EyeWorld is the official news magazine of the American Society of Cataract & Refractive Surgery.
Issue link: https://digital.eyeworld.org/i/111385
74 EW RESIDENTS February 2013 EyeWorld journal club Review of ���Residual lens cortex material: Potential risk factor for endophthalmitis by Christopher S��les, M.D., Charles Yu, M.D., Lingmin He, M.D., and Douglas Fredrick, M.D., residency program director Douglas Fredrick, M.D., residency program director, Department of Ophthalmology, Stanford University School of Medicine The February JCRS includes this interesting study regarding retained cortex and risk of endophthalmitis. I asked the Stanford residents whether the conclusions are valid. David F. Chang, M.D., chief medical editor T he risk of posterior capsular rupture with associated vitreous loss prompts cataract surgeons to leave lens cortex in the capsular bag if complete cortical cleanup is difficult and threatens undue complications. This calculus is based, in part, on the commonly held belief that leaving residual cortex is safe because it causes few clinically significant consequences beyond a transient increase in post-op intraocular inflammation. Lou and colleagues postulate whether this assumption should be re-evaluated, hypothesizing that ���residual lens cortex ��� promotes bacterial propagation in the anterior chamber [and increases] the risk for endophthalmitis after cataract surgery.���1 The authors conducted a study consisting of two experiments. In the in vitro experiment, they inoculated human aqueous humor and serial dilutions of homogenized lens cortex suspension with either S aureus or S epidermidis, the two most common causes of bacterial endoph- Residual lens cortex material: Potential risk factor for endophthalmitis after phacoemulsification cataract surgery Bingsheng Lou, M.D., Xiaofeng Lin, M.D., Lixia Luo, M.D., Yao Yang, M.D., Yanting Chen, M.D., Yizhi Liu, M.D. J Cataract Refract Surg (February) 2013; 39:250-257 Purpose: To evaluate the bacterial growth potential of residual lens cortex after phacoemulsification with the development of endophthalmitis. Setting: University medical center. Design: Experimental study. Method: Staphylococcus aureus and Staphylococcus epidermidis were dispensed into aqueous humor or serial dilutions of lens cortex that were obtained from cataract patients during phacoemulsification. After a 24-hour incubation, the colony-forming unit (CFU) was quantified. Eighty rabbits had phacoemulsification. Complete lens cortex removal was performed in 40 rabbits, while a quarter of lens cortex was retained in 40 rabbits. Staphylococcus aureus, with an inoculum size of 32 CFU, 56.3 CFU, and 108.6 CFU, was injected intracamerally at the conclusion of surgery and the production of endophthalmitis was measured 72 h later. The aqueous and vitreous humor samples were collected for microbiological diagnosis. Results: A statistically significant increase in bacterial growth of S aureus and S epidermidis was observed in each dilution of lens cortex compared with in aqueous humor (P<.001, Student t test). With inoculum of 32.0 CFU or 56.3 CFU of S aureus, there was a statistically higher incidence of bacterial culture-proven endophthalmitis in the residual lens cortex eye group than in the normal aphakic eye group (P<.05, Fisher exact test). Conclusions: Lens cortex was associated with a significant increase in bacterial growth compared with aqueous humor. Therefore, eyes with residual lens cortex seem more prone to develop into endophthalmitis if anterior chamber bacterial contamination occurs during phacoemulsification. Financial disclosure: No author has a financial or proprietary interest in any material or method mentioned. thalmitis after cataract surgery.1 Every dilution of lens cortex grew more colony forming units per mL than the aqueous control. In the in vivo experiment, the authors performed unilateral phacoemulsification cataract surgery on 80 New Zealand albino rabbits and left the eyes aphakic. Forty of the subjects received incomplete lens cortex removal and were designated the residual cortex group; the other 40 received complete cortex removal. After cataract surgery, 30 subjects from each group were injected with one of three predetermined aliquots of S aureus or S epidermidis, and the remaining 10 were injected with sterile balanced salt solution. The residual cortex group had a higher incidence of clinical and culture-positive endophthalmitis among eyes injected with bacteria, but also had a higher incidence of culture-negative endophthalmitis among eyes injected with balanced salt solution. No eyes grew evidence of intraoperative contamination prior to the injection of bacteria or balanced salt solution. Based on the above findings, the authors conclude that human lens cortex facilitates bacterial growth in vitro, possibly by acting as a nutrition source, and that as a result, residual cortex increases the incidence of endophthalmitis after phacoemulsification cataract surgery. ���To prevent endophthalmitis,��� they recommend, ���complete removal of the lens cortex ��� during surgery.��� Although we applaud the rigor of their experiments, we believe that the authors��� recommendations may be overstated because of limitations inherent to their chosen models. By homogenizing lens cortex into a suspension, the authors maximized the human tissue���s surface-to-volume ratio and, in turn, its bioavailability to bacteria. This model may be optimal for promoting bacterial growth, but it does not mimic what happens in vivo after incomplete cortical cleanup, where residual lens cortex is a discrete mass in a volume of sterile aqueous that is undergoing continuous turnover at a rate of 1% per minute. If pieces of whole lens cortex in aqueous were inoculated with bacteria, and the volume of aqueous was intermittently replaced with a sterile sample every 100 minutes, the experiment would have been more realistic and applicable to clinical practice. Given the model���s bias toward promoting bacterial growth, it is unsurprising that an increased number of bacterial colony forming units grew from the human lens cortex. However, perhaps any human tissue, if homogenized into a stagnant suspension, would increase the number of colony forming units after being inoculated with bacteria. It is also questionable whether the conclusions drawn from the authors��� rabbit model can be generalized to humans. The New Zealand albino rabbit is the most commonly used animal in lens and cataract surgery experiments because it is docile and its eyes are relatively large with easily visualized ocular structures.2 But there are known limitations to this model, including an inflammatory response that is more vigorous than that observed in humans if the rabbit is young and weighs less than or equal to 2 kg. Adult rabbits weighing 3-4 kg are thought to most closely replicate the human ocular inflammatory response to cataract surgery2; however, the authors used 1.75-2.25 kg rabbits. The model���s bias toward a vigorous inflammatory response is seen in the results of the residual cortex group, which had a higher incidence of sterile endophthalmitis even after no bacteria were injected. These findings validate previous studies of the rabbit model, which showed that homogenized cortex injected into rabbit eyes induces inflammation.3 The limitations inherent to rabbit models notwithstanding, we believe that cataract surgeons��� clinical experience confirms what has been demonstrated in animals���most would agree that leaving behind residual cortex can increase intraocular inflammation in the post-op period. Thus, sterile inflammation may be an independent reason to consider complete cortical removal even if such inflammation has not been shown to increase the risk of bacterial endophthalmitis in humans. The overall incidence of endophthalmitis after cataract surgery is approximately 0.06-0.27%,4-6