Rose bengal photodynamic antimicrobial therapy combined with amniotic membrane graft for the management of infectious keratitis in a patient with keratoprosthesis

Abstract

Purpose

To describe the use of Rose Bengal Photodynamic Antimicrobial Therapy (RB-PDAT) combined with amniotic membrane graft (AMG) as a novel surgical technique for infectious keratitis in a patient with Boston Keratoprosthesis (B-Kpro) type 1 Lucia.

Methods

A 91-year-old woman with a history of B-Kpro type 1 Lucia implantation in both eyes presented for routine follow-up complaining of pain and redness in the left eye. Slit-lamp examination revealed a 6 mm corneal infiltrate with an epithelial defect adjacent to the optic stem. After 2 weeks of topical antimicrobial therapy without any improvement, the decision was made to perform RB-PDAT in conjunction with an AMG. The infectious keratitis resolved within 21 days.

Conclusions

The combination of RB-PDAT and AMG may be considered as a surgical alternative in patients with keratoprosthesis complicated by severe infectious keratitis refractory to standard medical treatment.

1. Introduction

Keratoprosthesis (Kpro) has been employed as a therapeutic alternative for various corneal diseases causing blindness that cannot be addressed through conventional transplant methods.¹ Corneal melts and infectious keratitis are regarded as the most common complications following retrosprosthetic membrane (RPM) and glaucoma.² Infectious keratitis threatens the retention of the Kpro and exposes the patient to the risk of endophthalmitis, resulting in potential visual impairment.³ Prompt and proactive intervention is crucial to mitigate these risks. Risk factors for infectious keratitis include contact lens wear, ocular trauma, ocular surface diseases, lid diseases, corneal surgeries, systemic disease,4, 5, 6 and prolonged use of topical steroids, prevalent in Kpro patients.⁷ Long-term steroid use is weighed against the higher risk of complications related to ocular surface inflammation.⁸

Rose Bengal Photodynamic Antimicrobial Therapy (RB-PDAT) has been recognized for inhibiting in vitro growth of different organisms, including Methicillin-Resistant Staphylococcus aureus (MRSA),⁹ Fusarium solani¹⁰, Aspergillus fumigatus, Candida albicans¹¹, Pseudomonas aeruginosa¹² and Acanthamoeba. Positive outcomes have been reported in infectious keratitis caused by those agents, establishing it as a feasible and efficacious alternative, in those not responding to standard medical treatment.

The amniotic membrane (AM) is the placenta's inner layer; promotes cell growth and adhesion and reduces vascularization,^13,14 it serves as physical protection for the ocular surface.¹⁵ In refractory or persistent infections, it has proven effective as an adjuvant treatment.^15,16 Given our experience with both techniques, we opted to combine them to enhance the response to this critical case. This case describes a novel surgical technique used in a patient with severe infectious keratitis, combining RB-PDAT with an AM graft.

2. Case report

A 91-year-old woman with a history of severe glaucoma, multiple failed grafts and B-Kpro aphakic Type I Lucia design, implantation in both eyes (BE), presented to routine follow-up at the cornea service, two years after the Kpro implantation in the left eye (LE), complaining of pain and redness that started three days before her appointment. Her treatment consisted of topical fluorometholone 0.1 % twice daily (BID), timolol 0.5 %, brimonidine 0.2 %, dorzolamide 2 % BID, and moxifloxacin 0.5 % BID in BE. At presentation, the LE had conjunctival hyperemia, ciliary injection, and a 6 mm corneal infiltrate with an epithelial defect of the same size adjacent to the optic stem (Fig. 1). The anterior chamber and vitreous were quiet. Best corrected visual acuity (BCVA) was 20/200 in the right eye (RE) and 20/60 in the LE. A bandage contact lens that she was wearing was removed, fluorometholone 0.1 % was suspended and treatment was initiated with hourly fortified topical vancomycin 5 % and moxifloxacin 0.5 %. The rest remained the same and nothing was modified for the RE.

Fig. 1.

A, B. Clinical photograph of the LE showing conjunctival hyperemia and a corneal infiltrate adjacent to the KPro optic stem.

Corneal scraping revealed gram-negative bacilli, and after seven days the cultures demonstrated significant growth of P. aeruginosa sensitive to amikacin, gentamicin, moxifloxacin, and ceftazidime. Despite compliance of the treatment for two weeks, the clinical characteristics had worsened, there was a deterioration in vision, enlargement of the infiltrate, an increase in the epithelial defect, and the onset of melting.

Given continued worsening and the lack of response to topical broad-spectrum antimicrobial therapy during three weeks, a decision was made to perform RB-PDAT in conjunction with an AM graft due to the corneal lysis and thinning. Same medical treatment was continued. On day 21 after RB-PDAT, the epithelial defect and infiltrate had resolved (Fig. 2), and the patient reported complete resolution of pain. At this point, we reduced the frequency of topical antibiotics to four times daily (QID) and initiated topical corticosteroids QID (fluorometholone 0.1 %). Subsequently, Vancomycin was discontinued, and treatment was maintained with moxifloxacin 0.5 % BID alone. At the last follow-up visit, 6 months after clinical resolution, there were no signs of inflammation or infection, and a BCVA of 20/60 was achieved.

Fig. 2.

A, B. Postoperative (72 hours) clinical photograph showing the amniotic membrane graft in place. C. 1 week after RB-PDAT and AM graft showing improvement in the corneal infiltrate. D. A quiet ocular surface and resolution of the corneal infiltrate 3 weeks after the procedure.

2.1. Surgical technique

RB-PDAT was performed on day 25 after symptoms onset. The procedure was performed as described previously by Naranjo et al.¹⁷ by placing a lid speculum and applying topical anesthesia (tetracaine 0.5 %). The corneal epithelium was debrided within the area of the infiltrate. Then a sponge was soaked with 0.1 % rose bengal solution and placed over the de-epithelized cornea, followed by one drop every 3 minutes for 30 minutes. The sponge was then removed and the ocular surface was flushed with a balanced saline solution. Afterwards, the anterior corneal surface was irradiated with a 6 mW/cm2 custom-made green LED light source for 15 minutes to achieve a total energy exposure of 5.4 j/cm2. After RB-PDAT was completed, a 2 × 1 cm AMG was placed over the cornea with the epithelial side up, ensuring it remained well-positioned over the optical zone. The AMG was then secured with 10–0 nylon interrupted sutures placed on the cornea. Moxifloxacin 0.5 % eye drops were applied at the end of the procedure and the patient was told to continue with her standard medical treatment.

3. Discussion

Infectious keratitis exhibits its highest incidence in the two years following Kpro implantation.⁷ It poses a significant risk as it can lead to severe complications, including the potential development of endophthalmitis⁷. Reported incidences of infectious keratitis after Kpro implantation range from 3.2 % to 17.8 %.^2,7,18,19 One of the main risk factor for bacterial keratitis after Kpro surgery is failure to follow postoperative topical prophylactic antimicrobial regimens, eyelid abnormalities, compromised ocular surfaces, tear film alteration, prolonged use of contact lenses, long-term use of topical steroids, and increased antibiotic resistance caused by changes in the microbiota of the ocular surface.^7,20, 21, 22 Studies have indicated a heightened susceptibility to fungal and gram-positive bacterial keratitis in individuals undergoing prolonged vancomycin use.^7,18,23 Various bacterial strains are known as common agents of infectious keratitis in patients with KPro. In our case, Pseudomonas aeruginosa was the causative agent of the infection. A previous study identified it as the predominant cause of gram-negative endophthalmitis associated with BKPro implantation.²⁴

Clinical manifestations associated with infectious keratitis often include the formation of opacity or infiltration beneath the anterior plate of the Kpro along with an epithelial defect and corneal melting proximal to the optical cylinder,¹⁸ as presented in our patient.

Broad-spectrum antibiotics are recommended as prophylaxis against infectious keratitis in patients with Kpro.^6,25 In our practice, we choose moxifloxacin and/or vancomycin based on individual cases, additionally, a bandage contact lens is used, and each time it is replaced in the clinic, a single drop of 5 % iodopovidone is applied.

RB-PDAT introduces an alternative antimicrobial strategy, making it a potential consideration even in severe and refractory cases. Extensive effectiveness against various bacterial and fungal species has been documented and substantiated. Studies consistently show favorable outcomes against specific microbial agents,^9,11,12 with the heightened inhibition attributed to the increased singlet oxygen production by rose bengal when excited with green light.^26,27 RB-PDAT has several advantages as an alternative treatment, including the capacity to eliminate Pseudomonas species, oxidize virulence factors, enhance the tensile strength of the cornea, and remodel the extracellular matrix. This enhancement of corneal tensile strength proves beneficial in interrupting the corneal melting process that can occur during infection, thereby providing additional time for antimicrobial drugs to take effect.^12,17,28, 29, 30, 31

In addition to RB-PDAT, another option for treating severe infectious keratitis is photoactivated chromophore corneal cross-linking (PACK-CXL). Said et al.³² demonstrated that PACK-CXL effectively reduced the risk of severe complications, such as corneal perforation, in cases of advanced keratitis with corneal melting. Although PACK-CXL did not significantly reduce the time to healing compared to standard antimicrobial therapy, it could serve as a valuable additional treatment. Hafezi et al.³³ found that PACK-CXL was as effective as antimicrobial treatment in resolving early to moderate bacterial and fungal keratitis, regardless of the pathogen involved, making it particularly useful in cases of antimicrobial resistance. These findings support the potential use of PACK-CXL as a standalone and adjunctive treatment for severe or refractory infectious keratitis, expanding the available therapeutic options for challenging cases.

Despite PACK-CXL's effectiveness, it is important to note that studies have shown RB-PDAT to be more effective in vitro against several microorganisms compared to riboflavin-mediated PACK-CXL. For instance, RB-PDAT has demonstrated greater inhibition of bacterial strains such as Pseudomonas aeruginosa, with up to 93.9 % inhibition when combined with green light irradiation, compared to riboflavin's moderate effect.¹² The decision to select RB-PDAT over PACK-CXL was primarily based on its superior efficacy against Pseudomonas species and our extensive experience in performing the RB-PDAT procedure at our practice.

Previous studies suggest that RB-PDAT could serve as an adjunct therapy to treat corneal infections. In a pilot clinical study, Naranjo et al.¹⁷ reported a success rate of 72 %, which later increased to 77.4 % in an intermediate-term clinical outcomes study,³⁴ both without reported adverse effects. In both studies, infectious keratitis was caused by different microorganisms including Acanthamoeba, Fusarium, Curvularia, and Pseudomonas, although none involved patients with keratoprosthesis. The average time to clinical resolution exhibited variability, the first study reported a resolution time of 46.9 ± 26.4 days after RB-PDAT, while the second study indicated a time frame of 2.72 ± 1.85 months. In our case, the patient achieved clinical resolution 21 days after RB-PDAT.

The application of AM in ophthalmology has been thoroughly investigated, revealing its efficacy as an exceptional biological substrate. It is recognized for its ability to minimize water loss and apoptosis, harbor various growth factors, support the preservation of typical epithelial cell morphology, accelerate wound healing, and promote new cell growth.¹⁶ Moreover, the impact of surface material proves advantageous in cases of severe infectious keratitis, aiming to halt the progression of corneal melting and prevent perforation.³⁵ Additionally, the AM functions as an adjunctive tissue in the region between the corneal melt and the optical cylinder, aiding in the improvement of surface irregularities caused by the melt. These irregularities can lead to a poor distribution of the tear film, which deteriorates corneal healing, resulting in further thinning and subsequent perforation.

We acknowledge that performing two treatments simultaneously may create uncertainty about which intervention was primarily responsible for the success. However, an infection in a patient with a keratoprosthesis can lead to severe outcomes, so prompt and aggressive treatments are warranted. In this case, we believe both RB-PDAT and the AMG contributed to the therapeutic success. RB-PDAT likely played a role in halting the infection, while the AM graft helped reduce inflammation, minimize corneal lysis, and promote epithelialization.

The potential risk for devastating consequences in Kpro implantations requires regular monitoring, including searching for signs of infection in each visit. Prompt treatment should be initiated if needed to prevent severe complications. This case report documents a positive outcome using RB-PDAT in a patient with Kpro and infectious keratitis. RB-PDAT presents a promising and effective alternative for treating patients with keratoprosthesis who encounter infectious keratitis accompanied by corneal melting.

CRediT authorship contribution statement

Ruth Eskenazi-Betech: Writing – review & editing, Writing – original draft, Visualization, Project administration, Investigation, Data curation, Conceptualization. Guillermo Raul Vera-Duarte: Writing – review & editing, Writing – original draft, Methodology, Investigation, Conceptualization. Ana P. Murillo-López: Writing – original draft, Project administration, Methodology, Investigation, Conceptualization. Aldo Hernandez-Hernandez: Writing – original draft, Visualization, Investigation, Conceptualization. Arturo Ramirez-Miranda: Writing – review & editing, Writing – original draft, Visualization, Validation, Supervision. Alejandro Navas: Writing – review & editing, Writing – original draft, Visualization, Supervision, Data curation, Conceptualization. Enrique O. Graue-Hernandez: Writing – review & editing, Visualization, Validation, Supervision, Methodology, Investigation, Data curation, Conceptualization.

Patient consent

The patient consented to the publication of this case in writing.

Authorship

All authors attest that they meet the current ICMJE criteria for authorship.

Funding

This research received no external funding.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.