium should be spared.[23] Corneal endothelial damage although rare has been reported following collagen CXL.[65] Cases of persistent corneal edema following collagen CXL requiring penetrating keratoplasty have been reported.[66]
There are several studies showing successful treatment of bacterial keratitis using riboflavin with UVA as an adjunct treatment.[67],[68] Our study supports the role of riboflavin with UVA combination in the treatment of bacterial keratitis. The lack of in vitro activity of PACK-CXL against trophozoites may be due to different illumination (illuminator delivers 6.2 J per mL) and the wavelength (range 265–370 nm).[69] However, the use of riboflavin with UVA in fungal keratitis and Acanthamoeba keratitis needs further evaluation. We have summarized previous in vitro, in vivo, and human studies on PACK/CXL efficacy in microbial keratitis in [Supplementary Table 1] and [Supplementary Table 2].[38],[39],[41],[43],[44],[45],[46],[47],[48],[49],[50],[51],[52],[53]
There are some limitations of the present study. It is imperative that it is not possible to detect quantitative decrease in the number of viable microbes. The quantum of the inoculum of the microbes tested in this study has not been quantified. In addition, we did not carry out tests to detect quantitative decrease in the number of viable microbes. We could only detect growth of the microbes and motility of trophozoites in the case of Acanthamoeba. Another limitation is the authors did not add a drop of 20% dextran T-500 without riboflavin in the Petri dish for the second group or cover the Petri dish to avoid UV exposure (30 min) after 20 min diffusion with 0.1% riboflavin solution for the third group. However, our method is suitable to replicate solid state in vivo corneal conditions.
In most in vitro studies, the parameters used are higher than those of the Dresden protocol. The concentration of riboflavin solution used is higher, and the total UV-A radiation is also higher than one can use clinically for the benefit of patients. The in vitro study design used in this study is simple, unique, and different from the previous studies. This model may be useful for future investigations. More ex vivo studies using the Dresden protocol are needed as, although higher energy may yield positive results ex vivo, it may be toxic to the cornea when used in patients clinically.
ConclusionCombined riboflavin and UVA in the clinical dose inhibited the growth of P. aeruginosa and S. aureus. CXL was not found to be effective against C. albicans and Acanthamoeba. Our study indicates that CXL has the potential of being used as an adjunct treatment in bacterial keratitis.
Financial support and sponsorship
Nil.
Conflicts of interest
The authors declare that there are no conflicts of interests of this paper.
Supplementary Materials
References
Comments (0)