Endophthalmitis in the Modern Era

A literature review and proposed treatment algorithm

LAITH M. KADASI, MD • GAURAV GUPTA, MD, MBA

Endophthalmitis (Figure 1) is purulent inflammation of the aqueous and vitreous with a poor visual prognosis. In the more than 20 years since the original Endophthalmitis Vitrectomy Study (EVS), the landscape of endophthalmitis risk factors and treatment options continues to evolve.

Figure 1. Leukocornea as a result of massive corneal edema and complete hypopyon in advanced Staphylococcus aureus endophthalmitis.

This review will discuss the current body of literature and modern techniques in the surgical management of endophthalmitis.

PSEUDOPHAKIC POSTOPERATIVE ENDOPHTHALMITIS

In 1995, the EVS evaluated the role for early pars plana vitrectomy in comparison with intraocular antibiotic injection and systemic antibiotics.¹ Patients who were included in the study were diagnosed with acute pseudophakic postoperative endophthalmitis, with vision ranging from 20/50 to light perception. No difference was noted in final visual outcomes among patients with visual acuity better than light perception receiving tap and inject therapy or vitrectomy.

Patients with light perception vision receiving core vitrectomy were found to be three times more likely to achieve a minimum of 20/40 vision with a 50% reduction in the risk of severe vision loss.

Although this study is often considered in the management of endophthalmitis cases, the EVS protocol did not include an analysis of acute endophthalmitis from causes other than cataract surgery, such as endogenous or traumatic endophthalmitis, glaucoma filtration surgery, chronic endophthalmitis, or cases associated with the increasing application of intravitreal injections in modern vitreoretinal treatment algorithms.

CHRONIC POSTOPERATIVE ENDOPHTHALMITIS

Patients presenting more than six months postoperatively following cataract surgery may have a delayed onset of postoperative inflammation with capsular plaque noted on examination. These patients are frequently found to have Propionibacterium acnes chronic bacterial endophthalmitis, and they often suffer reduced rates of recurrent intraocular inflammation if they receive prompt PPV, central capsulectomy, and intravitreal vancomycin injection.²

GLAUCOMA FILTRATION SURGERY

Bleb-associated endophthalmitis has been described with an estimated incidence of 0.2% to 1.5%.³ Treatment strategies in bleb-associated endophthalmitis frequently target Gram-positive pathogens, such as Staphylococcus and Streptococcal species. There are conflicting data regarding the optimal treatment strategy in these challenging patients.

In a 2002 retrospective analysis, 49 patients undergoing prompt PPV for bleb-associated endophthalmitis had significantly worse final visual outcomes when compared with a tap and inject approach.⁴ Conversely, a 2004 retrospective analysis of 68 patients undergoing prompt PPV had better VA at 12 months and significantly lower rates of no light perception vision.⁵

INTRAVITREAL INJECTION-ASSOCIATED ENDOPHTHALMITIS

A 2011 meta-analysis of more than 100,000 intravitreal injections of anti-VEGF medications described endophthalmitis in 0.049% of all cases. Although a significant proportion of these patients were found to be culture-negative, Streptococcal isolates among culture-positive patients were found to be far more common following anti-VEGF injection, compared with intraocular surgery.⁶

The use of topical antibiotics is considered the standard of care in the postoperative period for many ocular surgeries. However, there is no clear benefit to the use of prophylactic antibiotics in intravitreal injections. A 2014 retrospective case control study examining more than 117,000 intravitreal anti-VEGF injections found no significant reduction in the risk of endophthalmitis with prophylactic topical antibiotic therapy.⁷

Furthermore, the repeated use of topical fluoroquinolones after intravitreal injection was associated with an increased rate of bacterial resistance compared with control eyes, estimated at a 63.6% resistance rate to fluoroquinolones among study eyes, compared with 32.1% in eyes that did not undergo antibiotic prophylaxis.

In 2011, Shah et al performed a retrospective case-control study of endophthalmitis outcomes involving 27,736 patients treated with intravitreal antibiotic injections, and they found a 78% rate of return to baseline VA within six months, with 17% of cases requiring delayed PPV.⁸ There are only limited studies at this time exploring the efficacy of early PPV in the treatment of intravitreal injection-associated endophthalmitis.

OBTAINING A CULTURE

A 2010 retrospective case-control study of 229 patients with either endogenous or exogenous endophthalmitis with both anterior-chamber and vitreous culture isolates revealed poor concordance between anterior-chamber and vitreous-chamber culture results.⁹ The discrepancy between the anterior and vitreous chambers was similarly noted in the EVS, as well as the European Society of Cataract and Refractive Surgery Endophthalmitis Study.^1,10

Overall, the most reliable source for culture appears to be postprocessing specimens obtained from vitrectomy fluid. A 60-patient study by Banu et al in 2011 compared aqueous humor and vitreous humor by vitreous tap, vitreous biopsy, or PPV.

Aqueous humor was obtained by paracentesis using a 26- or 27-gauge half-inch needle mounted on a tuberculin syringe; vitreous tap was obtained with a 22- or 27-gauge needle with aspiration of 0.1-0.3 mL of vitreous humor.

Vitreous biopsy was obtained with vitrectomy aspiration cutter removal of 0.2-0.3 mL through a pars plana approach, and vitrectomy fluid was obtained by processing the dilute vitreous fluid obtained intraoperatively during PPV through centrifugation and analysis of the sediment, following removal of the supernatant fluid in each sample.

The study group found that the highest sample yield was obtained using analysis of the concentrate obtained from vitrectomy fluid, followed by the vitreous tap or biopsy, with the lowest sample reliability from the anterior-chamber paracentesis group.¹¹

It is also beneficial for the microbiology laboratory to assess the biopsy in growth media associated with the most frequent exogenous endophthalmitis organisms, including thioglycollate broth, brain-heart infusion broth, blood agar, MacConkey agar, chocolate agar, and Sabouraud agar.

In addition to culture media, there have been several recent advances in pathogen detection with novel techniques in polymerase chain reaction and matrix-assisted laser desorption ionization–time of flight (MALDI-TOF) mass spectrometry.^12,13 Increasing application of these techniques in the future may allow for improvements in the detection of both fungal and bacterial pathogens.

Vitrectomy Techniques

Pars plana vitrectomy plays a critical role in the management of a large subset of endophthalmitis cases. The approach allows for debridement of inflammatory material, which is highly toxic to the retina, with a greater volumetric specimen than vitreous tap techniques, as well as increased intraoperative visualization using wide-angle viewing systems for better and more complete visualization of the periphery allowing thorough vitrectomy.

Transconjunctival sutureless vitrectomy (TSV) is an emerging technique with increased efficiency, improved recovery time, and greater patient comfort than traditional 20-gauge vitrectomy surgery.^14-16 However, there have been only limited data directly investigating the efficacy of sutureless vitrectomy in endophthalmitis management.

In 2002, Hilton et al reviewed 217 cases of endophthalmitis treated with office-based 23-gauge PPV resulting in a complication rate of approximately 6% for all cases included in the study. In 2008, Tan et al assessed four cases of acute postoperative endophthalmitis and two cases of chronic postoperative endophthalmitis with 23-gauge TSV resulting in VA of 20/40 or better in five of the six cases, with a single outcome resulting in counting fingers vision.¹⁷

Similarly, in 2012, Almanjoumi et al assessed 10 patients with hand motion vision or worse presenting with acute postcataract or postfiltration surgery endophthalmitis, with 20/50 VA or better in eight of 10 patients, noting two cases with 20/400 or worse vision, as well as two cases complicated by postoperative retinal detachment.¹⁸

Although the original EVS supported core vitrectomy in endophthalmitis treatment, the use of complete and early vitrectomy, including intraoperative detachment of the posterior vitreous cortex and shaving of the vitreous skirt, was performed in 2006 by Kuhn and Gini in a 47-patient retrospective series with the finding of favorable outcomes with 91% having final VA of 20/40 or better, zero cases of RD, and one case requiring silicone oil.¹⁹

The use of silicone oil has been shown to have antimicrobial properties in vitro, and the application of silicone oil tamponade with PPV in retrospective analysis of severe endophthalmitis has shown an association with rapid control of infection and improved VA, compared with PPV alone.^20,21

The use of intraoperative dexamethasone injection is controversial, with a presumed approach of suppressing the inflammatory response and its destructive potential on the retina. A review of 57 patients receiving combined treatment with intravitreal steroid and intravitreal antibiotic injection found a significantly reduced probability of 3-line visual improvement, compared with intravitreal antibiotic therapy alone.

Although intraoperative administration of intravitreal dexamethasone is frequently used as an adjunctive therapy in PPV cases, a 63-patient assessment of postoperative and post-traumatic endophthalmitis similarly found no significant improvement in VA at 12 weeks, compared with PPV and intravitreal antibiotic therapy alone.²²

TREATMENT RECOMMENDATIONS

Provided with the current body of data, we are able to build a treatment algorithm outlining a suggested framework for an approach to endophthalmitis (Figure 2). Notably, when applying this algorithm to patient care, the overall clinical picture must be taken into account. This consideration includes the individual’s medical comorbidities and goals of care.

Figure 2. Endophthalmitis treatment algorithm.

The outlined approach includes recommendations for acute pseudophakic, chronic pseudophakic, bleb-associated, and intravitreal injection-associated endophthalmitis, as well as traumatic endophthalmitis.

CONCLUSIONS

Vitreoretinal surgeons in the modern era face the challenge of integrating EVS guidelines with the more recent innovations in intraocular surgery. Treatment of endophthalmitis requires consideration of not only the EVS, but also more recent studies exploring the role of intravitreal steroids, TSV, and complete and early vitrectomy. RP

REFERENCES