By Richard R. Watkins, MD, MS, FACP, FIDSA

Associate Professor of Internal Medicine, Northeast Ohio Medical University; Division of Infectious Diseases, Cleveland Clinic Akron General, Akron, OH

Dr. Watkins reports that he has received research support from Allergan.

SYNOPSIS: A retrospective cohort study found that endogenous fungal endophthalmitis is associated with intravenous drug use and frequently results in poor visual outcomes despite appropriate surgical and antifungal therapy.

SOURCE: Tirpack AR, Duker JS, Baumal CR. An outbreak of endogenous fungal endophthalmitis among intravenous drug abusers in New England. JAMA Ophthalmol 2017;135:534-540.

Intravenous drug users (IVDUs) have an increased risk for a number of serious infectious disease complications, including infective endocarditis, osteomyelitis, HIV, and HCV. Less frequently, IVDUs can develop endogenous fungal endophthalmitis (EFE) through transient fungemia with dissemination to the eye. Tirpack and colleagues noticed an increase in cases of EFE at their institution (New England Eye Center, Boston) in IVDUs and sought to determine patient characteristics, management, and outcomes currently associated with EFE.

The study was retrospective and included IVDUs who had fundus findings characteristic of EFE, along with positive fungal ocular and/or blood cultures. Patients with exogenous endophthalmitis, such as post-surgical or from trauma, were excluded. The presenting visual acuity (VA) was assessed and the main outcome measured was the final best-corrected VA. All patients with EFE who presented to the investigators’ institution between May 2014 and May 2016 were screened for inclusion in the study.

Ten patients were included in the study, all of whom had unilateral eye involvement. Their mean age was 34 years (range, 24-60 years), and five patients were female. Nine patients lacked systemic signs or symptoms of infection, and one reported fevers and chills. The most common presenting symptom was that of floaters in the eye (n = 8), followed by reduced vision (n = 6), eye pain (n = 5), and photophobia (n = 3). The presenting VA ranged from 20/25 to detection of hand motions only. Dilated fundus examinations showed vitreous inflammation in all the eyes, with four having the classically described “string of pearls” formation. Nine of the patients were admitted to the hospital (one refused admission) and underwent intravitreal aspiration for cultures, followed by injection of an antifungal agent, with seven receiving voriconazole and two receiving amphotericin B. All patients received systemic antifungal therapy as well, with eight prescribed intravenous voriconazole and one prescribed oral voriconazole and oral fluconazole. The reason for the oral therapy was not provided. All patients were discharged home with oral antifungal therapy for a mean duration of 55 days (range, 35-76 days). Despite this, five patients developed worsening vitreitis and/or VA and required pars plana vitrectomy (PPV). The mean time to PPV was 18 days (range, 7 to 45 days). Cultures, including vitreous fluid and blood, were positive in only four cases and included Candida albicans (n = 2), Candida tropicalis (n = 1), and Candida dubliniensis (n = 1). The final VA ranged from 20/40 to 20/300, with 20/70 as the median final VA in the five patients who underwent PPV and 20/300 in those who did not have the procedure. All of the patients were eventually non-adherent to follow-up.


The diagnosis of EFE is made by a history of acute or progressive vision loss, characteristic physical examination findings, and laboratory confirmation (i.e., culture results). However, the diagnosis can be challenging, especially in IVDUs who may present with other signs or symptoms of embolic disease such that the eye symptoms are overlooked. Moreover, patients experiencing mental status changes might not be able to communicate changes in their vision. As the study by Tirpack et al highlights, a dilated funduscopic examination is a crucial step when the diagnosis of EFE is being considered. The low diagnostic yield of the cultures (only 4/10 were positive for Candida) was not surprising and has been observed in previous studies. This is believed to occur because Candida preferentially sequesters within inflammatory nodules and the initial vitreous tap is of low volume and from the middle of the vitreous cavity, when most of the focal inflammation is closer to the retinal surface. Moreover, even though a larger volume of fluid is available at the time of PPV, cultures obtained during the procedure often are negative because the patient already would have been started on systemic antifungal therapy.

Amphotericin B used to be the standard therapy for EFE, although it has fallen out of favor because it does not reach therapeutic intraocular levels when administered systemically and frequently causes retinal toxicity when given by intravitreal administration. Voriconazole has superseded amphotericin B because it achieves better blood-retinal barrier penetration, resulting in high intraocular concentration, and is safe for intravitreal therapy.

An area of controversy in the management of EFE pertains to the role of early vitrectomy in patients with worsening inflammation despite local and systemic antifungal therapy. Prior studies have shown improved visual outcomes when PPV is performed earlier in the course of EFE, which may be due to the procedure clearing the infecting organism, removing inflammatory mediators, and allowing increased penetration of systemic antifungal therapy. However, these benefits must be weighed carefully against the increased risk for retinal detachment when surgery is performed on an acutely inflamed eye. The final median VA was higher in the patients who had PPV, which also was associated with earlier resolution of inflammation and infectious lesions.

It is notable that none of the patients had 20/20 vision at the end of the study period, emphasizing the serious consequences of EFE. Thus, clinicians must maintain a high index of suspicion for EFE in IVDUs who present with visual changes, which might be the only clue that a serious fungal infection is present.