Eyeworld

EW CATARACT January 2011 31 Keying in on causative agents and outcomes O nly about half of patients with delayed-onset, bleb- associated endophthalmi- tis have acuity of 20/400 or better, even with timely treatment, according to Theodore Leng, M.D., clinical assis- tant professor, ophthalmology, Stan- ford University, Stanford, Calif. In the study e-published in the Septem- ber 2010 issue of Retina, investiga- tors set out to highlight the organisms and outcomes in such en- dophthalmitis cases. Investigators were spurred to consider this by the troubling out- comes that usually result from de- layed onset cases. "Infection after trabeculectomy for glaucoma sur- gery is a very devastating complica- tion, and it usually happens later on, years after mechanical surgery," Dr. Leng said. "We wanted to look at the clinical characteristics of the cases that had been presented from 1996 to 2008 at Bascom Palmer [Eye Institute, Miami]. We looked at the organisms involved, the antibiotic resistance patterns, and the clinical outcomes of those cases." Also, because a previous study at Bascom Palmer had looked at en- dophthalmitis cases at the institu- tion from 1969-1984 and from 1989-1995, investigators wanted to update the numbers. Causative organisms Included in the retrospective consec- utive case series were 71 eyes from 68 Bascom Palmer patients. "The Streptococcus species was the most common causative organism identi- fied, which was followed by gram negative organisms," Dr. Leng said. "Another key finding was that all those gram positive organisms re- sponded to vancomycin." Dr. Leng sees these results as offering practi- tioners a bit of relief because the standard of care for a case of en- dophthalmitis is to inject van- comycin. "We can continue giving vancomycin and know that all the gram positives will be covered by it," Dr. Leng said. However, this good news is tem- pered by poor eventual outcomes. "Even with timely treatment of the infection, either by tap injector or with vitrectomy, only about 50% or so of the patients had final visual acuity greater than 20/400," Dr. Leng said. "Even though we're able to cure the infection and treat the inflammation, only about half the patients had decent vision—that means it's a pretty devastating infec- tion to get." Investigators also considered which bacteria were the most preva- lent here and compared these to rates found in the earlier case series. "The number of Streptococcus species actually decreased percentage-wise from the first series," Dr. Leng said. "From 1969 to 1984 it was about 57%, and it was 30% in our recent series." However, the number of Staphylococcus species and gram neg- atives remained similar over the years. Dr. Leng thinks that this rep- resents a possible shift in the distri- bution of the microbial flora in southern Florida. Visual shift In terms of visual outcomes, there has only been a mild shift over the years. Dr. Leng found that results in the latest series were only slightly better than those in the earlier se- ries. In the earlier series, vision of greater than 20/400 was at 32% ver- sus 52% in the most recent investi- gation. Dr. Leng sees this decrease as possibly resulting from a couple of factors. "One is that vitrectomy sur- gery is more efficient now and safer than it was 20 years ago," he said. "Also, maybe because doctors are recognizing that this is a potentially devastating ocular infection and treating it more aggressively, the in- fection and inflammation is being reduced at a faster rate." Despite such improvements, Dr. Leng is surprised at how many pa- tients still had very poor vision. "In the end about half of the people had worse than 20/400 vision," he said. "I think that that speaks to the viru- lence and strength of the organisms involved versus Staphylococcus epider- midis, which is more associated with cataract and intravitreal injection in- fections." Staphylococcus epidermidis patients generally have much better visual acuity outcomes. Unfortunately, there is nothing that can be done to head off late bleb leaks. Dr. Leng thinks that this has been the main impetus for the development of new ways of doing glaucoma surgery. "At Bascom Palmer the rate of tube shunt sur- gery with Baerveldt glaucoma im- plants [Abbott Medical Optics, Santa Ana, Calif.] has increased," he said. "I think that glaucoma surgeons be- lieve they have less long-term com- plications like bleb leaks and thin cystic blebs, which tend to be a vehi- cle for the bacteria to cross into the eye." He pointed out that there are companies developing new surgical techniques like the Trabectome (Neomedix, Tustin, Calif.) and visco- canaloplasty, as well as newer tech- nologies, aimed at reducing the potential for a bleb and its associ- ated leakage. Overall, Dr. Leng stressed the importance of heading off delayed- onset endophthalmitis. "The take- home message is that you don't want patients to get one of these in- fections; if they do, the probability that they're going to have a good vi- sual outcome is poor," he said. "As a glaucoma surgeon you should try to prevent bleb leaks and thinning out of the bleb as much as possible." EW Editors' note: Dr. Leng has no financial interests related to his comments. Contact information Leng: 650-721-6888, tedleng@stanford.edu by Maxine Lipner Senior EyeWorld Senior Contributing Editor Eyeing late bleb-associated endophthalmitis Example of endophthalmitis; Dr. Leng et al. evaluated organisms and outcomes in several sim- ilar cases Source: Daljit Singh, M.D. continued from page 30 phisticated a theoretical formula, the more information it uses to try to come to a conclusion, but it is still based on prior experience and can come to the wrong conclusion. Therefore, I believe that theoret- ical formulas are not the way IOL calculations are going to be made for every single patient, as they still in- volve making assumptions. What ophthalmology as a specialty needs to do is abandon these older theoret- ical formulas based on the scientific model where every measurement needs to be accurate and every as- sumption needs to be correct, which is impossible. Instead, ophthalmology needs to switch over to what is called an engineering statistical- based for- mula, which is mathematically an entirely different exercise. This is where IOL calculation is going to be headed in the next few years. EW Editors' note: Dr. Hill has no financial interests related to this article. Contact information Hill: 480-981-6111, hill@doctor-hill.com

• Cover