Substance Use Disorder Interventions and Ophthalmic Outcomes for Injection Drug Use-Associated Endogenous...

Ophthalmologists who diagnose and treat injection drug use-associated endogenous endophthalmitis (IDU-EE) are on the frontlines of the ongoing opioid epidemic and can facilitate a multidisciplinary approach to the care of persons with substance use disorders (SUDs). To identify influential factors and opportunities for intervention, we examined recent SUD care and ophthalmic outcomes in the context of IDU-EE with attention to historical trends.

Retrospective cohort study SUBJECTS: Patients diagnosed with IDU-EE at a single, academic, tertiary-referral center from 2015 to 2021.

The clinical features, treatment, and outcomes of SUDs and IDU-EE were analyzed.

Addiction medicine outcomes: injection drug type, consultation by the addiction consult service (ACS), and initiation of medication(s) for opioid use disorder (MOUD). Ophthalmic outcomes: visual acuity (VA), ophthalmic intervention, ocular sequelae, and isolated organisms.

Sixty-two patients (65 eyes) with IDU-EE were included. Injected drugs included heroin (74.2%), cocaine (19.4%), and fentanyl (17.7%). Initial ophthalmic treatments were tap-and-inject (TAI) in 51 (78.5%) eyes and pars plana vitrectomy (PPV) with intravitreal antimicrobials in 9 (13.8%) eyes. Subsequent PPV was performed in 30 (46.2%) eyes. Intraocular cultures were positive in 34 (54.8%) eyes. Tap-and-inject was less likely to yield a positive culture (11/52) compared with PPV (26/39; P < 0.0001). Median VA improved from 1.30 (interquartile range [IQR]: 0.80-2.4) logarithm of the minimum angle of resolution (LogMAR) at presentation to 0.48 (IQR: 0.097-2.1) LogMAR at most recent follow-up (P = 0.0003); VA improved in 43 (68.2%) eyes. Retinal detachment occurred in 15 (23.1%) eyes. Initial TAI versus PPV did not influence final VA. Reported fentanyl use significantly increased the odds of off-chart (less than or equal to counting fingers) VA outcome (odds ratio: 5.25, 95% confidence interval: 1.3-20.9, P = 0.012) at the most recent follow-up. Among 54 patients who were admitted, 37 (68.5%) saw the ACS while hospitalized. All eligible patients who saw the ACS were offered a MOUD. Of these 24 patients, 13 (54.2%) accepted a new MOUD. No MOUDs were initiated for inpatients or outpatients who did not see the ACS.

Addiction consult services provide potentially life-saving care for patients with IDU-EE. The incidence and visual outcomes of this condition may be worsening in correlation with fentanyl use, which confers higher odds of poor visual outcomes. Initial vitrectomy may facilitate pathogen identification.

Proprietary or commercial disclosure may be found in the Footnotes and Disclosures at the end of this article.