Panophthalmitis With Orbital Cellulitis Following Glaucoma Drainage Implant Surgery in an Adult Patient

Introduction

Chronic angle-closure glaucoma (CACG) is a complex and challenging condition in ophthalmology, characterized by progressive optic nerve damage and visual field loss [1]. Despite advancements in treatment options, effectively managing difficult cases with chronic angle-closure glaucoma (CACG) is still a major challenge in clinical practice. In this case report, we present the clinical course of a patient with uncontrollable CACG who underwent a glaucoma drainage implant procedure as a last resort for intraocular pressure control. Unfortunately, during the postoperative period, the patient developed panophthalmitis with orbital cellulitis, which is a rare and serious complication. By reviewing the patient’s symptoms, diagnostic tests, treatment options, and results, our goal is to clarify the difficulties faced in treating unexpected and devastating complications that can occur after a glaucoma drainage implant procedure.

Case Report

A 77-year-old man with uncontrolled chronic angle-closure glaucoma presented with longstanding vision loss in the left eye and progressive visual decline in the right eye. The right eye surgical history included extracapsular cataract extraction with posterior chamber intraocular lens implantation performed several years prior, reflecting surgical practice at the time and the presence of a dense cataract that was not amenable to phacoemulsification given the technology available at that time. Additional procedures included trabeculectomy augmented with mitomycin C, 5-fluorouracil needling, and selective laser trabeculoplasty performed 9 months prior to presentation. Despite treatment with multiple topical medications – brinzolamide 1%-brimonidine 0.2% combination twice daily, betaxolol 0.25% twice daily, bimatoprost 0.01% once nightly, and pilocarpine 2% 4 times daily – with questionable compliance, intraocular pressure (IOP) remained uncontrolled. Visual acuity in the right eye declined from 20/40 to 20/160, prompting the addition of oral acetazolamide 250 mg twice daily and the decision to go ahead with glaucoma surgery.

Inactive non-granulomatous uveitis was also found during the preoperative assessment. The uveitis unit performed comprehensive investigations to determine the etiology, including complete blood count, erythrocyte sedimentation rate, C-reactive protein, QuantiFERON TB, serum angiotensin-converting enzyme, syphilis serology, hepatitis C and hepatitis B antibodies, and chest X-rays to exclude sarcoidosis and other systemic causes. No specific systemic etiology for the inactive non-granulomatous uveitis was found despite the completion of proper investigations. The patient was started on topical prednisolone acetate 1% every hour 2 days prior to surgery to reduce the risk of postoperative inflammatory reactivation.

The primary surgical team performed uneventful Paul glaucoma implant surgery. On postoperative day 1, IOP was 21 mmHg without medication, rising to 23 mmHg during the first week, for which the surgeon added tafluprost 0.0015% once at night. The decision to use a prostaglandin analog was made considering its potent IOP-lowering effect; however, we acknowledge that prostaglandin analogs have been associated with increased intraocular inflammation in susceptible eyes, particularly those with pre-existing uveitis, the rapid clinical progression, vitreous involvement, and orbital extension observed in this case were inconsistent with prostaglandin-induced inflammation alone.

On postoperative day 12, he presented to the emergency department with pain, swelling of the upper and lower eyelids, decreased vision, and restricted ocular motility. Vision in the right eye had deteriorated to light perception. Clinical examination revealed conjunctival injection, severe eyelid edema and erythema, corneal edema, and +3 anterior chamber cells without hypopyon. The fundus was not visible due to dense vitritis. B-scan ultrasonography showed vitreous opacities, shallow choroidal detachment, ocular wall thickening, and optic disc cupping. These findings were consistent with acute endophthalmitis (Figure 1).

We admitted the patient for vitreous tap followed by intravitreal vancomycin (1.0 mg/0.1 mL), intravitreal ceftazidime (2.25 mg/0.1 mL), and intravitreal dexamethasone (0.4 mg/0.1 mL). Adjunctive therapy included subconjunctival vancomycin (25 mg/mL) and subconjunctival ceftazidime (50 mg/mL), in addition to alternating hourly fortified topical vancomycin (50 mg/mL) and fortified topical ceftazidime (50 mg/mL). Anti-inflammatory and cycloplegic therapy consisted of topical prednisolone acetate 1% every hour and topical atropine 1% twice daily, along with systemic therapy using oral moxifloxacin 400 mg once daily. Although orbital pain improved slightly, the patient declined further treatment, including repeat intravitreal injection or pars plana vitrectomy (PPV), and left against medical advice, with vision of light perception.

Three weeks later, he returned with persistent symptoms and consented to PPV with repeat intravitreal vancomycin (1.0 mg/0.1 mL), intravitreal ceftazidime (2.25 mg/0.1 mL), and intravitreal voriconazole (50 mcg/0.1 mL), along with subconjunctival cefazolin (50 mg/mL) and subconjunctival dexamethasone (2 mg/0.5 mL) administered near the implant plate. We prescribed systemic cefazolin 1 g every 8 h and gentamicin 5 mg/kg/day for 8 days, plus vancomycin 1 g every 8 h for 2 days, which we discontinued after negative microbiological cultures, partial clinical stabilization, and continuation of broad-spectrum local therapy. The short duration of vancomycin was based on the absence of gram-positive organisms in cultures, clinical improvement with alternative antibiotics, and the desire to minimize potential nephrotoxicity in this elderly patient.

Topical therapy during hospitalization included fortified topical cefazolin (50 mg/mL), fortified topical ceftazidime (50 mg/mL), topical prednisolone acetate 1%, and topical atropine 1%. Prior to discharge, the patient completed a 3-day course of oral amoxicillin 500 mg administered 3 times daily. At discharge, he was kept on oral amoxicillin in addition to tapering fortified topical cefazolin and ceftazidime, along with topical prednisolone acetate and atropine. A clinical photograph obtained at the time of discharge showed residual conjunctival injection and mild corneal haze (Figure 2).

The vitreous sample underwent Gram staining and bacterial and fungal cultures in the microbiology laboratory, all of which were negative. Computed tomography imaging of the brain and orbit was performed during the same admission (postoperative day 12), which revealed right proptosis, diffuse soft tissue thickening extending pre- and post-septally, irregular uveoscleral thickening, orbital fat inflammation, and mild swelling of the anterior extraocular muscles, most prominent in the lateral rectus, with no optic nerve involvement. These findings were consistent with panophthalmitis and orbital cellulitis (Figure 3).

On follow-up, orbital inflammation and eyelid edema had completely resolved. Visual acuity in the right eye, initially limited to light perception, improved to 20/250.

Discussion

This case highlights the importance of a multidisciplinary approach and prompt recognition and management of tube drainage implant procedure complications to minimize the impact on visual function and overall patient well-being. The 77-year-old patient in this case developed postoperative pain, swelling in both the upper and lower eyelids, and a decrease in vision in the right eye 12 days after surgery; therefore, we diagnosed him with postoperative panophthalmitis with orbital cellulitis. Nevertheless, patients with glaucoma have reported favorable outcomes with Paul implant surgery for management of intraocular pressure (IOP). A retrospective cohort analysis, with a 1-year follow-up, showed a qualified success rate of 75% and an absolute success rate of 33% [2]. Infrequently, individuals undergoing glaucoma drainage implant therapy have infections. Koh et al documented notable postoperative complications among 82 patients who underwent the Paul glaucoma implant procedure, including 1 instance of endophthalmitis leading to visual loss [3].

Another long-term study with a 2-year follow-up found no serious postoperative complications, except for 1 case that showed retinal detachment 1 month after surgery and was unrelated to Paul glaucoma implants [4]. Zheng et al reported a case of orbital cellulitis following use of glaucoma drainage devices, but it is more common following Ahmed valve implantation [5].

There are 2 categories of infections that occur after glaucoma implant: early (acute), occurring within the first 4 weeks, and delayed, appearing after 6 weeks. Duan et al [6] identified the microbial cause of infective postoperative endophthalmitis in patients with GDI-related endophthalmitis: gram-positive organisms like Staphylococcus as the predominant isolates and they also found Pseudomonas aeruginosa, Staphylococcus, Aspergillus niger, and Neisseria meningitidis [7]. In another study, Streptococcus species were the most frequent isolated organisms [8]. Esporcatte et al identified Haemophilus influenzae and Streptococcus pneumoniae as the most frequently isolated organisms after use of glaucoma drainage devices, especially in children, which are normal components of the conjunctiva’s bacterial flora [9].

We considered toxic anterior segment syndrome (TASS) in the differential diagnosis, particularly given the negative microbiological cultures. However, several clinical features made TASS less likely in this case. TASS typically presents within 12 to 48 h following surgery and is confined to the anterior segment. In contrast, our patient developed symptoms 12 days postoperatively, with prominent vitreous involvement and radiologic evidence of orbital cellulitis. Furthermore, the presence of vitreous opacities, choroidal detachment, and progressive orbital inflammation favored an infectious etiology rather than a sterile postoperative inflammatory reaction. The clinical course and partial response to antimicrobial therapy further supported the diagnosis of panophthalmitis rather than TASS. The negative cultures may have been due to prior antibiotic exposure, the fastidious nature of potential organisms, or the low bacterial load at the time of sampling.

Treatment options for endophthalmitis resulting from glaucoma drainage devices include either pars plana vitrectomy or vitreous injection of antibiotics [10]. The highly resistant nature of methicillin led to a disagreement over whether to start broad-spectrum antibiotic treatment [11]. In the present case, the surgeon used vancomycin but did not extract the plate due to significant inflammation, as the patient initially refused this choice. However, there is no strong evidence to suggest that explanting the glaucoma drainage devices at the diagnosis of endophthalmitis is necessary for infection control, as it could serve as a reservoir for bacteria. However, explanting the glaucoma drainage devices can lead to permanent vision loss due to uncontrolled IOP [12]. However, Islam et al [7] showed that removing the glaucoma drainage device with immediate vitrectomy was an anatomically effective way to treat endophthalmitis. Zheng et al [13] conducted a case-series study on 14 cases of GDI-related endophthalmitis; they used vitreous tap and antibiotic injection as first-line therapies, performed pars plana vitrectomy in 4 eyes, and explanted the only eroded GDIs in 9 eyes. However, there was no statistically significant difference in the VA change from baseline to final VA or in the VA change from the time of endophthalmitis diagnosis to final VA between the patients who had the glaucoma drainage devices explanted and those who did not [14].

Although prostaglandin analogs have been associated with increased intraocular inflammation in susceptible eyes, their role in precipitating severe intraocular infection is still unproven. In our case, the rapid clinical progression, vitreous involvement, and orbital extension were inconsistent with prostaglandin-induced inflammation alone. Visual outcomes following device removal are difficult to interpret because retrospective studies lack randomization between retention and explantation, making comparison of visual results unreliable. Furthermore, early pars plana vitrectomy (PPV) in cases of glaucoma drainage device-related endophthalmitis dramatically reduces rates of evisceration/enucleation while keeping equal rates of phthisis. Therefore, early vitrectomy may be helpful in preventing eventual loss of the eye [7].

Conclusions

Panophthalmitis with orbital cellulitis following glaucoma drainage implant surgery is an extremely rare, devastating, and sight-threatening complication. Additionally, this complication poses diagnostic challenges because of overlapping clinical features. We hope that by reporting this case, clinicians can gain insights into the clinical presentation, diagnostic workup, and differential diagnosis, which will enhance diagnostic accuracy in similar cases. Moreover, the management of such patients requires a multidisciplinary approach involving glaucoma, retina, and oculoplastic surgeons, and sometimes infectious disease specialists. This case report provides valuable information on treatment strategies, including antimicrobial therapy, surgical intervention, and postoperative care, which can guide clinicians in managing similar cases effectively.