Risk factors and clinical outcomes of bacterial and fungal scleritis at a tertiary eye care hospital.

PURPOSE: The aim was to analyze demographics, risk factors, pathogenic organisms, and clinical outcome in cases with microbiologically proven bacterial or fungal scleritis.
MATERIALS AND METHODS: Retrospective review of all the medical records of patients with microbiologically proven infectious scleritis examined from March 2005 to December 2009 in the cornea services of L. V. Prasad Eye Institute, Hyderabad, India was done.
RESULTS: Forty-two eyes of 42 patients were included in this study. The mean age at presentation was 48.52 ± 14.10 years (range: 12-70). Surgery was the major risk factor seen in 24 eyes (58.5%). Scleral infection was noted after vitreoretinal surgery (with scleral buckle) in 15 eyes, cataract surgery in 3 eyes, pterygium surgery in 3 eyes, corneoscleral tear repair and scleral buckle surgery in 3 eyes. Sixteen eyes (39%) were on systemic or topical steroids at the time of presentation. History of injury was noted in 9 eyes (22%) and diabetes mellitus in 7 patients (17%). Associated keratitis was noted in 9 eyes (21.4%). The scleral abscess was unifocal in 33 eyes (78.5%), multifocal in 6 eyes (14.2%) and diffuse in 3 eyes (7.14%). The final follow-up ranged from 24 days to 37 months. The final visual acuity was better in 18 eyes (42.8%), stable in 13 (30.9%), and deteriorated in 7 eyes (16.6%). Recurrence was seen in 4 eyes (9.5%).
CONCLUSIONS: Surgery is a major risk factor for infectious scleritis in our series. Fungus was the most common organism isolated. Thorough debridement and intensive use of medications have improved the outcome.

INTRODUCTION

Inflammation of the sclera (scleritis) can be idiopathic or a manifestation of systemic immune-mediated disease, or less commonly due to infection. Prevalence of infectious scleritis varies between studies, ranging from 4.6%1 to 18%.2

Infectious scleritis can occur by secondary spread from an adjacent keratitis,3 severe infectious endophthalmitis,3 or primary scleral infection. Predisposing factors include systemic infection,14 systemic diseases,5 trauma,6 use of corticosteroid preparations,6 use of immunomodulators,7 previous scleritis8 or previous ocular surgery. Infectious scleritis was commonly reported after pterygium surgery associated with ocular irradiation or use of antimetabolites.9 Other surgeries include cataract,10 strabismus,11 glaucoma filtration surgery,12 suture removal,2 and vitreo-retinal procedures.13 Pseudomonas aeruginosa was the most commonly reported organism.14 Other organisms include Staphylococcus species,25 Streptococcus species,15 hemophilus influenza,8 Stenotrophomonas maltophilia,16 Serratia marcescens,7 Mycobacterium species,17 Nocardia,6 fungus,614 and virus.18

Diagnosis and management of infectious scleritis remain a challenge. Diagnosis is often masked by the close resemblance to immune-mediated scleritis clinically and therefore, there is often delay in the diagnosis. In suspected cases, microbiological evaluation of specimens obtained by scleral scraping in eyes with ulcerated lesions should be performed. Various modalities of treatment have been tried with variable success rates, which include topical medications alone,6 topical combined with systemic medications,19 antibiotic lavage,20 cryotherapy,21 amniotic membrane grafting,22 corneoscleral graft3 and finally enucleation or evisceration in nonreponsive or severe cases.323 The outcomes reported in the existing literature show varied clinical outcomes probably due to nonuniform clinical approach and practice.671424 Based on existing literature and our clinical experience we adopted to debridement (diagnostic and therapeutic) with intensive medications in all the cases, which led to overall good visual and complete anatomical outcomes. To the best of our knowledge, this is the largest series of bacterial and fungal scleritis showing good anatomical and functional outcomes. We hereby report a series of 42 patients with bacterial and fungal scleritis with various risk factors, microbiological profile, and clinical outcomes.

MATERIALS AND METHODS

We reviewed the medical records of all consecutive patients with microbiologically proven infectious scleritis presenting to the Cornea and Anterior Segment Service at the L. V. Prasad Eye Institute, India from March 2005 to December 2009. The study was approved by the Institutional Review Board of L. V. Prasad Eye Institute, Hyderabad, India, and was conducted in adherence to the tenets of the declaration of Helsinki. A total of 42 eyes of 42 patients were included in the study. All the eyes where the clinician strongly suspected infection underwent scleral deroofing of nodular lesions or debridement with exploration in ulcerated lesions.

Procedure

Surgical steps: After peribulbar anesthesia,

Nodular lesions

Conjunctiva around the nodule was incised circumferentially to expose the underlying scleral lesion. An incision was made using no. 15 surgical blade on a Bard Parker handle (Aspen Surgical™ Products, Caledonia, MI) and the nodule was deroofed.

Ulcerated lesions

Conjunctiva around the nodule was incised circumferentially to expose the lesion. Debridement was done using a no. 15 surgical blade on a Bard Parker handle.

In both the scenarios, the wound was explored until clinically normal sclera was seen, and thorough debridement was done until the underlying uveal tissue was just barely visible with taking care not to perforate the globe.

Necrotic tissue was excised and collected for smear and cultures. The base and edges of the lesions were scraped for microbiology. Bare sclera was left exposed to aid penetration of postoperative antibiotics.

Microbiology

The samples were inoculated directly onto blood agar, chocolate agar, nonnutrient agar with Escherichia coli overlay, Sabouraud dextrose agar, potato dextrose agar, thioglycollate broth, and brain heart infusion broth. In eyes of scleritis associated with scleral buckle infection, the buckle and surrounding necrotic material were also inoculated in culture media. In all the eyes, a thorough exploration was done on the table to rule out the presence of multifocal lesions.

Surgical debridement was repeated in eyes if there was, (1) Development of new scleral abscess or (2) No microbiological diagnosis established in previous attempt or (3) Nssssso clinical response in spite of use of sensitive antibiotics.

Treatment

Immediate treatment for all the eyes was based on smear report. If bacteria were noted in smears and also in smear-negative case, broad spectrum antibiotics (fortified cefazolin 5% and ciprofloxacin 0.03% eye drops) were administered starting half-hourly. Oral ciprofloxacin 500-750 mg twice a day was also administered. The treatment was further modified based on the culture and sensitivity report, obtained 24-72 h later. In cases of identification of fungal filaments on smears, 5% natamycin was commenced every half-hourly along with oral ketoconazole (200 mg twice a day). Clinically resolution was defined as when the patient complained of no symptoms, and any sign of scleral inflammation disappeared on slit-lamp examination.

Information including patient's age, risk factors, onset and course of the disease, pathogenic organism, methods of diagnosis, treatment, and outcomes were recorded. In the case of negative smear and cultures and no clinical improvement even at 48 h, the patient underwent repeat scleral scrapings for microbiology.

RESULTS

During the study period, 235 patients presented with scleritis, of which 42 patients were diagnosed to have an infection (17.87%). Demographic characteristics [Table 1]: Forty-two eyes of 42 patients (34 males and 8 females) were included in the study. The mean age at presentation was 48.52 ± 14.10 years (range: 12-70). Right eye was involved in 23 eyes and left eye in 19 eyes. The final follow-up ranged from 24 days to 37 months. Three patients were lost to follow-up, 2 patients before there was complete resolution of the primary infection and one patient after recurrence.

Table 1

Clinic characteristics, management and outcomes all the patients

Risk factors

The risk factors noted were previous surgery, use of topical and systemic steroids, injury and systemic illness like diabetes mellitus [Table 2]. Surgery was the major risk factor seen in 24 eyes (58.5%). All of these were referred cases, and the surgeries were performed in other institutes and hence the details were not available in the medical records. Forty percent of the patients with fungal scleritis were on either topical or systemic steroids or both at presentation. Eighty-five percent of patients with Staphylococcus species, 60% of patients with Streptococcus species and 40% of patients with Pseudomonas scleritis had surgery as the predisposing factor for the development of scleritis. Trauma was the predisposing factor in 50% of patients with Nocardia scleritis.

Table 2

Risk factors associated with bacterial and fungal scleritis

Clinical characteristics

The mean duration of symptoms before presentation was 31.79 ± 34.52 days (range: 2-180). The mean duration of the presentation was longer (54 days) after developing symptoms in patients with fungal scleritis compared with other organisms. Keratitis was noted in 9 eyes (21.42%). Scleral abscess was unifocal in 33 eyes (78.57%), multifocal in 6 eyes (14.28%) and diffuse in 3 eyes (7.14%).

Microbiological profile, management and clinical outcomes

Fungus

Fungus was the etiological agent in 10 eyes (23.8%) in our series. Nine of 10 had risk factors which included surgery in 3 eyes, injury in 3 and use of topical corticosteroids in 6 eyes at the time of presentation. All were treated with topical 5% natamycin and systemic ketoconazole. Topical itraconazole was added in eyes with recurrence (3 eyes). Amphotericin B (0.015%) wash was done in one eye at the time of therapeutic debridement. Multiple debridements were required in fungal scleritis for debulking the infection and also to improve penetration of medications. Mean duration of treatment was 92 days (range: 20 days to 1-year). After initial resolution, two eyes presented with recurrence of fungal scleritis. One patient that was lost to follow-up before complete resolution, [Figure 1].

Figure 1

Clinical photograph of right eye of a patient with pus point at 3O’clock position which revealed fungal etiology on scleral deroofing (a), at 3 weeks after scleral de-roofing and intensive antifungal medications (b), histopathology specimen of the scleral tissue stained with Gomori's methanamine silver showing fungal filaments (c, white stars depicting fungal filaments), clinical photograph with a well epithelized surface, senile cataract and no evidence of infection (d)

Bacterial

Staphylococcus species was isolated in 7 eyes (16.6%). Staphylococcus aureus was seen in 5 eyes, Staphylococcus epidermidis in one eye and species was not identified in one eye. Mean duration of treatment was 29.43 ± 6.73 days (range: 15-36). Complete resolution was noted in all. One eye that had no perception of light at presentation became phthisical

Pseudomonas aeruginosa was the cause of scleritis in 6 eyes (14.2%). Predisposing factor was surgery in 3 eyes (scleral buckling - 2 eyes, cataract surgery - 1 eye) and trauma in 2 eyes. One patient (patient-18) presented with postcataract surgery suspected endophthalmitis for which he underwent pars plana vitrectomy and vitreous biopsy, which did not yield any organism. One month later the patient developed scleral abscess at the site of cataract surgery incision, which was debrided and treated with topical and systemic ceftazidime, which was the only drug for which the organism was sensitive. In the other 5 cases, the organism was sensitive to ciprofloxacin and showed complete resolution with topical and systemic ciprofloxacin. Topical steroids were used in two eyes along with the antibiotics. Mean duration of treatment was 47.67 ± 30.44 days (range: 20-90). One eye underwent multilayered amniotic grafting for persistent epithelial defect. Poor visual acuity was seen in 2 eyes due to corneal scar [Figure 2]

Figure 2

Clinical photograph of the left eye of a patient showing Pseudomonas aeuroginosa corneal infiltrate and scleral abscess (a), at 1-week of scleral deroofing showing uveal show and a resolving corneal infiltrate (b), histopathology specimen of scleral tissue after H and E stain showing inflammatory cells and Gram-negative bacilli (c), clinical photograph of the patient with well epithelised surface, scleral thinning with uveal show, corneal scar is also seen (d). There is no evidence of infection

Nocardia asteroids were the etiological agent in 6 eyes (14.2%). Three eyes were on topical or systemic steroids at presentation. All eyes were treated with topical amikacin 2.5%, five eyes were treated with systemic amikacin for a period of 7-14 days depending on the response. Mean duration of treatment was 47.33 ± 21.76 days (range: 13-71). Surgical debridement was done 1-4 times depending on the response. Complete resolution of infection was seen in 5 eyes. Recurrence was seen in one eye that subsequently was lost to follow-up

Mycobacterium chelonae was isolated in 5 eyes (11.90%). The patients were treated with either only topical amikacin (3 eyes) or combined topical and systemic amikacin (1 eye). All the eyes were sensitive to amikacin. Complete resolution was seen in 3 eyes, and one patient was lost to follow-up. One eye developed secondary glaucoma that was subsequently controlled by topical and systemic antiglaucoma medications. One patient had a mixed infection with S. epidermidis, which could be a contaminant

Streptococcus species was isolated in 4 eyes (9.52%). The eye with history of injury underwent foreign body removal and application of tissue adhesive due to perforation at the limbus. Resolution of scleritis was seen in all the eyes with duration of treatment ranging from 16 to 50 days. One eye that presented with visual acuity of only accurate projection of rays after sclera buckling surgery had become phthisical

Corynebacterium diphtheriae was isolated in 2 eyes (4.76%). One patient underwent debridement, multilayered amniotic membrane transplantation and conjunctival autografting. Complete resolution was seen in 40 days in one patient and 30 days in the other

Brevibacterium species was isolated in one patient on examination buckle was exposed with adjoining scleral necrosis. Scleritis was successfully treated with buckle explantation, debridement and topical cephazolin (sensitive) eye drops for 15 days.

Visual outcomes

There was improvement in best spectacle-corrected visual acuity in 18 eyes (42.85%), visual acuity was unchanged in 13 eyes (30.9%), and deteriorated in 7 eyes (16.66%). At final follow-up 20/40 or better visual acuity was seen in 17 eyes (40.5%) and 20/50-20/200 in 9 eyes (21.5%) and worse than 20/200 in 16 eyes (38%), 14 of these 16 eyes had visual acuity of worse than 20/200 at presentation itself.

Complications

Recurrence was seen in 4 eyes (9.52%), of which 3 were seen in patients with resolved fungal scleritis. In 2 of these eyes recurrence was with the same organism but in one eye it was due to Brevibacterium species, which was treated successfully. Other complications included cataract in 3 eyes, secondary glaucoma in one and persistent epithelial defect in one eye.

DISCUSSION

The proportion of infectious scleritis (17.87%) in our study is a little higher than previous reports.6 In this series, majority of the cases had predisposing factors such as surgery, trauma and corticosteroid use. Surgery is said to be most common predisposing factor for infectious scleritis, which was also noted in the present series (58.5%). vitreoretinal surgery (43%, 18/42) was the major predisposing factor followed by cataract (7%, 3/42) and pterygium (7%, 3/42). Infectious scleritis had been reported either after vitrectomy or scleral buckling surgery.2425 In the scleral buckle infections, the patients, usually, present with chronic irritation and discomfort. The buckle is often colonized by microorganisms with minimal scleral necrosis. However in the present series, all patients had scleral necrosis extending beyond the scleral buckle which resolved after buckle explantation, debridement and appropriate antimicrobials, and therefore is not simple colonization.

Smiddy et al.26 have shown that Staphylococcus species accounted for most of the culture positive buckle infections and hypothesized that biofilm enveloping the exoplanet hinders the penetration of antibiotics and effectively sequestrating a reservoir of potential pathogenic microbes.

Though fungal scleritis is less common in western literature,27 it is the most common cause of scleritis in tropical countries like India.6 Incidence of fungal scleritis varied between different studies, 3-38%.628 Of all the fungi, scleritis is commonly reported with Aspergillus, other fungi include Metarrhizium anisopliae, Paecilomyces lilacinus, Scedosporium, Cephalosporium, Penicillium, Cladosporium, Candida parapsilosis, Petriellidium boydii, Fusarium and Rhizopus Acremonium.6293031 We also report Colletotrichum dematium and Aspergillus tereus that have not been reported so far as a cause of infectious scleritis.

Treating fungal scleritis is challenging due to the difficulty of diagnosis, the difficulty of drug penetration due to avascular nature of sclera and the limited number of antifungal agents. Existing literature does not provide any consensus on the class of antifungal or the route of administration to be followed. Moriarty et al.9 reported fungal hyphae in enucleated specimens from two patients in spite of aggressive topical and systemic antifungal therapy for 3 and 6 months, respectively. Surgical debridement helps debulk the infected deep stromal layers and also promotes penetration of antimicrobial agents into the affected tissue. It was hypothesized that recurring scleral abscesses are due to an immunologic response to fungal cell death but we have seen three recurrences in our series which were culture positive. Previous literature has shown poor outcome of fungal scleritis61429 In our case series of 10, nine cases were successfully treated with aggressive debridement both as diagnostic and therapeutic modality and prolonged topical and systemic antifungals. Thorough debridement helps in debulking the organism thus dislodging the nidus and also increasing the drug penetration. In our series, patients with fungal sceritis required multiple debridements.

Pseudomonas is the most common and fulminant organism causing infectious scleritis in the western literature.27 Due to poor success rate with medical therapy alone, clinicians resorted to other modalities such as subpalpebral irrigation with antibiotics, carbon dioxide laser, conjunctival recession and cryotherapy with variable success rates.332 In our series complete resolution was seen in all the six cases with surgical debridement, topical and systemic antibiotics. Topical steroids along with antibiotics were used in two cases. Role of corticosteroids has been controversial in infectious scleritis. Pseudomonas is known to exacerbate the issue destruction by upregulating multiple host cytokines and chemokines thus leading to worsening or recurrence by decreasing the ocular defense mechanism and increase the lytic enzyme activity.1933 Along with appropriate antibiotics steroids help in decreasing the proteolytic enzymes by the microorganisms, polymorphonuclear leucocytes, corneal epithelial cells and keratocytes.34 Huang et al.35 and Ma et al.22 have successfully treated 3 and 4 cases respectively using low-dose oral corticosteroids and postulated that low dose oral steroids eliminates the direct contact effect of topical steroids and reduces the likelihood of exacerbation or late recurrence of scleritis. A delicate balance exists between the induction of a protective and a destructive response to ocular infection. Steroids must be used only under close observation as while tissue destruction is reduced, host immune response to the infection will also be suppressed.36

Apart from routine microbiological evaluation specific emphasis should be laid in the diagnosis and management of fastidious organisms like nocardia. In our series, all the 6 cases were sensitive to amikacin. Though trimethoprim-sulfamethoxazole and amikacin are effective drugs in the management of nocardia scleritis, topical and systemic amikacin with thorough debridement may be the treatment of choice in the successful management of nocardia scleritis.

Brevibacterium species are gram-positive rods present in dairy products and human skin. It is a rare cause of ocular infection. Ghosheh et al. have reported a case of keratitis with Brevibacterium casei in a patient with aerosolized crack cocaine Use which showed complete resolution with sensitive antibiotics.37 Till date there are no cases of infectious scleritis due to Brevibacterium reported. We have seen two cases in our series. One case had scleral buckling as a risk factor, and the other case was primarily fungal scleritis, which was successfully treated with topical and systemic antifungal medications for 60 days. Twenty-two days after cessation of antifungals patient presented with recurrence, but on debridement and subsequent culture showed Brevibacterium species. Both cases were successfully treated with topical cefazolin and oral ciprofloxacin was used in one case.

Based on the outcomes reported in the existing literature and experience of treating previous cases we adopted to debridement with intensive medications in all the cases which have led to overall good visual and complete anatomical outcomes. Useful vision (better than 20/200) was seen in 26 cases (61.09%). Most of the cases with poor visual acuity were due to previous vitreo-retinal surgery, corneal scar or cataract. None of the eyes in this series required either enucleation or evisceration. This data may not be representative of the prevalence of bacterial and fungal scleritis because this was a retrospective study from a tertiary eye care hospital.

CONCLUSIONS

Bacterial and fungal scleritis is associated with a predisposing factor in the majority of cases. Pterygium surgery is the most common predisposing factor in western countries that may not be true always in our scenario. Fungi were the most common organism seen in our series, but outcome was good, contrary to poor results demonstrated in previous series. Thorough debridement, which may have to be repeated, can be both diagnostic and therapeutic, leaving bare sclera for penetration of antibiotics; and the intensive use of antimicrobials helped in eliminating the microbes and thus improving final outcome.