Abstract
Purpose :
Endophthalmitis is a severe complication following intravitreal injections that can cause irreversible blindness. Although complication rates are low, it is unknown to what extent a teaching setting affects the risk and outcomes of post-injection endophthalmitis. We conducted a retrospective case series of post-injection endophthalmitis cases at the Nashville Veterans Affairs (VA) Hospital where resident physicians perform intravitreal injections under the guidance of attending retina surgeons. Our hypothesis was that outcomes and complications of intravitreal injections in a VA teaching setting would be comparable to those previously reported in the ophthalmic literature.
Methods :
From July 1, 2015 to October 31, 2018, patients were identified based on CPT code for an intravitreal injection and ICD-9 and ICD-10 codes for endophthalmitis. Patient records were individually reviewed to confirm that an intravitreal injection preceded the onset of endophthalmitis. Data recorded included patient’s visual acuity at time of injection, final vision after endophthalmitis treatment, culture status, treatment, and training level of the physician who administered the injection.
Results :
From the 9,645 intravitreal injections performed, 6 cases (0.062%) of endophthalmitis occurred. Three cases were associated with an injection performed by a resident in his or her fourth year of post-graduate training while the remaining 3 were performed by an attending retina surgeon. Vitreous culture and gram stain confirmed an infectious etiology in four cases. The rate of culture growth was similar between cases associated with resident-performed and attending-performed injections (p = 1). All cases received intravitreal antibiotics at presentation, and surgical intervention was required in 3 cases. At final follow up, only two cases returned to baseline visual acuity. There was no difference in rate of culture growth between patients that regained baseline vision and those that did not (p = 0.4).
Conclusions :
In this teaching clinic, the rate of endophthalmitis following intravitreal injections was comparable to rates previously reported. Visual outcomes were variable and did not correlate with provider training level nor with culture-positivity. Further study will be required to examine other factors influencing rate of endophthalmitis and visual outcomes after injection-related endophthalmitis.
This abstract was presented at the 2019 ARVO Annual Meeting, held in Vancouver, Canada, April 28 - May 2, 2019.