He reported that 17 cases of presumed infectious endophthalmitis/TASS occurred among 9,510 cataract surgery cases performed over a 2-year period between April 2003 and March 2005. Fourteen surgeons contributed to the total caseload; seven of the surgeons had one or more cases of endophthalmitis/TASS.
Fourteen of the 17 cases were thought to be TASS, and the cases occurred in three clusters that each led to the introduction of new control methods. First there was a change from multi-to single-use phacoemulsification tips. After the second outbreak, wrapped cycle sterilization was instituted, and finally, a new protocol for culturing and cleaning of short-cycle sterilizer reservoirs was launched. During the last 7 months of the study since the latter changes were made, there were no new cases of endophthalmitis/TASS, reported Dr. Holland, clinical professor of ophthalmology and visual sciences, University of British Columbia, Vancouver, Canada.
"Short-cycle sterilization is bactericidal, but we know from our experience investigating diffuse lamellar keratitis (DLK) outbreaks that endotoxin survives. Our testing of the short-cycle sterilizer reservoirs showed they were heavily contaminated with endotoxin-producing bacteria in the biofilm, and so use of instruments coated with endotoxin may be a cause of TASS," Dr. Holland continued.
For the purpose of this study, any cases of postoperative inflammation with clinical features suggestive of classic infectious endophthalmitis or TASS were included as an event in the outbreak analysis. Based on the features of the cases and their outcomes, Dr. Holland suggested that 14 were likely TASS. Seven presented in the first 24 hours after surgery, 13 were associated with a final best spectacle-corrected visual acuity (BSCVA) of better than 20/40, there was no correlation found with intraocular vancomycin use, and of 12 cases where culture was performed, only one was positive.
The cases were performed at two hospitals. One had closed during month 12 of the study period and all subsequent cases were done at the second hospital. When the facility move was made, all equipment was transferred from the first hospital, including the instruments and tabletop sterilizers.
After a first case in the first month of the study, there were no more cases until months 6 and 7 when four cases occurred, all involving patients of a single surgeon. The hospital-based control investigation led to the recommendation for changing from multi-use to single-use phaco tips, and the transition was implemented in month 9. At the time, a suggestion was also made to reduce the use of short-cycle sterilization, based on the finding of gram-negative bacteria growth in cultures.
"Our sterilization techniques consisted of a combination of full-cycle and short-cycle steam sterilization, but the latter was the predominant method because it was quick, could be done with unwrapped instruments, and we had a limited number of instrument sets. However, no action was taken regarding the recommendation to reduce its use," Dr. Holland said.
The second outbreak occurred between months 10 and 13 and included six cases. Changes made in response to that cluster included use of disposable cannulas and sterilization of all sets with a wrapped cycle. Only about 5% of instruments were still flash sterilized. Between months 13 and 17, there were an additional six cases. Culturing of the three tabletop sterilizers again yielded heavy growth in all of gram-negative bacteria, including Sphingomonas paucimobilis, Ralstonia picketii, and Pseudomonas aeruginosa. The ultrasound bath was also contaminated with the same organisms.