The effect of speed of intravitreal antibiotic administration on the outcome of post-operative endophthalmitis treated at the Birmingham and Midland Eye Centre, United Kingdom

Session Details

Session Title: Inflammation in Ocular Infections

Session Date/Time: Wednesday 17/09/2014 | 08:00-09:30

Paper Time: 08:12

Venue: Capital Hall A

First Author: : A.Gao UK

Co Author(s): :    A. Gupta   S. Mirza           

Abstract Details

Purpose:

Cataract surgery is one of the most commonly performed surgical procedures in the United Kingdom (UK). Endophthalmitis is a rare but devastating complication, where a delay in the diagnosis and treatment is associated with worse visual prognosis for the patient. Prompt administration of intravitreal antibiotics is thought to be paramount in the treatment of postoperative endophthalmitis. The Birmingham and Midland Eye Centre (BMEC) is a large tertiary referral centre, where most cases of suspected endophthalmitis in the West Midlands are treated. In March 2010, a designated procedure room was set up in the eye casualty department to safely and promptly treat cases of suspected endophthalmitis with intravitreal antibiotics and to take aqueous and vitreous samples for microbiological testing. This study aims to compare the door-to-needle time, visual outcome, long-term complications, culture positive rates and the safety aspect between patients who received their intravitreal antibiotics in the casualty treatment room and those who received it in the operating theatre, at BMEC.

Setting:

The Birmingham and Midland Eye Centre, Birmingham, United Kingdom (UK).

Methods:

This was a retrospective study on all patients who were treated for post-cataract endophthalmitis at BMEC between May 2007 and April 2013 were included. Patient demographics, past ocular history, details of cataract surgery, door-to-needle time (i.e. time from registering at eye casualty to the time the patient received their intravitreal antibiotics), causative organisms and antibiotic sensitivity data, treatment, visual outcome and complications were collected. Patients were identified from the procedure room logbook and theatre logbook. Clinical notes and microbiological data were collated for each patient.

Results:

A total of 44 patients were included, 19 had their intravitreal antibiotics in Eye Casualty and 25 in the operating theatre. Mean time from surgery to presentation was 10 days (range 2-29 days) in the casualty group and 28 days (range 3-183 days) in the theatre group. 95% (18/19) of patients in the casualty group had uncomplicated phacoemulsification and lens implant whereas only 76% (19/25) of patients in the theatre group had uncomplicated surgery. The door-to-needle time was significantly less in the casualty group (mean=161 min., range 77-127 min) compared to the theatre group (mean=308 min, range 137-633 min). More patients in the theatre group had both aqueous and vitreous samples taken compared to the casualty group (88% vs 53%). Culture negative rate was higher in the casualty group (42%) compared to the theatre group (28%). Of all the culture positive cases, gram-positive organisms accounted for 72% (21/29). Difference between the final best corrected visual acuity (BCVA) and the worst BCVA during treatment was similar in both groups. 5% of patients in the casualty group went on to have vitrectomies whereas 20% of the theatre group had further surgical intervention. Long-term complications also occurred more frequently in the theatre cohort.

Conclusions:

Patients were managed much more promptly if treatment was initiated in the casualty setting, with a much shorter door-to-needle time. Combining the data for both groups, the proportion of patients who had visual improvement declined with longer door-to-needle time, and they were also more likely to have culture negative results. Patients were more likely to have both aqueous and vitreous samples taken if this was carried out in the operating theatre, suggesting that better microscope and equipment in the casualty procedure room may be required. Causative organisms were predominantly gram-positive, although the proportion was lower in our study compared to previous literature. The theatre group patients required more vitrectomies and had a higher percentage of long-term complications compared to the casualty group. This may be due to a delay in the administration of intravitreal antibiotics but other factors such as higher rate of intra-operative complications during cataract surgery, possibly more virulent organisms in this group, could also play a part. In conclusion, the treatment of infective endophthalmitis appears to be safe and prompt when carried out in the casualty treatment room compared to sampling in the theatre. Culture positive rate is still unsatisfactory and the use of polymerase chain reaction (PCR) for rapid diagnostic purposes may aid this in the future.

Financial Interest:

NONE